Antineoplastic agents. 410. Asymmetric hydroxylation of trans- combretastatin A-4

Article abstract of DOI:10.1021/jm9807149

The South African willow tree Combretum caffrum has yielded a number of potent cancer cell growth inhibitors. The present SAR studies of the antineoplastic agent combretastatin A-4 (1c) were focused mainly on the olefinic bridge to determine the effects o

Full text of DOI:10.1021/jm9807149

J . Med. Chem. 1999, 42, 1459-1465
1459
An tin eop la stic Agen ts. 410. Asym m etr ic Hyd r oxyla tion of tr a n s-Com br eta sta tin
A-41
George R. Pettit,*,† Brian E. Toki,† Delbert L. Herald,† Michael R. Boyd,‡ Ernest Hamel,‡ Robin K. Pettit,† and
J . Charles Chapuis†
Cancer Research Institute and Department of Chemistry, Arizona State University, P.O. Box 872404,
Tempe, Arizona 85287-2404, and Laboratory of Drug Discovery Research and Development, DTP/ DCTD,
National Cancer Institute, Frederick Cancer Research and Development Center, Frederick, Maryland 21702-1201
Received December 18, 1998
The South African willow tree Combretum caffrum has yielded a number of potent cancer cell
growth inhibitors. The present SAR studies of the antineoplastic agent combretastatin A-4
(1c) were focused mainly on the olefinic bridge to determine the effects on cancer cell growth
and, potentially, to better define the combretastatin A-4 binding site on tubulin. The geometric
trans-isomer 3a of combretastatin A-4 was converted to the (1S,2S)- and (1R,2R)-vicinal diols
4c and 4d , respectively, under Sharpless’ asymmetric dihydroxylation conditions. Cancer cell
line testing showed the (1S,2S)-diol 4c to be more potent than its enantiomer 4d . Diol 4c weakly
inhibited tubulin polymerization (IC50 ) 22 µM, versus 1.2 µM for combretastatin A-4), while
4d was inactive (IC50 > 40 µM). Esterification of either stereoisomer at the diol and/or phenolic
positions resulted in elimination of inhibitory activity.
In tr od u ction
For cancer to progress to metastatic disease, complex
biochemical and physiological changes leading to tissue
invasion and angiogenesis are required. Such neovas-
cularization is necessary for tumors to grow beyond a
spheroid of about 2 mm3, for diffusion of nutrients
cannot support a greater volume of neoplastic cells.2
Discovery and clinical development of antagonists of
such tumor angiogenesis should lead to considerable
improvements in cancer treatment.3,4 Currently some
10 antiangiogenic drugs are in clinical trials, including
substances as diverse as thalidomide,5 a fumagillin
derivative (TNP-470),3,6 and interleukin 12.3 One of the
most promising new antiangiogenic anticancer drugs
that began phase I clinical trials in November 1998 is
the cis-stilbene combretastatin A-4 prodrug (1a ).7
As part of an extensive structure-activity relation-
ship (SAR) study of combretastatin analogues, resulting
from our report of the first member of the series
(combretastatin, 2a ),8 we have been evaluating various
trans-stilbenes9,10 and their corresponding dihydro de-
rivatives (e.g., 2b, 2c). Recently, the trans-stilbenes
resveratrol11,12 and 3-hydroxy-4-methoxy-4′-nitro-trans-
stilbene have been found, respectively, to be chemopre-
ventive11 and to interfere with the phosphorylation of
certain proteins in the B-cell receptor cascade.13 The
current study was directed at ascertaining the effects
on cancer cell and microbial growth, inhibition of tubulin
polymerization,14,15 and eventually tumor angiogenesis
by the chiral vicinal diols related to combretastatin (2a ),
obtained from the poorly active trans-olefin 3a . Earlier
we had synthesized racemic combretastatin (2a ) as well
as isocombretastatin A (2b),8 and that approach was
later extended by others16,17 to related series of com-
pounds.
* To whom correspondence should be addressed.
Our previous studies suggested the need for a cis-
† Arizona State University.
‡ NCI.
olefin18,19 or other sp2-type20 unit linking the aryl groups
10.1021/jm9807149 CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/08/1999

1460 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8
Pettit et al.
F igu r e 2. Computer-generated perspective of the ethanediol
monohydrate 4d .
the preparation of the diols. The stereochemistry of the
trans-isomer 3c was confirmed by X-ray analysis, and
the molecule is shown in Figure 1. The trans-isomer 3c
was treated separately with the Sharpless AD mix-R
and AD mix-â to give the (1S,2S)- (4a ) and (1R,2R)- (4b)
diols, respectively, as colorless oils (Scheme 1). The
reactions were conducted in the dark in order to avoid
photochemical isomerization of the double bond. Depro-
tection using tetrabutylammonium fluoride afforded the
phenols 4c and 4d . Initial attempts at confirmation of
the absolute stereochemistry of the unprotected enan-
tiomers 4c and 4d , using X-ray structure determination
methods, failed. Although the relative stereochemistry
could be readily established via X-ray structure deter-
mination, the weak anomalous dispersion exhibited by
4d (for example) did not allow an unambiguous con-
figurational assignment for this compound (shown in
Figure 2) versus that of its enantiomer 4c. However,
the absolute stereochemical assignments of 4c and 4d
F igu r e 1. Computer-generated perspective of the silyl ether
of the trans-stilbene isomer 3c.
in order to attain high pharmacological activity. Cancer
cell line data have clearly shown a progressive decrease
in activity with cis-stilbenes > 1,2-diphenylethanes >
trans-stilbenes, as, for example, seen for combretastatin
A-1 (1b) > combretastatin B-1 (2c) > trans-isomer of
combretastatin A-1 (3b).18 Additional evidence was
furnished by the remarkable activity against cancer cell
lines of combretastatin A-4 (1c) vs its trans-isomer. The
Z-geometry also appears to be quite important for
inhibition of tubulin polymerization.
Resu lts a n d Discu ssion
Ch em istr y. The synthesis18 of combretastatin A-4
(1c) was repeated, and the intermediate silyl ether 3c
derivative of 3a was employed as the starting point for
Sch em e 1a
a
(a) TBAF, THF; (b) Ac2O, Et3N, DMAP, DCM; (c) 4-bromobenzoyl chloride, pyridine, DMF; (d) 4-bromobenzoyl chloride, Et3N, DMF.

Hydroxylation of trans-Combretastatin A-4
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8 1461
and 8 391, respectively. GI50 compare correlations,20
using 1a as the seed against 1a , 1c, 3a , 3c, 4a , 4b, 4c,
and 4d , were 1.00, 0.86, 0.84, <0.6, 0.74, <0.6, 0.62, and
<0.6, respectively. The consistent inhibitory activity
found for the (S,S)-diol (4c) compound compared to the
(R,R)-diol (4d ) chiral isomer provides another good
illustration of the considerable differences in biological
activity that can be expected for such enantiomeric
pairs.
