Dioxol and dihydrodioxin analogs of 2- and 3-phenylacetonitriles as potent anti-cancer agents with nanomolar activity against a variety of human cancer cells

Article abstract of DOI:10.1016/j.bmcl.2016.03.068

A small library of (Z)-2-(benzo[d][1,3]dioxol-5-yl) and (Z)-2,3-dihydrobenzo[b][1,4]dioxin-6-yl analogs of 2- and 3-phenylacetonitriles has been synthesized and evaluated for their anti-cancer activities against a panel of 60 human cancer cell lines. The

Full text of DOI:10.1016/j.bmcl.2016.03.068

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Dioxol and dihydrodioxin analogs of 2- and 3-phenylacetonitriles as
potent anti-cancer agents with nanomolar activity against a variety
of human cancer cells
Nikhil R. Madadi a, Amit Ketkar b, Narsimha R. Penthala a, April C. L. Bostian b, Robert L. Eoff b,
Peter A. Crooks a,
⇑
a Department of Pharmaceutical Sciences, College of Pharmacy, University of Arkansas for Medical Sciences, Little Rock, AR 72205, USA
b Department of Biochemistry and Molecular Biology, College of Medicine, University of Arkansas for Medical Sciences, Little Rock, AR 72205, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A small library of (Z)-2-(benzo[d][1,3]dioxol-5-yl) and (Z)-2,3-dihydrobenzo[b][1,4]dioxin-6-yl analogs of
2- and 3-phenylacetonitriles has been synthesized and evaluated for their anti-cancer activities against a
panel of 60 human cancer cell lines. The dihydrodioxin analog 3j and dioxol analogs 5e and 7e exhibited
the most potent anti-cancer activity of all the analogs synthesized in this study, with GI50 values of
<100 nM against almost all of the cell lines in the human cancer cell panel. Of these three, only compound
3j inhibited tubulin polymerization to any degree in vitro. The binding modes of 3j and the structurally
related tubulin-inhibitor DMU-212 were determined by virtual docking studies with tubulin dimer.
Compound 3j docked at the colchicine-binding site at the dimer interface of tubulin. The Full-Fitness
(FF) score of 3j was observed to be substantially higher than DMU-212, which agrees well with the
observed anti-cancer potency (GI50 values). The mechanism by which dioxol analogs 5e and 7e exert their
cytotoxic effects remains unknown at this stage, but it is unlikely that they affect tubulin dynamics.
Nevertheless, these findings suggest that both dioxol and dihydrodioxin analogs of phenylacrylonitrile
may have potential for development as clinical candidates to treat a variety of human cancers.
Ó 2016 Elsevier Ltd. All rights reserved.
Received 23 February 2016
Revised 15 March 2016
Accepted 16 March 2016
Available online xxxx
Keywords:
3,4-Methylenedioxycyanostilbenes
3,4-Ethylenedioxycyanostilbenes
Synthesis of cyanostilbenes
Anti-cancer activity
Molecular docking studies
In 1987, the US National Cancer Institute (NCI) identified the
cis-stilbene analog, combretastatin A-4 (CA-4; Fig. 1, structure A),
as the most potent anti-cancer agent of all the natural combretas-
tatins isolated from the African bush willow, Combretum caffrum.1
Early work showed that CA-4 targets tubulin and inhibits the pro-
liferation of both murine and human cancer cells.2 Unfortunately,
subsequent reports have indicated that CA-4 possesses unfavorable
properties, such as isomerization to the less active trans-stilbene
isomer (Fig. 1, structure B) in solution, low water-solubility and
vascular disruption,3–5 which has precluded its development as a
potential clinical candidate. Efforts to improve the drug-likeness
properties of CA-4 have resulted in the discovery of a water-sol-
uble prodrug, CA-4P, which is currently being evaluated in phase
II/III clinical trials for activity in anaplastic thyroid carcinoma,
non-small cell lung cancer, and ovarian cancer in combination with
conventional standard of care cytotoxic drugs that include pacli-
taxel, carboplatin, and the antiangiogenic agent, brevacizumab.6
Recently our laboratory has reported on some novel trans-CA-4
(Fig. 1, structure B) analogs as potent inhibitors of tubulin poly-
merization with growth inhibitory activities superior to cis-CA-
4.4,5 These molecules can also be considered as structural analogs
of resveratrol (Fig. 1, structure C), and include the anti-cancer
agent DMU-212 (Fig. 1, structure D), which binds to the colchicine
binding site on tubulin. Our work has demonstrated that a trans
double bond bearing a nitrile moiety can improve chemical stabil-
ity and can serve as an effective replacement for the cis-olefinic
moiety in CA-4 and its analogs.5 More recently, we have reported
on a series of diphenyl, 2-benzothio phenyl/phenyl, and 2-quino-
linyl/phenyl acrylonitrile derivatives (Fig. 1, structures E, F, and G
respectively) as potent anti-proliferative agents that have potential
as anti-tubulin therapeutics for treatment of both solid and hema-
tological tumors.7
In our current studies we have now synthesized a small library
of thirty one substituted phenylacrylonitrile analogs that incorpo-
rate benzo[d][1,3]dioxol-5-yl and 2,3-dihydrobenzo[b][1,4]dioxin-
6-yl moieties and have evaluated them against a panel of 60
human tumor cell lines. The most potent analogs identified have
been evaluated as inhibitors of tubulin polymerization.
