I 7H7: ; o

2607 LIBRARY
Michigan State
University

This is to certify that the
thesis entitled

THE SYNTHESIS OF ARYL PROPANOID COENZY ME A
THIOESTERS AND TAXOL ANALOGS FOR BIOCHEMICAL
INVESTIGATION OF A TAXOL PATHWAY ACYLTRANSFERASE

presented by

SANJIT SANYAL

has been accepted towards fulfillment
of the requirements for the

MASTER OF degree in CHEMISTRY
SCIENCE

 

 

/ Major Professor’s Signature

03/07/2007

 

Date

MSU is an Affirmative Action/Equal Opportunity Institution

_—a---O-.-I-I-.-.-O-o-a-O-.-D-.-.-.-I-.--.-.-C-l-.‘I-O-l-D-D-t-¢-O-C-C-n-O-I-I-h-I-Q-I—

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATEDUE

DAIEDUE

DAIEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p2/CIRC/DaleDueindd-pc1

 

THE SYNTHESIS OF ARYL PROPANOID COENZYME A THIOESTERS AND
TAXOL ANALOGS FOR BIOCHEMICAL INVESTIGATION OF A TAXOL
PATHWAY ACYLTRANSFERASE
By

Sanjit Sanyal

A THESIS
Submitted to
Michigan State University
In partial fiilfillment of the requirements
For the degree of
MASTER OF SCIENCE
Department of Chemistry

2007

ABSTRACT
THE SYNTHESIS OF ARYL PROPANOID COENZYME A THIOESTERS AND
TAXOL ANALOGS FOR BIOCHEMICAL INVESTIGATION OF A TAXOL
PATHWAY ACYLTRANSFERASE
By

Sanjit Sanyal

Coenzyme A thioesters have attracted considerable attention Since they represent an
important class of activated intermediates in various biological pathways. It is estimated
that approximately 4% of all known enzymes require CoA or CoA thioesters as
substrates. The present investigation describes the optimization of the synthesis of
surrogate coenzyme A thioesters that will be used to dissect the proposed mechanism of
an acyltransferase. Modifications of the previously described method by Walker et a1.
(Walker, K., Fujisaki, 8., Long, R. & Croteau, R. Proc. Natl. Acad. Sci, USA. 2002, 99,
12715) resulted in improvement of yields from submiligram quantity (~30%) to near
quantitative conversion (65-70%) with identical reaction conditions; this was most
apparent upon re-synthesis of two productive 3-amino-3-phenylpropanoid CoAs
synthesized in the previous study. Modification of an existing method for the coupling of
the 3'-amino-3'-phenylpropanoid sidechain of Taxol® onto C13 of baccatin III is
described that will give access to a library of Taxol® derivatives including second-
generation taxanes in a short synthetic route. Furthermore, synthesis of radiolabeled

baccatin III, an advanced taxol biosynthetic pathway intermediate, is also described.

T o my parents and wife for their support and love

III

ACKNOWLEDGEMENTS

First, I would like to thank my research advisor, Dr. Kevin D. Walker, for his guidance
and assistance that often includes humor and sometimes anger yet it is always honest and
insightful. Overall, as an advisor, he is appreciated by all who work for him. I
tremendously value the training he has provided thus far. He has taught me to think
critically and execute effective scientific research.

Secondly, I am thankful to Dr. Robert P. Hausinger, Dr. C. K. Chang and Dr. A.
Daniel Jones for their assistance and for serving as members of my guidance committee.
A special thank to Dr. Gavin E. Reid and Dr. Daniel Holmes for their valuable
suggestions and assistance in obtaining spectroscopic data.

Also, I feel very fortunate to work with talented people in my lab. My thanks go
out to senior research technician Karin Klettke for her patient teaching during my first
year in this lab. I am thankful to postdoctoral Dr. Catherine Gatzmeyer for her valuable
time and experience she shared with me. A special thanks goes to fellow-graduate
students, Mark, Irosha, Danielle, Sue and Washington, who not only challenged me with
excellent scientific discussion, but were also very good friends. I would also like to thank
talented undergraduates Erin Merriweather, Thomas J. Edwards, and Amanda Ward for
their assistance during the course of my research.

Last but not least, I would like to express my appreciation and love to my wife
Sulagna, and to my father, mother and sister for their love, encouragement and emotional

support.

TABLE OF CONTENTS

 

LIST OF TABLES AND FIGURES ............................................................................. VI
LIST OF ABBREVIATIONS ........................................................................................ IX
CHAPTER 1. THE SYNTHESIS OF ARYL PROPANOID COENZYME A
THIOESTERS .............. - - ................................... 1
1.1 INTRODUCTION .................................................................................................... 1
1.2 RESULTS AND DISCUSSION. ......................

 

 

CHAPTER 2. SYNTHESIS OF TAXOL ANALOG AND 13-3H-BACCATIN HI
TO ELUCIDATE THE MECHANISM OF A TAXOL PATHWAY BAHD

 

 

ACYLTRANSFERASE 25
2-1 INTRODUCTION .................................................................................................... 25
2-2 RESULTS AND DISCUSSION .............................................................................. 33
CHAPTER 3- EXPERIMENTAL METHODS ..................................... 41
BIBLIOGRAPHY ........................................................................................................... 57

LIST OF TABLES AND FIGURES

TABLE 1: Optimization of reaction time for synthesis of coenzyme Athioesters____10
TABLE 2: Optimization of amount of ethyl chloroformate ________________________________________ 17
TABLE 3: Conserved sequence motif in Taxol pathway acyltransferases ................. 29

FIGURE 1: Activation of carboxylic acids for the preparation of esters and

thioesters __________________________________________________________________________________________________________________________ 2
FIGURE 2: Synthesis of coenzyme A thioesters _________________________________________________________ 3
FIGURE 3: Commercially available coenzyme A’s ____________________________________________________ 3
FIGURE 4: Synthetic route developed by Walt et. a1 ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4
FIGURE 5: Synthesis of propionyl CoA by Strohl et. al _____________________________________________ 5
FIGURE 6: Synthesis of p-coumaryl CoA by Noel et. a1 ____________________________________________ 5
FIGURE 7: ,B-Phenylalanoyl- and phenylisoserinoyl-COA _________________________________________ 6
FIGURE 8: Target coenzyme A thioesters ___________________________________________________________________ 8
FIGURE 9: Protocol for synthesis of fl-phenylalanoyl CoA _______________________________________ 9
FIGURE 10: ESI-MS of ,B-phenylalanoyl CoA ____________________________________________________________ 11
FIGURE 11: Synthesis 3-hydroxy-3-phenylpropanoy1 CoA ______________________________________ 12
FIGURE 12: ESI-MS spectrum showing 3-hydroxy-3-phenylpropanoyl CoA (m/z
914.20) with impurity (838.27) ...................................................................................... 12
FIGURE 13: Structure elucidation of the by-product generating molecular ion m/z
839 ___________________________________________________________________________________________________________________________________ 13
FIGURE 14: Two possible ways of forming the by—product generating molecular ion
m/z 839 ____________________________________________________________________________________________________________________________ 13
FIGURE 15: Origin of differential reactivity of carbonate carbonyl and acyl carbonyl
__________________________________________________________________________________________________________________________________________ 14
FIGURE 16: ESI-MS data showing unreacted CoASH as a major product _______________ 15

VI

FIGURE 17: ESI-MS data showing the desired thioester (m/z 914.08) as major
product _____________________________________________________________________________________________________________________________ 16

FIGURE 18: Absence of the byproduct in synthesis of ,B-phenylalanoyl CoA __________ 18

FIGURE 19: Origin of differential reactivity of acyl carbon in fl-hydroxy- and N-

Boc-fl-aminophenylpropanoyl anhydride ______________________________________________________________________ 19
FIGURE 20: Synthesis of styryl-fl-alanoyl CoA __________________________________________________________ 20
FIGURE 21: ESI-MS spectrum of styryl-fl-alanoyl CoA ____________________________________________ 20
FIGURE 22: Synthesis of phenylisoserinoyl CoA ______________________________________________________ 21
FIGURE 23: ESI-MS spectrum of phenylisoserinoyl CoA _________________________________________ 22
FIGURE 24: Future synthesis of phenylisoserinoyl CoA ____________________________________________ 22
FIGURE 25: Synthesis 3-pyridyl-3-aminopropionic acid ___________________________________________ 23
FIGURE 26: Synthesis of methyl N-Boc-3-furyl-3-aminopropionic acid __________________ 23
FIGURE 27: Synthesis using ethyl benzoylacetate as starting material _____________________ 24
FIGURE 28: Structure of Taxol _____________________________________________________________________________________ 25
FIGURE 29: Semisynthesis of Taxol ____________________________________________________________________________ 27
FIGURE 30: Biosynthetic conversion of baccatin III to Taxol ___________________________________ 28

FIGURE 31: Schematic of proposed mechanism of active site motif C....HXXXD
during the acyl group transfer from CoA to the hydroxytaxane; complementary role
of the amino phenylpropanoid acyl group of the CoA thioester in the catalytic triad

FIGURE 32: Structure of vinon'ne synthase _________________________________________________________________ 31

FIGURE 33: Proposed complementary role of the amino phenylpropanoid acyl
group of the CoA thioester in the catalytsis by TAXO7 during the acyl group transfer
from CoA to the hydroxytaxane; Cys and Asp are not necessary for this type of

catalysis ___________________________________________________________________________________________________________________________ 32
FIGURE 34: Future plan to validate the SAC mechanism of TAXO7 ________________________ 32
FIGURE 35: Structure of 13-3H-baccatin III 33

VII

FIGURE 36: Synthesis of 7-TES-baccatin 1H from commercially available lO-DAB

oooooooooooooooooooooo

oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

FIGURE 37: Synthesis of 13-3H-baccatin III using sodium borotritide in THF ......... 35

FIGURE 38: Second generation Taxol derivatives _______________________________________________________ 36

FIGURE 39: Mukaiyama’s dehydration-condensation protocol between 7-TES-
baccatin III and protected N-benzoylphenylisoserines _________________________________________________

FIGURE 40:

FIGURE 41:

FIGURE 42:

FIGURE 43:

FIGURE 44:

FIGURE 45:

FIGURE 46:

FIGURE 47:

FIGURE 48:

FIGURE 49:

FIGURE 50:

Taxol biosynthetic pathway .....................................................................
Synthesis of 7-TES-baccatin III ______________________________________________________________
Structure of N-Boc-7-TES-2’-deoxytaxol ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
Synthesis of N-debenzoyl-2'-deoxytaxol ................................................
Synthesis of 2'-deoxytaxol .......................................................................
Structure of (3RS)-,6-phenylalanoyl CoA ...............................................
Structure of (R)-3-hydroxyl-3-phenylalanoyl COA ................................
Structure of (3R)-styryl-B-alanoyl CoA ..................................................
Structure of (2R, 3S)-3-phenylisoserinoyl CoA .....................................
Structure of 2'-deoxytaxol ........................................................................

Structure of 13-3H-baccatin III

VIII

37

38

38

39

39

40

44

46

47

50

51

54

THF

DCM

GC/MS

br

Boc

ESI-MS

CoA

TFA

NMR

TES

DMAP

DPTC

LIST OF ABBREVIATION

Tetrahydro furan
Dichloromethane
Gas Chromatography - Mass Spectrometry
Broad
Singlet
Doublet
Triplet
Quartet

tertiaryButyloxy Carbonyl

Electrospray Ionization Mass Spectrometry

Coenzyme A

Trifluoroacetic acid
Nuclear Magnetic Resonance Spectrum
Triethylsilyl
N, N-Dimethylaminopyridine

Dipyridyldithiocarbonate

IX

CHAPTER 1

SYNTHESIS OF ARYL PROPANOID COENZYME A THIOESTERS

1. 1 INTRODUCTION:
Natural products with interesting structural or biological function within the host
organism or that possess unique bioactivity against disease such as antimalarialsl'2 or

3'5 usually are initially isolated and structurally characterized by a variety of

anticancer,
analytical methods. Advances in molecular genetic techniques (e.g., EST library
construction)6'8 have facilitated the isolation of biosynthetic genes on the pathway of
corresponding natural products. Mutagenic methods to alter a single gene on the pathway
coupled with a metabolomic survey of the resulting pathway intermediates, particularly
after feeding labeled precursors to the host organisms, can often unequivocally establish
the function of the encoded enzyme.9 However, in systems where in vivo feeding of
substrates to auxotroph, or fully-functional organisms, is precluded by impermeability to
cell membranes or by complications in the culturability of the host organisms,lo
individual genes are alternatively expressed in a heterologous host, and the function of
the corresponding enzyme is validated by in vitro assay with rationally selected
substrate(s).

Accordingly, in vitro assay procedures have been used to determine the function

of coenzyme A (CoA)—dependent acyltransferases that impart diverse biological

processes throughout all kingdoms of organisms.11 Acyl CoA thioesters serve as acyl

group donors along the metabolic pathways of amino acids,”’13 lipids,'4'l7 secondary

”"19 and carbohydrates”. The acyltransferases on the corresponding pathways

metabolites,
utilize the relatively labile thioester bond to convey acyl groups to an acceptor molecule.
Examples are demonstrated in the formation of carbon-carbon bonds during iterative
Claisen condensation steps in fatty acid21 and polyketide biosyntheses,22 of carbon-
oxygen bonds during the acylation of hydroxyl groups in biosynthesis of vinorine23 and
paclitaxel,8 or of carbon-nitrogen bonds in amino group acylation of penicillin G24 and
N-acetylsphingosine25 biosyntheses.

Coenzyme A contains a reactive nucleophilic thiol functional group that interacts
with an activated carboxylic acid to produce acyl coenzyme A thioesters. One of the most

common synthetic methods to activate carboxylic acid involves treatment of the acid with

an alkyl/aryl chloroformate under basic condition to form a mixed anhydride (Figure 1)

 

° ° it JOL
. Et N
RJLO’H + Cl/Iko’R 3 4' R O O-R'
CHZCI2,rt
alky/aryl chloroformate mixed anhydride

Figure 1: Activation of carboxylic acids for the production of esters and thioesters
that will make the carbonyl group of the carboxylic acid more reactive towards an
incoming nucleophile, thiol function in this case leading to the formation of coenzyme A

thioesters (Figure 2).

