Inhibition of thioredoxin reductase by a novel series of bis-1,2-benzisoselenazol-3(2H)-ones: Organoselenium compounds for cancer therapy

Article abstract of DOI:10.1016/j.bmc.2012.04.033

Thioredoxin reductase (TrxR) is critical for cellular redox regulation and is involved in tumor proliferation, apoptosis and metastasis. Its C-terminal redox-active center contains a cysteine (Cys497) and a unique selenocysteine (Sec498), which are exposed to solvent and easily accessible. Thus, it is becoming an important target for anticancer drugs. Selective inhibition of TrxR by 1,2-(bis-1,2-benzisoselenazol-3(2H)-one)ethane (4a) prevents proliferation of several cancer cell lines both in vivo and in vitro. Using the structure of 4a as a starting point, a series of novel bis-1,2-benzisoselenazol-3(2H)-ones was designed, prepared and tested to explore the structure-activity relationships (SARs) for this class of inhibitor and to improve their potency. Notably, 1,2-(5,5′-dimethoxybis(1,2-benzisoselenazol-3(2H)-one))ethane (12) was found to be more potent than 4a in both in vitro and in vivo evaluation. Its binding sites were confirmed by biotin-conjugated iodoacetamide assay and a SAR model was generated to guide further structural modification.

Full text of DOI:10.1016/j.bmc.2012.04.033

Bioorganic & Medicinal Chemistry 20 (2012) 3816–3827
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Inhibition of thioredoxin reductase by a novel series of bis-1,2-
benzisoselenazol-3(2H)-ones: Organoselenium compounds for cancer therapy
Jie He a,b, , Dongdong Li c, , Kun Xiong a,b, Yongjie Ge a,b, Hongwei Jin a, Guozhou Zhang a,b
,
Mengshi Hong a,b, Yongliang Tian a,b, Jin Yin a,b, Huihui Zeng a,b,c,
⇑
a State Key Laboratory of Natural and Biomimetic Drugs, Peking University, Beijing 100191, China
b School of Pharmaceutical Sciences, Peking University, Beijing 100191, China
c Tianjin International Joint Academy of Biotechnology and Medicine, Tianjin 300457, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Thioredoxin reductase (TrxR) is critical for cellular redox regulation and is involved in tumor prolifera-
tion, apoptosis and metastasis. Its C-terminal redox-active center contains a cysteine (Cys497) and a
unique selenocysteine (Sec498), which are exposed to solvent and easily accessible. Thus, it is becoming
an important target for anticancer drugs. Selective inhibition of TrxR by 1,2-(bis-1,2-benzisoselenazol-
3(2H)-one)ethane (4a) prevents proliferation of several cancer cell lines both in vivo and in vitro. Using
the structure of 4a as a starting point, a series of novel bis-1,2-benzisoselenazol-3(2H)-ones was
designed, prepared and tested to explore the structure–activity relationships (SARs) for this class of
inhibitor and to improve their potency. Notably, 1,2-(5,50-dimethoxybis(1,2-benzisoselenazol-3(2H)-
one))ethane (12) was found to be more potent than 4a in both in vitro and in vivo evaluation. Its binding
sites were confirmed by biotin-conjugated iodoacetamide assay and a SAR model was generated to guide
further structural modification.
Received 16 March 2012
Revised 16 April 2012
Accepted 17 April 2012
Available online 23 April 2012
Keywords:
Organoselenium compound
Thioredoxin reductase
Anticancer drug
Structure–activity relationship
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Mammalian thioredoxin reductase (TrxR), together with
thioredoxin (Trx) and NADPH, constitutes a critical system for many
cellular pathways.1 Thioredoxins (from mammal to bacteria) are
small ubiquitous proteins containing a conserved pair of cysteine
residues (–CGPC–) and work as disulfide reductases maintaining
cytosolic proteins in their redox state.1,2 Mammalian TrxRs are
large selenoproteins (114 kDa or larger) with conserved C-terminal
sequences (–GCUG–) acting as the catalytically active sites.3,4 The
major biological function of TrxR is to reduce oxidized Trx to
maintain the cellular antioxidant system.1,4 It has been reported,5
however, that the Trx system is involved in many aspects of tumor
physiology such as proliferation, apoptosis and metastasis.
Researchers6,7 have also found that the level of TrxR in many tumor
cell lines is often more than 10-fold greater than the levels observed
in normal cells. Meanwhile, the C-terminal redox-active center of
TrxR is protruding away from the enzyme and hence easily
accessible for a wide range of substrates.8 Thus, targeting of the
Figure 1. Structure of ethaselen (4a).
C-terminal active site has become a promising concept for antican-
cer drug development.7,9 Many small molecules, such as metal
based complexes, curcumin and its analogues, arsenic trioxide
and quinines have been reported10–16 as potent TrxR inhibitors.
However, up to the end of 2011, only three anticancer drugs tar-
geting Trx system were under clinical trials: (1) Motexafin gadolin-
ium17 (MGd), phase III; (2) PX-1218 (1-methylpropyl 2-imidazolyl
disulfide), phase II; (3) Ethaselen (1,2-(bis-1,2-benzisoselenazol-
3(2H)-one)ethane), also named BBSKE (1,2-[bis(1,2-benzisosele-
nazolone-3(2H)-ketone)]ethane) (4a, Fig. 1), phase I. In particular,
MGd is a non-competitive inhibitor of rat TrxR (IC50 = 6 lM) and in-
duces enzymatic generation of reactive oxygen species (ROS) by
TrxR. It also inhibits ribonucleotide reductase (RNR) to exhibit anti-
cancer activity.19 PX-12 acts as an irreversible inhibitor of cytosolic
form thioredoxin (Trx-1) and can also reduce the level of several
other plasma proteins.20 Whereas, ethaselen (4a) which is a
organoselenium drugs developed by our group, was previously
Abbreviations: TrxR, thioredoxin reductase; Trx, thioredoxin; GR, glutathione
reductase; SAR, structure–activity relationship; DTNB, 5,50-dithiol-bis-2-nitroben-
zoic acid; LogPcal, calculated LogP; BIAM, biotin-conjugated iodoacetamide.
⇑
Corresponding author. Tel.: +86 10 82802878; fax: +86 10 62015584.
E-mail address: zenghh@bjmu.edu.cn (H. Zeng).
demonstrated21–24 to be
(IC50 = 0.35 M) that targeting both the Cys497 and Sec498 active
a potent inhibitor of TrxR in rats
l
These authors contributed equally to the work.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.04.033

J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
3817
sites of TrxR and showing remarkable anticancer activity in both
in vitro and in vivo experiments. Further reviews9,25 for this drug
are also available. Compared to MGd, ethaselen showed much
stronger inhibitory effects on TrxR, whilst they both had no obvious
interaction with Trx.26 In addition, ethaselen can be easily prepared
in high yield from readily available starting materials (see Section
4.2). In light of these reports, the novel selenium-containing struc-
ture 4a is considered to have unique biological properties, although
structural modification and structure-activity relationships (SARs)
study would be necessary to develop promising candidates for fu-
ture drug discovery.
centers inhibiting both the Cys497 and Sec498 residues of the
enzyme. To develop potent analogues of 4a with favorable
biological properties and to explore the SAR of bis-1,2-ben-
zisoselenazol-3(2H)-ones as TrxR inhibitors, a series of novel
substituted bis-1,2-benzisoselenazol-3(2H)-ones was designed
and synthesized. With the exception of compounds 12–16, the tar-
get compounds were prepared using a simple three-step method
(Scheme 1). First, sodium selenide was reacted with the corre-
sponding benzenediazonium salts, themselves prepared from
substituted 2-amino benzoic acids, to give 2,20-diselenobisbenzoic
acids. Subsequent heating under reflux with SOCl2 provided the
2-chloroselenobenzoyl chlorides, which were added to the corre-
sponding diamines to give the final products.
Herein, using the structure of 4a as starting point, we describe
the design and synthesis of
a novel series of bis-1,2-ben-
zisoselenazol-3(2H)-ones with the purpose of exploring any SARs.
In order to elucidate the relationship between the enzyme inhibi-
tory effects and cell growth inhibition activities of the active com-
pounds and to further confirm their binding sites at the enzyme,
several biological evaluations were employed to assess the proper-
ties of target compounds. Docking and pharmacological studies of
representative compound (12) were also carried out.
For 12 and 13, the hydroxyl group of compound 2i was methyl-
ated and acetylated, respectively, prior to the chlorination proce-
dure (Scheme 2). Compound 2j (intermediate of 12) was
prepared by a two-step procedure from 2i. Briefly, treatment of
2i with K2CO3 and CH3I afforded dimethyl 5,50-dimethoxy-2,20-
diselenobisbenzoate, which was used without purification and
subsequently hydrolyzed to yield 2j. Compound 2k (intermediate
of 13) was synthesized by treating 2i with acetic anhydride in pyr-
idine, whereas compound 14 was prepared from 13 by hydrolysis
of the acetoxy group. Methods for the preparation of 15 and 16
are described in Schemes S1 and S2 (see Supplementary data for
details).
