USF Libraries
USF Digital Collections

Studies in rhodium catalyzed intramolecular C-H insertion of amino acid derived α-diazo-β-(substituted)acetamides and it...

MISSING IMAGE

Material Information

Title:
Studies in rhodium catalyzed intramolecular C-H insertion of amino acid derived α-diazo-β-(substituted)acetamides and its application to the total synthesis of clasto-lactacystin β-lactone
Physical Description:
Book
Language:
English
Creator:
Flanigan, David L
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
γ-lactam
Natural product
C-H activation
Aldol
Proteasome inhibitor
Dissertations, Academic -- Chemistry -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Lactacystin is a microbial metabolite isolated by Omura that exhibits neurotrophic activity in neuroblastoma cell lines. Lactacystin and especially its β-lactone analog are the first examples of non-polypeptide small molecules capable of specifically inhibiting the 20S proteasome. Various asymmetric total syntheses of lactacystin and its analogs have been reported. The total synthesis of clasto-lactacystin β-lactone is achieved using L-serine methyl ester as the starting material and the sole source of stereochemical induction. The success of this synthesis hinges on two featured transformations.The first key step involves formation of the γ-lactam core via rhodium (II) catalyzed intramolecular C-H insertion of the α-diazo-α-(phenylsulfonyl)acetamide intermediate. The methodology for this transformation has been developed and applied to the synthesis of highly functionalized stereogenic γ-lactams from natural α-amino acids. Three control elements that govern γ-lactam formation are described. This step is highlighted by the simultaneous creation of two stereogenic centers of the γ-lactam core. The second key step involves the late stage aldol coupling for quaternary carbon formation and installation of the hydroxyisobutyl group. In all previously reported syntheses, this is the very first aspect which is addressed. The stereochemical outcome of this step is directed by the chiral environment of the enolate itself. Various attempts to achieve selectivity are explored and reported.Completion of the synthesis of clasto-lactacystin β-lactone requires 17 steps with an overall yield of 10%. Some general attempts for optimizing the synthetic scheme are discussed as well as the future direction of this research.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2004.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by David L. Flanigan Jr.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 179 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001479810
oclc - 56549620
notis - AJS3941
usfldc doi - E14-SFE0000418
usfldc handle - e14.418
System ID:
SFS0025110:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Studies in Rhodium Catalyzed Intramolecula r C-H Insertion of Amino Acid Derived Diazo-substituted)acetamides and its Application to the Total Synthesis of clastoLactacystin -Lactone by David L. Flanigan Jr. A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: Kyung Woon Jung, Ph.D. Edward Turos, Ph.D. Kirpal S. Bisht, Ph.D. Julie P. Harmon, Ph.D. Date of Approval: May 24, 2003 Keywords: -lactam, natural product, C-H activation, aldol, proteasome inhibitor Copyright 2004, David L. Flanigan Jr.

PAGE 2

For Liz, Bren and Kristen

PAGE 3

Acknowledgments First, I need to thank God for providing me w ith the ability to succeed in this field, without Him none of this is possible. Liz, you help me to realize my own poten tial and never underestimate my ability. Thank you for always believing in me and putting up with my “nonsense.” Bren, you constantly bring a smile to my face despite hard days of failu re and nothing has changed. My family, both immediate and extended, have been supportive in countless ways. They have been friends, hosts, sources of joy and voices of reason (and not :). Th ey have assisted with financial issues and furnished houses. Obviously, they have played an extraordinary part in making my life happier and easier and to them I am forever grateful. Dr. Jung, thank you for your encouragement and support during the last four years. Your guidance and criticism, though not always understood at the time, has changed me for the better both personally and professionally. I appreciate you br inging me into your lab and allowing me to study under your guidance. There are so many other people who deserve mention in this part that I’ll most likely run out of room. My committee members and faculty/staff friends in the department always support and assist me and I appreciate and value each and every one of them. Those friends that I met and spent so much time with are, first and foremost, Dr. Jay Parrish and Dr. Advait Nagle. You guys are truly special to me. Of course, Dr. Yoon taught me how to draw (really, it is important to us). There are also those who I saw every day and shared a laugh or vent session with. They are, in no special order, Young, Chiliu, Dr. Mishra, Rob, W oogie, Kisoo, Dr. Yoo, Michelle, Iris, Matt, Chong, Vince, Dr. Reddy, Dr. Reddy, Dr. Hwang, Dr. Riley, Christina, Thushara, Hla, Santhisree, Cindy and anyone else who slips my mind at the moment. Though the work was hard, we had some really good times!

PAGE 4

i Table of Contents List of Tables iii List of Figures iv List of Schemes v List of Abbreviations viii Abstract xv Chapter One: Introduction 1 1.1 Background 1 1.2 Kinetic Inhibition Studies 2 1.3 Outlook 4 Chapter Two: Previous Syntheses 6 2.1 Corey’s First Total Synthesis of Lactacystin 6 2.2 Corey’s Extended Methodology 9 2.3 Smith’s Total Synthesis of Lactacystin 10 2.4 Corey’s Expanded Studies 12 2.5 Baldwin’s Total Synthesis of Lactacystin 12 2.6 Chida’s Total Synthesis of Lactacystin 14 2.7 Corey’s Second Total Synthesis of Lactacystin 16 2.8 Kang’s First Total Synthesis of Lactacystin 18 2.9 Kang’s Second Total Synthesis of Lactacystin 19 2.10 Adam’s Total Synthesis of Lactacystin Analog PS-519 21 Chapter Three: Stereogenic -Lactams via Rhodium Catalyzed Intramolecular C-H Insertion 25 3.1 C-H Insertion of the Acyclic System 25 3.2 C-H Insertion of the Cyclic System 28 3.3 C-H Insertion of Amino Acid Derived -Diazo--(phenylsulfonyl) acetamides 31

PAGE 5

ii 3.4 C-H Insertion of -Diazo--(substituted)acetamides 35 3.5 C-H Insertion of a Conformationally Constrained Cyclic System 42 Chapter Four: Jung’s Total Synthesis of clastoLactacystin -Lactone 45 4.1 Retrosynthetic Analysis of clastoLactacystin -Lactone 45 4.2 First Generation Synthesis of the Bicyclic -Lactam Intermediate 46 4.3 Functionalization of the -Lactam Core 48 4.4 Second Generation Synthesis of the Bicyclic -Lactam Intermediate 50 4.5 Stereoselective Functionalization of the Bicyclic -Lactam 53 4.6 Elucidation of Aldol Stereoselectivity 56 4.7 Rationalization for Aldol Coupling Stereoselectivity 58 4.8 Endgame for the Total Synthesis of clastoLactacystin -Lactone 59 Chapter Five: Experimental Data 62 References 108 Appendices 113 Appendix A: Selected 1H NMR and 13C NMR Spectra 114 Appendix B: X-ray Crystallographic Data 157

PAGE 6

iii List of Tables Table 3-1 C-H Insertion of -Diazo Substrates at the Methylene Position 31 Table 3-2 C-H Insertion of -Diazo Substrates at Methyl and Methine Positions 33 Table 3-3 C-H Insertion of -Diazo Substrates at Electron Rich Methylene Positions 34 Table 3-4 C-H Insertion of -Diazo Substrates with Varied -Substituents 37 Table 3-5 C-H Insertion of -Diazo Norvaline Derivatives 41 Table 3-6 C-H Insertion of Conformationally Constrained N,N’Acetonide Substrates with Varied -Substituents 43

PAGE 7

iv List of Figures Figure 1-1 Schematic Diagram of the 20S Proteasome Upon Inhibition by Lactacystin 1 Figure 1-2 Kinetic Inhibition Studies of Lactacystin Analogs 3 Figure 1-3 Kinetic Inhibition Studies of -Lactone Analogs 4 Figure 2-1 Chelate Model for the Double Diastereoselective Mukaiyama Aldol Coupling 9 Figure 2-2 Stereochemical Rationalization of the Doubly Diastereoselective Aldol Coupling 23 Figure 4-1 Transition States for Quaternary C-5 Formation 58 Figure 4-2 Rationalization of Aldol Coupling Stereoselectivity 59

PAGE 8

v List of Schemes Scheme 2-1 Corey's Synthesis of the First Key Lactacystin Intermediate 7 Scheme 2-2 Completion of the First Total Synthesis of Lactacystin 8 Scheme 2-3 The Smith-Omura Total Synthesis of Lactacystin 10 Scheme 2-4 Baldwin's Total Synthesis of Lactacystin 13 Scheme 2-5 Chida's Total Synthesis of Lactacystin 15 Scheme 2-6 Corey's Second Enantioselective Total Synthesis of Lactacystin 17 Scheme 2-7 Kang's First Synthesis of a Key Intermediate of Lactacystin 18 Scheme 2-8 Kang's Second Synthesis of a Key Intermediate of Lactacystin 20 Scheme 2-9 Synthesis of the transOxazoline Coupling Partner 21 Scheme 2-10 Synthesis of the -Alkyl--formyl Amide Coupling Partner 22 Scheme 2-11 Endgame of Adams’ Total Synthesis of PS-519 23 Scheme 3-1 Rhodium Catalyzed Intramolecular C-H Insertion of -Diazo--(phenylsulfonyl)acetamides 26 Scheme 3-2 Conformational and Stereochemical Effects on C-H Insertion 27 Scheme 3-3 Synthesis Stereogenic -Lactams From -Amino Acids 28

PAGE 9

vi Scheme 3-4 Conformational and Stereochemical Effects on C-H Insertion of -Amino Acid Derivatives 29 Scheme 3-5 Conformational Requirements for Intramolecular C-H Insertion 30 Scheme 3-6 Functionalized Stereogenic -Lactams are Versatile Synthetic Intermediates 34 Scheme 3-7 Improved Synthetic Protocol for Analog Synthesis 36 Scheme 3-8 Pathway for C-H Insertion of -Diazo Substrates Resulting in Aromatic Cycloaddition 38 Scheme 3-9 Conformational Constraint of Cyclic System Inhibits -Lactam Formation 39 Scheme 3-10 Synthetic Protocol for Norvaline Analog Synthesis 40 Scheme 3-11 Synthesis of N,N’Acetonide Substrates with Varied -Substituents 42 Scheme 4-1 Retrosynthetic Analysis of the Total Synthesis of clastoLactacystin -Lactone 46 Scheme 4-2 Amide Coupling of (L)-Serine Methyl Ester 47 Scheme 4-3 Installation of the N,OCyclic System 47 Scheme 4-4 Intramolecular C-H Insertion of the (L)-Serine Derived -Diazo--(phenylsulfonyl)acetamide 48 Scheme 4-5 Functionalization of the -Lactam 48 Scheme 4-6 Functionalization of the C-5 center 49 Scheme 4-7 Unsuccessful Aldol Coupling 49 Scheme 4-8 Synthesis of the Acetonide Derivative of (L)-Serine 51 Scheme 4-9 Synthesis of the -Diazo--(phenylsulfonyl)acetamide 51 Scheme 4-10 -Lactam Formation via Rhodium Catalyzed Intramolecular C-H Insertion 52

PAGE 10

vii Scheme 4-11 Functionalization of the C-3 Center 53 Scheme 4-12 Selective -Methylation of the C-3 Center 54 Scheme 4-13 Functionalization of the C-5 Center 54 Scheme 4-14 Stereoselective Formation of the Quaternary C-5 Center 55 Scheme 4-15 Formation of the Quaternary C-5 Center of the 3 R Intermediate 55 Scheme 4-16 Oxidation of the C-9 Stereocenters for Structural Elucidation 57 Scheme 4-17 Acylation of the Iminoether Occurs with Enhanced Selectivity 57 Scheme 4-18 Endgame for the Total Synthesis of clastoLactacystin -Lactone 60

PAGE 11

viii List of Abbreviations 1,2-DCE 1,2-dichloroethane 13C NMR C-13 nuclear magnetic resonance 1H NMR proton nuclear magnetic resonance 2,2-DMP 2,2-dimethoxypropane 2,6-lut 2,6-lutidine Ac acetyl Ac2O acetic anhydride AcOH acetic acid AD-mixasymmetric dihydroxylation mix BF3OEt2 boron trifluoride diethyl etherate Bn benzyl BnBr benzyl bromide BOC tbutoxy carbonyl BOMCl benzyloxymethyl chloride BOPCl bis(2-oxo-3-oxazolidinyl)phosphinic chloride BrCH2COBr bromoacetyl bromide

PAGE 12

ix CBz benzyloxy carbonyl CDCl3 chloroformd CF3CH2OH trifluoroethanol (CH2O)n paraformaldehyde CH2(OMe)2 dimethoxymethane CH2Cl2 methylene chloride CH2N2 diazomethane CH3COBr acetyl bromide Cl3CCN trichloroacetonitrile ClCO2C(CH3)=CH2 isopropenyl chloroformate ClCOCH2CO2Et ethyl 4-chloroacetoacetate cm-1 wavenumbers (COCl)2 oxalyl chloride CsF cesium fluoride DBU 1,8-diazabicyclo[5.4.0]undec-7-ene DCC N,N’dicyclohexylcarbodiimide de diastereomeric excess DIBAL-H diisobutylaluminum hydride DIC N,N’diisopropylcarbodiimide DIPEA diisopropylethylamine DMF N,Ndimethylformamide DMS dimethylsulfide ee enantiomeric excess

PAGE 13

x Et ethyl Et2NH diethylamine EtOAc ethyl acetate EtOH ethanol H2 hydrogen H2CO formaldehyde H2CrO4 chromic acid H2SO4 sulfuric acid HCl hydrochloric acid HCO2H formic acid HCO2NH4 ammonium formate HF hydrofluoric acid Hg(O2CCF3)2 mercury (II) trifluoroacetate HONH2HCl hydroxylamine HCl salt HS(CH2)3SH 1,3-propanedithiol Imid imidazole IPA isopropanol i-Pr isopropyl i-PrCHO isobutyraldehyde i-PrMgBr isopropylmagnesium bromide Jones Reagent H2CrO4 solution K2CO3 potassium carbonate

PAGE 14

xi KMnO4 potassium permanganate LAH lithium aluminum hydride LDA lithium diisopropylamide LHMDS lithium bis(trimethylsilyl)amide LiBH4 lithium borohydride LiBr lithium bromide LiOH lithium hydroxide LiOOH lithium peroxide M molarity mCPBA 3-chloroperoxybenzoic acid Me methyl Me2AlCl dimethylaluminum chloride Me2CO acetone Me3OBF4 trimethyloxonium tetrafluoroborate MeCN acetonitrile MeI iodomethane MeNH2 methylamine MeOH methanol (MeSO2)2O methanesulfonic anhydride MHz megahertz MgI2 magnesium iodide mL milliliter MnO2 manganese dioxide

PAGE 15

xii MOM methoxymethyl MOMCl methoxymethyl chloride N Normality N nitrogen Na(Hg) sodium-mercury amalgam NaHCO3 sodium bicarbonate Na2SO4 sodium sulfate NaBH(OAc)3 sodium triacetoxyborohydride NaBH4 sodium borohydride NaClO2 sodium chlorite NaH sodium hydride NaIO4 sodium periodate NaN3 sodium azide NaOH sodium hydroxide NaOMe sodium methoxide nBu normal butyl NEt3 triethylamine NH3 ammonia NMO Nmethylmorpholine Noxide NMR nuclear magnetic resonance NPMBGlyME NPMB-glycine methyl ester O3 ozone

PAGE 16

xiii OsO4 osmium tetroxide P2O5 phosphorous pentoxide pABSA 4-acetamidobenzenesulfonyl azide Pd(OH)2/C palladium hydroxide on carbon Pd(Ph3P)4 tetrakis(triphenylphosphine)palladium(0) Pd/C palladium on carbon Ph phenyl PhCH3 toluene PhSCH2CO2H (phenylthio)acetic acid PhSeBr phenylselenyl bromide PhSO2Na benzenesulfinic acid sodium salt PivCl pivaloyl chloride PLE porcine liver esterase PMA phosphomolybdic acid PMB 4-methoxy benzyl pyr pyridine Rh2(cap)4 rhodium (II) caprolactamate dimer Rh2(OAc)4 rhodium (II) acetate dimer SnCl4 tin (IV) chloride SOCl2 thionyl chloride TBAF tetrabutylammonium fluoride TBSCl tbutyldimethylsilyl chloride

PAGE 17

xiv TBSOTf tbutyldimethylsilyl trifluoromethanesulfonate TBTU 2-(1 H -benzotriazol-1-yl)-1,1,3,3tetramethyluronium tetrafluoroborate tBu tert-butyl TEA triethylamine TEMPO 2,2,6,6-tetramethyl-1-piperidinyloxy TESCl triethylsilyl chloride TFA trifluoroacetic acid TfOH trifluoromethanesulfonic acid THF tetrahydrofuran Thr threonine TiCl4 titanium (IV) chloride TMOF trimethyl orthoformate TMSCl trimethylsilyl chloride Tol toluene TPAP tetrapropylammonium perruthenate TsOH toluenesulfonic acid

PAGE 18

xv Studies in Rhodium Catalyzed Intramolecular C-H Insertion of Amino Acid Derived -Diazo-substituted)acetamides and its Application to the Total Synthesis of clastoLactacystin -Lactone David L. Flanigan Jr. ABSTRACT Lactacystin is a microbial metabolite isolated by Omura that exhibits neurotrophic activity in neuroblastoma cell lines. Lactacystin and especially its -lactone analog are the first examples of non-polypeptide small molecules capable of specifically inhibiting the 20S proteasome. Various asymmetric total syntheses of lactacystin and its analogs have been reported. The total synthesis of clasto -lactacystin -lactone is achieved using L-serine methyl ester as the starting material and the sole source of stereochemical induction. The success of this synthesis hinges on two featured transformations. The first key step involves formation of the -lactam core via rhodium (II) catalyzed intramolecular C-H insertion of the -diazo--(phenylsulfonyl)acetamide intermediate. The methodology for this transformation has been developed and applied to the synthesis of highly functionalized stereogenic-lactams from natural -amino acids. Three control elements that govern -lactam formation are described. This step is highlighted by the

PAGE 19

xvi simultaneous creation of two stereogenic centers of the -lactam core. The second key step involves the late stage aldol coupling for quaternary carbon formation and installation of the hydroxyisobutyl group. In all previously reported syntheses, this is the very first aspect which is addressed. The stereochemical outcome of this step is directed by the chiral environment of the enolate itself. Various attempts to achieve selectivity are explored and reported. Completion of the synthesis of clastolactacystin -lactone requires 17 steps with an overall yield of 10%. Some general attempts for optimizing the synthetic scheme are discussed as well as the future direction of this research.

PAGE 20

1 Chapter One Introduction 1.1 Background Lactacystin ( 1 ) is a Streptomyces metabolite that was isolated and identified by mura and initially reported in 1991.1 It has been found to be an affector of neurite outgrowth in mouse neuroblastoma cell line Neuro 2A.2 The scientific community was excited by the discovery of the first non-protein microbial metabolite to exhibit neurotrophic activity and three total syntheses of the new agent were reported over the next 3 years.3-5 Corey reported that inhibition of the cell cycle of Neuro 2A and MG-63 human osteosarcoma cells past the G1 phase occurs upon treatment with lactacystin and related analogs 2-5 (Figure 1-2).6 The specific cellular target was identified by Schreiber Figure 1-1. Schematic Diagram of the 20S Proteasome Upon I n h i b i t i o n b y L ac t ac y s t i n Thr Lactacystin

PAGE 21

2 as the 20S proteasome using tritium-labled lactacystin analogs.7 The 20S proteasome is the catalytic core of the 26S proteasome responsible for the degradation of denatured and misfolded proteins. It is also instrumental in the degradation of regulatory proteins in charge of cellular growth and metabolism.8 Specifically it is known to exhibit chymotrypsin-like, trypsin-like and peptidylglutamyl-peptide hydrolyzing (PGPH) activities toward small peptides.7 The 20S proteasome is an arrangement of four stacked rings (and -units) consisting of seven protein subunits each (28 total protein subunits) reminiscent of a cylinder (Figure 1-1). Two threonine residues present on the -units of the 20S proteasome are responsible for much of its catalytic activity.9 Lactacystin and its analogs have shown to affect these sites by irreversibly acylating the N -terminus of the threonine residues rendering them inactive.7 Schreiber’s data is confirmed by X-ray crystallographic analysis of the 20S proteasome deactivated by lactacystin.10,11 1.2 Kinetic Inhibition Studies Extensive studies concerning the structural requirements for biological activity of the -lactam were performed. The obvious alteration points for lactacystin were the C-7 methyl, the hydroxyisobutyl group and variations of the N -acetylcysteine side chain. Schreiber’s initial report of the activity of lactacystin toward the 20S proteasome included a limited structural activity study of lactacystin ( 1 ) and related -lactone 2 dihydroxy acid 3 6-deoxy 5 6epi 7epi -lactacystin and deshydroxybutyl lactacystin analogs 4 (Figure 1-2).6 Kinetic inhibition data shows that the -lactam core and stereochemistry thereof are requirements for activity of the molecule. The dramatic

PAGE 22

3 increase in kinetic inhibition of the -lactone intermediate implies that the electrophilic nature of the carbonyl governs the rate of inhibition. For instance, the dihydroxy acid analog exhibits no activity presumably based on the absence of electrophilic character of the carboxyl moiety. The -lactone exhibits activity that is 15-fold more than that of lactacystin based on the increased electrophilicity of the lactone carbonyl as opposed to the carbonyl of the thioester. Smith preformed studies on analogs of the N -acetylcysteine side chain of lactacystin.12 Variations of the carboxylate function as well as deletion of the amide function had little effect on the activity of the analogs thus corroborating Schreiber’s presumption that the presence and reactivity of the C-4 carbonyl is crucial. Corey’s efforts toward elucidation of the ideal affector molecule focused on optimization of the C-7 alkyl and hydroxyisobutyl substituents of the more potent -lactone.13 Analogs prepared with variations to the hydroxyisobutyl substituent involved exchange of the isopropyl portion of the side chain for simple alkyls and unsaturated moieties and NH O HO O S CO2H AcHN HO NH O HO O S R' R HO 6 R = NHAc, R' = CO2Allyl 7 R = NHAc, R' = H 8 R = H, R' = CO2Et NH O OH O O NH O O S CO2H AcHN HO NH O HO O S CO2H AcHN H Smith Side-Chain Analogs C o r e y -S c h r eibe r A nalogs 1 Lactacystin 2 clasto -Lactacystin -lactone 4 Des(hydroxyisobutyl)lactacystin 5 6-Deoxylactacystin 3 clasto -Lactacystin dihydroxy acid Fi g u r e1-2. KineticInhibitionStudiesofLactac y stinAnalo g s NH HO O CO2H HO

PAGE 23

4 specifically 9-deoxy, 9-epi and 9-keto groups (Figure 1-3). Ultimately, the hydroxyisobutyl side chain proved to be nece ssary for sufficient inhibition, that which was 10-fold greater than the most active analog. Eventually, X-ray studies revealed that the isopropyl group of the hydroxyisobutyl side chain binds was bound by a hydrophobic pocket of the lactacystin-labeled proteasome subunit.14 Conversely, C-7 analogs of the lactone showed remarkable increase of inhibition when the methyl was exchanged for ethyl-, n -Bu and especially i-Pr alkyl substituents. Inhibition rates more than doubled with the introduction of longer alkyl chains. The 7-epi analog resulted in a decrease in activity reinforcing the notion that the original stereochemistry of the molecule is a requirement. 1.3 Outlook As the first non-protein neurotrophic fact or and proteasome inhibitor, lactacystin has garnered much attention from both the physical and life sciences. The effects this small molecule exerts on living systems makes it a valuable research tool for exploration NH O O O NH O R OH O O Corey's Hydroxyisobutyl Analogs R = R R = Et, i -Pr, n -Bu, epi -Me, Bn, gem -dimethyl HO H OH H OH H OH H O H O C o r e y 's C -7 A nalogs NH O n -Pr OH O O Adams' C-7 Target (PS-519) Figure 1-3. Kinetic Inhibition Studies of -Lactone Analogs79

PAGE 24

5 of protein biochemistry and molecular biol ogy. As Corey’s structure-activity project was submitted for publication, Adams was completing work that reported the synthesis of a highly active analog of lactacystin.15 A complimentary set of data supported Corey’s results and a -lactone incorporating an n -Pr C-7 substituent ( PS-519 ) was the highlight of Adams’ report. PS-519 went on to preclinical development by Millennium Pharmaceuticals for treatment of ischemia-reperfusion injury in stroke and myocardial infarction. The biosynthesis of lactacystin has been studied16 and more than 10 total syntheses of lactacystin and various analogs have been reported throughout the last decade.