 

NH2 NH2
N’ N
/ N
NK I \ o o k‘ I \
H30, CH3 9 O \N N H539. CHO—B-O-Ig-ON N
HO“. O-P-O-P-O 0 basic condition I I O
OH OH T o OH OH
O NH )0L 0 NH o o OH
JOK/I 0‘, ,o OH R 0’” R 5 NM *‘p;
N ,P\ . \lr Ho’ OH
H HO OH Activated form 0 H
Coenzyme A Coenzyme A Thioesters

Figure 2: Synthesis of coenzyme A thioesters

Commercially available CoA thioesters for biochemical studies are largely limited
to alkanoyl/alkenoyl/fatty acid-substitution, with the exception of benzoyl-CoA (Figure
3). Additionally the high cost of commercially available coenzyme As is due in large part
to production cost from natural resources.26 Consequently, only a limited selection of the

acyl thioesters is available.

0

\
MSCOA )LSCOA SCOA SCOA
Acetoacetyl CoA Acetyl CoA Benzoyl CoA 2-Butenoyl CoA Butyryl CoA

$1 8/mg $25/mg $28/mg Sac/mg $17/mg

 

O CoAS

O
\HJL SCoA o Arachidonoyl CoA
,, SCoA Ho W0 sza/mg

n = 13. Arachidoyl CoA ($19/mg) “mammary”, CoA N30

n = 16. Stearoyl CoA ($15/mg) $13/mg SCoA O O

= 15. n-heptadecanoyl CoA ($15/m9) 3-Hydroxy-3-methylglutaryl CoA M
n = 14, Palmitoyl CoA ($16/mg) 521/"10 HO n SCoA
n = 12. Myristoyl CoA ($16/mg) 0
n = 10. Laur0y|C0A($16/m9) SC°A n = 1. Malonyl CoA($11/mg)
n = 5. Octanoyl COA (515/1719) O O n = 2. Succinyl CoA ($20/mg)
n = 8. Decanoyl CoA ($16/m9) n = 3. Glutaryl CoA($15/mg)

n = 4, Hexanoyl CoA ($20/mg) isobutyryl CoA M LioMscoA
SCoA

n = 1. n-Propinoyl CoA (18/m9) $22/mg
Methylmalonyl CoA O
isovaleryl CoA $47lmg )4
o $13/mg SCoA
O 2-Methylcrotonyl CoA
scOA SCoA $14/mg
\
— n _
Linoleoyl CoA

n = 1.Palmitoleoyl CoA (SH/mg)
n = 3. Oleoyl CoA ($18/mg)

$20/mg
Figure 3: Commercially available coenzyme A (from Sigma Catalog, 2006)

The present acyl coenzyme A library lacks acyl groups that possess reactive functional
groups, and acyl-CoAs of this type remain a challenge to the synthetic community. The
synthesis and purification of various acyl CoAs in microgram scale was reported earlier
using CoA ligases27 but the substrate specificity of these enzymes restricts access to a
broad series of products. Recent methods incorporating catalytic RNA in the synthesis of
CoA thioesters28 emerge as an interesting approach, although the limited substrate

specificity of these RNA-based catalysts remains a major drawback toward practical

application of these methods. Synthetic procedures for acyl-CoA have also been

reported,”33 but most processes outline the activation of an alicyclic acyl module that is

void of functional groups with competing reactivity.

In 1991, Walt and his coworkers demonstrated that various acyl-CoAs can be

synthesized using S-acylthiocholine iodide as an acylating agent,34 but the CoAs they

synthesized are all commercially available and the acyl groups are mainly

unfunctionalized hydrocarbons (Figure 4).

HSH2CHZCN:

CH3
CH3

 

 

1.(RCO)2O, reflux, 10 min ii (EH3
cooled overnight A R SHZCHZC—N—CH3 I—
2. CH3I, Acetone, rt, 4 h I+
CH3
S-acylthiocholine Iodide
CoASH
in 0.1M phosphate buffer
pH = 7.1, rt, 3h
i
3

Figure 4: Synthetic route developed by Walt et al.

In 1995, Strohl and his coworkers reported a synthetic route35 using 1,l'-

carbodiimidazole as an activating agent (Figure 5) for synthesizing acetyl CoA and

propionyl CoA, which are relatively simple in complexity and, moreover, are

commercially available.

COOH 0 CW

 

 

 

\j: 65 °C, 30 min _ O
Q'Na CI CI THF, underAratm Cl Cl
0
@N—(
0V 0)
O N
\/U\ CoAS-H N THF, 65 °C, 90 min
SW = I ,> .__+
in aq. imidazole buffer N

pH = 7.0, rt, 15 min
Figure 5: Synthesis of propionyl CoA by Strohl et al.

The first example of the synthesis of a substituted phenylpropanoyl CoA came
from Noel and his coworkers in 2000.36 They have described a two-step synthesis of
coumaryl—COA involving the generation of an N-hydroxysuccinimide ester of 4-coumaric
acid followed by a thioester exchange with CoA (Figure 6). Regrettably, experimental

details, such as yield and spectroscopic characterizations are missing in this report.

0 0 0
Wm + QN-OH Wee“
HO O CoAS—H HO

p-Coumaryl CoA

 

 

Figure 6: Synthesis of p-coumaryl CoA by Noel et al.

The aforementioned syntheses describe the production of target CoA thioesters
used to investigate aspects of primary metabolism in various organisms. However, in
2002, the first synthetic procedure for making milligram quantities of amino- and
aminohydroxy phenylpropanoyl CoA was reported.37 In order to assess the biosynthetic
origin of the phenylpropanoid side chain of paclitaxel, several amino acid-CoA thioesters

were employed in a previous investigation by Walker and his coworkers.37 ,6-

Phenylalanoyl- and phenylisoserinoyl-CoA (Figure 7) were synthesized and identified as

productive substrates of the expressed taxane 13 ,8-O-phenylpropanoyltransferase.3 7

NH2

St"
KN N>
O

N
H3C,’ CH3 9 9

HO,,_ " o—B—o—B—o
R O OH OH
:2 UH o, ,o OH
S\/\N ’P\
i H HO OH

R10

 

R1 = NH2; R2 = H; ,B-Phenylalanoyl CoA
R1 = NH2; R2 = OH; Phenyisoseryl CoA

 

 

Figure 7: fi-phenylalanoyl- and phenylisoserinoyl-CoA

The apparent void in this area of research encouraged efforts to develop a direct
process for synthesizing quantitative amounts of novel acyl CoA thioesters. With an
emerging focus on biocatalysis, it becomes imperative to have access to a variety of
surrogate substrates in order to survey the substrate specificity of a particular enzyme.
Therefore to probe the specificity of the aminophenylpropanoyl transfer reaction in the
Taxol® biosynthetic pathway an expanded library of CoA thioesters bearing a variety
functional groups on the acyl moiety are explored and are described herein. The primary
goal was to optimize the reaction conditions of previously developed methods to make
the process more practical, economical, and general for the synthesis of a series of
unnatural CoA thioesters.

Herein is described a general procedure for synthesizing milligram quantities of
various aryl substituted propanoid coenzyme A thioesters. Also described is a modified

method to remove butoxycarbonyl from reactive amines by mild deprotection with

formic acid, which replaces the less selective trifluoroacetic acid-based deprotection that
partially degrades the CoA thioester.37 In general, the described methods, demonstrate the
incorporation of a mixed anhydride intermediate of a reactive carboxylic acid to convert
variously substituted aryl propanoic acids to their corresponding CoA thioester for
investigations on a phenylpropanoyltransferase enzyme on the paclitaxel biosynthetic
pathway. Several of the thioesters produced in this study can also be applied toward
investigation on phenylpropanoid metabolism that includes lignin and lignan

biosyntheses.”40

1. 2 RESULTS AND DISCUSSION:
The target coenzyme A thioesters were designed in such a way that can produce a

library of CoA thioesters of various functionalities with competing reactivity in them

(Figure 8).
NH2 0 OH O NH2 0
\
SCoA QMSCOA SCoA
NH2 0 o O NH2 0
SCoA GUSCOA \ SCoA
OH
NH o
2 NH2 0 NH2 0
\ SCoA O SCoA
I
N / \ / SCoA
/

Figure 8: Target coenzyme A thioesters

In order to investigate the substrate specificity of the 3-amino-3-
phenylpropanoyltransferase, the amounts of natural and surrogate CoA thioester
substrates derived synthetically need to consistently fall in milligram scale. These
relatively large gravimetric quantities are required to compare relative kinetic constants
of the enzyme for each substrate. Furthermore, this also ensures that large-scale
biosynthetic assays can be conducted to produce enough product for characterization by
sample-limited techniques, such as 1H- and l3C-NMR.

The synthesis of ,B-phenylalanoyl- and phenylisoserinoyl CoAs in a previous
study were not optimized for maximal yield.37 Instead, the focus of this earlier
investigation was framed in the context of characterizing the function of enzymes in a

library of acyltransferases expressed from genes derived from T axus plants. The assays

were developed to incorporate natural substrates, which are likely very catalytically
efficient. To assess the efficiency of surrogate substrates significant substrate may be
required to achieve apparent saturation of the enzyme for maximal turnover.

Following the literature procedure that describes the synthesis of fl-phenylalanoyl
CoA a trial synthesis was conducted to gain insight of the steps and processes of the
procedure that could compromise the yield. Various aspects of the synthesis were
modified to increase the yield from ~30%37 to ~66%. The commercially available ,8-

phenylalanine was first converted to its corresponding N—Boc protected amino methyl

 

 

 

 

ester (Figure 9).
NH2 0 BOC‘NH O BOC‘NH O
: socuz, MeOH : :
OH 70 °C, 2h _ OMe 1N NaOH. THF_ OH
then T Overnight, rt 7 7 6 o/
. . (800)20 70% then acidified °
(Rigjgg‘xagfiignw Py, Et3N, THF (in two steps) to pH 2.
CHCI3. Overnight, rt
Boc 806‘ Bocc
‘ NH O O
NH O ? A CoAS-H in NH O
OH CICOZEI. EI3N __7_ 0 CE 0.4 M aq. NaHCOa; SCoA
THF, rt, 1h,N2 atm 0.5 h, rt,
. Purified by C-18
"OI isolated silica column
solvent evaporated

and then dissoved in t-BuOH

B°°‘NH o NH2 0

I HCOOderopwise);
Wscm H20, 0 .0 EDA/kscm

66%
,B-Phenylalanoyl CoA

Figure 9: Protocol for synthesis of ,B-phenylalanoyl CoA

 

Then CoA as trilithiurn salt dissolved in a 0.4 N aqueous solution of sodium bicarbonate
was added into the reaction. Stirring the reaction for half hour and then quenching with
hydrochloric acid produced the N-Boc protected desired thioester. The N—Boc protected
acyl CoA was isolated by evaporating the solvent under reduced pressure and then

purified by anion exchange column using C-18 silica gel chromatography. To deprotect

the N-Boc group 88% aqueous formic acid was employed in place of trifluoroacetic acid
that was used in previous work.37 By virtue of formic acid being a weaker acid, the
deprotection step was slower than when more reactive trifluoroacetic acid used in the
previous method, however, degradation of the product was minimized upon deprotection
with the milder reagent.

Thus far described have been procedures to optimize the first two steps and the
last steps of the CoA thioester synthesis described in the previous investigation.37 Yet, the
formation of the mixed anhydride and the ensuing coupling reaction with CoA still
needed to be optimized. Increasing the CoA concentration from one to two equivalents
did not significantly improve the yield, which remained approximately at 61%. This
overall yield was sensitive to the reaction fine of the first step where the carboxylate

substrate is converted to the anhydride (Table 1).

Bocc .
BochH 0 NH 0 0 8°C NH 0

; ? A CoAS-H in F
OH CICOJEI. Eth 9, o OEt 0.4 M aq. NaHCQg> SCoA
THF, rt, 1h,N2 atm 0.5 h, it.
not isolated

solvent evaporated
and then dissoved in t-BuOH

 

 

 

Entry time for 1st step (min) time for 2nd step (min) % Yield

 

1 60 30 61
2 30 30 38
3 30 60 40
4 90 30 60
5 90 60 61

 

Table 1: Optimization of reaction time for synthesis of coenzyme A thioesters
From the results, it is apparent that the reaction time for anhydride formation is an

important factor to obtain a better yield. Since increasing the time for the trans-

10

thioesterification reaction did not significantly change the yield, it can be assumed that
the first step is rate-limiting, and is therefore guiding the overall product formation, i.e.,
the yield of the mixed anhydride is directly proportional to observed mass 913.27 in this
sequential reaction. The identity of the coenzyme thioester was verified by 1H-NMR and

confirmed with ESl-MS analysis (negative ion mode) (Figure 10).

100 913.27

Relative Abundance
8

30 935.27

10
455.40 1006.07
5 2143.13 “£11487.” 539,30 fl 833-5‘11WHV. 1106.00 _
[ I ' I I I ' ' ' I ' V ' I r " V ' I ' I ' V " I ' "
300 400 500 600 700 000 900 1000 1100 1200
mix

 

 

 

Figure 10: ESI-MS of fi-phenylalanoyl CoA
The next coenzyme thioester target had a hydroxy group instead of amino group
at the C3 of the ,B-phenylpropanoyl CoA. The general procedure described previously
was performed except that the last deprotection step was unnecessary since no Boc-
protection was needed for the starting material. The hydroxy group on the
phenylpropanoid sidechain was anticipated to not be as nucleophilic as the CoA thiol in
the polar protic solvent (t-BuOl—l) toward the anhydride. One equivalent of ethyl

chloroformate was used to convert the 3-hydroxylphenylpropanoic acid to the

corresponding CoA thioester outlined in Figure 11. Notably, while the thioesterification
step remained the same as in the synthesis of other CoA thioesters, the solvent system of
the initial step was modified from THF to a mixture of dichloromethane and THF to

solvate the starting material.
OH o OH o o OH o

-_: ? A CoAS-H in ?
OH ClcogEt. EtaN > Q 03 0.4 M aq. NaHCO3: SCoA
CHZCIz-THF (5:2, v/v). 0.5 h, rt.

rt, 1h,N2 atm

Figure 11: Synthesis 3-hydroxy-3-phenylpropanoyl CoA

 

 

However, ESl-MS analysis revealed the presence of an analyte (at m/z 838.27)
with an ion abundance similar to that of the target thioester product (at m/z 914.20)
(Figure 12).

