Synthesis and cytotoxic evaluation of combretafurazans

Article abstract of DOI:10.1021/jm049096o

Combretastatin A-4 is an antitumoral and antitubulin agent that is active only in its cis configuration. In the present manuscript, we have synthesized cis-locked combretastatins embodying a furazan ring (combretafurazans). To achieve this, we have developed a new strategy that exploits the dehydration of vicinal dioximes using the Mitsunobu reaction. Among the advantages of following such a strategy are the mild conditions used for the construction of the diarylfurazan derivatives, allowing for the presence of highly functionalized substrates and deactivated aromatic rings. Combretafurazans are more potent in vitro cytotoxic compounds compared to combretastatins in neuroblastoma cells, yet maintaining similar structure-activity relationship and pharmacodynamic profiles.

Full text of DOI:10.1021/jm049096o

3260
J. Med. Chem. 2005, 48, 3260-3268
Synthesis and Cytotoxic Evaluation of Combretafurazans
Gian Cesare Tron, Francesca Pagliai, Erika Del Grosso, Armando A. Genazzani, and Giovanni Sorba*
Dipartimento di Scienze Chimiche, Alimentari, Farmaceutiche e Farmacologiche, Universita´ del Piemonte Orientale “A.
Avogadro”, Via Bovio 6, 28100 Novara, Italy
Received November 9, 2004
Combretastatin A-4 is an antitumoral and antitubulin agent that is active only in its cis
configuration. In the present manuscript, we have synthesized cis-locked combretastatins
embodying a furazan ring (combretafurazans). To achieve this, we have developed a new
strategy that exploits the dehydration of vicinal dioximes using the Mitsunobu reaction. Among
the advantages of following such a strategy are the mild conditions used for the construction
of the diarylfurazan derivatives, allowing for the presence of highly functionalized substrates
and deactivated aromatic rings. Combretafurazans are more potent in vitro cytotoxic compounds
compared to combretastatins in neuroblastoma cells, yet maintaining similar structure-activity
relationship and pharmacodynamic profiles.
Introduction
Malignant tumors represent one of the most common
human diseases, and the clinical prognosis remains
relatively poor. This therefore leaves ample space for
the development of new therapeutic strategies for the
improvement of drugs that are currently in use.
Among the current targets for chemotherapy, along-
side DNA, microtubules have a prominent role. Since
tubulin-containing structures are crucial in a number
of cellular functions, such as chromosome segregation
Figure 1. Known SAR of combretastatin derivatives.
during cell division, intracellular transport, cell motility
and the maintenance of cell shape, microtubules have
known.3 CA-A4 has been shown to possess a powerful
represented a good target for chemotherapy, and a
cytotoxic activity against a broad panel of tumoral cell
number of drugs, mainly of natural origin, have
lines4,5 and MDR cells.6 In vivo, CA-A4 displays low or
emerged.1 Microtubules are at a dynamic equilibrium,
no antitumoral activity7 mainly because of low water
with monomeric and polymerized tubulin interchanging,
solubility, and therefore phosphate pro-drugs have
and this swaying process is thought to be fundamental
recently entered into clinical trials.8
for their cellular functions. The ability to disrupt
Because of the interest in these structures, several
microtubule assembly or disassembly makes tubulin
synthetic analogues of CA-A4 have been developed
inhibitors preferential for cells that have a high prolif-
allowing to sketch structure-activity relationships (Fig-
eration rate, such as tumors, since it results in cell
ure 1). The cis configuration of the double bond is
arrest during mitosis.1 Such a dynamic process is also
fundamental as is the 3,4,5-trimethoxy group on ring
highly pronounced in the neovasculature of solid tu-
A. However, recently Gaukroger questioned the impor-
mors, and this represents an additional attraction of
tance of the trimethoxyphenyl group for the antitubulin
tubulin inhibitors in chemotherapy.2 Since inhibition of
activity.9 Ring B seems more amenable to isosteric
polymerization or inhibition of the disassembly leads
substitutions. For instance, substituting the hydroxyl
to similar cellular events, drugs that target either
group with an amine leads to higher cytotoxicity and
process have been developed. For example, taxanes
an increased water solubility through salification.10
(such as paclitaxel and docetaxel) prevent the disas-
By comparing the minimal energy conformations for
sembly of the microtubule and are drugs of choice for
both colchicine and combretastatin, it has been dem-
ovarian and breast cancers, while other antimitotic
onstrated that it is possible to overcome the problem of
drugs of natural origin (such as vinca alkaloids) prevent
the isomerization of the active cis double bond to the
the assembly of the complex.
inactive trans by introducing five-membered hetero-
Among the most recent additions to this class of
cycles in place of the olefin group without substantial
antitumoral agents are combretastatins, first identified
loss of potency.11 These cis-locked analogues provide
in the bark of the South African willow tree Combretum
caffrum. Combretastatin A-4 (CA-A4) 1 appears to be
the most powerful antimitotic agent of this series and
three main advantages: (a) prevention of isomerization
of combretastatin from cis to trans; (b) increased
specificity, since the trans conformation might be rec-
is remarkably simple in its chemical structure (Figure
ognized by other cellular targets; and (c) the possibility
1), while being one of the most potent tubulin inhibitors
to use heterocyclic systems that might improve the
therapeutic potential of these drugs. This is strength-
ened by computational analysis showing that tubulin,
on its colchicine site, can host the compounds with
* To whom correspondence should be addressed. Tel: +39-0321-
375827.
Fax:
+39-0321-375821.
e-mail:
giovanni.sorba@
pharm.unipmn.it.
10.1021/jm049096o CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/13/2005

Evaluation of Combretafurazans
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9 3261
Scheme 1a
a Reagents and conditions: (a) TBDMSiCl, imidazole in CH2Cl2; (b) AD mix R (Sharpless asymmetric dihydroxylation reagent),
methanesulfonamide in H2O/tert-butyl alcohol; (c) NaOCl, KBr, TEMPO in CH2Cl2/H2O; (d) NH2OH‚HCl in pyridine/ethanol at 90 °C; (e)
NaOH in 1,2-propanediol at 160 °C; (f) PPh3, DIAD in toluene at 0 °C then reflux.
different spatial orientations and that three different
binding domains can be involved in the interaction with
these molecules.12 The strategy to improve the efficacy
or potency of combretastatin analogues has been widely
used and a number of heterocyclic compounds, as new
possible bioisosteres of the double bond of the combre-
tastatins, have been prepared.13 These studies have also
shown that loss of activity can be ascribed to an
incorrect orientation of two phenyl rings into the binding
site due to the specific five-membered ring.13 Therefore,
a key structural factor for the cytotoxic activity is the
presence of the double bond which forces the two
aromatic rings to be at an appropriate distance and to
have an optimal dihedral angle to maximize interactions
with tubulin.
