Structure-activity requirements for flavone cytotoxicity and binding to tubulin

Article abstract of DOI:10.1021/jm970842h

A series of 79 flavones related to centaureidin (3,6,4'-trimethoxy- 5,7,3'-trihydroxyflavone, 1) was screened for cytotoxicity in the NCI in vitro 60-cell line human tumor screen. The resulting cytotoxicity profiles of these flavones were compared for deg

Full text of DOI:10.1021/jm970842h

J . Med. Chem. 1998, 41, 2333-2338
2333
Str u ctu r e-Activity Requ ir em en ts for F la von e Cytotoxicity a n d Bin d in g to
Tu bu lin
J ohn A. Beutler, Ernest Hamel, Arnold J . Vlietinck,† Achiel Haemers,† Padinchare Rajan,† J ames N. Roitman,‡
J ohn H. Cardellina II, and Michael R. Boyd*
Laboratory of Drug Discovery Research and Development, Developmental Therapeutics Program,
Division of Cancer Treatment and Diagnosis, National Cancer Institute, Frederick, Maryland 21702-1201
Received December 15, 1997
A series of 79 flavones related to centaureidin (3,6,4′-trimethoxy-5,7,3′-trihydroxyflavone, 1)
was screened for cytotoxicity in the NCI in vitro 60-cell line human tumor screen. The resulting
cytotoxicity profiles of these flavones were compared for degree of similarity to the profile of 1.
Selected compounds were further evaluated with in vitro assays of tubulin polymerization and
[3H]colchicine binding to tubulin. Maximum potencies for tubulin interaction and production
of differential cytotoxicity profiles characteristic of 1 were observed only with compounds
containing hydroxyl substituents at C-3′ and C-5 and methoxyl groups at C-3 and C-4′.
In tr od u ction
ucts and synthetic and commercially available com-
pounds, all of which had either hydrogen, hydroxyl, or
methoxyl substituents at C-3. The compounds were
tested in the NCI primary in vitro human tumor cell
screen for cytotoxicity, and the resulting dose-response
curves were analyzed for the differential cytotoxicity
pattern12 characteristic of 1 and other antimitotic
standard agents. Selected compounds were further
examined in biochemical assays for inhibition of tubulin
polymerization and inhibition of [3H]colchicine binding
to tubulin.
The cytotoxic potencies (mean panel GI50) of 26
3-methoxyflavones in the NCI human tumor cell line
screen are reported in Table 1. Only compounds 1-3
showed significant cytotoxicity (GI50 < 2 µM) and TGI
(net total growth inhibition12)-Compare correlation
coefficients of >0.7. Several other compounds (e.g., 4-7)
showed modest cytotoxicity (GI50 7-10 µM) but did not
yield TGI-Compare correlation coefficients > 0.5 in
reference to 1.
Results from testing a second series, 15 compounds
with a 3-hydroxyl substituent, are summarized in Table
2. While many showed modest cytotoxicity (the GI50’s
of 27-33 were 4-10 µM), none yielded TGI-Compare
correlation coefficients of >0.5 in reference to 1.
Finally, the results for a third series of 38 flavones
with no substitution at C-3 are listed in Table 3. The
majority of the 3-unsubstituted flavones showed mini-
mal cytotoxicity in the NCI in vitro human tumor cell
assay, and none yielded TGI-Compare correlation
coefficients > 0.5. In this group, only compounds 42
and 43 exhibited GI50 < 10 µM.
Selected compounds were then evaluated for inhibi-
tion of tubulin polymerization in vitro (Table 4). Com-
pounds 1-3 significantly inhibited tubulin assembly at
low drug concentrations (substoichiometric to the assay
tubulin concentration of 10 µM). The IC50 values ob-
tained with 1-3 ranged from 0.8 to 3 µM, similar to
that obtained for colchicine (1.4 µM). Compound 3 was
essentially equipotent with 1 and only slightly less
potent than 2 in both tubulin assays, even though its
potency in the NCI primary screen was one-tenth that
Compounds such as vincristine, vinblastine, and
taxol, which interfere with tubulin polymerization and
cause mitotic arrest, are of proven clinical utility.1 In
addition to these established clinical agents, other
structurally diverse organic compounds are known to
interfere with tubulin polymerization. For example,
colchicine,2 stilbenes,3 and chalcones4 all appear to bind
to a common site on tubulin which partially overlaps a
binding site for podophyllotoxin.5 The Vinca alkaloids,6
dolastatin 10,7,8 the spongistatins,9 and taxol10 have also
been reported to bind to several apparently distinct sites
on tubulin.
We have explored the use of the NCI 60-cell line in
vitro screen, in conjunction with a pattern-matching al-
gorithm, Compare,11,12 as a means of identifying natural
product extracts potentially containing new tubulin-
interactive antimitotic leads.13,14 Stemming from that
effort, we recently reported the identification of the
flavone centaureidin, 1, as a cytotoxic principle of the
plant Polymnia fruticosa and further demonstrated
inhibitory effects of 1 on tubulin polymerization.13 That
first report was followed shortly thereafter by two
independent reports of a second flavone, 2, with similar
properties.15,16
The goal of the present study was to explore the
structural requirements for flavone interactions with
tubulin, both to ascertain potential directions for syn-
thetic lead-optimization studies, as well as to identify
an optimal candidate among currently available com-
pounds for in vivo xenograft studies. We report here
results of comparative in vitro antitumor screening of
79 flavones and subsequent tubulin polymerization
studies with 20 of those compounds.
Resu lts a n d Discu ssion
For this study, we selected a set of flavones from the
NCI repository, augmented by additional natural prod-
† Department of Pharmaceutical Sciences, University of Antwerp,
B-2610 Antwerpen, Belgium.
‡ Agricultural Research Service, USDA, Pacific West Area Regional
Research Center, Albany, CA 94710.