33“" 914.20

 

Relative Abundance
assesses

11 11111111

 

   

 

 

002 3 980.13
1008.07

'T350'I'1400 " '700 900 1000 1100 1200

 

 

uni-L
I l I
.5.
d
G
co
8:; a
.4
N
d
I»

Figure 12: ESI-MS showing desired 3-hydroxy-3-phenylpropanoyl CoA (m/z 914.20)
and an impurity (m/z 838.27)

Further analysis suggested that the compound is also a CoA thioester and was

putatively identified as a coupling product of ethyl chloroformate and the CoA salt

(Figure 13).

12

NH2

5’ N
0 Cl \ / \
\/
U21? HOH3C CH3 9 p N N)
j(4/O—P-O-P-O
I I O
OH OH
NH
O“ ,O OH
OOS\/\N/u\) /P\
H HO OH

Figure 13: Structure elucidation of the by-product generating molecular ion m/z 839

There are two possibilities for how this compound was generated. Either the
CoA-SH can attack the carbonate carbonyl instead of the acyl-carbonyl of the
intermediate anhydride or it can displace chloride from unreacted ethyl chloroformate to

form a thiocarbonate ethyl ester as shown in Figure 14.

 

 

 

 

 

 

 

 

 

 

"route A”
OH O OH O
? CoAS-H' 1n 7
OH ClCOzEt. Et3N o’lL OEt 0.4 M aq. NaHC03 SCoA
CHZCIZ-THF (5:2. v/v) 0.5 h.rt
rt. 1h.N2 atm Purified by C-18
Case 1 silica column
ABCoAStl-l i
"route B"
r /\o SCoA
Byproduct
9H 0 CoAS H in 9H 0
OH CICOzEt. EtaN oJL OEt 0 4 M aq NaHCOa SCoA
CHzclz-THF (5:2. vlv) 0. 5 h. rt
rt. 1h,N2 atm
0 CoAS-H In )1
0.4 M aq. NaHCO;,=
/\o’U\CI ,,5..,.., ”o sooA
Byproduct

Figure 14: Two possible ways of forming the by-product with molecular ion m/z 839
It is worth noting that in principle the acyl-carbonyl should be a more reactive
electrophilic center than the carbonate carbonyl. Contributing resonance structures

demonstrate that the lone pairs of electrons on the oxygen atoms flanking the carbonate

l3

carbonyl contribute electron density to this reactive carbon center, thus minimizing the
electrophilicity at this site (Figure 15). In contrast, the acyl carbonyl has only one oxygen
atom that donates electrons to reactive carbonyl carbon, and therefore the acyl carbonyl is

considered less deactivated than the carbonate carbonyl.

o “(o resonance

R\/II\-d ’>O-/\ 93‘“ Rig /\..___..R\/U\+//I\

acyl carbonyl—IQ I—— carbonate carbonyl

 

resonance
path B

O O

R \O/ILOA

Figure 15: Origin of differential reactivity of carbonate carbonyl and acyl carbonyl
If indeed CoASH is not regioselective for attack on a carbonyl in the asymmetric
anhydride, several options were considered to bias nucleophilic attack toward the acyl
carbonyl e. g. use of t-butyl chloroformate with a bulky alkyl group as the acid-activating
agent and t-butyl would likely diminish the CoASH attack towards the carbonate
carbonyl due to the steric hindrance of the t-butyl group.

Alternatively, less ethyl chloroformate (0.5 equivalent) was employed to
minimize the potential reaction of CoASH with excess chlorocarbonate reagent. ESI-MS
analysis of the product from the modified coupling reaction revealed that a major portion
of CoASH was unreacted ([M-H], m/z 766.27) and that the desired thioester ([M-H], m/z

914.20) was in relatively lower abundance (Figure 16).

14

766.27

75 _ 831.33

Relative Abundanc
8

30 — 475.13 84227
25
20

10 237.13 283.07 385.13 51020 .

5 I , l

0200 A 300 400 i 500 600 700 800 900 1000 1100 1200
ml:

 

 

Figure 16: ESI-MS showing unreacted CoA as a major component at m/z 766.27

A third option was attempted prior to converting the acid to a formate ester
bearing a bulky t—butyl group. Instead of carrying the mixed anhydride to the next step in
the reaction series, after only removing the methylene chloride/T HF solvents, the product
residue was dried under high vacuum for 24 h. This simple variation was considered to
potentially remove excess unreacted chloroformate(b. p. 95 °C), which is relatively
volatile. The resulting residue was isolated as a white solid compared to previous cases
where the residue containing the anhydride was obtained as a pale-white paste. Addition
of the CoASH under standard reaction conditions to the anhydride (as a white solid) in
the second step of reaction yielded the desired product as a single product, which was

verified by the abundant ion ([M—H], m/z 914.08) in the ESl-MS spectrum (Figure 17).

OHO

OH o OH O o ,
: : k CoAS-H In ?
OH CICOZEI, EISN > o 0E1 0.4 M aq. NaHCOJ,> SCoA
| Cchlg-THF (5:2, v/v). 0.5 h. rt,
Purified by C-18

rt, 1h,N2 atm
silica column
chromatography

/

 

 

\

dried under vacuum overnight

15

100' 914.08

 

 

 

 

 

 

90
801
C)
8
gm
3
560,
§
2
£54
455.06“:48
40‘ 566.47
573.51 876.01
30‘ 1025.93 11513512116111
429 56 13.02 753.10 103155
. 1304.64
20, 210.89 30454 I 32023 + .
................ 1..-,I.!..I,..J.,,I..I,,L,;,,,- , . l '. 1‘ I. 'I I I ,
100 200 300 400 500 600 700 000 900 1000 1100 1200 1300 1400

ml:
Figure 17: ESI-MS showing the desired thioester (m/z 914.08) as major product

Therefore, it is concluded that the production of the ethyl forrnate thioester of

CoA by-product in previous attempts at this reaction sequence with 3—hydroxy-3—

phenylpropanoic acid resulted from the reaction between unreacted ethyl chloroformate

and CoASH (Figure 14). Drying the reaction overnight under vacuum removed all of the

unreacted ethyl chloroformate; thus the CoASH nucleophile attacked only the activated
acid and not the competing chloroformate reagent.

Now that a convenient method for purifying the mixed anhydride was developed,

optimization of the ethyl chloroformate used in the reaction was pursued. Ethyl

chloroformate at 1.5 and 2.0 equivalents gave maximal yield at 68% (Table 2) indicating

that 1.5 equivalent of the activating agent is ideal for these reactions. These modified

steps were applied to synthesis of other phenylpropanoyl CoA thioesters.

16

X 0 X 0 O X 0

r ? /u\ CoAs-H in ?
OH Clcoggt, Et3N : o 051 0.4 M aq. NaHCO3: SCoA
CHZCIZ-THF (5:2. v/v). 0.5 h. rt,
Purified by C-18

TI. 1h.N2 atm . .
X = OH. NHBoc dried under vacuum overnight S'I'ca column

 

 

 

 

Chromatography

X Entry Equiv. of Chloroformate °/o Yield
OH 1 1.0 31
OH 2 1 .5 68
OH 3 2.0 68
NHBoc 1 1 .0 61
NHBoc 2 1.5 66

 

Table 2: Optimization of amount of ethyl chloroformate

The synthesis of fl-phenylalanoyl CoA, described in an earlier report3 I, was
reinvestigated, wherein the amount of ethyl chloroformate was optimized and the mixed
anhydride was dried overnight under vacuum, as described above to remove unreacted
chloroformate prior to conducting the next reaction step. However, the addition of excess
ethyl chloroformate (1.5 equivalents) showed no significant effect on the yield (Table 2).
Thus treatment was followed by the general procedure for CoASH coupling, and ESI-MS
analysis of the ensuing target product confirmed that the by—product ([M-H], m/z 838.27)
(cf. Figure 12), derived by the coupling of ethyl chloroformate with CoASH, was absent;
a single abundant peak corresponding to N-Boc-fl-phenylalanoyl CoA was observed ([M-

H], m/z 1013.40) (Figure 18).

17

1013.40

Relative Abundance
8

30' 1035.27

25
20 93347 1085.27

 

 

15
1°: 408.40 505-“ 586.33 666.4 096.47 J 110720
5 l I. r L L LA I 1 0. IL I l H
200 300 400 500 600 700 800 900 1000 1100 1200
ml:

 

Figure 18: Absence of the byproduct in synthesis of ,B-phenylalanoyl CoA

lntriguingly, even when excess chloroformate was not removed under vacuum
prior to initiating the final thioesterification step, the CoA nucleophile preferentially
attacked the B—phenylalanoyl mixed anhydride species. Relatively, none of the ethyl
carbonate thioester at [M—H], m/z 838.27 (cf. Figure 12) was observed by ESI—MS
analysis'(Figure 18). This observation is in stark contrast to that for the similar reaction
with fihydroxyphenylpropanoic acid where a significant amount of carbonate thioester
was produced. Conceivably, when the free OH-group of the flhydroxyphenylpropanoid
is not protected a non—bonded interaction (like H—bonding) might low the activity of the

acid towards the mixed anhydride formation (Figure 19).

H-bonding with
carbonyl oygen

H . 5 + I H \ _
| resonance Q (0 O
’DVIL /\ Acontributor 'bias'_ I \5JLO/\

 

  

electron density
decreased
less H-bonding interaction with carbonyl oygen
due to lower pKa of N-H relative to OH

Bot ,H.
C N (06+O

 

electron density not removed from reaction
center as much as in case of O-H group

Figure 19: Origin of differential reactivity of acyl carbonyl in fi-hydroxy- and N-
Boc-fi-aminophenylpropanoyl anhydride

Notably, the O-H bond is more labile than an N-H bond and, therefore, the O—H
group is more apt to engage in hydrogen bonding. Consequently, an intramolecular H-
bonding interaction with the carbonyl oxygen should increase the electrophilic character
of the carbonyl carbon and thus should accelerate the reaction rate. Ironically, this is not
Observed. The reason for this differential reactivity of the CoA nucleophile is unclear,
and experiments are currently being investigated to explore the phenomenon.

The [istyrylalanoyl CoA was made from fistyrylalanine that was produced
during the preparation of the product standard in the phenylaminomutase (PAM)
project.41 Since the [iamino acid was not commercially available, the first challenge was
to synthesize this amino acid derivative via modification of a literature procedure.42 (S)—3—
Amino-S-phenylpentanoic acid was converted first to its N—Boc methyl ester derivative.
Photochemical benzylic bromination of the N—Boc protected amino ester was proceeded

with N—bromosuccinimide treatment followed by a dehydrobromination to produce the N-

19

Boc styryl-flualanine methyl ester. Saponification of the ester and thioesterification of the
carboxylate with CoA followed the general procedure described earlier. Final N-Boc
deprotection with formic acid yielded styryl-flalanoyl CoA (Figure 20). The identity of
the desired CoA thioester was confirmed by ESI-MS analysis ([M-Hl, m/z 939.33)

(Figure 21).

 

 

B .
NH2 0 NH2 0 °C NH O
OH SOCIZ, MeOH OMe (Boc)2O OMe
70 °C, 2h Py. Et3N. THF
CHCI . O ht, rt 76%
(S)-3-amino-5-phenylpentanorc acid 3 vernIg in two steps

(purchased lrom Peptech)

1. N88. CCl4. 3h, 5-10°
with a 60wt tungsten lamp

8 4 B t
°° NH 0 0° NH O _
- placed ~2 cm away from the : -
OMe reactIon flask during reaction 2 \ OMe 1. 2N NaOH, THF, rt, 12 hé \ SCoA
2. DBU/ 70°C 2.ClCO;Et_ THF, 1 h then, 0
40% CoASH in 0.4 M Aq NaHCOa 60
3. HCOOH. rt

 

 

Figure 20: Synthesis of styryl-fl-alanoyl CoA

838.33

939.33

113.00

Relative Abundance

25 227.00

1° l 96127

11A-
vIv‘

1000

 

 

 

vv 'vvvv v—[VWV—r vvq

1400

l
Vva vv

 

., .1- “r
1200

Figure 21: ESI-MS spectrum of styryl-fl-alanoyl CoA ([M-H]", m/z 939.33)

20

The procedure for the synthesis phenylisoserinoyl COA is reported in the
literature.37 However, repeating this synthetic process revealed that trifluoroacetic acid-
mediated deprotection was harsh and difficult to control, resulting in compromised

yields. Formic acid, a milder deprotecting agent was used instead of TFA to remove the

 

 

Boc-group (Figure 22).
CH2N2- THF (Boc) )20 NaH
90% THF, rt, 20 min 0‘8
90%

1. CICOQEt THF
OMe 2N NaOH, THF 0- Na+ then CoASH SCoA
rt, 12 h 2. HCOOH, rt

30%

 

 

Figure 22: Synthesis of phenylisoserinoyl CoA
N -Boc-phenylisoserine was first converted to its methyl ester by diazomethane treatment
then C2 hydroxyl group was protected as O-Boc group with di—Iert-butoxydicarbonate
and sodium hydride treatment. The N- and O- protected intermediate was saponified and
converted to the thioester following the procedure described earlier. Final deprotection
with formic acid produced the desired CoA thioester, as verified by ESI-MS analysis

(IM-HI, m/z 928.8) (Figure 23).