PAGE 25

6 Chapter Two Previous Syntheses Newly discovered molecules usually generate interest throughout the synthetic community, particularly when they have novel structural scaffolds. Few synthetic targets, though, have demanded as much attention as lactacystin. Discovered by Omura in 1991 while screening thousands of soil samples for differentiation of the Neuro 2A cell line, lactacystin’s structure was elucidated using 1H and 13C NMR techniques. The absolute stereochemistry was determined using single-crystal X-ray analysis. The positive identification of lactacystin as a potential neurotrophic factor, its appeal as a synthetic target and its scant supply made the -lactam an extremely hot target for total synthesis. 2.1 Corey’s First Total Synthesis of Lactacystin One year after mura’s initial report, Corey published the first total synthesis of lactacystin.3 Starting with N -benzyl protected serine methyl ester, the amine and hydroxyl functionalities were simultaneously protected as the oxazolidine ( 10 ) resulting in a 9:1 ratio of diastereomers (Scheme 2-1). Aldol coupling of 10 with isobutyraldehyde utilizing the lithium bromide-lithium enolate complex yielded diastereomerically pure 11

PAGE 26

7 with a 51% yield after recrystallization. The si gnificance of this transformation lies in the control by which the eventual C-5 and C-9 stereocenters are formed. These seemingly simple transformations take advantage of the stereogenic environment indigenous to the enolate itself which controls the conformation of the enolate and facial approach of the aldehyde under non-chelated conditions resulting in formation of the quaternary C-5 center. With the C-5 and C-9 stereocenters established, the focus turned to the formation of the remaining stereocenters of the lactam ring. The oxazolidine was opened up under acidic conditions and the resulting primary hydroxyl was protected as silylether 12 Again, simultaneous protection of the amine and hydroxyl functionalities with paraformaldehyde resulted in an oxazolidine system ( 13 ) with diminished stereochemical bias. The methyl ester was reduced to the primary alcohol and oxidized back to the aldehyde ( 14 ) using LiBH4/MeOH and Swern conditions respectively. Intermediate 14 served as the key fragment and electrophile for the second aldol coupling. PirrungHeathcock17 anti-aldol conditions resulted in a mixture of diastereomers with 78% yield of a 1.5:1 favorable ratio. These results were obviously disappointing and an alternative NBn O CO2Me NBn O CO2Me OH O NHBn MeO2C OH TBS NBn O CO2Me OTBS NBn O CHO OTBS NBn O OTBS OH O O 1011i-PrCHO LDA, LiBr THF, -78 C i) TfOH, MeOH, ii) TBSCl, Imidazole 12 13 (CH2O)n, TsOH C6H6, i) LiBH4, MeOH/THF ii) Swern Ox. 14 O O LDA, THF, -78 C 15 Scheme 2-1. Corey's Synthesis of the First Key Lactacystin Intermediate5 9

PAGE 27

8 aldol coupling using Braun’s chiral controller/Cp2ZrCl enolate conditions,18 in light of good diastereoselection, were not practical for gram scale reactions. Moreover, desired product 15 and the product resulting from attack on the opposite face of the aldehyde were not easily separable by column chromatography. Catalytic hydrogenation of 15 resulted in a tandem debenzylation/cyclization sequence yielding the bicyclic -lactam ( 16 ) with C-6 and C-7 stereochemistry intact (Scheme 2-2). Deprotection of the primary hydroxyl yielded the diol intermediate. To avoid oxidation of both hydroxyls, selective oxidation of the primary hydroxyl was achie ved using a two-step process. Swern oxidation selectively oxidized the primary hydroxyl to the aldehyde followed by mild sodium hypochlorite oxidation to yield acid 17 Deprotection of the N/O acetal yielded dihydroxy acid 3 The first total synthesis of 1 was completed by coupling the N acetylcysteine allyl ester with the acid functionality of 3 using BOPCl and subsequent deallylation with triethylammonium formate and Pd(0). Corey’s total synthesis of lactacystin was a 15 step protocol with an overall yield of 6%, featuring two N O HO O TBSO N O HO O CO2H NH HO O CO2H HO NH O HO O S CO2H AcHN HO NH O HO O S CO2Allyl OH 15 H2, Pd/C EtOH AcHN i) HF, MeCN ii) Swern Ox. iii) NaClO2 HS(CH2)3SH, HCl, CF3CH2OH NHAc CO2Allyl HS BOPCl Pd(Ph3P)4HCO2H, Et3N 1 6 3 17 16 S c heme2-2. C ompletionoftheFi r stTotalSynthesisofLactacystin6 7

PAGE 28

9 diastereoselective aldol couplings, the second which suffered from mediocre selectivity. Another notable feature of the synthesis is th e evolution of the stereochemical elements of lactacystin from serine without the use of asymmetric methodologies. 2.2 Corey’s Extended Methodology Eventually, a follow-up to Corey’s lactacystin synthesis reported numerous improvements over the first, most notably of which was a superior methodology for the diastereoselective anti -aldol coupling as an alternative to the Pirrung-Heathcock conditions. A novel magnesium-catalyzed doubly diastereoselective anti -aldol coupling19 of aldehyde 14 and the tert -butyldimethylsilyl enol ether of methyl propionate under Mukiyama conditions resulted in a 90% yield of 15B with no evidence of the corresponding diastereomer (Figure 2-1). Though Mukiyama-type open transition state N O Ph H O OTBS H MgI2, CH2Cl2 N O OTBS H Ph Mg O H Me H OTMS OMe Me H OTMS OMe I N O OTBS H Ph Mg O H OMe I Me H TMSO N O Ph OTBS H N O Ph OTBS H CO2Me CO2Me HO Me HO Me synclinal antiperiplanar anti-aldol 15B syn aldol epi 15B 14 Figu r e 2-1. C helateModelfo r the D ouble D iaste r eoselectiveMukiyama A ldol C oupling

PAGE 29

10 aldol couplings typically occur antiperiplanar, the steric repulsions illustrated above make the synclinal transition state a much more energetically favorable arrangement. 2.3 Smith’s Total Synthesis of Lactacystin Six months later a collaboration between Smith and mura resulted in the second total synthesis of lactacystin ( 1 ).4 Touted as an easily accessible route to 1 and a variety of analogs, this 10-step protocol was e fficient and high yielding. Starting from the previously prepared unnatural amino acid 2 R ,3 S --hydroxyleucine methyl ester ( 18 ), oxazoline derivative 19 was prepared using methyl benzimidate (Scheme 2-3). Utilization of this starting material circumvented the installation of the crucial and synthetically challenging hydroxyisobutyl side chain. A stereoselective hydroxymethylation of the oxazoline using LHMDS/formaldehyde resulted in 20 with excellent yield and diastereoselectivity (>98% de ). This transformation installs the C-5 Scheme 2-3. The Smith-Omura Total Synthesis of Lactacystin NH2 CO2Me OH N O MeO2C Ph N O MeO2C Ph HO N O MeO2C Ph HO N O MeO2C Ph HO HO2C NH HO O CO2H HO 3 Ph(MeO)C=NH H2CO LHMDS i) Moffatt Ox. Ipc2B ii) i) O3, DMS ii) NaClO2 i) Pd, HCO2NH4ii) 0.1N NaOH 18 2221 20 195 7 6

PAGE 30

11 quaternary center with outstanding diastereocontrol. A two-step procedure including Moffatt oxidation of primary alcohol 20 to the corresponding aldehyde followed by Brown’s asymmetric allylation20 resulted in simultaneous formation of the C-6 hydroxyl and C-7 methyl substituents of 21 in good yield. Ozonolysis and selective oxidation of the resulting aldehyde yielded acid 22 Activation of the acid functionality using AcOH enabled spontaneous -lactam formation upon debenzylation of the oxazoline. Saponification of the ester moiety resulted in known intermediate 3. The synthesis was then completed utilizing Corey’s two-step protocol for esterification and deallylation. The highlights of the synthesis are the stereoselective hydroxymethylation and the asymmetric antiallylation steps resulting in a 13% overall yield. The authors also implied that the synthesis is designed in such a way as to allow for minor modifications of the protocol providing analogs for further study. Eventually, a full account of Smith and mura’s work including analog synthesis and biological assay data was published.12 The versatility for analog synthesis lay in the ability of the authors to obtain the unnatural amino acid in all four isomeric forms. Sharpless’ asymmetric epoxidation21 provided the initial stereochemical differentiation from which the remainder of the synthesis evolves. The bioassay studies focused on analogs of the N -acetyl cysteine side chain. Though a more active analog was not discovered, the authors state that a less cytotoxic yet more specific agent ( 8 ) was the ultimate target.

PAGE 31

12 2.4 Corey’s Expanded Studies Corey’s intense intertest in lactacystin was evident during 1993 when three simultaneous accounts of lactacystin analog syntheses were reported.22 Perhaps the most significant feature of these reports was the transformation of the dihydroxy acid ( 3 ) to the -lactone ( 2 ) using BOPCl as the activator for the lactonization. Initially, intermediate 3 was part of an alternate, higher yielding sequence for incorporation of the N -acetyl cysteine side chain. Lactonization of 3 occurred readily and nucleophilic substitution with the thiol containing side-chain was also an efficient process. These analogs, in addition to a few others, were the subject of Corey’s structure activity relationship and biological assay studies that identified clasto -lactacystin -lactone ( 2 ) as the most potent lactacystin analog known.6 2.5 Baldwin’s Total Synthesis of Lactacystin The first two total syntheses of lactacystin had similar strategies. Both syntheses utilize variations of Seebach’s oxazolidine/oxazoline alkylation protocol23 for formation of the C-5 quaternary center. The critical hydroxyisobutyl side chain was the very first structural and stereochemical issue addressed. Upon installation, the side chain was protected along with the amino functionality as a chiral oxazolidine/oxazoline from which the stereochemistry of the -lactam is derived. Baldwin’s synthesis of 1 employed a chiral bicyclic -lactam derived from ( R )-

PAGE 32

13 glutamate ( 23 ) for stereochemical induction (Scheme 2-4).24 A sequence of methylation, selenation and oxidation yielded the -unsaturated -lactam ( 24 ). Aromatization to siloxypyrrole 25 provided a key intermediate which was eventually utilized as the silyl enol ether in a Mukiyama type aldol coup ling. Using isobutyraldehyde as the electrophile and SnCl4 as the Lewis acid, the aldol coupling proceeded in only mediocre yield with a favorable 9:1 ratio of 26 and its C-9 epimer. Upon formation of the quaternary center, the secondary hydroxyl was protected through acetylation. Syn -dihydroxylation of the unsaturated lactam using osmylation proceeded stereoselectively via substrate control in good yield. Removal of the tertiary hydroxyl of the diol using Barton’s cyclic thiocarbonate radical methodology25 occurred in excellent yield, however, the resultant product was approximately a 1:1 ratio of C-6 epimers. Exposure to 0.5N NaOH resulted in a significant increase of the desired diastereomer ( 27 ), although substantial amounts of starting material (10%), epi -27 (10%) and elimination (5%) persisted. Cleavage of the hemiacetal using hydrogenation resulted in a mi xture of epimers at C-3 (formerly C-6). N O O Ph N O TBSO Ph N O O Ph N O O Ph Me Me N O O Ph OH OH ( R )-Glutamate i) LDA, MeI ii)LDA, PhSeBr iii) O3/pyr TBSOTf 2,6-lutidine i-PrCHO SnCl4 3 Me i) AcO2, pyr ii) OsO4, NMO iii) dehydroxylation iv) 0.5N NaOH Me i) Pd/C ii) TESCl/AcO2iii) HF/MeCN iv) H2CrO4 v) 0.2N NaOH 23 24 25 26 27 Scheme 2-4. Baldwin's Total Synthesis of Lactacystin9 6 OH

PAGE 33

14 The one-pot global protection was achieved by sequential addition of triethylsilylchloride and acetic anhydride/pyridine to mask the primary and secondary hydroxyls respectively. Deprotection of the silyl ether and oxidation using chromic acid yielded the carboxylic acid which was saponified with NaOH to yield 3 The synthesis of 1 was completed using Corey’s method of direct addition of the N -acetylcysteine sidechain. Baldwin’s synthesis utilized a unique route and was similar to Corey’s first synthesis in that he relied on the stereochemical nature of the starting amino acid for stereochemical induction, thus using no asymmetric methods. Drawbacks of the synthetic route were the low yields of the aldol coupling and the occurance of diastereomeric mixtures in no less than three steps of the synthesis. Overall, Baldwin’s synthesis of 1 was achieved in 20 steps with a 4.3% overall yield. 2.6 Chida’s Total Synthesis of Lactacystin Chida’s total synthesis of 126 was novel in the sense that the three previous reports started with amino acid derivatives. In this particular synthesis D-glucose serves as the starting material and the stereogenic template. The primary and secondary hydroxyls of the previously prepared 3-deoxy-1,2O -isopropylidene-3C -methyl--D-allofuranose ( 28 ) were benzylated and oxidized, respectively, to the protected ketone (Scheme 2-5). The Wittig reaction resulted in an inseparable mixture of ( E )and ( Z )-isomers ( 29 ) in a 1:1 ratio. Reduction of the ester and reaction with trichloroacetonitrile gave the Overman rearrangement27 substrate, trichloroacetimidate 30. The rearrangement was preformed in a sealed tube by heating 30 to 150 C in toluene for 89 hours to give terminal olefin 31

PAGE 34

15 The reaction gave a 60% yield of a mixture of C-5 epimers in a favorable 5:1 ratio. Hydrolysis of the acetonide and oxidative cleavag e of the diol affected formation of the hemiaminal. Subsequent chromic acid oxidation gave the -lactam which, upon treatment with NaBH4, was completely deprotected yielding 32 A sequence of reactions including protection of the secondary hydroxyl of 32 deprotection and oxidation of the benzyl ether to the aldehyde and addition of isopropylmagnesium bromide resulted in a complex mixture of C-9 epimers and reduced primary alcohol. Chromatographic separation of the mixture enabled recycling of the reduced product back into the sequence. The undesired C-9 epimer was subject to Moffatt oxidation and selectively reduced to yield desired alcohol 33 Desilylation, ozonolysis and selective oxidation using sodium chlorite provided the common intermediate 3 As was the case with the other syntheses, Chida also finished the synthesis of 1 using Corey’s protocol. The novelty of this synthesis, in D-glucose 28 29 31 30 NH O Me HO CH2OBn 32 O O O OH HOH2C Me O O O BnOH2C Me EtO2C O O O BnOH2C Me Cl3CCOH2C HN O O O BnOH2C Me H Cl3COCHN NH O Me HO HO 33 3 i) BnBr, CsF ii) H2CrO4iii) Ph3PCHCO2Et i) DIBAL-H ii) Cl3CCN, NaH 150 C, tol sealed tube i) aq. TFA ii) NaIO4iii) H2CrO4iv) NaBH4 i) TBSOTf, 2,6-lut ii) Na, NH3( l ) iii) Moffatt Ox. iv) i-PrMgBr S c heme2-5. C hida'sTotalS y nthesisofLactac y stin5 9

PAGE 35

16 addition to starting with D-glucose, lies in the formation of the C-5 quaternary center via Overman rearrangement. The three previous syntheses formed the C-5 quaternary center via stereoselective aldol coupling. The shortcom ings of the protocol are evident not only in the poor selectivity of the Wittig reaction and the Overman rearrangement, but especially in the necessity of a sealed tube for the latter transformation. The rearrangement, particularly, put an original twist on the protocol but renderd it impractical. 2.7 Corey’s Second Total Synthesis of Lactacystin Corey eventually published a new, shorter, more efficient synthesis of 1 employing a unique strategy for controlling diastereoselectivity utilizing a blocking group.14 This report featured an expedient construction of the -lactam moiety and installation of the isopropyl side-chain later in the synthesis to facilitate lipophilic group analog synthesis. The effectiveness of the blocking group itself relied on three requirements; (a) it had to be readily reducible, (b) it had to be sufficiently bulky to control the stereochemistry of the hydroxymethylation of the -keto ester and (c) it had to facilitate a scalable enantioselective process. The methylsulfide blocking group was installed by thioalkylation of dimethyl me thylmalonate giving the methylsulfanyl derivative 34 (Scheme 2-6). An enantioselective hydrolysis using porcine liver esterase gave the stereogenic monoester 35 in 62% yield and 95% ee after recrystallization. Exposure of the carboxylic acid to oxalyl chloride provided the stereogenic acid chloride which readily coupled with N -PMB-glycine methyl ester. Subsequent Dieckmann cyclization resulted in the cyclic -keto ester in a 1:1 ratio of diastereomers 36 At this

PAGE 36

17 point in the synthetic route the methylsulfide blocking group was intact, the -lactam skeleton was formed and the -carbon of the -keto ester was primed for aldol coupling. Enolization using DBU and addition of formaldehyde resulted in a 9:1 ratio of diastereomers which underwent stereoselective reduction of the keto moiety yielding 37 with a 95% yield and 99% ee The effectiveness of the methylsulfide blocking group was showcased in the two previous transformations for its ability to control selectivity through the stereogenic environment of the substrate alone without external asymmetric induction. Simultaneous protection and selec tive deprotection of the primary hydroxyl resulted in a silyl ether protected secondary hydroxyl and an oxidizable primary hydroxyl group. Diastereoselective desulfurization using Raney nickel resulted in a 10:1 ratio of epimers at the C-3 center. Dess-Martin oxidation of the primary alcohol to the aldehyde 38 set the system up for nucleophilic addition of the isopropyl side-chain or other nucleophiles leading to potential analogs. Grignard and organolithium additions of the isopropyl group proceeded slowly and resulted in the retro-aldol cleavage product. The MeS CO2Me Me CO2Me MeS CO2H Me CO2Me 34 35 N O O MeS Me PMB 36 N O HO MeS Me PMB 37 CO2Me OH N O TBSO PMB 38 CO2Me CHO Me N O TBSO PMB 39 CO2Me Me HO N O HO PMB 40 CO2H Me HO PLE, H2O ii) N -PMBGlyME i) (COCl)2, DMF iii) LDA, THF i) DBU, CH2O ii) NaBH(OAc)3 i) PivCl, pyr. ii) TBSOTf, 2,6-lut. iii) NaOMe, MeOH iv) Raney nickel v) Dess-Martin TMSCl ii) TFA aq. i) H2, Pd/C iii) LiOH, THF 2 Scheme 2-6. Corey's Second Enantioselective Total Synthesis of Lactacystin3 MgBr CO2Me

PAGE 37

18 use of 2-propenyl Grignard reagent in conjunction with TMSCl as an anion trap was efficient and the reaction occurred stereospecifically to give secondary alcohol 39 in high yield with no trace of retro-aldol cleavage. Corey proposed that the stereocontrol is a result of the steric blocking that exists when Mg2+ chelates the 1,3-dicarbonyl system. The completion of the synthesis involved hydrogenation of the isopropenyl group, desilylation of the secondary hydroxyl group and hydrolysis of the ester moiety yielding dihydroxy acid 40 As in previous work, BOPCl was used to affect lactonization and ceric ammonium nitrate oxidation of the N -PMB group yields -lactone 2 This protocol touts its applicability from an economic standpoint as well as its efficiency of synthetic steps and isolation of pure products. An additional point is the versatility it lends to analog synthesis for bioassay studies. 2.8 Kang’s First Formal Synthesis of Lactacystin Shortly after Corey’s second generation synthesis, Kang reported two novel routes28 to 1 both employing an intramolecular mercurioamidation of an allylic trichloroacetimidate.29 The first protocol started with the base promoted ring opening of Sharpless epoxide 4130 followed by functionalization of the primary hydroxyl as the trichloroacetimidate ( 42 ) using trichloroacetonitrile and DBU (Scheme 2-7). Treatment of 42 with mercuric trifluoroacetate and K2CO3 resulted in mercuration of the olefin and concomitant oxazoline formation via intramolecular mercurioamidation. Workup using TEMPO and LiBH4 gave the masked hydroxyl 43 Routine protection of the secondary hydroxyl, deprotection of the primary hydroxyl and sequential oxidation gave oxazoline 44 Reflux under acidic conditions followed by addition of zinc initiated oxazoline

PAGE 38

19 hydrolysis, concomitant -lactam cyclization and cleavage of the TEMPO group. Protection of the cisprimary and secondary hydroxyls as the acetonide occurred selectively (7:1) over the gemhydroxymethyl groups presumabley based on the cyclization of the fused 5,6-system ( 45 ) as opposed to the spiro system. Swern oxidation of the persisting primary hydroxymethyl group to the aldehyde and Grignard addition of i-PrMgBr resulted in a disappointing 1:1 ratio of C-9 epimers. Serendipituously, it was realized that with addition of excess Grignard reagent to the ester analog, the reduced desired product 46 was obtained exclusively. Intermediate 46 was the previously known trihydroxy intermediate of Baldwin’s synthesis, thus completing the first formal synthesis utilizing Kang’s intramolecular mercurioamidation methodology. 2.9 Kang’s Second Formal Synthesis of Lactacystin The second simultaneous report involved a completely original route to 1 The key step in the synthesis was, again, the intramolecular mercurioamidation of an allylic trichloroacetimidate. This time the functionality of the latent lactam is installed HO OTBDPS O O OTBDPS OH Cl3C NH OTBDPS OH O N Cl3C TEMPO CO2H OMOM O N Cl3C TEMPO NH O O O OH NH O HO HO OH i) LDA, THF ii) Cl3CCN, DBU i) Hg(O2CCF3)2ii) TEMPO, LiBH4 i) MOMCl, DIPE A ii) TBAF, THF iii) Swern Ox. iv) KMnO4 i) HCl, AcOH ii) Zn, EtOH iii) TsOH, Me2CO i) H2CrO4ii) CH2N2, THF iii) xs i-PrMgBr iv) TsOH, MeOH 1 46 45 44 43 42 41 S c heme 2-7. Kan g 'sFi r stS y nthesisofaKe y Inte r mediateofLactac y stin9

PAGE 39

20 throughout the first part of the synthesis and the key step was the endgame to 1 The Sharpless epoxide 47 was subjected to base promoted ring opening and the resulting allylic and secondary alcohols were selectively oxidized and benzoylated respectively (Scheme 2-8). Stereoselective crotylboration of -unsaturated aldehyde 48 provided alcohol 49 in a 50:1 diastereomeric ratio. The resultant secondary alcohol was protected as the MOM ether and the terminal olefin was oxidized to the aldehyde. Exposure of the aldehyde to hydroxylamine hydrochloride resulted in the oxime and upon treatment with methanesulfonic anhydride and DBU the latent nitrile ( 50 ) was unmasked. Hydrolysis of the benzoyl protecting group and functionalization with trichloroacetonitrile resulted in trichloroacetimidate 51 The key step of this synthesis is initiated by addition of mercuric acetate to the olefin of 51 Spontaneous cyclization to the oxazoline gives a mixture of separable diastereomers, the desired mercurate in 64%. As in the previous synthesis problematic oxidative demercuration issues were circumvented by exposing the oxazoline to LiBH4/TEMPO providing 52 Exposure to 6 N HCl followed by AcOH OH O O OBz OBz OH OBz CN OMOM O CN OMOM Cl3C NH N O Cl3C OMOM CN TEMPO 47 46 52 51 50 49 48 i) LDA, THF iii) (PhCO)2O, TEA ii) MnO2 asymmetric crotylboration i) CH2(OMe)2, P2O5ii) OsO4, NaIO4iii) HONH2HCl, pyr. iv) (MeSO2)2O, DBU i) K2CO3, MeOH ii) Cl3CCN, DBU i) Hg(O2CCF3)2 S c heme 2-8. Kang'sSecondSynthesisofaKeyInte r mediateofLactacystin ii) TEMPO, LiBH4