S0022-2623(97)00842-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 05/23/1998

2334 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
Beutler et al.
Ta ble 1. Structure and Cytotoxicity Data for
3-Methoxyflavone Derivatives
Ta ble 3. Structure and Cytotoxicity Data for Flavone
Derivatives Unsubstituted at Position 3
mean
compd C-5 C-6 C-7 C-8 C-3′ C-4′ C-5′ GI50 (µM)
mean
compd C-5
C-6
MeO HO
MeO MeO MeO HO
HO HO
MeO MeO MeO MeO MeO
MeO MeO MeO MeO HO
C-7
C-8 C-3′ C-4′ C-5′ GI50 (µM)
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
HO MeO MeO MeO HO HO
HO MeO HO HO MeO
MeO MeO MeO MeO i-PrO
HO MeO HO MeO MeO
HO MeO MeO MeO
H
H
H
H
4
6
11
15
17
18
19
22
24
26
28
30
30
32
33
36
40
42
43
45
48
54
55
58
59
60
62
68
78
79
89
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
HO
HO
HO
HO
HO
HO
HO
HO
Me
AcO
HO
H
HO
HO
HO
HO
HO
Me
HO
HO
H
HO
MeO
MeO
MeO
H
H
H
H
H
H
H
H
H
H
0.24
0.13
1.7
7.1
7.6
8.3
9.5
12
12
15
19
19
20
24
27
32
32
44
H
H
H
H
H
H
MeO MeO
MeO MeO MeO
HO MeO MeO MeO MeO MeO MeO
HO MeO MeO MeO MeO HO
H
H
H
H
H
HO
H
BnO MeO
H
MeO MeO MeO HO
H
H
Me
H
EtO
HO
AcO
H
H
H
H
HO
HO
H
MeO
MeO
MeO
MeO MeO MeO MeO
MeO MeO MeO MeO
2′,5′-di-MeO
2′,4′-di-MeO
MeO MeO MeO
H
H
H
MeO
2′,3′,4′-tri-MeO
MeO
MeO MeO MeO MeO MeO MeO MeO
HO MeO HO HO
MeO MeO MeO MeO penta-MeO
MeO MeO MeO MeO MeO MeO
H
MeO
MeO MeO MeO MeO
H
HO
H
MeO HO
MeO HO
H
HO
MeO
HO
MeO
H
H
H
H
H
MeO MeO MeO MeO MeO
H
MeO HO
H
HO
H
H
H
H
H
MeO HO
MeO MeO HO
HO MeO
MeO HO
HO
MeO MeO MeO MeO
H
MeO
H
H
MeO HO
MeO MeO
MeO
H
MeO
H
H
H
H
H
2′,3′,4′-tri-MeO
MeO
2′,3′,4′-tri-MeO
Me
MeO
H
H
MeO HO MeO
HO
HO
H
H
MeO HO
54
54
54
65
H
H
H
HO
H
MeO
MeO
MeO
MeO HO
MeO
H
H
H
H
HO
MeO MeO
MeO MeO
HO
MeO
MeO
H
H
H
H
H
H
H
MeO MeO
H
H
MeO HO
HO
HO
MeO
MeO MeO MeO MeO MeO MeO
HO MeO MeO MeO HO
MeO MeO MeO MeO
H
MeO
H
H
H
H
H
H
H
MeO MeO
HO
2′-MeO
>100
>100
>27
>33
H
2′,3′,4′-tri-MeO
MeO
2′,4′,5′-tri-MeO
HO
2′,3′,4′-tri-MeO
2′,3′,4′-tri-MeO
2′,3′,4′-tri-MeO
2′,3′,4′-tri-MeO
H
MeO
HO MeO HO MeO
HO MeO MeO MeO
MeO HO MeO
HO
H
MeO
MeO
HO MeO HO
HO HO MeO
HO Me
HO Me
MeO MeO MeO
HO Me
MeO MeO MeO MeO
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
HO
MeO
MeO
H
Ta ble 2. Structure and Cytotoxicity Data for
3-Hydroxyflavone Derivatives
H
MeO
HO MeO
H
H
H
H
H
H
H
>31
>100
>100
>100
>100
>100
>100
MeO
H
H
H
H
H
HO
MeO Me
H
H
H
MeO MeO
H AcNH
mean
C-7 C-8 C-3′ C-4′ C-5′ GI50 (µM)
compd C-5
C-6
Some structure-activity relationships for cytotoxicity
and associated inhibitory effects on tubulin polymeri-
zation are apparent from these results. The require-
ments for the B ring may be quite stringent. For
example, compounds 1-3 all have 3′-hydroxyl-4′-meth-
oxyl substituents, whereas compounds 5, 7, 15, and 26,
in which these substituents were reversed, were weakly
cytotoxic and showed no detectable effect on tubulin
polymerization or colchicine binding (Tables 1 and 4).
The case of 5 is striking because it differs from 2 only
in the reversal of B-ring substituents. The same was
true for compounds with 3′,4′-dimethoxyl substitution
(4, 22, 23), even though they had a substitution pattern
on the A and C rings identical to those of the more active
agents. The 3-methoxyl functionality also appears to
be essential for interference with tubulin polymeriza-
tion, as compounds with a 3-hydroxyl (27, 30-32, 34,
Table 2) or hydrogen (45-48, 73, Table 3) substituent
had no significant effect on in vitro tubulin polymeri-
zation or colchicine binding. The substitution on most
of the A ring is apparently not critical for activity, as
compounds 1-3 vary from 5,7-dihydroxy substitution
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
HO
HO
H
H
H
MeO MeO
MeO HO
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
HO
H
H
MeO
MeO
MeO
H
H
H
H
4
4.1
4.2
5.5
5.9
6.6
8.9
14
18
32
36
37
60
76
H
H
H
H
H
MeO
MeO
H
MeO
MeO
MeO MeO
2′,4′,6′-tri-MeO
MeO MeO
HO
HO
HO
MeO
HO
MeO MeO MeO
H
H
HO
HO
HO
H
H
H
H
H
H
HO
H
H
H
HO
MeO
HO
H
H
HO
H
MeO MeO
H
H
H
Me
H
HO
MeO
HO
H
HO
HO
HO
H
HO
MeO
HO
MeO
HO
HO
2′,4′-di-HO
>100
of 1. Only one other compound (31) showed any
inhibition of tubulin polymerization, albeit at an IC50
of 31 µM, or only slightly less than the highest test
concentration of 40 µM. The same subset of compounds
was also evaluated for inhibition of the binding of [3H]-
colchicine to tubulin. Only 1-3 had significant activity,
even when the flavone (at 50 µM) was present in 10-
fold molar excess to colchicine.