The diols 4c and 4d were also evaluated for potential
inhibitory effects on tubulin polymerization, in com-
parison with combretastatin A-4 (1c). Consistent with
its low cytotoxicity, diol 4d was inactive as an inhibitor
of tubulin assembly (IC50 value > 40 µM). In contrast,
diol 4c inhibited the reaction (IC50 22 ( 0.9 (SD) µM),
but it was much less potent than combretastatin A-4
(IC50 1.2 ( 0.03 (SD) µM).
In contrast to the human cancer cell line results, the
(1S,2S)- (4a ) and (1R,2R)- (4b) diols were more micro-
biologically active than any derived compound (Table
3). With the exception of the Gram-negative pathogen
Neisseria gonorrhoeae, the combretastatins show selec-
tivity for Gram-positive bacteria (Table 3). With its
abbreviated outer-membrane lipooligosaccharide, N.
gonorrhoeae is more susceptible to hydrophobes than
typical Gram-negative bacteria.
F igu r e 3. Computer-generated molecular structural drawing
of the tert-butyldimethylsilyl diacetate derivative 4e, showing
the absolute stereochemistry as S at chiral centers C1A and
C1′A.
were established by a subsequent X-ray crystal struc-
ture determination of one of the enantiomers as the tert-
butyldimethylsilyl diacetate derivative, i.e., 4e (Figure
3).
Compound 4e was prepared by treatment of the
TBDMS-protected phenol 4a with acetic anhydride and
triethylamine/DMAP. Thus, the absolute configuration
of 4e (and consequently all compounds derived from 4a )
was established as 1S,2S. Similarly, the corresponding
enantiomers of these compounds derived from 4b could
be assigned the 1R,2R absolute configuration.
Exp er im en ta l Section
Biologica l Eva lu a tion . The murine P388 lympho-
cytic leukemia cell line and human cancer cell line (GI50)
data (Table 2) proved to be quite useful. Results of the
asymmetric dihydroxylation of trans-stilbene 3c fol-
lowed by deprotection showed the (1S,2S)-diol 4c to be
significantly more active than the (1R,2R)-diol 4d . The
cancer cell growth inhibitory activity was markedly
reduced or eliminated by esterifying the hydroxyl group.
Testing of eight selected compounds in the NCI 60-cell
screen18 yielded similar results. Triplicate tests of 1a ,
1c, 3a , 3c, 4a , 4b, 4c, and 4d gave mean panel GI50
values (nM) of 4.9, 8.7, 686, 4 688, 6 098, >10 000, 3 833,
The trans-stilbene silyl ether (3c) was synthesized as
previously described.18 4-(Dimethylamino)pyridine, methane-
sulfonamide, tert-butyl alcohol, tetrabutylammonium fluoride,
p-bromobenzoyl chloride, and AD mixes-R and -â were pur-
chased from Sigma-Aldrich and used as received. Acetic
anhydride was purchased from Acros and used as received.
Pyridine and triethylamine were distilled and stored over KOH
pellets. Methylene chloride (DCM) and DMF were distilled and
stored over 4 Å molecular sieves. Organic extracts of aqueous
workups were dried over magnesium sulfate. Solid samples
and oils were dried under high vacuum (0.05 mmHg at 25 °C)
unless otherwise noted. Flash column chromatography was
performed on silica gel from EM Science (230-400 mesh).
Ta ble 1. Physical Properties of the Diols and Derivatives
yield
Rf values
(hexanes-EtOAc)
compd (%)a
cryst solvent
n/a
mp (°C)
[R]25 (deg)
formula analysis
(C24H36O7Si) C, H
D
4a
4b
4c
4d
4e
88
85
79
77
81
oil
oil
-74.5 (c)0.68, CHCl3)
+77.9 (c)0.42, CHCl3)
-87.8 (c)0.32, MeOH)
+89.7 (c)0.37, MeOH)
-1.30 (c)0.77, CHCl3)
0.18 (3:2)
0.18 (3:2)
0.07 (1:1)
0.07 (1:1)
0.14 (4:1)
n/a
(C24H36O7Si) C, H
ether
ether
hexanes
89-90 (of hydrate)
89-92 (of hydrate)
90-92
(C18H22O7‚H2O) C, H
(C18H22O7‚H2O) C, H
(C28H40O9Si) C; H: calcd, 7.35;
found, 7.80
4f
4g
4h
4i
85
89
80
93
hexanes
b
b
89.5-90.5
117-118
115-117
+2.24 (c)1.3, CHCl3)
-59.3 (c)0.40, MeOH)
+60.9 (c)1.0, MeOH)
-65.9 (c)0.98, CHCl3)
0.14 (4:1)
0.46 (2:1)
0.46 (2:1)
0.33 (4:1)
(C28H40O9Si) C, H
(C39H31O10Br3) C, H, Br
(C39H31O10Br3) C, H
(C38H42O9Br2Si) H, Br; C:
calcd, 54.95; found, 55.51
(C32H28Br2O9‚0.5H2O) C, H, Br
(C25H25BrO8) C, H, Br
(C29H29BrO10) C, H, Br
EtOAc-hexanes 59-60
4j
4k
4l
73
79
91
b
195-196 (of hydrate) -57.5 (c)0.32, CHCl3)
0.18 (2:1)
0.18 (1:1)
0.60 (1:1)
carbon disulfide 70-72
68-70
+104.5 (c)0.22, CHCl3)
+20.0 (c)0.30, CHCl3)
b
a
b
Percent yield from starting material. Crystallized following concentration of the flash column chromatography eluant.
