Design, synthesis, and identification of a novel napthalamide-isoselenocyanate compound NISC-6 as a dual Topoisomerase-IIα and Akt pathway inhibitor, and evaluation of its anti-melanoma activity

Article abstract of DOI:10.1016/j.ejmech.2017.04.052

Synthesis and anti-melanoma activity of novel naphthalimide isoselenocyanate (NISC) and naphthalimide selenourea (NSU) analogs are described. The novel agents were screened for growth inhibition of different human melanoma cell lines including those having BRAF^V600E mutation (UACC903, 1205Lu, and A375M) and BRAF^WT (CHL-1). In general, the NISC analogs (4a-d) were more effective in inhibiting the cell viability than the NSU analogs (7a-b). Overall, NISC-6 (4d), having a six-carbon alkyl chain, was identified as the most cytotoxic compound in both BRAF^V600E mutated and BRAF^WT cells. NISC-6 docked strongly into the binding sites of Akt1 and human topoisomerase IIα (Topo-IIα), and the docking results were supported by experimental findings showing NISC-6 to inhibit of both Akt pathway and Topo-IIα activity in a dose dependent manner. Furthermore, NISC-6 effectively induced apoptosis in human melanoma cells, inhibited tumor growth by ～69% in a melanoma mouse xenograft model, and showed excellent compliance with the Lipinski’ rule of five, suggesting both its efficacy and drug-like behavior under physiological conditions.

Full text of DOI:10.1016/j.ejmech.2017.04.052

Accepted Manuscript
Design, synthesis, and identification of a novel napthalamide-isoselenocyanate
compound NISC-6 as a dual Topoisomerase-IIα and Akt pathway inhibitor, and
evaluation of its anti-melanoma activity
Deepkamal N. Karelia, Ugir Hossain Sk, Parvesh Singh, A.S. Prakasha Gowda,
Manoj K. Pandey, Srinivasa R. Ramisetti, Shantu Amin, Arun K. Sharma
PII:
S0223-5234(17)30318-5
10.1016/j.ejmech.2017.04.052
EJMECH 9403
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 9 December 2016
Revised Date: 11 April 2017
Accepted Date: 20 April 2017
Please cite this article as: D.N. Karelia, U.H. Sk, P. Singh, A.S.P. Gowda, M.K. Pandey, S.R. Ramisetti,
S. Amin, A.K. Sharma, Design, synthesis, and identification of a novel napthalamide-isoselenocyanate
compound NISC-6 as a dual Topoisomerase-IIα and Akt pathway inhibitor, and evaluation of its anti-
melanoma activity, European Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.04.052.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Design, synthesis, and identification of a novel
napthalamide-isoselenocyanate compound NISC-6 as a
dual Topoisomerase-IIα and Akt pathway inhibitor, and
evaluation of its anti-melanoma activity
Deepkamal N. Karelia1, Ugir Hossain Sk2, Parvesh Singh3, A. S. Prakasha Gowda4, Manoj K.
Pandey1,5, Srinivasa R. Ramisetti1, Shantu Amin1, and Arun K. Sharma1,
1Department of Pharmacology; Penn State Cancer Institute, CH72; 4Department of Biochemistry
and Molecular Biology, Penn State College of Medicine, 500 University Drive, Hershey, PA
17033, USA.
2CSIR-Institute of Himalayan Bioresource Technology, Palampur, HP, India
3School of Chemistry and Physics, University of Kwa-Zulu Natal (UKZN), Westville campus,
Durban 4000, South Africa
5Department of Biomedical Sciences, Cooper Medical School of Rowan University, 401 S
Broadway, Camden, NJ 08103 (current address)
* Author to whom correspondence should be addressed
Arun K. Sharma, Ph.D.
Department of Pharmacology
Penn State College of Medicine
Penn State Cancer Institute, CH72
Penn State Milton S. Hershey Medical Center
500 University Drive
Hershey, PA 17033
Phone: 717-531-4563
Fax:
717-531-0244
E-mail: aks14@psu.edu
1

ACCEPTED MANUSCRIPT
Abstract
Synthesis and anti-melanoma activity of novel naphthalimide isoselenocyanate (NISC) and
naphthalimide selenourea (NSU) analogs are described. The novel agents were screened for
growth inhibition of different human melanoma cell lines including those having BRAFV600E
mutation (UACC903, 1205Lu, and A375M) and BRAFWT (CHL-1). In general, the NISC analogs
(4a-d) were more effective in inhibiting the cell viability than the NSU analogs (7a-b). Overall,
NISC-6 (4d), having a six-carbon alkyl chain, was identified as the most cytotoxic compound in
both BRAFV600E mutated and BRAFWT cells. NISC-6 docked strongly into the binding sites of
Akt1 and human topoisomerase IIα (Topo-IIα), and the docking results were supported by
experimental findings showing NISC-6 to inhibit of both Akt pathway and Topo-IIα activity in a
dose dependent manner. Furthermore, NISC-6 effectively induced apoptosis in human melanoma
cells, inhibited tumor growth by ~69% in a melanoma mouse xenograft model, and showed
excellent compliance with the Lipinski’ rule of five, suggesting both its efficacy and drug-like
behavior under physiological conditions.
Keywords: Naphthalimide, isoselenocyanate, selenourea, melanoma, apoptosis, anti-tumor
2

ACCEPTED MANUSCRIPT
1. Introduction
Melanoma accounts for less than 5% of skin cancer cases, but causes more than 75% of skin
cancer deaths. The vast majority of melanomas can be attributed to exposure to ultraviolet (UV)
radiation from the sun [1]. The incidence and mortality rate for melanoma continue to rise [2]; an
estimated 87,110 new cases of invasive melanoma will be diagnosed and an estimated 9,730
people will die of melanoma in the U.S. in 2017 [3]. However, no effective treatment currently
exists for patients suffering from the metastatic stage of this disease [4], which has poor prognosis
and is refractory to most conventional chemotherapies. Currently, the limited treatment options
for melanoma patients include dacarbazine and temozolomide, but both have a disappointing
response rate. The BRAFV600E inhibitor PLX4032 (vemurafenib) approved recently [5, 6], has
shown great therapeutic promise but suffers from onset of resistance in about 7 months [7-11]
leading to unsatisfactory outcomes. The combination therapies being tested in clinical trials have
also shown limited success. Multiple antibodies against programmed cell death-1 (PD-1) and its
ligand (PDL-1) are in development and have shown promise in melanoma and other
malignancies. Two of these, Opdivo (BMY) and Keytruda (Merck), were recently approved by
FDA for metastatic melanoma in 2014. However, despite clinical benefits, the checkpoint
inhibition is associated with a unique spectrum of side effects termed immune-related adverse
events (irAEs) that include dermatologic, gastrointestinal, hepatic, endocrine, and other
inflammatory events [12].
To address this unmet need, our laboratory has been involved extensively, over the last
eight years, in developing novel agents to treat melanoma [13-16]. We earlier developed
isoselenocyanate (ISC) compounds based on naturally occurring isothicyanates (ITCs) that
inhibited melanoma tumor growth by targeting the Akt signaling pathway [13, 14]. Naturally
3