Requirements for Flavone Cytotoxicity and Binding
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13 2335
Ta ble 4. Effects of Selected Flavones on in Vitro Tubulin
structure-activity observations in the present work, the
prediction of an antimitotic mechanism nonetheless has
proven true for a very limited set of flavone structures.
On the basis of the studies reported herein and on
the ready availability of 1 from a natural source,47
centaureidin has been selected for further evaluation
in the in vivo hollow fiber assay.48
Polymerization and [3H]Colchicine Binding
percent inhibition of
[3H]colchicine binding
tubulin polymerization
compd
IC50 (µM)
@5mM
@50 mM
1
2
3
4
5
7
15
16
19
25
27
30
31
32
34
45
46
47
48
73
2.0 ( 0.5
0.83 ( 0.2
3.0 ( 0.4
>40
35
59
43
76
89
75
4
Exp er im en ta l Section
>40
0
0
6
0
6
1
9
0
>40
Com p ou n d s. Centaureidin (1) (NSC #106969) was isolated
from P. fruticosa as previously described.13 Compound 2 (NSC
#642321) was submitted to the NCI by K. H. Lee.16 Com-
pounds 3, 6, 22, and 23 were purchased from Indofine
Chemicals. Eupatin (27, NSC #122412) was received from the
NCI repository, found to be a mixture of related flavones by
1H NMR, and resolved into three components, 27, 34, and 73,
by C18 HPLC on a Rainin Dynamax column (2.1 × 25 cm),
using a linear gradient of 60-100% MeOH at 16 mL/min. The
sample of eupatoretin (NSC #122414) was similarly resolved
into 27 and 45. All compounds thus obtained matched
literature values for UV, NMR, and MS data.18,27,28 Com-
pounds 13, 15, and 17 were isolated from Gutierrezia micro-
cephala (Compositae) as previously reported.49 Compound 4
was prepared by refluxing 13 with CH3I, K2CO3, and acetone
as described in the same paper. Compounds 20 and 38 were
prepared in like manner from kaempferol and quercetin,
respectively, and were purified by preparative centrifugal
circular TLC. Compound 25 was isolated from extracts of
Fairchild tangerine peels and purified chromatographically to
homogeneity. Sideritoflavone (42) was isolated from Hyptis
pectinata (Lamiaceae) and identified by comparison to litera-
ture data.50 Compounds 8, 9, 12, 18, 21, and 24 were synthe-
sized as detailed below. These novel compounds were isolated
as glassy solids and gave either satisfactory elemental analysis
or purity of g94% in two HPLC systems. All other compounds
were obtained directly from the NCI repository. The struc-
>40
>40
>40
>40
>40
>40
31 ( 5
>40
22
0
>40
2
6
5
0
>40
>40
>40
>40
0
>40
11
colchicine
podophyllotoxin
1.4 ( 0.3
82
to 5-hydroxy-6,7,8-trimethoxy, without major differences
in activity in the tubulin assays. However, the impor-
tance of a 5-hydroxyl was suggested by comparison of
12 with 1, where the only difference between the two
compounds is the lack of a 5-hydroxy group in 12.
Flavones having 3-methoxyl substitution have previ-
ously been reported as cytotoxic. In addition, antiviral
activity has been reported for certain members of this
group.17 Compounds 1 and 3 were identified as weakly
cytotoxic to KB cells by Kupchan.18 Other 3-methoxy-
flavones have also been reported as cytotoxic to KB
cells19-24 (ED50 values of 1-10 µg/mL) or to P388 leu-
kemia23,25,26 (ED50 values of 1-5 µg/mL). In contrast,
relatively few 3-hydroxyflavones have been found to be
cytotoxic. Kupchan identified compounds 25 and 32 as
cytotoxic to KB cells,27,28 while 3,5-dihydroxy-3′,4′,6,7-
tetramethoxyflavone was cytotoxic to five cell lines with
ED50’s of 0.5-2.5 µg/mL.29 Flavonoids unsubstituted at
the 3 position are more common in nature and have
been commonly reported as weak cytotoxins in a variety
of systems.19,21,27,28,30-39 Recently, Zahir et al.40 reported
two cytotoxic flavones which were inhibitory to topo-
isomerase I. Cushman and co-workers have synthesized
a variety of flavones as potential tyrosine kinase inhibi-
tors.41 Of an array of 55 flavone derivatives tested for
cytotoxicity,42 only 15 had an ED50 of <4 µg/mL in a
test panel of five cell lines. Other recent studies of
cytotoxic flavones have focused on polymethoxylated
flavones from citrus peels43 and synthetic 3-aminofla-
vones.44
1
tures of the compounds were all further confirmed by H and
13C NMR, UV, and HRMS.