Ta ble 2. P388 Mouse Leukemia Cell Line (ED50, µg/mL) and Human Tumor Cell Lines (GI50, µg/mL)
cell type
leukemia
P388
ovarian
OVCAR-3
CNS
SF-295
renal lung-NSC
colon
melanoma pancreas pharynx neuroblast
thyroid
prostate
DU-145
compd
A498 NCI-H460 KM20L2 SK-MEL-5 BXPC-3
FADU
SK-N-SH SW1736
1c
2b
4c
4d
2.7 × 10-4 <1.0 × 10-3 <1.0 × 10-3
5.6 × 10-4 6.1 × 10-2 2.0 × 10-4 3.9 × 10-1 6.5 × 10-4 2.6 × 10-4 7.1 × 10-4 7.6 × 10-4
32.9
9.0
0.91 2.5
>10 4.4
9.4
4.8
>10
0.28
1.5
>10
1.8
7.6
2.1
3.5
2.0
0.38
4.5
0.52
5.7
0.52
8.2
0.70
4.9
3.9
>10
0.60
2.1
19.0
6.1

1462 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8
Ta ble 3. Antimicrobial Activity of the Combretastatins
Pettit et al.
was 0.0624 for observed data and 0.0653 for all data. The
goodness-of-fit on F2 was 1.080. The corresponding Sheldrick
R values were wR2 of 0.1851 and 0.1892, respectively. A final
difference Fourier map showed minimal residual electron
density; the largest difference peak and hole being 0.891 and
-0.385 e/Å3, respectively. Final bond distances and angles
were all within acceptable limits. The structure of the trans-
isomer 3c is shown in Figure 1.25
microbe(s)
inhibited
minimum inhibitory
concentration (µg/disk)
compd
1a
1b
N. gonorrhoeae9
N. gonorrhoeae
S. aureus
50-100
25-50
50-100
25-50
1c
2c
3a
3c
N. gonorrhoeae18
N. gonorrhoeae
M. luteus9
12.5-25
25-50
Asym m etr ic Dih yd r oxyla tion (4a , 4b): Rep r esen ta tive
P r oced u r e. AD mix-R (8.4 g, 1.4 g/mmol of trans-stilbene silyl
ether, 3c) was added to a biphasic solution of butanol and
water (30 mL each). After this dissolved, methanesulfonamide
(0.57 g, 6.0 mmol, 1.0 equiv) was added, and the contents were
cooled to 0 °C. The trans-silyl ether stilbene (3c) (2.6 g, 6.0
mmol, 1.0 equiv) was added and the reaction mixture stirred
vigorously in the dark. After 24 h, additional AD mix-R (1.0
g) and another equivalent of methanesulfonamide were added.
The contents were warmed to room temperature, and stirring
continued until TLC (3:2 hexanes-EtOAc) showed no starting
material (72 h total). Sodium sulfite (4 g) was added, 45 min
later ethyl acetate (25 mL) was added, and the layers were
separated. The aqueous layer was further washed with EtOAc
(3 × 25 mL), the combined organic extracts were washed with
2 N KOH (50 mL), dried, and filtered, and solvent was removed
in vacuo to give a yellow oil that was subjected to flash column
chromatography (eluant 3:2 hexanes-EtOAc). Concentration
of the desired fractions yielded 4a as a clear oil: yield, 2.48 g;
IR νmax (Nujol) 3457w, 2955s, 2858s, 1593s, 1512s, 1462m,
1377m, 1236m, 1130m; 1H NMR δ (300 MHz, CDCl3) 0.07 (3H,
s), 0.09 (3H, s), 0.96 (9H, s), 3.73 (6H, s), 3.76 (3H, s), 3.79
(3H, s), 4.55 (1H, d, J 7.7 Hz), 4.60 (1H, d, J 7.7 Hz), 6.33 (2H,
s), 6.67 (1H, dd, J 2, 8.5 Hz), 6.72 (1H, s), 6.73 (1H, d, J 8.5
Hz); 13C NMR δ (75 MHz, CDCl3) 152.86, 150.79, 144.81,
137.40, 135.49, 132.62, 120.45, 119.73, 111.68, 103.91, 79.30,
78.75, 60.81, 56.00, 55.56, 25.69, 18.41, -4.73; EIMS m/z 464
(0.1%, M+), 449 (1%, M+ - CH3), 407 (6%, M+ - t-Bu), 267
(40%, M+ - HOCH - C6H2(OCH3)3), 198 (100%, M+ - HOCH
- C6H3(OCH3)(OTBDMS)).
a
4a
4b
4c-4l
Str. pneumoniae
Str. pneumoniae
a
25-50
50-100
a
At 100 µg/disk no inhibition was observed.
Melting points were obtained using an Electrothermal 9100
apparatus and are uncorrected. TLC was performed with
Analtech silica gel GHLF plates; all compounds were visible
under short-wave UV light (254 nm) and stained with 2% ceric
sulfate in ethanol. Mass spectra (FAB) were recorded using a
Varian MAT 311A. 1H NMR (300 MHz) was recorded on a
Varian Gemini 300 instrument using deuterated chloroform
as the solvent and referencing to TMS. 13C NMR (100 MHz)
was recorded on a Varian Gemini 400 instrument in CDCl3.
UV spectra were obtained employing a Perkin-Elmer Lambda
3B spectrometer interfaced with an IBM PC running PECSS
data acquisition software or a UV-vis-NIR scanning spec-
trophotometer interfaced with a Shimadzu UV-3101PC. IR
spectra were obtained with a Mattson 2020 Galaxy series FTIR
instrument using NaCl plates. Combustion analyses were
determined by the Galbraith Microanalytical Laboratory and
are within (0.3% of the calculated value. X-ray experiments
were performed with an Enraf-Nonius CAD4 diffractometer.
Tubulin polymerization was evaluated as described previ-
ously.20
X-r a y Cr ysta l Str u ctu r e Deter m in a tion . Silyl ether of
the trans-stilbene isomer, 3c: Large, thick plates of the
compound could be grown from hexane. A large crystal (∼0.52
× 0.42 × 0.40 mm) of 3c was used in the data collection, which
was performed at 299 ( 1 K. Accurate cell dimensions were
determined by least-squares fitting of 25 carefully centered
reflections in the range of 35° < Θ < 40° using Cu KR
radiation.
Similarly, the resultant diol from AD mix-â gave 4b as a
clear oil (2.40 g): IR νmax (Nujol) 3408w, 2924s, 2854s, 1593s,
1512s, 1462m, 1377m, 1236m, 1130m; H NMR δ (300 MHz,
1
CDCl3) 0.07 (3H, s), 0.09 (3H, s), 0.96 (9H, s), 3.73 (6H, s),
3.76 (3H, s), 3.80 (3H, s), 4.55 (1H, d, J 7.7 Hz), 4.60 (1H, d, J
7.7 Hz), 6.33 (2H, s), 6.67 (1H, dd, J 2, 8.5 Hz), 6.72 (1H, s),
6.73 (1H, d, J 8.5 Hz); 13C NMR δ (100 MHz, CDCl3) 152.85,
150.76, 144.79, 137.40, 135.42, 132.52, 120.41, 119.67, 111.63,
103.84, 79.27, 78.73, 60.78, 55.98, 55.53, 25.66, 18.41, -4.74;
EIMS m/z 464 (0.5%, M+), 449 (0.5%, M+ - CH3), 407 (6%,
M+ - t-Bu), 267 (40%, M+ - HOCH - C6H2(OCH3)3), 198
(100%, M+ - HOCH - C6H3(OCH3)(OTBDMS)).