21

100'
90
80

70 792.8
475-9 776.9

498.9
50 520.9

341.0 656.9

640.9
461.0 912.8

Relative Abundance
8

928 .8
30
363.
9 1048.7

I 760.9 1064.7
903.0

227.9 ,
624.9
10 895-8 1064.6

 

8 9.1

 

 

 

 

 

 

 

 

 

 

 

 

. ,..,. , .11.,..,... . .1” ., !.,.-l.,.., .14, ,.!, .1 , .-,
0 250 300 350 $00 450 500 550 000 650 700 750 800 850 N0 950 10001050
mtz

 

Figure 23: ESI-MS spectrum of phenylisoserinoyl CoA ([M-H]", m/z 928.8)
Encouraged by the synthesis of 3-hydroxy-3-phenylpropanoyl CoA without
hydroxy group protection has prompted a reinvestigation of the synthesis of

phenylisoserinoyl CoA without O-Boc protection (Figure 24).

 

1. CICOZEt, THF,Et3N
OH then CoASH

2. TFA, H20, 0 °c

   

 

Figure 24: Future synthesis of phenylisoserinoyl CoA thioester
Currently, experiments are ongoing in order to synthesize a series of coenzyme A
thioesters in which the acyl moiety possess a variety of structural features and functional
groups that have competing nucleophilic reactivities. The main challenge is gaining
access to the corresponding flamino acids. For example, 3-pyridyl-3-aminopropionic

acid is not available commercially, and, consequently, the synthesis of this pyridyl amino

22

acid was attempted by following a literature procedure.43 The first step of the reaction
involved the formation of an enamine with excess ammonium acetate dissolved in
ethanol under reflux. The water, by-product of this reaction, hydrolyzed the enamine
back to the starting material, ultimately, even when attempts were made to remove the
water with Dean-Stark trap and by adsorption with 4A molecular sieves. The challenge
was overcome by using tetraethoxy orthosilicate that serves as a very powerful
dehydrating agent during the reaction (Figure 25). The product will be converted to its

corresponding CoA thioester.

 

 

 

Si(OEt)4, reflux, 6h
1 \ QB 79% > | \ OEt I \ SCoA
N / 2142031027 EtOH N / 81993 N /
30%

 

Figure 25: Synthesis of 3-pyridyl-3-aminopropionic acid
3-Furyl-3-aminopropionic acid was purchased and was converted to its N-Boc

methyl ester derivative (Figure 26).

 

 

 

Boc(
””2 O SOCI2, MeOH NH O NH2 0
o OH 70 °C, 12 h = 0 OMe O SCoA
\ I then(Boc)zO \ I steps \ I
Py,Et3N,THF

 

CHCI3, Overnight, rt
Figure 26: Synthesis of methyl N-Boc—3-furyl-3-aminopropionic acid
Ethyl benzoylacetate was used to synthesize benzoylacetic acid following a
literature procedure.44 It was also used to prepare a unique enamino acid (Figure 27)
following a literature procedure.43 All these ,B-amino acids will be converted into

corresponding CoA thioesters in near future.

23

O O

I \ OEt
NH40AC,SI(OEt)4 / J 2N NaOH,THF,rt.12h

EtOH, reflux, 24 h and then acidified to pH = 2
NH2 0 O O
\ OEt OH

 

 

2N NaOH, THF, rt, 12 h
then ClCOzEt, THF

NH2 0 i NH2 0
\
steps

Figure 27: Synthesis using ethyl benzoylacetate as starting material

Isteps

 

 

 

Conclusion:

The routes toward the production of milligram quantities of various aryl
substituted propanoid compounds have been described. Also reported is a modified
method that removes t-butoxycarbonyl from reactive amines by mild deprotection with
formic acid in place of the less selective trifluoroacetic acid that partially degrades the
CoA thioester, used in a previous procedure.37 Presently efforts are ongoing to synthesize
a series of unnatural CoA thioesters possessing different functional groups in the acyl
side chain. Foreseeably, application of these CoA thioesters toward elucidating the

function and mechanism of various secondary metabolite pathway enzymes is feasible.

24

CHAPTER 2

SYNTHESIS OF A TAXOL® ANALOG AND 13-3H-BACCATIN 111 TO
ELUCIDATE THE MECHANISM OF A TAXOL® PATHWAY BAHD

ACYLTRANSFERASE

2. 1 INTRODUCTION:

Taxol® (Figure 28), a taxane diterpenoid (generically known as paclitaxel),
isolated from various yew plant species (T axus), is a potent pharmaceutical agent4546 that

has current and potential application in the treatment of 3 of the top 4 deadliest diseases

7

in the world, including heart disease4 , cancer48 and Alzheimer’s disease”, and thus has

became one of the best selling drugs in history, (netting $3 billion in sales in 2000).45“I6

O

Ph/ILNH O
: 2,
3' , 0"-

Figure 28: Taxol®

A OH

O O
9.1.

C
Ho AcO

OBz

    

Despite approval by FDA in 1992 for clinical trials the poor solubility of Taxol® in blood
plasma affected drug delivery and efficacy. Improvement in infirsion adjuvant brought
Taxol® to the forefront as a chemotherapeutic against ovarian and breast cancers.

Recently, however, improvements in Taxol® delivery in clinical use have been hugely

25

bolstered by the production of AbraxaneTM (Abraxis Oncology, Bridgewater, NJ). This
drug surrogate is derived by a formulation process that suspends the drug in a protein
albumin-based emulsion that has fewer side effects than direct delivery of the drug with
toxic chemical solvents, such as Cremophor® EL.50 Still, the production of Taxol® was
suffering from the isolation of the drug from the bark of the limited natural resource, the
pacific yew.46 That isolation procedure was not only inefficient (yield of Taxol® from the
bark extract was 0.014%)46 but was also a destructive isolation technique that killed the
tree. The projected increase in the use of Taxol® for basic research, cancer chemotherapy,
arterial stent treatment, and potential Alzheimer’s disease application warrants effort to
improve existing production processes. The total synthesis of the drug,51 while a
significant accomplishment, is costly and low yielding, and thus provides an unrealistic

52-5 3 currently the major route to

alternative for commercial supply. Semisynthesis,
Taxol®, involves a limited number of synthetic steps to convert more abundant, advanced
Taxol® pathway intermediates (for example, lO-deacetylbacatin III isolated from various

T axus needles) (Figure 29) to the target drug.

26

 

         

 

OH OTES
Ho--- ' 1.TES-Cl. Py, rt, 17 h, 91% : HO,._
; H s o 2.Ac20, Py. rt, 48 h, 82% . H i o
HO 2 AcO HO 5 AcO
082 082
10-Deacetylbaccatin Ill 7-TES-baccatin III
(isolated & purified
fmmthe Taxus "66d'681 3.0 equiv of NaN(SiMe3)2 71530., ,Ph
3.5 equiv of ,B-IactamA ' '
THF, 0 °C, 0.5 h, \ A
86% O COPh
Ph
OANH o
2 HF-Pyridine =
TTHF, 25 °C, 1.25 h a
80% OTES

   

Figure 29: Semisynthesis of Taxol®- the process begins with the isolation of 10-

deacetylbacatin 111 from the needles of T axus.

However, while the semisynthetic process addresses the supply issue, several
challenges54 remain obvious in terms of technical difficulties like the isolation and
separation of lO-deacetylbacatin 111 from other metabolites and cellular organelles. Also
consistency in crop production remains a concern with regards to the product yield of the
necessary starting material.

Since the production of Taxol® will, for now, depend on biological means of
production, it remains imperative to understand the biosynthetic pathway of Taxol® in
planta. Advances in molecular genetic engineering techniques and biotechnology,
coupled with the recent isolation and characterization of sixteen Taxus—derived
acyltransferases, six of which are functionally characterized, include the C10-
acyltransferase, the C2-hydroxyl and C3'-amino benzoyl transferases, and C13-0-

phenylpropanoyl transferase (Figure 30).54

27

 

Figure 30: Biosynthetic conversion of 10-deacetylbaccatin III to Taxol‘D — the last
few steps involve BAHD superfamily acyltransferases including C10-
acetyltransferase, the C2-hydroxyl and C3'-amino benzoyltransferases, and C13-0-
phenylpropanoyl transferase
The acyltransferases involved in Taxol® biosynthesis all belong to the
Benzyl alcohol acetyl-, Anthocyanine—O-hydroxy-cinnamoyl-, anthranilate-N-Hydroxy-
cinnamoyl/benzoyl-, Deacetylvindoline acetyltransferase (BAHD) enzyme superfamilf5
that have played a principal role in the evolution of secondary metabolism in several
plant species. BAHD acyltransferases are involved in the biosynthesis of diverse natural
plant products including volatile flower scent in Clarkia, capsaicin biosynthesis in
Capsicum, and the biosynthesis of morphine and the anti-cancer drug vincristine.”59
Despite the importance of BAHD acyltransferases and the products they create,
little is known about their catalytic mechanism, subtrate specificity, and evolution.
Several reports have highlighted that sequence similarity does not correlate with enzyme
function. For example, even when two acyltransferases share greater than 90% similarity,
the reaction they catalyze can be completely different.60 These enzymes require an acyl

coenzyme A cosubstrate as the acyl donor, and have a signature conserved sequence

motif (HXXXD) in the putative active site (Table 3).61

28

Regan A B AcylLGroup Transferred

 

Consensus ...C ......... HXXXDG

TAX01 ...c ......... HGVCDG 5-OAcetyl
TAX02 ...C ......... HSVSDG 2-O-Benzoyl
TAX06 ...c ......... HGICDG 10-O—Acetyl
TAX10 ...C ......... HSVCDG 3'-N-Benzoyl
TAX07 ...c ......... GSACDA 13-0(3-Amino-3-

Phenylpropanoyl)

Table 3: Conserved sequence motif in Taxol® pathway acyltransferases

Three residues (C, H, D) in this motif are postulated to constitute a catalytic triad
that is directly involved in acyl group transfer from the CoA donor to an amino,
hydroxyl, or thio group of the acceptor (Figure 31).“

However, TAXO7 (3-amino-3-phenylpropanoyltransferase) contains a natural
Gly—>His and Ala—>Gly substitutions in this motif (Table 3); this latter exchange (A—-)G)
should be relatively benign. Although based on the obligatory function of the histidine
residue in proposed mechanism (Figure 31 A), the Gly substitution for His would be
expected to dramatically abrogate or at least compromise acyl-group transfer catalysis for
this indeed-functional transcyclase. Therefore, an alternative mechanism can be
suggested for TAXO7 that incorporates the NHz—nitrogen of the TAX07 cosubstrate 3-
amino-3-phenylpropanoyl CoA as crucial acid-base catalyst in place of imidazole of the

missing His residue (Figure 31 B).

29

 

N r .
I
. , H
/Ou\' H O /Taxane O i
l SCoA “Q Asp "O Asp
Cysvs\ H ______ N/AN’l-Il

 

 

 

 

Figure 31: A) Schematic of proposed mechanism of the active site motif
C....HXXXD motif during the acyl group transfer from CoA to the hydroxytaxane;
B) Complementary role of the amino phenylpropanoid acyl group of the CoA
thioester in the catalytic triad

The first crystal structure of a BAHD acyltransferase, vinorine synthase (VS; EC
2.3.1.160), was reported recently62, and thus serves as a structural model for the T axus
transacylases. Interestingly, while the putative catalytic residues, HXXXD, are located
within or proximate to the active site, another highly conserved region DFGWGKP is
distally located, thus raises an interesting question regarding the function of this motif
during catalysis. Analysis of the 3D-structural data of the acetyl coenzyme A-dependent
acetyltransferase, vinorine synthase, from Rauwolfia plant, reveals that the enzyme is
composed of two structural domains (Figure 32), but shares only moderate amino acid
sequence relatedness (~45% similarity; ~24% identity) with the T axus acyltransferases.
However, like the T axus transferases, vinorine synthase and all of the catalytic residues
are found in domain 1, including the signature HXXXD motif. This structural

conservation suggests that substrate variability might be encoded in the mutable domain

II.

30

   

   

A 2 13 t

Figure 32: Structure of vinorine synthsase. A and B represent orthogonal views of
the VS structure as depicted in ribbon representation. N and C denote the termini of
VS. The secondary structure elements are labeled (a1—al3 and bl—bl4), and
domains 1 and 2 are indicated. The large crossbver loop (amino acids 201-213)
connects both domains. The conserved and catalytic residues His'w and Asp‘“ are
shown in ball-and-stick representation.

Furthermore, reports on the VS mechanism suggest that the His residue is
postulated to be the only catalytic residue, whereas the Asp residue is likely for
maintaining the enzyme geometry; the Cys residue is not considered in the proposed
mechanism for VS. This latter postulate deviates from the previously proposed catalytic
mechanism involving an acid/base triad for hydrogen bonding (cf. Figure 31).62 Since the
3-amino-3-phenylpropanoyl transferase, designated as TAXO7 lacks this purported His
residue, a new mechanism can be proposed wherein the amino nitrogen of the co-

substrate (3-amino-3phenylpropanoyl coenzyme A) serves as the sole general base

involved in catalysis (Figure 33).

31

 

H‘TT
H.
“ o
CoAS : ,NH,
1 / H’
Taiane

 

 

 

Figure 33: Proposed complementary role of the amino phenylpropanoid acyl group
of the CoA thioester in the catalytsis by TAX07 during the acyl group transfer from
CoA to the hydroxytaxane; Cys and Asp are not necessary for this type of catalysis

Validation of the proposed mechanism will add another example in the list of
substrate-assisted catalysis (SAC),63'65 which marks a unique mode of catalysis that will
be explored.

A series of unnatural coenzyme A thioesters will be used to screen TAX07 in

order to test whether the hypothesized SAC mechanism is viable (Figure 34).