Syn th esis of 3′-Hyd r oxy-3,4′-m eth oxyfla von oid s. Su m -
m a r y (See Sch em e 1). Flavone 12, lacking a C-5 substituent,
was prepared from 2,4-dihydroxyacetophenone (Acros). Ben-
zylation with benzyl bromide afforded phenol 80. The latter
was oxidized in an Elbs reaction into the parahydroxy ana-
logue 81 and methylated selectively to afford hydroxyac-
etophenone 82. Compound 82 was used to prepare chalcone
83 in alkaline medium with 3-(benzyloxy)-4-methoxybenzal-
dehyde. This chalcone gave rise to the 3-hydroxyflavonoid 84
in an AFO oxidation with hydrogen peroxide. Methylation of
the 3-hydroxy group and debenzylation by hydrogenolysis
afforded the target 12, 3′-hydroxy-3,4′-methoxyflavone.
Flavones with 5-substitution (8, 9, 18, 21, 24) were synthe-
sized from the corresponding phenols (86-88). Phenol 86 was
prepared by aromatization of 5,5-dimethyl-1,3-cyclohexanedi-
one. Phenols 87 and 88 were commercially available (Acros).
Compounds 86-88 were converted in a Houben-Hoesch
reaction into the corresponding methoxy ketones 89-91.
Compound 91 was further oxidized in an Elbs oxidation with
potassium persulfate into the parahydroxy analogue 92.
Compounds 89-92 were treated with 3-(benzyloxy)-4-meth-
oxybenzoic anhydride and the potassium salt of 3-(benzyloxy)-
4-methoxybenzoic acid (Allan-Robinson condensation) to give
rise to the corresponding 3-methoxyflavones 93 and 95-97.
Compound 93 was further methylated into the methoxy
analogue 94. Ethylation of 95 under analogous conditions
yielded the monoethyl derivative 98. Intermediates 93-98
were debenzylated by hydrogenolysis to the desired 3′-hydroxy-
3,4′-dimethoxyflavones (8, 9, 18, 21, 24).
Edwards et al.45 reviewed NCI in vivo and in vitro
screening data for 139 flavones, 15 isoflavones, and 62
flavanones in 1979. On the basis of the data available
at that time, they concluded that flavonoids were not a
promising group of antitumor compounds. It was strik-
ing, however, that they suggested that “It is also
possible that some of the compounds may inhibit mitotic
spindle formation, owing to their possession of contigu-
ous alkoxy-groups deviating from strict coplanarity.”45,46
While their prediction was not consistent with the
4-(Ben zyloxy)-2-h yd r oxya cetop h en on e (80). 2′,4′-Di-
hydroxyacetophenone (10 g, 66 mmol) was dissolved in 50 mL
of dry acetone and refluxed for 3 h with 8 g of powdered
anhydrous K2CO3 and 9.4 mL (13.5 g, 79 mmol) of benzyl
bromide. The mixture was cooled, and the acetone was
removed under reduced pressure; 150 mL of H2O was added,

2336 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
Beutler et al.
Sch em e 1a
a
Reaction conditions: (a) K2S2O8, NaOH, aq pyridine; (b) (CH3O)2SO2, K2CO3, acetone; (c) 3-(benzyloxy)-4-methoxybenzaldehyde, NaOH,
EtOH; (d) H2O2, NaOH, MeOH; (e) H2, Pd/C, MeOH; (f) MeOCH2CN, HCl, ZnCl2, ether; (g) 3-(benzyloxy)-4-methoxybenzoic anhydride,
potassium 3-(benzyloxy)-4-methoxybenzoate; (h) C2H5I, K2CO3, acetone, DMF.
and the products were acidified and extracted with CH2Cl2.
The extract was dried over MgSO4, filtered, and evaporated.
The residue was purified through a silica column (EtOAc-
petroleum ether) to obtain 13.5 g (85%) of the benzyloxy
derivative 80.
4-(Ben zyloxy)-2,5-d ih yd r oxya cetop h en on e (81). Com-
pound 81 was prepared by the oxidation of 80 as described50,51
(39% yield).
4-(Ben zyloxy)-2-h ydr oxy-5-m eth oxyacetoph en on e (82).
Compound 81 (5 g, 18.4 mmol) was dissolved in 50 mL of dry
acetone and refluxed with 6 g of powdered anhydrous K2CO3
and 2.6 g (20.5 mmol) of dimethyl sulfate for 15 h. The acetone
was removed under reduced pressure, and the residue was
taken up in EtOAc (150 mL) and washed with dilute NaOH.
The aqueous layer was acidified and extracted with CH2Cl2.
The extract was dried over MgSO4, filtered, and evaporated.
The residue was purified on silica (EtOAc-petroleum ether)
to yield 3.7 g (70%) of crystalline 82.
2-Hyd r oxy-â,4,6-tr im eth oxya cetop h en on e (91). Com-
pound 91 was prepared by the acylation of 3,5-dimethoxyphe-
nol as described.52 Silica column separation (EtOAc-petro-
leum ether) gave 45% yield of the unsymmetrical product 91.
2,5-Dih yd r oxy-â,4,6-t r im et h oxya cet op h en on e (92).
Compound 92 was prepared by the Elbs oxidation of 91 as
described53,54 (22% yield).
3′-(Ben zyloxy)-7-h yd r oxy-3,4′-d im et h oxy-5,6-d im et h -
ylfla von e (93). Compound 93 was synthesized by the Allan-
Robinson reaction involving 84, 3-(benzyloxy)-4-methoxyben-
zoic anhydride and potassium 3-(benzyloxy)-4-methoxybenzoate
as described17 (85% yield).
3′-(Ben zyloxy)-3,7,4′-t r im et h oxy-5,6-d im et h ylfla von e
(94). Compound 94 was obtained by methylation of 93 using
dimethyl sulfate as described for the preparation of 82 (80%
yield).
3′-(Ben zyloxy)-5,7-dih ydr oxy-3,4′-dim eth oxyflavon e (95).