Cr ysta l d a ta : C24H34O5Si1, FW ) 430.60, orthorhombic,
P212121; a ) 8.131(2), b ) 9.441(1), c ) 32.997(6) Å; V )
2533.0(8) Å3, Z ) 4, Fc ) 1.129 Mg/M3; µ(Cu KR) ) 1.053 mm-1
λ ) 1.54180 Å.
,
All reflections corresponding to more than a complete octant
(0 < )h < )9, 0 < )k < )11, -3 < )l < )38) were collected
over the range of 0 < 2Θ < 130° using the ω/2Θ scan
technique. Three intensity control reflections were also mea-
sured for every 60 min of X-ray exposure time and showed a
maximum variation of +4.7% over the course of the collection.
Subsequent statistical analysis of the complete reflection data
set using the XPREP21 program, coupled with density mea-
surements, verified that the space group was P212121. After
Lorentz and polarization corrections, merging of equivalent
reflections, and rejection of systematic absences, 2 538 unique
reflections (R(int) ) 0.0311) remained, of which 2 385 were
considered observed (Io > 2σ(Io)) and were used in the
subsequent structure determination and refinement. Linear
and anisotropic decay corrections were applied to the intensity
data as well as an empirical absorption correction (based on a
series of psi-scans).22 Structure determination was readily
accomplished with the direct-methods program SIR92.23 All
non-hydrogen atom coordinates were located in a routine run
using default values in that program. The remaining H atom
coordinates were calculated at optimum positions. All non-
hydrogen atoms were refined anisotropically in a full-matrix
least-squares refinement using SHELXL-97.24 The H atoms
were included; their Uiso thermal parameters were fixed at 1.2
the Uiso of the atom to which they were attached and forced to
ride that atom. The final standard residual R1 value for 3c
(1S,2S)- a n d (1R,2R)-1-(3-Hyd r oxy-4-m eth oxyp h en yl)-
2-(3′,4′,5′-tr im eth oxyph en yl)-1,2-eth a n ediol (4c, 4d ): Rep -
r esen ta tive P r oced u r e. The TBDMS-protected phenol 4b
(1.20 g, 2.59 mmol) from the AD mix-â was dissolved in THF
(50 mL), and tetrabutylammonium fluoride (2.59 mL, 2.59
mmol, 1.0 M) was added to the solution. The solution was
stirred under Ar and monitored by TLC (3:2 hexanes-EtOAc).
After 30 min, ice (20 g) was added and the resultant solution
extracted with EtOAc (100 mL). The organic layer was washed
successively with water (50 mL) and brine (50 mL), dried, and
filtered, and solvent was removed in vacuo to give a creamy
white/yellow solid. Flash chromatography (eluant 1:1 DCM-
EtOAc) resulted in isolation of a white amorphous solid (4d ):
yield, 0.75 g; IR νmax (Nujol) 3457w, 2922, 2854s, 1589s; 1H
NMR δ (300 MHz, CDCl3) 3.74 (6H, s), 3.81 (3H, s), 3.86 (3H,
s), 4.57 (1H, d, J 7 Hz), 4.64 (1H, d, J 7 Hz), 5.58 (1H, s, broad),
6.36 (2H, s), 6.58 (1H, dd, J 2, 8 Hz), 6.72 (1H, d, J 8 Hz), 6.88
(1H, d, J 2 Hz); EIMS m/z 350 (10%, M+), 332 (5%, M+ - H2O),
198 (100%, M+ - HOCH - C6H3(OCH3)(OH)).
Synthesis of the monohydrate (1S,2S)-enantiomer 4c from
4a gave a white powder after recrystallization: IR νmax (Nujol)
3457w, 2922, 2854s, 1589s; 1H NMR δ (300 MHz, CDCl3) 3.74
(6H, s), 3.81 (3H, s), 3.86 (3H, s), 4.57 (1H, d, J 7 Hz), 4.63
(1H, d, J 7 Hz), 5.60 (1H, s, broad), 6.36 (2H, s), 6.57 (1H, dd,

Hydroxylation of trans-Combretastatin A-4
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8 1463
J 2, 8 Hz), 6.72 (1H, d, J 8 Hz), 6.87 (1H, d, J 2 Hz); EIMS
Hz), 6.33 (1H, d, J 9 Hz), 6.42 (2H, s), 6.84 (1H, d, J 8 Hz),
6.99 (1H, dd, J 2, 8 Hz), 7.18 (1H, d, J 2 Hz), 7.54 (4H, d, J 8
Hz), 7.66 (2H, d, J 8 Hz), 7.87 (2H, d, J 8 Hz), 7.88 (2H, d, J
8 Hz), 8.03 (2H, d, J 8 Hz); EIMS m/z 900 (5%, M+), 183 (100%,
[OCC6H4Br]+).
4h was prepared similarly: 1H NMR δ (300 MHz, CDCl3)
3.74 (6H, s), 3.76 (3H, s), 3.78 (3H, s), 6.29 (1H, d, J 9 Hz),
6.33 (1H, d, J 9 Hz), 6.42 (2H, s), 6.84 (1H, d, J 8 Hz), 6.99
(1H, dd, J 2, 8 Hz), 7.18 (1H, d, J 2 Hz), 7.54 (4H, d, J 8 Hz),
7.66 (2H, d, J 8 Hz), 7.87 (2H, d, J 8 Hz), 7.88 (2H, d, J 8 Hz),
8.03 (2H, d, J 8 Hz); EIMS m/z 900 (5%, M+), 183 (100%,
[OCC6H4Br]+).
(1S,2S)- a n d (1R,2R)-1,2-Dia cet oxy-1-(3-[ter t-b u t yld i-
m eth ylsilyloxy]-4-m eth oxyp h en yl)-2-(3′,4′,5′-tr im eth oxy-
p h en yl)eth a n e (4e, 4f): Rep r esen ta tive P r oced u r e. Acetic
anhydride (0.51 mL, 5.38 mmol, 2.5 equiv) was added to a
solution of the diol (1.0 g, 2.15 mmol, 1.0 equiv) with triethy-
lamine (0.90 mL, 6.46 mmol, 3.0 equiv) and DMAP (53 mg,
0.043 mmol, 0.2 equiv) in DCM (5 mL) under Ar at room
temperature. The yellow solution was stirred for 50 min, and
TLC (3:2 hexanes-EtOAc) showed no residual starting mate-
rial. Methanol (5 mL) was added and the reaction mixture
concentrated to a yellow oil that was washed with ether (10
mL) and evaporated. The oil was partitioned with ether (25
mL) and 1 N HCl (25 mL). The organic layer was further
washed with 10% aqueous sodium bicarbonate (25 mL) and
water (25 mL) and dried. Filtration followed by removal of
solvent in vacuo gave rise to a yellow oil that was subjected to
flash column chromatography (4:1 hexanes-EtOAc). The
desired fractions were concentrated to a clear oil which
crystallized upon standing: yield, 0.95 g. 4e: 1H NMR δ (300
MHz, CDCl3) 0.04 (3H, s), 0.07 (3H, s), 0.95 (9H, s), 2.09 (3H,
s), 2.10 (3H, s), 3.73 (6H, s), 3.74 (3H, s), 3.78 (3H, s), 5.90
(1H, d, J 9 Hz), 5.93 (1H, d, J 9 Hz), 6.33 (2H, s), 6.68 (3H,
m); EIMS m/z 548 (10%, M+), 491 (70%, M+ - t-Bu), 267 (100%,
M+ - AcOCHC6H2(OCH3)3 - Ac), 197 (80%, M+ - AcOCHC6H3-
(OCH3)(OTBDMS) - Ac).