ACCEPTED MANUSCRIPT
occurring and synthetic phenylalkyl ITCs[17-19] are well established anti-cancer agents against a
variety of organ sites [20, 21]. Our structural activity relationship (SAR) studies directed towards
the isosteric replacement of sulfur (S) in ITCs by selenium (Se) and varying the alkyl chain length
connecting the phenyl ring and the ISC functionality, led to the identification of phenylbutyl
isoselenocyanate (ISC-4) as the most promising agent in this series of compounds [13]. ISC-4 was
found to be an effective anti-cancer agent [13, 16, 22, 23], being particularly effective against
melanoma [13, 14, 16]. It inhibited Akt signaling pathway and significantly retarded melanoma
tumor growth without any systemic toxicity [13, 14, 16]. Exploiting the known anti-cancer
properties of naturally occurring ITCs, we recently also reported on development of
naphthalimide-isothiocyanate (NITC) compounds [15] with promising activity against melanoma.
Several naphthalimide analogs with varying efficacy have been reported in literature as having
promising anti-cancer activities [24-29]. Mitonafide, a naphthalimide analog and a well-known
Topoisomerase-IIα (Topo-IIα) inhibitor, showed antitumor activity both preclinically [30] and in
phase I and phase II clinical trials [31-33], but failed due to systemic toxicity issues [31-34]. We
conducted a SAR study by introducing ITC (-N=C=S) functionality and altering the chain length
to generate a series of NITC analogs that were effective in inhibiting melanoma tumor growth,
while substantially reduced the systemic toxicity associated with mitonafide, resulting in more
efficacious agents with higher therapeutic index. Our studies showed that the incorporation of -
N=C=S functionality into mitonafide structure drastically reduced its systemic toxicity issues
[15], possibly because of the enhanced reactivity of the compound towards glutathione and
cysteine residues leading to the formation of conjugates that would excrete relatively faster from
the body.
4

ACCEPTED MANUSCRIPT
In the present study, we hypothesized that introducing the ISC (-N=C=Se) functionality of
Akt pathway inhibitor ISC-4, into napthalimide moiety, as in Topo-IIα inhibitor mitonafide,
would result in a dual Akt pathway/Topo-IIα activity inhibitor with enhanced antitumor activity.
Therefore, we synthesized a series of naphthalimide-ISC (NISC) compounds. In addition, since –
N=C=Se follows similar mechanism as –N=C=S functionality [35], the new NISC compounds
were also expected to have a similar safety profile as NITC compounds at effective doses, thus
having a high therapeutic index. Furthermore, we also synthesized a series of naphthalimide
analogs by incorporating a selenourea functionality in place of N,N-dimethyl in mitonafide, with
or without nitro substitution to generate napthalimide-selenourea (NSU) compounds. Both the
NISC and NSU compounds were screened for their potency in inhibiting human melanoma cells
viability, and the best identified compound, NISC-6 (4d), was evaluated for its anti-apoptotic
activity on both BRAFV600E mutant and BRAFWT melanoma cells. Further, NISC-6’s mechanism
of action was characterized in vitro, and its tumor inhibitory activity was evaluated in melanoma
xenograft mouse model.
2. Results and discussion
2.1.Design.
To create a dual inhibitor of Topo-IIα and Akt signaling pathway, we intended to adapt a
fragment based approach by blending –N=C=Se functionality of Akt pathway inhibitor ISC-4
and naphthalimide moiety of Topo-IIα inhibitor mitonafide (Figure 1).
5

ACCEPTED MANUSCRIPT
Figure 1. Fragment based design of dual inhibitor NSIC-6 using key functionalities of
Mitonafide and ISC-4
In addition to this, we also took advantage of our previously published SAR results on NITC
compounds. Two compounds, the NNITC-2 obtained by replacement of N,N-dimethylamino
group of mitonafide by the –N=C=S functionality, and NITC-6 obtained by extending the alkyl
chain length to six carbons (Figure 2), were identified as the most potent agents that effectively
inhibited melanoma cell viability, induced apoptosis, and inhibited tumor growth in a melanoma
xenograft model [15]. The study also showed that increasing the alkyl chain length from 2
carbons (in NITC-2) to 6 carbons (in NITC-6), and substituting NITC-2 by a nitro group (as in
NNITC-2) enhanced the potency of the compounds. Therefore, we first synthesized
corresponding isosteric Se analogs, NISC-6 and NNISC-2, of the two NITC compounds (Figure
2). Interestingly, while the potency increased as expected in case of NISC-6, NNISC-2 was in
fact less cytotoxic than NNITC-2. This suggested that unlike nitro substituted NNITC-2, which
6

ACCEPTED MANUSCRIPT
was more potent than unsubstituted NITC-2, the nitro substitution at 3-position of naphthalimide
ring system was not favorable in case of Se analogs. Therefore, we next focused on studying the
chain length effect on unsubstituted NISC scaffold and synthesized compounds with two carbon
(NISC-2) and four carbon (NISC-4) alkyl chains (Figure 2). However, in case of NISC analogs
the increase in chain length did not seem to have a consistent contribution towards activity,
although, NISC-6 with six carbon alkyl chain was identified as the most effective compound.
O
O
O
Nitro group
incorporation
N=C=S
N=C=S
Increase in Chain length
Increases in potency
N
N
N=C=S
N
O
O
Increases in potency
O
NITC-6
NITC-2
NO2
NNITC-2
Se incorporation
Se incorporation
O
O
O
9
1
N=C=Se
N=C=Se
O
8
7
N=C=Se
2N
N
N
3
O
O
6
4
5
NISC-4
NISC-6
NO2
O
NNISC-2
O
O
N
O
Se NH2
.HBr
S
NH2
.HBr
Se NH2
Se incorporation
R = H
N
Se incorporation
R = NO2
N
NH
NH
NH
.HBr
O
O
NSU-2
NNSU-2
R
R = H (NTU-2)
R = NO2 (NNTU-2)
Figure 2. Rational designing of novel compounds.
Previously, we also synthesized naphthalimide-thiourea (NTU) analogs designed by in
substituting N,N-dimethylamino group in mitonafide with a thiourea functionality, with or without
7

