Synthesis and antimicrotubule activity of combretatropone derivatives.

Article abstract of DOI:10.1016/S0968-0896(02)00052-4

Combretatropone is a hybrid of combretastatin and colchicine in which the o-methoxyphenol of dihydrocombretastatin A-4 is replaced by an alpha-methoxytropone. Derivatives of combretatropone have been synthesized and evaluated for antimicrotubule activity.

Full text of DOI:10.1016/S0968-0896(02)00052-4

Bioorganic & Medicinal Chemistry 10 (2002) 1895–1903
Synthesis and Antimicrotubule Activity of
Combretatropone Derivatives
Mark E. Janika and Susan L. Baneb,*
aDepartment of Chemistry, SUNY-Binghamton, Binghamton, NY 13902, USA
bDepartment of Chemistry, SUNY-Fredonia, Fredonia, NY 14063, USA
Received 21 September 2001; accepted 20 December 2001
Abstract—Combretatropone is a hybrid of combretastatin and colchicine in which the o-methoxyphenol of dihydrocombretastatin
A-4 is replaced by an a-methoxytropone. Derivatives of combretatropone have been synthesized and evaluated for antimicrotubule
activity. All combretatropones were less active than the corresponding colchicine derivatives, supporting the idea that loss of ligand
conformational entropy upon tubulin binding results in decreased potency for colchicinoid ligands. The structure–activity rela-
tionship of the combretatropone series was different than that of the colchicine series. These data indicate that conformationally
mobile and conformationally rigid colchicinoids do not interact with the receptor site in the same manner. # 2002 Elsevier Science
Ltd. All rights reserved.
Introduction
substituent on the B ring plays at least two important
roles. In the absence of the C-7 substituent, the colchicine
ring system exists in twoconformational isomers with
different axial chiralities. The amide in natural S-colchi-
cine holds the aryl-tropone rings in the aR configuration,
while the amide in unnatural R-colchicine places the
aryl-tropone rings in the inactive aS conformation.8À10
The C-7 substituent alsoaffects the kinetics of colchici-
noids binding to tubulin, presumably through contact
with the protein in the transition state. The presence of an
amine, substituted amine or amide at the C-7 position
Microtubules are the cellular target for a number of useful
chemotherapeutic agents.1 These drugs are generally
divided intothree classes defined by the three known
drug binding domains of tubulin, the major protein
component of microtubules. Drugs that interact with
the Taxol and vinblastine regions of the protein are used
clinically in the treatment of cancer. Molecules that bind
to the colchicine domain of tubulin, however, are not
yet in standard use as antineoplastic agents.2 The promise
of the antiangeogenic compound combrestatin A-4, a
colchicine site ligand, has sparked renewed interest in
colchicine site drugs as potential anticancer drugs.3
raises the activation energy of the association reaction
11,12
by up to5 kcal/mol.
The B ring itself, however,
seems to play no enthalpic role in either the ground or
transition state. Removal of the entire B ring decreases
the affinity of the molecule for tubulin (MTC, 2), but
the loss of activity is attributed to the conformational
properties of the ligand and not directly to binding site
interactions. Thermodynamic analyses of colchicine and
MTC binding totubulin have shwon that the two
ligands have identical binding enthalpies, indicating that
the contacts made between colchicine and tubulin
involve just the A and C rings.13À15
The most studied ligand for the colchicine binding site
on tubulin is colchicine itself (Fig. 1). Several hundred
derivatives and analogues of colchicine (1) have been
tested for antimicrotuble activity.4 Attempts togain a
mechanistic understanding have taken different forms,
from conventional SAR studies to intricate QSAR and
thermodynamic analyses.5À8 The primary pharmaco-
phore appears to consist of the A and C rings of the parent
molecule, but it is also clear that the B ring portion of the
molecule contributes to the efficacy of the drug. The
If the sole role of the B ring is to limit the conformational
mobility of the A–C ring system, structure–activity rela-
tionships in the colchicinoid and MTC series should
have a parallel correlation. For example, substituents
that decrease the activity of the colchicine derivative
*Corresponding author. Tel.: +1-607-777-2927; fax: +1-607-777-
4478; e-mail: sbane@binghamton.edu
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00052-4

1896
M. E. Janik, S. L. Bane / Bioorg. Med. Chem. 10 (2002) 1895–1903
Chemistry
The synthesis of 10-chloro-10-demethoxycombretatro-
pone (9) is shown in Scheme 1. Compounds 5–8 were
previously reported by Andres et al.18 and have been
synthesized here using some procedural modifications.
3-Hydroxybenzaldehyde was protected, reduced and
converted to the phosphonium salt (5) in 63% yield.
Subsequent Wittig condensation of 5 with 3,5-dime-
thoxy-4-benzyloxybenzaldehyde (6) provided a mixture
of cis- and trans-stilbene (7) in 58% yield. Ring expan-
sion of the 3-silyloxyphenol of 7 intoan a-chlorotropone
was accomplished as modified from Andres et al.18 as fol-
lows. Stilbene 7 was subjected tocatalytic hydrogenation
followed by Birch reduction. Methylation of the phenol
was accomplished in high yield with diazomethane rather
than methyl iodide. Subsequent dichlorocyclopropa-
nation, desilylation and epoxidation provided 8 in 30%
overall yield from 7. Acid catalyzed ring expansion of 8
Figure 1. Structures of colchicine and analogues.
should cause an equivalent decrease in the activity of
the MTC derivative. Structure–activity studies of tro-
pone-substituted analogues in both the colchicine and
MTC series have been performed. It was found that the
nature of the C-10 substituent of colchicine has little effect
on its ability to inhibit in vitro microtubule assembly.16 In
contrast, a similar series of MTC derivatives displayed a
wide variation in activity as a function of the tropone
substituent.17
was then achieved with pTSA in benzene toproduce
in 57% yield.
9
Compounds 10–14 were prepared directly from com-
pound 9 (Scheme 2). Combretatropone (4) was produced
in 80% yield by reaction of magnesium methoxide with
9 at room temperature. The remaining alkoxy com-
bretatropones (10–12) were prepared in the following
manner. Compound 9 was reacted separately with either
ethoxymagnesium bromide or propoxymagnesium bro-
mide under reflux to afford the alkoxycombretatropones
10 and 11 in yields of 73 and 76%, respectively. The
isopropoxy derivative 12 was prepared in 35% yield by
the reaction of compound 9 with sodium isopropoxide.
The remaining derivatives shown in Scheme 2 were pre-
pared as follows. 10 - Demethoxy - 10 - thiomethylcom-
bretatropone (13) was produced in 78% yield by the
reaction of 9 with sodium thiomethoxide. Conversion of
compound 9 into the tropone derivative 14 was accom-
plished using sodium borohydride in DMSO in a yield
of 67%. Synthesis of combretatropolone (15) was carried
out by acid catalyzed hydrolysis of 4 in 86% yield.
In order to explore the role of ligand conformational
entropy in the binding of colchicinoids to tubulin, a
structure–activity study of tropone-containing ligands
with additional degrees of conformational freedom was
undertaken. It was previously shown that a hybrid of the
conformationally mobile combretastatin (3) and the tro-
pone-containing colchicine, termed ‘combretatropone’
(4), retained tubulin-binding ability.18 This skeleton
provides a suitable scaffold on which to evaluate the
relative role of conformational entropy in the affinity of
colchicinoids for tubulin. Therefore, in this study, a series
of substituted tropone derivatives of combretatropone
were prepared and their ability toinhibit microtubule
assembly in vitrowas evaluated. The structure–activity
relationship in the combretatropone series followed the
SAR for bicyclic (MTC) compounds rather than the SAR
in the corresponding tricyclic (colchicinoid) molecules.