In the present manuscript, we have rigidified com-
bretastatin by replacing the olefin group with the
furazan system (1,2,5-oxadiazole) maintaining the tri-
methoxy group on ring A and synthesizing derivatives
(e.g. 3-hydroxy-4-methoxy, 4-methoxy, 3-amino-4-meth-
oxy) on phenyl ring B (combretafurazans). To ac-
complish this task, we have developed a new synthetic
procedure which employs the Mitsunobu reaction to
close the vicinal dioximes. This has also allowed us to
synthesize substituted diphenylfurazans using mild
reaction conditions.
the compounds when tested in neuroblastoma cells,
while maintaining similar structure-activity relation-
ship and pharmacodynamic profiles of combretastatin
A-4.
Chemistry
The combretafurazans derivatives were synthesized
starting from corresponding stilbene derivatives ob-
tained using the Perkin reaction.14 The furazan ana-
logue of CA-A4 was synthesized according to Scheme
1. Protection of the phenolic group of 1 with tert-
butyldimethylsilyl chloride gave a compound which was
dihydroxylated using the protocol developed by Pettit15
to obtain the desired diol 2 along with the diketone 3.
Our attempt to carry out the dihydroxylation using
catalytic osmium tetroxide and morpholine N-oxide
failed, as in our hands we were only able to isolate the
two aldehydes resulting from the cleavage of the double
bond. The diol 2 was oxidized to diketone 3 using
TEMPO and sodium hypochlorite, and the following
transformation into bis-oxime 4 was performed in an
excess of hydroxylamine hydrochloride and pyridine,
heated for 3 days at 90 °C. Higher reaction times and
temperatures gave only one of four possible isomers of
vicinal dioximes, the thermodynamically more stable
isomer (E,E bis-oximes or anti form), and this was
We hereby show that replacing the olefin group with
the furazan moiety increases the cytotoxic potency of

3262 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9
Tron et al.
Scheme 2. Proposed Mechanism for the Formation of Nitrile Byproducts
confirmed using the analytical methods based on the
complex formations between bis-oximes and nickel(II)
salts.16
work successfully,21 and therefore the use of diphenyl-
dioximes which have pKa between 8 and 11 should yield
5a.22
We then attempted to close the vicinal dioximes to
furazan. The traditional synthesis of the furazan ring
system starting from vicinal dioximes involves harsh
reaction conditions and is performed only on unfunc-
tionalized or poorly functionalized substrates.17 Such
difficulty is highlighted by the existence of few such
reactions described in chemical databases (SciFinder
Scholar and Beilstein CrossFire). The two main methods
that have been developed in the literature consist in the
dehydration of vicinal dioximes in basic or acidic
medium18 or in the deoxygenation of the corresponding
furoxans (1,2,5-oxadiazole 2-oxide) with reducing agents
such as triethyl phosphite.19
The dehydrating conditions require a molar excess of
sodium hydroxide in 1,2-ethanediol or a strong acid,
such as p-toluenesulfonic acid at high temperature (ca.
160 °C). Our attempt to carry out this reaction using
p-toluenesulfonic acid at reflux led to the formation of
a plethora of hardly separable compounds. Dehydration
in basic medium successfully formed two main prod-
ucts: the desired compound 5a and a demethylated
product 5b. Nonetheless, due to the presence of byprod-
ucts, even after column chromatography, it was not
possible to obtain 5a with an acceptable purity. Fur-
thermore, since these reactions are unlikely to be
tenable with functionalized substrates or deactivated
aromatic rings, we abandoned such strategies.
We then developed a new protocol that uses the
Mitsunobu reaction20 for the synthesis of furazan via
vicinal dioximes. It is well-known that the Mitsunobu
reaction carried out on 1,2 diols brings epoxide forma-
tion deriving from an intramolecular Mitsunobu reac-
tion. Our hypothesis was that the closure of the vicinal
dioximes, which mimic the 1,2 diol function (4), via the
intramolecular Mitsunobu reaction should directly yield
the furazan ring. This new strategy would allow the use
of milder reaction conditions which would then be more
compatible with functionalized substrates. Usually, the
acidic component of the reaction has to possess a pKa
not higher than 11 to allow protonation of the triphen-
ylphosphine-DIAD adduct in order for the reaction to
The reaction with 4 was performed using toluene as
solvent. After the addition of DIAD at 0 °C, the reaction
was heated at reflux and was completed in 1 h, giving
the desired product 5a. Although the anti configuration
of 4 should not favor the ring closure, it has been shown
that the isomerization of dioximes can occur easily
under various conditions (acid or basic medium, heat,
UV light), and this might provide an explanation for the
success of the reaction.23 Furthermore, during the
course of the reaction, the removal of the protecting
group was also observed. Last, the formation of com-
bretafurazan 5a was accompanied by the formation of
two nitriles derivatives as byproducts (5c, 24%; 5d 30%)
deriving from a Beckmann fragmentation.23
The proposed mechanism for the formation of the two
nitriles is outlined in Scheme 2. The ring closing
reaction competes with the Beckmann fragmentation,
which yields a nitrile derivative and a carbocationic
species which is reduced to nitrile via nitrile N-oxide.24
This hypothesis is supported by the knowledge that the
Beckmann fragmentation competes with transposition
when the carbocation formed is relatively stable.23 This
occurs in the reaction described since the trimethoxy-
phenyl carbocation is stabilized by the three electron-
releasing methoxy groups.
To prove this hypothesis we carried out the Mit-
sunobu reaction on diphenyl dioximes displaying in the
para position: an electron donating group (OMe, 6), no
functional group (7) or an electron-withdrawing group
(Br, 8). Such reactions yielded the desired furazan
systems 6a (51%), 7a (59%), 8a (68%) in fair yield
(Scheme 3). The best yield was obtained, as hypoth-
esized, when the electron-withdrawing group was present
(8a), suggesting that the Beckmann fragmentation
pathway was depleted.25
The synthesis of combretafurazan bearing only a
methoxy group on the B ring is reported in Scheme 4.
We chose to prepare the diketone moiety 10 starting
from corresponding stilbene 9 using Sharpless proto-
col.26 The stilbene 9 was prepared starting from the
corresponding carboxylic acid 3,4,5,-trimethoxy-R-[(4-

Evaluation of Combretafurazans
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9 3263
Scheme 3a
Scheme 5a
a Reagents and conditions: (a) PPh3, DIAD in toluene at 0 °C
then reflux.
methoxyphenyl)methylene]-(E)-benzeneacetic acid.7b Af-
ter the formation of the dioxime 11, we successfully
obtained the corresponding furazan 12a via the Mit-
sunobu reaction. Parallel execution of this reaction with
the dehydration in basic medium gave a significantly
lower yield (18% vs 43%). Interestingly, demethylation
of the 4-methoxy group to give 12b was, of course, not
observed in the Mitsunobu conditions, while it was being
predominant (25% yield) in the basic dehydration
process.