ACCEPTED MANUSCRIPT
nitro substitution [15]. Although, NTU analogs [15] were less effective than the corresponding
NITC compounds, literature reports have shown that selenourea functionality when incorporated
into certain moieties in place of thiourea can significantly enhance the potency of the compound
[36-38]. For example, isosteric Se analog of S,S′-1,4-phenylenebis(1,2-ethanediyl)bis-isothiourea
(PBIT), a well-known chemopreventive agent that exerts its chemopreventive efficacy by
inhibiting inducible nitric oxide synthase (iNOS) [39], by selenourea, as in PBISe [36, 38],
significantly enhanced the cytotoxicity and tumor inhibitory action of the compound, particularly
in melanoma xenograft models [36]. We have also recently shown various heterocyclic-
selenoureas to be more potent anticancer agents than their corresponding thiourea analogs [37].
Therefore, to evaluate the effect of selenourea group when attached to a naphthalimide moiety,
we also synthesized NSU compounds (7a-b) by replacing N,N-dimethylamino group in
mitonafide by a selenourea functionality, with or without nitro substitution. Although, literature
reports have shown potency to increase when thiourea functionality is replaced by selenourea in
certain scaffolds [36-38], in the present case the NSU compounds were no better than their NTU
counterparts.
2.2.Chemistry
The NISC compounds (4a-d) were synthesized as outlined in Scheme 1 following our
earlier reported method for phenylalkyl isoselenocyanates [13]. The key intermediates,
aminoalkyl-naphthalimides 2a-d, were synthesized by refluxing 1,8-naphthalic anhydride or 3-
nitro-1,8-naphthalic anhydride with appropriate diaminoalkane following previously reported
methods [15, 40, 41]. Formylation of compounds 2a-d using ethyl formate and potassium
carbonate under refluxing conditions gave formamides 3a-d, which on treatment with
8

ACCEPTED MANUSCRIPT
triphosgene and selenium powder in the presence of triethylamine in refluxing methylene
chloride [13, 42] furnished the desired NISCs 4a-d in good yields. The structural features for the
formamides can be explained by taking specific example 3a, which exhibited presence
1
characteristic peaks for –CHO at
δ
7.94 (s, 1H) and –NH at
δ
8.14 (br s, 1H) as singlets in H
NMR spectrum. Further in the 13C NMR spectrum, the presence of carbonyl carbon of -NH-
CHO functionality at 164.1 ppm along with two carbonyl carbons of thalimide –(C=O)2-N-
functionality at
δ
δ
161.7 ppm supported the structure of the compound 3a. In similar way, the
1
structures of all other formamides 3b-d were confirmed by H NMR, 13C NMR and mass
spectral analysis.
Similarly, the structural features of naphthalimide isoselenocyanates 4a-d can be
explained by taking specific example 4a, which exhibited the absence of characteristic peak for –
NH (as in precursor compound 3a) along with de-shielding of CH2 protons attached to the
1
−N=C=Se functionality at
δ
3.89 (t, 2H) as triplet integrating for two protons in H NMR
spectrum. Further the presence of two nearly overlapping carbonyl carbons of naphthalimide –
164.06 and 164.02 ppm in the 13C NMR spectrum confirmed the
(C=O)2-N- functionality at
δ
structure of the compound 4a.The structures of isoselenocyanates 4b-d were also confirmed
1
13
based on H NMR, C NMR and mass spectral analysis. It is interesting to note that while two
naphthalimide carbonyl group peaks of compounds 3a and 4a as well as other symmetric analogs
3c,d and 4c,d overlap or nearly overlap, compounds 3b and 4b showed well separated carbonyl
peak signals due to the presence of the nitro group.
9

ACCEPTED MANUSCRIPT
O
O
O
O
NHCHO
NH2
N
n
N
N
C
O
N
n
n
Se
(i)
(iii)
(ii)
O
O
O
O
R
R
R
R
1
4a, R = H, n = 1 (NISC-2)
b, R = NO2, n = 1 (NNISC-2)
c, R = H, n = 3 (NISC-4)
d, R = H, n = 5 (NISC-6)
3a, R = H, n = 1
b, R = NO2, n = 1
c, R = H, n = 3
d, R = H, n = 5
R = H, NO2
2a, R = H, n = 1
b, R = NO2, n = 1
c, R = H, n = 3
d, R = H, n = 5
(iv)
O
N
O
N
O
N
O
OH
Br
Se NH2
NH
(v)
(vi)
.HBr
O
O
R
R
R
5a, R = H
b, R = NO2
7a, R = H (NSU-2)
b, R = NO2 (NNSU-2)
6a, R = H
b, R = NO2
Scheme 1. Synthesis of naphthalimide-isoselenocynate (4) and selenourea (7) derivatives.
Reagents and conditions: (i) diaminoalkane, EtOH, reflux, 2 h (ii) ethyl formate, K2CO3, reflux,
6 h (iii) thiophosgene, Se powder, Et3N, CH2Cl2, reflux (iv) ethanolamine, EtOH, reflux, 1 h (v)
PBr3, EtOAc, 80 ˚C, 2 h (vi) selenourea, CH3CN, reflux, 4 h.
The NSU compounds (7a-b) were synthesized as shown in Scheme 1. The precursor
bromoalkyl naphthalimides 6a-b were synthesized following our previously reported methods in
two steps: refluxing 1,8-naphthalic anhydride or 3-nitro-1,8-naphthalic anhydride with
ethanolamine in ethanol gave alcohols 5a-b which on treatment with PBr3 in refluxing ethyl
acetate yielded compounds 6a-b. The desired NSUs 7a and 7b were finally obtained by refluxing
6a-b with selenourea in acetonitrile for 4 h. The compounds were purified by silica gel column
chromatography and characterized on the basis of NMR and high-resolution MS. For example,
1H NMR spectrum of 7a exhibited the de-shielded protons of CH2 functionality attached to
isoselenouronium moiety at
δ 3.54 (t, 2H) as triplet integrating for two protons. Further the
presence of two carbonyl carbons of naphthalimide –(C=O)2-N- functionality at
δ
167.4 and
10