These data support the notion that conformationally
mobile ligands interact with the colchicine binding site
differently than the conformationally rigid colchicinoids.
The synthesis of the fluorocombretatropones 16–17 and
alkylcombretatropone derivatives 18–21 is shown in
Scheme 3. Reaction of combretatropolone (15) with
diethylaminosulfur trifluoride (DAST) at 0 ꢀC resulted
in the formation of two isomeric products (16 and 17) in
a combined yield of 83%. The fluorocombretatropones
Scheme 1. (a) nBuLi, THF, 58%; (b) H2, Pd/C, EtOH, 96%; (c) Li/NH3, t-BuOH, THF, reflux, 90%; (d) CH2N2, CH2Cl2, 97%; (e) NaO2C2Cl3,
DME, reflux; (f) 20% HF, CH3CN, 53%; (g) mCPBA, CH2Cl2, 67%; (h) pTSA, C6H6, reflux, 57%.

M. E. Janik, S. L. Bane / Bioorg. Med. Chem. 10 (2002) 1895–1903
1897
Table 1. Inhibition of in vitro microtubule assembly by C-11 sub-
stituted combretatropone derivatives
Compound
Substituenta
I50 (mM)b
4
9
OCH3
Cl
OCH2CH3
OCH2CH2CH3
OCH(CH3)
SCH3
H
54.3Æ3.23c
300Æ4.04
148Æ1.06
235Æ3.76
254Æ3.61
64.1Æ2.02
375Æ5.12
175Æ4.82
182Æ3.81
187Æ3.70
92.2Æ2.28
95.4Æ2.56
158Æ2.70
165Æ4.09
180Æ1.41
38.2Æ1.69
174Æ2.08
10
11
12
13
14
15
16
17
18
20
19
21
22
23
24
OH
F
Fd
Scheme 2. (a) Mg, I2, CH3OH, 66%; (b) EtOH, EtMgBr, THF, 73%;
(c) PrOH, PrMgBr, THF, 76%; (d) Na, iPrOH, 38%; (e) NaSCH3,
H2O, CH3OH, 78%; (f) NaBH4, DMSO, 67%; (g) 2 N HCl, CH3OH,
reflux, 86%.
CH3
CH3
CH2CH3
d
d
CH2CH3
NH2
NHCH3
N(CH3)2
aSubstituent on the tropone ring.
bI50 is the concentration of ligand required to effect a 50% reduction in
microtubule polymerization.
cErrors are expressed as standard deviation of the mean. Each com-
pound was tested a minimum of four times.
dIsomer of compound in which the relative location of the C-10 and
C-11 groups are interchanged; see Scheme 3.
vitro assembly of microtubule protein (Table 1). The I50
value is the concentration of ligand required to decrease
the extent of microtubule polymerization to 50% of the
control.
Scheme 3. (a) DAST, CH2Cl2, 83%; (b) CH3MgBr, THF, 0 ꢀC;
(c) CH3CH2MgBr, THF, 0 ꢀC.
It has been shown that both the absolute and relative
magnitude of I50 values for antimicrotubule drugs are
dependent on the conditions employed in the assay.19
The assay conditions and procedures used in different
laboratories must be taken into account when data from
different sources are compared. The assay conditions
used toevaluate the MTC analogues are significantly
different from the standard assay employed in our
laboratory. In order to make a precise comparison
between the bicyclic and combretatropone SAR, some of
the previously studied MTC derivatives were synthesized
in our laboratory and subjected to our standard assay
conditions. These values are included in Table 2.
16 and 17 were then reacted with the appropriate Grignard
reagent toproduce alkylcombretatropones
18–21. Reac-
tion of compound 16 with methyl magnesium bromide in
THF at 0 ꢀC produced 10-demethoxy-10-methylcom-
bretatropone (18) in a yield of 51%. Similarly, reaction
of 16 with ethyl magnesium bromide in THF at 0 ꢀC
produced 10-demethoxy-10-ethylcombretatropone (19)
in a yield of 59%. The remaining two isomeric alkyl-
combretatropones (20 and 21) were prepared in an
analogous manner from fluorocombretatropone 17 in
yields of 48 and 56%, respectively.
The synthesis of the amine derivatives (22–24) was
accomplished in the following manner. Combreta-
tropone (4) was reacted with ammonia in a pressure
bottle for 12 h at 85 ꢀC to produce 10-amino-10-de-
methoxycombretatropone (22) in a yield of 82%. 10-
Demethoxy-10-methylaminocombretatropone (23) was
prepared in a yield of 84% through the reaction of 4
with methylamine for 24 h at 35 ꢀC in a pressure bottle.
Reaction of 4 with dimethylamine at 80 ꢀC for 3 days in
a pressure bottle afforded 10-demethoxy-10-dimethyl-
aminocombretatropone (24) in a yield of 65%.
There are a few trends in the polymerization data that
are notable. First, the combretatropones as a class are
significantly less active than the analogous bicyclic col-
chicine derivatives, which are themselves less active than
C-10 substituted colchicine analogues. The most active
combretatropone is 23, which has an I50 value of 38 mM.
In contrast, the I50 values obtained under similar con-
ditions for the colchicine derivatives corresponding to
combretatropones 4, 9, 10, 13, 14, 16, 18, and 19 fall
between 2 and 8 mM. The least active colchicinoids,
which correspond to the propoxy and isopropoxy com-
bretatropones, have I50 values of 11 and 22mM, respec-
tively.16 The I50 values for the analogous MTC derivatives
tend tobe in between those found for colchicine and the
combretatropones (Table 2 and ref 17). These data
support the conclusion of Menendez et al.13 that the
Biological Results and Discussion
The antimicrotubule activities of the combretatropone
analogues were assessed by their ability to inhibit in

1898
M. E. Janik, S. L. Bane / Bioorg. Med. Chem. 10 (2002) 1895–1903
Table 2. Inhibition of in vitro microtubule assembly by selected
colchicinoids
methylene bridge in sufficient toallow the ‘isocolchicine-
like’ isomer to interact with tubulin in a ‘colchicine-like’
manner. So, with respect to the tropone ring isomers,
the entropic loss may be counterbalanced by the
enthalpic gain.
Conclusions
Substituent
Colchicine (I)
MTC (II)
Combretatropone (III)
A series of substituted combretatropones have been
synthesized and evaluated for antimicrotubule activity.
The SAR results support the idea that loss of ligand
conformational entropy upon tubulin binding results in
decreased potency for colchicinoid ligands. These results
also support the notion that conformationally mobile
colchicinoids may interact with the receptor site differ-
ently than conformationally restricted colchicinoids.14
This possibility needs to be considered in 3-D-QSAR
studies of colchicine site ligands.
I50, mMa
I50, mMb
I50, mM
OCH3
NHCH3
SCH3
CH2CH3
Cl
5.1
2.0
2.5
4.0
4.0
6.9Æ2.0c
3.5Æ1.0
11Æ1.7
63Æ3.0
ꢁ500d
54
38
64
160
300
aFrom ref 16.
bDetermined in this laboratory.
cErrors are expressed as standard deviation of the mean. Each com-
pound was tested a minimum of four times.
dValue from ref 17.
Experimental
conformational entropy of the ligand is an important
factor in the potency of colchicinoid drugs.
All NMR spectra were recorded on a Bruker AM 360
spectrometer in deuterochloroform. Chemical shifts are
reported in ppm (d) and J coupling constants are reported
in Hz. Infrared spectra were measured in spectroscopy
grade chloroform on a Perkin-Elmer 1600 series FTIR.