PAGE 40

21 hydrolyzes the oxazoline, initiates cyclization and removes the TEMPO group resulting in the known triol 46 Overall, both of Kang’s syntheses of 1 rely on Sharpless asymmetric epoxidation early in the synthetic scheme and utilize the intramolecular mercurioamidation of a trichloroacetimidate to set up a zinc promoted cyclization to the -lactam. 2.10 Adams’ Total Synthesis of Lactacystin Analog PS-519 Lastly, Adams’ team at LeukoSite conducted an independent synthesis and structure activity study of 2 and its analogs. Confirming and building upon Corey’s data, Adams found that increasing the bulk of the C-7 alkyl substituent the activity increases 2fold. Ultimately, the group settled on the npropyl C-7 substituent/-lactone combination to give them the best activity and labled the experimental compound PS-519 The novelty of this synthesis lies in the use of a semi-convergent route. All other syntheses to date have been strictly linear routes based on the compact nature of 1 and 2 The key step of the synthesis is the convergence of a stereogenic aldehyde and a trans -oxazoline via a S c heme2-9. Synthesisofthe trans -Oxazoline C ouplingPa r tne r CO2Me CO2Me OH OH CO2Me O B r Ph O CO2Me O NH2 Ph O NH O CO2Me Ph AD-mixPhC(OCH3)3 BF3OEt2then CH3COBr CO2Me O N3 Ph O NaN3 H2Pd(OH)2/C 53 58 57 56 55 545TsOH

PAGE 41

22 doubly diastereoselective aldol coupling. The construction of the oxazoline fragment begins with Sharpless’ asymmetric dihydroxylation31 of the Wittig product of isobutyraldehyde and methyl-triphenylphoranylideneacetate ( 53 ). The quatitative conversion to diol 54 occurred with only 77% ee but was improved to >99% ee after recrystalization. Upon addition of trimethyl orthobenzoate the diol was tied up as the cyclic orthoester and subsequently opened with the addition of acetylbromide resulting in the bromohydrin 55 Azide 56 is the product of bromide displacement and hydrogenation yields amine 57 Refluxing in toluene with toluenesulfonic acid provided stereogenic oxazoline 58 that would ultimately provide the hydroxyisobutyl and C-5 portions of PS519 The coupling partner of 58 is a chiral -alkyl--formyl amide that will provide the chelation control for the doubly diastereoselective aldol coupling. The coupling fragment is derived from acyloxazolidinone 59 (Scheme 2-10). Alkylation of the titanium enolate using benzyloxymethyl chloride gave oxazolidinone 60 which, after peroxide hydrolysis, yields the carboxylic acid 61 Transformation of the acid to the diethyl amide 62 was achieved using triethylamine and TBTU to add diethylamine. Hydrogenolysis of the benzyl ether yielded the hydroxy amide ( 63 ) which, upon exposure to Dess-Martin Scheme 2-10. Synthesis of the -Alkyl--formyl Amide Coupling Partner O N O O Bn O N O O Bn OBn O OH OBn O NEt2 OBn O NEt2 OH O NEt2 O H TiCl4, DIPEA BOMCl LiOOH Et2NH TEA TBTU H2Pd(OH)2/C Dess-Martin 59 64 63 62 61 60

PAGE 42

23 oxidation conditions resulted in the -alkyl--formyl amide ( 64 ) coupling partner. After exploring a multitude of conditions for the aldol coupling of 58 and 64 it was determined that enolization of the oxazoline and sequential addition of dimethylaluminum chloride followed by the aldehyde gave the 6 S alcohol ( 65 ) exclusively. The stereochemical rationale was explained using the transition state model shown in Figure 2-2. Chelation of the 1,3-dicarbonyl system of 64 by the Lewis acid requires the n -propyl substituent to adopt a siface blocking position. This leaves only approach from the reface possible in anti -Felkin-Ahn-Eisenstein fashion.32 While the stereochemistry of C-6 was a direct result of the chelation model, the C-5 stereochemistry is purely based on steric bias of the isopropyl substituent of the oxazoline. Apart from obtaining alcohol 65 exclusively, the significance of this transformation lies in the use of a dialkylaluminum chloride as a Figure 2-2. Stereochemical Rationalization of Adams' Doubly Diastereoselective Aldol Coupling O N MeO OM1 Ph H N R1 O O R2 M2 + O N Ph N O OH MeO2C R1 R2 H O n -Pr H O M2 NR1R2 re 64 656 5S c heme2-11. End g ameofAdams' T otalS y nthesisofPS-519 H NEt2 O O O N Ph O OH MeO2C 66 LHMDS, 10 Me2AlCl NEt2 i) H2, Pd(OH)2/C ii) NaOH, H2O NH HO n -Pr O CO2H HO NEt3ClCO2C(CH3)=CH2 9 67 NH O n -Pr OH O O 5 6(PS-519)

PAGE 43

24 bidentate Lewis acid as opposed to its typical utility as a monodentate coordinator. With alcohol 65 in hand, three routine transformations remained to complete the synthesis of PS-519 (Scheme 2-11). As seen in previous syntheses, hydrogenolysis of the oxazoline initiates spontaneous cyclization to the -lactam 66 which was recrystalized and structurally confirmed using X-ray diffraction analysis. Saponification of the methyl ester using NaOH yielded dihydroxy acid 67 which was then activated by conversion to the mixed anhydride using isopropenyl chloroformate. Subsequent cyclization resulted in lactone 9 an analog of the highly potent -lactone 2 Adam’s synthesis of 9 was also applied to the total synthesis of 2 using a parallel method with only slight variations in starting material preparation. This particular synthesis is highlighted by a semiconvergent protocol requiring 10 operations resulting in a 20% overall yield of 9 starting from readily available oxazoline 58. Overall, lactacystin ( 1 ) and its various analogs have garnered much attention based not only on the pharmacological significance of the target but on the synthetic challenges they present. The syntheses above are the premier works preformed on this family of molecules and do not include the many unsuccessful attempts or works in progress that have not yet been reported.

PAGE 44

25 Chapter Three Stereogenic -Lactams via Rhodium Catalyzed Intramolecular C-H Insertion Of the many natural products that possess -lactam cores,33 lactacystin ( 1 ) and its intermediates are quite possibly the most outstanding in terms of biological activity, structural originality and synthetic appeal. The -lactone ( 2 ) is particularly intriguing based on its bicyclic [3.2.0] arrangement, four contiguous stereogenic centers (one a quaternary center) and high concentration of heteroatoms. Considering the methodologies we developed, 2 appeared to be an ideal synthetic target for its validation. In all previous syntheses of 1 and its analogs the strategy is to asymmetrically functionalize a linear molecule followed by formation of the -lactam by nucleophilic addition of an amine moiety to an activated ester. Our proposed synthesis involved -lactam formation via rhodium catalyzed intramolecular C-H insertion followed by stereoselective functionalization of the -lactam core. 3.1 C-H Insertion of the Acyclic System Our first tier of this methodology originated from our discovery of rhodium catalyzed intramolecular C-H insertion of -diazo--(phenylsulfonyl)acetamides to give -lactams

PAGE 45

26 regioand stereoselectively.34 Previous work showed that this general transformation, when applied to -amino acid derivatives, typically resulted in mixtures of and lactam regioisomers.35 Padwa and Wee reported that the ratios of the regioisomers could be affected by the electronic nature of the carbenoid -substituent.35d,e Our model system utilizes an -phenylsulfonyl substituent for alteration of the electron density of the metallocarbenoid and also exerts a steric effect for enhancement of regioand stereoselectivity. The showcase example of this transformation from the -diazo-(phenylsulfonyl)acetamide to the trans--lactam is shown in Scheme 3-1. Reflux conditions using rhodium acetate dimer provide the -lactams exclusively in excellent yield with no -lactam or aromatic cycloaddition side products. Rationalization of the regioand stereochemical selectivitiey is shown in Scheme 3-2. The first control element of this transformation is the conformational effect. Rhodium carbenoid 70 can adopt the s-cis and the s-trans conformations. The s-cis conformer is favored due to the severe nonbonded interaction between the tbutyl substituent and the rhodium ligands that exist in the s-trans conformer. As a result of the confor mational effect, two possible transition states exist through which insertion can occur. Reaction via 5-membered transition state 71 and 6-membered transition state 72 would yield -lactam 73 and -lactam 74 respectively. Padwa and Doyle performed similar transformations using the -acetyl Scheme 3-1. Rhodium Catalyzed Intramolecular C-H Insertion of -Diazo--(phenylsulfonyl)acetamides N t-Bu N PhSO2 Ph N2 O PhSO2 Ph O t-Bu Rh2(OAc)4CH2Cl2, 12h 95% 6869

PAGE 46

27 substituent with various rhodium catalysts.35e All attempts resulted in mixtures of and -lactam with no ratio of regioisomers exceeding 3:7 respectively. In the same report it was proposed that when a catalyst with an electron donating ligand is used the rhodium carbenoid is stabilized thus proceeding through a late transition state.36 As shown by the results, when the -phenylsulfonyl substituent is incorporated into the insertion precursor we achieve complete selectivity for the lactam. Presumably, the -phenylsulfonyl substituent lends further stabilization to the carbenoid allowing the cyclization to occur via 6-membered transition state exploiting th e stereoelectronic effect. Also explained by the transition state is the transgeometry at C-3 and C-4. For insertion to occur the carbon-rhodium “bond” and the target C-H bond must be arranged parallel to each other. Two conformations exist since insertion is occurring at a methylene center. The more favorable conformation in this case is that in which the C4/C5 bond is oriented so that the phenyl substituent is in the pseudoequitorial position resulting in trans-lactam 74 .37 The outstanding selectivity of this transformation makes it an attractive method of Scheme 3-2. Conformational and Stereochemical Effects on C-H Insertiont-Bu N PhSO2 R Rh O N PhSO2 t-Bu Rh O R H N H N Rh Rh O O t-Bu H R t-Bu H R PhSO2PhSO2N PhSO2 R O t-Bu N O t-Bu R PhSO2 s-cis 70 s-trans 70 71 74 73 72 ConformationalStereoelectronic3 4 5

PAGE 47

28 constructing -lactams. 3.2 C-H Insertion of the Cyclic System In an attempt to expand the general applicability of this methodology we considered utilizing amino acids as highly versatile, stereogenic substrates. To our surprise an extensive literature search revealed that previous attempts using amino acids as C-H insertion substrates were unsatisfactory due to poor regioand stereoselectivities and side reactions.38 With an original approach, our methodology was modified to accommodate amino acid derived precursors for the synthesis of highly functionalized chiral lactams.39 Synthesis of the prototype phenylalanine derived -diazo-(phenylsulfonyl)acetamide 77 began with esterification of the acid function of (L)phenylalanine followed by amide coupling using (phenylthio)acetic acid resulting in amide 75 (Scheme 3-3). Reduction of the methyl ester using LiBH4 generated in situ gave the primary alcohol. With protection of the alcohol and amide moieties as our focus, we sought to simultaneously tie-up both functional groups as acetonide 76 In toluene, the Scheme 3-3. Synthesis Stereogenic -Lactams From Amino Acid Precursors NH2 CO2H Ph (L)-Phenylalanine NH CO2Me Ph O PhS N Ph O PhS O N Ph O PhSO2 O N2 N O O PhSO2 Ph H i) HCl(g), MeOH ii) PhSCH2CO2H DIC, Imid. i) LiBH4, MeOH/THF ii) 2,2-DMP,TsOH, PhCH3, Dean-Starke i) m -CPBA ii) ABSA, DBU Rh2(OAc)4CH2Cl2, 91% 75 78 77 76

PAGE 48

29 alcohol was refluxed with 2,2-DMP and catal ytic TsOH. A Dean-Starke apparatus was instrumental in water removal. This particular transformation was quite disappointing based on the inconsistency of product yield and side reactions that occurred. After a tedious purification using silica gel column the phenylsulfide was completely oxidized to the phenylsulfone using multiple equivalents of m -CPBA. Finally, the -position was activated for carbenoid formation by installation of a diazo group using Davies’ p -ABSA reagent and DBU yielding 77 Intramolecular C-H insertion occurs efficiently under reflux in methylene chloride with a catalytic amount of rhodium acetate dimer. This transformation provides the trans-lactam 78 as a single diastereomer, in excellent yield without formation of -lactam or aromatic cycloaddition products. The regioand stereoselectivity of this cyclization is explained using a modified transition state theory congruent to our aforementioned acyclic system and control elements previously discussed (Scheme 3-4). Metallocarbenoid 79 adopts the favorable s-cis conformation as dictated by the conformational effect. Based on the severe nonbonded interaction Scheme 3-4. ConformationalandStereochemicalEffectsonC-HInsertionof A mino A cid Derivatives N PhSO2 Rh O N PhSO2 Rh O H N Rh O Ph H PhSO2N PhSO2 Ph O s-cis 79 s-trans 79 80 78 813 4 5O Ph O Ph O H N Rh O H Ph PhSO2 O O H

PAGE 49

30 between the rhodium ligands and the gem -dimethyl moiety of the acetonide, s-trans 79 is the unfavorable conformation. Insertion then occurs via the 6-membered transition state ( 81 ) as directed by the stereoelectronic effect and proven by the absence of -lactam formation. The transstereochemistry of the -lactam is established by the preference of the C-4 substituent to orient itself in the pseudoequatorial position, thus relieving the 1,3diaxial interaction present in transition state 80 The profound significance of the conformational effect is exemplified by attempting to perform C-H insertion on an analog of 77 that has a formaldehyde derived acetal ( 82 ) in place of the acetonide (Scheme 3-5). The s-trans 83 conformation of the amide is more favored in the presence of the methylene group. In this case, the typical insertion center is oriented away from the rhodium carbenoid, therefore, no insertion occurs. Intramolecular C-H insertion of the (L)-phenylalanine derivative resulted in -lactam 78 as a single enantiomer whose stereochemistry was governed by the existing stereocenter of the original amino acid. The stereochemistry of the -lactam ( 78 ) was confirmed by X-ray crystallographic analysis. Ultimately, two new stereocenters are formed in a single synthetic operation providing highly versatile, functionalized, stereogenic -lactams selectively. Scheme 3-5. Conformational Requirements for Intramolecular C-H Insertion N PhSO2 N2 O O Ph N PhSO2 Rh O O Ph Rh(II) No Cyclization Products 82 s-trans 83

PAGE 50

31 3.3 C-H Insertion of Amino Acid Derived -Diazo--(phenylsulfonyl)acetamides With a highly effective protocol for converting (L)-phenylalanine into a highly functionalized stereogenic -lactam in hand, we decided to apply it to a series of -amino acids. It is well established that the trend for preference of C-H insertion to occur at a particular center is methine>methylene>methyl. The electron deficient carbenoid carbon complexes with the most electron rich C-H bond via the 6-membered transition state and insertion immediately ensues. The substrates were specifically chosen to examine our stereoelectronic and substituent effect hypotheses by containing multiple insertion sites and varying degrees of electron donating and withdrawing substituents, respectively. Table 3-1 shows the results of C-H insertion on -amino acid derived -diazo sbstrates with various alkyl substituents and multiple potential insertion centers. In most cases the -lactam was formed exclusively and in high yield. Entries 1 and 2 are perfect examples of the influence that the stereoelectronic effect has on the course of the reaction. With Table 3-1. C -HInse r tionof D iazoSubst r atesattheMeth y lenePosition N R O PhSO2 O N2 N O O PhSO2 R H Rh2(OAc)4CH2Cl2, reflux entryRreactantyield (%)product 1 2 3 4 5 6 Me Eti-Pr Bn CH2CO2Me 4-(MeO)Ph 87 92 87c87 64 86 Ph 791 84a85 86a87 88b89 90a91 9293 94d95 7778a Starting with the corresponding racemic -amino acid. b Rh2(pfb)4 was used. c -lactam was also obtained (6%). d Starting with 4-methoxytyrosine.

PAGE 51

32 two electronically similar methylene centers present on the ethyl ( 84 ) and propyl ( 86 ) alkyl chains as the possible insertion sites, the stereoelectronic effect dictates the formation of the -lactam in favor of the -and -lactams. Entry 3 shows the preference for insertion to occur at a methine center. Though insertion at the methylene center of the isobutyl ( 88 ) side chain provides the -lactam in high yield, six percent of -lactam is formed as well. In comparison, the only difference that exists between the analogs is the presence of a methine center in 88 as opposed to an additional methylene center in 86 In the case of 86 no -lactam is formed and it can be concluded that the stereoelectronic effect dictates -lactam formation exclusively. However, the preference for insertion at a methine center is significant enough to provide the -lactam, albeit as the minor product. The goal of entries 4 and 6 were to compare the results of insertion on analogs of our showcase example (entry 7). Presumably, the facile conversion of 77 to 78 was enhanced by the phenyl substituent directly adjacent to the methylene center. Incorporation of an additional methylene center to the insertion precursor ( 90 ) had no effect on the transformation and only -lactam was formed. In an attempt to enhance the transformation of 77 to 78 a tyrosine analog with an electron donating 4-methoxy substituent was derived ( 94 ). No appreciable improvement was noticed in the reaction in light of the modest increase in electron density. -Diazo amide 92 is an (L)-glutamic acid derivative that possesses an electron withdrawing carbomethoxy group. This reaction proceeded with only a moderate yield as a result of the decreased reactivity of the insertion center. This particular reaction was intriguing since there was no -lactam, lactam or ylide formation detected and no starting material recovered. The -lactam was

PAGE 52

33 the only product recovered and the other side products were indecipherable. A series of analogs with potential methyl and methine insertion centers were also subject to our methodology (Table 3-2). -Diazo amide 96 derived from (L)-alanine underwent intramolecular C-H insertion to yield -lactam 97 exclusively, though only in moderate yield. Insertion into the methyl C-H is an unfavorable process especially in the presence of a methine center from which the -lactam would form. This example corroborates the strong influence that the stereoelectronic effect exerts on this transformation. In entries 2 and 3 both insertion precursors possess methine centers from which the -lactam, via C-H insertion, will potentially form. C-H insertion of -diazo amides 98 an (L)-valine derivative, and 100 an (L)-isoleucine derivative, both proceeded with excellent yield and result in exclusive -lactam formation ( 99 and 101 respectively) as anticipated. Disastereomeric mixtures were formed in both cases. This is a result of the substituents at the methine centers having no significant steric bias (two methyls in the case of (L)valine and a methyl and an ethyl in the case of (L)-isoleucine) to achieve a particular conformation at the insertion center. A point of note is that upon desulfonation of 101 using Na(Hg) a single diastereomer ( 102 ) was recovered which suggests that the insertion Table 3-2. C-H Insertion of -Diazo Substrates at Methyl and Methine Postitions N O PhSO2 O N2 N O O PhSO2 H Rh2(OAc)4CH2Cl2, entryR2reactantyield (%)product 1 2 3 Me Me 67 95 93 9697 9899 100101 R2 R1 R1 R2 R1Me Et HH

PAGE 53

34 proceeds with complete retention of configura tion of the inserton center. In light of the success of methylene insertion we attempted insertion of electronically activated methylene centers. A set of -amino acid derivatives that were suitable for this transformation were those of (L)-serine ( 103 ) and two different (L)-threonine analogs ( 105 and 107 ) (Table 3-3). We anticipated an enhancement in efficiency of insertion into the methylene C-H adjacent to a heteroatom due to electronic inductive effects. As expected, the efficiency of insertions were improved as yields were comparable with data from methine C-H insertions. Entries 1 and 2, which possessed essentially identical insertion sites, both proceeded with excellent yield and resulted in exclusive trans-lactam formation. The alternate (L)-threonine derivative 107 also underwent C-H insertion in high yield but resulted in a diastereomeric mixture of lactams at the C-3 Table 3-3. C-H Insertion of -Diazo Substrates at Electron Rich Methylene Positions N O PhSO2 O N2 N O O PhSO2 H Rh2(OAc)4CH2Cl2, entryR2reactantyield (%)product 1 2 3 Me 97 94 92 103104 105106 107108 R1 OTBS TBSO R1 R1Me HH R2 R2 H H104 108 N O O PhSO2 TBSO H N O O PhSO2 H TBSO NH O HO O S CO2H AcHN HO 1 NH O OH O O Cl 109 Salinosporamide A Lactacystin Scheme 3-6. Functionalized Stereogenic-Lactams are Versatile Synthetic Intermediates

PAGE 54

35 position as experienced previously with the (L)-valine and (L)-isoleucine derivatives. Presumably, the steric bias of the methyl group rivals that of the tertbutyldimethylsilyl ether since the bulk of the substitution on the silicon is one oxygen atom removed from the reaction center. Nevertheless, insertion at the chiral center of the side-chain of (L)threonine results in retention of configuration. Moreover, the success of insertion at the methylene position adjacent to a heteroatom adds a degree of functionality to the lactam. An intriguing point and an impetus for further pursuit and application of this methodology is the striking resemblance that -lactams 104 and 108 have with synthetic targets 1 and 109 and it is, therefore, anticipated that they will be valuable synthetic intermediates (Scheme 3-6). In general this methodology is useful for the synthesis of highly functionalized -lactams from stereogenic -amino acids. In one synthetic operation, two additional chiral centers are formed in a stereoselective fashion and result in -lactams that are functionalized at each carbon center. 3.4 C-H Insertion of -Diazo--(substituted)acetamides The success of C-H insertion of -amino acid derivatives encouraged us to explore alternate pathways for diversified functionality and a streamlined synthetic sequence. Our initial route for construction of the -diazo--(phenylsulfonyl)acetamides derived from -amino acids was relatively costly, suffered from yield inconsistency and required multiple laborious purification steps. Partic ularly, the acetonide formation step, which involved tying-up the amide nitrogen and the primary alcohol, was impetuously

PAGE 55

36 inconsistent. Side products, incomplete cyclizations and a rather formidable flash column purification rendered this process impractical. An improved sequence that was one synthetic operation shorter, utilized commodity reagents and required no purification, until after the diazotransfer step, was formulated. The new protocol called for the reduction of the (L)-phenylalanine using lithium aluminum hydride to provide the corresponding amino alcohol (Scheme 3-7). The Fischer protocol was utilized to complex and precipitate the unreacted LAH as a means of filtering the product and avoiding the use of aqueous work-up procedures. A s uperior procedure for acetonide formation involved refluxing the amino alcohol in a 1: 1 mixture of 1,2-DCE and acetone with the addition of Na2SO4 as a water scavenger. This reaction was complete in less than one hour and simple filtration of the solids followed by evaporation of the solvent provided acetonide 110 cleanly in near quantitative yield. This improved protocol also provided options for functionalization of this system. The secondary amine, in this case, could be Scheme 3-7. Improved Synthetic Protoc ol for Analog Synthesis NH2 CO2H Ph (L)-Phenylalanine HN O Ph 110 i) LAH, THF, ii) Me2CO, 1,2-DCE Na2SO4 i) BrCH2COBr, TEA ii) PhSO2Na, DMF i) diketene, THF iii) ABSA, DBU ii) ABSA, DBU i) ClCOCH2CO2Et, THF ii) ABSA, DBU N O Ph N O Ph N O Ph N O Ph PhSO2 EtO O O O O O O N2 H N2 N2 N2 LiOH MeCN/H2O 113 112 111 77

PAGE 56

37 easily acylated using a variety of reagents. Toward obtaining our -diazo-(phenylsulfonyl)acetamide intermediate, a three-step procedure consisting of Nacylation using bromoacetyl bromide followed by bromide displacement with benzenesulfinic acid sodium salt proceeded cleanly and in very good yield providing the known phenylsulfonyl intermediate ( 77 ). With our goal of highlighting the profound effect that the -phenylsulfonyl substituent plays in -lactam formation via C-H insertion, a series of -diazoacetamide analogs with -substituents of varying electron-withdrawing capacities were prepared. The three variations used to contrast to the -phenylsulfonyl substituent were the -ethoxycarbonyl, -acetyl and, simply, -diazo substituents (Table 3-4). Our improved protocol for synthesis of the -substituted acetamides was applied and construction of the three analogs was possible from acetonide 110 Nacylation using ethyl-4-chloroacetoacetate and diketene afforded the -ethoxycarbonyl 111 and -acetyl 112 acetamides respectively in high yield with no need for purification. Diazotransfer of both 111 and 112 although occurring at a decreased rate as compared to the phenylsulfonyl acetamides, proceeds in very high yield. Analog 113 which possessed a proton as the -substituent, is a product of base promoted decarbonylation of 112 using Table 34 C -HInse r tionof D iazoSubst r ateswithVa r ied -Substituents N Ph O R O N2 N O O R Ph H Rh2(OAc)4CH2Cl2, entryR reactantyield (%) 1 2 3 4 MeCO EtO2C PhSO284 93 95 91 113 78 112 114 111 115 77 116 N O O H R + H-lactam-lactam