2. Results and discussion
2.1. Chemistry
Although many studies of 4a concerning its biological and phar-
maceutical properties have been published, this is our first time to
report the synthesis of 4a and its analogues. The core bis-1,2-ben-
zisoselenazol-3(2H)-one structure was retained and attention was
focused predominantly on the introduction of substituents on the
aromatic rings, because 4a was designed to contain two active
2.2. Inhibition of TrxR in vitro
The inhibitory effects of all the target compounds were exam-
ined using a 5,50-dithiol-bis-2-nitrobenzoic acid (DTNB) reduction
assay.8,27 All compounds tested inhibited the DTNB-reducing
Scheme 1. Synthesis of compounds 2–11. Reagents: (a) H2O, NaNO2, HCl, Na2Se2; (b) SOCl2, DMFcat; (c) NH2-R3-NH2, triethylamine, CH3CN or CH2Cl2.

3818
J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
Scheme 2. Synthesis of compounds 12, 13 and 14. Reagents: (a) DMF, K2CO3, CH3I, then H2O, NaOH; (b) pyridine, CH3COOCOCH3, then H2O; (c) SOCl2, DMFcat; (d) CH2Cl2,
NH2CH2CH2NH2, triethylamine; (e) DMSO, HCl.
activity of TrxR (TrxR1 of rat liver) in a dose-dependent manner.
Their IC50 values ranged from 0.1 to more than 50 M, as shown
in Table 1. Several compounds (4b, 6a, 6b, 9, 10a, 12 and 14)
showed good inhibitory effects (IC50 <1 M) which were compara-
inhibitory activities of TrxR in vitro (IC50 <1 lM), such as 6a, 12
l
and 14, generally showed comparable activities of cell growth inhi-
bition with regard to 4a. This suggests a possible link between TrxR
inhibition and cell growth inhibition of the tested compounds (see
Section 2.4). For compound 12, improvement of growth inhibition
activity in all tested cell lines can be observed in comparison to 4a.
However, compounds containing chloro or bromo substituents (7a,
8a and 11a) exhibited very poor inhibitory effects on cancer cell
l
ble to or even better (6a, 12, 14) than that of 4a. Some SAR infor-
mation could be derived from these data. It is clear that the
enzyme inhibitory activities of the compounds tested were re-
duced significantly when the selenium atom was replaced by
either an oxygen (15) or sulfur (16) atom. This may result from
the Se–N bond being a more electrophilic site for attack from the
selenolthiols and thiols of proteins.28 Furthermore, published re-
search29 has shown that benzisoselenazolone exhibits a much
higher level of reactivity than its benzisothiazolone analogue. In
addition, in order to obtain good level of activity, the R3 group is
better to be an alkyl chain because replacement of the alkyl chains
(4a–g, 5a–e and 6a–d) with either cyclohexyl (4h, 5f and 6e) or
aryl (4i and 4j) groups led a significant decrease in the enzyme
inhibitory activities. In contrast, among the compounds containing
alkyl chains mentioned above, a slight trend of decreased activities
can be observed when the number of carbons increased from 2 to
more than 5 (except 8a, b). Therefore, the placement of short alkyl
chains at R3 position was preferred for this series of compounds.
Substituents at the R1 and R2 positions also influenced the bio-
logical activities of the compounds tested. Compound 12 (methox-
lines (IC50 >50
lM), whereas their fluoro analogue (6a and 10a)
were much more active (IC50 between 3 and 13
lM in all tested cell
lines, Table 2). Subsequent calculation of the LogP values for all of
the compounds, revealed that compounds with halogen atoms (ex-
cept for fluoro) showed higher LogPcal with the LogPcal values of 7a,
8a and 11a being 4.45, 5.71 and 4.49, respectively. In contrast, the
LogPcal of all the other compounds (tested in cell growth inhibi-
tion) were less than 4. Precipitation of small amounts of com-
pounds (7a, 8a and 11a) was also observed when they were
incubated with cells in medium for 48 h during the MTT assay. It
was postulated that the poor inhibitory activities on cell growth
of the chloro or bromo containing compounds were caused by their
high lipophilicity (large LogP) and high molecular weight and that
these properties made them less active in vivo than in vitro be-
cause limited solubility and large molecular weight hindered their
cellular permeability.
yl in R1) showed the best inhibitory effect (IC50 = 0.13
whereas compounds 6a (fluoro in R1) and 14 (hydroxyl in R1) were
less active (IC50 = 0.28 M and 0.32 M, respectively) but still bet-
lM),
2.4. Inhibition of cellular TrxR
l
l
ter than 4a. Among compounds with halogen atom substituents,
fluoro containing ones (e.g. 6a and 10a) were more active than
their chloro and bromo analogues (e.g. 7a, 8a and 11a), although
To substantiate the assumption made above and to clarify the
possible link between TrxR inhibition and cell growth inhibition,
U87 MG cell line was selected to investigate the inhibitory effects
of five representative compounds (12, 4a, 5a, 8a and 9) on the
they all showed good inhibitory activities (IC50 <3 lM).
activity of cellular TrxR. Exposure of U87 MG cells to 1–20 lM
2.3. Growth inhibition of cancer cells
solutions of different compounds showed that cellular TrxR activ-
ity was attenuated to varying degrees ( Fig. 2A). Encouragingly,
compound 12 showed great inhibitory effect on cellular TrxR with
In order to identify novel anticancer agents, the inhibitory
effects of selected compounds on the growth of human cancer cell
lines, including U87 MG (glioma), MIA PaCa-2 (pancreatic carci-
noma) and LoVo (colon cancer), were examined. The results are
summarized in Table 2. As expected, compounds with good
an IC50 value of approximately 2
fect of 4a (IC50 ꢀ5 M). TrxR activity in U87 MG cell extracts was
inhibited by more than 90% when the concentration of 12 reached
20 M. In fact, compounds 4a, 5a, 8a and 9 exhibited dose-depen-
lM, which was better than the ef-
l
l

J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
3819
Table 1
In vitro TrxR inhibitory activities of the novel series of bis-1,2-benzisoselenazol-3(2H)-onesa
Compound
R1
R2
R3
IC50
(l
M)
4a
4b
4c
4d
4e
4f
4g
4h
4i
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
F
CH2CH2
CH2CH2CH2
0.35b
0.78 0.03
1.03 0.05
1.95 0.15
1.66 0.29
5.85 0.57
4.49 0.76
12.79 1.21
47.68 10.28
34.30 7.45
1.26 0.31
1.16 0.21
2.38 0.03
3.16 0.05
4.91 0.31
>50
0.28 0.04
0.86 0.03
1.64 0.19
3.87 1.02
10.49 2.14
1.31 0.11
1.80 0.74
1.58 0.52
1.59 0.29
0.57 0.01
0.82 0.11
2.91 0.80
2.87 0.51
4.54 0.91
0.13 0.02
22.60 4.05
0.32 0.02
CH2CH2CH2CH2
CH2CH2CH2CH2CH2
CH2CH2CH2CH2CH2CH2
CH2CH2CH2CH2CH2CH2CH2
CH2CH2CH2CH2CH2CH2CH2CH2
Cyclohexane
PhCH2
Ph
CH2CH2
CH2CH2CH2
CH2CH2CH2CH2
CH2CH2CH2CH2CH2
CH2CH2CH2CH2CH2CH2
Cyclohexane
CH2CH2
H
H
4j
5a
5b
5c
5d
5e
5f
6a
6b
6c
6d
6e
7a
7b
8a
8b
9
10a
10b
11a
11b
12
13
14
CH3
CH3
CH3
CH3
CH3
CH3
F
F
F
F
F
Cl
Cl
Br
Br
NO2
H
H
H
H
CH2CH2CH2
CH2CH2CH2CH2
CH2CH2CH2CH2CH2
Cyclohexane
CH2CH2
CH2CH2CH2CH2CH2
CH2CH2
CH2CH2CH2CH2CH2
CH2CH2
CH2CH2
CH2CH2CH2CH2CH2
CH2CH2
CH2CH2CH2CH2CH2
CH2CH2
CH2CH2
CH2CH2
F
Cl
Cl
H
H
H
OCH3
OCOCH3
OH
15
16
O
S
CH2CH2
CH2CH2
>100
48.09 1.91
a
Values are means and SD of at least two experiments.