UV spectra were obtained on a Hewlett Packard 8453
diode ray spectrometer. Extinction coefficients were
determined in PME buffer (PME buffer=100mM Pipes,
2 mM EGTA, 1 mM MgSO4, pH 6.90 at 23 ꢀC) and are
reported as MÀ1 cmÀ1. Melting points were determined
using an Electrothermal IA6304 open capillary melting
point apparatus and are uncorrected. EI high-resolution
mass spectra were obtained by Dr. Gordon Nicol at the
University of Delaware.
Differences in conformational mobility alone, however,
cannot explain the trends in the combretatropone data.
Table 2 compares the antimicrotubule activity of the
analogous derivatives in the colchicine, MTC and com-
bretatropone series. The nature of the tropone sub-
stituent has little effect on the potency of colchicinoids.
In contrast, the biological activity of the MTC and
combretatropone derivatives varied considerably with
the tropone substituent. The most dramatic difference is
seen in the chlorotropone series. 10-Chloro-10-deme-
thoxycolchicine is highly potent—in fact, it is slightly
more active than colchicine itself. The corresponding
bicyclic and combretatropone compounds are con-
siderably less active than the parent molecules. Fur-
thermore, performing an isosteric substitution of ethyl
for methoxy produced little effect on activity in the col-
chicine series but decreased activity in the bicyclic and
combretatropone analogues. It appears that steric and
electrostatic properties of the substituent play a role in
its antimicrotubule activity in the absence of the tethering
B ring. This observation is an important consideration
in 3-D-QSAR analyses of colchicine site ligands. 3-D-
QSAR analyses using CoMFA require the generation of
alignment rules.20,21 Our results indicate that an align-
ment rule generated for conformationally restricted
ligands may not be suitable for the conformationally
mobile ligands.
All reactions were conducted under a protective atmo-
sphere of either dry nitrogen or argon gas unless otherwise
stated. All solvents were dried and distilled prior to use.
All reagents were obtained pure from Aldrich Chemical
Corporation or were purified according to known lit-
erature purification procedures before use. Column
chromatography was performed using Baker silica gel
60–200 or 200–400 mesh. A chromatatron (Harrison
Research) was used for radial chromatrography.
Purity criteria were established using twodiverse systems
of HPLC. The first method performed was reverse
phase chromatography using a Waters Bondapak C18
column (3.9Â300 mm). The solvent system employed
with this technique was methanol/acetonitrile/water,
12.5:37.5:50. The second method used was liquid/solid
adsorption chromatography. The adsorbent used was
Keystone Silica 60 (particle size 10 nanometers, pore size
60 A). The dimensions of the column were 40Â5 mm, and
the solvent system employed was 70:30 hexanes/iso-
propanol. The detection method for both systems was
an ISCO V4 variable wavelength detector, connected to
a Spectra-Physics SP4270 integrator. No impurities
were detected in any combretatropone tested in the
biological assays.
A somewhat surprising outcome from the combreta-
tropone series is found in compounds 17, 20, and 21
(Table 1). The structural isomers in each pair of com-
pounds are analogous to the colchicine and iso-
colchicine pair for the parent molecule. In the tricyclic
series, isocolchicine isomers are inactive as inhibitors of
microtubule assembly at sub-millimolar concentra-
tions.22,23 In the combretatropone series, though, the
activities of the two isomers are essentially identical. It
appears that the conformational mobility imparted by the

M. E. Janik, S. L. Bane / Bioorg. Med. Chem. 10 (2002) 1895–1903
1899
3-(tert-Butyldimethylsiloxy)benzyl-triphenylphosphonium
bromide (5). The procedure of Andres et al.18 was used.
Briefly, 3-hydroxybenzaldehyde was treated with imida-
zole and t-butyldimethylchlorosilane in THF to produce
3-(tert-butyldimethylsiloxy)benzaldehyde in 87% yield.
The aldehyde was reduced to3-( tert-butyldimethylsi-
loxy)benzyl alcohol by catalytic hydrogenation (97%).
3-(tert-Butyldimethylsiloxy)benzyl bromide was pre-
pared in two steps by reaction of the benzyl alcohol with
trifluoroacetic anhydride (96%) followed by LiBr in THF
(87%). Treatment of the benzyl bromide with triphenyl-
phosphine in xylenes yielded compound 5 (88%). The
structures of all products were confirmed by 1H NMR.
and lithium wire (1.40 g) was then added tothe reaction.
The hydrogenated stilbene (3.12 g, 8.04 mmol) was
added and the reaction allowed to reflux for 1 h. The
reaction was quenched and worked up to yield 2.81 g
(90%) of a brown oil. The product was confirmed by 1H
NMR.
1,1,2,2-Tetramethyl-1-sila-1-{5-[2-(3,4,5-trimethoxyphe-
nyl)ethyl]cyclohexa-1,4-dienyloxy}propane (reaction d).
The cyclohexadienyl compound (3.05 g, 7.9 mmol) was
dissolved in 20 mL of methylene and cooled to 0 ꢀC with
an ice bath. To this solution was added an excess of an
etherial diazomethane solution (220 mL). The solution
was allowed to stir for 6 h at 0 ꢀC. The ice bath was then
removed and the solution allowed to stir at room tem-
perature for an additional 24h. The solution was then
evaporated and purified to yield 3.07g (97%) a pure light-
yellow oil. The compound was confirmed by 1H NMR.
4-Benzoyl-3,5-dimethoxybenzaldehyde (6). Syringe alde-
hyde (3.12 g, 17.1 mmol) was dissolved in 30 mL of THF
and to the stirred solution was added potassuim
hydroxide (1.07 g, 18.8 mmol). The resulting solution
was refluxed for 24 h and five crystals of 18-crown-6 and
benzyl bromide (2.24 mL, 18.8 mmol) were then added
to the solution and refluxed continued for 24 h. The
reaction was cooled to room temperature, worked up
and purified using flash chromatography to yield 3.30 g
of a pure yellow liquid. The product was confirmed by
1H NMR.
7,7-Dichloro-3-[2-(3,4,5-trimethoxyphenyl)ethyl]bicyclo
[4.1.0]hept-3-en-1-ol (reactions e–f). Sodium trichloro-
acetate (1.10 g, 6.38 mmol) was added to a 25 mL two-
neck round-bottom flask. To this flask was added the
methylated dihydroaromatic (1.22 g, 3.1 mmol in 3 mL
of dimethoxyethane). The resulting solution was then
refluxed for 5 h. The reaction was cooled to room tem-
perature and worked up. The crude product was then
dissolved in 30 mL of acetonitrile and to this solution
was added 6 mL of a 20% HF solution. The solution
was allowed to stir at room temperature for 12 h,
worked up and purified using flash chromatography to
yield 0.590 g (53%) of a yellow oil. The product was
1-(3-{(1Z and 1E)-2-[3,5-Dimethoxy-4-(phenylmethoxy)-
phenyl]vinyl}phenoxy)-1,1,2,2-tetramethyl-1-silapropane
(7). Phosphonium bromide 5 (6.2 g, 11.1 mmol) was
dissolved in 75 mL of dry THF and cooled to 0 ꢀC. To
this solution was added n-BuLi (7.6 mL, 12.2 mmol) and
the resulting solution was stirred at 0 ꢀC for 1.75 h.
Aldehyde 6 (2.52 g, 9.23 mmol) was then added and the
reaction was allowed to stir for 2 h at room temperature.