ACCEPTED MANUSCRIPT
164.0 ppm confirmed the structure of the compound 7a. In similar way, the structure of other
NSU compound 7b was confirmed by 1H NMR, 13C NMR and mass spectral analysis.
2.3. Biology
2.3.1. Evaluation of the anti-proliferative activities of the novel molecules and SAR
analysis
The effects of all the novel agents on cancer cell viability were evaluated using MTT assay. The
compounds were screened in the four melanoma cell lines (human melanoma cells with
BRAFV600E mutation (UACC903, 1205Lu, and A375M) and cells having BRAFWT (CHL-1)) at
different concentration, in the range of 0 to 50 µM, and at 24, 48, 72 h time points. The precursor
compound mitonafide, and clinically used BRAFV600E inhibitor PLX-4032 were used as controls.
Table 1 lists the IC50 for all the compounds in the four cell lines at three different time points.
NISC compounds (4a-d) were in general more effective in inhibiting cell viability than the NSU
derivatives (7a-b). Nitro substitution (compare NISC-2 vs. NNISC-2 and NSU-2 vs. NNSU-2) in
fact reduced the potency of the compounds. This was in contrast to our previous report where
nitro-substituion at 3-position of the naphthalimide ring enhanced the potency of the isosteric S
analogs (NITC) of NISC compounds [15], suggesting that replacing S by Se can change the
overall properties of the molecules. Furthermore, in previously published NITC compounds, the
cytotoxicity increased with increasing alkyl chain length, while in the present case the chain
effect was not consistent [15]. As indicated by Table 1, two compounds stood out as the most
potent at reducing the cell viability, were NISC-2 and NISC-6, having the naphthalimide and
isoselenocyanate moieties separated by 2 and 6 carbon alkyl chains, respectively. Considering
11

ACCEPTED MANUSCRIPT
IC50 values at different time points, NISC-6 was identified as the most effective compound.
NISC-6 was more effective than PLX-4032 at early time point (24 h) in all cell lines tested
(Table 1) and also exhibited a better potency at 48 and 72 h treatment in BRAFV600E mutated
UACC903 cells. In BRAFV600E mutated 1205Lu and A375M cell lines, PLX-4032 was relatively
more effective at 48 and 72 h treatment points. It is noteworthy that NISC-6 was ~15 times more
effective in BRAFWT CHL-1 cells compared to PLX-4032, indicating that NISC-6 was equally
effective on both BRAF mutant and WT cells.
Table 1. Effect of novel compounds on different melanoma cancer cell lines.
IC50 (µM)
NISC-2 NISC-4 NISC-6 NNISC-2
PLX-
4032
>50
NSU-2
NNSU-2 Mitonafide
13.1±1.4 13.4±2.5
15.9±0.9 10.6±4.0
15.4±1.6 13.4±2.6
13.1±1.4 13.4±2.5
9.2±1.3 2.5±0.4
12.0±1.3 1.7±0.3
19.4±1.4 11.8±2.1
21.0±1.6 2.7±0.4
28.4±5.4 4.4±0.5
9.7±1.3 13.7±2.9
8.0±1.2 1.5±0.1
8.9±1.4 0.8±0.0
21.6±2.3
19.9±1.3
20.3±4.2
21.6±2.3
14.4±0.8
15.9±2.5
24.8±1.3
21.0±0.8
24.0±3.7
15.3±0.0
12.6±0.0
11.0±3.2
46.8±8.6
11.5±4.8
8.2±1.2
46.8±8.6
12.8±1.2
12.1±1.5
>50
11.1±3.9
0.9±0.1
0.6±0.1
11.4±3.9
0.8±0.1
0.5±0.0
3.8±0.4
1.5±0.1
1.3±0.1
2.8±0.4
0.9±0.1
0.5±0.0
11.0±0.1
10.7±0.1
9.4±0.1
6.3±0.0
5.5±0.0
5.9±0.0
9.7±1.2
7.8±0.4
8.0±0.9
6.3±0.3
6.8±0.7
7.6±0.5
>50
>50
39.6±9.5
>50
24
48
72
24
48
72
24
48
72
24
48
72
1205Lu
UACC903
A375M
7.6±1.6
2.0±0.1
>50
18.1±2.8
14.0±1.9
>50
12.3±3.2
2.7±1.00
>50
2.05±0.6
0.5±0.2
>50
17.9±2.9
28.7±9.1
17.9±3.8
5.2±0.5
5.1±0.0
>50
25.8±4.5
>50
8.5±1.5
13.5±1.6
CHL-1
20.0±3.5
12.7±1.9
Data represent mean IC50 values (± SD) of % cell viability determined by the MTT assay in
triplicates. IC50 values were obtained by performing non-linear regression (variable slope)
analysis using GraphPad Prism.
Figure 3A shows the cell viability curves for NISC-6 in UACC903 cell line at 24, 48,
and 72 h time points. Comparison of NISC-6 with its precursor compounds, mitonafide and ISC-
4, revealed that NISC-6 was more cytotoxic than ISC-4 but was slightly less effective than
12

ACCEPTED MANUSCRIPT
mitonafide (Figure 3B). Since our goal was also to make the compound more tolerable than
mitonafide, we compared the effects of NISC-6 with mitonafide on normal cells viability.
Human dermal fibroblast cells (nHDFs) were used as the representative normal cells. Effect of
NISC-6 and mitonafide on viability of nHDFs was evaluated using MTT assay. As represented
by the bar graph in figure 4, at the concentration of 2.5 and 5 µM, NISC-6 showed more than
70% viable nHDF cells, while mitonafide showed less than 20% viable cells. Overall, this data
indicate that although NISC-6 is relatively less potent, it is more tolerable than mitonafide, and is
more selective towards cancer cells than normal cells.
Figure 3. NISC-6 potently inhibited cell viability of melanoma cells. (A) Melanoma cells
(UACC903) were treated for 24, 48 and 72h at different concentrations of NISC-6. MTT assay
was performed as described in Method and Material section. Cell viability curves were obtained
by performing non-linear regression using GraphPad Prism software. Calculated IC50 for all the
time points are shown in Table 1. (B) All three compound (NISC-6, ISC-4 and Mitonafide) were
treated for 48h and growth curves were obtained in a similar manner as (A).
13