 

   

81 IO
Aroylpropanoyl CoA : ' Ov-
TAX07 1'2,
R1 = NH,, R, = H 0
R1 = NH2, R2 = OH
NH, o NH, o o o
Wscm W30“ SCoA
1 3 5
NH, O OH 0 0

Figure 34: Future plans to validate the SAC mechanism. According to that
mechanism (cf. Figure 33) CoA thioesters with 1, 2, 3 will be probably productive
substrates but not the 4, 5, 6 CoA thioesters

32

2. 2 RESULTS AND DISCUSSION:

The initial challenge to investigate the mechanism of the TAX07 acyltransferase
was to synthesize radiolabeled baccatin III substrate (Figure 35) at ~10 Ci/mmol specific
activity in order to gain umol-scale sensitivity of the product formed in the assays and
analyzed by high performance liquid chromatography and radioactivity detection.
Another objective was to synthesize product standards [2'-deoxyTaxol, 13-0—(3'-
hydroxy-3'-phenylpropanoyl)baccatin III, 13-0-(3'-keto-3'-phenylpropanoyl)baccatin III,
13-0-(3'-phenylpropanoyl)baccatin III, and 13-0-(3'-imino-3'-phenylpropanoyl)baccatin
III] that will confirm the formation of a putative biosynthetic product derived from C13
acylation of baccatin III by TAX07 catalysis with variously substituted
phenylpropanoids.

Synthesis of 13-3H-Baccatin III:

    

OAc 0 OH
T
HO ' PI: 0
HO § AcO
OCOPh

Figure 35: 13-3H-Baccatin 111
Since the discovery in 1971 of the diterpenoid Taxol®, with its anticancer activity
and unusual ability to stabilize the assembly of microtubules, there has been a need for
radiolabeled Taxol® and related compounds to facilitate pharmacological studies.52
Recent developments in Taxol® chemistry have provided methods for regiospecifically

tritium-labeling of the phenylpropanoyl sidechain or the baccatin III ring system.

33

In 1993, Kepler and his coworkers demonstrated the syntheses of [3"-3H] Taxol®
and [13-3H] Taxol®.66 They found that adaptations of the methods by Holton67'68 and
Ojima69 were the most effective for the small-scale required for synthesis of radiolabeled
Taxol® at high specific activity. These methods were applied in the present synthesis of
13-3H-baccatin III. The C7 hydroxy group of commercially available lO-deacetylbaccatin
III (IO-DAB) (Natland Inc., Research Triangle Park, NC) was first protected with the
triethylsilyl ether using triethylsilylchloride and pyridine. The resulting 7—TES-baccatin

III was acetylated at C10 using acetyl chloride and pyridine (Figure 36).

0H0

OHO

  

OTES

   
 

TES-Cl. Py A? no»

CHacOC|.Py 4 Ho--~
o 25 °C, 17 n, 70%

5 °C, 48 h, 60%

 

 

 

5copn ocoph

10-Deacetylbaccatin Ill 7-TES-10-Deacetylbaccatin Ill 7-TES-Baocatin Ill
5165/9. Natland Inc

Figure 36: Synthesis of 7-TES-baccatin [II from commercially available lO-DAB

In the next steps, 7-TES-baccatin III was oxidized to its 13-keto analogue using
neutral, activated manganese dioxide as per literature procedure.66 Various methods are
described for reducing the keto group to the hydroxyl and each were considered.
Treatment of that ketone with sodium borohydride in ethanol reported66 to give 35%
yield of the silylbaccatin III with considerable side reactions resulting from cleavage of
the ester groups on the baccatin 111 core. Borane-tetrahydrofiiran complex is reported70 to
selectively reduce the on, B-unsaturated ketone of progesterone. Based on this latter
report, the 13-oxobaccatin III was treated with sodium borotritide (ARC, Inc., St. Louis,
MO) in tetrahydrofuran and afforded quantitative conversion yielding 13-3H-7-TES-
baccatin III that was finally converted to the desired 13-3H-baccatin III by removing the

triethylsilyl group with HF °pyridine (Figure 37).

34

OAc o

  

 

 

    

OTES OH
1. NaBT4. THF. 0 °c
Ho-c- 6 . MnOz. CH20I2A 2. HF-Py, THF. 0 °c _ T 6’.
; n.100°/ T o ' .
. H i O o >90/omtwosteps HO ‘ H _-' O
HO é A°° Ho ; AcO
OCOPh OCOPh
7-TES-Baccatin Ill 13-Oxo-7-TES-baccatin Ill 13-3H-Baccatin Ill

Figure 37 : Synthesis of 13-3H-Baccatin III using sodium borotritide in THF

Synthesis of 2'-DeoxyTaxol®:

N—Debenzoyl-2'-deoxytaxol and 2'-deoxytaxol were synthesized by modifying a
previous method.71 The former metabolite will serve as an intermediate to investigate the
substrate specificity of the Taxus-derived N-acyl transferase”, and likely serve as a
productive substrate to screen for a putative cytochrome P450-dependent 2'-hydroxylase.
Also, the N-Debenzoyl-2'-deoxytaxol can serve as a synththetic precursor toward the
development of modified second-generation taxanes.

The second-generation taxanes mostly have modifications at 3'-amino and 13-
hydroxyl functions. For example, non-aromatic 3'-N-acy1 analogs 1 are more cytotoxic
than Taxol®73, and thiocarbamates 2 and 3 have greater cytotoxicity and tubulin
polymerization effectiveness than the unmodified drug74 (Figure 38). Perhaps the most
promising propanoyl sidechain analog is the 3'-tert-butyl-3'-N-thienoyl 4, which in
addition to possessing 2-fold better tubulin assembly properties and cytotoxicity against
816 melanoma cells relative to Taxol®, is 25 times more water soluble than the parent

drug.75

35

-H _
W 2' i 0% \ 2' 3." 032
- \ -
2 OH 03 OH
R: OX R= n-Bu ort-Bu R= n-Bu,i—Bu,t—Bu ori-Pr

o
RJLrgH o
a 3. 035.
1 OH
0
\\ IgH o R NH 0 \
SW03. R'Mok R: @ Cg
4 OH ‘ K:

5OH

Figure 38: Second generation Taxol® derivatives
Although most of efforts have been directed at modifications of the ring system,
there are very few alternate ways of coupling the side chain to baccatin HI. It has been
reported76 that the C-13 hydroxyl group of baccatin H1 is very resistant to acylation, in
part because of its hindered location at the concave face the molecule. Thus the acylating

l77 is the most

agents need to have compact structure. Though Ojima’s fl-Lactam protoco
extensively used method for the sidechain attachment to baccatin III, oxazoline78 and
dioxo-oxaisothiazole79 ring systems have also been utilized for this purpose. Less
obviously, it is also possible to prepare modified Taxol® analogues by direct modification
of an existing sidechain. This approach was demonstrated few years ago by developing a
conversion of cephalomannine to Taxol®8°, but unfortunately Taxol® itself proved
resistant to such manipulation. Another approach, that involves a dehydration

condensation step, utilizes 0,0'-di(2-pyridyl) thiocarbonate (DPTC) and DMAP to add

the side chain on baccatin 111 (Figure 39).“

36

1. DPTC. DMAP

 

 

 

0
JL toluene. 73 °C
Ph NH O (100% based on
7 66% conversion) 4
+ V
: OH HO 2. Pd(OH)2/C. H2,
OBn EtOH. rt (76%)
3. HFopyridine. THF
rt (100%)
1. DPTC, DMAP
ph O toluene. 73 °C
)W (IL (95% based on
.-‘ o -
BZN 0H 93 /o conversnon) :
r0 ; 2. TFA. H20.
. H ;‘ 0 °C, 100°/
PMP HO A00 0
> 8 steps OCOPh

Figure 39: Mukaiyama’s dehydration-condensation protocol between 7-TES-
baccatin III and protected N-benzoylphenylisoserines

Analysis of Mukaiyama’s protocol has indicated that direct attachment of the
protected sidechain (13't case in Figure 39) is difficult and very low yielding (100% based
on 66% conversion after 4 times operation, therefore one individual step yields less than
20% of product)“. It involves two deprotection steps (7-O—TES & 2'-0—Bn) leading to
Taxol® whereas any further manipulation at 3'-N, that will be necessary to generate
second generation taxanes, needs an extra deprotection steps of the 3'-N-benzoyl group
on the sidechain. Secondly, the sidechain protected as N,0-cyclic acetal (2nd case in
Figure 39) seemed to be more reactive, because of their less hindered structure, and very
efficient (the reaction yielded 95% based on 93% conversion and involves only one
deprotection steps“. But the major limitation of this option lies into the fact the synthesis
of that N,0—cyclic acetal involves more than eight steps and again the 3'-N-benzoyl group
needs to be deprotected for any further manipulation towards other Taxol® analogs.

By adapting and modifying the above dehydation-condensation protocol our goal
was to synthesize 2'-deoxyTaxol® that can be utilized both as biosynthetic as well as
synthetic precursor of modified Taxol® analogues. Whereas it can be used as a product

standard of TAX07 and TAX10 acyltransferases, its N-debenzoyl analog (3'-N-

37

2and

debenzoyl-2'-deoxyTaxol®) can also be used as a substrate for 2'-hydroxylase8
TAX10” (Figure 40). Moreover, after 2'-hydroxylation, the free 3'-amino group can be
further manipulated, synthetically via a known acylation method“, or biochemically
using TAX10, to synthesize several second-generation taxanes.
PhfifiTax 1O

nHz o

3.

   

Figure 40: Taxol® biosynthetic pathway
First step was to synthesis of 7-TES-baccatin III that was done by the method
mentioned before (Figure 41). Attempted coupling of 7-TES-baccatin III with N-benzoyl-
fl-phenylalanine using DPTC and DMAP failed to produce the Taxol® analog indicating
towards the fact that 2'-oxygen might be a condition for this coupling reaction.

TES-Cl, Pyridine
O 25 °C, 17 h. 70%

CH3COCI, Pyridine:
5 °C, 48 h. 60%

 

 

> OH“

     

6coph

10-Deacetylbaccatin Ill 7-TES-10—Deacetylbaccatin III 7-TES-Baocatin Ill
5165/9, Nattand Inc

Figure 41: Synthesis of 7-TES-baccatin 111
However, after several unsuccessful attempts with N—benzoyl-fl-phenylalanine at
varying reaction conditions, it was observed that changing the 3'-amino protecting group
from benzoyl to t-butyloxycarbonyl (Boc) allow the reaction to proceeds successfully
with good yield (Figure 42). It is important to note that excess (6.0 equivalent) DPTC and

N-Boc-fl-phenylalanine was used instead of 2.0 equivalents of each as reported earlier.85

38

ores
B°°‘NH o

DMAP, DPTC :
N-Boc-fl-Phenylalanine _ ' 0' ..
Toluene, 70 °C, overnight
74%

0H---

 

  

O

H .

   

HO ; Acé
OCOPh
7-TES-Baccatin Ill N-Boc-7-TES-2 -deoxytaxol

Figure 42: N-Boc-7-TES-2'-deoxyTaxol®

HF°pyridine was introduced to deprotect the TES group at C7. After complete
deprotection of 7-hydroxy function deprotection of 3'-N-Boc group was attempted.
Trifluoroacetic acid was tried for this purpose. Surprisingly complete decomposition of
the substrate under TFA condition was observed. Reaction at low temperature (0 °C) was
also attempted which resulted in no change of substrate. Altering the deprotection
sequence (3'-N-Boc followed by 7-TES) caused decomposition at the very beginning
(during deprotection of 3'-N—Boc). It was apparent that deprotection with TFA was
decomposing the substrate perhaps due to its very high acidity. To avoid use of this
strong acid, 80% aqueous formic acid was employed and it was observed that both 7-TES
and 3'-N-Boc group were deprotected with formic acid (Figure 43). Use of formic acid
thus not only afforded desired product (3'-N-debenzoy1-2'-deoxyTaxol®) but also

shortened the synthetic path by eliminating one extra deprotection step.

80% HCOOH :
n, 100%

 

   

N-Boc-7-TES-2'-deoxytaxol N-Debenzoyl-2'-deoxytaxol
Figure 43: Synthesis of N-debenzoyl-2'-deoxyTaxol®
Whereas the intermediate 3'-N-debenzoyl-2'-deoxyTaxol® can be used as a

substrate for TAX10, the benzoylation was done chemically also to have a substrate for

39

2'-hydroxylase and product standard for TAX07 and TAX10 Taxol® pathway
acyltransferases. The benzoylation of the 3'-amino group was done by a literature
method84 to obtained quantitative conversion of the substrate to 2'-deoxyTaxol®, the

target compound (Figure 44).

OAc o

    

PhCOCI, pH 9 >
rt. 100%

 

 

N-Debenzoyl-2'-deoxytaxol 2'-deoxytaxol

Figure 44: Synthesis of 2'-deoxyTaxol®

Conclusion:

Modification of an existing method for the coupling of the 3'-amino-3'-
phenylpropanoid sidechain of Taxol® onto C13 of baccatin III is described. Although
many methods for the attachment of C13 sidechain to baccatin III for the synthesis of
Taxol® and related compounds were developed in the last few years by several groups,
the described method, will give access to a library of Taxol® derivatives including
second-generation taxanes in a short synthetic route. Furthermore, synthesis of
radiolabeled baccatin 111, an advanced taxol biosynthetic pathway intermediate, is also
described. Foreseeably, application of these Taxol® analogues toward elucidating the

putative SAC mechanism of TAX07 acyltransferase is feasible.

40

CHAPTER 3

EXPERIMENTAL METHODS

General Methods:

Coenzyme A as trilithium salt was purchased from Sigma. (R,S)-3-(tert-
butoxycarbonylamino)—3-phenylpropanoic acid and (R)-3-hydroxy-3-phenylpropanoic
acid was purchased from Alfa Aesar. Benzoylacetic acid was prepared per literature
procedure.86 (Z)-ethyl-3-(tert-butoxycarbonylamino)-3-phenylacrylic acid was prepared
as per literature procedure87 from ethyl isonicotinoylacetate (purchased from AK
Scientific, Inc, Mountain View, CA). (R,E)-3-(tert-butoxycarbonylamino)-5-phenylpent-
4-enoic acid was prepared from (S)-3-amino-5~phenylpentanoic acid (purchased from
Chem-Impex International, Inc. Wood Dale, IL) as per literature procedure.88 N—Boc-
(2R,3.S')-3-phenylisoserine was purchased from Chem-Impex International, Inc. (Wood
Dale, IL). Ethyl chloroformate, ethyl benzoylacetate and all other reagents were
purchased from Aldrich unless notified otherwise. C18 (carbon 11%) reversed phase
silica gel was purchased from Silicycle, (QC, Canada).