The mixture was diluted with ether (40 mL), and washed
first with NaHSO3 (20 mL), followed by NaHCO3
(20 mL), and finally with water (35 mL). The organic
layer was dried with sodium sulfate, filtered and evapo-
rated to yield a dark-yellow oil. The product was purified
using flash chromatography (80:20 hexane/acetone) to
yield 2.57 g (58%) of a clear oil. The structure of the
1
confirmed by H NMR.
Compound 8 (reaction g). The cyclopropyl alcohol
(0.62 g, 1.6 mmol) was dissolved in 50 mL of dry
methylene chloride and to this solution was added
recrystalized MCPBA (0.432 g, 1.5 mmol). The reaction
was allowed to stir at room temperature for 3.5 h,
worked up and purified to yield 0.430 g (67%) of a yel-
1
low oil. The product was confirmed by H NMR.
1
compound was confirmed by H NMR.
11-Chloro-11-demethoxycombretatropone (9). Toa stirred
solution of epoxide (8, 0.42 g, 1.1 mmol) in 40 mL of
distilled benzene was added recrystalized pTSA (15 mg).
The resulting solution was allowed to reflux for 5 h. The
reaction was cooled to room temperature and diluted
with ether (30 mL). The mixture was washed with water
(25 mL) and then brine (20 mL). Dried the organic layer
with sodium sulfate, filtered and stripped of solvent to
yield a brown oil. The product was purified first by flash
chromatography followed by radial chromatography
(90:10 CH2Cl2/acetone) to yield 0.205 g (57%) of a
white solid: e323=7.01Â103; mp 114–116 ꢀC. 1H NMR d
7.68 (d, 1H, H-12, J=9.3 Hz), 7.16 (s, 1H, H-9), 6.89 (d,
1H, H-14, J=9.3 Hz) 6.79 (t, 1H, H-13, J=9.3 Hz), 6.34
(s, 2H, H-1 and H-5), 3.81 (s, 9H), 2.83 (s, 4H, H-7 and
H-8); 13C NMR d 179.3, 153.2, 150.0, 148.7, 137.8,
137.0, 136.6, 135.2, 134.7, 129.9, 105.3, 60.7, 56.0, 42.3,
37.6. IR 3052–2812, 1720, 1620, 1595cmÀ1. LR-MS (m/z):
336 (10), 335 (5), 334 (M+, 20), 182 (15), 181 (100), 148
(10), 135 (10), 69 (40). HRMS calcd for C18H19O4Cl:
334.0970. Found: 334.0956.
8,8-Dichloro-4-oxa-3-[2-(3,4,5-trimethoxyphenyl)ethyl]-
tricyclo[5.1.0.0<3,5>]octan-1-ol (8). A modification of
the procedure of Andres et al.18 was employed. The letters
b–g refer tothe individual reactions shown in Scheme 1.
2,6-Dimethoxy-4-{2-[3-(1,1,2,2-tetramethyl-1-silapropoxy)
phenyl]ethyl}phenol (reaction b). The mixture of E+Z
stilbenes 7 (3.24 g, 6.8 mmol) was dissolved in 15 mL of
ethanol in a pressure bottle. To this solution was added
10 mol% Pd/C and the stilbene was allowed to react at
45 psi/H2 over a 24-h time period. The solution was
then vacuum filtered and stripped of solvent, yielding
2.52 g (96%) of a pure light-brown oil. The compound
1
was confirmed by H NMR.
2,6-Dimethoxy-4-{2-[5-(1,1,2,2-tetramethyl-1-silapropoxy)
cyclohexa-1,4-dienyl]ethyl}phenol (reaction c). A 1000 mL
three-neck round-bottom flask was cooled to À78 ꢀC
and ammonia (230 mL), t-butanol (20 mL) and 100 mL
of dry THF were added to the flask. Stirring was initiated

1900
M. E. Janik, S. L. Bane / Bioorg. Med. Chem. 10 (2002) 1895–1903
2 - Methoxy - 6- [2 - (3,4,5 - trimethoxyphenyl)ethyl]cyclo-
hepta-2,4,6-trien-1-one (combretatropone, 4). Oven dried
magnesium metal (0.057 g, 24 mmol) was added to a
two-neck round-bottom flask. Dry methanol (6 mL) and
five crystals of iodine were then added to the flask. The
reaction was then refluxed for 1 h, at which point the
magnesium had dissolved. Chlorocombretatropone 9
(80 mg, 0.24 mmol) was dissolved in 3 mL of dry
methanol and syringed into the two-neck flask. The
reaction was allowed to stir at room temperature for
7.5 h. The reaction was diluted with chloroform (20 mL)
and neutralized with 1 N HCl (10 mL). The layers were
seperated and the aqueous layer extracted five times
with chloroform (5Â15 mL). Combined organic layers,
dried with sodium sulfate, and stripped of solvent to
yield a dark-yellow oil. The product was purified first by
flash chromatography followed by radial chromato-
graphy (80:20 CH2Cl2/acetone) to yield 52 mg (66%) of
a light yellow oil: e333=7.20Â103. 1H NMR d 7.16 (s, 1H,
H-9), 6.93 (t, 1H, H-13, J=10.2 Hz), 6.70 (d, 1H, H-14,
J=10.2 Hz), 6.62(d, 1H, H-12, J=10.2 Hz), 6.33 (s, 2H,
H-1 and H-5), 3.92 (s, 3H, 11-OCH3), 3.83 (s, 9H), 2.81
(m, 4H, H-7 and H-8); 13C NMR d 179.9, 160.7, 148.9,
146.7, 132.2, 131.7, 127.1, 126.4, 124.1, 107.2, 101.1,
56.5, 51.8, 51.7, 38.3, 33.2. IR 3040–2820, 1720, 1627,
1600 cmÀ1. LR-MS (m/z): 331 (15), 330 (M+, 35), 315
(20), 312 (10), 300 (35), 182 (10), 181 (100), 148 (10).
HRMS calcd for C19H22O5: 330.1462. Found: 330.1458.
solution was slowly syringed in and the resulting solu-
tion was allowed to stir at room temperature for 20 min.
Chlorocombretatropone 9 (50 mg, 0.15 mmol) was dis-
solved in 2 mL of THF and then syringed into the reac-
tion flask. The reaction was then allowed to stir at room
temperature for 24 h. The solution was diluted with
chloroform (20 mL) and water (15 mL) and the aqueous
layer was neutralized with 0.5 N HCl (1 mL). The layers
were separated and the aqueous layer was extracted
four additional times with chloroform. The product was
purified by flash chromatography followed by radial
chromatography (80:20 CH2Cl2/acetone) to yield 40 mg
(76%) of a white solid: e323=6.49Â103; mp 103–105 ꢀC.
1H NMR d 7.20 (s, 1H, H-9), 6.93 (t, 1H, H-13, J=
10.4 Hz), 6.69 (d, 1H, H-14, J=10.3 Hz), 6.65 (d, 1H,
H-12, J=10.4 Hz), 6.36 (s, 2H, H-1 and H-5), 4.02 (t,
2H, 11-OCH2CH2CH3, J=6.8 Hz), 3.81 (s, 9H), 2.83
(m, 4H, H-7 and H-8), 1.98–1.87 (m, 2H, 11-OCH2
CH2CH3), 1.05 (t, 3H, 11-OCH2CH2CH3, J=7.1 Hz);
13C NMR d 179.7, 164.7, 153.3, 150.8, 136.6, 136.2,
135.3, 131.7, 130.5, 112.5, 105.5, 70.8, 60.9, 56.1, 42.7,
37.7, 22.0, 10.4. IR 3041–2830, 1714, 1627, 1595 cmÀ1
.
LR-MS (m/z): 359 (5), 358 (M+, 20), 182 (20), 181
(100), 148 (15). HRMS calcd for C21H26O5: 358.1772.