Allan-Robinson reaction of (methoxyacetyl)phloroglucinol 90
with 3-(benzyloxy)-4-methoxybenzoic anhydride and potassium
3-(benzyloxy)-4-methoxybenzoate as described17 gave 33% yield
of 95.
2′-Hyd r oxych a lcon e (83). The chalcone 83 was prepared
from 82 and 3-(benzyloxy)-4-methoxybenzaldehyde following
published procedures17 (80% yield).
7,3′-B is (b e n zy lo x y )-3-h y d r o x y -4′,6′-d im e t h o x y fla -
von e (84). The 3-hydroxyflavone 84 was prepared by the
Algar-Flynn-Oyamada (AFO) oxidation of the chalcone 83
as described17 (66% yield).
7,3′-Bis(ben zyloxy)-3,6,4′-tr im eth oxyflavon e (85). Meth-
ylation of 84 with dimethyl sulfate in dry acetone following
the procedure described17 gave 85 in 76% yield.
1,3-Dih yd r oxy-4,5-d im eth ylben zen e (86). Compound 86
was prepared as described51 from 5,5-dimethyl-1,3-cyclohex-
anedione.
2,4-Dih yd r oxy-â-m et h oxy-5,6-d im et h yla cet op h en on e
(89). Compound 89 was prepared from 86 and methoxyace-
tonitrile following the Houben-Hoesch procedure as de-
scribed52 (71% yield).
2,4,6-Tr ih yd r oxy-â-m eth oxya cetop h en on e (90). The
Houben-Hoesch acylation of phloroglucinol with methoxyace-
tonitrile as described52 gave 90 in 79% yield.
3′-(Ben zyloxy)-3,5,7,4′-tetr a m eth oxyfla von e (96). Com-
pound 96 was prepared by the Allan-Robinson method from
91, 3-(benzyloxy)-4-methoxybenzoic anhydride, and potassium
3-(benzyloxy)-4-methoxybenzoate as described17 (68% yield).
3′-(Be n zyloxy)-6-h yd r oxy-3,5,7,4′-t e t r a m e t h oxyfla -
von e (97). Compound 97 was prepared by the Allan-
Robinson procedure as described17 (52% yield).
3′-(Ben zyloxy)-7-eth oxy-5-h yd r oxy-3,4′-d im eth oxyfla -
von e (98). Compound 95 (0.5 g, 1.2 mmol) was dissolved in
50 mL of dry acetone and 10 mL of dry N,N-dimethylforma-
mide and stirred with 5 g of powdered anhydrous K2CO3.
Ethyl iodide (0.25 g, 1.6 mmol) was added, and the mixture
was refluxed for 2 h. The acetone was removed under reduced
pressure and the residue dissolved in EtOAc. The solution
was washed with H2O, dried over MgSO4, and evaporated. The
product was purified through a silica column (EtOAc-toluene)
to obtain 0.33 g (60%) of crystalline 98.

Requirements for Flavone Cytotoxicity and Binding
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13 2337
(3) Cushman, M.; Nagarathnam, D.; Gopal, D.; Chakraborti, A. K.;
Lin, C. M.; Hamel, E. Synthesis and Evaluation of Stilbene and
Dihydrostilbene Derivatives as Potential Agents that Inhibit
Tubulin Polymerization. J . Med. Chem. 1991, 34, 2579-2588.
(4) Edwards, M. L.; Stemerick, D. M.; Sunkara, P. S. Chalcones: A
New Class of Antimitotic Agents. J . Med. Chem. 1990, 33, 1948-
1954.
(5) Brewer, C. F.; Loike, J . D.; Horwitz, S. B.; Sternlicht, H.; Gensler,
W. J . Conformational Analysis of Podophyllotoxin and its
Congeners. Structure-Activity Relationship in Microtubule
Assembly. J . Med. Chem. 1979, 22, 215-221.
(6) Safa, A. R.; Hamel, E.; Felsted, R. L. Photoaffinity Labeling of
Tubulin Subunits with a Photoactive Analogue of Vinblastine.
Biochemistry 1987, 26, 97-102.
(7) Bai, R.; Pettit, G. R.; Hamel, E. Binding of Dolastatin 10 to
Tubulin at a Distinct Site for Peptide Antimitotic Agents Near
the Exchangeable Nucleotide and Vinca Alkaloid Sites. J . Biol.
Chem. 1990, 265, 17141-17149.
(8) Bai, R.; Pettit, G. R.; Hamel, E. Dolastatin 10, a Powerful
Cytostatic Peptide Derived from a Marine Animal. Inhibition of
Tubulin Polymerization Mediated through the Vinca Alkaloid
Binding Domain. Biochem. Pharmacol. 1990, 39, 1941-1949.
(9) Bai, R.; Cichacz, Z. A.; Herald, C. L.; Pettit, G. R.; Hamel, E.
Spongistatin 1, a Highly Cytotoxic, Sponge-Derived, Marine
Natural Product that Inhibits Mitosis, Microtubule Assembly,
and the Binding of Vinblastine to Tubulin. Mol. Pharmacol.
1993, 44, 757-766.
(10) Rao, S.; Krauss, N. E.; Heerding, J . M.; Swindell, C. S.; Ringel,
I.; Orr, G. A.; Horwitz, S. B. 3′-(p-Azidobenzamido)taxol Photo-
labels the N-terminal 31 Amino Acids of â-Tubulin. J . Biol.
Chem. 1994, 269, 3132-3134.
(11) Paull, K. D.; Lin, C. M.; Malspeis, L.; Hamel, E. Identification
of Novel Antimitotic Agents Acting at the Tubulin Level by
Computer-assisted Evaluation of Differential Cytotoxicity Data.
Cancer Res. 1992, 52, 3892-3900.