⇑
Corresponding author. Tel.: +1 501 686 6495; fax: +1 501 686 6057.
E-mail address: pacrooks@uams.edu (P.A. Crooks).
http://dx.doi.org/10.1016/j.bmcl.2016.03.068
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Madadi, N. R.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.03.068

2
N. R. Madadi et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Figure 1. Structures of combretastatin A-4 (CA-4), trans-CA-4, DMU-212, (Z)-diphenyl, (Z)-2-benzothiophene/phenyl, and (Z)-2-quinolinyl/phenyl acrylonitrile derivatives.
An initial series of eleven (Z)-3-(benzo[d][1,3]dioxol-5-yl)-2-
phenylacrylonitrile (3a–3e) and (Z)-3-(2,3-dihydrobenzo[b][1,4]-
dioxin-6-yl-2-phenylacrylonitrile (3f–3k) analogs were synthe-
sized by reacting benzo-[d][1,3]dioxole-5-carbaldehyde (2a) or
2,3-dihydrobenzo[b][1,4]dioxine-6-carbaldehyde (2b) with an
appropriately substituted phenylacetonitrile (1a–1h) at reflux
temperature in 5% sodium methoxide/methanol to yield the
desired compounds in yields ranging from 70% to 95%
(Scheme 1).7,8
yield the desired compound in yields ranging from 80% to 90%
(Scheme 2).
A third series of substituted (Z)-2-(benzo[d][1,3]dioxol-5-yl)-3-
phenyl acrylonitrile analogs (7a–7h) were synthesized by reacting
appropriate substituted aromatic and hetero aromatic aldehydes
(6a–6h) with 2-(benzo[d][1,3]dioxol-5-yl)acetonitrile (1i) at reflux
temperature in 5% sodium methoxide/methanol. The reaction was
carried out as described previously7 for analogs 3a–3k8 to yield the
desired compound in yields ranging from 80% to 90% (Scheme 3).
The NCI employs an effective triage system for the submitted
compounds based on duplicates already screened and ADME algo-
rithm results, prior to selecting them for initial single dose and
subsequent five dose screening assays.9 The in vitro screening of
the above compounds was carried out utilizing the procedure
described by Rubinstein et al.9–11 Of the thirty one phenylacetoni-
A
second series of twelve substituted (Z)-2-(benzo[d][1,3]
dioxol-5-yl)-3-phenylacrylonitrile analogs (5a–5l) were synthe-
sized by reacting appropriately substituted aromatic aldehydes
(4a–4l) with 2-(benzo[d][1,3]dioxol-5-yl)acetonitrile (1i) at reflux
temperature in 5% sodium methoxide/methanol. The reaction
was carried out as described previously7 for analogs 3a–3k8 to
R1
R1
CN
OHC
O
5% NaOCH3
Methanol / reflux
R2
R2
O
+
(CH2)n
(CH2)n
CN
O
R4
R3
O
R4
R3
1a-1h
2a-2b
3a-3k
1
: R = R3 = OCH3, R2 = R4 = H
2a: n=1
2b
3a: R1 = R3 = OCH3, R2 = R4 = H, n=1
3b: R1 = R2 = R3= OCH3, R4 = H, n=1
3c: R4 = OCH3, R1 = R2 = R3 = H, n=1
1a
1b
: R1 = R2 = R3 = OCH3, R4 = H
: n=2
1c: R1 = R2 = R3 = H, R4 = OCH3
1d: R1 = R3 = R4 = H, R2 = NO2
1e: R1 = R3 = R4 = H, R2 = OCH3
1f: R1 = R2 = R3 = R4 = H
: R2 = NO2,R1 = R3 = R4 = H, n=1
: R2 = OCH3,R1 = R3 = R4 = H, n=1
3d
3e
3f
3g
: R1 = R2 = R3 = R4 = H, n=2
1
: R1 = R2 = R3 = H, R4 = CF3
: R1 = R2 = OCH3, R3 = R4 = H
1g
1h
: R = R3 = H, R4 = CF3, n=2
3h: R1 = R2 = OCH3, R3 = R4 = H, n=2
3i: R1 = R3 = R4 = H, R2 = OCH3, n=2
3j: R1 = R2 = R3 = OCH3, R4 = H, n=2
3k: R2 = NO2, R1 = R3 = R4= H, n=2
Scheme 1. Synthesis of (Z)-2-(benzo[d][1,3]dioxol-5-yl)-2-phenylacrylonitrile (3a–3e) and (Z)-2,3-dihydrobenzo[b][1,4]dioxin-6-yl-2-phenylacrylonitrile (3f–3k) analogs.