Found: 358.1794.
11-Isopropoxy-11-demethoxycombretatropone (12). Into
a two-neck round-bottom was added distilled iso-
propanol (3 mL) and sodium metal (0.030 g, 1.35 mmol).
The resulting solution was then refluxed for 45 min at
which point the sodium had dissolved. Chlorocom-
bretatropone 9 (0.030 g, 0.090 mmol) was dissolved in
2 mL of isopropanol and syringed into the reaction flask
and the reaction was allowed to stir at room tempera-
ture for 20 min. The reaction was diluted with methyl-
ene chloride (25 mL) and water (15 mL) and the
aqueous layer was neutralized with 0.5 N HCl (0.5 mL).
The layers were separated and the aqueous layer was
extracted three additional times with methylene chlor-
ide. The organic layers were combined, dried with
sodium sulfate, filtered and stripped of solvent. The
product was purified first with flash chromatography
followed by radial chromatography (80:20 CH2Cl2/ace-
11-Ethoxy-11-demethoxycombretatropone (10). Toa
round-bottom flask was added 3 mL of dry THF and
excess dry ethanol (0.5 mL). To this stirred solution
(3.74 mL, 3.75 mmol) of a 1 M ethylmagnesium bromide
solultion was slowly syringed in and the resulting solu-
tion was allowed to stir at room temperature for 20 min.
Chlorocombretatropone 9 (50 mg, 0.15 mmol) was dis-
solved in 2 mL of dry THF and syringed into the reac-
tion flask. The reaction was allowed to stir at room
temperature for 12 h. The reaction was then diluted with
chloroform (15 mL) and water (15 mL). The aqueous
layer was neutralized with 0.5 N HCl (1 mL). The layers
were separated and the aqueous layer was extracted
four times with chloroform. The organic extracts were
combined, dried with sodium sulfate, filtered and strip-
ped of solvent. The product was first purified by flash
chromatography followed with radial chromatography
(80:20 CH2Cl2/acetone) to yield 37 mg (73%) of a col-
orless oil: e334=6.98Â103. 1H NMR d 7.18 (s, 1H, H-9),
6.92 (t, 1H, H-13, J=10.5 Hz,), 6.69 (d, 1H, H-14,
J=10.5 Hz), 6.63 (d, 1H, H-12, J=10.5 Hz), 6.34 (s, 2H,
H-1 and H-5), 4.12 (q, 2H, 11-OCH2CH3, J=7.1 Hz),
3.80 (s, 9H), 2.82 (m, 4H, H-7 and H-8), 1.51 (t, 3H, 11-
OCH2CH3, J=7.0 Hz); 13C NMR d 175.4, 160.1, 148.8,
146.3, 133.5, 132.2, 131.8, 127.2, 126.1, 107.9, 101.1,
60.4, 56.4, 51.7, 38.3, 33.3, 9.9. IR 3040–2810, 1720,
1627, 1600 cmÀ1. LR-MS (m/z). 345 (5), 344 (M+, 10),
316 (10), 207 (15), 182 (15), 181 (100), 148 (10). HRMS
calcd for C20H24O5: 344.1626. Found: 344.1637.
tone) to yield 12 mg (38%) of a white solid: "324
=
4.36Â103; mp 99–102 ꢀC. H NMR d 7.18 (s, 1H, H-9),
6.92 (t, 1H, H-13, J=10.5 Hz), 6.69 (d, 1H, H-14,
J=10.5 Hz), 6.67 (d, 1H, H-12, J=10.4 Hz), 6.36 (s, 2H,
H-1 and H-5), 4.67 (m, 1H, 11-OCH(CH3)2), 3.81 (s, 9H),
2.83 (m, 4H, H-7 and H-8), 1.42 (d, 6H, 11-OCH(CH3)2,
J=6.1 Hz), 13C NMR d 180.1, 163.6, 153.1, 150.5,
136.5, 136.0, 132.3, 130.7, 129.1, 114.2, 105.4, 71.4, 60.7,
56.5, 42.5, 37.5, 21.5. IR 3040–2820, 1720, 1627,
1600cmÀ1. LR-MS (m/z): 359 (5), 358 (M+, 20), 316 (10),
301 (15), 285 (20), 197 (15), 182 (20), 181 (100), 148 (5).
HRMS calcd For C21H26O5: 358.1772. Found: 358.1758.
1
11-Demethoxy-11-thiomethylcombretatropone (13). In a
round-bottom flask, sodium methane thiolate (15 mg,
2.0 mmol) was dissolved in water (1mL) and stirring was
initiated. Chlorocombretatropone 9 (26 mg, 0.078 mmol)
was dissolved in methanol (3 mL) and then syringed into
the reaction flask. The reaction was allowed to stir at
room temperature for 4 h. The reaction was then diluted
11-Demethoxy-11-propoxycombretatropone (11). Toa
round-bottom flask was added 3 mL of dry THF and
excess dry 1-propanol (0.4 mL). To the stirred solution,
1.9 mL (3.75 mmol) of a 2 M propylmagnesium bromide

M. E. Janik, S. L. Bane / Bioorg. Med. Chem. 10 (2002) 1895–1903
1901
with methylene chloride (25 mL) and water (15 mL) and
the aqueous layer was acidified with 0.5 N HCl (0.5 mL).
The layers were separated and the aqueous layer was
extracted three additional times with methylene chlor-
ide. The organic extracts were combined, dried with
sodium sulfate, filtered and evaporated. Purification of
the product was first by flash chromatography followed
by radial chromatography (60:40 hexanes/acetone) yield-
ing 21mg (78%) of a light-yellow oil: e346=8.87Â103. 1H
NMR d 7.04 (s, 1H, H-9), 6.97 (d, 1H, H-12, J=9.8 Hz),
6.90 (t, 1H, H-13, J=9.8 Hz), 6.74 (d, 1H, H-14,
J=9.8 Hz), 6.35 (s, 2H, H-1 and H-5), 3.81 (s, 9H), 2.84
(s, 4H, H-7 and H-8), 2.37 (s, 3H, 11-SCH3); 13C NMR
d 178.0, 155.6, 148.9, 146.3, 132.3, 131.6, 129.8, 128.0,
126.8, 121.7, 101.1, 56.5, 51.7, 38.2, 33.3, 10.9. IR 3030–
2840, 1670, 1620, 1590 cmÀ1. LR-MS (m/z): 348 (5), 347
(10), 346 (M+, 30), 341 (20), 339 (10), 315 (15), 299 (20),
230 (10), 182 (15) 181 (100), 148 (10), 137 (15). HRMS
calcd for C19H22O4S: 346.1237. Found: 346.1236.
OH); 13C NMR d 177.0, 155.6, 153.1, 139.3, 136.5, 136.0,
128.6, 126.9, 124.0, 122.0, 105.3, 60.7, 56.0, 43.4, 38.3. IR
3564–3346, 3052–2825, 1714, 1610 cmÀ1. HRMS calcd
for C19H23O5: 316.1310. Found: 316.1320.
11-Fluoro-11-demethoxycombretatropone and 10-fluoro-
10-demethoxyisocombretatropone (16and 17).
Com-
bretatropolone (15, 30 mg, 0.095 mmol) was dissolved in
dry methylene chloride (3 mL) and cooled to 0 ꢀC. Di-
ethylaminosulfur trifluoride (38 mL, 0.285 mmol) was
then slowly syringed into the reaction and the solution
allowed to stir at room temperature for 7 h. The reaction
was then diluted with methylene chloride (40 mL) and
washed twotimes with water (2 Â10mL). The organic
layer was dried with sodium sulfate, filtered and evapo-
rated. The product was purified first by flash chromato-
graphy followed by radial chromatography (50:50 ethyl
acetate/petroleum ether) to yield 13 mg of a white solid
(16) and 12 mg of a second white solid (17) (83% overall
yield): Compound 16: e328=7.24Â103; mp 104–107 ꢀC.