Cite from our previous study (see Ref. 22).
b
dent inhibition of the target. Interestingly, comparison of these re-
sults with the cell growth inhibition activities of the selected com-
pounds ( Fig. 2B) revealed good correlation (Table 3) and
demonstrated that compounds with higher cell growth inhibition
activities ultimately displayed better percentile inhibitions of
intracellular TrxR and vice versa. This result, combined with obser-
vations from our previous study21 of 4a, indicated that the cell
growth inhibitory activities of this series of bis-1,2-ben-
zisoselenazol-3(2H)-ones could be attributed to their inhibitory ef-
fects of intracellular TrxR. Again, 8a showed weak inhibitory
activity consolidating our assumptions, whereas compound 9,
Table 2
Growth inhibition activities of selected compoundsa.
Compound
IC50 (lM) of cell lines
Mia-paca2
LoVo
U87
4a
4d
4h
4i
3.79 0.19
3.27 0.20
11.50 0.74
>50
7.67 0.23
5.50 0.96
17.22 0.48
>50
6.41 0.30
6.05 0.67
>50
ND
>50
17.77 0.09
12.25 2.39
6.95 0.41
>50
>50
>50
12.60 1.25
>50
1.64 0.25
20.76 2.78
6.60 1.11
4j
>50
ND
5a
5d
6a
7a
8a
9
10a
11a
12
13
14
5.94 0.89
6.71 0.92
3.07 0.45
>50
6.12 0.63
16.56 0.21
4.55 0.13
NDb
which was a potent TrxR inhibitor in vitro (IC50 = 0.57 lM), exhib-
ited very poor activity in both cell growth inhibition and TrxR inhi-
bition in vivo, in spite of a low LogP (LogPcal = 2.98) and low
molecular weight in comparison to 8a. The reason for this abnor-
mal property of 9 would require further study.
>50
ND
>50
ND
8.47 0.57
>50
ND
ND
2.84 0.30
10.23 1.23
5.30 0.10
5.28 0.54
ND
6.53 1.77
2.5. Detection of binding sites
a
All values are shown in IC50 (lM) values and expressed as means SD of trip-
According to our initial design, it was assumed that compound
12 would interact with the enzyme at both the Sec498 and Cys497
licate experiments.
b
ND = No data.

3820
J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
Figure 3. Binding sites detection of 12 at TrxR. Different concentrations of 12 were
added to reduced TrxR (1.1 M) and incubated at room temperature for 2 h. The
same amounts of DMSO were added to the control experiments. Lanes 1–3, pH 8.5;
Lanes 4–6, pH 6.5; Lanes 2, 5, 10 M; Lanes 1, 6, 20 M; Lanes 3, 4, control.
l
l
l
thiols.28 Subsequent reaction of the other isoselenazolone ring
with the neighboring –SH group of Cys497 leads to a double-block-
ing of the redox-active site of TrxR. This hypothesis has been sup-
ported either by glutathione reductase (GR) inhibition assays in
which only minor inhibition was observed for the compounds
tested (see Supplementary data, Fig. S1) or by our previous study26
in which 4a exhibited poor or no inhibitory effects of Sec498 mu-
tant (Sec to Cys) or C-terminal truncated human TrxR.
2.6. Docking study
To better understand the potency of compound 12 and to guide
further SAR study, we carried out docking study of it with crystal-
lographic structure of TrxR. There was evidence34 showing that
small molecules could bind to TrxR in the cavity between the C-
terminal and N-terminal active sites. For the current docking study,
the crystallographic structure of human thioredoxin reductase I
(PDB code: 2J3N) was selected and the conformation of the enzyme
used for the experiment was presumed to be the favored and sol-
vent stable form.34 A docking sphere in the cavity between the two
redox-active centers (Cys59, Cys 64 at the N-terminal and Cys497,
Sec498 at the C-terminal) was created. As illustrated in Figure 4,
following the hydrogen bond formation between one of its car-
bonyl group and hydrogen of His472, one unit of benzisoselenazo-
lone heterocycle of 12 inserted into the presumed ‘pocket’
consisting of His108, Trp407, Pro408 and Leu409 while another
unit of benzisoselenazolone formed a hydrogen bond with Lys29.
Furthermore, hydrogen bonds between the two methoxy groups
of 12 and Lys123 and Trp407 were also observed. It was demon-
strated34,36,37 that His472, Tyr116, Trp407 and Lys29 played a very
important role for catalysis of TrxR by means of facilitating its
reduction. This result, along with biological assays date, suggests
that compound 12 is a potential inhibitor of TrxR.
Figure 2. Comparison between cellular TrxR inhibitory effects and cell growth
inhibitory activities of selected compounds (12, 4a, 5a, 8a and 9). (A) Dose-
dependent curves of the inhibitory effects of tested compounds of cellular TrxR in
U87 MG cell lines. (B) Dose-dependent curves of U87 MG cell growth inhibitory
activities of tested compounds. Symbols were used for 12 (ꢀ), 4a (h), 5a (4), 8a (d)
and 9 (N). See Section 2.4 for discussion.
Table 3
The correlation coefficient between cell growth inhibi-
tion and TrxR inhibition of the compounds tested
Compounds
Correlation coefficient (r)a
12
4a
5a
8a
9
0.98
0.96
0.85
0.92
0.89
a
The values of r were calculated and represented by
pearson product-moment correlation coefficient.
2.7. Pharmacological properties of 12
residues of the C-terminal active sites of TrxR. In order to confirm
its binding sites, a biotin-conjugated iodoacetamide (BIAM) assay
was conducted. It was reported30 that BIAM can react with TrxR
selectively by adjusting the pH value. At a pH of 6.5 only the
Sec498 residue was alkylated by BIAM, whereas at a pH of 8.5, both
the Sec498 and Cys497 residues were alkylated by BIAM. In the
current study, the blotting result confirmed that 12 can react with
both sites of the C-terminal redox-active center because the bands
of the experimental groups were much weaker than the control
groups at both pH 8.5 and 6.5 (Fig. 3). Moreover, 12 blocked both
active sites of TrxR at either pH (8.5 or 6.5) in a dose-dependent
manner. Under physiological conditions, the pKa value of the –
SeH group in Sec is 6.531 whereas that of the –SH in Cys is 8.5.32
It is clear that the selenocysteine is more reactive than cysteine un-
der physiological conditions because the former is present as the
predominant form of selenide. Therefore, it was speculated that
12 inhibits TrxR in a two-step mechanism which is similar to that
of other reported TrxR inhibitors.8,33 First, the ionized –SeH group
of Sec498 acts as a nucleophilic agent and attack the selenium
atom of a single isoselenazolone ring in 12 because the Se–N bond
of isoselenazolones reacts in a facile manner with selenothiols and
Pharmacological studies revealed that 12 inhibited TrxR in a
rapid, mixed-type and reversible mechanism (Fig. 5). Michaelis–
Menten analysis ( Fig. 5A) and Lineweaver–Burk plot ( Fig. 5B)
suggested that 12 acts as a mixed-type inhibitor of TrxR with re-
gard to DTNB, as opposed to a pure competitive inhibitor. This
can be explained by the proposed double-blocking inhibition in
which both Cys497 and Sec498 are modified covalently by 12.
When the concentration of inhibitor increases, the Michaelis–Men-
ten Constant (Km) increases (the increasing x-intercept in Fig. 5B)
as a higher concentration of DTNB is required to reach the Km, be-
cause the inhibitor and DTNB react with the same active sits of
TrxR. This type of inhibition, however, cannot be overcome by high
concentrations of DTNB as covalent bonds are formed, so the Max-
imal Rate of Reaction (Vmax) decreases (increasing y-intercept in
Fig. 5B). Furthermore, 12 inhibits TrxR in a rapid but time-depen-
dent manner ( Fig. 5C). However, the overall reaction is largely
reversible by ultrafiltration assay (Fig. 5D). This may due to the
recyclization of the benzisoselenazolone ring or cleavage of
the Se–Se bond between the selenium atoms of the inhibitor and
the Sec498 residue by the sulfur atom of Cys497, before an

J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
3821
Figure 6. SAR summary for bis-1,2-benzisoselenazol-3(2H)-ones as TrxR inhibitors.
docking studies described above. Briefly, selenium is essential to
the biological activity of this series of compounds and the linker
between the two benzisoselenazolone heterocycles should be a
short alkyl group. Furthermore, small substituents on the benzene
ring with the potential for hydrogen bond formation are preferred
in order to reduce the molecular lipophilicity. As TrxR receives
increasing attention as drug target, especially for anticancer ther-
apy, these SARs and docking studies may provide models for the
future development of organoselenium compounds as TrxR inhib-
itors. Encouragingly, the discovery of 12 substantiates the asser-
tion that structural modification of 4a was necessary and feasible
to improve the biological activity of this series of compounds. This
study also confirms that both Cys497 and Sec498 are binding
targets for 12. In addition, the enzyme inhibitory effects of the
compounds tested demonstrates the close relationship with their
anticancer activities according to our comparison between the
Figure 4. Representation of the putative binding modes of compound 12 (cyan
carbon) in the ‘pocket’ between the two redox-active centers of TrxR. The ligand
and Residues closed to it are displayed as sticks. Hydrogens are hidden for clarity.