ACCEPTED MANUSCRIPT
Figure 4. NISC-6 is more tolerable than Mitonafide in normal fibroblast cells, while at the
same time effective on cancer cells. Melanoma cells (UACC903) and normal human fibroblasts
(nHDFs) were treated for 48h at given concentrations of NISC-6 and mitonafide. MTT assay was
used to measure cell viability. Bar graph represents mean ± SD.
2.3.2. Molecular docking study of NISC-6 in the active sites of Akt and Topo IIα
NISC-6 was designed to be a dual inhibitor of Topo-IIα activity and Akt signaling pathway. In
order to afford some guidance for its design, in silico docking simulations were conducted using
the CDOCKER algorithm [43] in Discovery Studio (DS) program. NISC-6 was docked into both
Akt1 and Topo-IIα structures and compared with ISC-4 and mitonafide, respectively. Prior to
docking these compounds, the re-docking of the co-crystallized ligand into the binding site of
each protein was conducted to check the predictive efficiency of the docking protocol. For this
purpose, the native ligands, EVP1 (etoposide) and XM11 were docked with Topo-IIα and Akt1,
respectively, and their best poses were identified based on the scoring function (-CDOCKER
energy). The predicted docking pose and its X-ray structure showed a very good structural
correlation (Figure S1a-b, Supplementary Material) based on the favorable computed root mean
square deviation (< 1Å). Moreover, the predicted poses displayed interactions with similar amino
acid residues of Akt1 (Glu234, Arg4) and Topo-IIα (Arg503, DG13, Asp479, DA12) as observed
14

ACCEPTED MANUSCRIPT
in their native ligands, and thus validated the reproducibility efficiency of the protocol. NISC-6
(4d) and the reference compounds were subsequently docked into the binding sites of Akt1 and
Topo-IIα (details in method section) using the same docking protocol (Figure 5). The computed
negative binding energies (BEs) suggested the tendency of each compound to inhibit the proteins
under investigation. NISC-6 (BE = -76.7 kcal mol-1) was found to be the stronger inhibitor of
Topo-IIα than mitonafide (BE = -37.8 kcal mol-1). NISC-6 (BE= -63.7 kcal mol-1) also exhibited
stronger interaction with Akt1 as compared to ISC-4 (BE = -27.4 kcal mol-1). NISC-6 thus
demonstrated stronger interactions with both Akt1 and Topo-IIα as compared to the reference
controls ISC-4 and mitonafide, respectively, and therefore promise to be a potent dual inhibitor.
The ligand-protein complexes visualized further to understand their binding modes
revealed the predominance of electrostatic and hydrophobic forces accounting for their host-
guest relationship. NISC-6 (Figure 5A), for instance, exhibited electrostatic interactions with
Asp292 (cation-π) and Lys179 (anion-π), a hydrophobic interaction (π−π stacking) with Phe161
and a non-classical hydrogen bond (2.6Å) with Glu234 of Akt1 protein. ISC-4 in contrast missed
all those interactions observed in case of NISC-6 and mitonafide, and displayed only
hydrophobic (π−π stacking) interaction with Phe442 through its phenyl ring (Figure 5B).
In case of Topo-IIα, NISC-6 (Figure 5C) fitted nicely into the binding cavity via two
non-classical hydrogen bonds (2.98 Å, 2.99Å) with Arg503 through its alkyl moiety (-CH2-) and
multiple hydrophobic interactions (π−π stacking and π-alkyl) with the guanine (DG13) and
adenine (DA12) nucleotides. Mitonafide exhibited hydrophobic forces with cytosine (DC8),
thymine (DT9) and guanine (DG13) nucleotides of topo-II in addition to two non-classical
hydrogen bonds with Leu502 (Figure 5D). The nitro group of mitonafide was also found to be
engaged in electrostatic interaction with the adenine (AD12) nucleotide. Although, both the
15

ACCEPTED MANUSCRIPT
NISC-6 and mitonafide showed good binding affinity for topo-II, the binding conformation of
the former was observed to be more folded resulting in its tighter fitting into the binding groove
of the protein and in turn more stability of its complex (BE = -76.7 kcal mol-1) relative to the
latter. Furthermore, the NISC-6 displayed interactions with similar amino acids as observed in
case of etoposide (Arg503, DG13), and suggested that this compound could probably inhibit the
enzyme by same mechanism as etoposide. Overall, the docking results suggest that the aromatic
ring system present in NISC-6 is vital for locking its geometry in the binding site of both the
proteins.
Figure 5. NISC-6 docks into the binding pockets of both Akt and Topo-IIα. Docked pose of
NISC-6 (A) and ISC-4 (B) into the binding cavity of Akt1. Docked pose of NISC-6 (C) and
mitonafide (D) into the binding cavity of Topo-IIα. Ligand poses are shown in sticks (lemon
color), while the protein residues are shown in line ribbon (sky blue) format. Interacting amino
acids of protein as green lines, non-classical hydrogen bonds as blue dotted lines, electrostatic
interactions as black dotted lines and hydrophobic interactions are shown as magenta dotted
lines.
16

ACCEPTED MANUSCRIPT
2.3.3. NISC-6 inhibited in vitro Topo-II
α
activity and phosphorylation of Akt.
Based on structural design, NISC-6 possesses both the active site of ISC-4 and mitonafide, which
are the Akt pathway and Topo-IIα activity inhibitors, respectively. Our docking studies showed
that NISC-6 is able to fit in the active sites of both the proteins, more strongly than the parent
molecules. In order to prove the dual inhibitory property of NISC-6 experimentally, we first
tested the effect NISC-6 on Topo-IIα activity, in vitro. Topo-IIα enzyme, can convert supercoiled
DNA into relaxed DNA strands. Hence, using these principle, we tested NISC-6’s ability to
inhibit Topo IIα activity in vitro. Purified Topo IIα enzyme was incubated with super coiled
DNA in presence of DMSO, NISC-6, mitonafide or etoposide. Mitonafide and etoposide were
used as positive controls. As shown in figure 6A, super coiled DNA ran faster than relaxed DNA
strand or opened circular DNA, followed by relaxed DNA strand and Nicked open DNA. In the
presence of just the enzyme and supercoiled DNA, almost all the super coiled DNA was
converted into nicked open DNA. However, in presence of the positive control, increasing dose
of etoposide or mitonafide, the super coiled DNA is not converted in to nicked open DNA.
Similar results were observed for NISC-6, which showed very similar response as mitonafide at
inhibiting Topo-IIα activity.
17