All substrates were used without further purification. Tetrahydrofuran and
dichloromethane were obtained from dry still packed with activated alumina that was

pressurized with nitrogen gas. Silica gel (230-400 Mesh) and aluminium backed silica gel

41

6O TLC plates, embedded with A254 chromophores, were purchased from EMDTM
Chemicals Inc (Gibbstown, NJ).

All reported yields are for isolated materials. 1H and 13C NMR spectra were
recorded on a Varian Inova-300 (300.11 and 75.47 MHz respectively), Varian VXR-SOO
or Varian Unity-SOO-Plus spectrometer (499.74 and 125.67 MHz respectively) and were
referenced to residual solvent signals either at 7.24 ppm or at 4.67 ppm for CDC13 and
D20 respectively. All apparent coupling constants (J values) were measured at the
indicated field strengths.

Tandem gas chromatography/mass spectrometry analysis was conducted by
loading 1 pl of sample onto an HP SHS GC column (0.25-mm inner diameter x 30 m,
0.25-um film thickness) (Agilent, Palo Alto, CA) coupled to mass selective detector
(model 5973 inert®, Agilent) in ion scan mode from 50-300 atomic mass units.
Electospray Ionization Mass Spectrometry (ESI-MS) analysis was conducted by loading
0.1 mM solution of thioester samples in water (pH = 5) onto a Q-Tof UltimaTMAPI
(Micromass, Beverly, MA) mass spectrometer coupled with a LC-system (model 2795,

Waters, Milford, MA) at the Mass Spectrometry Facility, Michigan State University.

General Procedure:

The carboxylic acid (1 equivalent) was suspended in THF was added ethyl
chloroformate (1.5 equivalents) and 1M Et3N in THF (1.2 equivalents) under N2 to form
the mixed anhydride. The reaction was stirred for one hour at room temperature with
monitoring by analytical TLC. After completion of the reaction, the solvents were

evaporated under reduced pressure followed by vacuum and the residue was dissolved in

42

t-BuOH. CoA (as trilithium salt, 1.2 equivalents) in 0.4 M NaHCO3 was added to the
solution and the mixture was stirred for half hour at room temperature. Then quenched
with 1M HCl and adjusted to pH 5. The solvents were evaporated under reduced pressure
at room temperature. The residue was purified by flash column chromatography using
C13-silica gel that was first loaded and washed with methanol and then with water
(pH=5). The residue, after loading on the column, was first eluted with water (10-20 m1)
and then with increasing concentrations of methanol (5-100%) in water (pH=5). The
product comes with about 10-20% methanol. The derived product was then lyophilized.
To remove the N-Boc protection, the residue was dissolved in 1 ml water, cooled to 0 °C,
and 1 ml of trifluoroacetic acid was added dropwise with stirring for 1h to deprotect the
amino group. The mixture was then warmed to room temperature and stirred an
additional 1 h. The progress of decarbonylation of the N-Boc compound was monitored
by silica gel analytical TLC (l-butanol/HZO/AcOH, 523:2, vol/vol/vol) with detection by
UV absorbance. After complete deprotection, the reaction was diluted with 50 ml of
water and concentrated to 0.5 ml under vacuum; this dilution and evaporation'process
was repeated three times to remove residual trifluoroacetic acid. Finally the sample was
concentrated to dryness and resuspended in 5 ml of water. The product was purified by
C18 silica gel flash column chromatography as described above. The CoA thioesters were
eluted with 15-20% methanol, which was removed in vacuo and finally lyophilized to
obtain a white solid. Purity of the product was assessed by analytical TLC. Compound

was then confirmed by ESI-MS and 1H NMR spectroscopy.

43

Experimental Details and Spectroscopic Data:

tfiN
N __N
1o'a11- A \ N/>\He

/
H3C” CH3 9 9| HC N
" o-fi-o-fi-o 5
c»1 OH 4 1~

(3RS)-fl-Phenylalanoyl coenzyme A.

 

O OOH

\‘ I

HO OH

NHZC)
Figure 45: (3RS')-fl-Phenylalanoyl coenzyme A.

Following the general procedure, N-Boc-fl-phenylalanine (26 mg, 98 umol)
suspended in THF (1.4 ml) was added ethyl chloroformate (18 ul, 15.9 mg, 147 umol)
and 1M Et3N in THF (13 pl, 118 umol) under N2 to form the mixed anhydride. The
reaction was stirred for one hour at room temperature with monitoring by analytical TLC.
After completion, the solvents were evaporated under reduced pressure followed by
vacuum and the residue was dissolved in 2 ml t-BuOH. CoA (as trilithium salt, 83.7 mg,
108 umol) in 2 ml 0.4 M NaHCO3 was added to the solution and the mixture was stirred
for another half hour at room temperature. Then quenched with 1M HCl and adjusted to
pH 5. The solvents were evaporated under reduced pressure at room temperature. The
residue was purified by the described method using C13 silica gel chromatography. The
product comes with about 10-15% methanol. The derived N-Boc-fl-phenylalanoyl
Coenzyme A was then lyophilized, and the Boc group was deprotected by above
described method and the residual trifluoroacetic acid was removed. Finally the sample

was concentrated to dryness and resuspended in 5 ml of water. The product was purified

44

by C13 silica gel flash column chromatography as described above. The CoA thioester
was eluted with 20% methanol, which was removed in vacuo and finally lyophilized to
obtain a white solid. The purity was assessed by analytical TLC which was found to be
98-100%. Compound was then confirmed by ESI—MS and lHNMR spectroscopy. Yield =
66% based on CoA. ESI-MS: Calculated 914.18, Experimental [M—H] 913.27, [M+H]
was not possible. 1H-NMR (500 MHz, D20) (see Figure 45 for numbering) 5 (in ppm) :
0.66 (3H, s, H-lO'), 0.78 (3H, s, H-1 1'), 2.16 (2H, t, H-4'), 2.8 (2H, m, H—l'), 3.1 ( 2H, t,
H-2'), 3.26-3.28 (4H, m, H-2 & H-5'), 3.44 (1H, dd, J = 4.8 and 9.6 Hz, Ha-S"), 3.66 (1H,
dd, J = 4.8 and 9.6 Hz, Hb-S"), 3.86 (1H, s, H-7'), 4.06 (2H, s, H—9'), 4.43 (1H, ddd, J =
2.7 and 5.3 Hz, H-4"), 4.56-4.66 (2H, m, H-2" and H-3"), 6.0-6.1 (two doublets; one set
from each stereoisomer, J = 6.9 Hz for both, H-l"), 7.21-7.27 (5H, m, HA phenyl
protons), 8.26 (1H, 3, HC adenine-CH), 8.33 (1H, 3, H3 adenine-CH).

(R)-3-Hydroxy-3-phenylpropanoyl coenzyme A.

H2N
—.N
I I N
10&11 X \N/)\He
N
0

H304 CH3 0 0 HC

H H01. " O-llii-O—l'ji-O 5"
A o7 6. 9' OH OH 4 1~

O NH
3 1 S 1' - O“ ’0 OH
2 W” 3' 4 5' HO/ \OH
OH 0

Figure 46: (R)-3-Hydroxy-3-phenylpropanoyl coenzyme A.
Following the general procedure, (R)-3:hydroxy-3-phenylpropanoic acid
(21 mg, 125 pmol) suspended in THF (1.8 ml) was added ethyl chloroformate (23 pl,

20.3 mg, 188 umol) and 1M Et3N in THF (17 pl, 150 umol) under N2 to form the mixed

anhydride. The reaction was stirred for one hour at room temperature with monitoring by

45

analytical TLC. After completion, the solvents were evaporated under reduced pressure
followed by vacuum and the residue was dissolved in 2.5 ml t—BuOH. CoA (as trilithium
salt, 107 mg, 138 umol) in 2.5 ml 0.4 M NaHCO3 was added to the solution and the
mixture was stirred for another half hour at room temperature. Then quenched with 1M
HCl and adjusted to pH 5. The solvents were evaporated under reduced pressure at room
temperature. The residue was purified by the described method using C13 silica gel
chromatography. The product comes with about 15% methanol which was evaporated
under vacuo and the residue was finally lyophilized to white solid. The purity was
assessed by analytical TLC which was found to be 98-100%. Compound was then
confirmed by ESI-MS and 1HNMR spectroscopy. Yield = 68% based on CoA. ESI-MS:
Calculated 915.17, Experimental [M—H] 914.09, [M+H] was not possible. 1H-NMR (500
MHz, D20) ( see Figure 46 for numbering) 8 (in ppm) : 0.66 (3H, s, H—10'), 0.78 (3H, s,
H-1 1‘), 2.39 (2H, t, H-4'), 2.80 (2H, m, H-l’), 3.20 ( 2H, t, H-2'), 3.36-3.38 (4H, m, H—2
& H-5'), 3.46 (1H, dd, J = 4.8 and 9.6 Hz, Ha-S"), 3.80 (1H, dd, J= 4.8 and 9.6 H2, H1,-
5"), 3.96 (1H, s, H-7'), 4.06 (2H, s, H-9'), 4.46 (1H, ddd, J = 2.7 and 5.3 Hz, H-4"), 4.60-
4.82 (2H, m, H-2" and H-3"), 5.80 (1H, m, H-3), 6.40 (1H, d, J= 6.9 Hz H-l"), 7.21-7.27
(5H, m, HA phenyl protons), 8.20 (1H, 5, HC adenine-CH), 8.43 ( 1H, 3, H3 adenine-CH).
(3R)- Styryl-fl—alanoyl coenzyme A.
Synthesis of Sodium (3R, 4E)-N-Boc-3-amino-5-phenylpent-4-enoate:

Thionyl chloride (0.1 ml, 1.5 mol) at 0°C was added to the reaction mixture
dissolved in methanol at a rate to keep the stirred suspension of (S)-3-amino-5-
phenylpentanoic acid (193 mg, 1 mol) at reflux. After the initial refluxing ceased, the

mixture was heated at 70 °C for an additional 2 h, then the reaction was cooled, and the

46

methanol was removed under reduced pressure to give a crude methyl ester hydrochloride
salt of (S)-3-amino-5—phenylpentanoate, which was used in next reaction step without
further purification. For reference, a small amount of the amine (Rf = 0.2) was evaluated

on analytical silica gel TLC (95:5 CH2Cl2/methanol, v/v).

HZN
N ...N
X \ NATHB

ooHcN

    

NH2 0
Figure 47 : (3R) Styryl—fl-alanoyl coenzyme A.

To the crude mixture of the fl-amino methyl ester intermediate (193 mg, 0.8
mmol), triethylamine (0. 12 ml, 0.84 mmol), pyridine (0.06 ml, 0.84 mmol) dissolved in
THF (0.6 ml) and CHCl3 (0.6 ml) at 0 °C was added solid di-tert-butyldicarbonate (185
mg, 0.84 mmol). The solution was stirred at 0 °C for half hour, and then overnight at
room temperature. The solution was washed with 15% phosphoric acid, aqueous
NaHCO3 solution and saturated NaCl solution. The organic fraction was dried over
Na2SO4, filtered and the solvent evaporated to dryness under reduced pressure. The crude
product was purified by silica gel flash column chromatography (90: 10 hexane/ethyl
acetate, v/v); the fractions containing methyl (S)-N-Boc-3-amino-5-phenylpentanoate
were combined, and the solvent was evaporated to give the product as a colorless oil
which solidified upon overnight drying in vacuo (76% yield, Rf = 0.35 in 90:10
hexane/ethyl acetate, v/v on silica gel TLC). 1H-NMR (300 MHz, CDCl3) (it ppm 1.42

[9H, s, C(CH3)3], 1.77-1.91 (2H, m, PhCH2CH2), 2.50-2.85 (4H, m, PhCH2CH2 and

47

CH2CO), 3.68 (3H, s, OCH3), 3.90-4.01 (1H, m, NHCH), 4.97 (1H, d, J = 9.9 Hz, NH)
and 7.13-7.31 (5H, m, aromatic protons).

The N—Boc protected amine, described above, was converted to the unsaturated
methyl (3R, 4E)-N-Boc-3-amino-5-phenylpent-4-enoate according to a literature
procedure that describes a sequential and selective benzylic bromination, by N-
bromosuccinimide treatment followed by a dehydrobromination.88 Under an inert
atmosphere of nitrogen, a solution of N—Boc ester (0.5 mmol, 0.153 g) in dry CC14 (2 ml)
was treated with N-bromosuccinimide (0.56 mmol, 0.097 g). The reaction mixture was
illuminated for 3 hour under a 60 W tungsten lamp held ~ 2 cm away from the flask
while maintaining the temperature at less than 10 °C (in cold room). The product mixture
was purified by silica gel flash column chromatography (90:10 hexane/ethyl acetate, v/v);
the fractions containing the alkene intermediate were combined, and the solvent was
evaporated to give the product as a white solid (40 % yield, Rf = 0.52 in 70:30
hexane/ethyl acetate, v/v on silica gel TLC). 1H-NMR (300 MHz, CDC13) é; ppm 1.46
[9 H, s, C(CH3)3], 2.67-2.74 ( 2 H, m, CH2CO), 3.70 (3 H, s, OCH3), 4.62-4.78 (1 H, m,
NHCH), 5.23-5.37 (1 H, m, NH), 6.19 (1 H, dd, J= 6.2 and 15.7, PhCH=CH), 6.55 (l H,
d, J = 15.7, PhCH) and 7.20-7.38 (5H, m, aromatic protons).