Please cite this article in press as: Madadi, N. R.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.03.068

N. R. Madadi et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
3
O
R1
CN
R1
R2
O
O
R2
R3
5% NaOCH3
O
OHC
R3
+
Methanol / reflux
CN
R5
5a-5l
R5
R4
4a-4l
4
R4
1i
2
3
1
5
5a: R2= R3= R4= OCH3, R1= R5= H
4a:
4b:
4c:
R = R = R = OCH3, R = R = H
R
R
2 = R4= OCH3, R1 = R3 = R5 = H
2 = R4= Br, R1= R3 = R5 = H
5b: R2 = R4= OCH3, R1 = R3 = R5 = H
: R2 = R4= Br, R1= R3 = R5 = H
5c
5d
5e
5f
4d: R2 = R4 = F, R1 = R3= R5 = H
4e: R3= OCH3, R1= R2 = R4 = R5 = H
4f: R2= R3 = OCH3, R1= R4 = R5= H
: R2 = R4 = F, R1 = R3= R5 = H
: R = OCH3, R = R = R4 = R5 = H
3
1
2
: R = R3 = OCH3, R1= R4 = R5= H
2
4g: R2 = R3= Cl, R1= R4= R5 = H
5g: R2 = R3= Cl, R1= R4= R5 = H
5h: R1= R5 = Cl, R2= R3 = R4= H
5i: R2 = Br, R3= F, R1= R4 = R5= H
5j: R3 = R5 = OCH3, R1 = R2 = R4 = H
: R = R5 = Cl, R2= R3 = R4= H
: R2 = Br, R3= F, R1= R4 = R5= H
: R3 = R5 = OCH3, R1 = R2 = R4 = H
1
4h
4i
4j
4k: R3 = R5= Cl, R1 = R2 = R4 = H
: R3 = R5= Cl, R1 = R2 = R4 = H
5k
3
4
: R = R = OCH3, R5 = Cl, R1= R2= H
4l: R3= R4= OCH3, R5 = Cl, R1= R2= H
5l
Scheme 2. Synthesis of substituted (Z)-2-(benzo[d][1,3]dioxol-5-yl)-3-phenylacrylonitrile analogs (5a–5l).
Comp
ound
Ar
6a/7a
6b/7b
6c/7c
6d/7d
6e/7e
6f/7f
6g/7g
6h/7h
Scheme 3. Synthesis of (Z)-2-(benzo[d][1,3]dioxol-5-yl-3-heteroarylacrylonitrile analogs (7a–7h).
trile analogs synthesized, fourteen analogs (3e, 3h–3j, 5a–5h, 5k,
5l and 7a–7h) were selected and evaluated for anti-cancer activity
against a panel of 60 human tumor cell lines. Compounds were ini-
HCT-116 (colon), SNB-75 (CNS), M14 (melanoma), SK-MEL-5 (mel-
anoma), UACC-62 (melanoma), NCI/ADR-RES (ovarian), A498
(renal) and MDA-MB-468 (breast) cancer cell lines. When the
3,4,5-methoxyphenyl group of compound 3j was replaced with a
3,4-dimethoxyphenyl moiety (3h), the average GI50 values
tially screened at 10À5 M to determine growth inhibition (GI50
)
(single dose results for compounds 3e, 3h–3j, 5a–5l, 7a–7h are
provided in the supporting information). The 10 M single dose
l
declined from 97 nM to 3.7 lM. Also, when the 3,4,5-trimethoxy-
screening results for these compounds are presented in the supple-
mentary data section. From the 14 compounds selected for single
dose screening, eight compounds (3h–3j, 5e, 5j, and 7c–7e)
showed promising anti-cancer activity and were selected for sub-
sequent five dose studies.