ACCEPTED MANUSCRIPT
A
C
NISC-6
Etoposide
Mitonafide
NISC-6
2
0 2.5 5 µM
Compound
pHOT
Liner DNA
Topo II alpha
0
+
-
0
-
+
-
0
+
-
+
0.5
+
1
+
2
+
5
+
0.5
+
-
1
2
5
0.5
+
-
+
1
5 ( µM)
+
-
+
phospho-AKT
+
-
+
-
+
-
+
+
+
-
+
-
-
-
+
-
-
pan-AKT
-
+
+
+
+
+
+
+
NOC DNA
RL DNA
phospho-Erk1/2
Erk1/2
SC DNA
cleaved caspase-3
NOC : Nick opened circular DNA; RL: relaxed DNA; SC: supercoiled DNA
β-actin
Merged
p-H2A.X
F-Actin
DAPI
B
Figure 6. NISC-6 inhibited TopoIIα activity in vitro and also inhibits p-AKT levels in
melanoma cells. (A) NISC-6 inhibits Topo IIα from relaxing the supercoiled plasmid. In brief,
pBR322 plasmid DNA was incubated with Topo-IIα in presence of different drug concentration,
along with reaction mixture for 1 h at 37 °C. After incubation, the samples were subjected to gel
electrophoresis on 1% agarose gel. Gel was stained with ethidium bromide for visualizing the
DNA bands. (B) UACC903 cells were treated with NISC-6 at given concentration for 24 h.
Whole cell lysates were subjected to Western blot analysis. Resulting blots were probed with
phospho-Akt and pan-Akt. β-actin was used as loading control. (C) NISC-6 increase DNA
damage marker, p-H2A.X, expression in melanoma cells. UACC 903 cells were seeded on a
glass coverslip and treated with NISC-6 (5 µM) for 6 h. After treatment, cells were fixed and
probed with anti-body against p-H2A.X. DAPI (blue) was used to stain the nucleus and
phalloidin (red) was used to stain F-actin.
18

ACCEPTED MANUSCRIPT
It is very well established that phosphorylation of histone H2A.X is indicative of DNA
damage, particularly in case of DNA double strand breaks [44]. Further, Topo-II inhibitors are
known to increase p-H2AX expression [45]. Since, NISC-6 inhibited Topo-IIα activity, we
evaluated its ability to increase p-H2A.X expression in UACC 903 cells. Phospho-H2A.X (DNA
damage marker) expression increases upon DNA damage by external agents or by apoptosis (cell
death). However, DNA damage by Topo-IIα inhibitors is an early event compared to apoptosis
[46]. Therefore, we investigated the levels of p-H2A.X at early time point of 6 h rather than 24 h.
As shown in figure 6B, p-H2A.X expression increased when cells were incubated with 5 µM
dose of NISC-6, compared to the vehicle control. This finding further confirmed that NISC-6
induces DNA double strand breaks by inhibiting Topo-IIα activity.
To test the potential of NISC-6 to inhibit the phosphorylation of Akt, UACC903 cells
were treated with increasing dose of NISC-6 for 24 h and phosphorylation of Akt was observed
using Western blot. NISC-6 potently inhibited phosphorylation of Akt in a dose dependent
manner (Figure 6C). p-Erk was not affected showing specificity. Results obtained from both the
Topo IIα activity assay and Western blot analysis of p-Akt, along with our docking studies
strongly suggest that NISC-6 possesses both the Topo-IIα inhibitory activity and ability to inhibit
Akt pathway as per the structural design.
2.3.4. NISC-6 induced apoptotic cell death in melanoma cells.
Since both the Topo-IIα and Akt are well-known markers of resistance towards cancer cell death
and have been associated with poor prognosis in clinic [47, 48], we wanted to see whether the
dual inhibition of Topo-IIα and Akt pathway by NISC-6 translates to apoptotic cell death.
19

ACCEPTED MANUSCRIPT
Hallmark of apoptosis is the activation of caspase 3/7 and presence of phosphatidylserine (PS) on
the outer surface of the plasma membrane [49]. Based on our MTT data, we selected two doses
of NISC-6 to treat UACC903 cells for 24 h. Treated cells were tested for the presence of PS on
the outer leaflet of plasma membrane by annexin-V staining, and presence of caspase 3/7
activity. This was achieved by using Muse Live & Dead kit, and Caspase 3/7 activity kit. Both
the kits contain 7-AAD, which can only enter the cells when plasma membrane has lost its
integrity. Therefore, if the cells are dying in a necrotic manner, they will be stained only with 7-
AAD. Muse Live & Dead kit contains annexin-V dye which cannot enter intact cells and can
only bind to PS on the outer leaflet of the plasma membrane, while caspase 3/7 activity kit
contains a molecule which can travel freely inside the cells and fluoresce only when cleaved in
presence of active caspase 3/7. The fluorescence present in both kits can then be detected via
MUSE cell analyzer. Figure 7A-B shows the graphs obtained from Muse cell analyzer. The
bottom left corner of the graph represents healthy live cells (Annexin-V negative, Casapse 3/7
negative, 7-AAD negative), bottom right corner represents early apoptotic cells (Annexin-V
positive, Caspase 3/7 positive, 7-AAD negative), top right corner represents late apoptotic cells
(Annexin-V positive, Casapse 3/7 positive, 7-AAD positive) and top left corner represents
necrotic cells (Annexin-V negative, Caspase 3/7 negative, 7-AAD positive). As shown in figure
7A-B, increasing dose of NISC-6 translocated the cells from bottom left corner to bottom right
corner to top right corner in a dose dependent manner. There were no cells present in top left
corner at the doses tested, indicating that treatment with NISC-6 does not induce necrosis. It is
clear from figure 7A-B, that NISC-6 dose dependently increases apoptosis in melanoma cells.
Taken together, it can be concluded that NISC-6 induces cells death in an apoptotic manner.
20

ACCEPTED MANUSCRIPT
Figure 7. NISC-6 induced apoptotic cell death melanoma cells. Melanoma cells (UACC903) were
treated with NISC-6 for 24 h at given concentrations. At the end of the experiment, cells were subjected
to (A) Caspase 3/7 activity assay and (B) Annexin-V (Live and Dead) assay using a Muse cell analyzer.
Performing the assay as described in the methods section, generated a histogram as shown above. Bottom
left corner represents live cells (Caspase 3/7 (–), Annexin-V (–) and 7-AAD (–)), bottom right corner
represents early apoptotic cells (Caspase 3/7 (+), Annexin-V (+) and 7-AAD (–)), top right corner
represents late apoptotic cells (Caspase 3/7 (+), Annexin-V (+) and 7-AAD (+)) and top left corner
represents necrotic cells (Caspase 3/7 (–), Annexin-V (–) and 7-AAD (+)).
2.3.5. NISC-6 inhibited tumor growth in a melanoma xenograft mouse model.
21