(12) Boyd, M. R.; Paull, K. D. Some Practical Considerations and
Applications of the NCI in vitro Drug Discovery Screen. Drug
Dev. Res. 1995, 34, 91-109.
(13) Beutler, J . A.; Cardellina, J . H., II; Lin, C. M.; Hamel, E.; Cragg,
G. M.; Boyd, M. R. Centaureidin, A Cytotoxic Flavone from
Polymnia fruticosa, Inhibits Tubulin Polymerization. BioMed.
Chem. Lett. 1993, 3, 581-584.
(14) Beutler, J . A.; Cardellina, J . H., II; Gray, G. N.; Prather, T. R.;
Shoemaker, R. H.; Boyd, M. R.; Lin, C. M.; Hamel, E.; Cragg,
G. M. Two New Cytotoxic Chalcones from Calythropsis aurea.
J . Nat. Prod. 1993, 56, 1718-1722.
(15) Lichius, J . J .; Thoison, O.; Montagnac, A.; Pais, M.; Gue´ritte-
Voegelein, F.; Se´venet, T.; Cosson, J . P.; Hadi, A. H. A. Antimi-
totic and Cytotoxic Flavonols from Zieridium pseudobtusifolium
and Achronycha porteri. J . Nat. Prod. 1994, 57, 1012-1016.
(16) Shi, Q.; Chen, K.; Li, L.; Chang, J . J .; Autry, C.; Kozuka, M.;
Konoshima, T.; Estes, J . R.; Lin, C. M.; Hamel, E.; McPhail, A.
T.; McPhail, D. R.; Lee, K. H. Antitumor Agents, 154. Cytotoxic
and Antimitotic Flavonols from Polanisia dodecandra. J . Nat.
Prod. 1995, 58, 475-482.
(17) De Meyer, N.; Haemers, A.; Mishra, L.; Pandey, H. K.; Pieters,
L. A. C.; Vanden Berghe, D. A.; Vlietinck, A. J . 4′-Hydroxy-3-
methoxyflavones with Potent Antipicornavirus Activity. J . Med.
Chem. 1991, 34, 736-746.
7,3′-Dih yd r oxy-3,6,4′-tr im eth oxyfla von e (12). The bis-
(benzyloxy)flavone 85 (0.5 g, 1 mmol), dissolved in 50 mL of
methanol, was stirred with 0.5 g of Pd/C catalyst (10%).
A
steady stream of H2 gas was passed over the suspension for
90 min. The catalyst was removed by filtration and the filtrate
evaporated. The residue was crystallized from methanol (yield
0.28 g, 85%). 1H NMR: δ 3.80 (3H, s, OCH3), 3.93 (6H, s, 2 ×
OCH3), 5.70 (1H, s, OH), 6.40 (1H, br s, OH), 6.90 (2H, s),
7.45-7.75 (3H, m). HREIMS: m/z 344.0899; calcd for C18H16O7,
344.0896.
7,3′-Dih yd r oxy-3,4′-d im eth oxy-5,6-d im eth ylfla von e (9).
Flavone 9 was prepared from 93 by hydrogenolysis, as
described for the preparation of 12 (91% yield). 1H NMR: δ
2.20 (3H, s), 2.82 (3H, s), 3.78 (3H, s, OCH3), 3.93 (3H, s,
OCH3), 6.80 (1H, s), 7.06 (1H, d, J ) 10 Hz), 7.60 (1H, d, J )
10 Hz), 7.95 (1H, s). HREIMS: m/z 342.1106; calcd for
C
19H18O6, 342.1103.
3′-Hyd r oxy-3,7,4′-tr im eth oxy-5,6-d im eth ylfla von e (18).
Compound 94 was debenzylated by hydrogenolysis following
the procedure described for the preparation of 12 (92% yield).
1H NMR: δ 2.17 (3H, s), 2.78 (3H, s), 3.25 (3H, s, OCH3), 3.90
(3H, s, OCH3), 3.95 (3H, s, OCH3), 6.80 (1H, s), 7.00 (1H, d, J
) 10 Hz), 7.55, 7.65 (2H, m). HREIMS: m/z 356.1263; calcd
for C20H20O6, 356.1259.
3′-Hyd r oxy-3,5,7,4′-tetr a m eth oxyfla von e (24). Deben-
zylation of 96 by hydrogenolysis as described for the prepara-
tion of 12 gave 24 (yield 80%). 1H NMR: δ 3.77 (3H, s, OCH3),
3.88 (3H, s, OCH3), 3.90 (3H, s, OCH3), 3.93 (3H, s, OCH3),
6.35 (1H, s), 6.55 (1H, s), 6.95 (1H, d, J ) 9 Hz), 7.50-7.65
(2H, m). HREIMS: m/z 358.1052; calcd for C19H18O7, 358.1052.
6,3′-Dih yd r oxy-3,5,7,4′-tetr a m eth oxyfla von e (21). De-
benzylation of 97 by hydrogenolysis in the presence of Pd/C
(10%) catalyst, as described for the preparation of 12, yielded
89% of 21. 1H NMR: δ 3.75 (3H, s, OCH3), 3.89 (3H, s, OCH3),
3.92 (3H, s, OCH3), 3.98 (3H, s, OCH3), 6.90 (1H, s, arom),
7.45-7.65 (3H, m). HREIMS: m/z 374.1007; calcd for C19H18O8,
374.1002.
7-Eth oxy-5,3′-dih ydr oxy-3.4′-dim eth oxyflavon e (8). Fla-
vone 8 was prepared by the debenzylation of 98 by hydro-
genolysis in the presence of Pd/C (10%) catalyst, as described
for the preparation of 12 (82% yield). 1H NMR: δ 1.35 (3H, t,
J ) 8.5 Hz), 3.72 (3H, s, OCH3), 3.87 (3H, s, OCH3), 4.01 (2H,
q, J ) 8.5 Hz), 6.21 (1H, s), 6.36 (1H, s), 6.88 (1H, d, J ) 8.5
Hz), 7.45-7.60 (2H, m). HREIMS: m/z 358.1041; calcd for
C
19H18O7, 358.1052.