Standard atom color codes: carbons in residues, green; oxygen, pink; nitrogen, blue;
sulfur, yellow; selenium, orange. Hydrogen bonds between ligands and residues are
symbolized by grey dash lines.
intermolecular bond between Cys497 and another selenium atom
of the inhibitor can be formed. As a result, therefore, the inhibitor
can be released from the active sites.
3. Conclusion
In Figure 6, the SARs for bis-1,2-benzisoselenazol-3(2H)-ones as
TrxR inhibitors is summarized based on the biological assays and
Figure 5. Characterization of TrxR inhibition by 12 (A) Michaelis–Menten analysis of DTNB reduction assay, control (no inhibitor, j); 50 nM (d); 200 nM (N) and 500 nM (.);
concentrations of DTNB were set as 25, 50, 100 and 500
M. (B) Mixd-type inhibition of TrxR was determined by Lineweaver-Burk plots, control (no inhibitor, }); 50 nM (4);
200 nM (s) and 500 nM (h). (C) TrxR is inhibited in a time-dependent manner; concentrations of DTNB were set as 25, 50, 100 and 500 M. (D) Reversibility of TrxR
inhibition was tested by ultrafiltration assay, black, before ultrafiltration; grey, after ultrafiltration. ⁄P <0.005 vs before ultrafiltration (0.5 M); ⁄⁄P <0.001 vs before
ultrafiltration (1 M). Data was obtained by triplicate experiments and presented as mean SD.
l
l
l
l

3822
J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
results of MTT assays and cellular TrxR inhibition assays. Although
the exact molecular mechanism of TrxR inhibition by these com-
pounds requires further clarification, given the complex kinetic
behavior exhibited by 12 during its reaction with the enzyme,
the development of this novel series of bis-1,2-benzisoselenazol-
3(2H)-ones offers a solid basis for designing potent organosele-
nium compounds as anticancer drugs by targeting TrxR.
4.2.1.4. 5,50-dichloro-2,20-diselenobisbenzoic acid (2d).
Com-
pound 1d (17.1 g, 0.1 mol) was used as reactant to give 20.0 g of 2d
1
as yellow solid (85%). H NMR (400 MHz, DMSO-d6) d 14.12 (br, 2H,
COOH), 8.00 (d, J = 2.2, 2H, ArH), 7.64 (d, J = 8.6, 2H, ArH), 7.60 (dd,
J1 = 8.6, J2 = 2.2, 2H, ArH). 13C NMR (100 MHz, DMSO-d6) d 167.92,
144.59, 133.82, 132.55, 132.04, 131.26, 130.89. MS (EI) m/z: 469.8
(M+).
4.2.1.5. 5,50-dibromo-2,20-diselenobisbenzoic acid (2e).
Com-
4. Experimental
pound 1e (21.5 g, 0.1 mol) was used as reactant to give 21.5 g of 2e
1
as yellow solid (77%). H NMR (400 MHz, DMSO-d6) d 14.06 (br, 2H,
4.1. Chemicals and methods
COOH), 8.10 (d, J = 2.3, 2H, ArH), 7.69 (dd, J1 = 8.6, J2 = 2.3, 2H, ArH),
7.55 (d, J = 8.6, 2H, ArH). 13C NMR (100 MHz, DMSO-d6) d 171.55,
145.11, 138.92, 136.65, 130.36, 128.77, 120.31. MS (EI) m/z: 557.7
(M+).
All chemical reagents were commercially available (purchased
from Sigma–Aldrich, Alfa Aesar and TCI in high quality) and were
used without further purification unless otherwise stated. SOCl2
was distilled prior to use. 1H and 13C NMR spectra were obtained
on a Bruker Avance III instrument, 400 MHz for 1H spectra and
100 MHz for 13C spectra, in DMSO-d6 or CDCl3. Mass spectra were
run on an APEX IV FT-MS (7.0T) instrument. The purity of all com-
pounds tested was determined by HPLC analysis (Agilent 1100,
Agilent Technologies, Inc., USA) with a flow rate of 1 ml/min at
4.2.1.6. 5,50-dinitro-2,20-diselenobisbenzoic acid (2f).
Com-
pound 1f (18.2 g, 0.1 mol) was used as reactant to give 17.0 g of
2f as brown solid (69%). 1H NMR (400 MHz, DMSO-d6) d 13.40
(br, 2H, COOH), 8.58 (d, J = 2.3, 2H, ArH), 8.07 (dd, J1 = 8.8,
J2 = 2.3, 2H, ArH), 7.91 (d, J = 8.8, 2H, ArH). 13C NMR (100 MHz,
DMSO-d6)
d 168.90, 151.5, 149.60, 132.03, 130.95, 128.76,
25 °C (Ultimate XB-C18 column, 5
l
m, 4.6 ꢁ 250 nm, Welch Mate-
125.42. MS (EI) m/z: 461.9 (M+).
rials, Inc.). The samples were eluted with Milli-Q water containing
0.1% phosphate and methanol, using UV monitor at 254 nm for
detection. Retention times (tR) were calculated in minutes, and
purity was calculated as % total area.
4.2.1.7. 4,40-difluoro-2,20-diselenobisbenzoic acid (2g).
Com-
pound 1g (15.5 g, 0.1 mol) was used as reactant to give 17.1 g of 2g
as yellow solid (78%). 1H NMR (400 MHz, DMSO-d6) d 13.92 (br, 2H,
COOH), 8.12 (dd, J1 = 8.4, J2 = 6.0, 2H, ArH), 7.40 (dd, J1 = 9.8, J2 = 2.3,
2H, ArH), 7.23 (td, J1 = 8.4, J2 = 2.3, 2H, ArH). 13C NMR (100 MHz,
DMSO-d6) d 167.76, 162.52 (d, JC–F = 250.5), 136.64 (d, JC–F = 8.5),
134.46 (d, JC–F = 20.8), 125.77 (d, JC–F = 9.3), 116.09 (d, JC–F = 3.5),
114.05 (d, JC–F = 25.0). MS (EI) m/z: 438.0 (M+).
4.2. Synthesis
4.2.1. General procedure for the synthesis of 2(a–i)
2-Amino benzoic acid (0.1 mol), HCl (20 ml) and H2O (20 ml)
were added to a beaker and the mixture was cooled at 0 °C. An
aqueous solution of sodium nitrite (0.11 mol) (30%) was added
slowly added and the reaction mixture was stirred for 30 min.
The mixture was then poured into an aqueous solution of disodium
diselenide (0.05 mol) and the resulting solution was stirred at
room temperature for another 3 h before the pH was adjusted with
aqueous HCl (10%) to a value of 2–3. Precipitation occurred and the
final products were obtained by filtration.
4.2.1.8. 4,40-dichloro-2,20-diselenobisbenzoic acid (2h).
Com-
pound 1h (17.1 g, 0.1 mol) was used as reactant to give 20.7 g of
2h as yellow solid (88%). 1H NMR (400 MHz, DMSO-d6) d 14.08 (br,
2H, COOH), 8.03 (d, J = 8.3, 2H, ArH), 7.60 (d, J = 1.7, 2H, ArH), 7.44
(dd, J1 = 8.3, J2 = 1.7, 2H, ArH). 13C NMR (100 MHz, DMSO-d6) d
171.56, 150.02, 137.26, 137.07, 133.18, 129.34, 125.58. MS (EI) m/
z: 469.9 (M+).
4.2.1.1. 2,20-diselenobisbenzoic acid (2a).
Compound 1a
4.2.1.9. 5,50-dihydroxy-2,20-diselenobisbenzoic acid (2i).
Com-
(13.7 g, 0.1 mol) was used as reactant to give 19.6 g of 2a as yellow
solid (95%). 1H NMR (400 MHz, DMSO-d6) d 13.67 (br, 2H, COOH),
8.00 (d, J = 7.6 Hz, 2H, ArH), 7.66 (d, J = 7.6 Hz, 2H, ArH), 7.45 (t,
J = 7.6 Hz, 2H, ArH), 7.33 (t, J = 7.6 Hz, 2H, ArH). 13C NMR
(100 MHz, DMSO-d6) d 168.62, 133.43, 133.15, 131.41, 129.65,
129.43, 126.32. MS (EI) m/z: 401.9 (M+).
pound 1i (15.3 g, 0.1 mol) was used as reactant to give 14.5 g of 2i
as brown solid (67%). 1H NMR (400 MHz, DMSO-d6) d 13.16 (br, 2H,
COOH), 9.82 (s, 2H, OH), 7.44 (d, J = 2.8, 2H, ArH), 7.42 (d, J = 8.7, 2H,
ArH), 6.93 (dd, J1 = 8.7, J2 = 2.8, 2H, ArH). 13C NMR (100 MHz, DMSO-
d6) d 168.39, 156.24, 130.75, 129.28, 121.34, 121.18, 117.71. MS (EI)
m/z: 433.9 (M+).