Last, the amine-substituted combretafurazan was
prepared starting from the stilbene derivative 13 as
outlined in Scheme 5. Decarboxylation of 13 with copper
in quinoline gave the desired stilbene 14. After forma-
tion of the diketone 15, we were not able to obtain a
single isomer of the dioxime, but only a mixture (16).
a Reagents and conditions: (a) Cu, quinoline 180 °C; (b) KMnO4
in Ac2O; (c) NH2OH‚HCl in pyridine/ethanol 90 °C; (d) NaOH in
1,2-propanediol at 160 °C; (e) PPh3, di-tert-butyl azodicarboxylate
in toluene at 0 °C then reflux; (f) H2, Pd/C 5% in EtOH.
Attempts to separate the isomers failed, so we used the
mixture without further purification to perform the
dehydrating reaction. The basic dehydration applied to
electron-withdrawing substituted vicinal dioximes led
to a reaction mixture from which no furazan derivative
was isolated. The most likely explanation for this is that
the nitro group deactivates the aromatic ring and
promotes nucleophilic attack both on the phenyl group
and on the methoxy group to give unstable phenolic
derivatives, which are subsequently oxidized. On the
other hand, the desired product (17) was obtained in
Scheme 4a
a Reagents and conditions: (a) KMnO4 in Ac2O; (b) NH2OH‚HCl in pyridine/ethanol at 90 °C; (c) PPh3, DIAD in toluene at 0 °C then
reflux; (d) NaOH in 1,2-propanediol at 160 °C.

3264 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9
Tron et al.
Scheme 6a
compound 12b was assessed, which displays both the
demethylation at the para position of ring A and the
dehydroxylation in meta of ring B, a further substantial
loss of activity was identifiable compared to 5a (140-
fold). Finally, the meta position of ring B was either
substituted with an electron-withdrawing nitro group
(17) or an electron-donating amino group (18). Indeed,
compound 17 displayed a 50-fold drop in activity while
the potency of compound 18 reflected that of 5a. In
conclusion, this preliminary assay shows that the
substitution of the double bond with the furazan ring
increases the potency of combretastatin. Furthermore,
the combretafurazans display a similar structure-
activity relationship and relative rank order of potency
compared to the original structure.13 For example,
demethylations on the A ring generate a loss of potency
both in combretafurazans (5b) and combretastatins.10
Similarly, substituting the hydroxyl with the Grimm’s
isostere amino group (18) does not result in any
significant difference in the activity of either structure,10
while substituting the same position with an electron-
withdrawing group (17) causes a loss of potency in both
combretastatin and combretafurazan.10 Such similarity
would suggest that combretafurazans elicit their cyto-
toxicity with a similar mechanism of action compared
to combretastatin. To establish a qualitative link be-
tween combretastatin and these compounds, we per-
formed immunocytochemistry on cells treated with the
two most potent combretafurazans, 5a and 18, to
establish whether these were able to mimic the reported
ability of combretastatin to inhibit microtubule struc-
ture in cultured cells.2,13 Indeed, while control cells
displayed the typical weblike tubulin architecture (Fig-
ure 3a), cells treated with 1, 5a or 18 displayed a
disorganized tubulin structure and a change in gross
morphology (Figure 3b-d). To confirm this finding, cells
were grown in the presence of 1, 5a or 18, and proteins
were extracted in the presence of paclitaxel, which
prevents further tubulin rearrangements. Western blot-
ting of the pelletable fraction and the soluble fraction
then allows distinction between the polymerized and
free form of tubulin.27 Indeed, while in control cells
tubulin was present in similar amounts in the polym-
erized (pellet) and free form (supernatant), the equilib-
rium was significantly shifted toward the free form
when cells were treated with 1 or with combretafura-
zans (Figure 3e). This is yet further evidence that
inhibition of tubulin polymerization represents the
primary mechanism of action of combretafurazans.
a Reagent and conditions: (a) TBAF sol. 1 M in THF.
good yield (40%) with the Mitsunobu reaction. This
strongly suggests that this strategy is an improvement
to the published reactions to perform dehydrating
reactions to form furazan ring systems from phenyl
dioximes bearing a deactivating group. In this last
reaction, the Mitsunobu reaction was performed replac-
ing DIAD with di-tert-butyl diazodicarboxylate to allow
an easier purification of the furazan from hydrazodi-
carboxylate. Reduction of the nitro compound with
hydrogen and Pd/C 5% gave the final amino derivative
18.
Biological Results and Discussion
All final products (5a, 5b, 12a, 12b, 17, 18) were
tested in cytotoxicity assays. Furthermore, we also
decided to test the two main intermediates, the diketone
19 and the bis-oxime 20, after removal of their protect-
ing groups (Scheme 6). Cytotoxicity was evaluated on a
neuroblastoma cell line, SH-SY5Y, and cells were
treated for 48 h with increasing concentrations of
combretastatin A-4 (1) or combretafurazans to obtain
an approximate IC50 value. Preliminary data on other
experimental time points (i.e. 24 and 72 h) yielded
similar results (data not shown). The results (Table 1)
show that treatment with combretastatin A-4 was
capable of significantly reducing tumoral cell viability
at nM concentrations with an approximate IC50 of 5.8
nM. The synthetic intermediates 19 and 20 lost ap-
proximately 90- and 40-fold activity compared to 1,
although they still remained cytotoxic. In our opinion
this signifies that the vicinal dioxime or diketone
reduces the affinity of the compound for tubulin, reveal-
ing the importance of the correct alignment of the two
phenyl rings. Strikingly, compound 5a, where the di-
oxime was closed via the Mitsunobu reaction, displayed
an IC50 approximately 4-fold lower than combretastatin,
suggesting that the conformational block of ring A and
B elicited by the furazan ring favors tumoral cell
cytotoxicity. Compound 5a also displayed a similar Hill
slope, vaguely suggesting a similar mechanism of action
(Figure 2). When compound 5b was tested, which bears
demethylation at the para position of ring A, a signifi-
cant loss of activity compared to 5a was evident (ap-
proximately 12-fold). When compound 12a was tested,
which bears meta dehydroxylation of ring B, a similar
loss of potency was observed. As might be expected when
Last, the cytotoxic nature of these compounds and the
ability to inhibit tubulin might suggest that these drugs
might trigger apoptosis in a cell-cycle specific manner,13
causing cell arrest in the G2/M phase. To establish if
these assumptions were correct, we aimed at verifying
whether treatment of cells with these compounds could
activate caspase-3, an effector of apoptosis.28 This
enzyme is present in cells as a 33 KDa protein, and is
cleaved to an active 17 KDa protein when the apoptotic
program is initiated.28 In control cells, while the inactive
protein was present, there was no band corresponding
to the active fragment. On the contrary, cells treated
with combretastatin or combretafuranzans displayed
the active fragment suggesting that programmed cell
death is responsible for their cytotoxic effects (Figure

Evaluation of Combretafurazans
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9 3265
Table 1. Cytotoxicity of Combretastatin and Combretafurazans on SH-SY5Y Neuroblastoma Cellsa
1
5a
5b
12a
12b
17
18
19
20
IC50 (nM) n
5.8 (16)
1.4 (12)
17.5 (12)
8.9 (12)
196 (12)
72.5 (12)
1.6 (20)
524 (8)
254 (8)
a IC50 values have been generated by a curve-fitting program (Kaleidograph). Full concentration-response curves for compounds 1, 5a
and 18 are illustrated in Figure 2. The number of 8-point concentration-response curves performed for each sample is in parentheses.