PAGE 57

38 LiOH in aqueous media. In anticipation of consistent results with previous reports which showed poor stereoand regioselectivities, we were surprised to find that trans--lactam formation was the exclusive product of C-H insertion with exception to the -diazo substrate 113 Cyclization of 111 and 112 occurred efficiently and in very high yield providing chiral -lactams 114 and 115 with functionality that is more versatile than the phenylsulfone substituent. -Diazo substrate 113 when subjected to insertion conditions, resulted in aromatic cycloaddition yielding cycloheptatriene 116 The latter result was not at all surprising and is, in fact, quite common when aromatic systems are in the presence of carbenes. The peculiar aspect of this data is the complete selectivity of the transformations. We expected a certain degree of -lactam formation and obtained exclusively -lactam. In the case of the -Diazo substrate 113 we expected a mixture of products including -lactam, but recovered the aromatic cycloaddition product exclusively. The obvious explanation for this data is the absence of an electronwithdrawing -substituent. Padwa and Doyle have shown, using competition experiments, that the rhodium catalyst ligands play a major role in reaction preference.35e Therefore, in an attempt to switch the preference in reactivity from aromatic cycloaddition to -lactam formation, we used the rhodium caprolactamate catalyst which has a more electron donating ligand. Under these conditions the reaction yielded mostly Scheme 3-8. Pathway for C-H Insertion of -Diazo Substrates Resulting in Aromatic Cycloaddition N Ph O H O N2 Rh2(cap)4CH2Cl2, reflux N O O H H N O O H H 113117116 silica gel

PAGE 58

39 aromatic cycloaddition product 116 along with a minor product that was eventually elucidated as the norcaradiene tautomer ( 117 ) of the major product (Scheme 3-8). Two explanations for the persistence of aromatic cycloaddition in the absence of an substituent have appeared in the literature. Wee suggests that the electronic differences of the -substituents dictate the reaction pathway and, in the case of the -diazo substrate, preference for aromatic cycloaddition dominates.35d Padwa and Doyle reason that conformational influences of the -substituents (other than -diazo, of course) inhibit the approach of the phenyl group to the reactive carbene center.35e This theory is further corroborated by the fact that highly electron rich phenyl rings, as in the case of dimethoxy phenyl groups, do not undergo aromatic cycloaddition despite their increased vulnerability to carbene attack.40 Overall, our methodology is an improvement over previous methodologies based on the regioand stereoselectivities obtained from C-H insertion. It was originally assumed Scheme 3-9. ConformationalConstraintoftheCyclicSystemInhibits -LactamFormation N R Rh O N R Rh O H N Rh O Ph H R N R Ph O s-cis 118 s-trans 118 120 114 and 115 121 O Ph O Ph O H N Rh O H Ph R O O H R = MeCO, EtO2CN O H Ph R Rh O 119N O O R Ph 122

PAGE 59

40 that the -phenylsulfonyl group was responsible for the selectivity of the methodology, and it is most definately the case with our initial, acyclic example. Clearly, it is not the only factor in play since comparable success was observed in the presence of the ethoxycarbonyl and -acetyl substituents. Our latest rationalization is that the acetonide moiety present in all of the amino acid derived systems must play a significant role in lending a degree of conformational constraint to the transition state of the transformation. The aromatic cycloaddition product is not observed when electron withdrawing groups are present and the unfavored 5-membered transition state ( 119 ) required for -lactam formation would experience severe non-bonded interactions. With this particular system used for C-H insertion of -amino acid derivatives -lactams are the only likely product (Scheme 3-9). At this point the phenyl group had participated in the reaction and had a substantial Scheme 3-10. Synthetic Protocol for Norvaline Analog Synthesis NH2 CO2H (DL)-Norvaline HN O 123 i) LAH, THF, ii) Me2CO, 1,2-DCE Na2SO4 i) BrCH2COBr, TEA ii) PhSO2Na, DMF i) diketene, THF iii) ABSA, DBU ii) ABSA, DBU i) ClCOCH2CO2Me, THF ii) ABSA, DBU N O N O N O N O PhSO2 MeO O O O O O O N2 H N2 N2 N2 LiOH MeCN/H2O 126 125 124 86

PAGE 60

41 influence on the results. To dispel any skepticism about the success of this project we applied the same methodology to an amino acid system that included an npropyl side chain. In this case multiple equivalent insertion sites were available without the influence of the phenyl group. We initiated this leg of the project with the racemic, unnatural amino acid norvaline. The improved protocol for synt hesis of the C-H insertion precursors was applied to the construction of the norvaline analogs (Scheme 3-10). Reduction of the amino acid with LAH followed by acetonide formation yielded versatile intermediate 123 Next, the diversification using the various Nacylation conditions was carried out followed by diazo transfer resulting in insertion precursors 86, 124 and 125 Deacylation of 125 using aqueous LiOH provided -diazo substrate 126 Exposure of the diazosubstrates to the standard C-H insertion conditions yielded -lactams in all cases (Table 3-5). In the case of the -phenylsulfonyl ( 86 ), -methoxycarbonyl ( 124 ), and acetyl ( 125 ) compounds, insertion occurred with complete regioand stereoselectivity. Yields were all very good and no minor products were detected. On the other hand, diazo substrate 126 formed a mixture of products with very good yield but, in only a 2:1 ratio which were separated and identified as the -and -lactams respectively. Table 3-5. C-H Insertion of -Diazo Norvaline Derivatives N O R O N2 N O O R H Rh2(OAc)4CH2Cl2, entryR reactantyield (%) 1 2 3 4 MeCO MeO2C PhSO291 90 93 84a126 87 125 127 124 128 86 130 N O O H + H-lactam-lactam 129 R a Product ratio was 2:1 favoring the -lactam

PAGE 61

42 3.5 C-H Insertion of a Conformationally Constrained Cyclic System An additional series of analogs were prepared that had a varied cyclic system incorporated into the insertion precursors (scheme 3-11). The N,O -acetonide that was utilized throughout all of the previous work was replaced with an N,N’acetonide in which both nitrogens were of the secondary amide type. The goal was to probe the effects of a more rigid cyclic system on the course of insertion. The replacement of an sp3 hybridized center with an sp2 hybridized center would provide the increase in rigidity we sought. Starting from (L)-phenylalanine, es terification followed by amide formation using methylamine resulted in the -amino amide, which, upon exposure to refluxing acetone/1,2-DCE in the presence of Na2SO4, was transformed to the N,N’acetonide 131. Acylation followed by diazo transfer using conditions previously delineated achieved construction of the -phenylsulfonyl-,-ethoxycarbonyland -acetyl-diazo substrates 132 133 and 134 respectively. Our typical C-H insertion conditions yielded -lactams in Scheme 3-11. Synthesis of N,N' -Acetonide Substrates with Varied -Substituents NH2 CO2H Ph (L)-Phenylalanine HN N Ph 131 i) SOCl2, MeOH iii) Me2CO, 1,2-DCE Na2SO4 i) BrCH2COBr, TEA ii) PhSO2Na, DMF i) diketene, THF iii) ABSA, DBU ii) ABSA, DBU i) ClCOCH2CO2Me, THF ii) ABSA, DBU N N Ph N N Ph N N Ph PhSO2 MeO O O O O O N2 N2 N2 134 133 132 ii) MeNH2, MeOH O O O O

PAGE 62

43 all cases in excellent yield, regioand stereoselectively (Table3-6). In general, a 2-6% increase in yield of C-H insertion of the N,N’acetonide analogs over the N,Oacetonide system was obtained. This was attributed to the increase in rigidity of the N,N’acetonide system over the N,Oacetonide and the marginal boost in conformational constraint that it lends to the transition state of the transformation. After an extensive review of the literature, there seems to be no clear-cut rationalization concerning the effects of -substituents on C-H insertion. Conversely, extensive work has been done describing the diversity of this chemistry when the catalyst ligands are varied. The presently accepted transition state theory for C-H insertion implies that the selectivity of the reaction depends on the electrophilicity of the carbenoid itself. Highly electrophilic, more reactive carbenes proceed via an early transition state resulting in decreased selectivity. Therefore, greater selectivity can be achieved by using a less reactive carbenoid system, namely the carbene intermediate from rhodium acetate or any other catalyst ligand that is not electron withdrawing in nature. One would then assume that electron withdrawing -substituents would increase the reactivity of the carbene and make the C-H insertion process less selective, but this is just not the case. As Table 3-6. C -H Inse r tion of C onfomationaly C ont r ained A cetonideSubst r ateswithVa r ied -Substituents N Ph O R N N2 N N O R Ph H Rh2(OAc)4CH2Cl2, entryR reactantyield (%) 1 2 3MeCO MeO2C PhSO293 97 99 135 134 136 133 137 132lactam O O

PAGE 63

44 seen in our initial acyclic example of C-H insertion of -diazo-(phenylsulfonyl)acetamides, the excellent regioselectivity is a result of the electron withdrawing -phenylsulfonyl substituent. It is clear that the selectivity of the amino acid methodology is not based solely on the mitigating effects of the -phenylsulfonyl substituent since excellent selectivity is achieved with other -substituents as well. The conformational constraints exerted by the acetonide moiety must, therefore, influence the regioselectivity and most likely enhance the overall success of the transformation. The premise that the acetonide moiety plays a significant role in this transformation is corroborated by an alternate cyclic system that was used. The increased rigidity of the N,N’acetonide system does result in an increase in the efficiency of the tansformation. Overall, carbenoid insertions are complex transformations that have various possible reaction pathways and outcomes that can be tuned by variation of catalyst ligands and substituents. What we have elaborated is the formation of -lacams in a regioand stereoselective manner from -amino acids utilizing a variety of -substituents. The utility of this methodology is directly applied to natural product synthesis of -lactam containing targets.

PAGE 64

45 Chapter Four Jung’s Total Synthesis of clastoLactacystin Lactone Early on in the development of our rhodium catalyzed intramolecular C-H insertion methodology, we discovered just how promising and effective this transformation was. The applicability of our initial, acylic methodology was profoundly broadened by its adaptation to accommodate -amino acids as starting materials.39 Using only the -amino acid as the source of stereogenic bias, -lactams were synthesized regioand stereoselectively yielding a single enantiomer. Two additional stereocenters were formed simultaneously as dictated by the original stereocenter of the starting material and the three control elements delineated in chapter three. Immediately, we turned our full attention to the total synthesis of clastoLactacystin -Lactone ( 2 ). 4.1 Retrosynthetic Analysis of clastoLactacystin Lactone We had taken notice that application of this methodology to (L)-serine yielded a highly functionalized -lactam that possessed many of the characteristics of 2 Retrosynthetically, we envisioned an expedient route to the target molecule featuring two key steps (Scheme 4-1). -Lactone 2 is readily available via lactonization of the cis -

PAGE 65

46 arranged, masked hydroxyl and acid functionalities of intermediate 138 The alcohol derivative of 138 is the product of the second key step of the synthesis, namely a quaternary carbon-forming late stage aldol-coupling of the iminoether 139 and isobutyraldehyde. The iminoether is derived from -lactam 140 The bicyclic -lactam is a product of our rhodium catalyzed intramolecular C-H insertion of -amino acid derived -diazo--(phenylsulfonyl)acetamides methodology which is the first key step in the synthetic route. The -diazo insertion precursor 143 is efficiently derived from (L)serine. 4.2 First Generation Synthesis of the Bicyclic -Lactam Intermediate Our initial endeavor into the synthesis of 2 began with the natural -amino acid (L)-serine (Scheme 4-2). Esterification of the amino acid was conducted by bubbling HCl gas into a methanolic suspension of (L)-serine. (Phenylthio)acetic acid was then dissolved in methylene chloride and activated using the carbodiimide reagent DCC. Scheme 4-1. Retrosynthetic Analysis of the Total Synthesis of clastoLactacystin Lactone NH O OH O O N PO O CO2Me HO N PO O CO2Me N O O PO H N O OP PhSO2 O N2 NH2 CO2H OH (L)-serine Aldolcoupling C-H Insertion 2 141140 139 138 PhSO2

PAGE 66

47 Addition of the HCl salt of serine methyl ester and triethylamine resulted in amidecoupling of amide 142 The primary alcohol and the amide nitrogen were then simultaneously protected as the acetonide by refluxing 142 in toluene with 2,2-DMP and pTsOH yielding ester 143 (Scheme 4-3). For this transformation to occur effectively, removal of water using a Dean-Starke trap was required to yield the desired products. This particular reaction provided inconsistent results due to extensive side-product formation and incomplete cyclizations. The impracticality of this transformation was compounded by the necessity of a tedious purification via flash column chromatography. Nevertheless, the ester function of the recovered phenylsulfide ( 143 ) was reduced to the primary alcohol 144 using LiBH4 generated in situ from excess equivalents of NaBH4 and LiCl in a 1:1 solution of THF and methanol. Protection of the hydroxyl as the silyl ether 145 using TBSCl and imidazole in DMF occurred in high yield. Complete oxidation of phenylsulfide 145 using excess mCPBA yielded phenylsulfone 146 in nearly quantitative yield (Scheme 4-4). Diazo transfer using pABSA and DBU provided Scheme 4-2. Amide Coupling of (L)-Serine Methyl Ester NH2 CO2H OH (L)-serine i) HCl(g), MeOH, ii) PhSCH2CO2H, DCC, TEA, CH2Cl2 NH OH MeO2C PhS O 142Scheme 4-3. Installation of the N,OCyclic System 142 2,2-DMP pTsOH, tol Dean-Starke N O PhS MeO2C O 144 143 N O PhS O OTBS NaBH4/LiCl N O PhS O 145 OH THF/MeOH TBSCl imid, DMF

PAGE 67

48 the -diazo--(phenylsulfonyl)acetamide insertion precursor 103 Our standard conditions for rhodium catalyzed intramolecular C-H insertion were carried out and the transbicyclic -lactam 104 was the sole product of the transformation. As one of the key steps of the synthetic route to 2 the -lactam was produced in 97% yield as a single enantiomer. This highly effective methodology provides -lactams with two additional stereocenters in one synthetic operation in a regioand stereoselective manner. 4.3 Functionalization of the -Lactam Core The goal of the next leg of the synthesis was to functionalize the -lactam at the C-3 and C-5 positions. Meyers has shown that alkylations of bicyclic systems typically occur on the -face (convex face in this case) of the molecule. In our bicyclic system the C-3 position is highly activated for alkylation. Methylation of this center using iodomethane and sodium hydride in DMF occured at 0 C. Elimination of the protected Scheme 4-4. Intramolecular C-H Insertion of the (L)-Serine Derived Diazo-(phenylsulfonyl)acetamide 145 N O PhSO2 O OTBS N2 N O O H TBSO PhSO2 mCPBA pABSA 103 104 N O PhSO2 O OTBS 146 CH2Cl2 DBU, MeCN Rh2(OAc)4CH2Cl2, Scheme 4-5. Functionalization of the Lactam 104 MeI, NaH DMF, -20 C N O O H TBSO PhSO2 TBAF, THF BnBr, NaH 148 147 N O O H BnO PhSO2 N O O H OH PhSO2 1493

PAGE 68

49 secondary C-4 hydroxyl substituent was a competing pathway that was ultimately inhibited by performing the reaction at -20 C (Scheme 4-5). The methylation of the bicyclic system occurred stereoselectively in nearly quantitative yield ( 147 ). This particular reaction benefits from the concave/convex nature of the bicyclic -lactam. We then turned our attention to hydrolysis of the acetonide followed by oxidation of the resulting primary alcohol in preparation fo r the aldol coupling of the hydroxyisobutyl side-chain. To circumvent reactivity issues with the silyl ether protecting group throughout the remainder of the synthesis, the protected secondary hydroxyl 147 was unmasked using TBAF and reprotected as the benzyl ether 149 Subsequent hydrolysis via reflux in methanol and catalytic pTsOH yielded the primary alcohol 150 which was then oxidized using Jones reagent to acid 151 (Scheme 4-6). Methylation of the carboxylic acid using iodomethane and DBU provided the ester functionality at the C-5 position of 152 which would prove integral in the aldol coupling of the hydroxyisobutyl side-chain. At this point in the synthesis a multitude of conditions to achieve aldol Scheme 4-6. Functionalization of the C-5 center 149 NH O PhSO2 BnO pTsOH 150 MeOH, NH O PhSO2 BnO CO2H 151 OH H2CrO4Me2CO NH O PhSO2 BnO CO2Me 152 MeI DBU, MeCN 5S c heme4-7. UnsuccessfulAldolC ou p lin g NP O PhSO2 RO CO2Me P = BOC, CBz, Bn NP O PhSO2 RO CO2Me OH "Aldol Conditions" isobutyraldehyde 152 R = Bn, Me, PMB

PAGE 69

50 coupling of the ester with isobutyraldehyde were tried. Various benzyl, carbonate and carbamate Nprotecting groups along with numer ous C-4 hydroxyl protecting groups were employed to direct the reactivity toward enolization of the C-5 position (Scheme 47). All attempts at aldol coupling of the ester and isobutyraldehyde and/or formaldehyde using amide bases were unsuccessful resulting in elimination products or no reaction altogether. A complete reevaluation of our proposed synthesis followed the unsuccessful attempts to achieve the second key-step aldol coupling. The synthetic route to the bicyclic -lactam had a few shortcomings, namely inconsistencies of the acetonide forming step and the amide coupling reaction (not to mention cost of the acid reagent). It had also been a consequence of the proposed synthetic route that the natural amino acid would ultimately yield the enantiomer of the na tural product, thus requiring the use of the much less economically feasible (D)-serine as the starting material to obtain the correct enantiomer. This synthesis would eventually benefit from the discovery of the synthetically superior second generation route to -diazo--(phenylsulfonyl) acetamides applied to our earlier methodologies in chapter 3. 4.4 Second Generation Synthesis of the Bicyclic -Lactam Intermediate Additional concerns that we had were the lack of significant stereochemical bias to affect the aldol-coupling selectively. Originally, the bulky silyl ether was used anticipating a need for stereochemical induction. Eventually, this group proved to be incompatible with subsequent steps of the synthesis. It had been exchanged for other alkyl ethers such as benzyl and methyl although the effectiveness of both toward

PAGE 70

51 induction of stereogenic differentiation at th e aldol step were ineffective based on the distal orientation of the phenyl ring and th e size of the methyl group, respectively. We settled on using a tertbutyl ether based on its steric presence, resiliency to most conditions and its facile introduction and cleavage. The HCl salt of (L)-serine methyl ester was subject to the ether forming conditions using liquefied isobutylene and pTsOH in methylene chloride (Scheme 4-8). Initially,this transformation required three days to complete. However, our endeavors into the realm of process chemistry provided us with optimized conditions for this etherification providing the tertbutyl ether 153 Dehydration of the ptoluenesulfonic acid monohydrate shortened the reaction time to 24 hours and eliminated the need for a closed, pressurized system. Reduction of the ester using LAH at 0 C provided amino alcohol 154 cleanly and in high yield. The acetonide formation ( 155 ) was achieved using the conditions from the previous methodologies and occurred in nearly quantitative yield. NAcylation using bromoacetyl bromide proceeded readily but suffered from mediocre yield a nd required flash column chromatography for S c heme 4 -8. S y nthesiso f the A cetonideDe r i v ati v eo f ( L ) -Se r ine NH2 CO2H OH (L)-Serine i) HCl(g), MeOH, ii) isobutylene, pTsOH, CH2Cl2 153 NH2 CO2Me Ot-Bu 154 NH2 Ot-Bu OH LAH THF, 0 C Me2CO/1,2-DCE Na2SO4, HN O Ot-Bu 155Scheme 4-9. Synthesis of the -Diazo--(phenylsulfonyl)acetamide N O Ot-Bu X O X = N O Ot-Bu PhSO2 O N O Ot-Bu PhSO2 O N2 159 158 Br Cl 156 157X X O TEA, CH2Cl2 NaSO2Ph DMF pABSA DBU, MeCN 155

PAGE 71

52 purification of the bromide ( 156 ) (Scheme 4-9). Once again, we were dissatisfied with the means by which the bromide was recovered and the yield of the product. We replaced the acid bromide with its chloride analog and a much cleaner reaction ensued with improved yield. Recovery of chloride 157 consisted of simple recrystallization. Displacement of the chloride with benzenesulfinic acid sodium salt occurred readily and sulfone 158 was purified by recrystallization. Diazo transfer to the -position of the amide occurred readily using pABSA and DBU in acetonitrile providing the -diazo-(phenylsulfonyl)acetamide C-H insertion precursor 159 which was also purified by recrystallization. Application of our standard C-H insertion conditions yielded bicyclic lactam 160 in high yield regioand stereoselectively (Scheme 4-10). At this point in the synthetic scheme significant improvements over the first generation synthesis were evident by the requirement of less synthetic operations to the lactam intermediate, product availability from the natural amino acid, incorporation of the tertbutyl ether which protects the hydroxyl function and will lend stereochemical bias to further transformations. Nowhere were the benefits of our experience in process chemistry more apparent than in the techniques used for purification throughout the first leg of the synthesis. The first generation synthetic route called for flash column purification of six intermediates. Our second generation synthesis of the bicyclic -lactam Scheme 4-10. Lactam Formton via Rhodium Catalyzed Intramolecular C-H Insertion N O O t-BuO H PhSO2 159 160 Rh2(OAc)4CH2Cl2,

PAGE 72

53 calls for only one flash column purification directly following the insertion step. This enabled us to scale-up lactam production without concern for costly and time consuming flash column purifications. 4.5 Stereoselective Functionalization of the Bicyclic -Lactam With the bicyclic -lactam in hand, functionalization of the C-3 and C-5 centers became the focus of the next leg of the synthesis. Stereoselective methylation of the C-3 center occurred readily using NaH and iodomethane in DMF resulting in alkylated bicyclic -lactam 161 in nearly quavtative yield (Scheme 4-11). We were confident that methylation at the C-3 center occurred from the -face (the convex face) of the -lactam based on x-ray crystal data obtained from the product of an analogous transformation involving the phenylalanine derivative 78 Though, without irrefutable evidence we were forced to consider the possibility of and -alkylated products. Following methylation, reduction of the phenylsulfone group using 10% Na(Hg) in methanol occurred readily resulting in an equal mixture of C-3 epimers 162 and epi162 This result was not at all surprising as we somewhat anticipated a degree of diastereoselection based on the single electron transfer mechanism through which this reaction proceeded. We did expect a more favorable ratio resulting from the stereochemical and steric bias of the -lactam Scheme 4-11. Functionalization of the C-3 Center N O O t-BuO H 160 PhSO2 161 MeI, NaH DMF, 0 C N O O t-BuO H 162 & epi162 1:1 10% Na(Hg) MeOH, 0 C 3

PAGE 73

54 itself. An alternate route involving reduction of the phenylsulfone group prior to methylation was explored (Scheme 4-12). The -methylene product ( 163 ) of Na(Hg) reduction was methylated using LHMDS and iodomethane resulting in the C-3 methylated -lactam with complete diastereoselectivity. Moreover, the original methylation-reduction conditions were optimized to yield the desired epimer ( 162 ) in a diastereoselective manner as well.41 Despite the ambiguity of the C-3 epimers, both were carried through the synthesis separately. A one-pot procedure for the hydrolysis of the N,Oacetonide and subsequent oxidation of the resultant primary alcohol was devised (Scheme 4-13). Essentially, simple Jones oxidation conditions are sufficiently acidic to effect hydrolysis of the acetonide moiety and oxidation of the acid ( 164 and epi164 ) promptly follows. Methylation of the acid usi ng trimethyl orthoformate in methanol with catalytic sulfuric acid resulted in methyl esters 165 and epi165 Again, numerous attempts at forming the quaternary center via aldol coupling of Nprotected methyl esters were unsuccessful resulting in elimination products and recovered starting materials. In Scheme 4-12. Selective -Methylation of the C-3 Center N O O t-BuO H 160 N O O t-BuO H epi162 10% Na(Hg) MeOH, 0 C 163 MeI LHMDS, -78 C S c heme 4 -13. Functionalizationo f the C -5 C ente r 162 & epi162 H2CrO4 NH O t-BuO CO2H NH O t-BuO CO2Me TMOF MeOH, H2SO4 165 & epi165 164 & epi164 Me2CO