4.2.1.2. 5,50-dimethyl-2,20-diselenobisbenzoic acid (2b).
Com-
4.2.1.10. 5,50-dimethoxy-2,20-diselenobisbenzoic acid (2j).
The
pound 1b (15.1 g, 0.1 mol) was used as reactant to give 20.3 g of 2b
1
preparation of 2j was a two-step process. Step 1: 2i (4.3 g, 0.01 mol)
was dissolved in 100 ml DMF containing 6.5 g K2CO3, to which
7.2 ml CH3I was added dropwise under stirring and the mixture was
stirred for 14 h at 50 °C. Then the mixture was poured into 200 ml
water and this solution was extracted by diethyl ether (5 ꢁ 50 ml)
and washed by 0.1 M NaOH (3 ꢁ 20 ml) and water (50 ml). The ether
extract was dried by Na2SO4 followed by the removal of solvent under
vacuum to give 3 g yellow powder without further purification. Step 2:
the yellow powder was dissolved in 50 ml methanol and 80 ml water
containing 2 g NaOH. The reaction mixture was stirred under reflux
and the solution became transparent after 1 h. The solution was stirred
for another 1 h. After that, HCl was added to adjust the pH to 2. The
mixture was filtrated to give 2.3 g of 2j. Yellow solid (50%). 1H NMR
(400 MHz, DMSO-d6) d 12.13 (br, 2H, COOH), 7.60 (d, J = 8.8, 2H,
as yellow solid (90%). H NMR (400 MHz, DMSO-d6) d 13.55 (br, 2H,
COOH), 7.84 (d, J = 1.4 Hz, 2H, ArH), 7.50 (d, J=8.2 Hz, 2H, ArH), 7.29
(dd, J1 = 8.2 Hz, J2=1.4 Hz, 2H, ArH), 2.29 (s, 6H, CH3). 13C NMR
(100 MHz, DMSO-d6) d 168.52, 135.94, 134.33, 131.82, 129.84,
129.45, 128.52, 20.07. MS (EI) m/z: 429.9 (M+).
4.2.1.3. 5,50-difluoro-2,20-diselenobisbenzoic acid (2c).
Com-
pound 1c (15.5 g, 0.1 mol) was used as reactant to give 19.9 g of 2c
as yellow solid (91%). 1H NMR (400 MHz, DMSO-d6) d 14.07 (br, 2H,
COOH), 7.78 (dd, J1 = 9.1, J2 = 2.8, 2H, ArH), 7.65 (dd, J1 = 8.5, J2 = 5.2,
2H, ArH), 7.42 (td, J1 = 8.5, J2 = 2.8, 2H, ArH). 13C NMR (100 MHz,
DMSO-d6) d 167.54, 160.91 (d, JC–F = 244.5), 131.65 (d, JC–F = 7.2),
130.34 (d, JC–F = 6.7), 128.18 (d, JC–F = 2.5), 120.96 (d, JC–F = 21.6),
117.85 (d, JC–F = 23.0). MS (EI) m/z: 437.9 (M+).

J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
3823
ArH), 7.20 (d, J = 2.7, 2H, ArH), 7.13 (dd, J1 = 8.8, J2 = 2.7, 2H, ArH), 3.75
(s, 6H, OCH3). 13C NMR (100 MHz, DMSO-d6) d 168.20, 158.18, 130.75,
129.64, 123.26, 120.22, 115.96, 55.52. MS (ESI) m/z: 460.9 (M+ꢂH).
4.2.2.7. 4-Fluoro-2-(chloroseleno)benzoyl chloride (3g).
Com-
pound 2g (4.4 g, 0.01 mol) was used as reactant to give 4.5 g of 3g
as orange crystal (83%). 1H NMR (400 MHz, CDCl3) d 7.86 (dd,
J1 = 9.9, J2 = 2.3, 1H, ArH), 7.80 (dd, J1 = 8.4, J2 = 5.7, 1H, ArH), 7.19 (td,
J1 = 8.4, J2 = 2.3, 1H, ArH). 13C NMR (100 MHz, CDCl3) d 171.83,
160.44, 147.83, 144.46, 135.26, 127.44, 127.10.
4.2.1.11. 5,50-diacetoxy -2,20-diselenobisbenzoic acid (2k).
Com-
pound 2i (4.3 g, 0.01 mol), 40 ml acetic anhydride and 3 ml pyridine
were used as reactants and the mixture was under reflux at 100 °C
for half an hour. Then it was poured into 100 ml water and was stirred
for another 1 h, after which it was extracted by ethyl acetate and the or-
ganic solvent was dried and removed to give 3.0 g of 2k. White solid
(59%). 1H NMR (400 MHz, DMSO-d6) d 13.98 (br, 2H, COOH), 7.78 (d,
J = 2.6, 2H, ArH), 7.67 (d, J = 8.7, 2H, ArH), 7.31 (dd, J1 = 8.7, J2 = 2.6,
2H, ArH), 2.25 (s, 6H, CH3). 13C NMR (100 MHz, DMSO-d6) d 169.13,
167.81, 151.37, 149.17, 130.67, 130.02, 129.50, 127.32, 20.77. MS (EI)
m/z: 518.0 (M+).
4.2.2.8. 4-Chloro-2-(chloroseleno)benzoyl chloride (3h).
Com-
pound 2h (4.7 g, 0.01 mol) was used as reactant to give 5.0 g of 3h
as orange crystal (87%). 1H NMR (400 MHz, CDCl3) d 8.13 (d, J = 1.8,
1H, ArH), 7.74 (d, J = 8.1, 1H, ArH), 7.40 (dd, J1 = 8.1, J2 = 1.8, 1H,
ArH). 13C NMR (100 MHz, CDCl3) d 171.61, 151.20, 136.24, 135.47,
132.15, 128.54, 124.58.
4.2.2.9. 5-Methoxy-2-(chloroseleno)benzoyl chloride (3j).
Com-
pound 2j (4.6 g, 0.01 mol) was used as reactant to give 3.9 g of 3j as or-
ange crystal (68%). 1H NMR (400 MHz, CDCl3) d 7.90 (d, J = 8.8, 1H, ArH),
7.39 (d, J = 2.7, 1H, ArH), 7.33 (dd, J1 = 8.8, J2 = 2.7, 1H, ArH), 3.80 (s, 3H,
OCH3). 13C NMR (100 MHz, CDCl3) d 172.53, 158.31, 134.70, 127.28,
127.22, 121.64, 112.50, 55.64.
4.2.2. General procedure for the synthesis of 3(a–k)
2,20-Diselenobisbenzoic acid (0.01 mol) was added to thionyl
chloride (20 ml) and DMF (0.5 ml) was added as a catalyst. The
reaction mixture was stirred at reflux (85 °C) for 3 h. The solvents
were evaporated under vacuum and the crude products were puri-
fied by recrystallization from cyclohexane.
4.2.2.10. 5-Acetoxy-2-(chloroseleno)benzoyl chloride (3k).
Com-
pound 2k (5.2 g, 0.01 mol) was used as reactant to give 4.1 g of 3k as or-
ange crystal (65%). 1H NMR (400 MHz, CDCl3) d 8.13 (d, J = 2.4, 1H, ArH),
8.10 (d, J = 8.8, 1H, ArH), 7.57 (dd, J1 = 8.8, J2 = 2.4, 1H, ArH), 2.40 (s, 3H,
CH3). 13C NMR (100 MHz, CDCl3) d 172.02, 169.05, 149.47, 143.04,
130.23, 129.60, 128.29, 126.90, 21.11.
4.2.2.1. 2-(Chloroseleno)benzoyl chloride (3a).
Compound
2a (4 g, 0.01 mol) was used as reactant to give 4.1 g of 3a as orange
crystal (81%). 1H NMR (400 MHz, CDCl3) d 8.09 (d, J = 7.7, 1H, ArH),
7.85 (d, J = 7.7, 1H, ArH), 7.67 (t, J = 7.7, 1H, ArH), 7.39 (t, J = 7.7, 1H,
ArH). 13C NMR (100 MHz, CDCl3) d 164.51, 136.19, 134.54, 131.29,
129.03, 127.41, 126.42.
4.2.3. General procedure for the synthesis of 4(a–j)
A solution of chloride 3a (1.27 g, 5 mmol) in dry acetonitrile
was added dropwise to a stirred solution of the corresponding
amine in dry acetonitrile containing triethylamine (1.1 g, 11 mmol)
over 30 min at room temperature. Stirring was continued for addi-
tional 3 h. Upon completion of the reaction, the solvent was evap-
orated under vacuum and the residue was treated with water and
stirred for 1 h. The insoluble solid was filtered off and washed with
water to provide the crude product, which was recrystallized from
DMSO/H2O.