Tables 1 and 2 provide the 50% Growth Inhibitory (GI50) data
from the five-dose study of the above eight compounds and for
the positive control, DMU-212, a structurally related stilbene ana-
log, against the panel of 60 human tumor cell lines. DMU-212 is an
anti-tubulin agent which has been shown to possess potent anti-
proliferative/proapoptotic activities in a variety of cancer cells,
including K562 (leukemia), HT29 (colo-rectal), and HePG2 (hep-
atoma) HeLa (cervical), LnCaP (prostate), HepG2 (hepatoma) and
MCF-7 (breast) cancer cells.12–14
From the five dose studies of the (Z)-2,3-dihydro-benzo[b][1,4]-
dioxin-6-yl-2-phenylacrylonitrile analogs 3h–3j (Scheme 1), com-
pound 3j was found to be a very effective anti-cancer agent with
an average GI50 value of 97 nM against all 60 human cancer cell
lines in the panel. In particular, 3j exhibited a GI50 value of
20 nM against cancer cell lines SF-295 (CNS), SF-539 (CNS), and
MDA-MB-435 (melanoma) and a GI50 value of 30 nM against
phenyl group of compound 3j was replaced with a 4-methoxyphe-
nyl moiety (3i) the average GI50 values deteriorated even further
(ꢀ50% of the GI50 values were >100
lM). The growth inhibition
activities of analogs 3h, 3i and 3j suggest that the presence of a
3,4,5-trimethoxyphenyl group affords significant anti-cancer activ-
ity against most of the human cancer cell lines in the panel.
From the five dose studies carried out on the (Z)-2-(benzo[d]
[1,3]dioxol-5-yl-3-arylacrylonitrile analogs 5e, 5j, and 7c–7e, com-
pounds 5e and 7e were the most potent anti-cancer agents in this
series. Compound 7e was particularly effective against three speci-
fic cancer cell lines: NCI-H522 (non-small cell lung), SNB-75 (CNS)
and MDA-MB-435 (melanoma) with GI50 values of 20 nM and also
exhibited GI50 values of 30 nM against HL-60(TB) (leukemia), SR
(leukemia), COLO 205 (colon), HT29 (colon) and A498 (renal) can-
cer cell lines. Substitution of a 1-naphthyl or 4-methoxy-1-naph-
thyl moiety for the 2-benzthiophenyl moiety in 7e resulted in
significant loss of anti-cancer activity (i.e. compounds 7c and 7d,
respectively, Table 2). Compound 5e exhibited significant anti-can-
cer activity against HOP-62 (non-small cell lung) and MDA-MB-
435 (melanoma) cancer cell lines with GI50 values of 30 nM, and
afforded a GI50 value of 40 nM against K-562 (leukemia) and NCI-
Please cite this article in press as: Madadi, N. R.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.03.068

4
N. R. Madadi et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Table 1
Growth inhibition (GI50
Table 2
/
l
M)a data for compounds 3h–3j and DMU-212 against a
Growth inhibition (GI50/l
M)a data for compounds 5e, 5j, and 7c–7e against a panel of
panel of 60 human cancer cell lines
60 human cancer cell types
Panel/cell line
3h
3i
3j
DMU-212
Panel/cell line
5e
5j
7c
7d
7e
GI50
GI50
GI50
GI50
GI50
GI50
GI50
GI50
GI50
(lM)
(
lM)
(l
M)
(lM)
(l
M)
(
lM)
(lM)
(lM)
(lM)
Leukemia
CCRF-CEM
HL-60(TB)
K-562
MOLT-4
RPMI-8226
SR
Leukemia
CCRF-CEM
HL-60(TB)
K-562
MOLT-4