PAGE 74

55 an attempt to activate the C-5 center the amide function was converted to the iminoether ( 166 and epi166 ) using Meerwein’s salt in methylene chloride. Addition of iminoether 166 and epi166 to LDA in THF at -78 C resulted in enolization of the ester. Isobutyraldehyde was added after 10 minutes followed by saturated ammonium chloride solution which resulted in formation of the quaternary C-5 center of 167 Upon purification of the products it was revealed that the aldol-coupling of the iminoether product derived from the reduction-methyla tion protocol yielded a single isomer ( epi167 ) as a crystalline solid. The methylation-reduction iminoether intermediate ( 166 ) provided a 2:1 mixture of diastereomers as an oil (Scheme 4-15). X-ray crystal analysis of epi167 revealed that the quaternary center had indeed formed selectively with the correct stereochemistry at the C-5 quaternary center and the C-9 center as well. Unfortunately, the C-3 center which remained ambiguous until this point was shown to Scheme 4-14. Stereoselective Formation of the Quaternary C-5 Center epi165 N O t-BuO CO2Me epi166 N t-BuO O CO2Me HO epi167 Me3OBF4CH2Cl2 isobutyraldehyde LDA, -78 C 5 5 9Scheme 4-15. Formation of the Quaternary C-5 Center of the 3 R Intermediate 165 N O t-BuO CO2Me 166 N t-BuO O CO2Me HO 167 5 R 9 S / 167 5 S Me3OBF4CH2Cl2 isobutyraldehyde LDA, -78 C 2:1 ratio of diastereomers

PAGE 75

56 be alkylated from the -face establishing absolute 3 S stereochemistry. Epimerization of the C-3 methyl center was unsuccessful and this leg of the project was, therefore, discontinued. Although, not the results we were hoping for, this data proved invaluable for elucidation of the absolute stereochemistry of the C-3 centers of the bicyclic methylated compounds 162 and epi162 and displayed a new and efficient option for stereoselective C-3 alkylation of bicyclic -lactams. The diastereomeric mixture ( 167 5 R 9 S & 167 5 S ) resulting from aldol coupling of the 3 R epimer ( 166 ) was now the focus and elucidation of the absolute stereochemistry was our goal. 4.6 Elucidation of Aldol Stereoselectivity Throughout the course of the aldol coupling two new stereocenters are formed. In this case the C-5 and the C-9 stereocenters were those in question. Based on the selectivity of the aldol coupling of epi166 we felt confident that the tertbutyl ether “blocking group” was performing its function. A simple test to distinguish which center had formed selectively was to oxidize the secondary diastereomeric alcohols ( 167 5 R 9 S & 167 5 S ) to ketones (Scheme 4-16). Obviously, if C-9 was the nonselectively-formed center, the product of oxidation would be a single isomer. Interestingly, oxidation of the mixture using TPAP/NMO conditions resulted in oxidation of one of the diastereomers and left the other unreacted as the secondary alcohol, a piece of data which would become useful later in the synthesis. Exposure of the mixture to Dess-Martin oxidation conditions yielded two different ketone products ( 168 R & S ), thus confirming

PAGE 76

57 diastereomers at the C-5 center. This result proved that our inclusion of a bulky ether substituent failed to induce selectivity for the aldol coupling reaction. Various conditions were used to improve the selectivity of the second key-step of the synthesis with no success. The best selectivity that could be achieved was a 2:1 ratio of 167 R & S diastereomers respectively. An alternate route for formation of the quaternary carbon center which holds promise is acylation of iminoether 166 under the same conditions using isobutyrylchloride (Scheme 4-17). Ketone 168 S is recovered as a 10:1 diastereomeric mixture that matches spectral data from that previously generated in the oxidation of the 167 5 R 9 S & 167 5 S mixture. Reduction of the ketone ( 168S ) at this and the two ensuing Scheme 4-16. Oxidation of the C-9 Stereocenters for Structural Elucidation N t-BuO O CO2Me O 2:1 ratio of diastereomers TPAP, NMO H2O, MeCN, CCl4 Dess-Martin N t-BuO O CO2Me HO 167 5 S 168 S N t-BuO O CO2Me O 168 R & S 167 5 R 9 S /167 5 S9 9Scheme 4-17. Acylation of the Iminoether Occurs with Enhanced Selectivity 166 N t-BuO O CO2Me O 168 S isobutyrylchloride LDA, -78 C 10:1 diastereomeric ratio reduction conditions N t-BuO O CO2Me HO 167 5 R 9 R

PAGE 77

58 deprotected intermediates resulted in perfect selectivity for 9 R secondary alcohol 167 5 R 9 R This alternate route may eventually prove successful under the selectivity inducing reduction conditions or inversion of the selectively formed C-9 center itself. 4.7 Rationalization for Aldol Coupling Stereoselectivity With our first successful aldol transformation from epi166 in hand, we were excited by the high degree of selectivity we experienced. Unfortunately, the same degree of selectivity was not obtained for the reaction of 166 Initially, we were confounded by the complete lack of stereoselectivity for the latter aldol transformation. The tertbutyl ether, which was intended to act as a blocking group, was effective in the case of epi166 but rather ineffective as in the case of 166 Assuming that the aldol coupling would proceed through a 6-membered transition state, we expected the highly populated face of enolate of 166 present with a methyl at C-3 and tbutyl ether at C-4 would only reinforce the bias for facial-selectivity during the tran sformation (Figure 4-1). Therefore, approach of the aldehyde would be more likely to occur from the -face of the enolate as in 166. Our conclusion and explanation for the results we obtained are based on the conformation of the enolate substituents in their minimized energy states (Figure 4-2). O-Li+O N iPr H OMe MeO OtBu O Li+ON OMe MeO OtBu H iPr 166 166 Figure 4-1. Transition States for Quaternary C-5 Formation

PAGE 78

59 MM2 calculations using Chem 3D Pro 7.0 show that when the C-3 methyl is on the face of the enolate, the tert-butyl ether can adopt a position which is more facially blocking as determined by the dihedral angle of 72 (t-BuO-C-C=C). When the C-3 methyl is on the -face of the enolate, a steric repulsion exists between the two substituents disallowing the blockage of the -face by the tert-butyl ether evidenced by the dihedral angle of 51. Obviously, st eric consequences of the overpopulated -face of the enolate had a profoundly detrimental effect on this transformation. 4.8 Endgame for the Total Synthesis of clastoLactacystin -Lactone The 2:1 mixture of diastereomers ( 167 5 R 9 S & 167 5 S ) was carried forward through the deprotection stage of the synthesi s using anhydrous TFA for cleavage of the tertbutyl protecting group yielding a mixture of diols ( 169 5 R 9 S /169 5 S ). Exposure of the diols to 1% HCl in methanol resulted in the known amide 170 5 R 9 S as a mixture of N MeO t-BuO LiO OMe N MeO t-BuO LiO OMe Dihedral Angle( t-BuO-C-C=C): 72Dihedral Angle( t-BuO-C-C=C): 51Figu r e 4 -2. R ationalizationo f Aldol C ouplingSte r eoselecti v ity

PAGE 79

60 diastereomers with 170 5 S The major component of the resulting mixture was identical to the 1H NMR spectral data reported by Adam’s and Smith for the amide 169 5 R 9 S This data was useful by confirming that aldol product 167 5 R 9 S was formed with a favorable yet mediocre ratio. Combined with the results of oxidation of the aldol diastereomers, we were able to distinguish the stereochemical configuration of the product of the acylation methodology and declare it a viable alternative to the aldol coupling. Completion of the synthesis of the -lactone 2 was accomplished using Adams’ method of basic hydrolysis of the methyl ester to the dihydroxy acid followed by activation to the mixed anhydride which spontaneously lactonized resulting in 2 as a single isomer. The corresponding diastereomer was most likely hydrolyzed but was not fit for lactonization based on the transarrangement of the hydroxyl and carboxylic acid substituents and subsequently lost during work-up of 2 The synthetic -lactone 2 was identical spectroscopically to reported data and an x-ray crystal analysis confirmed the absolute stereochemistry of the final product. The total synthesis of 2 was accomplished in 17 steps with a 10% overall yield. Made obvious by the elegant and elaborate syntheses highlighted above, lactacystin ( 1 ) and its analogs have been intensely pursued by some of the most Scheme 4-18. Endgame for the Total Synthesis of clastoLactacystin Lactone NH HO O CO2Me HO 167 5 R 9 S /167 5 S 2:1 ratio of diastereomers i)TFA ii)1% HCl NH HO O CO2H HO O Cl O TEA, THF 169 5 R 9 S /169 5 S 170 5 R 9 S /170 5 S 2 LiOH MeOH

PAGE 80

61 prestigious and successful synthetic groups in recent years. Moreover, it is a perfect example of the significant role natural product synthesis plays in the scientific community. Seldom do we experience the discovery of a natural target that possesses such novel biological activity and specificity. Because of scant supply, the only practical means of obtaining material for further research is through total synthesis.

PAGE 81

62 Chapter Five Experimental Data All experiments were carried out under nitrogen atmosphere using oven dried glassware (or flame dried when necessary). All chemicals were purchased from Aldrich Chemical Co. and/or Acros Organics and used without further purification unless otherwise noted. Methylene chloride was distilled over calcium hydride. THF and diethyl ether were distilled over sodium metal. Proton nuclear magnetic resonance (250 MHz) and 13C (63 MHz) spectra were recorded at room temperature in CDCl3 unless otherwise noted. All chemical shifts are reported as relative to CHCl3 ( H 7.26 ppm) and CDCl3 ( C 77.0 ppm) as internal standards, respectively, using a Bruker DPX 250 spectrometer. Infrared spectra were recorded using a Nicolet Magna FTIR 550 spectrometer and are reported in reciprocal centimeters (cm-1). Elemental analysis was preformed by Atlantic Microlab, Inc., Norcross, GA. Thin layer chromatography (TLC) was preformed on EMD precoated silica plates with silica gel 60 , 250 m thickness. Visualization of TLC was accomplished using a UV lamp (254 nm), iodine or charring solutions (ninhydrin and PMA). Flash column chromatography was performed on Whatman Purasil 60 (230-400 mesh) silica gel.

PAGE 82

63 Preparation of Phenylthioacetamide 75 from (L)-Phenylalanine NH2 CO2H Ph (L)-Phenylalanine i) HCl(g), MeOH, ii) PhSCH2CO2H, DIC, Imid., DMF NH CO2Me PhS O Ph 75 Thionyl chloride (3.65 mL, 50 mmol) was added drop wise to a solution of (L)phenylalanine (7.6 g, 46 mmol) in methanol (60 mL) at 0 C, and the mixture was heated under reflux 1 h. The solvent was evaporated to give (L)-phenylalanine methyl ester, hydrochloride as white solid. To a solution of (L)-phenylalanine methyl ester hydrochloride (46 mmol), phenylthioacetic acid (8.5 g, 51 mmol) and imidazole (4.1 g, 60 mmol) in DMF ( 50 mL, C = 1.0 M), was slowly added N,N’ -diisopropylcarbodiimide (8 mL, 51 mmol) at 0 C. The mixture was stirred briefly at 0 C and then at r.t. for 12 h. After filtration of the precipitate, the filtrate was diluted with EtOAc and the organic layer was washed twice with water and dried over Na2SO4. After evaporation of solvent, the resulting residue was recrystallized with hexanes-EtOAc to afford desired product 75 (15 g, 97%) as white solid: 1H NMR (250 MHz, CDCl3) 7.31-7.18 (m, 10 H), 6.95 (m, 1 H), 4.85 (m, 1 H), 3.68 (s, 3 H), 3.61 (s, 2 H), 3.06 (d, 2 H, J = 5.9 Hz), 13C NMR (62.5 MHz, CDCl3) 171.4, 167.7, 135.5, 134.5, 129.2, 129.1, 128.6, 128.3, 127.1, 126.7, 53.3, 52.3, 37.7, 37.4; IR (thin film, cm-1) 1744, 1676, 1512, 1265.

PAGE 83

64 Preparation of Alcohol 75A NH CO2Me PhS O Ph 75 NH PhS O Ph NaBH4/LiCl OH THF/MeOH 75A To a solution of phenylthioacetylmide methyl ester 75 (10 g, 30 mmol) in THFMeOH (150 ml, C = 0.2 M, 1:1 vol. ratio), were slowly added NaBH4 (3.4 g, 90 mmol) followed by LiCl (3.7 g, 90 mmol) at 0 C. The resulting mixture was stirred for 1 h at r.t., and the solvent was evaporated. The residue was partitioned with EtOAc and brine, and the organic layer was dried over Na2SO4 and concentrated. The resulting residue was recrystallized with hexanes-EtOAc to afford primary alcohol 75A (8.6 g, 95%) as white solid: 1H NMR (250 MHz, CDCl3) 7.31-7.10 (m, 10 H), 7.03 (br d, 1 H), 4.14 (m, 1 H), 3.66 and 3.56 (ABq, 2 H, J = 17.1 Hz), 3.53 (m, 2 H), 2.79 (m, 2 H), 2.14 (t, 1 H, J = 5.7 Hz); 13C NMR (62.5 MHz, CDCl3) 168.3, 137.1, 135, 129.3, 129.1, 128.1, 126.7, 64.1, 53.1, 37.5, 36.8; IR (thin film, cm-1) 3290, 1661, 1540, 1265.

PAGE 84

65 Preparation of Acetonide 76 NH PhS O Ph OH 2,2-DMP, pTsOH PhCH3, reflux Dean-Starke N O PhS O Ph 76 77A Primary alcohol 77A (6 g, 20 mmol) is added to a mixture of 2,2dimethoxypropane (4.8 mL, 40 mmol), and catalytic p -toluenesulfonic acid in toluene (70 ml, C = 0.3 M). The solution is heated under reflux for 1 hr using a Dean-Starke apparatus for the removal of water. The reaction mixture was poured into saturated NaHCO3 and extracted with EtOAc. The combined organic layer was washed with brine, then dried over Na2SO4, filtered and concentrated to give crude product as an oil, which was chromatographed to afford acetonide 76 (4.8 g, 70 %): 1H NMR (250 MHz, CDCl3) 7.52-7.11 (m, 10 H), 4.04 (m, 1 H), 3.79 (m, 2 H), 3.48 and 3.40 (ABq, 2 H, J = 13.7 Hz), 2.89 (m, 2 H), 2.05 (s, 3 H), 1.47 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 165.3, 137.2, 134.7, 131.1, 129.2, 129.0, 127.3, 127.1, 95.8, 66.9, 59.4, 40.7, 38.9, 26.9, 22.7; IR (thin film, cm-1) 1652, 1496, 1265.

PAGE 85

66 Preparation of Phenylsulfone 77 N O PhS O Ph 76 mCPBA CH2Cl2, 0 C N O PhSO2 O Ph 76A To a solution of phenylthioacetamide 76 (3.7 g, 10.8 mmol) in dry CH2Cl2 (54 mL, C = 0.2 M) m -CPBA (6.7 g, 27 mmol) was slowly added and stirred for 1 hr at 0 C. The reaction mixture was poured into 1 N NaOH, and extracted with CH2Cl2. The organic layer was washed with brine, dried over Na2SO4, filtered and concentrated yielding pure phenylsulofone 76A : 1H NMR (250 MHz, CDCl3) 7.84-7.50 (5 m, 5 H), 7.36-7.17 (m, 5 H), 4.39 (m, 1 H), 4.01 (dd, 1H, J = 5.0, 8.9 Hz), 3.87 (d, 1H, J = 8.9 Hz), 3.81 and 3.41 (ABq, 2 H, J = 13.8 Hz), 2.93 (m, 2 H), 1.71 (s, 3H), 1.51 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 158.3, 138.7, 137.0, 134.2, 129.4, 129.2, 129.0, 128.4, 127.3, 96.1, 67.5, 61.7, 60.0, 40.8, 26.6, 22.5; IR (thin film, cm-1) 1738, 1373, 1245, 1046. General Preparation of -Diazo--(phenylsulfonyl)acetamides. To a mixture of a phenylsulfonyl)acetamide (5.0 mmol) and p acetamidobenzenesulfonyl azide (1.3 g, 5.5 mmol) in dry CH3CN (25 mL, C = 0.2 M), was slowly added DBU (1.64 mL, 11 mmol) at 0 C. The resulting mixture was stirred for 1 hr at 0 C, and the solvent was evaporated. The residue was diluted with Et2O, and the mixture was washed successively with 1 N NaOH, water, and brine. The yellow organic layer was dried over Na2SO4, filtered, and concentrated. The residue was

PAGE 86

67 chromatographed to give diazo --( phenylsulfonyl)acetamide. N O PhSO2 O Ph 77 N2 77: 1H NMR (250 MHz, CDCl3) 7.96-7.51 (m, 5 H), 7.35-7.17 (m, 5 H), 4.39 (m, 1 H), 3.85 (d, 2 H, J = 3.6 Hz), 3.08 (dd, 1 H, J = 5.3, 13.5 Hz), 2.81 (dd, 1 H, J = 9.3, 13.5 Hz), 1.69 (s, 3 H), 1.34 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 153.9, 141.9, 136.5, 133.9, 129.4, 129.2, 128.8, 127.5, 127.1, 96.7, 74.8, 67.1, 58.7, 40.0, 26.9, 23.4; IR (thin film, cm-1) 2086, 1734, 1641, 1371, 1265, 1047. N O PhSO2 O 84 N2 84: 1H NMR (250 MHz, CDCl3) 8.03-7.53 (m, 5 H), 4.01 (dd, 1 H, J = 5.0, 9.0 Hz), 3.82 (dd, 1 H, J = 2.5, 9.0 Hz), 3.79 3.73 (m, 1 H), 1.75-1.65 (m, 2 H) 1.64 (s, 3 H), 1.42 (s, 3 H), 0.91 (t, 3 H, J = 7.4 Hz); 13C NMR (62.5 MHz, CDCl3) 153.8, 142.0, 133.8, 129.1, 127.8, 96.6, 74.3, 67.3, 58.6, 26.9, 26.2, 23.6, 10.1; IR (thin film, cm-1) 1696, 1419, 1367, 1304, 1141, 1080, 1026.

PAGE 87

68N O PhSO2 O 86 N2 86: 1H NMR (250 MHz, CDCl3) 8.02-7.53 (m, 5 H), 4.0 (m, 1 H), 3.83 (m, 2 H), 1.64 (m, 2 H) 1.61 (s, 3 H), 1.41 (s, 3 H), 1.31 (m, 2 H), 0.96 (t, 3 H, J = 7.2 Hz); 13C NMR (62.5 MHz, CDCl3) 153.6, 141.9, 133.7, 129.0, 127.7, 96.4, 74.1, 67.6, 57.2, 36.1, 26.2, 23.5, 19.1, 13.8; IR (thin film, cm-1) 2089, 1638, 1388, 1153, 1085. N O PhSO2 O 88 N2 88: 1H NMR (360 MHz, CDCl3) 8.01-7.55 (m, 5 H), 4.06 3.95 (m, 2 H), 3.81 (dd, 1 H, J = 2.7, 6.1 Hz) 1.63 (s, 3 H), 1.41 (s, 3 H), 0.97 (d, 3 H, J = 3.5 Hz), 0.96 (d, 3 H, J = 3.8 Hz); 13C NMR (62.5 MHz, CDCl3) 153.5, 141.9, 133.8, 129.0, 127.6, 96.1, 73.6, 67.8, 56.2, 43.1, 26.3, 25.5, 23.6, 21.3; IR (thin film, cm-1) 2092, 1645, 1388, 1266, 1155, 1084.

PAGE 88

69N O PhSO2 O 90 N2 Ph 90: 1H NMR (250 MHz, CDCl3) 8.02-7.52 (m, 5 H), 7.36 7.18 (m, 5 H), 4.02 (dd, 1 H, J = 5.7, 8.7 Hz) 3.88 (d, 1 H, J = 8.7 Hz), 3.82 (m, 1 H), 2.74-2.50 (m, 2 H), 2.02 (m, 2 H), 1.63 (s, 3 H), 1.41 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 153.7, 142.0, 140.0, 133.8, 129.1, 128.8, 128.3, 127.8, 126.5, 96.5, 74.0, 67.5, 56.8, 35.8, 32.3, 26.4, 23.7; IR (thin film, cm-1) 2102, 1617, 1374, 1308, 1142, 1083. N O PhSO2 O 92 N2 MeO2C 92: 1H NMR (250 MHz, CDCl3) 8.0-7.54 (m, 5 H), 4.08-3.97 (m, 1 H), 3.81 (d, 1 H, J = 9.0 Hz), 3.73 (s, 3 H), 2.36 (m, 2 H), 1.99 (m, 2 H), 1.64 (s, 3 H), 1.39 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 172.4, 153.9, 141.8, 133.8, 129.1, 127.6, 96.4, 74.3, 67.3, 56.5, 51.9, 30.0, 29.0, 26.6, 23.4; IR (thin film, cm-1) 2096, 1730, 1630, 1391, 1317, 1147, 1080.

PAGE 89

70N O PhSO2 O 94 N2 4-(MeO)Ph 94: 1H NMR (250 MHz, CDCl3) 7.98-7.57 (m, 5 H), 7.13 (d, 2 H, J = 8.5 Hz), 6.88 (d, 1 H, J = 8.5 Hz), 4.27 (m, 1 H), 3.87 (d, 2 H, J = 3.2 Hz), 3.8 (s, 3 H), 3.03 (dd, 1 H, J = 5.0, 13.0 Hz), 2.79 (dd, 1 H, J = 8.9, 13.0 Hz), 1.69 (s, 3 H), 1.36 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 158.6, 154, 141.8, 133.9, 130.5, 129.3, 128.4, 127.5, 114.3, 96.7, 74.8, 67.2, 59.1, 55.2, 39.3, 26.9, 23.4; IR (thin film, cm-1) 2087, 1623, 1512, 1363, 1244, 1150, 1083. N O PhSO2 O 96 N2 96: 1H NMR (250 MHz, CDCl3) 8.0-7.51 (m, 5 H), 4.09-3.94 (m, 2 H), 3.68 (dd, 1 H, J = 3.0, 8.5 Hz), 1.60 (s, 3 H), 1.40 (s, 3 H), 1.26 (d, 3 H, J = 6.1 Hz); 13C NMR (62.5 MHz, CDCl3) 153.7, 142.0, 133.8, 129.1, 127.7, 96.7, 74.2, 69.9, 52.8, 26.2, 23.8, 20.1; IR (thin film, cm-1) 2097, 1641, 1257, 1156.

PAGE 90

71N O PhSO2 O 98 N2 98: 1H NMR (250 MHz, CDCl3) 8.0-7.51 (m, 5 H), 3.95-3.79 (m, 3 H), 2.10 (m, 1 H), 1.64 (s, 3 H), 1.40 (s, 3 H), 0.95 (d, 3 H, J = 7.0 Hz), 0.90 (d, 3 H, J = 6.7 Hz); 13C NMR (62.5 MHz, CDCl3) 154.5, 142.0, 133.8, 129.2, 127.6, 96.6, 74.8, 64.9, 62.3, 30.0, 25.9, 23.6, 19.1, 16.8; IR (thin film, cm-1) 2097, 1641, 1268, 1048. N O PhSO2 O 100 N2 100: 1H NMR (250 MHz, CDCl3) 8.01-7.26 (m, 5 H), 3.98-3.82 (m, 3 H), 1.85 (m, 1 H), 1.61 (s, 3 H), 1.41 (s, 3 H), 1.4 1.0 (m, 2 H), 0.95 (t, 2 H, J = 7.2 Hz), 0.85 (d, 3 H, J = 6.8 Hz); 13C NMR (62.5 MHz, CDCl3) 154.4, 141.9, 133.8, 129.1, 127.6, 96.6, 74.5, 64.5, 61.0, 36.3, 26.2, 25.3, 23.9, 13.3, 12.1; IR (thin film, cm-1) 2097, 1641, 1257, 1156.