4.2.2.2. 5-Methyl-2-(chloroseleno)benzoyl chloride (3b).
Com-
pound 2b (4.3 g, 0.01 mol) was used as reactant to give 4.7 g of 3b
as orange crystal (87%). 1H NMR (400 MHz, CDCl3) d 7.96 (d, J = 8.2,
1H, ArH), 7.65 (s, 1H, ArH), 7.48 (d, J = 8.2, 1H, ArH), 2.37 (s, 6H,
CH3). 13C NMR (100 MHz, CDCl3) d 172.74, 141.63, 135.72, 134.17,
129.61, 127.44, 126.38, 20.66.
4.2.2.3. 5-Fluoro-2-(chloroseleno)benzoyl chloride (3c).
Com-
pound 2c (4.4 g, 0.01 mol) was used as reactant to give 4.1 g of 3c as
orange crystal (76%). 1H NMR (400 MHz, CDCl3) d 7.78 (dd, J1 = 9.0,
J2 = 2.7, 1H, ArH), 7.65 (dd, J1 = 8.3, J2 = 5.4, 1H, ArH), 7.42 (td,
J1 = 8.3, J2 = 2.7, 1H, ArH). 13C NMR (100 MHz, CDCl3) d 168.84,
161.41 (d, JC–F = 248.6), 131.65 (d, JC–F = 7.5), 130.34 (d, JC–F = 8.7),
128.18 (d, JC–F = 2.2), 120.96 (d, JC–F = 24.3), 117.85 (d, JC–F = 22.5).
4.2.3.1.1,2-(Bis-1,2-benzisoselenazol-3(2H)-one)ethane (4a).
Com-
pound 3a and 1,2-diaminoethane (0.16 g, 2.6 mmol) were used as reac-
tants to give 0.98 g of 4a as yellow solid (92%). 1H NMR (400 MHz,
DMSO-d6) d 7.96 (d, J = 8.0, 2H, ArH), 7.80 (d, J = 7.7, 2H, ArH), 7.56 (t,
J = 8.0, 2H, ArH), 7.39 (t, J = 7.7, 2H, ArH), 4.01 (s, 4H, CH2). 13C NMR
(100 MHz, DMSO-d6) d 166.59, 139.69, 131,38, 127.65, 127.31, 125.82,
125.64, 43.12. HRMS Calcd for C16H12N2O2Se2: 423.9229. Found
424.9302 (M++H). HPLC tR = 18.06 min (water/methanol (40:60), purity
99.99%).
4.2.2.4. 5-Chloro-2-(chloroseleno)benzoyl chloride (3d).
Com-
pound 2d (4.7 g, 0.01 mol) was used as reactant to give 4.4 g of 3d
as orange crystal (76%). 1H NMR (400 MHz, CDCl3) d 8.12 (d, J = 8.6,
1H, ArH), 7.69 (d, J = 2.2, 1H, ArH), 7.65 (dd, J1 = 8.6, J2 = 2.2, 1H,
ArH). 13C NMR (100 MHz, CDCl3) d 171.70, 144.40, 136.12, 133.58,
133.06, 130.02, 128.51.
4.2.3.2.
(4b).
1,3-(Bis-1,2-benzisoselenazol-3(2H)-one)propane
Compound 3a and 1,3-diaminopropane (0.2 g, 2.7 mmol)
were used as reactants to give 0.95 g of 4b as yellow solid (87%). 1H
NMR (400 MHz, DMSO-d6) d 8.05 (d, J = 8.4, 2H, ArH), 7.80 (d, J = 7.6,
2H, ArH), 7.60 (t, J = 8.4, 2H, ArH), 7.41 (t, J = 7.6, 2H, ArH), 3.79 (t,
J = 6.9, 4H, CH2), 2.00 (quint, J = 6.9, 2H, CH2). 13C NMR (100 MHz,
4.2.2.5. 5-Bromo-2-(chloroseleno)benzoyl chloride (3e).
Com-
pound 2e (5.6 g, 0.01 mol) was used as reactant to give 4.7 g of 3e as
yellow crystal (71%). 1H NMR (400 MHz, CDCl3) d 8.14 (d, J = 2.2, 1H,
ArH), 7.65 (dd, J1 = 8.6, J2 = 2.2, 1H, ArH), 7.49 (d, J = 8.6, 1H, ArH).
13C NMR (100 MHz, CDCl3) d 171.55, 145.11, 138.90, 136.60, 130.34,
128.70, 120.33.
DMSO-d6)
d 166.89, 139.76, 131.92, 128.41, 127.77, 126.36,
126.24, 41.55, 31.55. HRMS Calcd for C17H14N2O2Se2: 437.9386.
Found 438.9461 (M++H). HPLC tR = 24.03 min (water/methanol
(50:50), purity 98.87%).
4.2.2.6. 5-Nitro-2-(chloroseleno)benzoyl chloride (3f).
Com-
pound 2f (4.9 g, 0.01 mol) was used as reactant to give 4.0 g of 3f
as yellow crystal (67%). 1H NMR (400 MHz, CDCl3) d 8.32 (d, J = 2.1,
1H, ArH), 8.01 (dd, J1 = 8.5, J2 = 2.1, 1H, ArH), 7.89 (d, J = 8.5, 1H,
ArH). 13C NMR (100 MHz, CDCl3) d 170.05, 153.96, 146.17, 129.59,
127.93, 126.01, 122.91.
4.2.3.3. 1,4-(Bis-1,2-benzisoselenazol-3(2H)-one)butane (4c).
Com-
pound 3a and 1,4-diaminobutane (0.24 g, 2.7 mmol) were used as reac-
tants to give 0.96 g of 4c as yellow solid (85%). 1H NMR (400 MHz,
DMSO-d6) d 8.03 (d, J = 8.3, 2H, ArH), 7.79 (d, J = 7.3, 2H, ArH), 7.58 (t,
J = 8.3, 2H, ArH), 7.40 (t, J = 7.3, 2H, ArH), 3.76 (s, 4H, CH2), 1.66 (s,

3824
J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
4H,CH2). 13C NMR (100 MHz, DMSO-d6) d 166.27, 139.07, 131.33, 127.94,
127.26, 125.81, 125.68, 42.73, 27.06. HRMS Calcd for C18H16N2O2Se2:
451.9542. Found 452.9618 (M++H). HPLC tR = 25.57 min (water/methanol
(50:50), purity 96.22%).
CH2). 13C NMR (100 MHz, DMSO-d6) d 166.30, 164.95, 139.30,
138.90, 138.81, 135.82, 135.20, 132.20, 128.73, 127.88, 125.81,
124.63, 46.20. HRMS Calcd for C21H14N2O2Se2: 485.9385. Found
486.9457 (M++H). HPLC tR = 5.27 min (water/methanol (20:80), pur-
ity 96.83%).
4.2.3.4.
(4d).
1,5-(Bis-1,2-benzisoselenazol-3(2H)-one)pentane
Compound 3a and 1,5-diaminopentane (0.29 g, 2.8 mmol)
4.2.3.10.
(4j).
1,4-(Bis-1,2-benzisoselenazol-3(2H)-one)benzene
Compound 3a and 1,4-diaminobenzene (0.28 g, 2.6 mmol)
were used as reactants to give 0.98 g of 4d as yellow solid (84%). 1H
NMR (400 MHz, DMSO-d6) d 8.02 (d, J = 8.1, 2H, ArH), 7.74 (d, J = 7.3,
2H, ArH), 7.52 (t, J = 8.1, 2H, ArH), 7.41 (t, J = 7.3, 2H, ArH), 3.71 (t,
J = 7.0, 4H, CH2), 1.68 (quint, J = 7.0, 4H, CH2), 1.35 (quint, J = 7.0,
2H, CH2). 13C NMR (100 MHz, DMSO-d6) d 166.71, 139.54, 131.83,
128.52, 127.77, 126.29, 126.20, 43.57, 30.04, 23.77. HRMS Calcd for
were used as reactants to give 0.89 g of 4j as yellow solid (75%). 1H
NMR (400 MHz, DMSO-d6) d 8.10 (d, J = 8.0, 2H, ArH), 7.91 (d, J = 7.7,
2H, ArH), 7.72 (s, 4H, ArH), 7.68 (t, J = 8.0, 2H, ArH), 7.49 (t, J = 7.7,
2H, ArH). 13C NMR (100 MHz, DMSO-d6) d 165.01, 138.90, 137.14,
132.27, 128.44, 127.92, 126.50, 125.27. HRMS Calcd for
C
19H18N2O2Se2: 465.9699. Found 466.9785 (M++H). HPLC
tR = 23.87 min (water/methanol (50:50), purity 99.45%).
C
20H12N2O2Se2: 471.9229. Found 472.9301 (M++H). HPLC
tR = 5.70 min (water/methanol (20:80), purity 95.14%).