RPMI-8226
SR
3.76
2.07
NAb
8.85
4.81
0.93
>100
>100
NA
>100
>100
2.76
0.21
0.04
NA
0.62
0.06
0.06
2.89
3.14
NA
3.22
6.42
3.93
0.25
0.24
0.04
0.45
0.49
0.05
1.56
1.49
0.39
2.81
3.23
0.53
3.65
3.50
0.79
4.99
4.49
1.60
1.91
1.57
0.53
3.71
3.85
5.44
0.05
0.03
0.04
0.07
0.19
0.03
Non-small cell lung cancer
A549/ATCC
HOP-62
HOP-92
NCI-H23
Non-small cell lung cancer
A549/ATCC
HOP-62
HOP-92
NCI-H23
4.12
2.98
9.64
6.61
2.80
21.7
0.06
0.05
0.12
0.16
0.05
3.70
3.14
7.23
3.37
3.70
0.27
0.30
4.22
0.86
0.04
2.65
3.30
7.42
6.33
2.58
5.00
3.77
4.55
6.45
2.89
2.65
2.14
12.0
5.09
0.09
0.07
17.00
0.32
0.02
>100
>100
>100
>100
NCI-H522
NCI-H522
2.37
Colon cancer
COLO 205
HCC-2998
HCT-116
HCT-15
HT29
KM12
SW-620
Colon cancer
COLO 205
HCC-2998
HCT-116
HCT-15
HT29
KM12
SW-620
2.77
8.85
2.71
0.73
NA
>100
7.78
6.33
2.10
NA
0.12
0.14
0.03
0.04
NA
2.07
3.59
3.23
2.82
2.32
3.63
2.07
0.14
5.26
0.22
0.07
0.06
0.07
0.12
1.59
11.9
2.68
0.87
0.71
2.25
1.23
2.58
11.0
3.72
1.77
3.02
3.77
3.72
1.61
5.42
2.41
0.75
1.60
2.39
0.72
0.03
0.27
0.04
0.04
0.03
0.05
0.05
1.59
1.04
4.26
3.93
0.05
0.04
CNS cancer
SF-268
SF-295
SF-539
SNB-19
SNB-75
U251
CNS cancer
SF-268
SF-295
SF-539
SNB-19
SNB-75
U251
5.38
2.49
2.56
4.08
2.16
3.97
>100
10.6
73.1
>100
21.9
>100
0.11
0.02
0.02
0.19
0.03
0.05
7.59
2.18
2.18
4.85
1.88
3.07
0.73
0.20
0.25
0.75
NAb
0.43
9.36
1.75
1.61
6.95
NA
8.99
2.96
2.71
6.29
2.45
3.85
9.34
1.91
1.58
4.24
1.14
2.71
1.64
0.04
0.04
0.23
0.02
0.06
3.59
Melanoma
LOX IMVI
M14
MDA-MB-435
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-62
Melanoma
LOX IMVI
M14
MDA-MB-435
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-62
4.92
1.63
0.26
1.84
4.59
0.84
1.04
NA
0.06
0.03
0.02
NA
0.06
0.03
0.03
4.89
2.81
1.04
3.95
3.86
2.50
2.37
0.57
0.12
0.03
0.21
0.45
0.21
0.59
5.33
1.32
0.30
0.72
4.47
2.58
2.32
4.22
3.13
0.43
3.18
5.84
3.20
2.63
3.96
1.68
0.31
2.90
4.52
1.52
1.51
0.05
0.04
0.02
0.04
0.24
0.06
0.06
3.87
0.51
36.8
>100
3.54
8.68
Ovarian cancer
IGROV1
OVCAR-3
OVCAR-4
NCI/ADR-RES
SK-OV-3
Ovarian cancer
IGROV1
OVCAR-3
OVCAR-4
NCI/ADR-RES
SK-OV-3
9.40
2.16
11.2
1.52
3.24
>100
37.1
24.9
4.45
>100
0.63
0.04
0.26
0.03
0.04
5.29
3.45
4.00
3.01
3.34
3.60
0.27
NA
0.08
0.40
8.80
3.10
1.65
2.41
2.75
5.95
4.18
8.25
2.52
3.58
4.07
3.04
3.79
1.65
2.35
0.09
0.04
1.27
0.04
0.06
Renal cancer
786-0
A498
ACHN
CAKI-1
UO-31
Renal cancer
786-0
A498
ACHN
CAKI-1
UO-31
3.88
1.35
7.61
2.92
5.01
35.8
0.04
0.03
0.08
0.05
NA
5.42
0.74
4.51
3.00
3.69
18.90
0.20
0.52
NA
4.93
1.33
4.92
NA
5.62
1.17
4.52
4.18
6.48
4.38
0.41
4.24
2.06
6.24
0.88
0.03
0.06
0.05
0.08
>100
>100
>100
>100
0.58
6.66
Prostate cancer
PC-3
DU-145
Prostate cancer
PC-3
DU-145
4.57
4.69
>100
>100
0.06
0.13
3.22
4.23
0.32
0.59
3.03
3.62
5.52
6.94
3.04
3.49
0.06
0.21
Breast cancer
MCF7
MDA-MB-231/ATCC
HS 578T
Breast cancer
MCF7
MDA-MB-231/ATCC
HS 578T
0.86
4.69
5.56
1.11
3.01
69.3
>100
4.29
0.04
0.17
0.07
0.03
1.66
3.74
3.09
2.22
0.08
2.11
0.46
0.28
0.85
3.36
3.62
0.69
1.99
5.11
3.54
2.11
0.61
2.43
2.63
0.51
0.09
0.18
0.08
0.38
MDA-MB-468
MDA-MB-468
GI50 values <1
l
M are bolded.