Figure 2. Concentration-response curve of combretastatin
(1) and of the two most potent combretafurazans synthesized.
Values are mean ( SEM of 12-24 determinations in 3-6
separate experimental days. * p < 0.01 if compared to 5a and
vs 18 (ANOVA followed by Student’s t-test)
4). These cells also displayed an increased expression
of procaspase-3, suggesting that up-regulation of this
protein might also take place. Last, cell cycle effects of
these compounds were assessed using propidium iodide
staining. Indeed, while control cells were mainly in G1
phase, cells treated with combretafurazans or combre-
tastatins resulted in G2/M cell arrest (Figure 5). Such
arrest was time-dependent with accumulation in the
tetraploid peak observable after 6 h and reaching a
plateau after 16 h, after which cell death rendered the
experiment difficult to interpret and perform (data not
shown).
Figure 3. Combretastatin and combretafurazans affect tu-
bulin polymerization (a-d). Bright field and immunfluores-
cence using an anti-tubulin antibody of control cells (a) or cells
treated for 24 h with combretastatin (3 nM, b), 5a (3 nM, c),
or 18 (3 nM, d). Brightfield images have been obtained with
an 20× objective, while immunofluorescence images have been
obtained with a 100×/oil immersion objective. (e) Western blot
of tubulin extracted in the presence of paclitaxel from neuro-
blastoma cells treated with the indicated compounds all at 50
nM for 16 h. Results are representative of three separate
experiments which yielded comparable results. P represents
the pelletable (polymerized) fraction while S represents su-
pernatant (free) fraction of tubulin.
Conclusions
In conclusion, we have successfully used a new
protocol to synthesize functionalized diphenylfurazan
compounds starting from vicinal dioximes using mild
reaction conditions. This method represents the first
example of using the Mitsunobu reaction for the con-
struction of furazan rings. Replacing the double bond
with the furazan moiety leads to slightly more potent
compounds as assessed in cytotoxicity assays. Further-
more, in neuroblastoma cells, rigidifying combretastatin
with a furazan group does not change the reported
structure-activity relationship of this natural product.10
Last, these compounds appear to elicit their tumor
cytoxicity in a fashion similar to combretastatin, via
inhibition of tubulin polymerization, which then leads
to cell cycle arrest in G2/M and subsequent apoptosis.
Melting points were determined in open glass capillary with
a Stuart scientific SMP3 apparatus and are uncorrected. All
the compounds were checked by IR (FT-IR THERMO-NICO-
1
LET AVATAR); H and 13C APT (JEOL ECP 300 MHz) and
mass spectrometry (Thermofinningan LCQ-deca XP-PLUS and
Thermofinningan TRACE GC-POLARIS Q). Chemical shifts
are reported in part per million (ppm). Column chromatogra-
phy was performed on silica gel (Merck Kieselgel 70-230 mesh
ASTM) using the indicated eluants. Thin-layer chromatogra-
phy (TLC) was carried out on 5 × 20 cm plates with a layer
thickness of 0.25 mm (Merck Silica gel 60 F254). When
necessary they were developed with KMnO4 or vanillin re-
agent. Elemental analysis (C, H, N) of the target compounds
was performed by REDOX (Monza, Italy) and are within
Experimental Section
Chemistry. Commercially available reagents and solvents
were used without further purification and were purchased
from Fluka-Aldrich or Lancaster. Tetrahydrofuran (THF) was
distilled immediately before use from Na/benzophenone under
a slight positive atmosphere of N2, and toluene was dried by
distillation from sodium and stored on activated molecular
sieves (4 Å). When needed the reactions were performed in
flame- or oven-dried glassware under a positive pressure of
dry N2.
(0.4% of the calculated values unless otherwise noted.
29
Compounds 1,14
6
29, 7
and 137b were synthesized as
described previously.
(1S,2R)-1-(3-{[tert-Butyl(dimethyl)silyl]oxy}-4-methoxy-
phenyl)-2-(3,4,5-trimethoxyphenyl)-1,2-ethanediol (2). AD
mix-R (83 g) was added to a biphasic solution of 200 mL of

3266 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9
Tron et al.
of the solvent, the crude vicinal dioxime was purified by
column chromatography using PE/EtOAc 4:6 as eluant to
obtain 3.3 g of 4 as a colorless gum. (85%).
2-Methoxy-5-[4-(3,4,5-trimethoxyphenyl)-1,2,5-oxadia-
zol-3-yl]phenol (5a) and 4-[4-(3-Hydroxy-4-methoxyphen-
yl)-1,2,5-oxadiazol-3-yl]-2,6-dimethoxyphenol (5b). Basic
Dehydration Method. To a solution of 300 mg of 4 (0.61
mmol) in 1,2-propanediol (4 mL) was added 244 mg of sodium
hydroxide (6.1 mmol: 10 equiv). The resulting suspension was
heated at 150 °C for 16 h and worked up by diluition with
EtOAc and washed sequentially with 2 N HCl and brine. After
drying over sodium sulfate and evaporation of the solvent, the
residue was purified by column chromatography using PE/
EtOAc 8:2 to elute 5a and 5b. The obtained compounds were
then washed with EtOH to give 5a (63 mg, 29%) and 5b (14
mg, 7%) as white powders. Mitsunobu Dehydration Method.
To a cooled (0 °C) suspension of 4 (300 mg; 0.61 mmol) in dry
toluene (4 mL) was added triphenylphosphine (400 mg, 1.53
mmol, 2.5 equiv). DIAD (301 µL, 309 mg, 1.53 mmol, 2.5 equiv)
was then added dropwise, and the resulting solution was
heated at reflux for 1 h. The reaction was worked up by
evaporation of the solvent, and the residue was purified by
column chromatography. Elution with PE/EtOAc 9:1 gave 35
mg of 3,4,5-trimethoxybenzonitrile (5d, 30%). The solvent was
then changed to PE/EtOAc 8:2 to elute 42 mg of 3-hydroxy-
4-methoxybenzonitrile (5c, 24%). Finally, the eluant was
changed to PE/EtOAc 7:3 to give 52 mg of 5a as a white powder
(24%).
Figure 4. Combretastatin and combretafurazans activate
caspase-3. Western blot of samples extracted from neuroblas-
toma cells treated for 24 h with the indicated compounds (all
at 3 nM). Results are representative of three separate experi-
ments.