PAGE 91

72N O PhSO2 O OTBS 103 N2 103: 1H NMR (250 MHz, CDCl3) 8.0-7.53 (m, 5 H), 3.98 (m, 3 H), 3.62 (m, 2 H), 1.58 (s, 3 H), 1.41 (s, 3 H), 0.86 (s, 9 H), 0.059 (s, 6 H); 13C NMR (62.5 MHz, CDCl3) 154.1, 142.1, 133.7, 129.0, 127.8, 96.8, 74.6, 65.9, 62.7, 58.2, 26.5, 25.7, 23.5, 18.2, -5.58, IR (thin film, cm-1) 2092, 1634, 1448, 1378, 1261, 1155, 1087. N O PhSO2 O OTBS 105 N2 105: 1H NMR (250 MHz, CDCl3) 8.0-7.53 (m, 5 H), 4.24 (dq, 1 H, J = 6.3, 6.3 Hz), 3.92 (dd, 1 H, J = 5.5, 10.0 Hz), 3.73 (dd, 1 H, J = 2.5, 10.0 Hz), 3.55 (m, 1 H), 1.56 (s, 3 H), 1.42 (s, 3 H), 1.36 (d, 3 H, J = 6.1 Hz), 0.89 (s, 9 H), 0.092 (s, 3 H), 0.082 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 154.9, 141.9, 133.8, 129.1, 127.5, 96.4, 75.5, 73.7, 64.6, 61.0, 60.2, 25.7, 25.6, 24.5, 20.9, 19.1, 18.2, 14.1, -5.68, -5.79; IR (thin film, cm-1) 2090, 1646, 1344, 1253, 1154, 1086.

PAGE 92

73 General Procedure for Rhodium Catalyzed Intramolecular C-H Insertion To a solution of an diazo-(phenylsulfonyl)acetamide (1 mmol) in dry CH2Cl2 (20 mL), was added Rh2(OAc)4 (11 mg, 2.5 mol%). The mixture was refluxed for 12 h under N2, cooled to r.t., and concentrated. The residue was chromatographed to give pure lactam. N O O Ph PhSO2 H 78 78: 1H NMR (250 MHz, CDCl3) 7.93-7.49 (m, 5 H), 7.35-7.23 (m, 5 H), 4.50 (d, 1 H, J = 9.1 Hz), 4.14-4.05. (m, 2 H), 3.88 (dd, 1 H, J = 6.2, 9.1 Hz), 3.68 (dd, 1 H, J = 7.9, 8.0 Hz), 1.65 (s, 3 H), 1.45 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) ; IR (thin film, cm-1) 1734, 1373, 1266, 1246, 1046. N O O PhSO2 H 85 85: 1H NMR (250 MHz, CDCl3) 8.01-7.54 (5 H, m) 4.12 (dd, 1 H, J = 5.5, 8.5 Hz), 4.00 (d, 1 H, J = 9.9 Hz), 3.73 (m, 1 H), 3.48 (dd, 1 H, J = 8.5, 9.0 Hz), 2.81 (m, 1 H), 1.51 (s, 3 H), 1.42 (d, 3 H, J = 6.7 Hz), 1.38 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 161.7, 137.8, 134.1, 129.8, 128.8, 92.4, 76.2, 68.8, 64.9, 34.1, 26.4, 23.5, 18.7; IR (thin

PAGE 93

74 film, cm-1) 1696, 1419, 1367, 1304, 1142, 1080, 1023. N O O Et PhSO2 H 87 87: 1H NMR (250 MHz, CDCl3) 8.01-7.54 (5 H, m), 4.10 (dd, 1 H, J = 4.1, 8.6 Hz), 4.05 (d, 1 H, J = 9.6 Hz), 3.77 (m, 1 H), 3.51 (dd, 1 H, J = 8.6, 9.0 Hz), 2.68 (m, 1 H), 2.2 (m, 1 H), 1.54 (s, 3 H), 1.40 (s, 3 H), 0.97 (t, 3 H, J = 7.4 Hz); 13C NMR (62.5 MHz, CDCl3) 161.6, 137.7, 134.0, 129.7, 128.7, 91.9, 74.7, 69.4, 63.6, 40.2, 26.4, 26.3, 23.3, 11.6; IR (thin film, cm-1) 1698, 1416, 1306, 1142, 1082, 1028. N O O PhSO2 H 89 89: 1H NMR (250 MHz, CDCl3) 8.03-7.58 (5 H, m), 4.16 (d, 1 H, J = 8.2 Hz), 4.07 (dd, 1 H, J = 5.5, 7.9 Hz), 3.87 (m, 1 H), 3.59 (dd, 1 H, J = 7.9, 9.3 Hz) 2.79 (1H, m), 2.22 (m, 1 H), 1.62 (s, 3H), 1.45 (s, 3H), 1.05 (d, 3 H, J = 6.7 Hz), 1.04 (d, 3 H, J = 6.9 Hz); 13C NMR (62.5 MHz, CDCl3) 162.4, 137.9, 134.1, 129.7, 128.8, 92.4, 72.7, 69.6, 59.9, 42.3, 29.1, 26.9, 23.4, 20.6, 17.5 ; IR (thin film, cm-1) 1706, 1407, 1309, 1266, 1150, 1085.

PAGE 94

75N O O Bn PhSO2 H 91 91: 1H NMR (250 MHz, CDCl3) 8.08-7.50 (5 H, m), 7.37-7.21 (m, 5 H), 4.22 (d, 1 H, J = 10.2 Hz), 3.78 (m, 1 H), 3.55 (dd, 1 H, J = 3.9, 13.3 Hz) 3.21-3.01 (m, 2 H), 2.70 (dd, 1 H, J = 11.6, 13.0 Hz), 1.45 (s, 3H), 1.35 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 161.2, 137.8, 134.2, 129.8, 129, 128.9, 128.7, 127.2, 91.9, 74.3, 68.9, 63.1, 41.8, 38.9, 26.3, 23.4; IR (thin film, cm-1) 1694, 1316, 1308, 1145, 1083, 1025. N O O PhSO2 H MeO2C 93 93: 1H NMR (250 MHz, CDCl3) 8.03-7.26 (5 H, m), 4.23-4.16 (m, 2 H), 3.81 (m, 1 H), 3.72 (s, 3 H), 3.58 (dd, 1 H, J = 9.0, 9.1 Hz), 3.24 (dd, 1 H, J = 3.3, 17.2 Hz), 3.09 (m, 1 H), 2.67 (dd, 1 H, J = 10.5, 17.2 Hz), 1.49 (s, 3H), 1.39 (s, 3H); 13C NMR (62.5 MHz, CDCl3) 171.6, 160.9, 137.1, 134.3, 129.9, 128.8, 92.1, 73.5, 69.9, 63.8, 52.0, 36.7, 35.3, 26.4, 23.3; IR (thin film, cm-1) 1734, 1705, 1281, 1143, 1080, 1036.

PAGE 95

76N O O PhSO2 H 4-(MeO)Ph 95 95: 1H NMR (250 MHz, CDCl3) 7.95-7.50 (m, 5 H, 7.19 (d, 2 H, J = 8.6 Hz), 6.89 (d, 1 H, J = 8.6 Hz), 4.46 (d, 1 H, J = 8.6 Hz), 4.16-4.07 (m, 2 H), 3.86 (m, 1 H), 3.81 (s, 3 H), 3.70 (m, 1 H), 1.66 (s, 3 H), 1.46 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 161.7, 159.2, 137.7, 134.1, 130.7, 129.7, 128.8, 128.5, 114.6, 92.9, 69.0, 65.9, 55.3, 42.8, 26.7, 23.5; IR (thin film, cm-1) 1711, 1365, 1217. N O O PhSO2 H 97 97: 1H NMR (250 MHz, CDCl3) 8.0-7.34 (m, 5 H), 3.42-4.02 (m, 3 H), 3.39 (m, 1 H), 2.89-2.17 (m, 2 H), 1.56 and 1.53 (s, 3H), 1.38 (s, 3 H); IR (thin film, cm-1) 1700, 1405, 1310, 1265, 1150, 1085, 1039. N O O PhSO2 H 99 99: 1H NMR (250 MHz, CDCl3) 8.06-7.49 (m, 5 H), 4.49 and 3.78 (m, 3 H), 4.07 and 3.54 (s, 1 H), 1.60 and 1.59 (s, 3 H), 1.50 (s, 3 H), 1.43 and 1.42 (s, 3 H), 1.32

PAGE 96

77 and 1.11 (s, 3 H); IR (thin film, cm-1) 1701, 1448, 1393, 1147, 1058, 1214, 1085, 1036. N O O PhSO2 H Et Me 101 101: 1H NMR (250 MHz, CDCl3) 8.10-7.53 (m, 5 H), 4.56 and 3.80 (m, 3 H), 4.12 and 3.61 (s, 1 H), 2.30 (m, 1 H), 1.63 and 1.52 (s, 3 H), 1.47 (s, 3 H), 1.37 and 1.11 (s, 3 H), 0.94 and 0.91 (t, 3 H, J = 7.5 Hz); IR (thin film, cm-1) 1706, 1422, 1369, 1265, 1150, 1084. N O O TBSO PhSO2 H 104 104: 1H NMR (360 MHz, CDCl3) 8.01-7.55 (m, 5 H), 4.83 (dd, 1 H, J = 4.8, 6.7 Hz), 4.26 (d, 1 H J = 6.8 Hz), 4.17 (dd, 1 H, J = 5.8, 8.2 Hz), 3.97 (ddd, 1 H, J = 4.8, 5.8, 8.4 Hz), 3.60 (dd, 1 H, J = 8.4, 8.4 Hz), 1.57 (s, 3 H), 1.41 (s, 3 H), 0.91 (s, 9 H), 0.24 (s, 3 H), 0.17 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 160.7, 138.1, 134.1, 129.6, 128.8, 92.7, 77.8, 69.3, 67.6, 66.8, 26.6, 25.6, 23.4, 17.9, -4.85, -5.15; IR (thin film, cm-1) 1713, 1447, 1371, 1265, 1153, 1082.

PAGE 97

78N O O TBSO PhSO2 H 106 106: 1H NMR (360 MHz, CDCl3) 7.97-7.54 (m, 5 H), 4.85 (dd, 1 H, J = 3.5, 4.8 Hz), 4.12 (d, 1 H J = 4.8 Hz), 3.82 (m, 1 H), 3.51 (dd, 1 H, J = 3.5, 9.3 Hz), 1.59 (s, 3 H), 1.43 (s, 3 H), 1.35 (d, 3 H, J = 6.0 Hz), 0.91 (s, 9 H), 0.27 (s, 3 H), 0.23 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 161.5, 137.9, 134.2, 129.6, 129.2, 128.8, 92.5, 79.2, 74.4, 73.8, 67.0, 27.5, 25.6, 23.6, 17.9, 16.8, -4.51, -5.17 ; IR (thin film, cm-1) 1718, 1310, 1250, 1151, 1083. Preparation of -lactam 105 N O O PhSO2 H Et Me 101 N O O H Et Me 102 Na(Hg) NaH2PO4MeOH To a solution of lactam 101 (62 mg, 0.18 mmol) and anhydrous disodium hydrogen phosphate (104 mg, 0.72 mmol) in 2 mL of methanol cooled to the 0 C was added pulverized 10% sodium amalgam (168 mg). The mixture was stirred for 30 min., poured into water and extracted with ether. The organic layer was dried over Na2SO4, filtered and concentrated. The residue was chromatographed to give the reduced lactam in 83% yield as a single isomer: 1H NMR (250 MHz, CDCl3) 3.94 (dd, 1 H, J = 6.2, 8.7

PAGE 98

79 Hz), 3.85 (dd, 1 H, J = 6.2, 8.5 Hz), 3.68 (dd, 1 H, J = 8.5 8.7 Hz), 2.87 and 1.94 (ABq, 2 H, J = 16 Hz) 1.62 (s, 3 H), 1.44 (s, 3 H), 0.99 (s, 3 H), 0.84 (t, 3 H, J = 7.4 Hz); 13C NMR (62.5 MHz, CDCl3) 170.0, 90.3, 68.6, 63.8, 51.4, 40.6, 32.8, 26.0, 23.4, 19.0, 9.5; IR (thin film, cm-1) 1700. General Preparation of -Amino Acid Derived N,OAcetonides To a suspension of LAH (3.8 g, 100 mmol) and 250 mL of anhydrous THF (0.2 M) cooled to 0 C was slowly added 50 mmol of amino acid. The entire suspension was then refluxed under N2 atmosphere for 8 hrs. The reaction mixture was cooled to 0 C and, while stirring, 3.8 mL of water was added via dropping funnel. It was followed by 3.8 mL of 20% NaOH and 7.6 mL of water. The entire solution was then stirred for an additional half hour and filtered through a sintered glass funnel with a celite pad. The filter cake was rinsed with a portion of anhydrous THF and the filtrate was then evaporated yielding the crude amino alcohol. The crude product was then combined with 250 mL of a 1:1 mixture of acetone and 1,2-DCE and 10 grams of anhydrous Na2SO4 and refluxed under N2 atmosphere for 1 hr. The solution was then cooled to r.t. and filtered through a sintered glass funnel with a pad of celite. The filter cake was rinsed with a portion of acetone and the filtrate was evaporated yielding the crude acetonide.

PAGE 99

80HN O Ph 110 110: 1H NMR (250 MHz, CDCl3) 7.33-7.19 (m, 5 H), 3.87 (dd, 1 H, J = 7.1, 7.4 Hz), 3.69 (m, 1 H), 3.39 (dd, 1 H, J = 7.7, 7.8 Hz), 3.00 (dd, 1 H, J = 5.8, 13.6 Hz), 2.72 (dd, 1 H, J = 7.8, 13.6 Hz) 1.44 (s, 3 H), 1.31 (s, 3H); 13C NMR (62.5 MHz, CDCl3) 137.7, 128.3,128.1, 126.0, 94.6, 69.7, 58.6, 39.0, 27.3, 26.1. General Preparation of -Amino Acid Derived N,N’Acetonides Thionyl chloride (3.65 mL, 50 mmol) was added drop wise to a solution of (L)phenylalanine (7.6 g, 46 mmol) in methanol (60 mL) at 0 C, and the mixture was heated under reflux 1 h. The solvent was evaporated to give (L)-phenylalanine methyl ester hydrochloride as white solid. The solid was then redissolved in MeOH and methylamine was added and stirred for 4 hrs. The solvent was evaporated and the residue was dissolved in EtOAc and neutralized with 1 N HCl. The organic solution was dried and concentrated yielding crude aminoamide. The crude product was then combined with 250 mL of a 1:1 mixture of acetone and 1,2-DCE and 10 grams of anhydrous Na2SO4 and refluxed under N2 atmosphere for 1 hr. The solution was then cooled to r.t. and filtered through a sintered glass funnel with a pad of celite. The filter cake was rinsed with a portion of acetone and the filtrate was evaporated yielding the crude acetonide.

PAGE 100

81HN N Ph 131 O 131: 1H NMR (250 MHz, CDCl3) 7.25-7.00 (m, 5 H), 3.64 (dd, 1 H, J = 4.5, 7.6 Hz), 2.98, (dd, 2 H, J = 4.4, 14.1 Hz), 2.57 (s, 3 H), 1.09 (s, 3 H), 0.99 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 172.7, 136.6, 128.9, 127.9, 126.1, 75.0, 58.6, 36.6, 26.5, 24.5. General Preparation of -Diazo--(ethoxycarbonyl)acetamides To a solution of crude acetonide (10 mmol) in anhydrous THF (50 mL) and TEA (0.153 mL, 11 mmol) at 0 C was added me thyl-3-chloro-3-oxopropionate (0.118 mL, 11 mmol) and stirred under N2 atmosphere for 1 hr. The reaction mixture was evaporated and the residue was diluted with EtOAc and washed with water, dried and concentrated. The crude product was then diluted with MeCN (50 mL) and pABSA (360 mg, 15 mmol) added. The solution was cooled to 0 C, DBU was added drop wise and the mixture was stirred for 12 hrs. The reaction mixture was then evaporated, the residue dissolved in equal parts EtOAc and water and the organic phase was then washed with water twice, dried and concentrated. The re d-brown residue was purified by flash column and the pale yellow product was recovered as an oil.

PAGE 101

82N O Ph EtO O O N2 111 111: 1H NMR (250 MHz, CDCl3) 7.12-6.92 (m, 5 H), 3.99 (q, 2 H, J = 7.2 Hz), 3.78 (m, 1 H), 3.77 (dd, 1 H, 5.0, 8.5 Hz), 3.62 (d, 1 H, J = 8.5 Hz), 2.78 (dd, 1 H, J = 7.9, 12.9 Hz), 2.57 (dd, 1 H, J = 6.8, 13.1 Hz), 1.78 (s, 3 H), 1.31 (s, 3 H), 1.06 (t, 3 H, J = 7.0 Hz); 13C NMR (62.5 MHz, CDCl3) 161.8, 157.2, 137.5, 129.1, 128.4, 126.7, 96.2, 68.3, 67.8, 60.8, 58.6, 40.7, 27.4, 23.0, 14.2. N O MeO O O N2 124 124: 1H NMR (250 MHz, CDCl3) 4.15 (ddd, 1 H, J = 2.3, 5.8, 13.9 Hz), 3.88 (dd, 1 H, J = 5.8, 8.6 Hz), 3.65 (s, 3 H), 3.67 3.61 (m, 1 H), 1.53 (s, 3 H), 1.43 (m, 2 H), 1.36 (s, 3 H), 1.12 (m, 2 H), 0.75 (t, 3 H, J = 7.3 Hz); 13C NMR (62.5 MHz, CDCl3) 162.5, 156.8, 95.9, 67.7, 67.5, 57.0, 52.1, 36.3, 26.7, 23.3, 19.2, 13.6.

PAGE 102

83N N Ph MeO O O N2 133 O 133: 1H NMR (250 MHz, CDCl3) 7.26-6.94 (m, 5 H), 4.97 (t, 1 H, J = 4.5 Hz), 3.66 (s, 3 H), 3.04 (d, 2 H, J = 4.6 Hz), 2.63 (s, 3 H), 1.43 (s, 3 H), 0.87 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 167.0, 161.2, 158.1, 135.6, 129.2, 128.1, 126.7, 80.0, 68.4, 59.6, 51.8, 37.3, 23.8, 23.0, 22.7. General Preparation of -Diazo--(acetyl)acetamides To a solution of acetonide (10 mmol) in anhydrous THF (50 mL) at 0 C was added diketene (0.848 mL, 11 mmol) and stirred under N2 atmosphere for 1 hr. The reaction mixture was evaporated and the residue was diluted with EtOAc and washed with water, dried and concentrated. The crude product was then diluted with MeCN (50 mL) and pABSA (360 mg, 15 mmol) added. The solution was cooled to 0 C, DBU was added drop wise and the mixture was stirred for 12 hrs. The reaction mixture was then evaporated, the residue dissolved in equal parts EtOAc and water and the organic phase was then washed with water twice, dried and concentrated. The red-brown residue was purified by flash column and the pale yellow product was recovered as an oil.

PAGE 103

84N O Ph O O N2 112 112: 1H NMR (250 MHz, CDCl3) 7.37-7.12 (m, 5 H), 4.34-4.25 (m, 1 H) 4.30 (dd, 1 H, J = 6.3, 13.2 Hz), 3.87 (d, 1 H, J = 8.8 Hz), 2.98 (dd, 1 H, J = 8.5, 13.3 Hz), 2.81 (dd, 1 H, J = 6.3, 13.4 Hz), 2.12 (s, 1 H), 1.85 (s, 1 H), 1.53 (s, 1 H); 13C NMR (62.5 MHz, CDCl3) 186.9, 156.6, 137.0, 128.8, 128.3, 126.6, 96.1, 67.8, 58.4, 40.6, 27.0, 26.2, 22.8. N O O O N2 125 125: 1H NMR (250 MHz, CDCl3) 3.88 (dd, 1 H, J = 5.6, 8.6 Hz), 3.73 (m, 1 H), 3.62 (dd, 1 H, J = 2.7, 8.7 Hz), 2.16 (s, 3 H), 1.49 (s, 3 H), 1.41 (m, 2 H), 1.35 (s, 3 H), 1.08 (m, 1 H); 0.72 (t, 3 H, J = 7.3 Hz); 13C NMR (62.5 MHz, CDCl3) 188.9, 156.3, 95.9, 74.3, 67.7, 57.1, 36.2, 27.1, 26.3, 23.4, 19.1, 13.6.

PAGE 104

85N N Ph O O N2 134 O 134: 1H NMR (250 MHz, CDCl3) 7.31-7.06 (m, 5 H), 4.74 (t, 1 H, J = 4.5 Hz), 3.14 (d, 2 H, J = 4.8 Hz), 2.77 (s, 3 H), 2.24 (s, 3 H), 1.56 (s, 3 H), 1.05 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 186.8, 166.8, 157.8, 135.5, 129.1, 128.2, 126.8, 80.0, 59.6, 53.1, 37.6, 26.4, 23.9, 23.0, 22.7. General Preparation of -Diazo-acetamides. To a mixture of 5:1 water/MeCN (13.3 mL) was added the diazo -( acetyl)acetamide (10 mmol) followed by LiOH (1.47 g, 35 mmol). The reaction mixture was then stirred at r.t. for 20 hrs. The solvent was evaporated and the residue dissolved in EtOAc. The organic solution was then washed with water twice and dried. Flash column chromatography was used for purification of the product.

PAGE 105

86N O Ph O N2 H 113 113: 1H NMR (250 MHz, CDCl3) 7.37-7.19 (m, 5 H), 4.83 (s, 1 H), 3.87 (s, 2 H), 3.70 (m, 1 H), 3.02 (dd, 1 H, J = 2.0, 12.8 Hz), 2.82 (dd, 1 H, J = 10.3, 13.1 Hz), 1.73 (s, 3 H), 1.58 (s, 3 H). N O O N2 H 126 126: 1H NMR (250 MHz, CDCl3) 4.76 (s, 1 H), 3.80 (dd, 1 H, J = 5.3, 8.9 Hz), 3.67 (d 1 H, 8.8 Hz), 3.36 (m, 1 H), 1.51 (s, 3 H), 1.46 (m, 2 H), 1.40 (s, 3 H), 1.18 (m, 2 H), 0.83 (t, 3 H, J = 7.0 Hz); 13C NMR (62.5 MHz, CDCl3) 161.8, 95.2, 66.8, 57.2, 47.9, 35.8, 26.9, 23.3, 19.5, 13.7. N O O EtO2C Ph H 114 114: 1H NMR (250 MHz, CDCl3) 4.30 (m, 1 H), 4.25 (q, 2 H, J = 7.6 Hz), 4.16 (dd, 1 H, J = 5.8, 8.1 Hz), 4.04 (d, 1 H, J = 11.9 Hz), 3.83 (dd, 1 H, J = 8.6, 12.6 Hz),

PAGE 106

87 3.77 (t, 1H, J = 8.6 Hz), 1.71 (s, 3 H), 1.50 (s, 3 H), 1.25 (t, 3 H, J = 7.6 Hz); 13C NMR (62.5 MHz, CDCl3) 168.5, 165.1, .137.4, 129.1, 127.8, 127.0, 92.1, 69.0, 65.1, 61.7, 60.9, 48.7, 26.5, 23.7, 14.1. N O O MeOC Ph H 115 115: 1H NMR (250 MHz, CDCl3) 7.36-7.20 (m, 5 H), 4.27 (m, 1 H), 4.19 (m, 1 H), 4.13 (m, 1 H), 3.90 (dd, 1 H, J = 8.6, 11.1 Hz) 3.67 (t, 1 H, J = 8.6 Hz) 2.39 (s, 3 H), 1.65 (s, 3 H), 1.51 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 202.2, 165.3, 138.3, 129.0, 127.6, 127.3, 91.9, 69.2, 67.4, 64.7, 45.4, 31.1, 26.6, 23.7. N O H H O 116 116: 1H NMR (250 MHz, CDCl3) 6.58 (m, 2 H), 6.27 (d, 2 H, J = 9.1 Hz), 6.07 (s, 1 H), 5.12 (dd, 1 H, J = 5.4, 9.0 Hz), 4.20 (dd, 1 H, J = 5.1, 7.5 Hz), 3.70 (m, 2 H), 2.76 (dd, 1 H, J = 2.0, 14.8 Hz), 2.47 (m, 2 H), 1.68 (s, 3 H), 1.62 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 168.1, 131.7, 130.4, 129.9, 127.2, 121.4, 119.4, 94.9, 69.1, 57.0, 47.7, 34.7, 25.4, 24.0.