4.2.3.5.
(4e).
1,6-(Bis-1,2-benzisoselenazol-3(2H)-one)hexane
Compound 3a and 1,6-diaminohexane (0.30 g, 2.6 mmol)
4.2.4. General procedure for synthesis of 5(a–f), 6(a–e), 7(a, b), 8
(a, b), 9, 10(a, b), 11(a, b), 12, 13
were used as reactants to give 1.03 g of 4e as yellow solid (86%). 1H
NMR (400 MHz, DMSO-d6) d 8.04 (d, J = 8.1, 2H, ArH), 7.84 (d,
J = 7.6, 2H, ArH), 7.59 (t, J = 8.1, 2H, ArH), 7.44 (t, J = 7.6, 2H, ArH),
3.71 (t, J = 7.0, 4H, CH2), 1.63 (quint, J = 7.0, 4H, CH2), 1.36 (quint,
J = 7.0, 4H, CH2). 13C NMR (100 MHz, DMSO-d6) d 166.69, 139.51,
131.83, 128.54, 127.77, 126.88, 126.21, 43.63, 30.31, 26.25. HRMS
Calcd for C20H20N2O2Se2: 479.9855. Found 480.9936 (M++H). HPLC
tR = 37.26 min (water/methanol (50:50), purity 96.10%).
A solution of chloride 3(b–k) (5 mmol) in dry dichloromethane
was added dropwise to a stirred solution of the corresponding
amine in dry dichloromethane containing triethylamine (1.1 g,
11 mmol) over 30 min at room temperature. Stirring was contin-
ued for additional 4 h. Upon completion of the reaction, the solvent
was evaporated under vacuum and the residue was treated with
water and stirred for 1 h. The insoluble solid was filtered off and
washed with water to give the crude product, which was recrystal-
lized from DMSO/H2O.
4.2.3.6.
(4f).
1,7-(Bis-1,2-benzisoselenazol-3(2H)-one)heptane
Compound 3a and 1,7-diaminoheptane (0.38 g, 2.9 mmol)
4.2.4.1. 1,2-(5,50-Dimethyl(bis-1,2-benzisoselenazol-3(2H)-one))
were used as reactants to give 0.99 g of 4f as yellow solid (80%). 1H
NMR (400 MHz, DMSO-d6) d 8.03 (d, J = 8.0, 2H, ArH), 7.79 (d, J = 7.5,
2H, ArH), 7.51 (t, J = 8.0, 2H, ArH), 7.40 (t, J = 7.5, 2H, ArH), 3.71 (t,
J = 7.2, 4H, CH2), 1.63 (quint, J = 7.0, 4H, CH2), 1.33 ꢀ 1.28 (m, 6H,
CH2). 13C NMR (100 MHz, DMSO-d6) d 166.15, 138.98, 131.31,
128.02, 127.25, 125.77, 125.69, 43.12, 29.81, 28.25, 26.01. HRMS
Calcd for C21H22N2O2Se2: 494.0012. Found 495.0085 (M++H). HPLC
tR = 6.10 min (water/methanol (20:80), purity 96.37%).
ethane (5a).
Compound 3b (1.34 g, 5 mmol) and 1,2-diami-
noethane (0.16 g, 2.6 mmol) were used as reactants to give 0.99 g
of 5a as yellow solid (88%). 1H NMR (400 MHz, DMSO-d6) d 7.82
(d, J = 8.2, 2H, ArH), 7.61 (d, J = 1.4, 2H, ArH), 7.39 (dd, J1 = 8.2, J2=
1.4, 2H, ArH), 3.97 (s, 4H, CH2), 2.38 (s, 6H, CH3). 13C NMR
(100 MHz, DMSO-d6) d 166.55, 136.42, 135.19, 132.70, 129.26,
127.38, 126.06, 125.68, 43.20, 20.57. MS (EI) m/z: 452.1 (M+). HPLC
tR = 9.23 min (water/methanol (40:60), purity 99.58%).
4.2.3.7.
(4g).
1,8-(Bis-1,2-benzisoselenazol-3(2H)-one)octane
Compound 3a and 1,8-diaminooctane (0.36 g, 2.5 mmol)
4.2.4.2. 1,3-(5,50-Dimethyl(bis-1,2-benzisoselenazol-3(2H)-one))
propane (5b).
Compound 3b (1.34 g, 5 mmol) and 1,3-diami-
were used as reactants to give 0.98 g of 4g as yellow solid (77%). 1H
NMR (400 MHz, DMSO-d6) d 8.03 (d, J = 8.0, 2H, ArH), 7.79 (d,
J = 7.5, 2H, ArH), 7.51 (t, J = 8.0, 2H, ArH), 7.40 (t, J = 7.5, 2H, ArH),
3.70 (t, J = 7.3, 4H, CH2), 1.61 (quint, J = 7.0, 4H, CH2), 1.29 (s, 8H,
CH2). 13C NMR (100 MHz, DMSO-d6) d 166.91, 137.98, 131.55,
128.88, 127.43, 125.12, 124.56, 44.42, 28.55, 27.15, 24.82. HRMS
Calcd for C22H24N2O2Se2: 508.0168. Found 509.0242 (M++H). HPLC
tR = 7.08 min (water/methanol (20:80), purity 97.31%).
nopropane (0.19 g, 2.6 mmol) were used as reactants to give 0.97 g
of 5b as yellow solid (83%). 1H NMR (400 MHz, DMSO-d6) d 7.90 (d,
J = 8.1, 2H, ArH), 7.69 (d, J = 1.9, 2H, ArH), 7.43 (dd, J1 = 8.1, J2= 1.9,
2H, ArH), 3.75 (t, J = 7.0, 4H, CH2), 2.39 (s, 6H, CH3). 13C NMR
(100 MHz, DMSO-d6) d 166.33, 135.90, 135.21, 132.59, 127.90,
127.26, 125.56, 41.08, 31.00, 20.41. MS (EI) m/z: 466.0 (M+). HPLC
tR = 15.90 min (water/methanol (40:60), purity 98.84%).
4.2.4.3. 1,4-(5,50-Dimethyl(bis-1,2-benzisoselenazol-3(2H)-one))
4.2.3.8.
(4h).
1,6-(Bis-1,2-benzisoselenazol-3(2H)-one)cyclohexane
Compound 3a and 1,6-diamino- cyclohexane (0.33 g,
butane (5c).
Compound 3b (1.34 g, 5 mmol) and 1,4-diamin-
obutane (0.23 g, 2.6 mmol) were used as reactants to give 0.95 g of
5c as yellow solid (79%). 1H NMR (400 MHz, DMSO-d6) d 7.90 (d,
J = 8.0, 2H, ArH), 7.72 (d, J = 1.8, 2H, ArH), 7.41 (dd, J1 = 8.0, J2=
1.8, 2H, ArH), 3.73 (s, 4H, CH2), 2.38 (s, 6H, CH3), 1.63 (s, 4H,
CH2). 13C NMR (100 MHz, DMSO-d6) d 166.54, 134.65, 134.71,
131.88, 127.54, 127.76, 125.98, 42.79, 29.97, 20.34. HRMS Calcd
2.9 mmol) were used as reactants to give 0.85 g of 4h as yellow so-
lid (71%). 1H NMR (400 MHz, DMSO-d6) d 8.01 (d, J = 8.2, 2H, ArH),
7.65 (d, J = 7.3, 2H, ArH), 7.45 (t, J = 8.2, 2H, ArH), 7.33 (t, J = 7.3, 2H,
ArH), 4.53 (s, 2H, CH), 2.10–2.00 (m, 8H, CH2). 13C NMR (100 MHz,
DMSO-d6) d 166.20, 138.90, 133.06, 128.27, 127.37, 126.03, 125.84,
49.81, 29.15. HRMS Calcd for C20H18N2O2Se2: 477.9699. Found
478.9764 (M++H). HPLC tR = 34.30 min (water/methanol (50:50),
purity 95.58%).
for
C
20H20N2O2Se2: 479.9855. Found 480.9923 (M++H). HPLC
tR = 17.31 min (water/methanol (40:60), purity 97.96%).
4.2.4.4. 1,5-(5,50-dimethyl(bis-1,2-benzisoselenazol-3(2H)-one))
4.2.3.9.
(4i).
a
,4-(Bis-1,2-benzisoselenazol-3(2H)-one)toluene
pentane (5d).
Compound 3b (1.34 g, 5 mmol) and 1,5-diam-
Compound 3a and ,4-diaminotoluene (0.33 g, 2.7 mmol)
a
inopentane (0.27 g, 2.6 mmol) were used as reactants to give
0.94 g of 5d. as yellow solid (76%). 1H NMR (400 MHz, DMSO-d6)
d 7.88 (d, J = 8.1, 2H, ArH), 7.60 (s, 2H, ArH), 7.41 (d, J = 8.1, 2H,
ArH), 3.67 (t, J = 6.9, 4H, CH2), 2.39 (s, 6H, CH3), 1.65 (quint,
J = 6.9, 4H, CH2), 1.33 (quint, J = 6.9, 2H, CH2). 13C NMR (100 MHz,
were used as reactants to give 0.79 g of 4i as yellow solid (65%).