1H NMR d 7.24 and 7.19 [dd, 1H, H-12, J=22 Hz (F-
H), J=10.1 Hz (H-H], 7.21 (t, 1H, H-13, J=10.1 Hz),
7.12 (s, 1H, H-9), 6.83 (d, 1H, H-14, J=10.1 Hz), 6.37
(s, 2H, H-1 and H-5), 3.83 (s, 6H, 2-OCH3 and 4-OCH3),
3.81 (3H, 3-OCH3), 2.86 (s, 4H, H-7 and H-8); 13C
NMR d 177.6, 163.7, 153.2, 138.4, 138.2, 136.8, 135.2,
131.0, 122.3, 122.0, 105.4, 60.7, 56.0, 42.2, 37.1. IR
3040–2835, 1720, 1630, 1610 cmÀ1. LR-MS (m/z): 319
(5), 318 (M+, 45), 290 (25), 182 (10), 181 (100), 148 (5),
109 (10). HRMS calcd for C19H22O4F: 318.1266.
Found: 318.1271. Compound 17: e320=4.34Â103; mp
109–112 ꢀC. 1H NMR d 7.25 [d, 1H, H-9, J=26 Hz (F–H)],
7.16 (t, 1H, H-13, J=9.9 Hz), 7.11 and 7.08 [dd, 1H, H-
12, J=9.9 Hz (H–H) and 7.9 Hz (F–H)], 6.89 (d, 1H, H-14,
J=9.9 Hz), 6.35 (s, 2H, H-1 and H-5), 3.82 (s, 9H), 2.87 (s,
4H, H-7 and H-8); 13C NMR d 177.2, 164.2, 153.2,
139.5, 139.3, 135.4, 129.8, 129.6, 118.7, 118.3, 105.4,
60.7, 56.0, 42.8, 37.2. IR 3040–2810, 1725, 1640,
1610 cmÀ1. LR-MS (m/z): 319 (15), 318 (M+, 55), 305
(30), 301 (20), 287 (40), 207 (10), 182 (15), 181 (100), 148
(10), 109 (15). HRMS calcd for C19H22O4F: 318.1266.
Found: 318.1250.
11-Demethoxycombretatropone (14). In a round-bottom
flask, chlorocombretatropone 9 (17 mg, 0.051 mmol) was
dissolved in distilled DMSO (2 mL) and stirring initiated.
To this solution was added sodium borohydride (5.1 mg,
0.134 mmol). The reaction was allowed to stir at room
temperature for 6 h. The reaction was then diluted with
methylene chloride (15 mL) and water (10 mL). The
layers were separated and the aqueous layer extracted
three additional times with methylene chloride. Com-
bined the organic extracts, dried with sodium sulfate,
filtered and evaporated. The product was purified first
by flash chromatography followed by radial chromato-
graphy (90:10 ethyl acetate/hexanes) yielding 10 mg
1
(67%) of a colorless oil: e312=6.19Â103. H NMR d
7.05 (s, 1H, H-9), 7.03 (d, 1H, H-11, J=11.5 Hz), 7.01
(t, 1H, H-12, J=11.5 and 11.3 Hz), 6.86 (t, 1H, H-13,
J=11.5 and 11.3 Hz), 6.82 (d, 1H, H-14, J=11.3 Hz),
6.35 (s, 2H, H-1 and H-5), 3.82 (s, 9H), 2.83 (m, 4H, H-
7 and H-8); 13C NMR d 187.1, 153.1, 150.2, 142.0,
141.0, 137.9, 136.6, 135.6, 135.4, 133.1, 105.4, 60.7, 56.0,
42.4, 37.1. IR 3020–2835, 1650, 1610, 1595 cmÀ1. LR-
MS (m/z): 301 (5), 300 (M+, 40), 285 (25), 269 (20), 194
(10), 182 (15), 181 (100), 148 (10), 91 (10). HRMS calcd
for C18H20O4: 300.1360. Found: 300.1361.
11-Demethoxy-11-methylcombretatropone (18). Fluoro-
combretatropone 16 (30 mg, 0.094 mmol) was dissolved
in dry THF (5 mL) and cooled to 0 ꢀC. Tothe cooled
solution was slowly added 3 M methylmagnesium bro-
mide (130 mL, 0.188 mmol). The reaction was allowed to
stir at 0 ꢀC for 45 min. The reaction was diluted with
methylene chloride (30 mL) and water (10 mL) and the
aqueous layer was acidified with 0.5 N HCl (1 mL). The
layers were separated and the aqueous layer was extrac-
ted three additional times with methylene chloride. The
organic extracts were combined, dried with sodium sul-
fate and evaporated. Purification of the product was
first by flash chromatography followed by radial chro-
matography (90:10 CH2Cl2/acetone) yielding 15 mg
(51%) of a white solid: e326=8.65Â103; mp 98–101 ꢀC.
1H NMR d 7.23 (s, 1H, H-9), 7.03 (d, 1H, H-14,
J=9.3 Hz), 6.99 (t, 1H, H-13, J=9.3 Hz), 6.71 (d, 1H, H-
12, J=9.3 Hz), 6.34 (s, 2H, H-1 and H-5), 3.82 (s, 9H),
2.83 (s, 4H, H-7 and H-8), 2.29 (s, 3H, 11-CH3); 13C
NMR d 186.5, 153.1, 151.2, 147.7, 139.1, 138.2, 137.7,
Combretatropolone (15). Combretatropone (4, 60 mg,
0.18 mmol) was dissolved in a 1:1 ratio of methanol/2 N
HCl (5 mL of each solvent). Stirring was initiated and
the solution was refluxed for 16 h. The reaction was
cooled to room temperature and diluted with methylene
chloride (25 mL) and water (15 mL). The aqueous layer
was neutralized with 5% sodium bicarbonate (5 mL)
and the layers were separated. The aqueous layer was
extracted four additional times with methylene chloride
and the organic extracts were combined, dried with
sodium sulfate, filtered and evaporated. Purification of
the product was first by flash chromatography followed
by radial chromatography (80:20 CH2Cl2/acetone) to
yield 49 mg (86%) of a yellow solid: e330=1.02Â104; mp
107–111 ꢀC. H NMR d 7.49 (s, 1H, H-9), 7.41 (d, 1H,
1
H-14 J=9.3 Hz), 7.37 (t, 1H, H-13, J=9.3 Hz), 6.94 (d,
1H, H-12, J=9.4 Hz), 6.33 (s, 2H, H-1 and H-5), 3.80 (s,
9H), 2.89 (m, 4H, H-7 and H-8), 1.66 (broad s, 1H, 11-

1902
M. E. Janik, S. L. Bane / Bioorg. Med. Chem. 10 (2002) 1895–1903
135.8, 135.3, 130.9, 105.5, 60.7, 56.0, 42.6, 37.5, 23.1. IR
3041–2830, 1720, 1610 cmÀ1. LR-MS (m/z): 315 (5), 314
(M+, 10), 182 (15), 181 (100), 148 (15). HRMS calcd for
C20H25O4: 314.1516. Found: 314.1519.