The N-Boc methyl ester, methyl (3R, 4E)-N-Boc-3-amino-5-phenylpent-4-enoate
(20 mg, 65 umol), described above, in THF (1.2 ml) was hydrolyzed for 12 h with NaOH
(65 umol, 32.5 ul of a 2M aqueous solution) to the yield the sodium salt of (3R, 4E)-N-

Boc-3-amino-5-phenylpent-4-enoic acid.

48

Synthesis of (3R)-Styryl-fl—alan0yl Coenzyme A:

This carboxylate sodium salt was suspended in THF (1.4 ml) to which was added
ethyl chloroformate (7.0 ul, 7.8 mg, 72 umol) under nitrogen to form the mixed
anhydride. The mixture was stirred vigorously at room temperature for 1 h. The
transesterification of the mixed anhydride with CoA (as trilithium salt, 60 mg, 78 mo] in
1.4 ml 0.4 M NaHCO;;), the purification of N-Boc protected CoA ester, and the N-
deprotection with trifluoroacetic acid were al.] performed as described in general
procedure. The final product (3R)-styryl-fl—alanoyl coenzyme A, was eluted fiom the C13
silica gel column with 15-20% methanol which was evaporated and the residue was
lyophilized to give the product as white solid. The purity was assessed by analytical TLC
which was found to be 98—100%. Compound was then confirmed by ESI-MS and
lHNMR spectroscopy. Yield = 60% based on CoA. ESI-MS: Calculated 940.20,
Experimental [M—H] 939.33, [M+H] was not possible. ‘H NMR (500 MHz, D20) (see
Figure 47 for numbering) 6 (in ppm) : 0.68 (3H, s, H-10'), 0.76 (3H, s, H-1 1'), 2.16 (2H,
t, H-4'), 2.82 (2H, m, H-l'), 3.11 (2H, t, H-2'), 3.26-3.28 (4H, m, H-2 & H-5'), 3.44 (1H,
dd, J= 4.8 and 9.6 Hz, Ha-S"), 3.66 (1H, dd, J= 4.8 and 9.6 Hz, Hb-S"), 3.86 (1H, s, H-
7'), 4.06 (2H, s, H-9'), 4.43 (1H, ddd, J = 2.7 and 5.3 Hz, H-4"), 4.56-4.66 (2H, m, H-2"
and H-3"), 4.67-4.78 (1H, m, H-3) 6.0-6.1 (1H, d, J: 6.9 Hz, H-l"), 6.19 (1H, dd, J= 6.2
and 15.7, H-4), 6.55 (1H, d, J = 15.7, H-S), 7.21-7.27 (5H, m, HA phenyl protons), 8.26

( 1H, s, Hc adenine-CH), 8.33 (1H, 3, HB adenine-CH).

49

(2R, 35)-3-phenylisoserinoyl coenzyme A.

HZN
10511‘ :{N ATHB
H30 CH3 0 Hc/k
H HO,_ o- P'— o— P'- o 5
A o7 6. 9' OH OH 4 1-
OH NH
3 1 S\1'/\ O 4' 5' O“ ’0 OH
2 2' N 3' F’\
H HO’ OH

NH2 0
Figure 48: (2R, 3S)-3-phenylisoserinoyl coenzyme A
N-Boc-(2R,3S)-3-phenylisoserine (62 mg, 0.2 mmol) in 2.5 ml of THF was
esterified by diazomethane treatment. The crude ester was subjected to silica gel flash
column chromatography (50:50 hexane/ethyl acetate, v/v) to yield the pure ester
derivative (0.18 mmol, 90% yield). This methyl ester (55 mg, 0.18 mmol) was dissolved
in THF (3.7 ml) and added to a stirred suspension of sodium hydride ( 1.1 mmol) in THF
(2.5 ml) under nitrogen. To the mixture was added di-t-butyl dicarbonate (46 mg, 0.2
mmol) in THF (5 ml), and the suspension was stirred at room temperature for 20 minute.
The mixture was then chilled on ice, quenched with 1 ml of water, and finally filtered.
The filtrate was collected and the solvent evaporated. The product was purified by silica
gel flash column chromatography (65:35 hexane/ethyl acetate, v/v) to yield pure N,0-di-
Boc-(2R,3S)-3-phenylisoserine methyl ester (0.16 mmol, 90% yield). This methyl ester
(26 mg, 65 umol), described above, in THF (1.2 ml) was hydrolyzed for 12 h with NaOH
(65 umol, 32.5 ul of a 2M aqueous solution) to the yield the sodium salt of N,0-di-Boc-
(2R,3S)-3-phenylisoserine. This carboxylate sodium salt was suspended in THF (1.4 ml)
to which was added ethyl chloroformate (7.0 u], 7.8 mg, 72 umol) under nitrogen to form

the mixed anhydride. The mixture was stirred vigorously at room temperature for 1 h.

50

The transesterification of the mixed anhydride with CoASH (as trilithium salt, 60 mg, 78
umol in 1.4 ml 0.4 M NaHCO3), the purification of N,0-di-Boc protected CoA ester, and
the N,0-deprotection with trifluoroacetic acid were all performed as described in general
procedure. The final product (2R,3S)-3-phenylisoserinoyl coenzyme A, was eluted from
the C13 silica gel column with 15-20% methanol which was evaporated and the residue
was lyophilized to give the product as white solid. The purity was assessed by analytical
TLC which was found to be 98-100%. Compound was then confirmed by ESI-MS and
1HNMR spectroscopy. Yield = 67% based on CoA. ESI-MS: Calculated 930.1,
Experimental [M—H] 928.8, [M+H] was not possible. 1H NMR (300 MHz, CD3OD) (see
Figure 48 for numbering) 5 (in ppm) : 0.83 (3H, s, H-10'), 1.05 (3H, s, H-1 1'), 2.44 (2H,
t, H-4'), 2.79 (2H, m, H-l'), 3.45 ( 2H, m, H-2'), 3.47 (2H, dd, J = 6.6 Hz, H-5'), 3.57
(1H, dd, J= 6.6 and 10.5 Hz, Ha-S"), 3.98 (1H, dd, J= 5.4 and 9.9 Hz, Hb-S"), 4.06 (1H,
s, H-7'), 4.25 (2H, s, H-9'), 4.49 (1H, br ddd, H-4"), 4.69-4.90 (4H, m, H-2, H-3, H-2" and
H-3"), 6.13 (d, J = 6.0 Hz, H-l"), 7.25-7.41 (5H, m, HA phenyl protons), 8.18 (1H, 5, HC
adenine-CH), 8.57 ( 1H, 3, H3 adenine-CH).

2'-Deoxytaxol.

 

R = H, N-Debenzoyl-2'-deoxytaxol
R = COPh, 2'-Deoxytaxol

Figure 49: N-Debenzoyl-2'-deoxytaxol and 2'-Deoxytaxol.

51

Preparation of 7-TES-10-deacetylbaccatin II]: To a solution of commercially available
10-deacetylbaccatin III (0.45 g, 0.8 mmol) in pyridine (20 mL) was added TES-Cl (2.5
mL, 16 mmol, 20 equivalents) dropwise. The solution was stirred at 25 °C for 17 h. After
dilution with ether (100 ml), the solution was washed with aqueous CuSO4 solution (3 x
30 mL) and brine (3 x 20 mL). The organic layer was dried over MgSO4, concentrated
and purified by flash column chromatography (65:35 hexanes/ethyl acetate, v/v) to
produce 7-TES-10-deacetylbaccatin III as white solid (yield = 70%).

Preparation of 7-TES-baccatin 111: To a solution of 7-TES-lO-deacetylbaccatin III (0.3 g,
0.46 mmol) in pyridine (10 mL) at 0 °C was added acetyl chloride (0.17 mL, 2.3 mol, 5
equivalents) dropwise. The solution was stirred at 5-10 °C (in cold room) for 48 h. After
dilution with ether (100 ml), the solution was washed with aqueous CuSO4 solution (3 x
30 mL) and brine (3 x 20 mL). The organic layer was dried over MgSO4, concentrated
and purified by flash column chromatography (65:35 hexanes/ethyl acetate, v/v) to
produce 7-TES-baccatin III as white solid (yield = 60%).

Preparation of N-Boc- 7—TES-2’-deoxytaxol: Di-pyridylthiourea (0.3 g, 1.3 mmol) was
added to a solution of 3-(R,.S')-N-Boc-,B-phenylalanine (Alfa Aesar, Ward Hill, MA) (0.3
g, 1.3 mmol), N,N-dimethylaminopyridine (0.167 g, 1.3 mmol) and 7-TES-baccatin III
(0.15 g, 0.22 mol) in toluene (24 mL). The reaction was stirred at 70 °C for overnight.
The solvent was evaporated under reduced pressure and the crude residue was purified by
flash column chromatography using 65:35 hexanes/ethyl acetate, v/v. Fractions
containing the product were combined and solvent was evaporated under reduced

pressure to obtain the desired product as pale white solid (yield = 74%).

52

Preparation of N—debenzoyl-Z'-deoxytaxol: 0.15 mL of formic acid (88% aqueous
solution) was added to a solution of N-Boc-7-TES-2'-deoxytaxol (50 mg, 53 mol) in 1
mL dichloromethane and stirred at room temperature with monitoring by TLC. Residual
formic acid was removed on a vacuum pump upon completion of the reaction indicated
by TLC (no starting material was observed). The reaction mixture was then diluted with
ethyl acetate and the solution was washed with 5% NaHC03 (2 x 10 mL), water (2 x 10
mL), brine (2 x 10 mL) and then dried over sodium sulfate. The solution was then filtered
and solvent was removed under reduced pressure. The residue was purified by PTLC (90:
10 ethyl acetate/methanol, v/v) to obtain the desired product as yellow solid (yield =
100%). 1H NMR (500 MHz, CDC13) (see Figure 49 for numbering) 8 (in ppm): 1.14 (3H,
s, H-16), 1.24 (3H, s, H-17), 1.68 (3H, s, H-19), 1.79 (3H, d, J= 1.5 Hz, H-18), 1.88 (1H,
ddd, J = 2.3 Hz, 11.0 Hz, 14.7 Hz, Hb-6), 2.23 (3H, s, 10-OAc), 2.28 (2H, s H-14), 2.38
(3H, s, 4-OAc), 2.54 (1H, ddd, J: 6.7 Hz, 9.7 Hz, 14.8 Hz, Ha-6), 2.98 (1H, d, J= 2.7,
H-2'), 3.24 (1H, br (1, H-3') 3.79 (1H, dd, J= 1.0 Hz, 7.0 Hz, H-3), 4.19 (1H, dd, J= 1.0
Hz, 8.5 Hz, Hb-20), 4.30 (1H, dd, J = 1.0 Hz, 8.4 Hz, Ha-20), 4.40 (1H, dd, J = 6.7 Hz,
10.9 Hz, H-7), 4.94 (1H, d, J= 0.8, H-S), 5.67 (1H, d, J= 7.1 Hz, H-2), 6.37 (1H, s, H-
10), 7.35 (1H, t, t, p-Ph-2), 7.42 (2H, m, m-Ph-2), 7.48 (2H, m, o-Ph-2), 7.51 (2H, m, m-
Ph-l), 7.61 (1H, t, t, p-Ph-l), 8.13 (2H, two doublets, o-Ph-l).

Preparation of 2 '-de0xytaxol .'

In a screw-cap tube 8.3 mg (10 umol) of N-debenzoy1-2'-deoxytaxol was taken and
dissolved in 0.1 N aqueous NaOH solution to maintain the pH at 9.0. Benzoyl chloride
(50 uL) was then added to it. The mixture was stirred vigorously on a vortex for 30

minute. The aqueous layer was then extracted with ethyl acetate (5 mL x 2). The organic

53

fraction was then washed with brine (5 mL x 2), dried over sodium sulfate, filtered.
Solvent was evaporated under reduce pressure and the residue was purified by PTLC
(65:35, hexanes/ethyl acetate, v/v) to obtain the desired product as a pale white solid
(yield = 90%). 1H NMR (500 MHz, CDC13) (see Figure 49 for numbering) 5 (in ppm):
1.14 (3H, s, H-16), 1.24 (3H, s, H-17), 1.68 (3H, s, H-19), 1.79 (3H, d, J= 1.5 Hz, H-18),
1.88 (1H, ddd, J: 2.3 Hz, 11.0 Hz, 14.7 Hz, Hb-6), 2.23 (3H, s, lO-OAc), 2.28 (2H, s H-
14), 2.38 (3H, s, 4-OAc), 2.54 (1H, ddd, J = 6.7 Hz, 9.7 Hz, 14.8 Hz, Ha-6), 2.98 (1H, d,
J: 2.7, H-2'), 3.79 (1H, dd, J= 1.0 Hz, 7.0 Hz, H-3), 4.19 (1H, dd, J= 1.0 Hz, 8.5 Hz,
Hb-20), 4.30 ( 1H, dd, J = 1.0 Hz, 8.4 Hz, Ha-20), 4.40 (1H, dd, J = 6.7 Hz, 10.9 Hz, H-7),
4.94 (1H, d, J= 0.8, H-S), 5.24 (1H, br d, H-3'), 5.67 (1H, d, J= 7.1 Hz, H-2), 6.37 (1H,
s, H-10), 7.35 (1H, t, t, p-Ph-2), 7.40 (2H, m, m-Ph-N), 7.42 (2H, m, m-Ph-2), 7.48 (2H,
m, o-Ph-2), 7.49 (1H, m, p-Ph-N), 7.51 (2H, m, m-Ph-l), 7.61 (1H, t, t, p-Ph-l), 7.74
(2H, two doublets, o-Ph-N), 8.13 (2H, two doublets, o-Ph-l).

13-3H-Baccatin III.