m/z 350 (1%, M+), 332 (1%, M+ - H2O), 198 (100%, M+
HOCH - C6H3(OCH3)(OH)).
-
X-r a y Cr ysta l Str u ctu r e Deter m in a tion . 1-(3-Hydroxy-
4-methoxyphenyl)-2-(3′,4′,5′-trimethoxyphenyl)-1,2-ethanediol,
4d : Plate-shaped crystals of the monohydrate of this com-
pound were grown from an ether solution. A crystal (∼0.32 ×
0.30 × 0.18 mm) of 4d was used in the data collection, which
was performed at 298 ( 1 K. Accurate cell dimensions were
determined by least-squares fitting of 25 carefully centered
reflections in the range of 35° < 0 < 40° using Cu KR radiation.
Cr ysta l d a ta : C18H22O7‚1H2O, FW ) 368.37, orthorhombic,
P212121; a ) 6.1050(2), b ) 13.677(3), c ) 21.478(4) Å; V )
1793.4(5) Å3, Z ) 4, Fc ) 1.364 Mg/m3; µ(Cu KR) ) 0.907 mm-1
λ ) 1.54180 Å.
,
All reflections corresponding to slightly more than a com-
plete octant (0 < )h < )7, 0 < )k < )16, -1 < l < )25) were
collected over the range of 0 < 2Θ < 130° using the ω/2Θ scan
technique. The Friedel reflections were also collected (when-
ever possible) immediately after each reflection. Three inten-
sity control reflections were also measured for every 60 min
of X-ray exposure time and showed a maximum variation of
-7.7% over the course of the collection. Statistical analysis of
the complete reflection data set using the XPREP21 program,
coupled with density measurements, verified the space group
as P212121. After Lorentz and polarization corrections, merging
of equivalent reflections, and rejection of systematic absences,
2 992 independent reflections (R(int) )0.1019) remained, of
which 2 899 were considered observed (Io > 2σ(Io)) and were
used in the subsequent structure determination and refine-
ment. Linear and anisotropic decay corrections were applied
to the intensity data as well as an empirical absorption
correction (based on a series of psi-scans).22 Structure deter-
mination was readily accomplished with the direct-methods
program SIR92.23 All non-hydrogen atom coordinates were
located in a routine run using default values in that program.
The remaining H atom coordinates were either determined by
difference maps or calculated at optimum positions. All non-
hydrogen atoms were refined anisotropically in a full-matrix
least-squares refinement using SHELXL-93.26 The H atoms
were included; their Uiso thermal parameters were fixed at 1.5
the Uiso of the atom to which they were attached and forced to
ride that atom. The final standard residual R1 value for 4d
was 0.0745 for observed data and 0.0757 for all data. The
goodness-of-fit on F2 was 0.973. The corresponding Sheldrick
R values were wR2 of 0.2079 and 0.2098, respectively. A final
difference Fourier map showed minimal residual electron
density; the largest difference peak and hole being +0.355 and
-0.351 e/Å3, respectively. The final bond distances and angles
were all within acceptable limits. The absolute structure of
4d could not be assigned with certainty, since the value of the
Flack parameter (0.23) deviated significantly from 0.00 (in-
dicative of the correct enantiomer) and exhibited a large esd
(0.31). Consequently, the structure of the 1-(3-hydroxy-4-
methoxyphenyl)-2-(3′,4′,5′-trimethoxyphenyl)-1,2-ethanediol
monohydrate (4d ), as shown in Figure 2,25 represents one of
the two possible enantiomeric configurations.
4f was isolated as large glassy cubes: 1H NMR δ (300 MHz,
CDCl3) 0.04 (3H, s), 0.07 (3H, s), 0.95 (9H, s), 2.09 (3H, s),
2.10 (3H, s), 3.73 (6H, s), 3.74 (3H, s), 3.78 (3H, s), 5.90 (1H,
d, J 9 Hz), 5.93 (1H, d, J 9 Hz), 6.33 (2H, s), 6.68 (3H, m);
EIMS m/z 548 (5%, M+), 491 (60%, M+ - t-Bu), 267 (100%,
M+ - AcOCHC6H2(OCH3)3 - Ac), 197 (80%, M+ - AcOCHC6H3-
(OCH3)(OTBDMS) - Ac).
X-r a y Cr yst a l St r u ct u r e Det er m in a t ion . tert-Butyldi-
methylsilyl diacetate derivative, 4e: Large, thick plates of the
compound were grown from hexane. A large crystal (∼0.52 ×
0.44 × 0.40 mm) of 4e was used in the data collection, which
was performed at 296 ( 1 K. Accurate cell dimensions were
determined by least-squares fitting of 25 carefully centered
reflections in the range of 35° < Θ < 40° using Cu KR
radiation.
Cr ysta l d a ta : C56H80O18Si2, FW ) 1087.38, triclinic, Pl; a
) 9.103(5), b ) 12.782(5), c ) 14.164(5) Å; R ) 111.29(2), â )
97.92(2), γ ) 91.00(2); V ) 1516.8(12) Å, Z ) 1, Fc ) 1.201
Mg/m3; µ(Cu KR) ) 1.088 mm-1, λ ) 1.54180 Å.