1,2-Bis(4-bromophenyl)-1,2-ethanedione Dioxime (8).
To a solution of 1,2-bis(4-bromophenyl)-1,2-ethanedione (2 g,
5.45 mmol) in dry pyridine (20 mL) and ethanol (20 mL) was
added an excess of hydroxylamine hydrochloride (3.75 g, 54.5
mmol; 10 equiv). The resulting solution was heated at 90 °C
for 2 days and worked up by dilution of EtOAc and washing
with 2 N HCl (× 3), water (× 1) and brine (× 1). After drying
over sodium sulfate and evaporation of the solvent, the crude
product was purified by column chromatography using PE/
EtOAc 8:2 as eluant to obtain 0.76 g of 8 as a colorless gum
(35%).
Figure 5. Combretastatin and combretafurazans induce cell
cycle arrest. Cell cycle analysis of neuroblastoma cells treated
for 16 h with vehicle (a), 1 (b, 10 nM), 5a (c, 10 nM), and 18
(d, 10 nM). Data are representative of three separate experi-
ments. Similar results were obtained in separate experiments
when drugs were incubated for 8 or 24 h. The Y-axis represents
cell number and the X-axis represents fluorescence on a linear
scale.
2,3-Dimethoxy-5-[(Z)-2-(4-methoxyphenyl)ethenyl]phen-
yl Methyl Ether (9). To 3,4,5,-trimethoxy-R-[(4-methoxyphen-
yl)methylene]-(E)-benzeneacetic acid7b (19 g, 0.055 mol) dis-
solved in quinoline (200 mL) was added copper powder (16.7
g; 5 equiv). The resulting mixture was heated at 220 °C for 3
h. After cooling, the copper was filtered off through a Celite
pad. The filtrate was washed with concentrated hydrochloric
acid (20 mL), and the aqueous layer was filtered again through
a Celite pad and extracted several times with EtOAc. The
combined organic layers were washed with saturated aqueous
sodium carbonate and brine. After drying over sodium sulfate
and removal of the solvent, the residue was purified by column
chromatography using as eluant PE/EtOAc 9:1 to give 9 as
an amorphous solid (14.7 g, 89%).
1-(4-Methoxyphenyl)-2-(3,4,5-trimethoxyphenyl)-1,2-
ethanedione (10). To a cooled (0 °C) and stirred solution of
9 (3 g, 0.02 mol) in acetic anhydride (60 mL) was added
portionwise potassium permanganate (12.64 g, 0.08 mol, 4
equiv). During the addition the temperature was increased to
20 °C. After 1 h the reaction was cooled at 0 °C and was worked
up by dilution with water. Sodium sulfite was added until all
potassium permanganate was reduced. The mixture was
extracted with EtOAc, and the organic phases were washed
respectively with saturated aqueous NaHCO3 and brine and
dried over sodium sulfate. Evaporation of the solvent gave a
crude product, which was recrystallized with PE/EtOAc 8:2,
obtaining 1.84 g of diketone 10 as a yellow solid (28%).
tert-butyl alcohol and 200 mL of water. After its dissolution,
methansulfonamide (4.6 g) was added and the contents were
cooled to 0 °C. cis-Silyl ether combretastatin A-4 (20 g; 0.046
mol) was added and the reaction mixture stirred vigorously
in the dark. After 30 h, sodium sulfite was added (10 g), 30
min later the reaction was diluted with EtOAc, and the layers
were separated. The aqueous layer was further washed with
EtOAc (× 4), and the combined organic extracts were washed
with 2 M NaOH (× 2) and brine (× 1). After drying over sodium
sulfate and evaporation of the solvent, the crude product was
purified by column chromatography using PE/EtOAc 8:2 to
give 5.19 g of corresponding diketone 3 (25%). Finally the
eluant was replaced with PE/EtOAc 5:5 to elute 6.2 g of diol 2
as an amorphous gum (29%).
1-(3-{[tert-Butyl(dimethyl)silyl]oxy}-4-methoxyphenyl)-
2-(3,4,5-trimethoxyphenyl)-1,2-ethanedione (3). To a solu-
tion of 2 (1.5 g; 3.23 mmol) in dry CH2Cl2 were added TEMPO
(50 mg, 0.323 mmol, 0.1 equiv) and KBr (576 mg, 4.84 mmol,
1.5 equiv). The resulting solution was then cooled at 0 °C, and
commercial sodium hypochlorite at 5% was added under
vigorous stirring (573.5 mg, 7.75 mmol, 2.4 equiv) to yield a
yellow solution. After 10 min, the reaction was worked up by
dilution with CH2Cl2 and washed sequentially with water, 2
N HCl and brine. The solution was then evaporated and
purified by column chromatography using PE/EtOAc 9:1.
Compound 3 was obtained as a yellow powder (1.4 g, 94%).
1-(3-{[tert-Butyl(dimethyl)silyl]oxy}-4-methoxyphenyl)-
2-(3,4,5-trimethoxyphenyl)-1,2-ethanedione Dioxime (4).
To a solution of 3 (5.86 g, 12.75 mmol) in dry pyridine (58 mL)
and ethanol (76 mL) was added an excess of hydroxylamine
hydrochloride (8.92 g, 127.54 mmol; 10 equiv). The resulting
solution was heated at 90 °C for 3 days, worked up by dilution
of EtOAc and washed with 2 N HCl (× 3), water (× 1) and
brine (× 1). After drying over sodium sulfate and evaporation
1-(4-Methoxyphenyl)-2-(3,4,5-trimethoxyphenyl)-1,2-
ethanedione Dioxime (11). To a solution of 10 (3 g, 9.0
mmol) in dry pyridine (30 mL) and ethanol (40 mL) was added
an excess of hydroxylamine hydrochloride (6.25 g, 90 mmol;
10 equiv). The resulting solution was heated at 90 °C for 3
days and worked up by dilution of EtOAc and washing with 2
N sulfuric acid (× 3), water (× 1) and brine (× 1). After drying
over sodium sulfate and evaporation of the solvent, the crude

Evaluation of Combretafurazans
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9 3267
product was purified by column chromatography using PE/
EtOAc 5:5 as eluant to obtain 3 g of 11 as a colorless gum
(95%).
EtOAc 9:1 afforded 3,4,5-trimethoxybenzonitrile along with di-
tert-butyl hydrazoazodicarboxylate. The eluant was then
changed to PE/EtOAc 8:2 and PE/EtOAc 7:3 to give 17 as a
yellow solid (453 mg, 40%). The solid was crystallized from
ethanol.
2-Methoxy-5-[4-(3,4,5-trimethoxyphenyl)-1,2,5-oxadia-
zol-3-yl]aniline (18). A mixture of 17 (166 mg, 0.428 mmol),
5% palladium on carbon (100 mg) and 20 mL of EtOH was
equipped with a balloon of hydrogen gas and stirred at room
temperature. After 2 h, the reaction mixture was filtered
through a pad of Celite. The filtrate was evaporated and the
residue was purified by column chromatography using as
eluant PE/EtOAc 7:3 to give 18 as a white powder (123 mg,
80%).