 

Figure 50: 13-3H-Baccatin III
Preparation of 13-0xo- 7-TES-baccatin III: 7-TES-baccatin (0.1 g, 0.14 mmol), prepared
by the method described above, was dissolved in dichloromethane (10 mL) and

manganese (IV) oxide powder was added and stirred slowly with monitoring the reaction

54

by TLC. Upon completion of the reaction, (i.e. when TLC showed no starting material
present), the reaction mixture was filtered and solvent was evaporated under reduced
pressure. The residue was then purified by flash column chromatography using 65:35,
hexanes/ethyl acetate, v/v. Fraction containing the desired product were then combined
and solvent was evaporated under reduced pressure to obtain the product as white solid
(yield = 100%).

Preparation of 13-3H—baccatin III:

In a vial l3-Oxo-7-TES-baccatin III (10 mg, 0.014 mmol) in THF (0.5 mL) was taken
and stirred at 0 °C. In another vial excess 3H-NaBH4 (50 mCi, specific activity 100-500
mCi/mmol), solid, ARC Inc.) was dissolved in minimum amount of 0.01N sodium
hydroxide solution which was then slowly added by a syringe to the vial containing the
substrate. The reaction was then stirred at 0 °C for 10 minute and then at room
temperature with occasional monitoring by TLC. The reaction was quenched by adding
the starting material (10 mg) and was stirred for another 2 h. Finally water (0.5 ml) was
added to the reaction. The reaction mixture was then extracted with ethyl acetate (5 mL x
2) and the solvent was evaporated under flow of nitrogen gas. Without further
purification (to avoid loss of radioactivity), the residue was taken to the next deprotection
step and the 7—TES-group was deprotected in the same way as described above. The
residue obtained after solvent evaporation was then purified by PTLC (60:40,
hexanes/ethyl acetate, v/v) to get the product as white solid (8 mg, 0.014 mmol, 100%
chemical yield). The radioactivity of the product was then determined by a liquid
scintillation counter (model 14/4C Wallace Instruments, Copley, OH) (0.493 x 109 DPM,

0.22 mCi, specific activity 16 mCi/mmol). The identity of the product was confirmed by

55

eluting it in HPLC, which, under same condition, indicated same retention time with

commercially available authentic baccatin III sample.

56

10.

11.

12.

13.

14.

15.

16.

17.

BIBLIOGRAPHY

Haynes, R. K. Current Topics in Medicinal Chemistry (Sharjah, United Arab
Emirates) 2006, 6, 509.

Lommen, W. J. M.; Schenk, E., Bouwmeester, H. J .; Verstappen, F. W. A. Planta
Med. 2006, 72, 336.

Larkin, J. M. G.; Kaye, S. B. Expert Opinion on Investigational Drugs 2006, 15,
691.

Bergstralh, D. T.; Ting, J. P. Y. Cancer Treatment Reviews 2006, 32, 166-179.
Garcion, E.; Lamprecht, A.; Heurtault, B.; Paillard, A.; Aubert-Pouessel, A.;
Denizot, B.; Menei, P.; Benoit, J. -P. Molecular Cancer Therapeutics 2006, 5,
1710.

Galis, 1.; Simek, P.; Narisawa, T.; Sasaki, M.; Horiguchi, T.; Fukuda, H.;
Matsuoka, K. Plant J. 2006, 46, 573.

Tan, J .; Wang, H. -L.; Yeh, K. -W. Biotechnology Letters 2005, 27, 1517.

Jennewein, S.; Wildung, M. R.; Chan, M.; Walker, K.; Croteau, R. Proc. Natl.
Acad. Sci, USA. 2004, 101, 9149.

Kim, J.; DellaPenna, D. Proc. Natl. Acad. Sci., USA. 2006, 103, 3474.
Kaeberlein, T.; Lewis, K.; Epstein, S. S. Science 2002, 296, 1127.

Coleman, T. M.; Huang, F. Chem. Biol. 2002, 9, 1227.

Takahashi, N.; Yamada, T. Oral Microbiology and Immunology 2000, 15, 96.
Perez, A. G.; Olias, R.; Luaces, P.; Sanz, C. J. Agric. Food Chem. 2002, 50, 4037.

Zou, J .; Zheng, Z. (20020923, 2003) Patent No: 2002-CA1445
2003025165.

Wiberg, E.; Banas, A.; Stymne, S. Planta 1997, 203, 341.
Vogel, G.; Browse, J. Plant Physiol. 1996, 110, 923.

Bisanz, C.; Bastien, 0.; Grando, D.; Jouhet, J .; Marechal, E.; Cesbron-delauw, M.
-F. Biochem. J. 2006, 394, 197.

57

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Walker, K.; Croteau, R. Phytochemistry 2001, 58, 1.

Stewart, C., Jr., Kang, B.-C., Liu, K., Mazourek, M., Moore, S. L., Yoo, E. Y.,
Kim, B. -D.; Paran, 1.; Jahn, M. M. Plant J. 2005, 42, 675.

Vigeolas, H.; van Dongen, J. T.; Waldeck, P.; Huhn, D.; Geigenberger, P. Plant
Physiol. 2003, 133, 2048-2060.

White, S. W.; Zheng, J.; Zhang, Y. -M.; Rock, C. 0. Annual Review of
Biochemistry 2005, 74, 791.

Jeong, J. -C.; Srinivasan, A.; Grueschow, S.; Bach, H.; Sherman, D. H.; Dordick,
J. S. J. Am. Chem. Soc. 2005, 127, 64.

Ma, X.; Koepke, J.; Bayer, A.; Fritzsch, G.; Michel, H.; Stoeckigt, J. Acta
Crystallographica, Section D: Biological Crystallography D61, 694.

Alonso, M. J.; Bermejo, F.; Reglero, A.; Fernandez-Canon, J. M.; Gonzalez de
Buitrago, G.; Luengo, J. M. J. Antibiot. 1988, 41, 1074.

Momoi, M.; Tanoue, D.; Sun, Y.; Takematsu, H.; Suzuki, Y.; Suzuki, M.; Suzuki,
A.; Fujita, T.; Kozutsumi, Y. Biochem. J. 2004, 381, 321.

Sigma catalog, 2006-2007.
Pichersky, E.; Beurle, T. Anal. Biochem. 2002, 302, 305.
Coleman, T. M.; Huang, Faqing. Chem. Biol. 2002, 9, 1227.

Jonsson, S., Ricagno, S., Lindqvist, Y. & Richards, N. G. J. (2004) J. Biol. Chem.
279, 36003-36012.

Jia, Y., Yuan, W., Wodzinska, J ., Park, 0, Sinskey, A. J. & Stubbe, J. (2001)
Biochemistry 40, 1011-1019.

Wu, J ., Bell, A. F., Jaye, A. A. & Tonge, P. J. (2005) J. Amer. Chem. Soc. 127,
8424-8432.

Wong, B. J. & Gerlt, J. A. (2004) Biochemistry 43, 4646-4654.

Rao, K. S., Vander Velde, D., Dwyer, T. M., Goodman, S. I. & Frerrnan, F. E.
(2002) Biochemistry 41, 1274-1284.

0uyang, T.; Walt, D. R. J. Org. Chem. 1991, 56, 3752.

58

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

Raj gharia, V. B.; Priestley, N. D.; Strohl, W. R. Anal. Biochem. 1995, 224, 159.

Jez, J. M.; Ferrer, J.; Bowman, M. E.; Dixon, R. A.; Noel, J. P. Biochemistry
2000, 39, 890.

Walker, K., Fujisaki, S., Long, R. & Croteau, R. Proc. Natl. Acad. Sci, USA.
2002, 99, 12715.

Chen, R, Duran, A. L., Blount, J. W., Sumner, L. W. & Dixon, R. A. (2003)
Phytochemistry (Elsevier) 64, 1013-1021.

Ye, Z. H., Zhong, R., Morrison, W. H., 111 & Himmelsbach, D. S. (2001)
Phytochemistry 57, 1 177-1 185.

Davin, L. B. & Lewis, N. G. (2004) Phytochem. Rev. 2, 257-288.

Klettke, K.; Sanyal, S.; Mutatu, W.; Walker, K. unpublished results 2006,
(manuscript submitted).

Gani, D.; Sanvoisin, J.; O’Donnell, M. E. J. Chem. Soc., Perkin Trans. 1, 2001,
1696-1708.

Z Zhao, Y.; Zhao, J .; Zhou, Y.; Lei, Z.; Li, L.; Zhang, H. New. J. Chem. 2005, 29,
769.

Rosenfeld, 8.; William, A. J. Chem. Edu. 1991, 68, 66.

Kingston, D. G. I. J. Nat. Prod. 2000, 63, 726.

Kingston, D. G. 1. Chem. Comm. 2001, 867.

Stone, G. W & coworkers. J. Am. Coll. Cardio. 2005. 45, 1186 and1201.
Garber, K. J. Natl. Cancer Inst. 2004, 96, 90.

Holzgrabe, U. Pharmazie in unserer Zeit. 2005, 34, 96.

Garber, K. J. Natl. Cancer Inst. 2004, 96, 90.

Nicolaou, K. C.; Nanterrnet, H. U.; Guy, R. K.; Couladouros, E. A.; Sorensen, E.
J. J. Am. Chem. Soc. 1995, 117, 624.

Kingston, D. G. I. J. Nat. Prod. 1999, 62, 1448.

http://www.patentstrom.us/patents/614482.html

59

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

Walker, K.; Croteau, R. Phytochemistry, 2001, 58, 1.

St-Pierre, B.; De Luca, V.; Recent Adv. Phytochem. 2000, 34, 285.
D’Auria, J. C.; Chen, R; Pichersky, E. Plant Physiology 2002, 130, 466.
Grothe, T.; Lenz, R.; Kutcham, T. M. J. Biol. Chem. 2001, 276, 30717.

St. Pierre, B.; Laflamme, P.; Alarco, A. —M.; De Luca, V. Plant Journal 1998, 14,
703.

Stewart Jr, C.; Kang, B. —C.; Liu, K.; Mazourek, M.; Moore, S. L.; Yoo, E. Y.;
Kim, B. —D.; Paran, 1.; Jahn, M. M. The Plant Journal 2005, 42, 675.

Okada, T.; Hirai, M.; Suzuki, H.; Yamazaki, M.; Saito, K. Plant Cell Physiology
2005, 46, 233.

St-Pierre, B.; De Luca, V.; Recent Adv. Phytochem. 2000, 34, 285.

Ma, X.; Koepke, J .; Panjikar, S.; Fritzsch, G.; Stockigt, J. J. Biol. Chem. 2005,
280, 13576.

Carter, P.; Wells, J. A. Science 1987, 23 7, 394.
Carter, P.; Abrahmsen, L.; Wells, J. A. Biochemistry 1991, 30, 6142.

Corey, D. R.; Willet, W. S.; Coombs, G. S.; Craik, C. S. Biochemistry 1995, 34,
11521.

Taylor, G. F.; Thorton, S. S.; Tallent, C. R.; Kepler, J. A. J. Label. Comp.
Radiopharm. 1993, 33, 501.

Holton, R. A. European Patent Application 903 05845.1 , 1990.
Holton, R. A.; Liu, J. A.; Gentile, L. N.; Beidiger, R. J. Presented at the Second
National Cancer Institute Workshop on T axol and T axus, September 1992.

Ojima, 1.; Habus, I.; Zhao, M.; Georg, G. 1.; Jayasinghe, L. R. J. Org. Chem.1986,
56, 1681.

Stefanovic, M.; Lajsic, S. Tetrahedron Lett. 1967, 19, 1777.
Shiina, I.; Saito, K.; Frechard-Ortuno, 1.; Mukaiyama, T. Chemistry Lett. 1998, 3.

Walker, K.; Long, R.; Croteau, R. Proc. Natl. Acad. Sci, USA. 2002, 99, 9166.

60

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

Roh, E. J.; Song, C. E.; Kim, D.; Pae, H. 0.; Chung, H. T.; Lee, K. S.; Chai, K.;
Lee, C. 0.; Un Choi, S. Bioorg. Med. Chem. 1999, 7, 2115.

Xue, M.; Long, B. H.; Fairchild, C.; Johnston, K.; Rose, W. C.; Kadow, J. F.;
Vyas, D. M.; Chen, S. H. Bioorg. Med. Chem. Lett. 2000, 10, 1327.

Ali, S. M.; Hoemann, M. Z.; Aube, J .; Georg, G. 1.; Mitscher, L. A.; Jayasinghe,
L. R. J. Med. Chem. 1997, 40, 236.

Kingston, D. G. I. J. Nat. Prod. 2000, 63, 726.

Ojima, I.; Habus, I.; Zhao, M.; Georg, G. 1.; Jayasinghe, L. R. J. Org. Chem.1986,
56, 1681.

Kingston, D. G. 1.; Chaudhary, A. G.; Gunatilaka, A. A. L.; Middleton, M. L.
Tetrahedron Lett. 1994, 35, 4483.

Baloglu, E.; Kingston, D. G. I. J. Nat. Prod. 1999, 62, 1068.

Gunatilaka, A. A. L.; Chordia, M. D.; Kingston, D. G. I. J. Org. Chem. 1997, 62,
3775.

Shiina, 1.; Saito, K.; Frechard-Ortuno, I.; Mukaiyama, T. Chemistry Lett. 1998, 3.
Long. M. R.; Croteau, R. Biochem. Biphys. Res. Comm. 2005, 338, 410.
Walker, K.; Long, K.; Croteau, R. Proc. Natl. Acad. Sci. 2002, 99, 9166.

Georg, G. I.; Boge, T. C.; Cheruvallath, Z. S.; Harriman, G. C. B.; Hepperle, M.;
Park, H.; Himes, R. H. Bioorg. Med. Chem. Lett. 1994, 4, 335.

Saitoh, K.; Shiina, I.; Mukaiyama, T. Chemistry Lett. 1998, 679.
Rosenfeld, 8.; William, A. J. Chem. Edu. 1991, 68, 66.

Z Zhao, Y.; Zhao, J .; Zhou, Y.; Lei, Z.; Li, L.; Zhang, H. New. J. Chem. 2005, 29,
769.

Gani, D.; Sanvoisin, J.; O’Donnell, M. E. J. Chem. Soc., Perkin Trans. 1, 2001,
1696-1708.

61

 

1111111111111111