2.82 (m, 4H, H-7 and H-8), 2.70 (m, 2H, 11-CH2CH3
J=7.2Hz), 1.15 (t, 3H, 11-CH2CH3, J=7.2Hz); 13C
NMR d 184.3, 149.0, 148.9, 146.1, 140.1, 132.3, 132.1,
131.7, 126.8, 125.4, 101.1, 56.5, 51.8, 35.0, 32.8, 24.5,
9.7. IR 3045–2825, 1714, 1610, 1595cmÀ1. LR-MS (m/z):
329 (5), 328 (M+, 10), 221 (10), 182 (20), 181 (100), 148
(10). HRMS calcd for C21H27O4: 328.1672. Found:
328.1684.
11-Ethyl-11-demethoxycombretatropone (19). Fluoro-
combretatropone 16 (28 mg, 0.088 mmol) was dissolved
in dry THF (4mL) and cooled to 0ꢀC. Tothe cooled
solution was slowly added 1 M ethylmagnesium bromide
(0.415 mL, 0.194 mmol) and the reaction was allowed to
stir at 0 ꢀC for 10 min. The ice bath was then removed
and the reaction allowed to stir at room temperature for
an additional 10 min. The reaction was diluted with
methylene chloride (25 mL) and water (10 mL) and the
aqueous layer was acidified with 0.5 N HCl (0.5 mL).
The layers were separated and the aqueous layer was
extracted three additional times with methylene chloride.
The organic extracts were combined, dried with sodium
sulfate and evaporated. The product was purified first
by flash chromatography followed by radial chromato-
graphy (98:2 CH2Cl2/acetone) to yield 17 mg (59%) of a
11 - Amino - 11 - demethoxycombretatropone (22). In a
pressure tube, combretatropone (4, 22 mg, 0.067 mmol)
was dissolved in distilled methanol (3 mL) and cooled to
0 ꢀC. Ammonia (3 mL) was then condensed into the
pressure tube and the reaction was stirred on an oil bath
at 85 ꢀC for 12 h. The solution was then evaporated to
yield a brown oil. The product was purified first by flash
chromatography followed by radial chromatography
(70:30 ethyl acetate/acetone) yielding 17 mg (82%) of a
yellow oil: e336=9.32Â103. 1H NMR d 7.23 (s, 1H, H-9),
7.02 (t, 1H, H-13, J=10.1 Hz), 6.78 (d, 1H, H-14, J=
10.1 Hz), 6.61 (d, 1H, H-12, J=10.1 Hz), 6.36 (s, 2H, H-
1 and H-5), 3.81 (s, 9H), 2.85 (s, 4H, H-7 and H-8); 13C
NMR d 174.6, 156.5, 153.0, 151.9, 136.4, 136.2, 135.5,
131.3, 127.6, 112.6, 105.4, 60.7, 56.0, 43.0, 38.1. IR
3510, 3357, 3030–2825, 1676, 1610, 1595 cmÀ1. LR-MS
(m/z): 316 (10), 315 (M+, 40), 300 (20), 298 (15), 284
(25), 182 (10), 181 (100), 148 (10). HRMS calcd for
C19H24O4N: 315.1469. Found: 315.1465.
light-yellow solid: e332=6.68Â103; mp 104–107 ꢀC. H
1
NMR d 7.07 (d, 1H, H-14, J=9.2 Hz), 7.05 (s, 1H, H-9),
6.67 (d, 1H, H-12, J=9.2 Hz), 6.41 (t, 1H, H-13, J=
9.2 Hz), 6.32 (s, 9H), 3.82 (s, 6H, 2-OCH3 and 4-OCH3),
3.80 (s, 3H, 3-OCH3), 2.82 (s, 4H, H-7 and H-8), 2.69
(m, 2H, 11-CH2CH3, J=7.4 Hz), 1.17 (t, 3H, 11-CH2
CH3, J=7.4 Hz); 13C NMR d 181.3, 149.3, 148.8, 148.4,
140.9, 135.4, 132.2, 131.8, 128.9, 125.9, 101.2, 56.4, 51.7,
37.8, 34.5, 25.3, 9.5. IR 3030–2825, 1725, 1610, 1595cmÀ1
.
11-Demethoxy-11-methylaminocombretatropone (23). In
a pressure tube, combretatropone (4, 23mg, 0.070 mmol)
was dissolved in distilled methanol (5 mL). To this solu-
tion was added 2 M methylamine (2.8 mL, 0.140 mmol)
and the reaction was then stirred at 35 ꢀC on an oil bath
for 24 h. The solution was then evaporated to yield a
black solid. Purification of the product was carried out
first by flash chromatography followed by radial chro-
matography (80:20 CH2Cl2/acetone) yielding 19 mg
(84%) of a yellow solid: e338=1.04Â104; mp 111–
113 ꢀC. 1H NMR d 7.23 (broad s, 1H, NH), 7.16 (s, 1H,
H-9), 7.12 (t, 1H, H-13, J=10.0 Hz), 6.57 (d, 1H, H-14,
J=9.9 Hz), 6.39 (d, 1H, H-12, J=10.0 Hz), 6.36 (s, 2H,
H-1 and H-5), 3.81 (s, 9H), 3.04 (d, 3H, NHCH3, J=
5.1 Hz), 2.84 (m, 4H, H-7 and H-8); 13C NMR d 172.4,
152.6, 148.8, 147.6, 132.3, 132.1, 130.8, 125.0, 120.0,
103.0, 101.1, 56.4, 51.7, 38.9, 34.0, 25.1. IR 3335, 3020–
2825, 1670, 1610cmÀ1. LR-MS (m/z): 330 (10), 329
(M+, 60), 312 (10), 298 (15), 182 (15), 181 (100), 148
(10). HRMS calcd for C20H26O4N: 329.1625. Found:
329.1637.
LR-MS (m/z): 329 (5), 328 (M+, 10), 182 (15), 181 (100),
148 (15). HRMS calcd for C21H27O4: 328.1672. Found:
328.1682.
10-Demethoxy-10-methylisocombretatropone (20). This
compound was synthesized as described 18 except that
the isocombretatropone isomer 17 (27 mg, 0.085 mmol)
was used as the starting compound. The product was
purified first by flash chromatography followed by
radial chromatography (90:10 CH2Cl2/acetone) to yield
13 mg (48%) of a white solid: e321=3.72Â103; mp 107–
ꢀ
1
109 C. H NMR d 7.22 (d 1H, H-12, J=8.8 Hz), 7.04
(s, 1H, H-9), 6.99 (d, 1H, H-14, J=8.7 Hz), 6.81 (t, 1H,
H-13, J=8.8 Hz), 6.35 (s, 2H, H-1 and H-5), 3.82 (s,
9H), 2.81 (m, 4H, H-7 and H-8), 2.27 (s, 3H, 10-CH3);
13C NMR d 186.4, 153.1, 149.4, 146.6, 138.9, 136.5,
135.9, 134.2, 132.2, 130.7, 105.3, 60.7, 56.0, 42.2, 37.3,
22.5. IR 3040–2830, 1720, 1610 cmÀ1. LR-MS (m/z):
315 (5), 314 (M+, 10), 207 (10), 182 (15), 181 (100),
148 (10). HRMS calcd for C20H25O4: 314.1516. Found:
314.1529.
11-Demethoxy-11-dimethylaminocombretatropone (24).