Biologica l Assa ys. In vitro tubulin polymerization and
colchicine binding assays were performed as previously re-
ported.2,55 Tubulin polymerization results shown in Table 4
are averages ( SD from three determinations, using a tubulin
concentration of 10 µM; the colchicine inhibition assay results
were derived from three determinations. The 60-cell cytotox-
icity screen has been described previously.12 Due to limited
compound availability, results shown in Tables 1-3 were
derived from single tests of each compound in the full 60-cell
panel.
(18) Kupchan, S. M.; Bauerschmidt, E. Cytotoxic Flavonols from
Baccharis sarothroides. Phytochemistry 1971, 10, 664-666.
(19) LeQuesne, P. W.; Menachery, M. D.; Raffauf, R. F. Antitumor
Plants. Part IX. Structural Reassignments for Three Flavonoids
from Lychnophora affinis Gardn. J . Nat. Prod. 1979, 42, 320-
321.
Ack n ow led gm en t. We thank M. E. Wall (Research
Triangle Institute), K. H. Lee (University of North
Carolina), E. R. Silveira (Universidade Federale do
Ceara´, Brazil), and R. A. Smith (ICN Pharmaceuticals)
for permission to quote NCI testing data for compounds
which they had supplied to the NCI; A. Monks and D.
Scudiero for the 60-cell testing; G. N. Gray for mass
spectroscopy; and J . J ohnson for assistance in arranging
antitumor screening of NCI repository compounds.
(20) Arisawa, M.; Hayashi, T.; Shimizu, M.; Morita, N.; Bai, H.; Kuze,
S.; Ito, Y. Isolation and Cytotoxicity of Two New Flavonoids from
Chrysosplenum grayanum and Related Flavonols. J . Nat. Prod.
1991, 54, 898-901.
(21) Kingston, D. G. I.; Rao, M. M.; Zucker, W. V. Plant Anticancer
Agents. IX. Constituents of Hyptis tomentosa. J . Nat. Prod. 1979,
42, 496-499.
(22) Raffauf, R. F.; Menachery, M. D.; LeQuesne, P. W.; Arnold, E.
V.; Clardy, J . Antitumor Plants. 11. Diterpenoid and Flavonoid
Constituents of Bromelia pinguin L. J . Org. Chem. 1981, 46,
1094-1098.
(23) Hou, R. S.; Duh, C. Y.; Wang, S. K.; Chang, T. T. Cytotoxic
Flavonoids from the Leaves of Melicope triphylla. Phytochemistry
1994, 35, 271-272.
Refer en ces
(24) Bittner, M.; Silva, M.; Vargas, J .; Bohlmann, F. Biologically
Active Flavones from Gutierrezia resinosa. Phytochemistry 1983,
22, 1523-1524.
(25) Dong, X. P.; Che, C. T.; Farnsworth, N. R. Cytotoxic Flavonols
from Gutierrezia microcephala. J . Nat. Prod. 1987, 50, 337.
(26) Saleh, A. A.; Cordell, G. A.; Farnsworth, N. R. Isolation of 3,6-
Dimethoxy-4′,5,7-trihydroxyflavone from Acanthospermum gla-
bratum. Lloydia 1976, 39, 456-458.
(1) Rowinsky, E. K.; Donehower, R. C. The Clinical Pharmacology
and Use of Antimicrotubule Agents in Cancer Chemotherapeu-
tics. Pharmacol. Ther. 1991, 52, 35-84.
(2) Muzzafar, A.; Brossi, A.; Lin, C. M.; Hamel, E. Antitubulin
Effects of Derivatives of 3-Demethylthiocolchicine, Methylthio
Ethers of Natural Colchicinoids, and Thioketones Derived from
Thiocolchicine. Comparison with Colchicinoids. J . Med. Chem.
1990, 33, 567-571.

2338 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
Beutler et al.
(27) Kupchan, S. M.; Sigel, C. W.; Hemingway, R. J .; Knox, J . R.;
Udayamurthy, M. S. Tumor Inhibitors XXXIII. Cytotoxic Fla-
vones from Eupatorium species. Tetrahedron 1969, 25, 1603-
1615.
(28) Kupchan, S. M.; Sigel, C. W.; Knox, J . R.; Udayamurthy, M. S.
Tumor Inhibitors XXXVI. Eupatin and Eupatoretin, Two Cyto-
toxic Flavonols from Eupatorium semiserratum. J . Org. Chem.
1969, 34, 1460-1463.
(29) Zheng, G. Q. Cytotoxic Terpenoids and Flavonoids from Arte-
misia annua. Planta Med. 1994, 60, 54-57.
(30) Hayashi, T.; Uchida, K.; Hayashi, K.; Niwayama, S.; Morita, N.
A Cytotoxic Flavone from Scoparia dulcis L. Chem. Pharm. Bull.
1988, 36, 4849-4851.
(31) Woerdenbag, H. J .; Merfort, I.; Passreiter, C. M.; Schmidt, T.
J .; Willuhn, G.; van Uden, W.; Pras, N.; Kampinga, H. H.;
Konings, A. W. T. Cytotoxicity of Flavonoids and Sesquiterpene
Lactones from Arnica Species Against the GLC4 and the COLO
320 Cell lines. Planta Med. 1994, 60, 434-437.
(32) Parmar, V. S.; Bisht, K. S.; Sharma, S. K.; J ain, R.; Taneja, P.;
Singh, S.; Simonsen, O.; Boll, P. M. Highly Oxygenated Bioactive
Flavones from Tamarix. Phytochemistry 1994, 36, 507-511.