ACCEPTED MANUSCRIPT
NISC-6 effectively inhibited both Topo-IIα activity as well as phosphorylation of Akt, and
induced apoptotic cancer cell death in melanoma cells. To determine if the in vitro efficacy of
NISC-6 translates to in vivo models, we performed tumor inhibition studies in xenograft mouse
model of melanoma. Athymic nude mice were subcutaneously injected with UACC 903 cells and
tumor development was allowed to occur for six days, by which time tumors undergo
vascularization (angiogenesis). Mice were then injected i.p with NISC-6 or control DMSO, three
times a week on Mondays, Wednesdays and Fridays at a dose of 0.6 µmole in DMSO (100 µL).
Tumor size was measured periodically and at the same time animal weights were also measured.
As shown in figure 8A, NISC-6 inhibited tumor growth and by the end of the experiment (40th
day) it was able to reduce tumor burden by ~69% as compared to vehicle treated mice (Mean
±SEM). These results complimented our in vitro cell viability and apoptosis assays data. Body
weight loss is often used as a measure of drug-induced toxicity. No evidence of systemic toxicity
was observed at the dose used as evidenced by the body weights; NISC-6 treated mice did not
show any significant weight loss as compared to the vehicle treated mice (Figure 8B). Hence,
tumor inhibition along with in vitro results strongly suggest the potential of NISC-6 as a novel
dual targeting therapy for melanoma.
22

ACCEPTED MANUSCRIPT
Figure 8. NISC-6 potently inhibited melanoma tumor growth in athymic nude mice (A) Six
days after subcutaneous injection of UACC 903 melanoma cells, nude mice were treated i.p.
with DMSO or NISC-6 (0.6 µmoles) thrice per week. NISC-6 significantly reduced tumor
growth (~69%) compared to DMSO control. (B) Body weights of treated mice compared to the
control DMSO vehicle treated mice showed no significant differences between groups. Data
represents the mean tumor volume or body weight with error bars representing SEM.
23

ACCEPTED MANUSCRIPT
2.3.6. Computation of physiochemical properties of NISC-6
NISC-6 promises to be an efficacious and safe drug candidate to develop further since it
effectively killed melanoma cells while being non-toxic to normal cells, and reduced tumor
growth in vivo without any apparent systemic toxicity. Our in vivo results further suggest that
NISC-6 must be bioavailable to reduce tumor growth so effectively. However, to access its drug
likeness, we computed its physiochemical and ADME (Absorption, Distribution, Metabolism
and Extraction) properties. The knowledge of physicochemical and ADME properties of
prospective drug candidates is crucially important in the drug design programs. The Lipinski’s
“rule of five” has widely been employed to filter the chemical entities, which have potential of
acting as drugs [50, 51]. According to this rule, any scaffold with the molecular weight (MW)
greater than 500, hydrogen bond donors (HBDs) more than five, hydrogen bond acceptors
(HBAs) more than 10 and an octanol-water partition coefficient (logP) value more than 5, is
likely to have poor permeation or absorption under physiological conditions. Therefore, the non-
compliance of any one of these rules may lead to the poor bioavailability of the compound of
interest. In order to check the drug-likeness of NISC-6, we subjected it to the analysis of Lipinski
(or Pfizer) rule of five. Computation was performed using the “Filter by Lipinski and Veber rule”
algorithm embedded in DS 4.0 Client. The predicted Lipinski’s properties of NISC-6, for
instance, MW; 385, HBDs; 0, HBAs; 4 and LogP; 3.2, exhibited excellent compliance with the
Lipinski’s rule of five, and favored the tendency of this compound to act as a drug
physiologically. Finally, the topological polar surface area (TPSA), which is another very useful
property to measure the hydrogen bonding capability of compounds, was computationally
predicted using the same tool in DS. Any compound with TPSA <140 Å2 is considered to have
24

ACCEPTED MANUSCRIPT
high oral availability [52, 53]. The computed TPSA of value 49.7 Å2 for NISC-6 also supported
its drug-likeness.
3. Conclusion
Novel naphthalimide analogs bearing isoselenocyanate (4a-d) and selenourea (7a-b)
functionalities have been developed using a fragment-based design. Both, NISC (4a-d) and NSU
(7a-b) analogs exhibited efficient inhibition of melanoma cell growth, NISC compounds showed
relatively better potency. Based on cell viability assays in BRAFV600E mutant and BRAFWT
human melanoma cells, NISC-6 was identified as the most effective compound that was toxic to
cancer cells while sparing the normal nHDF cells. As expected from its fragment-based design,
incorporating naphthalimide moiety of Topo-IIα inhibitor mitonafide and ISC moiety of Akt
pathway inhibitor ISC-4, NISC-6 was found to be a dual inhibitor of Topo-IIα and Akt pathway
as evidenced by both the docking studies and the experimental results. However, more in-depth
mechanistic studies would be required to establish whether these are the real targets of NISC-6
or are inhibited as a consequence of some upstream target(s). NISC-6 was effective in inducing
apoptosis in human melanoma cells and effectively inhibited subcutaneous melanoma tumor
growth (~69%) without any apparent systemic toxicity. NISC-6 thus seems to be a promising
drug candidate to develop further as a potential melanoma therapeutic.
4. Experimental
4.1. Chemistry
4.1.1. General
25

ACCEPTED MANUSCRIPT
Melting points were recorded on a Fischer-Johns melting point apparatus and are uncorrected.
1H-NMR and 13C-NMR spectra were recorded on a Bruker Advance 500 instrument in either
D2O or CDCl3, operating at 500 and 125 MHz, respectively. Chemical shifts are reported in
δ
values (ppm) and J values are reported in hertz (Hz). The signals are quoted as s (singlet), d
(doublet), t (triplet), q (Quartet), m (multiplet), and dd (doublet of doublet). High resolution (ESI
or EI) MS were carried out at the Chemistry Instrumentation Center, State University of New
York at Buffalo, NY. Experiments involving moisture and/or air sensitive components were
performed under a nitrogen atmosphere in oven-dried glassware. Reagents, starting materials,
and anhydrous solvents were purchased from commercial suppliers and were used as received.
Reaction courses were monitored by thin-layer chromatography (TLC) on pre-coated silica gel
60 F254 aluminum sheets (Merck, Darmstadt, Germany), and the spots were visualized under UV
light. The crude reaction products were purified by silica gel column chromatography using
silica gel 60 Ȧ (Merck, 230-400 mesh). The mitonafide was synthesized following a literature
method [41]. Naphthalimide linked alkylamines (2a-d) [40, 54], alcohols (5a,b) and bromide
(6a,b) [55] were synthesized using reported methods [15]. The purity of the final compounds
(≥97%) was quantified by analytical high performance liquid chromatography analysis by
comparing the peak areas of the product relative to any impurities.
4.1.2. General method for synthesis of napthalimide-alkyl formamides. Formamides (3a-d)
were synthesized following a literature method [56].
4.1.2.1. 2-(2-Formamido-ethyl)-benzo[de]isoquinoline-1,3-dione (3a).
Ethyl formate (3.1 g, 41.7 mmol) was added dropwise to a mixture of napthalimide alkylamine
2a (1.0 g, 4.2 mmol) in ethanol (10 mL) at room temperature and the resulting mixture was
o
heated at 60 C for 18 h. The solvent was removed under reduced pressure to yield the
26