All reflections corresponding to a complete hemisphere (0
< )h < )10, -15 < )k < )15, -16 < )l < )16) were collected
over the range of 0 < 2Θ < 130° using the ω/2Θ scan
technique. The Friedel reflections were also collected (when-
ever possible) immediately after each reflection. Three inten-
sity control reflections were also measured for every 60 min
of X-ray exposure time and showed a maximum variation of
+6.2% over the course of the collection. Subsequent statistical
analysis of the complete reflection data set using the XPREP21
program verified the space group as P1, with two independent
molecules in the asymmetric unit cell. After Lorentz and
polarization corrections, merging of equivalent reflections, and
rejection of systematic absences, 10 168 unique reflections
(R(int))0.0447) remained, of which 9 928 were considered
observed (Io > 2σ(Io)) and were used in the subsequent
structure determination and refinement. Linear and aniso-
tropic decay corrections were applied to the intensity data as
well as an empirical absorption correction (based on a series
(1S,2S)- a n d (1R,2R)-1,2-Di(p-br om oben zoyloxy)-1-(3-
[p -b r om ob en zoyloxy]-4-m et h oxyp h en yl)-2-(3′,4′,5′-t r i-
m eth oxyp h en yl)eth a n e (4g, 4h ): Rep r esen ta tive P r oce-
d u r e. The asymmetric triol 4c (0.5 g, 1.43, 1.0 equiv) was
dissolved in 2:1 pyridine-DMF (10 and 5 mL, respectively)
while under Ar, and p-bromobenzoyl chloride (1.92 g, 8.56
mmol, 6.0 equiv) was added. The resultant suspension was
stirred for 2 h, when TLC analysis showed no residual starting
material and a single product. Ice (20 g) was added, and the
contents were filtered. The residue was taken up in ethyl
acetate (50 mL) and washed successively with 1 M HCl, water,
10% sodium bicarbonate, water, and brine. The organic layer
was dried and filtered, and removal of solvent gave a yellow
solid that was subjected to flash chromatography (eluant 4:1
hexanes-EtOAc). Concentration of the desired fractions led
to a white solid (4g): yield, 1.14 g; 1H NMR δ (400 MHz,
CDCl3) 3.74 (6H, s), 3.76 (3H, s), 3.78 (3H, s), 6.29 (1H, d, J 9

1464 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8
Pettit et al.
of psi-scans).22 Structure determination was readily accom-
plished with the direct-methods program SIR92.23 All non-
hydrogen atom coordinates were located in a routine run using
default values in that program. The remaining H atom
coordinates were calculated at optimum positions. All non-
hydrogen atoms were refined anistropically in a full-matrix
least-squares refinement using SHELXL-97.24 The H atoms
were included; their Uiso thermal parameters were fixed at 1.5
the Uiso of the atom to which they were attached and forced to
ride that atom. The final standard residual R1 value for 4e
was 0.0634 for observed data and 0.0642 for all data. The
goodness-of-fit on F2 was 1.036. The corresponding Sheldrick
R values were wR2 of 0.1726 and 0.1732, respectively. A final
difference Fourier map showed minimal residual electron
density; the largest difference peak and hole being 0.280 and
-0.163 e/Å3, respectively. The absolute structure of 4e could
be assigned with certainty based on the value of the Flack
parameter (0.06, esd 0.03) for the enantiomer shown in Figure
3.25 Consequently, the absolute stereochemistry of structure
of the silyl-protected diacetate derivative 4d can be assigned
the S configuration at both chiral carbons, C1A and C1′A. Final
bond distances and angles were all within acceptable limits.
concentrated, yielding a white solid. This was recrystallized
from carbon disulfide to give a white powder: yield, 0.60 g;
1H NMR δ (300 MHz, CDCl3) 3.70 (6H, s), 3.72 (3H, s), 3.75
(3H, s), 4.54 (2H, s (coalesced doublet)), 6.29 (2H, s), 6.78 (1H,
d, J 8 Hz), 6.83 (1H, dd, J 2, 8 Hz), 7.03 (1H, d, J 2 Hz), 7.61
(2H, d, J 8 Hz), 8.00 (2H, d, J 8 Hz); EIMS m/z 532 (1%, M+),
517 (5%, M+ - CH3), 514 (5%, M+ - H2O), 335 (5%, M+ - OCH
- C6H3(OCH3)3), 198 (100%, M+ - HOCH - C6H3(OCH3)-
(O2C6H4Br)), 183 (45%, [OCC6H4Br]+).
(1R,2R)-1,2-Diacetoxy-1-(3-[p-br om oben zoyloxy]-4-m eth -
oxyp h en yl)-2-(3′,4′,5′-tr im eth oxyp h en yl)eth a n e (4l). Ace-
tic anhydride (84 µL, 0.89 mmol, 2.5 equiv) was added to a
solution of the diol 4k (0.19 g, 0.36 mmol, 1.0 equiv) with
DMAP (8.8 mg, 0.071 mmol, 0.2 equiv) and triethylamine (0.15
mL, 1.07 mmol, 3.0 equiv) in DCM (1 mL) under Ar at room
temperature. The contents were stirred for 3 h, and TLC (1:1
hexanes-EtOAc) showed no residual starting material. Metha-
nol (5 mL) was added and the reaction mixture concentrated
to a yellow oil that was washed with ether, which was removed
by evaporation. The oily residue was partitioned with ether
(15 mL) and 1 N HCl (15 mL). The organic layer was further
washed with 10% aqueous sodium bicarbonate (15 mL). The
combined aqueous layers were back-extracted with DCM (5
mL), and the organic layers were combined and dried. Filtra-
tion and evaporation of solvents gave an off-white solid that
was subjected to flash chromatography (eluant 1:1 hexanes-
EtOAc). The desired fractions were collected and concentrated,
(1S,2S)-1,2-Di(p-b r om ob en zoyloxy)-1-(3-[ter t-b u t yld i-
m eth ylsilyloxy]-4-m eth oxyp h en yl)-2-(3′,4′,5′-tr im eth oxy-
p h en yl)eth a n e (4i). The resultant diol from AD mix-R 4a
(0.75 g, 1.61 mmol, 1.0 equiv) was dissolved in DMF (15 mL),
and pyridine (15 mL, 1:1 DMF-pyridine) was added at room
temperature. The reaction was stirred until homogeneous, and
4-bromobenzoyl chloride (0.89 g, 4.04 mmol, 2.5 equiv) was
added all at once. The yellow mixture was stirred under Ar at
room temperature. A white precipitate formed within 30 min,
and after 2 h the temperature was raised to 45 °C; 16 h later,
the temperature was raised to 85 °C, and another 2.5 equiv of
4-bromobenzoyl chloride was added. No starting material was
visualized following TLC. Ice (30 g) and ether (25 mL) were
added. The ethereal layer was successively washed with 1 M
HCl (25 mL), 10% sodium bicarbonate (25 mL), and water (25
mL). The organic phase was dried and filtered, and solvent
was removed to give a yellow oil that was subjected to flash
chromatography (9:1 hexanes-EtOAc). The desired product
(Rf ) 0.33, 4:1 hexanes-EtOAc) was isolated as an off-white
solid that was recrystallized from EtOAc-hexane to afford
small needles: yield, 0.70 g; 1H NMR δ (300 MHz, CDCl3)
-0.02 (3H, s), 0.04 (3H, s), 0.92 (9H, s), 3.70 (6H, s), 3.73 (3H,
s), 3.76 (3H, s), 6.23 (1H, d, J 9 Hz), 6.29 (1H, d, J 9 Hz), 6.41
(2H, s), 6.70 (1H, d, J 8 Hz), 6.78 (1H, s), 6.79 (1H, d, J 8 Hz),
7.52 (4H, d, J 8 Hz), 7.84 (4H, d, J 8 Hz); EIMS m/z 830 (2%,
M+2), 828 (1%, M+), 774 (15%, M+ - t-Bu + H), 183 (100%,
[OCC6H4Br]+).