GI50 values <1 lM are bolded.
a
a
GI50: 50% growth inhibition, concentration of drug resulting in a 50% reduction
GI50: 50% growth inhibition, concentration of drug resulting in a 50% reduction
in net protein increase compared with control cells.
in net protein increase compared with control cells.
b
b
NA: not analyzed.
NA: not analyzed.
In vitro tubulin polymerization experiments were performed, as
described previously15 in order to test the ability of compounds 3j,
5e and 7e to exert their cytotoxic effects through inhibition of
tubulin dynamics. The tubulin polymerization assay measures
changes over time in optical density (O.D.) at 340 nM. An increase
H522 (non-small cell lung) cancer cells. The anti-cancer activity
declined significantly (only ꢀ25% of the GI50 values were <1
lM)
when the 4-methoxyphenyl group of compound 5e was replaced
with a 2,4-dimethoxyphenyl moiety (5j).
Please cite this article in press as: Madadi, N. R.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.03.068

N. R. Madadi et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
5
in the O.D. results from tubulin polymerization. The resulting curve
typically exhibits three phases: an initial nucleation phase, a poly-
merization phase, and a steady-state phase where there is slow
depolymerization of tubulin as the GTP substrate is depleted
(Fig. 2A).
tion rate increased from 1.3 ( 0.1) Â 10À5 A.U.s/s for the DMSO
control reaction to 1.6 ( 0.2) Â 10À5 A.U.s/s (P = 0.11) and 3.2
( 0.3) Â 10À5 A.U.s/s (P < 0.0001) in the presence of 10
lM and
100 lM 3j, respectively (where n = 3–6 experiments and values
reported represent the mean std. error) (Fig. 2C).
Compound 3j appeared to affect both the rapid polymerization
phase and the slow depolymerization phase in a concentration-
dependent manner (Fig. 2B and C). The tubulin polymerization rate
decreased from 8.6 ( 0.4) Â 10À4 A.U.s/s for the DMSO control
reaction to 6.9 ( 0.9) Â 10À4 A.U.s/s (P = 0.083) and 2.0 ( 1.2) Â
Compound 5e, on the other hand, did not alter the initial rate of
polymerization to any degree, changing from 8.6 ( 0.4) Â 10À4 A.U.
s/s for the DMSO control reaction to 8.0 ( 0.7) Â 10À4 A.U.s/s
(P = 0.51) and 8.9 ( 2.0) Â 10À4 A.U.s/s (P = 0.87) in the presence
of 10 lM and 100 lM 5e, respectively (where n = 3–6 experiments
10À4 A.U.s/s (P < 0.0001) in the presence of 10
lM and 100
l
M 3j,
and values reported represent the mean std. error) (Fig. 2B).
Compound 5e exhibited a slightly stronger effect on tubulin
depolymerization, as the tubulin depolymerization rate increased
from 1.3 ( 0.1) Â 10À5 A.U.s/s for the DMSO control reaction to
1.8 ( 0.6) Â 10À5 A.U.s/s (P = 0.0003) and 2.0 ( 0.3) Â 10À5 A.U.s/s
respectively (where n = 3–6 experiments and values reported rep-
resent the mean std. error) (Fig. 2B). The tubulin depolymeriza-
(P < 0.0001) in the presence of 10 lM and 100 lM 5e, respectively
(where n = 3–6 experiments and values reported represent the
mean std. error) (Fig. 2C). Finally, compound 7e did not alter
tubulin dynamics in any measurable way (Fig. 2B and C). Taken
together, it would appear that while 3j is able to actively disrupt
tubulin dynamics, 5e only destabilizes tubulin after GTP is
depleted, and 7e is without effect.
Among the most potent anti-cancer compounds identified in
this study, only 3j showed a significant effect in our tubulin
polymerization assay. In order to determine the binding mode of
compounds 3j and to compare it with that of the known tubulin-
inhibitor DMU-212, virtual docking of these compounds was
performed with the tubulin dimer. All docking studies were per-
formed using SwissDock, as described earlier.15 Briefly, atomic
coordinates of all the ligands were generated using MarvinSketch
(ChemAxon), and coordinates for tubulin were derived from the
crystal structure PDB 1SA0. All coordinates files were prepared
for docking using UCSF-Chimera, and submitted for docking to
the SwissDock server. Resulting binding modes were ranked based
on Full-Fitness (FF) scores generated by SwissDock, and the top-
scoring poses for each compound were examined in detail manu-
ally to identify their interactions with tubulin.