1-(3-Hydroxy-4-methoxyphenyl)-2-(3,4,5-trimethoxyphen-
yl)-1,2-ethanedione (19). To a solution of 3 (379 mg, 0.824
mmol) in dry THF (3 mL) was added TBAF (1 M solution in
THF) (824 µL, 0.824 mmol, 1 equiv). The solution became
rapidly red. After 5 min, the reaction was worked up by
diluition of EtOAc and washed with saturated aqueous NH4-
Cl. The water phase was extracted once again with EtOAc.
The combined organic phases were then washed with brine
and dried over sodium sulfate. After evaporation, the crude
dione was purified by column chromatography using as eluant
PE/EtOAc 8:2 to give 19 as yellow powder (150 mg, 40%).
1-(3-Hydroxy-4-methoxyphenyl)-2-(3,4,5-trimethoxyphen-
yl)-1,2-ethanedione Dioxime (20). To a solution of 4 (150
mg, 0.31 mmol) in dry THF (1.5 mL) was added TBAF (1 M in
THF) (380 µL, 0.31 mmol, 1 equiv). The solution became
rapidly red. After 2 h, the reaction was worked up by dilution
with EtOAc and washed with saturated aqueous NH4Cl. The
water phase was extracted a second time with EtOAc. The
organic phases collected were then washed with brine and
dried over sodium sulfate. After evaporation, the crude was
purified by column chromatography using as eluant PE/EtOAc
8:2 to give 20 as amorphous solid (108 mg, 93%).
Cell Culture and Cytotoxicity Assay. The SH-SY5Y
human neuroblastoma cell line was obtained from ATCC (LGC
Promochem Teddington, UK) and cultured in 50% DMEM and
50% F-12 supplemented with 10% foetal bovine serum, 2 mM
L-glutamine, penicillin (100 µg/mL), and streptomycin (100 µg/
mL). For cytotoxicity assay, cells were plated on 24-well plates
and grown for 24, 48 or 72 h in the presence or absence of
combretastatin or combretafurazans. On the experimental day,
cells were washed twice in Locke’s solution (134 mM NaCl, 5
mM KCl, 4 mM NaHCO3, 10 mM HEPES [pH 7.6], 2.3 mM
CaCl2, 1 mM MgCl2, 5 mM sucrose) and incubated for 1 h with
MTT (250 µg/mL in Locke’s solution) at 37 °C. Reactions were
then stopped and the crystals solubilized in isopropyl alcohol/
HCl before being read at 570 nm in a spectrophotometer. To
determine IC50 values, data were plotted and fitted using the
Kaleidagraph software (Synergy software, Reading, PA).
Immunocytochemistry. SH-SY5Y were plated on glass
coverslips and grown to subconfluency in the presence or
absence of drugs for 24 h. Cover slips were then fixed in
paraformaldehyde (3.7%) for 30 min at 4°C. Cells were then
washed once in phosphate buffered saline (PBS) and perme-
abilized in 0.1% Triton-X100 for 10 min at room temperature.
Cells were then washed twice in PBS and blocked with 10%
horse serum. Anti-tubulin primary antibody (Sigma-Aldrich;
1:1000) was incubated overnight, and after three further
washes, coverslips were incubated with a chicken anti-mouse
secondary Texas Red-conjugated antibody (Molecular Probes,
1:250). Slides were then visualized in a Nikon upright fluo-
rescence microscope.
3-(4-Methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)-1,2,5-
oxadiazole (12a) and 2,6-Dimethoxy-4-[4-(4-methoxyphen-
yl)-1,2,5-oxadiazol-3-yl]phenol (12b). Basic Dehydration
Method. To a solution of 300 mg of 11 (0.833 mmol) in 1,2-
propanediol (4 mL) was added 333 mg of sodium hydroxide
(8.33 mmol; 10 equiv). The resulting suspension was heated
at 150 °C for 16 h and worked up by dilution with EtOAc and
washed sequentially with 2 N HCl and brine. After drying over
sodium sulfate and evaporation of the solvent, the residue was
purified by column chromatography (PE/EtOAc 9:1 to elute
12a and PE/EtOAc 8:2 to elute 12b). The compounds obtained
were then washed with EtOH to give 12a (52 mg, 18%) and
12b (67 mg, 25%) as white powders. Mitsunobu Dehydra-
tion Method. To a cooled (0 °C) suspension of 11 (200 mg;
0.55 mmol) in dry toluene (3 mL) was added triphenylphos-
phine (216.3 mg, 0.825 mmol, 1.5 equiv). DIAD (162.4 µL, 166.8
mg, 0.825 mmol, 1.5 equiv) was then added dropwise, and the
resulting solution was heated at reflux for 1 h. The reaction
was worked up by evaporation of the solvent, and the residue
was purified by column chromatography. Elution with PE/
EtOAc 95:5 gave 15 mg of 4-methoxybenzonitrile (34%). The
solvent was then changed to PE/EtOAc 9:1 to elute 18 mg of
3,4,5-trimethoxybenzonitrile (17%) and 80 mg of 12a as a white
powder (43%).
2,3-Dimethoxy-5-[(Z)-2-(4-methoxy-3-nitrophenyl)ethen-
yl]phenyl Methyl Ether (14). To the carboxylic acid 13 (5
g, 0.013 mol) dissolved in quinoline (50 mL) was added copper
powder (4.26 g). The resulting mixture was heated at 180 °C
for 2.5 h. After cooling, the copper was filtered off through a
Celite pad. The filtrate was washed with concentrated hydro-
chloric acid (20 mL), and the aqueous layer was filtered again
through a Celite pad and extracted several times with EtOAc.
The combined organic layers were washed with saturated
aqueous sodium carbonate and brine. After drying over sodium
sulfate and removal of the solvent, the residue was purified
by column chromatography using as eluant PE/EtOAc 7:3 to
give 14 as brownish solid (1.83 g, 41%).
1-(4-Methoxy-3-nitrophenyl)-2-(3,4,5-trimethoxyphenyl)-
1,2-ethanedione (15). To a cooled (0 °C) and stirred solution
of 14 (1 g, 2.89 mol) in acetic anhydride (22 mL) was added
portionwise potassium permanganate (1.8 g, 11.59 mmol, 4
equiv). During the addition the temperature increased to 4 °C.
After 30 min the reaction was cooled at 0 °C and was worked
up by dilution with water; solid sodium sulfite was added until
all potassium permanganate was reduced. The compound was
then extracted twice with EtOAc, and the combined organic
extracts were washed with saturated aqueous NaHCO3 (× 2)
and brine and dried over sodium sulfate. Evaporation of the
solvent gave a crude semisolid residue, to which ethanol was
added to decompose excess acetic anhydride. The residue so
obtained was then crystallized with MeOH to give 475 mg of
diketone 15 as a yellow solid (44%).