(33) Afifi, M. S. A.; Ahmed, M. M.; Pezzuto, J . M.; Kinghorn, A. D.
Cytotoxic Flavolignans and Flavones from Verbascum sinaiticum
Leaves. Phytochemistry 1993, 34, 839-841.
(34) Mori, A.; Nishino, C.; Enoki, N.; Tawata, S. Cytotoxicity of Plant
Flavonoids Against HeLa Cells. Phytochemistry 1988, 27, 1017-
1020.
(35) Malterud, K. E.; Hanche-Olsen, I. M.; Smith-Kielland, I. Fla-
vonoids from Orthosiphon spicatus. Planta Med. 1989, 55, 569-
570.
(36) Ryu, S. H.; Ahn, B. Z.; Pack, M. Y. The Cytotoxic Principle of
Scutellariae Radix Against L1210 Cell. Planta Med. 1985, 355-
356.
(41) Cushman, M.; Nagarathnam, D.; Geahlen, R. L. Synthesis and
Evaluation of Hydroxylated Flavones and Related Compounds
as Potential Inhibitors of the Protein-Tyrosine Kinase p56lck. J .
Nat. Prod. 1991, 54, 1345-1352.
(42) Cushman, M.; Nagarathnam, D. Cytotoxicities of Some Fla-
vonoid Analogues. J . Nat. Prod. 1991, 54, 1656-1660.
(43) Chen, J .; Montanari, A. M.; Widmer, W. W. Two New Poly-
methoxylated Flavones, a Class of Compounds with Potential
Anticancer Activity, Isolated from Cold Pressed Dancy Tangerine
Peel Oil Solids. J . Agric. Food Chem. 1997, 45, 364-368.
(44) Dauzonne, D.; Folle´as, B.; Martinez, L.; Chabot, G. G. Synthesis
and in vitro Cytotoxicity of a Series of 3-Amino flavones. Eur.
J . Med. Chem. 1997, 32, 71-82.
(45) Edwards, J . M.; Raffauf, R. F.; LeQuesne, P. W. Antineoplastic
Activity and Cytotoxicity of Flavones, Isoflavones, and Fla-
vanones. J . Nat. Prod. 1979, 42, 85-91.
(46) Kupchan, S. M.; Stevens, K. L.; Rohlfing, E. A.; Sickles, B. R.;
Sneden, A. T.; Miller, R. W.; Bryan, R. F. New Cytotoxic Lignans
from Aniba megaphylla Mez. J . Org. Chem. 1978, 43, 586-590.
(47) Christensen, L. P.; Lam, J . Flavones and Other Constituents
from Centaurea species. Phytochemistry 1991, 30, 2663-2665.
(48) Hollingshead, M. G.; Alley, M. C.; Camalier, R. F.; Abbott, B.
J .; Mayo, J . G.; Malspeis, L.; Grever, M. R. In Vivo Cultivation
of Tumor Cells in Hollow Fibers. Life Sci. 1995, 57, 131-141.
(49) Roitman, J . N.; J ames, L. F. Chemistry of Toxic Range Plants.
Highly Oxygenated Flavonol Methyl Ethers from Gutierrezia
microcephala. Phytochemistry 1985, 24, 835-848.
(50) Novelo, M.; Cruz, J . G.; Herna´ndez, L.; Pereda-Miranda, R.;
Chai, H.; Mar, W.; Pezzuto, J . M. Cytotoxic Constituents from
Hyptis verticillata. J . Nat. Prod. 1993, 56, 1728-1736.
(51) Kablouki, M. S. The Aromatization of Cyclic Ketones. II. Novel
Synthesis of Substituted Dihydroxybenzenes. J . Org. Chem.
1974, 39, 3696-3698.
(52) Muller, E., Ed. Methoden de Organischen Chemie, Houben-Weyl;
Georg Thieme Verlag: Stuttgart, 1973; Ketonen Teil 1, pp 389-
421.
(53) Sethna, S. M. The Elbs Persulfate Oxidation. Chem. Rev. 1951,
49, 91-101.
(54) Muller, E., Ed. Methoden der Organischen Chemie, Houben-Weyl;
Georg Thieme Verlag: Stuttgart, 1976; Phenole Teil 1, pp 43-
53.
(37) Ulubelen, A.; O¨ ksuz, S. Cytotoxic Flavones from Centaurea
urvillei. J . Nat. Prod. 1982, 45, 373.
(38) Kaneda, N.; Pezzuto, J . M.; Soejarto, D. D.; Kinghorn, A. D.;
Farnsworth, N. R.; Santisuk, T.; Tuchinda, P.; Udchachon, J .;
Reutrakul, V. Plant Anticancer Agents, XLVIII. New Cytotoxic
Flavonoids from Muntingia calabura Roots. J . Nat. Prod. 1991,
54, 196-206.
(39) Wall, M. E.; Wani, M. C.; Fullas, F.; Oswald, J . B.; Brown, D.
M.; Santisuk, T.; Reutrakul, V.; McPhail, A. T.; Farnsworth, N.
R.; Pezzuto, J . M.; Kinghorn, A. D.; Besterman, J . M. Plant
Antitumor Agents. 31. The Calycopterones, a New Class of
(55) Borisy, G. G. A Rapid Method for Quantitative Determination
of Microtubule Protein Using DEAE-Cellulose Filters. Anal.
Biochem. 1972, 50, 373-385.
Biflavonoids with Novel Cytotoxicity in
a Diverse Panel of
Human Tumor Cell Lines. J . Med. Chem. 1994, 37, 1465-1470.
(40) Zahir, A.; J ossang, A.; Bodo, B.; Provost, J .; Cosson, J . P.;
Se´venet, T. DNA Topoisomerase I Inhibitors: Cytotoxic Flavones
from Lethedon tannaensis. J . Nat. Prod. 1996, 59, 701-703.
J M970842H