Both 3j and DMU-212 docked at the colchicine-binding site at
the dimer interface of tubulin (Fig. 3A). Table 3 shows the docking
(FF) scores for these compounds. Interestingly, the FF score for the
tubulin inhibitor DMU-212 was observed to be significantly lower
(À2198.7 kcals/mol) than compound 3j in our docking studies.
None of the compounds make any polar contacts with any resi-
dues of tubulin. Both 3j and DMU-212 are stabilized through van
der Waals’ interactions at the colchicine-binding site on tubulin.
DMU-212 interacts with residues of b-tubulin only, whereas com-
pound 3j makes additional van der Waals’ contacts with four resi-
dues of a-tubulin, besides sharing all the b-tubulin contacts made
by DMU-212 (Fig. 3B). This would explain the low FF score of
DMU-212 as compared to 3j.
In conclusion, a series of (Z)-2-(benzo[d][1,3]dioxol-5-yl) and
(Z)-2,3-dihydrobenzo[b][1,4]dioxin-6-yl analogs of 2- and
3-phenylacetonitriles has been synthesized and evaluated for their
anti-cancer activities against a panel of 60 human cancer cell lines.
Eight compounds (3h–3j, 5e, 5j, and 7c–7e) were identified as
molecules of interest from a single dose anti-cancer screening
assay. These compounds were then evaluated for dose-dependent
growth inhibition and cytotoxicity against the 60 human cancer
cell panel. From the eight compounds selected, analogs 3j and 7e
exhibited the most potent anti-cancer activity with GI50 values of
<100 nM against a significant number of cell lines in the panel.
Figure 2. In vitro tubulin polymerization in the presence of 3j, 5e, and 7e. Panel A.
Representative tubulin polymerization results. The reaction was initiated by raising
the temperature of tubulin from 4 °C to 37 °C and then monitoring changes in O.D.
at 340 nm. The three phases of the tubulin polymerization assay are noted. The rate
of tubulin polymerization (Panel B) and depolymerization (Panel C) was measured
in the presence of the indicated concentrations of 3j, 5e, and 7e. DMSO served as a
Compound 5e exhibited GI50 values <1 lM against most of the
human cancer cell lines in the panel. These analogs also showed
superior growth inhibition when compared to the structurally
related tubulin inhibitor, DMU-212. Tubulin polymerization
was inhibited in vitro by 3j, and molecular docking studies
control reaction. Asterisks represent
Student’s t-test.
P values, which were calculated using a
Please cite this article in press as: Madadi, N. R.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.03.068

6
N. R. Madadi et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
are potent anticancer agents but exhibit little or no inhibition of
tubulin polymerization. The mechanism by which 5e and 7e exert
their cytotoxic effects remains unknown at this stage, but it is unli-
kely that they affect tubulin dynamics. Nevertheless, the results
from this study suggest that (Z)-2-(benzo[d][1,3]-dioxol-5-yl) and
(Z)-2,3-dihydrobenzo[b][1,4]-dioxin-6-yl analogs of 2- and
3-arylacetonitriles may have potential for development as clinical
candidates to treat a variety of human cancers.
Acknowledgements
We are grateful to the NCI/NIH, United States (Grant Number
CA 183895 and COBRE Award Number P20GM109005), to the
Arkansas Research Alliance Scholar award (to P.A.C.) and to the
NCI drug screening program.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.03.
068.
References and notes
1. Pettit, G. R.; Singh, S. B.; Niven, M. L.; Hamel, E.; Schmidt, J. M. J. Nat. Prod. 1987,
50, 119.
2. Pettit, G. R.; Singh, S. B.; Hamel, E.; Lin, C. M.; Alberts, D. S.; Garcia-Kendall, D.
Experientia 1989, 45, 209.
3. Greene, L. M.; Wang, S.; O’Boyle, N. M.; Bright, S. A.; Reid, J. E.; Kelly, P.;
Meegan, M. J.; Zisterer, D. M. Oncol. Rep. 2013, 29, 2451.
4. Madadi, N. R.; Penthala, N. R.; Song, L.; Hendrickson, H. P.; Crooks, P. A.
Tetrahedron Lett. 2014, 55, 4207.