Recrystallized pTSA (5 mg) was added toa pressure
tube. To the tube was added combretatropone (4,
25 mg, 0.076 mmol) dissolved in dry THF (2 mL). To
the solution was added 2 M dimethylamine (0.5 mL,
0.152 mmol) and the reaction was stirred at 80 ꢀC fo r
72 h. The reaction was cooled to room temperature and
diluted with methylene chloride. The organic layer was
washed first with 1% sodium bicarbonate (15 mL) and
then water (15 mL). Dried the organic layer with sodium
sulfate, filtered and evaporated. The product was puri-
fied using flash chromatography followed by radial
10 - Ethyl - 10 - demethoxyisocombretatropone (21). This
compound was synthesized as described 19 except that
the isocombretatropone isomer 17 (26 mg, 0.082 mmol)
was employed as the starting material. The product was
purified first by flash chromatography followed by
radial chromatography (98:2 CH2Cl2/acetone) to yield
15 mg (56%) of a light-yellow solid: e319=3.80Â103; mp
111–114 ꢀC. H NMR d 6.99 (d, 1H, H-12, J=9.0 Hz),
1
6.73 (d, 1H, H-14, J=9.1 Hz), 6.71 (s, 1H, H-9), 6.40
(t, 1H, H-13, J=9.0 Hz), 6.36 (s, 2H, H-1 and H-5),
3.83 (s, 6H, 2-OCH3 and 4-OCH3), 3.81 (s, 3H, 3-OCH3),

M. E. Janik, S. L. Bane / Bioorg. Med. Chem. 10 (2002) 1895–1903
1903
chromatography (80:20 CH2Cl2/acetone) yielding 17 mg
1
for obtaining the mass spectrometry data. This work
was supported by NSF MCB-9406242.
(65%) of a yellow oil: e354=9.08Â103. H NMR d 7.04
(s, 1H, H-9), 6.96 (t, 1H, H-13, J=9.8 Hz), 6.52 (d, 1H, H-
14, J=9.8Hz), 6.41 (d, 1H, H-12, J=9.8 Hz), 6.37 (s, 2H,
H-1 and H-5), 3.82 (s, 6H, 2,4-OCH3), 3.81 (s, 3H, 3-
OCH3), 3.10 (s, 6H, N(CH3)2) 2.85 (m, 4H, H-7 and H-8);
13C NMR d 180.4, 158.9, 153.0, 149.4, 136.5, 136.4,
132.6, 132.3, 125.4, 113.2, 105.4, 60.7, 55.9, 42.1, 41.7,
37.7. IR 3040–2812, 1710, 1610 cmÀ1. LR-MS (m/z): 344
(10), 343 (M+, 65), 207 (10), 182 (10), 181 (100), 162
(60), 148 (10), 134 (20). HRMS calcd for C21H28O4N:
343.1781. Found: 343.1799.
References and Notes
1. Jordan, M. A.; Wilson, L. Curr. Opin. Cell Biol. 1998, 10, 123.
2. Hamel, E. Med. Res. Rev. 1996, 16, 207.
3. Shi, Q.; Chen, K.; Morris-Natschke, S. L.; Lee, K. H. Curr.
Pharm. Des. 1998, 4, 219.
4. Lee, K. H. Med. Res. Rev. 1999, 19, 569.
5. Quinn, F. R.; Neiman, Z.; Beisler, J. A. J. Med. Chem.
1981, 24, 636.
6. (a) See, for example: Brown, M. L.; Rieger, J. M.; Macdo-
nald, T. L. Bioorg. Med. Chem. 2000, 8, 1433. (b) Zhang, S. X.;
Feng, J.; Kuo, S. C.; Brossi, A.; Hamel, E.; Tropsha, A.; Lee,
K. H. J. Med. Chem. 2000, 43, 167. (c) Polanski, J. SAR
QSAR Environ. Res. 2000, 11, 245.
Routine preparation of MTP and determination of protein
concentration. Microtubule protein (MTP), consisting
of tubulin and microtubule associated proteins, was
purified from bovine brain by two cycles of assembly/
disassembly.24 The purified MTP was drop frozen and
stored in liquid nitrogen until use. The frozen MTP was
prepared daily for polymerization assays by desalting
the thawed protein into PME buffer.
7. Timasheff, S. N.; Andreu, J. M.; Na, G. C. Pharmac. Ther.
1991, 52, 191.
8. Brossi, A.; Boye, O.; Muzaffar, A.; Yeh, H. J. C.; Toome,
V.; Wegrzynski, B.; George, C. FEBS Lett. 1988, 229, 82.
9. Berg, U.; Bladh, H. Helv. Chim. Acta 1999, 82.
10. Shi, Q.; Chen, K.; Chen, X.; Brossi, A.; Verdier-Pinard,
P.; Hamel, E.; McPhail, A. T.; Tropsha, A.; Lee, K. H. J. Org.
Chem. 1998, 63, 4018, and references therein.
11. Pyles, E. A.; Hastie, S. B. Biochemistry 1993, 32, 2329.
12. Chakrabarti, G.; Sengupta, S.; Bhattacharyya, B. J. Biol.
Chem. 1996, 271, 2897.
Microtubule assembly assays. MTP at a concentration
of 2 mg/mL was incubated with the appropriate ligand
at the desired concentration in PME buffer for 20 min at
room temperature. GTP was then added to achieve the
final concentrations of 0.1 mM. Assembly was initiated
by warming the solution to 37 ꢀC and the polymeri-
zation process was monitored by observing the change
in turbidity of the solution (apparent absorption at
400 nm on a Beckman Model 25/UV–visible spectro-
photometer containing a thermostated cell holder). The
I50 (ligand concentration required to produce 50%
inhibition of microtubule assemble relative to a control
without ligand) was determined by interpolation from a
plot of percent inhibition versus ligand concentration.
The percent inhibition was determined using the plateau
value after polymerization of the MTP was at steady
state. The I50 value for each compound was measured a
minimum of four times. The activities reported are the
mean and standard deviation of these data.
13. Menendez, M.; Laynez, J.; Medrano, F. J.; Andreu, J. M.
J. Biol. Chem. 1989, 264, 16367.
14. Perez-Ramirez, B.; Andreu, J. M.; Gorbunoff, M. J.;
Timasheff, S. N. Biochemistry 1996, 35, 3277.
15. Rossi, M.; Gorbunoff, M. J.; Caruso, F.; Wing, B.; Perez-
Ramirez, B.; Timasheff, S. N. Biochemistry 1996, 35, 3286.
16. Staretz, M. E.; Hastie, S. B. J. Med. Chem. 1993, 36, 758.
17. Hahn, K. M.; Humphreys, W. G.; Helms, A. M.; Hastie,
S. B.; Macdonald, T. L. Bioorg. Med. Chem. Lett. 1991, 1, 471.
18. Andres, C. J.; Bernardo, J. E.; Macdonald, T. L.; Yan, Q.;
Hastie, S. B.; Macdonald, T. L. Bioorg. Med. Chem. Lett.
1993, 3, 565.
19. Getahun, Z.; Jurd, L.; Chu, P. S.; Lin, C. M.; Hamel, E. J.
Med. Chem. 1992, 35, 1058.
20. Cramer, R. D.; Patterson, D. E.; Bunce, J. D. J. Am.
Chem. Soc. 1998, 110, 5959.
21. Lemmen, C.; Lengauer, T. J. Comp.-Aid. Mol. Des 2000,
14, 215.
22. Hastie, S. B.; Williams, R. C., Jr.; Puett, D.; Macdonald,
T. L. J. Biol. Chem. 1989, 264, 6682.
23. Boye, O.; Brossi, A. In The Alkaloids; Brossi, A., Cordell,
G. A., Eds.; Academic: San Diego, 1992; Vol. 41, p 125.
24. Williams, R. C., Jr.; Lee, J. C. Methods Enzymol. 1982, 85,
376.
Acknowledgements
We thank Dr. James L. Potenziano, Jr. for assistance in
synthesis of the MTC derivatives and Dr. Gordon Nicol