(1S ,2S )-1,2-Di(p -b r om ob e n zoyloxy)-1-(3-h yd r oxy-4-
m eth oxyp h en yl)-2-(3′,4′,5′-tr im eth oxyp h en yl)eth a n e (4j).
4i (1.24 g, 1.49 mmol) was converted to 4j in an identical
fashion to the deprotection procedure of 4c. After flash
chromatography (2:1 hexanes-EtOAc), the desired fractions
were concentrated to give a fine woolly solid (4j): yield, 0.79
g; 1H NMR δ (400 MHz, CDCl3) 3.72 (6H, s), 3.78 (3H, s), 3.84
(3H, s), 5.56 (1H, s), 6.29 (1H, d, J 9 Hz), 6.29 (1H, d, J 9 Hz),
6.43 (2H, s), 6.67 (1H, dd, J 2, 8 Hz), 6.68 (1H, d, J 8 Hz), 6.97
(1H, d, J 2 Hz), 7.54 (4H, m), 7.87 (4H, m); EIMS m/z 716
(10%, M+), 516 (5%, M+ - HO2CC6H4Br), 183 (100%, [OCC6H4-
Br]+).
1
producing a white solid: yield, 0.20 g; H NMR δ (300 MHz,
CDCl3) 2.10 (3H, s), 2.12 (3H, s), 3.75 (3H, s), 3.76 (6H, s),
3.78 (3H, s), 5.93 (1H, d, J 9 Hz), 5.97 (1H, d, J 9 Hz), 6.34
(2H, s), 6.80 (1H, d, J 8 Hz), 6.89 (1H, dd, J 2, 8 Hz), 7.08 (1H,
d, J 2 Hz), 7.65 (2H, d, J 8 Hz), 8.03 (2H, d, J 8 Hz); EIMS
m/z 616 (15%, M+), 559 (1%, M+ - O2CCH3), 498 (5%, M+
-
(O2CCH3)2), 434 (6%, M + H - OCC6H4Br), 197 (100%,
[HOCHC6H2(OCH3)3]+), 183 (55%, [OCC6H4Br]+).
An tim icr obia l Su scep tibility Testin g. Compounds were
screened against the bacteria Stenotrophomonas maltophilia,
Micrococcus luteus, Staphylococcus aureus, Escherichia coli,
Enterobacter cloacae, Enterococcus faecalis, Streptococcus pneu-
moniae, and Neisseria gonorrhoeae and the fungi Candida
albicans and Cryptococcus neoformans, according to estab-
lished disk susceptibility testing protocols.27
Ack n ow led gm en t. Appreciation and thanks for
support of this research are extended to Outstanding
Investigator Grant CA 44344-05-9 with the Division of
Cancer Treatment and Diagnosis, NCI, DHHS, the
Arizona Disease Control Research Commission, Virginia
Piper, Diane Cummings Halle, Gary L. and Diane R.
Tooker, Polly Trautman, J ohn and Edith Reyno, the
Caitlin Robb Foundation, and the Robert B. Dalton
Endowment. We also thank Drs. Cherry L. Herald,
Fiona Hogan, J ean M. Schmidt, and Michael D. Wil-
liams, as well as David M. Carnell, Laura Crews, and
Lee Williams, the National Science Foundation for
equipment Grant CHE-8409644, and the NSF Regional
Instrumentation Facility in Nebraska (Grant CHE-
8620177).
Su p p or tin g In for m a tion Ava ila ble: Expanded Table 2
containing the murine P388 and human cancer cell line results
for the less active and inactive compounds 4a , 4b, and 4g-4l;
X-ray crystallographic tables of experimental detail, atomic
coordinates, bond lengths and angles, and anisotropic thermal
parameters for 3c, 4d , and 4e. This information is available
free of charge via the Internet at http://pubs.acs.org.
(1R,2R)-1,2-Dih yd r oxy-1-(3-[p -b r om ob e n zoyloxy]-4-
m eth oxyph en yl)-2-(3′,4′,5′-tr im eth oxyph en yl)eth an e (4k).
The chiral phenol 4d (0.5 g, 1.43 mmol, 1.0 equiv) was
dissolved in DMF (20 mL), and triethylamine (0.40 mL, 2.85
mmol, 2.0 equiv) was added. After 10 min, addition of p-
bromobenzoyl chloride (0.31 g, 1.43 mmol, 1.0 equiv) resulted
in
a yellow solution as the reaction proceeded at room
temperature under Ar. Starting material could no longer be
visualized by TLC after 2 h. Ice was added, and a precipitate
formed that was removed by filtration. The aqueous filtrate
was washed with ethyl acetate (25 mL), and the concentrated
residue was added to the filtered solid and flash chromato-
graphed (1:1 hexanes-EtOAc). The desired fraction were
Refer en ces
(1) For contribution 409, refer to: Pettit, G. R.; Tan, R.; Melody,
N.; Kielty, J . M.; Pettit, R. K.; Herald, D. L.; Tucker, B.; Mallavia,
L. P.; Doubek, D. L.; Schmidt, J . M. Antineoplastic Agents 409.
Isolation and Structure of Montanastatin from a Terrestrial
Actinomycete. Bioorg. Med. Chem., in press.

Hydroxylation of trans-Combretastatin A-4
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 8 1465
(2) Weidner, N.; Folkman, J . Tumor Vascularity as a Prognostic
Factor in Cancer. In Cancer: Principles & Practice of Oncology
Updates; DeVita, V. T., Hellman, S., Rosenberg, S. A., Eds.; J .
B. Lippincott: Philadelphia, 1997; Vol. 11, pp 1-24.
(3) Pluda, J . M. Tumor-Associated Angiogenesis: Mechanisms,
Clinical Implications, and Therapeutic Strategies. Semin. Oncol.
1997, 24, 203-218.
(4) Folkman, J . New Perspectives in Clinical Oncology from Angio-
genesis Research. Eur. J . Cancer 1996, 32A, 2534-2539.