PAGE 107

88 N O O MeOC H 128 128: 1H NMR (250 MHz, CDCl3) 4.08 (dd, 1 H, J = 5.6, 8.2 Hz), 3.74 (m, 1 H), 3.60 (d, 1 H, J = 11.0 Hz), 3.48 (t, 1 H, J = 9.1 Hz), 2.54 (m, 1 H), 2.37 (s, 3 H), 1.54 (s, 3 H), 1.41 (s, 3 H), 1.33 (m, 1 H), 0.81 (t, 3 H, J = 4.3 Hz); 13C NMR (62.5 MHz, CDCl3) 203.0, 166.0, 91.3, 69.5, 66.3, 64.1, 42.0, 30.8, 26.5, 25.3, 23.6, 12.1. N O O MeO2C H 127 127: 1H NMR (250 MHz, CDCl3) 3.99 (dd, 1 H, J = 5.8, 8.3 Hz), 3.71 (m, 1 H), 3.64 (s, 3 H), 3.46 (dd, 1 H, J = 8.8, 8.8 Hz), 3.40 (d, 1 H, J = 11.3 Hz), 2.38 (qt, 1 H, J = 2.5, 8.5 Hz), 1.49 (s, 3 H), 1.44 (m, 2 H), 1.30 (s, 3 H), 0.73 (t, 3 H, J = 7.3 Hz); 13C NMR (62.5 MHz, CDCl3) 169.5, 165.7, 91.3, 69.1, 64.4, 60.0, 52.4, 45.3, 26.3, 25.0, 23.5, 11.8.

PAGE 108

89N N O PhSO2 Ph H 135 O 135: 1H NMR (250 MHz, CDCl3) 7.90-7.44 (m, 5 H), 7.29-7.23 (m, 5 H), 4.51 (d, 1 H, J = 9.1 Hz), 4.14 (m, 2 H), 2.80 (s, 3 H), 1.78 (s, 3 H), 1.46 (s, 3 H). N N O MeO2C Ph H 136 O 136: 1H NMR (250 MHz, CDCl3) 7.45-7.13 (m, 5 H), 4.30 (d, 1 H, J = 9.2 Hz), 4.10 (dd, 1 H, J = 9.2, 10.5 Hz), 3.95 (d, 1 H, J = 11.7 Hz), 3.74 (s, 3 H), 2.82 (s, 3 H), 1.84 (s, 3 H), 1.47 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 168.7, 167.6, 167.2, 136.7, 128.9, 127.7, 127.5, 76.9, 63.1, 58.0, 52.8, 47.8, 24.6, 24.5, 23.2. N N O MeOC Ph H 137 O 137: 1H NMR (250 MHz, CDCl3) 7.39-7.24 (m, 5 H), 4.30 (d, 1 H, J = 9.0 Hz), 4.13 (d, 1 H, J = 10.7, 33.0 Hz), 4.11 (d, 1 H, J = 13.1 Hz), 2.80 (s, 3 H), 2.39 (s, 3 H), 1.79 (s, 3 H), 1.48 (s, 3 H); 13C NMR (62.5 MHz, CDCl3) 201.3, 167.5, 167.0, 137.1, 128.4, 128.2, 127.2, 127.1, 76.4, 64.0, 62.3, 44.7, 30.7, 24.1, 22.7.

PAGE 109

90 Preparation of Ot-Bu-(L)-Serine Methyl Ester NH2 CO2H OH (L)-serine i) HCl(g), MeOH, ii) isobutylene, pTsOH, CH2Cl2 153 NH2 CO2Me Ot-Bu Thionyl chloride (38 mL, 0.53 mol) was added to a solution of (L)-serine (50 g, 0.48 mol) in MeOH (0.5 L) at 0 C and the resulting mixture was refluxed for 2 hours. After concentration, the residue was solidified from ether to give (L)-serine methyl ester HCl salt as white solid. Liquid isobutylene (200 mL) was added to a mixture of (L)serine methyl ester HCl salt (20 g, 0.12 mol), p -TsOH (40 g, 0.12 mol) and CH2Cl2 (400 mL) in a thick-walled, well-stoppered flask at –78C and the resulting mixture was stirred for 72 hours at room temperature. After degassing, the reaction mixture was concentrated and the residue was diluted with EtOAc, washed with saturated NaHCO3, dried over anhydrous Na2SO4, and concentrated to afford 153 which was used for the next step without further purification: 1H NMR (250 MHz, CDCl3) 3.72 (s, 3 H), 3.59 (br s, 3 H), 1.82 (br s, 2 H, NH2), 1.15 (s, 9 H); 13C NMR (62.5 MHz, CDCl3) 173.9, 72.4, 63.5, 54.7, 51.3, 26.9.

PAGE 110

91 Preparation of Amino Alcohol 154 153 NH2 CO2Me Ot-Bu 154 NH2 Ot-Bu OH LAH THF, 0 C A solution of 153 (24 g, 0.14 mol) in anhydrous THF was added to the mixture of lithium aluminum hydride (31g, 0.27 mol) in THF (450 mL) at 0 C. The resulting reaction mixture was stirred for 8 hrs. at room temperature and quenched by successive addition of water (30 mL)-20% NaOH (30 mL)-water (60 mL). The precipitate was filtered and washed with THF. The filtrate was dried and concentrated to give amino alcohol 154 as an oil, which was used for the next step without further purification: 1H NMR (250 MHz, CDCl3) 3.58 (m, 2 H), 3.39 (d, 2 H, J = 5.3 Hz), 3.00 (m, 1 H), 1.19 (s, 9 H); 13C NMR (62.5 MHz, CDCl3) 72.6, 64.5, 64.3, 51.9, 26.9.

PAGE 111

92 Preparation of N,OAcetonide 155 154 NH2 Ot-Bu OH Me2CO/1,2-DCE Na2SO4, HN O Ot-Bu 155 The mixture of the amino alcohol 154 (12 g, 80 mmol), acetone (200 mL), and anhydrous Na2SO4 (68 g, 0.48 mol) in 1,2-dichloroethane (200 mL) was refluxed with stirring for 2 hrs. The reaction mixture was filtered and concentrated to give acetonide intermediate 155 Preparation of NAcylated N,OAcetonides 156 and 157 N O Ot-Bu X O X =Br Cl 156 157X X O TEA, CH2Cl2 HN O Ot-Bu 155 Bromoacetyl bromide or chloroacetyl chloride (80 mmol) was added to a mixture of acetonide 155 (15 g, 80 mmol) and TEA (17 mL, 160 mmol) in CH2Cl2 (160 mL) at 0 C. After stirring for 4 hrs. at room temperature, the reaction mixture was washed with water, and the aqueous layer was extracted with CH2Cl2. The combined organic layers were washed with brine, dried over Na2SO4 and evaporated to give the corresponding haloamide 156/157 which was used for the next step without further purification. For

PAGE 112

93 compound 156 : 1H NMR (250 MHz, CDCl3) 4.32 (ABq, 1 H, JAB = 10.7 Hz), 4.10 (m, 1 H), 4.00 (m, 1 H), 3.83 (d, 1 H, J = 9.7 Hz), 3.79 (ABq, 1 H, JAB = 10.7 Hz), 3.43 (d, 2 H, J = 7.0 Hz), 1.66 (s, 3 H), 1.53 (s, 3 H), 1.18 (s, 9 H); 13C NMR (62.5 MHz, CDCl3) 164.7, 95.6, 73.7, 65.7, 63.5, 58.0, 29.8, 27.3, 26.9, 22.0; IR (thin film, cm-1) 2976, 1648, 1415, 1365, 1264, 1193, 726, 705. Anal. Calcd. for C12H22BrNO3: C, 46.76; H, 7.19; N, 4.54. Found: C, 46.29; H, 7.21; N, 4.58, 161: Preparation of -Phenylsulfonylacetamide 158 N O Ot-Bu X O X = N O Ot-Bu PhSO2 O 158 Br Cl 156 157 NaSO2Ph DMF PhSO2Na (84 mmol) was added to a solution of haloacetyl amide 156/157 (80 mmol) in DMF (160 mL) at room temperature. After stirring for 6 hrs. the reaction mixture was diluted with EtOAc, washed with water twice, dried over Na2SO4 and concentrated. The residue was purified by column chromatography to give phenylsulfonyl acetamide 158: 1H NMR (250 MHz, CDCl3) 7.90-7.52 (m, 5 H), 5.10 (A Bq, 1 H, JAB = 14.0 Hz), 4.47 (m, 1 H), 4.04 (A BX, 1 H, JAB = 9.2 Hz, JAX = 5.3 Hz), 3.91 (AB q, 1 H, JAB = 14.0 Hz), 3.77 (AB X, 1 H, JAB = 9.2 Hz, JAX = 0 Hz), 3.42 (d, 2 H, J = 7.3 Hz), 1.57 (s, 3 H), 1.49 (s, 3 H), 1.15 (s, 9 H); IR (thin film, cm-1) 2980, 1651, 1423, 1366, 1320, 1157, 741. Anal. Calcd. for C18H27NO5S: C, 58.51; H, 7.37; N, 3.79;

PAGE 113

94 S, 8.68. Found: C, 58.41; H, 7.20; N, 3.81; S, 8.72. Preparation of -Diazo--(phenylsulfonyl)acetamide 163. N O Ot-Bu PhSO2 O N O Ot-Bu PhSO2 O N2 159 158 pABSA DBU, MeCN To a solution of phenylsulfonylacetamide 158 (6.8 g, 18 mmol) in acetonitrile (90 mL) was added p -ABSA (5.2 g, 21 mmol) followed by DBU (6.8 mL. 45 mmol). The reaction mixture was stirred for 1 hr. at 0 C. After evaporation of acetonitrile the residue was diluted with EtOAc. The organic layers were washed with water twice, dried over Na2SO4 and then evaporated. The residue was purified by column chromatography to give diazo-(phenylsulfonyl)acetamide 159 as yellow solid: 1H NMR (250 MHz, CDCl3) 8.02 (d, 2 H, J = 7.1 Hz), 7.57 (m, 3 H), 3.99-3.91 (m, 3 H), 3.37 (m, 2 H), 1.59 (s, 3 H), 1.41 (s, 3 H), 1.16 (s, 9 H); 13C NMR (62.5 MHz, CDCl3) 154.2, 142.2, 133.6, 129.0, 127.7, 96.8, 74.7, 73.5, 66.3, 61.9, 57.1, 27.2, 26.6, 23.5; IR (thin film, cm-1) 2976, 2092, 1628, 1395, 1267, 1150, 1085, 745.

PAGE 114

95 Preparation of -Lactam 160 N O O t-BuO H PhSO2 160 Rh2(OAc)4CH2Cl2, N O Ot-Bu PhSO2 O N2 159 To a solution of the diazo--(phenylsulfonyl)acetamide 159 (6.0 g, 15 mmol) in dry CH2Cl2 (300 mL, C = 0.05 M) was added catalytic amount of Rh2(OAc)4 (66 mg, 0.15 mmol). The mixture was refluxed with stirring for 12 hrs. under N2, cooled to room temperature, and was then concentrated. The residue was purified by column chromatography to give lactam 160 as a single isomer: 1H NMR (250 MHz, CDCl3) 7.95 (d, 2 H, J = 7.3 Hz), 7.67-7.51 (m, 3 H), 4.69 (dd, 1 H, J = 3.75, 6.0 Hz), 4.25 (d, 1 H, J = 6.0 Hz), 4.14 (A BX, 1 H, JAB = 8.9 Hz, JAX = 5.8 Hz), 3.96 (m, 1 H), 3.62 (AB X, JAB = 8.9 Hz, JBX = 8.0 Hz), 1.58 (s, 3 H), 1.39 (s, 3 H), 1.25 (s, 9 H); 13C NMR (62.5 MHz, CDCl3) 161.1, 138.1, 134.1, 129.5, 128.7, 92.9, 76.9, 76.1, 67.6, 67.4, 67.3, 28.3, 26.9, 23.4; IR (thin film, cm-1) 2977, 1705, 1401, 1321, 1266, 1153, 1080, 889, 840, 737. Anal. Calcd. for C18H25NO5S: C, 58.83; H, 6.86; N, 3.81; S, 8.73. Found: C, 58.80; H, 6.78; N, 3.98; S, 8.68.

PAGE 115

96 Preparation of -Methylated -Lactam 161 N O O t-BuO H 160 PhSO2 161 MeI, NaH DMF, 0 C N O O t-BuO H PhSO2 MeI (2.6 mL, 42 mmol) was added to a mixture of lactam 160 (5.0 g, 14 mmol) and NaH (1.1 g, 28 mmol) in DMF (28 mL) at 0 C. After stirring for 1 hour at 0 C, the reaction was quenched with saturated NH4Cl, extracted with EtOAc, dried over Na2SO4 and concentrated to give methylatedlactam 161 : 1H NMR (250 MHz, CDCl3) 7.907.51 (m, 5 H), 4.84 (d, 1 H, J = 4.0 Hz), 4.18 (A BX, 1 H, JAB = 8.7 Hz, JAX = 5.3 Hz), 3.92 (m, 1 H), 3.79 (AB X, 1 H, JAB = 8.7 Hz, JBX = 7.7 Hz), 1.68 (s, 3 H), 1.51 (s, 3 H), 1.47 (s, 3 H), 1.28 (s, 9 H); 13C NMR (62.5 MHz, CDCl3) 165.6, 135.6, 134.0, 131.3, 128.2, 92.6, 78.9, 76.0, 67.6, 67.5, 67.4, 28.4, 26.9, 23.2, 14.6; IR (thin film, cm-1) 2979, 1707, 1405, 1306, 1265, 1189, 1151, 1114, 742. Anal. Calcd. for C19H27NO5S: C, 59.82; H, 7.13; N, 3.67; S, 8.41. Found: C, 59.87; H, 7.00; N, 3.73; S, 8.42.

PAGE 116

97 Preparation of Reduced -Lactam 161 N O O t-BuO H PhSO2 161 N O O t-BuO H 162 10% Na(Hg) MeOH, 0 C 10% Sodium amalgam (12 g) was added to a solution of methylated lactam 161 (5.0 g, 13 mmol) and anhydrous disodium hydrogen phosphate (11 g) in dry methanol (250 mL) at 0 C. The mixture was stirred for 2 hrs., quenched with water, and extracted twice with ether. The combined organic layers were dried over Na2SO4, filtered and concentrated to give 162 : 1H NMR (250 MHz, CDCl3) 4.14 (m, 1 H), 4.03 (m, 2 H), 3.58 (br t, 1 H), 2.67 (m, 1 H), 1.66 (s, 3 H), 1.44 (s, 3 H), 1.23 (d, 3 H, J = 7.5 Hz), 1.17 (s, 9 H); 13C NMR (62.5 MHz, CDCl3) 173.2, 91.5, 74.2, 72.5, 68.2, 64.8, 48.9, 28.0, 26.5, 23.7, 11.0; IR (thin film, cm-1) 2983, 1699, 1456, 1402, 1364, 1263, 1191, 1101, 1039, 745. Anal. Calcd. for C13H23NO3: C, 64.70; H, 9.61; N, 5.80. Found: C, 64.19; H, 9.59; N, 3.79.

PAGE 117

98 Preparation of Carboxylic Acid 164 H2CrO4 NH O t-BuO CO2H 164 Me2CO N O O t-BuO H 162 Jones reagent (1.0 M, 103 mL, 103 mmol) was added to a solution of 162 (4.2 g, 17 mmol) in acetone (340 mL) at 0 C and the resulting mixture was stirred for 2 hrs. at that temperature. The reaction was quenched with IPA, extracted three times with CH2Cl2, dried over Na2SO4 and concentrated to give carboxylic acid 164 which was used for the next step without further purification. Preparation of Methyl Ester 165 NH O t-BuO CO2H NH O t-BuO CO2Me TMOF MeOH, H2SO4 165 164 A mixture of the carboxylic acid 164 and trimethyl orthoformate (19 mL, 170 mmol) in methanol (340 mL) containing a catalytic amount of sulfuric acid (2-3 drops) was stirred for 4 hours at room temperature. The reaction mixture was neutralized by addition of solid NaHCO3 and evaporated. The residue was diluted with water, extracted with CH2Cl2, dried over Na2SO4, concentrated and purified by column chromatography

PAGE 118

99 to afford 165 : 1H NMR (250 MHz, CDCl3) 5.91 (br s, 1 H), 4.36 (br d, 1 H, J = 7.0 Hz), 4.04 (br s, 1 H), 3.77 (s, 3 H), 2.50 (dq, 1 H, J = 7.0 Hz, 7.3 Hz), 1.23 (s, 9 H), 1.13 (d, 3 H, J = 7.3 Hz); 13C NMR (62.5 MHz, CDCl3) 179.4, 171.3, 74.9, 73.0, 63.2, 52.4, 39.8, 27.9, 8.8. Anal. Calcd. for C11H19NO4: C, 57.62; H, 8.35; N, 6.11. Found: C, 56.71; H, 8.31; N, 5.93. Preparation of Iminoether 166 N O t-BuO CO2Me 166 Me3OBF4CH2Cl2 NH O t-BuO CO2Me 165 To a mixture of 165 (2.6 g, 11 mmol) and 4 molecular sieve in dry CH2Cl2 (110 mL) was added a solution of trimethyloxonium tetrafluoroborate (4.8 g, 31 mmol) in CH2Cl2 (60 mL) at 0 C. The reaction mixture was stirred for 3 hrs., filtered and poured into an aqueous solution of K2CO3. The aqueous layer was extracted with CH2Cl2 and the combined organic layers were dried over Na2SO4, concentrated and chromatographed to give 166 : 1H NMR (250 MHz, CDCl3) 4.38 (dd, 1 H, J = 4.3, 7.3 Hz), 4.29 (d, 1 H, J = 4.3 Hz), 3.89 (s, 3 H), 3.74 (s, 3 H), 2.79 (dq, 1 H, J = 7.3 Hz, 7.3 Hz), 1.25 (s, 9 H), 1.08 (d, 3 H, J = 7.3 Hz); 13C NMR (62.5 MHz, CDCl3) 176.7, 173.0, 75.2, 74.4, 74.0, 55.3, 52.0, 42.0, 28.0, 10.0; Anal. Calcd. for C12H21NO4: C, 59.24; H, 8.70; N, 5.76. Found: C, 58.65; H, 8.75; N, 5.77.

PAGE 119

100 Preparation of -Methylene -Lactam. N O O t-BuO H 10% Na(Hg) MeOH, 0 C 163 N O O t-BuO H 160 PhSO2 10% Sodium amalgam (7.8 g) was added to a solution of lactam 160 (3.0 g, 8.4 mmol) and anhydrous disodium hydrogen phosphate (7.0 g) in dry methanol (160 mL) at 0 C. The mixture was stirred for 2 hrs., concentrated, diluted with water and extracted with ether twice. The combined organic layers were dried over Na2SO4, filtered and concentrated. The residue was chromatographed to give reduced lactam 163 : 1H NMR (250 MHz, CDCl3) 4.09 (m, 3 H), 3.54 (dd, 1 H, J = 7.1, 7.9 Hz), 2.73 (m, 2 H), 1.62 (s, 3 H), 1.42 (s, 3 H), 1.14 (s, 9 H). Preparation of -Methylated -Lactam epi162 N O O t-BuO H N O O t-BuO H epi162 163 MeI LHMDS, -78 C To a solution of reduced lactam 163 (250 mg, 1.0 mmol) in THF (5.0 mL) was added a solution of LHMDS in THF (1.2 mmol) at -78C. After stirring for 15 min at that temperature, MeI (11 L, 1.2 mmol) was added to the reaction mixture, which was

PAGE 120

101 further stirred for 30 min. and quenched with saturated NH4Cl. The aqueous layer was extracted with EtOAc and the combined organic layers were washed with brine, dried over Na2SO4 and concentrated. The residue was chromatographed to give epi162 : 1H NMR (250 MHz, CDCl3) 4.15 (dd, 1 H, J = 5.7, 8.2 Hz), 3.96 (m, 1 H), 3.55 (m, 2 H), 2.78 (dq, 1 H, J = 7.1, 9.7 Hz), 1.62 (s, 3 H), 1.46 (s, 3 H), 1.19 (d, 3 H, J = 7.1 Hz), 1.17 (s, 9 H). Preparation of Carboxylic Acid epi164 H2CrO4 NH O t-BuO CO2H e p i 164 Me2CO N O O t-BuO H epi162 Following the same procedures for the preparation of 164 epi164 was obtained from epi162 in a 56% yield: 1H NMR (250 MHz, CDCl3) 6.2 (br s, 1 H), 4.1 (d, 1 H, J = 3.0 Hz), 4.0 (br s, 1 H), 3.77 (s, 3 H), 2.34 (dq, 1 H, J = 3.0, 7.6 Hz), 1.21 (d, 3 H, J = 7.6 Hz), 1.21 (s, 9 H).

PAGE 121

102 Preparation of Methyl Ester epi165 NH O t-BuO CO2Me epi165 MeOH, H2SO4NH O t-BuO CO2H epi164 TMOF Following the same procedures for the preparation of 165 epi165 was obtained from epi164 in a 70% yield: 1H NMR (250 MHz, CDCl3) 6.23 (s, 1 H), 4.04 (m, 2 H), 3.77 (s, 3 H), 2.34 (qd, 1 H, J = 3.0, 7.8 Hz), 1.21 (m, 12 H). Preparation of Iminoether epi166 N O t-BuO CO2Me epi166 Me3OBF4CH2Cl2 NH O t-BuO CO2Me epi165 Following the same procedures for the preparation of 166 epi166 was obtained from epi165 in a 70% yield: 1H NMR (250 MHz, CDCl3) 4.29 (d, 1 H, J = 3.2 Hz), 4.09 (dd, 1 H, J = 3.3, 3.6 Hz), 3.83 (s, 3 H), 3.75 (s, 3 H), 2.65 (m, 1 H), 1.22 (d, 3 H, J = 7.6 Hz), 1.18 (s, 9 H).

PAGE 122

103 Preparation of Aldol Coupling Product epi167 N O t-BuO CO2Me epi166 N t-BuO O CO2Me HO e p i 167 isobutyraldehyde LDA, -78 C To a solution of iminoether epi166 (250 mg, 1.0 mmol) in THF (5.0 mL) was added a solution of LDA in THF (1.2 mmol) at -78 C. After stirring for 15 min. at that temperature, isobutyraldehyde (11 L, 1.2 mmol) was added to the reaction mixture, which was further stirred for 30 min. and quenched with saturated NH4Cl. The aqueous layer was extracted with EtOAc and the combined organic layers were washed with brine, dried over Na2SO4 and concentrated. The residue was chromatographed to give aldol product epi167 as a single isomer: 1H NMR (250 MHz, CDCl3) 4.22 (d, 1 H, J = 4.1 Hz), 3.94 (dd, 1 H, J = 3.4, 11.1 Hz), 3.85 (s, 3 H), 3.69 (s, 3 H), 2.64 (m, 1 H), 1.95 (m, 1 H), 1.23 (d, 3 H, 7.4 Hz), 1.15 (s, 9 H), 1.00 (d, 6 H, J = 6.8 Hz).