1-(4-Methoxy-3-nitrophenyl)-2-(3,4,5-trimethoxyphenyl)-
1,2-ethanedione Dioxime (16). To a solution of 15 (2 g, 5.33
mmol) in dry pyridine (58 mL) and ethanol (76 mL) was added
an excess of hydroxylamine hydrochloride (3.73 g, 53.33 mmol;
10 equiv). The resulting solution was heated at 90 °C for 2
days and worked up by dilution with EtOAc and washed
respectively with 2 N HCl (× 3), water (× 1) and brine (× 1).
After drying over sodium sulfate and evaporation of the
solvent, the crude dioxime was purified by column chroma-
tography using PE/EtOAc 5:5 as eluant to obtain 1.86 g of 22
as a reddish gum (86%).
Tubulin Polymerization Assay. To measure the degree
of tubulin polymerization, we used an adaptation of the
methods described by Minotti et al.27 In brief, cells were grown
in 75 cm2 flasks in the presence or absence of drugs for 16 h.
Cells were then tripsinised and centrifuged twice at 600g for
5 min. Cells were then resuspended in 70 µL of of hypotonic
buffer (20 mM Tris-Hcl pH 6.8, 1 mM MgCl2, 2 mM EGTA,
protease inhibitors, 0.5% Igepal) containing 4 µg/mL paclitaxel.
Lysates were incubated for 10 min at room temperature and
3-(4-Methoxy-3-nitrophenyl)-4-(3,4,5-trimethoxyphenyl)-
1,2,5-oxadiazole (17). To a cooled (0 °C) suspension of 16
(1.205 g; 2.97 mmol) in dry toluene (20 mL) were added
triphenylphosphine (1.56 g, 5.95 mmol, 2 equiv) and di-tert-
butyl azodicarboxylate (1.37 g, 5.95 mmol, 2 equiv). The
resulting solution was heated at reflux for 1 h. The reaction
was worked up by evaporation of the solvent, and the residue
was purified by column chromatography. Elution with PE/

3268 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 9
Tron et al.
(8) (a) Chaplin, D. J.; Pettit, G. R.; Hill, S. A. Anti-vascular
approaches to solid tumor therapy: evaluation of combretastatin
A4 phosphate. Anticancer Res. 1999, 19, 189-196. (b) West CM,
Price P. Combretastatin A-4 phosphate. Anticancer Drugs 2004,
15, 179-187.
(9) Gaukroger, K.; Hadfield, J. A.; Lawrence, N. J.; Nolan, S.;
McGown, A. T. Structural requirements for the interaction of
combretastatins with tubulin: how important is the trimethoxy
unit? Org. Biomol. Chem. 2003, 1, 3033-3037.
(10) For SAR studies on CA-4A see: (a) Cushman, M.; Nagarathnam,
D.; Gopal, D.; Chakraborti, A. K.; Lin, C. M.; Hamel, E. Synthesis
and Evaluation of Stilbene and Dihydrostilbene Derivatives as
Potential Anticancer Agents That Inhibit Tubulin Polymeriza-
tion. J. Med. Chem. 1991, 34, 2579-2588. (b) Cushman, M.;
Nagarathnam, D.; Gopal, D.; He, H. M.; Lin, C. M.; Hamel, E.
Synthesis and evaluation of analogues of (Z)-1-(4-methoxyphen-
yl)-2-(3,4,5-trimethoxyphenyl)ethane as potential cytotoxic and
antimitotic agents. J. Med. Chem. 1992, 35, 2293-2306. (c)
Medarde, M.; Ramos, A.; Caballero, E.; Pelaez-Lamamie de
Clairac, R.; Lopez, J. L.; Gravalos, D. G.; San Feliciano, A.
Synthesis and antineoplastic activity of combretastatin ana-
logs: heterocombretastatins. Eur. J. Med. Chem. 1998, 33, 71-
77. (d) Hatanaka, T.; Fujita, K.; Ohsumi, K.; Nakagawa, R.;
Fukuda, Y.; Nihei, Y.; Suga, Y.; Akiyama, Y.; Tsuji, T. Novel
B-ring modified combreastatin analogues: syntheses and anti-
neoplastic activity. Bioorg. Med. Chem. Lett. 1998, 8, 3371-3374.
(11) Ohsumi, K.; Hatanaka, T.; Fujita, K.; Nakagawa, R.; Fukuda,
Y.; Nihei, Y.; Suga, Y.; Morinaga, Y.; Akiyama, Y.; Tsuji, T.
Syntheses and antitumor activity of cisrestricted combretast-
atins: 5-membered heterocycles analogs. Bioorg. Med. Chem.
Lett. 1998, 8, 3153-3158.
(12) ter Haar, E.; Rosenkranz, H. S.; Hamel, E.; Day, B. W.
Computational and Molecular Modeling Evaluation of the
Structural Basis for Tubulin Polymerization Inhibition by
Colchicine Site Agents. Bioorg. Med. Chem. 1996, 4, 1659-1671.
(13) Nam, N. H.; Combretastatin A-4 Analogues as Antimitotic
Antitumor Agents. Curr. Med. Chem. 2003, 10, 1697-1722 and
references therein.
(14) Gaukroger, K.; Hadfield, J. A.; Hepworth, L. A.; Lawrence, N.
J.; McGown, A. T. Novel Syntheses of Cis and Trans Isomers of
Combretastatin A-4. J. Org. Chem. 2001, 66, 8135-8138.
(15) Pettit, G. R.; Toki B. E.; Herald, D. L.; Boyd, M. R.; Hamel, E.;
Pettit, R. K.; Chapuis, J. C. Antineoplastic agents. 410. Asym-
metric hydroxylation of trans-combretastatin A-4. J. Med. Chem.
1999, 42, 1459-1465.
then vortexed. Lysates were then corrected for protein amounts
(Bradford assay, Sigma-Aldrich) and 50 µg (corresponding to
50 µls) was centrifuged at 13000 rpm for 15 min at room
temperature. Supernatant and pellet were then resuspended
in equal volumes of SDS-loading buffer and run on a 10%
SDS-PAGE polyacrylamide gel. After transfer of proteins to
nitrocellulose (blocked in 5% milk), tubulin was identified with
an anti-tubulin primary antibody (1:1000, Sigma-Aldrich) and
anti mouse secondary antibody peroxidase-conjugated (1:8000,
Amersham Bioscience) and visualized by chemiluminescence
(Supersignal WestPico, Pierce).
Assessment of Caspase-3 Activity. Cells were treated for
24 h in the presence or absence of the desired compounds. Cells
were then resuspended in 10 mM Tris [pH 7.5], 150 mM NaCl,
5 mM EDTA [pH 8.0], 10 µg/mL of aprotinin, 10 µg/mL of
leupeptin and 10 µg/mL of pepstatin A, 1 mM PMSF, 0.2 M
sodium orthovanadate, 0.2 M sodium fluoride and 1% Triton
X-100 and underwent three freeze-thaw cycles in dry ice.