(5) D′Amato, R. J .; Loughnan, M. S.; Flynn, E.; Folkman, J .
Thalidomide is an Inhibitor of Angiogenesis. Proc. Natl. Acad.
Sci. U.S.A. 1994, 91, 4082-8085.
(6) Sin, N.; Meng, L.; Wang, M. Q. W.; Wen, J . J .; Bornmann, W.
G.; Crews, C. M. The Anti-angiogenic Agent Fumagillin Co-
valently Binds and Inhibits the Methionine Aminopeptidase,
MetAP-2. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 6099-6103.
(7) Dark, G. G.; Hill, S. A.; Prise, V. E.; Tozer, G. M.; Pettit, G. R.;
Chaplin, D. J . Combretastatin A-4, an Agent That Displays
Potent and Selective Toxicity Toward Tumor Vasculature.
Cancer Res. 1997, 57, 1829-1834.
(8) Pettit, G. R.; Smith, C. R.; Singh, S. B. Recent Advances in the
Chemistry of Plant Antineoplastic Constituents. In Biologically
Active Natural Products; Hostettmann, K., Lea, P. J ., Eds.;
Clarendon Press: Oxford, 1987; pp 105-116.
(9) Pettit, G. R.; Rhodes, M. R.; Herald, D. L.; Chaplin, D. J .;
Stratford, M.; Pettit, R. K.; Chapuis, J .-C.; Oliva, D.; Hamel, E.
Antineoplastic Agents 393. Synthesis of the trans-Isomer of
Combretastatin A-4 Prodrug. Anti-Cancer Drug Des., in press.
(10) Bedford, S. B.; Quarterman, C. P.; Rathbone, D. L.; Slack, J . A.;
Griffin, R. J .; Stevens, M. F. G. Synthesis of Water-soluble
Prodrugs of the Cytotoxic Agent Combretastatin A-4. BioMed.
Chem. Lett. 1996, 6, 157-160.
(11) J ang, M.; Cai, L.; Udeani, G. R.; Slowing, K. V.; Thomas, C. F.;
Beecher, C. W. W.; Fong, H. H. S.; Farnsworth, N. R.; Kinghorn,
A. D.; Mehta, R. G.; Moon, R. C.; Pezzuto, J . M. Cancer
Chemopreventive Activity of Resveratrol, a Natural Product
Derived from Grapes. Science 1997, 275, 218-220.
(15) Cushman, M.; He, H.-M.; Katzenellenbogen, J . A.; Varma, R.
K.; Hamel, E.; Lin, C. M.; Ram, S.; Sachdeya, Y. P. Synthesis of
Analogues of 2-Methoxyestradiol with Enhanced Inhibitory
Effects on Tubulin Polymerization and Cancer Cell Growth. J .
Med. Chem. 1997, 40, 2323-2334.
(16) Mannila, E.; Talvitie, A. Combretastatin Analogues via Hydra-
tion of Stilbene Derivatives. Liebigs Ann. Chem. 1993, 1037-
1039.
(17) Medarde, M.; Pela´ez-Lamamie´ de Clairac, Ramos, A. C.; Cabal-
lero, E.; Lo´pez, J . L.; Gra´valos, D. G.; Feliciano, A. S. Synthesis
and Pharmacological Activity of Combretastatin Analogues,
Naphthylcombretastatins and Related Compounds. BioMed.
Chem. Lett. 1995, 5, 229-232.
(18) Pettit, G. R.; Singh, S. B.; Boyd, M. R.; Hamel, E.; Pettit, R. K.;
Schmidt, J . M.; Hogan, F. Antineoplastic Agents 291. Isolation
and Synthesis of Combretastatins A-4, A-5, and A-6. J . Med.
Chem. 1995, 38, 1666-1672.
(19) Pettit, G. R.; Singh, S. B.; Niven, M. L.; Hamel, E.; Schmidt, J .
M. Isolation, Structure, and Synthesis of Combretastatins A-1
and B-1, Potent New Inhibitors of Microtubule Assembly,
Derived from Combretum caffrum. J . Nat. Prod. 1987, 50, 119-
131.
(20) Pettit, G. R.; Toki, B.; Herald, D. L.; Verdier-Pinard, P.; Boyd,
M. R.; Hamel, E.; Pettit, R. K. Antineoplastic Agents 379.
Synthesis of Phenstatin Phosphate. J . Med. Chem. 1998, 41,
1688-1695.
(21) XPREPsThe automatic space group determination program in
the SHELXTL PC program package (see ref 25).
(22) North, A. C. T.; Phillips, D. C.; Mathews, F. S. Acta Crystallogr.
1968, A24, 351-359.
(23) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;
Burla, M.; Polidori, G.; Camalli, M. SIR92 - A Program for
Automatic Solution of Crystal Structures by Direct Methods;
Dipartimento Geomineralogico, University of Bari: Bari, Italy.
(24) Sheldrick, G. M. SHELXL-97. Crystal Structure Refinement -
DOS/WIN95/NT Version, Release 97-2; Institut Anorg. Che-
mie: Go¨ttingen, Germany, 1998.
(12) Teguo, P. W.; Decendit, A.; Krisa, S.; Deffieux, G.; Vercauteren,
J .; Me´rillon, J .-M. The Accumulation of Stilbene Glycosides in
Vitis vinifera Cell Suspension Cultures. J . Nat. Prod. 1996, 59,
1189-1191.
(13) Bishop, A. C.; Moore, D.; Scanlan, T. S.; Shokat, K. M. Screening
a Hydroxystilbene Library for Selective Inhibition of the B Cell
Antigen Receptor Kinase Cascade. Tetrahedron 1997, 53, 11995-
12004.
(25) Preparation of Figures 1-3 was done with SHELXTL-PC version
5.03, 1994, an integrated software system for the determination
of crystal structures from diffraction data; Siemens Industrial
Automation, Inc., Analytical Instrumenation, Madison, WI
53719.
(26) Sheldrick, G. M. SHELXL93. Program for the Refinement of
Crystal Structures; Institut Anorg. Chemie: Go¨ttingen, Ger-
many, 1993.
(14) Chen, K.; Kuo, S.-C.; Hsieh, M.-C.; Mauger, A.; Lin, C. M.;
Hamel, E.; Lee, K.-H. Antitumor Agents. 178. Synthesis and
Biological Evanuation of Substituted 2-Aryl-1,8-naphthyridin-
4(1H)-ones as Antitumor Agents that Inhibit Tubulin Polymer-
ization. J . Med. Chem. 1997, 40, 3049-3056.
(27) National Committee for Clinical Laboratory Standards. 1997.
Performance standards for antimicrobial disk susceptibility tests,
6th ed.; approved standard M2-A6; NCCLS: Wayne, PA.
J M9807149