Figure 3. Docking poses at the colchicine binding site of tubulin. a. The tubulin
dimer is shown as a cartoon, with a-tubulin in black and b-tubulin in light gray. Top
binding poses of the compounds 3j (red) and DMU-212 (green) are shown at the
dimer interface. b. Zoomed view for the top poses of DMU-212 (red) and 3j (green)
are shown at the colchicine-binding site of tubulin in each panel. Residues of
tubulin involved in van der Waals’ interactions with the two compounds are labeled
in each panel.
5. Penthala, N. R.; Janganati, V.; Bommagani, S.; Crooks, P. A. Med. Chem. Commun.
2014, 5, 886.
6. Lee, L.; Robb, L. M.; Lee, M.; Davis, R.; Mackay, H.; Chavda, S.; Babu, B.; O’Brien,
E. L.; Risinger, A. L.; Mooberry, S. L.; Lee, M. J. Med. Chem. 2010, 53, 325.
7. Penthala, N. R.; Sonar, V. N.; Horn, J.; Leggas, M.; Yadlapalli, J. S.; Crooks, P. A.
Med. Chem. Commun. 2013, 4, 1073.
8. General synthetic procedure synthesis of (Z)-2-(benzo [d][1,3]dioxol-5-yl) and (Z)-
2,3-dihydrobenzo[b] [1,4]dioxin-6-yl analogs of 2- and 3-phenylacetonitriles (3a–
3k, 5a–5l, and 7a–7h): the aromatic carbaldehyde (1 mole) and the appropriate
substituted phenylacetonitrile (1.1 mol equiv) were added to 5% sodium
methoxide in methanol and the mixture was stirred at reflux temperature
for 2–3 h. The resulting solid was filtered off, washed with water, and finally
washed with cold methanol. The obtained crude solid was recrystallized from
Table 3
Docking scores of 3j and DMU-212 with tubulin
Compound
FF score (kcal/mol)
3j
À4217.6
À2198.7
DMU-212
methanol to afford the desired product as
a pure crystalline solid. The
demonstrated that 3j has a higher affinity for the colchicine-bind-
ing site on tubulin when compared to DMU-212. Compounds 5e
and 7e do not appear to exert cytotoxicity through inhibition of
tubulin dynamics.
It should be noted that tubulin inhibitor 3j is a member of the
(Z)-2,3-dihydrobenzo[b]-[1,4]dioxin-6-yl-2-arylacrylonitrile series
of analogs, whereas compounds 5e and 7e, belong to the
(Z)-2-(benzo[d][1,3]dioxol-5-yl-3-arylacrylonitrile series of ana-
logs. Molecular docking studies show that the 3-(3,4-ethylene-
dioxyphenyl)-acetonitrile moiety in the dihydrodioxin analog 3j
is recognized by tubulin and similar to DMU-212 binds to the col-
chicine binding site on tubulin. Dioxol compounds 5e and 7e both
contain a 2-(3,4-methylenedioxyphenyl)-acetonitrile moiety, and
characterization data for the active compounds is given in reference section.
9. Rubinstein, L. V.; Shoemaker, R. H.; Paull, K. D.; Simon, R. M.; Tosini, S.; Skehan,
P.; Scudiero, D. A.; Monks, A.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82, 1113.
10. Penthala, N. R.; Yerramreddy, T. R.; Madadi, N. R.; Crooks, P. A. Bioorg. Med.
Chem. Lett. 2010, 20, 4468.
11. Janganati, V.; Penthala, N. R.; Madadi, N. R.; Chen, Z.; Crooks, P. A. Bioorg. Med.
Chem. Lett. 2014, 24, 3499.
12. Androutsopoulos, V. P.; Ruparelia, K. C.; Papakyriakou, A.; Filippakis, H.;
Tsatsakis, A. M.; Spandidos, D. A. Eur. J. Med. Chem. 2011, 46, 2586.
13. Sale, S.; Verschoyle, R. D.; Boocock, D.; Jones, D. J. L.; Wilsher, N.; Ruparelia, K.
C.; Potter, G. A.; Farmer, P. B.; Steward, W. P.; Gescher, A. J. Br. J. Cancer 2004,
90, 736.
14. Madadi, N. R. C.; Lett, P. A. Drug Design Discovery 2015, 12, 521.
15. Madadi, N. R.; Penthala, N. R.; Howk, K.; Ketkar, A.; Eoff, R. L.; Borrelli, M. J.;
Crooks, P. A. Eur. J. Med. Chem. 2015, 103, 123.
Please cite this article in press as: Madadi, N. R.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.03.068