ACCEPTED MANUSCRIPT
corresponding napthalimide-akylformamide 3a as a crude product which was purified by
crystallization from a mixture of CH2Cl2:hexanes (9:1) to yield 0.7 g (62.5%) as pale yellow
solid, m.p. 163–165oC. 1H NMR (500 MHz, DMSO-d6): δ 3.45 (m, 2 H, CH2NH), 4.17 (t, 2 H, J
= 6.0 Hz, CH2N), 7.87 (t, 2 H, J = 7.5 Hz, Ar-H5,H8), 7.94 (s, 1 H, CHO), 8.14 (br s, 1H, NH),
13
8.45 (d, 2 H, J = 8.0Hz, Ar-H6,H7), 8.49 (d, 2 H, J = 7.0 Hz, Ar-H4,H9); C NMR (125 MHz,
DMSO-d6) δ 164.1 (C=O), 161.7 (2 × C=O), 134.7, 134.6, 131.7, 131.16, 131.11, 127.9, 127.6,
122.6 (aryl), 39.2 (NCH2), 35.5 (CH2NH);HRMS (EI) m/z calcd for C15H12N2O3, 268.0842;
found: 268.08514.
4.1.2.2. 3-Nitro[2-(2-formamido-butyl)]-benzo[de]isoquinoline-1,3-dione (3b).
Ethyl formate (2.55 g, 34.4 mmol) was added dropwise to a mixture of napthalimide alkylamine
2b (1.0 g, 3.44 mmol) in ethanol (10 mL) at room temperature and the resulting mixture was
o
heated at 60 C for 18 h. The solvent was removed under reduced pressure to yield the
corresponding napthalimide-akylformamide 3b as a crude product which was purified by
crystallization from a mixture of CH2Cl2:hexanes (9:1) to yield 0.6 g (56.0%) as yellow solid,
m.p. 215–217oC. 1H NMR (500 MHz, DMSO-d6): δ 3.47 (m, 2 H, CH2NH), 4.19 (t, 2 H, J = 6.0
Hz, CH2N), 7.9 (1H, s, CHO), 8.07 (t, 1 H, J = 7.5 Hz, Ar-H8), 8.10 (br s, 1 H, NH), 8.70 (d, 1 H,
J = 7.0 Hz, Ar-H7), 8.78 (d, 1 H, J =7.5 Hz, Ar-H9), 8.97 (d, 1 H, J = 2.5 Hz, Ar-H4), 9.49 (d, 1
H, J = 2.0 Hz, Ar-H6); 13C NMR (125 MHz, DMSO-d6) δ 163.4 (C=O), 163.0 (C=O), 161.9
(C=O), 146.3, 136.7, 134.3, 131.3, 130.1, 130.0, 129.7, 124.5, 123.2, 123.1 (aryl), 39.5 (NCH2),
35.4 (CH2NH); HRMS (EI) m/z calcd for C15H11N3O5, 313.0693; found: 313.06946.
4.1.2.3. (2-Formamido-butyl)-benzo[de]isoquinoline-1,3-dione (3c).
27

ACCEPTED MANUSCRIPT
Ethyl formate excess (2.7 g, 37.3 mmol) was added dropwise to a mixture of napthalimide
alkylamine 2b (1 g, 3.7 mmol) in ethanol (10 mL) at room temperature and the resulting mixture
o
was heated at 60 C for 18 h. The solvent was removed under reduced pressure to yield the
corresponding napthalimide-akylformamide 3c as a crude product. This was purified by
crystallization from CH2Cl2/hexanes 9:1 to yield 0.75 g (68.8%) as pale brown solid, m.p. 160–
162oC. H1 NMR (500 MHz, DMSO-d6): δ 1.46-1.51 (m, 2 H, CH2), 1.62- 1.67 (m, 2 H, CH2),
4.27 (m, 2 H, , CH2NH), 4.05 (t, 2 H, J = 6.5 Hz, CH2N), 7.86 (t, 2 H, J = 6.5 Hz, Ar-H5,H8),
7.98-7.99 (m, 2 H, NH, CHO), 8.44 (dd, 2H, J = 8.0, 1.0 Hz, Ar-H6,H7), 8.49 (dd, 2 H, J = 7.5,
1.0 Hz, Ar-H4,H9); 13C NMR (125 MHz, DMSO-d6) δ 163.9 (C=O), 161.40 (C=O), 161.38
(C=O), 134.7, 131.7, 131.2, 127.8, 127.6, 122.5 (aryl), 41.0 (NCH2), 37.2 (CH2NH); 26.9 (CH2),
25.6 (CH2); HRMS (EI) m/z calcd for C17H16N2O3, 296.1155; found: 296.11569
4.1.2.4. 2-(2-Formamido-hexyl)-benzo[de]isoquinoline-1,3-dione (3d).
Ethyl formate excess (2.5 g, 33.7 mmol) was added to a mixture of napthalimide alkylamine 2d
(1.0 g, 3.4 mmol) in ethanol (10 mL) at room temperature and the resulting mixture was heated
o
at 60 C for 18 h. The solvent was removed under reduced pressure to yield the corresponding
napthalimide-akylformamide as a crude product. This was purified by column chromatography
from CH2Cl2/hexanes 7:3 to yield 0.6 g (54.5%) as pale yellow solid, m.p. 129–131oC. NMR
(500 MHz, DMSO-d6): δ 1.50-1.30 (m, 6 H, CH2CH2CH2), 1.65-1.60 (m, 2H, CH2), 3.10-3.05
(m, 2 H, CH2NH), 4.05 (t, 2 H, J = 7.5 Hz, CH2N), 7.85 (t, 2 H, J = 7.5 Hz, Ar-H5,H8), 8.00-7.95
(m, 2 H, NH and CHO), 8.43 (d, 2 H, J = 8.0 Hz, Ar-H6,H7), 8.48 (d, 2 H, J = 7.0 Hz, Ar-H4,H9);
13C NMR (125 MHz, DMSO-d6) δ 163.8 (C=O), 161.3 (2 × C=O), 134.7, 131.7, 131.1, 127.7,
28