Lysates were the centrifuged at 13 000 rpm for 30 min at 4°
C. Supernatants were then collected, and protein was deter-
mined. A 25 µg amount of protein was run on a 15% SDS-
PAGE polyacrylamide gel and transferred onto a nitrocellulose
membrane. Procaspase-3 and caspase-3 were identified with
an anti-caspase 3 primary antibody (1:1000, BD Pharmingen)
and anti rabbit secondary antibody peroxidase-conjugated (1:
10000, Amersham Bioscience) and visualized by chemilumi-
nescence (Supersignal WestPico, Pierce).
Flow-Cytometric Analysis of Cell-Cycle Status. SH-
SY5Y grown in the presence or absence of compounds for 6,
12, 18 or 24 h were washed once in PBS and resuspended in
1 mL of 30:70 ice cold PBS/EtOH and stored at -20 °C. Cells
were then washed twice in PBS and resuspended in PBS
containing RNAse (100 µg/mL) for 1 h at 37°. DNA was then
stained with a PBS solution containing 5 mM EDTA and 100
µg/mL propidium iodide for 30 min at 4 °C in the dark. Cell
cycle analysis was determined with a FACSVantage SE DiVa
(Becton Dickinson).
Acknowledgment. We dedicate this work, with
deep respect, to Professor Alberto Gasco (University of
Turin) our “maestro” and pioneer in the pharmacochem-
istry of 1,2,5-oxadiazole ring.
(16) Toul, J.; Padrta, J.; Okac, A. Thin-layer and paper chromatog-
raphy of some 1,2-dioximes and their monoximes. J. Chroma-
trogr. 1971, 57, 107-119.
(17) Angeli, A. Zur Kenntnis der Verbindungen Welche die Gruppe
C2N2O2 enthalten. Ber. 1893, 527-530.
(18) Friedrichsen, W. Methoden der Organischen Chemie/(Houben-
Weyl); Bd. E8c.Stuttgart; Thieme: New York, Teil 3, 1994; p 648.
(19) Grundman, C. The specific reduction of furoxans to furazans.
Ber. 1964, 97, 575-578.
(20) Mitsunobu, O. The use of diethyl azodicarboxilate and triph-
enylphosphine in synthesis and transformation of natural
products. Synthesis 1981, 1, 1-28.
(21) Wada, M.; Mitsunobu, O. Intermolecular dehydration between
alcohols and active hydrogen compounds by means of diethyl-
azodicarboxilate and triphenylphosphine. Tetrahedron Lett.
1972, 13, 1279-1282.
(22) Bambenek, M. A.; Pflaum, R. T. The reaction of nickel with
dioximes. Inorg. Chem. 1963, 2, 289-292.
(23) McCarty, C. G. In The Chemistry of the carbon-nitrogen double-
bond; Patai, S., Ed.; Interscience Publishers: New York, 1970;
pp 383-392
(24) (a) Smith, L. I. Aliphatic diazo compounds, nitrones, and
structurally analogous compounds. Chem. Rev. 1938, 23, 193-
285. (b) Grundman, C.; Frommeld, H. D. Nitrile oxides. III.
Reduction of Nitrile Oxides to Nitriles. J. Org. Chem. 1965, 30,
2077-2078.
Supporting Information Available: Characterization
(mp, IR, MS, and 1H and 13C NMR data) of all new compounds
and elemental analyses of all target compounds are available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Jordan, A.; Hadfield, J. A.; Lawrence, N. J.; McGown, A. T.
Tubulin as a target for anticancer drugs: agents which interact
with the mitotic spindle. Med. Res. Rev. 1998, 18, 259-96.
(2) Griggs, J.; Metcalfe, J. C.; Hesketh, R. Targeting tumor vascu-
lature: the development of combretastatin A4. Lancet Oncol.
2001, 2, 81-87.
(3) Pettit, G. R.; Cragg, G. M.; Singh, S. B. Antineoplastic agents,
122. Constituents of Combretum caffrum. J. Nat. Prod. 1987,
50, 386-391.
(4) Pettit, G. R.; Singh, S. B.; Hamel, E.; Lin, C. M.; Alberts, D. S.;
Garcia-Kendall, D. Isolation and structure of the strong cell
growth and tubulin inhibitor combretastatin A-4. Experientia
1989, 45, 209-211.
(5) Nabha, S. M.; Mohammad, R. M.; Wall, N. R.; Dutcher, J. A.;
Salkini, B. M.; Pettit, G. R.; Al Katib, A. M. Evaluation of
combretastatin A-4 prodrug in a non-Hodgink’s lymphoma
xenograft model: preclinical efficacy. Anti-Cancer Drugs 2001,
12, 57-63.
(25) Tron, G. C.; Del Grosso, E.; Sorba, G. Unpublished data.
(26) Sharpless, K. B.; Lauer, R. F.; Repic, O.; Teranishi, A. Y.;
Williams, D. R. Permanganate in Acetic Anhydride. R-Diketones
Directly from Olefins. J. Am. Chem. Soc. 1971, 93, 3303-3304.
(27) Minotti, A. M.; Barlow S. B.; Cabral, F. Resistance to antimitotic
drugs in chinese hamster ovary cells correlates with changes in
the level of polimerized tubulin. J. Biol. Chem. 1991, 266, 3987-
3994
(28) Boatright, K. M.; Salvesen, G. S.; Guy, S. Mechanism of caspase
activation. Curr. Opin. Cell. Biol. 2003, 15, 725-731
(29) Balasubrahmanyam, S. N.; Rajendran, N.Carbon NMR line
assignments in 3,4-di(4′- and/or 4′′-substituted)phenyl-1,2,5-
oxadiazole 2-oxides. Indian J. Chem., Sect. B. 1988, 27B (9),
793-796.
(6) McGown, A. T.; Fox B. W. Differential cytotoxicity of combret-
astatins A-1 and A-4 in two daunorubicin-resistant P388 cell
lines. Cancer Chemother. Pharm. 1990, 26, 79-81.
(7) (a) Dorr, R. T.; Dvorakova, K.; Snead, K.; Alberts, D. S.; Salmon,
S. E.; Pettit, G. R. Antitumor activity of combretastatin-A4-
phosphate, a natural product tubulin inhibitor. Invest. New
Drug. 1996, 14, 131-137. (b) Ohsumi, K.; Nakagawa, R.;
Fukuda, Y.; Hatanaka, T.; Morinaga, Y.; Nihei, Y.; Ohishi, K.;
Suga, Y.; Akiyama, Y.; Tsuji, T. Novel Combretastatin Analogs
Effettive against Murine Solid Tumors: Design and Structure-
Activity Relationships. J. Med. Chem. 1998, 41, 3022-3032.
JM049096O