PAGE 123

104 Preparation of Aldol Coupling Products 167 5 R 9 S / 167 5 S N O t-BuO CO2Me 166 N t-BuO O CO2Me HO 167 5 R 9 S / 167 5 S isobutyraldehyde LDA, -78 C 2:1 ratio of diastereomers Following the same procedure for epi167 aldol reaction of 166 afforded an inseparable mixture 167 5 R 9 S and 167 5 S in 70% yield. 167 5 R 9 S : 1H NMR (250 MHz, CDCl3) 4.64 (d, 1 H, J = 8.67 Hz), 3.95 (dd, 1 H, J = 3.0, 10.9 Hz), 3.85 (s, 3 H), 3.67 (s, 3 H), 2.68 (m, 1 H), 1.92 (m, 1 H), 1.15 (s, 9 H), 1.07 (d, 3 H, J = 7.6 Hz), 1.01 (d, 3 H, J = 6.76 Hz), 0.99 (d, 3 H, J = 6.7 Hz). 167 5 S : 1H NMR (250 MHz, CDCl3) 4.36 (d, 1 H, J = 9.5 Hz), 4.08 (dd, 1 H, J = 3.4, 5.5 Hz), 3.87 (s, 3 H), 3.70 (s, 3 H), 3.23 (d, 1 H, J = 3.4 Hz), 2.73 (m, 1 H), 1.85 (m, 1 H), 1.22 (s, 9 H), 1.11 (d, 3 H, J = 7.5 Hz), 1.04 (d, 3 H, J = 6.7 Hz), 0.94 (d, 3 H, J = 6.9 Hz).

PAGE 124

105 Preparation of -Keto Esters 172R and 172S. Dess-Martin N t-BuO O CO2Me O 168 R & S N t-BuO O CO2Me HO 167 5 R 9 S / 167 5 S 2:1 ratio of diastereomers The mixture of keto esters 167 5 R 9 S and 167 5 S were dissolved in CH2Cl2 and cooled to 0 C. Excess DMP solution was added and the reaction was quenched after 5 min. using saturated sodium bicarbonate solution. The organic solution was washed with water, dried and concentrated. The products were separated using flash column chromatography. 168 5 R : 1H NMR (250 MHz, CDCl3) 5.10 (d, 1 H, J = 8.5 Hz), 3.86 (s, 3 H), 3.74 (s, 3 H), 3.06 (m, 1 H), 2.71 (m, 1 H), 1.15 (s, 9 H + 3 H), 1.11 (d, 3 H, J = 7.6 Hz), 1.08 (d, 3 H, J = 6.8 Hz). 168 5 S : 1H NMR (250 MHz, CDCl3) 5.11 (d, 1 H, J = 8.0 Hz), 3.90 (s, 3 H), 3.75 (s, 3 H), 2.87 (m, 2 H), 1.15 (s, 9 H + 3 H), 1.09 (d, 3 H, J = 7.8 Hz), 1.04 (d, 3 H, J = 6.8 Hz).

PAGE 125

106 Preparation of Dihydroxy Esters 169 5 R 9 S and 169 5 S NH HO O CO2Me HO 167 5 R 9 S /167 5 S 2:1 ratio of diastereomers TFA 1% HCl NH HO O CO2H HO 169 5 R ,9 S / 169 5 S 170 5 R 9 S / 170 5 S LiOH MeOH A solution of aldol products 167 5 R 9 S and 167 5 S (160 mg, 0.51 mmol) in trifluoroacetic acid (2.5 mL) was stirred for 1 hr. at room temperature. After concentration, the crude diol compounds were dissolved in a solution of 1% HCl in EtOH. The mixture was stirred for 1 hour, concentrated, diluted with EtOAc, washed with saturated NaHCO3 and dried over Na2SO4. After concentration, the residue was chromatographed to afford an inseparable mixture of dihydroxy lactams 169 5 R 9 S and 169 5 S 169 5 R 9 S : 1H NMR (250 MHz, CD3OD) 4.43 (d, 1 H, J = 6.0 Hz), 3.90 (d, 1 H, J = 7.3 Hz), 3.72 (s, 3 H), 2.94 (m, 1 H), 1.65 (m, 1 H), 1.06 (d, 3 H, J = 7.6 Hz), 0.97 (d, 3 H, J = 6.6 Hz), 0.83 (d, 3 H, J = 6.8 Hz). 169 5 S : 1H NMR (250 MHz, CD3OD) 4.61 (d, 1 H, J = 5.8 Hz), 3.95 (d, 1 H, J = 5.1 Hz), 3.74 (s, 3 H), 2.48 (m, 1 H), 1.76 (m, 1 H), 1.10, (d, 3 H, J = 7.3 Hz), 1.01 (d, 3 H, J = 6.8 Hz), 0.93 (d, 3 H, J = 6.68 Hz). Preparation of clastoLactacystin -Lactone 2 A mixture of dihydroxy lactams 169 5 S & 169 5 R 9 S in 0.1 N NaOH/EtOH solution was stirred for 4 hrs. at r.t. The reaction mixture was neutralized with saturated NH4Cl and concentrated to give a mixture of dihydroxy acids. To the mixture of dihydroxy acids and TEA in THF was added isopropenyl chloroformate at 0 C and

PAGE 126

107 stirred for 1 hour. The reaction mixture was diluted with water, extracted with CH2Cl2, dried over anhydrous. Na2SO4, concentrated and purified by column chromatography to afford 2 1H NMR (250 MHz, Pyridine d5) 10.47 (s, 1 H), 7.87 (d, 1 H, J = 6.8 Hz), 5.69 (d, 1 H, J = 6.1 Hz), 4.36 (dd, 1 H, J = 3.7, 6.8 Hz), 3.06 (m, 1 H), 2.12 (m, 1 H), 1.48 (d, 3 H, J = 7.5 Hz), 1.13 (d, 3 H, J = 6.9 Hz), 1.02 (d, 3 H, J = 6.7 Hz).

PAGE 127

108 References 1 Omura, S.; Fujimoto, T.; Otoguro, K.; Matsuzaki, K.; Moriguchi, R.; Tanaka, H.; Sasaki, Y. J. Antibiot. 1991 44 113. 2 Omura, S.; Matsuzaki, K.; Fujimoto, T.; Kosuge, K.; Furuya, T.; Fujita, S.; Nakagawa, A. J. Antibiot. 1991 44 117. 3 Corey, E. J.; Reichard, G. A. J. Am. Chem. Soc. 1992 114 10677. 4 Sunazuka, T.; Nagamitsu, T.; Matsuzaki, K.; Tanaka, H.; Omura, S.; Smith, A. B. III. J. Am. Chem. Soc. 1993 115 5302. 5 Hidemitsu, U.; Baldwin, J. E.; Russell, A. T. J. Am. Chem. Soc. 1994 116 2139. 6 Fenteany, G.; Standaert, R. F.; Reichard, G. A.; Corey, E. J.; Schreiber, S. L. Proc. Natl. Acad. Sci. U.S.A. 1994 91 3358. 7 Fenteany, G.; Standaert, R. F.; Lane, W. S.; Choi, S.; Corey, E. J.; Schreiber, S. L. Science 1995 268 726. 8 Corey, E. J.; Le, W. –D. Z. Chem. Pharm. Bull. 1999 47 1. 9 Dick, L. R.; Cruikshank, A. A.; Grenier, L.; Melandri, F. D.; Nunes, S. L.; Stein, R. L. J. Biol. Chem. 1996 271 7273. 10 Groll, M.; Ditzel, L.; Lwe, J.; Stock, D.; Bochtler, M.; Bartunik, H. D.; Huber, R. Nature 1997 386 463.

PAGE 128

109 11 Bogyo, M.; McMaster, J. S.; Gaczynska, M.; Tortorella, D.; Goldberg, A. L.; Ploegh, H. Proc. Natl. Acad. Sci. U.S.A. 1997 94 6629. 12 Nagamitsu, T.; Sunazuka, T.; Tanaka, H.; mura, S.; Sprengeler, P. A.; Smith, A. B. III. J. Am. Chem. Soc. 1996 118 3584. 13 Corey, E. J.; Le, W. –D. Z.; Nagamitsu, T.; Fenteany, G. Tetrahedron 1999 55 3305 and references therein. 14 Corey, E. J.; Le, W. –D. Z.; Nagamitsu, T. Angew. Chem. Int. Ed. 1998 37 1676. 15 Soucy, F.; Grenier, L.; Behnke, M. L.; Destree, A. T.; McCormack, T. A.; Adams, J.; Plamondon, L. J. Am. Chem. Soc. 1999 121 9967. 16 Nakagawa, A.; Takahashi, S.; Uchida, A.; Matsuzaki, K.; mura, S.; Nakamura, A.; Kurihara, N.; Nakamatsu, T.; Miyake, Y.; Take, K.; Kainosho, M. Tetrahedron Lett. 1994 35 5009. 17 Pirrung, M. C.; Heathcock, H. C. J. Org. Chem. 1980 45 1727. 18 Braun, M.; Sacha, H. Angew. Chem. Int. Ed. 1991 10 1383. 19 Corey, E. J.;Li, W. Reichard, G. A. J. Am. Chem. Soc. 1998 120 2330. 20 Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986 108 293. 21 Gao, Y.; Hanson, R.; Klunder, J. M.; Ko, S. Y.; Masamune, H.; Sharpless, K. B. J. Am. Chem. Soc. 1987 109 5765. 22 (a) Corey, E. J.; Choi, S. Tetrahedron Lett. 1993 34 6969. (b) Corey; E. J.; Reichard, G. A. Tetrahedron Lett. 1993 34 6973. Corey; E. J.; Reichard, G. A.; Kania, R. Tetrahedron Lett. 1993 34 6977. 23 Seebach, D.; Aebi, J. D. Tetrahedron Lett. 1983 24 3311.

PAGE 129

110 24 Uno, H.; Baldwin, J. E.; Russell, A. T. J. Am. Chem. Soc. 1994 116 2139. 25 Barton, D. H. R.; Subramanian, R. J. Chem. Soc., Perkin Trans. 1 1977 1718. 26 (a) Chida, N.; Takeoka, J.; Tsutsumi, N.; Ogawa, J. J. Chem. Soc. Chem. Commun. 1995 793. (b) Chida, N.; Takeoka, J.; Ando, K.; Tsutsumi, N.; Ogawa, J. Tetrahedron 1997 53 16287. 27 Overman, L. E. J. Am. Chem. Soc. 1978 98 2901. 28 (a) Kang, S. H.; Jun, H. –S. J. Chem. Soc. Chem. Commun. 1998 1929. (b) Kang, S. H.; Jun, H. –S.; Youn, J. –H. Synlett 1998 1045. 29 Kang, S. H.; Lee, S. B. J. Chem. Soc. Chem. Commun. 1998 761. 30 Jacobsen, E. N.; Marko, I.; Mungall, W. S.; Schroder, G; Sharpless, K. B. J. Am. Chem. Soc. 1988 110 1968. 31 Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.; Hartung, J.; Jeong, K. –S.; Kwong, H. –L.; Morikawa, K.; Wang, Z. –W.; Xu, D.; Zhang, X. – L. J. Org. Chem. 1992 57 2768. 32 Ahn, N. T. Topics in Current Chemistry ; Springer-Verlag: New York, 1980; Vol. 88, p 145. 33 (a) Hartwig, W.; Born, L. J. Org. Chem. 1987 52 4352. (b) Paquette, L. A.; Macdonald, D.; Anderson, L. G.; Wright, J. J. Am. Chem. Soc. 1989 111 8037. (c) Meyers, A. I.; Snyder, L. J. Org. Chem. 1993 58 36. (d) Hayasshi, Y.; Kanayama, J.; Yamaguchi, J.; Shoji, M. J. Org. Chem. 2002 67 9443 and references herein. (e) Barrett, A. G. M.; Head, J.; Smith, M. L.; Stock, N. S.; White, A. J. P.; Williams, J. P. J. Org. Chem 1999 64 6005. (f) Roberson, C. W.; Woerpel, K. A. J. Org. Chem. 1999 64 1434. (g) Roberson, C. W.; Woerpel,

PAGE 130

111 K. A. Org. Lett. 2000 2 621. (h) Roberson, C. W.; Woerpel, K. A. J. Am. Chem. Soc. 2002 124 11342. (i) Xia, Q.; Ganem, B. Org. Lett. 2001 3 485. (j) Fenteany, G.; Standaert, R. F.; Lane, W. S.; Choi, C.; Corey, E. J.; Schreiber S. L. Science 1995 268 726 and references therein. (k) Corey, E. J.; Li, W. –D. A. Chem. Pharm. Bull. 1999 47 1. (l) Masse, C. J.; Morgan, A. J.; Adams, J.; Panek, J. S. Eur. J. Org. Chem. 2000 2513. (m) Yoon, C. H.; Flanigan, D. L.; Jung, Y. C.; Jung, K. W. submitted J. Am. Chem. Soc. (n) Yoon, C. H.; Flanigan, D. L.; Chong, B. –D.; Jung, K. W. J. Org. Chem. 2002 67 6582. (o) Yoon, C. H.; Nagle, A.; Chen, C.; Gandhi, D.; Jung, K. W. In press Org. Lett. 2003 (p) Ousmer, M.; Braun, N. A.; Bavoux, C.; Perrin, M.; Ciufolini, M. A. J. Am. Chem. Soc. 2001 ; 123 7534. (q) Snider, B. B.; Lin, H. Org. Lett. 2000 2 643. (r) Wardrop, D. J.; Basak, A. Org. Lett. 2001 3 1053. (s) Macdonald, S. J. F.; Clarke, G. D. E.; Dowle, M. D.; Harrison, L. A.; Hodgson, S. T.; Inglis, G. G. A.; Johnson, M. R.; Shah, P.; Upton, R. J.; Walls, S. B. J. Org. Chem. 1999 64 5166. (t) Andrews, D. M.; Carey, S. J.; Chaignot, H.; Coomber, B. A.; Gray, N. M.; Hind, S. L.; Jones, P. S.; Mills, G.; Robinson, J. E.; Slater, M. J. Org. Lett. 2002 4 4475. (u) Andrews, D. M.; Jones, P. S.; Mills, G.; Hind, S. L.; Slater, M. J.; Trivedi, N.; Wareing, K. J. Bioorg. Med. Chem. Lett. 2003 13 1657. 34 Yoon, C. H.; Zaworotko, M. J.; Moulton, B.; Jung, K. W. Org. Lett. 2001 3 3539. 35 (a) Doyle, M. P.; McKervey, M. A.; Ye, T. Modern Catalytic Methods for Organic Synthesis with Diazo Compounds ;Wiley-Interscience: New York, 1998 and references therein. (b) Doyle, M. P.; Taunton, J.; Pho, H. Q. Tetrahedron

PAGE 131

112 Lett. 1989 30 5397. (c) Doyle, M. P.; Pieters, R. J.; Taunton, J.; Pho, H. Q. J. Org. Chem. 1991 56 820. (d) Wee, A. G. H.; Liu, B.; Zhang, L. J. Org. Chem. 1992 57 4404. (e) Padwa, A. P.; Austin, D. J.; Price, A.T.; Semones, M. A.; Doyle, M. P.; Protopopova, M. N.; Winchester, W. R.; Tran, A. J. Am. Chem. Soc. 1993 ; 115 8669. (f) Wee, A. G. H.,; Slobodian, J. J. Org. Chem. 1996 61 2897. (g) Doyle, M. P.; Kalinin, A. V. Tetrahedron Lett. 1996 37 1371. (h) Hashimoto, S. –I.; Anada, M. Tetrahedron Lett. 1998 39 79. (i) Wee, A. G. H.; Liu, B.; McLeod, D. D. J. Org. Chem. 1998 63 4218. (j) Mood, C. J.; Miah, S.; Slawin, A. M. Z., Mansfield, D. J.; Richards, I. C. Tetrahedron 1998 54 9689. 36 Davies, H. M. L.; Panaro, S. A. Tetrahedron 2000 56 4871. 37 Taber, D. F.; You, K. K.; Rheingold, A. L. J. Am. Chem. Soc. 1996 118 547. 38 (a) Zaragoza, F.; Zahn, G. Prakt. Chem 1995 292. (b) Doyle,M. P.; Kalinin, A. V. Tetrahedron Lett. 1996 37 1371. 39 Yoon, C. H.; Flanigan, D. L.; Chong, B. –D.; Jung, K. W. J. Org. Chem. 2002 67 6582 40 .Doyle, M. P.; Shanklin, M. S.; Oon, S. –M.; Pho, H. Q.; van der Heide, F. R.; Veal, W. R. J. Org. Chem. 1988 53 3384. 41 Ultra dilute conditions were required for this reduction to occur diastereoselectively. Originally, using a 0.05M methanol solution resulted in a mixture of epimers. Further diluting the concentration to 0.025M provides the methyl product selectively.

PAGE 132

113 Appendices

PAGE 133

114 Appendix A: Selected 1H NMR and 13C NMR Spectra

PAGE 134

115 Appendix A: (Continued)

PAGE 135

116 Appendix A: (Continued)

PAGE 136

117 Appendix A: (Continued)

PAGE 137

118 Appendix A: (Continued)

PAGE 138

119 Appendix A: (Continued)

PAGE 139

120 Appendix A: (Continued)

PAGE 140

121 Appendix A: (Continued)

PAGE 141

122 Appendix A: (Continued)

PAGE 142

123 Appendix A: (Continued)

PAGE 143

124 Appendix A: (Continued)

PAGE 144

125 Appendix A: (Continued)

PAGE 145

126 Appendix A: (Continued)

PAGE 146

127 Appendix A: (Continued)

PAGE 147

128 Appendix A: (Continued)

PAGE 148

129 Appendix A: (Continued)

PAGE 149

130 Appendix A: (Continued)

PAGE 150

131 Appendix A: (Continued)

PAGE 151

132 Appendix A: (Continued)

PAGE 152

133 Appendix A: (Continued)

PAGE 153

134 Appendix A: (Continued)

PAGE 154

135 Appendix A: (Continued)

PAGE 155

136 Appendix A: (Continued)

PAGE 156

137 Appendix A: (Continued)

PAGE 157

138 Appendix A: (Continued)

PAGE 158

139 Appendix A: (Continued)

PAGE 159

140 Appendix A: (Continued)

PAGE 160

141 Appendix A: (Continued)

PAGE 161

142 Appendix A: (Continued)

PAGE 162

143 Appendix A: (Continued)

PAGE 163

144 Appendix A: (Continued)

PAGE 164

145 Appendix A: (Continued)

PAGE 165

146 Appendix A: (Continued)

PAGE 166

147 Appendix A: (Continued)

PAGE 167

148 Appendix A: (Continued)

PAGE 168

149 Appendix A: (Continued)

PAGE 169

150 Appendix A: (Continued)

PAGE 170

151 Appendix A: (Continued)

PAGE 171

152 Appendix A: (Continued)

PAGE 172

153 Appendix A: (Continued)

PAGE 173

154 Appendix A: (Continued)

PAGE 174

155 Appendix A: (Continued)

PAGE 175

156 Appendix A: (Continued)

PAGE 176

157 Appendix B: X-Ray Crystallographic Data Table 1. Crystal data and structure refinement for ( 2 ) bdm1. Identification code bdm1 Empirical formula C10 H15 N O4 Formula weight 213.23 Temperature 100(2) K Wavelength 0.71073 Crystal system Orthorhombic Space group P2(1)2(1)2(1) Unit cell dimensions a = 7.2118(6) = 90. b = 8.2964(7) = 90. c = 18.2095(16) = 90. Volume 1089.51(16) 3 Z 4 Density (calculated) 1.300 Mg/m 3 Absorption coefficient 0.101 mm -1 F(000) 456 Crystal size 0.20 x 0.20 x 0.03 mm 3 Theta range for data collection 2.24 to 28.31. Index ranges -9<=h<=9, -10<=k<=11, -21<=l<=24 Reflections collected 9626 Independent reflections 2608 [R(int) = 0.0586] Completeness to theta = 28.31 98.2 % Absorption correction None Refinement method Full-matrix least-squares on F 2 Data / restraints / parameters 2608 / 0 / 139 Goodness-of-fit on F 2 0.944 Final R indices [I>2sigma(I)] R1 = 0.0404, wR2 = 0.0782 R indices (all data) R1 = 0.0495, wR2 = 0.0806 Absolute structure parameter 0.8(11) Largest diff. peak and hole 0.286 and -0.169 e. -3

PAGE 177

158 Appendix B: (Continued) 2

PAGE 178

159 Appendix B: (Continued) Table 7. Crystal data and structure refinement for ( epi-167 ) df01m. Identification code df01m Empirical formula C16 H29 N O5 Formula weight 315.40 Temperature 100(2) K Wavelength 0.71073 Crystal system ? Space group ? Unit cell dimensions a = 6.15(2) = 90. b = 15.20(5) = 104.57(4). c = 9.89(3) = 90. Volume 895(5) 3 Z 2 Density (calculated) 1.171 Mg/m 3 Absorption coefficient 0.086 mm -1 F(000) 344 Crystal size 0.30 x 0.20 x 0.10 mm 3 Theta range for data collection 2.13 to 28.37. Index ranges -7<=h<=8, -20<=k<=20, -13<=l<=13 Reflections collected 13194 Independent reflections 4231 [R(int) = 0.0481] Completeness to theta = 28.37 97.0 % Max. and min. transmission 0.9915 and 0.9747 Refinement method Full-matrix least-squares on F 2 Data / restraints / parameters 4231 / 1 / 229 Goodness-of-fit on F 2 0.951 Final R indices [I>2sigma(I)] R1 = 0.0436, wR2 = 0.0887 R indices (all data) R1 = 0.0514, wR2 = 0.0917 Absolute structure parameter -0.6(8) Largest diff. peak and hole 0.271 and -0.170 e. -3

PAGE 179

160 Appendix B: (Continued) epi -167


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200421Ka 4500
controlfield tag 001 001479810
005 20080428132309.0
006 m d s
007 cr mn
008 040817s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000418
035
(OCoLC)56549620
9
AJS3941
b SE
040
FHM
c FHM
090
QD31.2 (ONLINE)
1 100
Flanigan, David L.
0 245
Studies in rhodium catalyzed intramolecular C-H insertion of amino acid derived -diazo--(substituted)acetamides and its application to the total synthesis of clasto-lactacystin -lactone
h [electronic resource] /
by David L. Flanigan Jr.
260
[Tampa, Fla.] :
University of South Florida,
2004.
502
Dissertation (Ph.D.)--University of South Florida, 2004.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 179 pages.
520
ABSTRACT: Lactacystin is a microbial metabolite isolated by Omura that exhibits neurotrophic activity in neuroblastoma cell lines. Lactacystin and especially its -lactone analog are the first examples of non-polypeptide small molecules capable of specifically inhibiting the 20S proteasome. Various asymmetric total syntheses of lactacystin and its analogs have been reported. The total synthesis of clasto-lactacystin -lactone is achieved using L-serine methyl ester as the starting material and the sole source of stereochemical induction. The success of this synthesis hinges on two featured transformations.The first key step involves formation of the -lactam core via rhodium (II) catalyzed intramolecular C-H insertion of the -diazo--(phenylsulfonyl)acetamide intermediate. The methodology for this transformation has been developed and applied to the synthesis of highly functionalized stereogenic -lactams from natural -amino acids. Three control elements that govern -lactam formation are described. This step is highlighted by the simultaneous creation of two stereogenic centers of the -lactam core. The second key step involves the late stage aldol coupling for quaternary carbon formation and installation of the hydroxyisobutyl group. In all previously reported syntheses, this is the very first aspect which is addressed. The stereochemical outcome of this step is directed by the chiral environment of the enolate itself. Various attempts to achieve selectivity are explored and reported.Completion of the synthesis of clasto-lactacystin -lactone requires 17 steps with an overall yield of 10%. Some general attempts for optimizing the synthetic scheme are discussed as well as the future direction of this research.
590
Adviser: Jung, Kyung Woon
653
-lactam.
Natural product.
C-H activation.
Aldol.
Proteasome inhibitor.
690
Dissertations, Academic
z USF
x Chemistry
Doctoral.
773
t USF Electronic Theses and Dissertations.
949
FTS
SFERS
ETD
QD31.2 (ONLINE)
nkt 8/30/04
4 856
u http://digital.lib.usf.edu/?e14.418