1H NMR (400 MHz, DMSO-d6) d 8.09 (d, J = 8.0, 2H, ArH), 8.03 (d,
J = 8.0, 2H, ArH), 7.89 (d, J = 7.7, 2H, ArH), 7.87 (d, J = 7.7, 2H, ArH),
7.70 ꢀ 7.60 (m, 4H, ArH), 7.50 ꢀ 7.40 (m, 4H, ArH), 4.94 (s, 2H,

J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
3825
DMSO-d6) d 166.66, 136.27, 135.69, 133.05, 128.54, 127.77, 126.01,
43.57, 30.03, 28.7, 20.92. MS (EI) m/z: 494.0 (M+). HPLC
tR = 6.42 min (water/methanol (20:80), purity 96.94%).
2H, ArH), 7.51 (dd, J1 = 8.6, J2 = 2.5, 2H, ArH), 7.47 (td J1 = 8.8,
J2 = 2.5, 2H, ArH), 3.69 (t, J = 6.9, 4H, CH2), 1.66 (quint, J = 6.9, 4H,
CH2), 1.33 (quint, J = 6.9, 2H, CH2). 13C NMR (100 MHz, DMSO-d6)
d 165.55, 162.12 (d, JC–F = 241.2), 133.56, 128.41 (d, JC–F = 6.9),
127.51 (d, JC–F = 7.1), 119.96 (d, JC–F = 21.5), 112.49 (d, JC–F = 21.8),
43.24, 29.37, 23.13. MS (EI) m/z: 501.9 (M+). HPLC tR = 5.28 min
(water/methanol (20:80), purity 97.89%).
4.2.4.5. 1,6-(5,50-Dimethyl(bis-1,2-benzisoselenazol-3(2H)-one))
hexane (5e).
Compound 3b (1.34 g, 5 mmol) and 1,6-diami-
nohexane (0.30 g, 2.6 mmol) were used as reactants to give
1.04 g of 5e as yellow solid (82%). 1H NMR (400 MHz, DMSO-d6)
d 7.88 (d, J = 8.2, 2H, ArH), 7.62 (s, 2H, ArH), 7.39 (d, J = 8.2, 2H,
ArH), 2.39 (s, 6H, CH3), 3.70 (t, J = 6.9, 4H, CH2), 1.60 (s, 4H, CH2),
1.33 (s, 4H, CH2). 13C NMR (100 MHz, DMSO-d6) d 166.13, 135.63,
135.20, 132.53, 128.04, 127.26, 125.49, 43.11, 29.79, 25.71, 20.40.
4.2.5.1. 1,4-(5,50-Difluoro(bis-1,2-benzisoselenazol-3(2H)-one))
cyclohexane (6e).
Compound 3c (1.36 g, 5 mmol) and 1,4-
diaminocyclohexane (0.32 g, 2.8 mmol) were used as reactants to
give 1.14 of 6e as yellow solid (89%). 1H NMR (400 MHz, DMSO-
d6) d 8.11 (dd, J1 = 8.8, J2 = 5.1, 2H, ArH), 7.58 (dd, J1 = 8.7, J2 = 2.7,
2H, ArH), 7.51 (td J1 = 8.8, J2 = 2.7, 2H, ArH), 4.51 (s, 2H, CH), 2.09
ꢀ 2.00 (m, 8H, CH2). 13C NMR (100 MHz, DMSO-d6) d 165.97,
161.76 (d, JC–F = 244.3), 132.59, 128.87 (d, JC–F = 6.9), 126.88 (d, J
C–F = 7.7), 117.54 (d, JC–F = 22.3), 112.61 (d, JC–F = 22.1), 50.12,
28.89. MS (EI) m/z: 513.9 (M+). HPLC tR = 6.75 min (water/methanol
(20:80), purity 97.44%).
HRMS Calcd for
C22H24N2O2Se2: 508.0168. Found 509.0231
(M++H). HPLC tR = 7.44 min (water/methanol (20:80), purity
97.30%).
4.2.4.6. 1,4-(5,50-Dimethyl(bis-1,2-benzisoselenazol-3(2H)-one))
cyclohexane (5f).
Compound 3b (1.34 g, 5 mmol) and 1,4-
diaminocyclohexane (0.30 g, 2.6 mmol) were used as reactants to
give 0.93 g of 5f as yellow solid (74%). 1H NMR (400 MHz, DMSO-
d6) d 7.90 (d, J = 8.4, 2H, ArH), 7.73 (s, 2H, ArH), 7.41 (d, J = 8.4, 2H,
ArH), 4.49 (s, 2H, CH), 2.40 (s, 6H, CH3), 2.02 ꢀ 1.90 (m, 8H, CH2).
13C NMR (100 MHz, DMSO-d6) d 166.87, 135.21, 134.80, 132.78,
127.55, 126.87, 125.12, 49.82, 29.02, 20.42. MS (EI) m/z: 505.0
(M+). HPLC tR = 7.41 min (water/methanol (20:80), purity 98.14%).
4.2.5.2. 1,2-(5,50-Dichloro(bis-1,2-benzisoselenazol-3(2H)-one))
ethane (7a).
Compound 3d (1.44 g, 5 mmol) and 1,2-diami-
noethane (0.16 g, 2.6 mmol) were used as reactants to give 1.01 g
of 7a as yellow solid (82%). 1H NMR (400 MHz, DMSO-d6) d 7.98
(d, J = 8.6, 2H, ArH), 7.74 (d, J = 2.0, 2H, ArH), 7.62 (dd, J1 = 8.6,
J2 = 2.0, 2H, ArH), 4.00 (s, 4H, CH2). 13C NMR (100 MHz, DMSO-
d6) d 166.96, 141.51, 132.77, 131.87, 130.59, 129.27, 129.02,
41.86. MS (EI) m/z: 492.2 (M+). HPLC tR = 6.54 min (water/methanol
(20:80), purity 99.17%).
4.2.4.7. 1,2-(5,50-Difluoro(bis-1,2-benzisoselenazol-3(2H)-one))
ethane (6a).
Compound 3c (1.36 g, 5 mmol) and 1,2-diami-
noethane (0.16 g, 2.6 mmol) were used as reactants to give 1.05 g
of 6a as yellow solid (91%). 1H NMR (400 MHz, DMSO-d6) d 7.98
(dd, J1 = 8.8, J2 = 4.9, 2H, ArH), 7.54 (dd, J1 = 8.6, J2 = 2.7, 2H, ArH),
7.48 (td J1 = 8.8, J2 = 2.7, 2H, ArH), 4.00 (s, 4H, CH2). 13C NMR
(100 MHz, DMSO-d6) d 165.94, 160.48 (d, JC–F = 249.1), 130.21 (d,
JC–F = 5.9), 128.34 (d, JC–F = 6.9), 126.12 (d, JC–F = 2.1), 120.45 (d, J
C–F = 22.1), 116.81 (d, JC–F = 20.5), 43.21. HRMS Calcd for
4.2.5.3. 1,5-(5,50-Dichloro(bis-1,2-benzisoselenazol-3(2H)-one))
pentane (7b).
Compound 3d (1.44 g, 5 mmol) and 1,5-diam-
inopentane (0.29 g, 2.8 mmol) were used as reactants to give
1.15 g of 7b as yellow solid (86%). 1H NMR (400 MHz, DMSO-d6)
d 8.06 (d, J = 8.5, 2H, ArH), 7.73 (d, J = 2.2, 2H, ArH), 7.65 (dd,
J1 = 8.5, J2 = 2.2, 2H, ArH), 3.71 (t, J = 6.8, 4H, CH2), 1.67 (quint,
J = 6.8, 4H, CH2), 1.33 (quint, J = 6.8, 2H, CH2). 13C NMR (100 MHz,
C
16H10N2O2F2Se2: 459.9041. Found 460.9121 (M++H). HPLC
tR = 4.72 min (water/methanol (20:80), purity 95.30%).
DMSO-d6)
d 166.77, 142.54, 133.51, 132.00, 130.41, 129.85,
4.2.4.8. 1,3-(5,50-Difluoro(bis-1,2-benzisoselenazol-3(2H)-one))
128.70, 43.20, 29.60, 23.43. MS (EI) m/z: 533.9 (M+). HPLC
tR = 9.37 min (water/methanol (20:80), purity 97.15%).
propane (6b).
Compound 3c (1.36 g, 5 mmol) and 1,3-diami-
nopropane (0.19 g, 2.6 mmol) were used as reactants to give 1.07 g
of 6b as yellow solid (90%). 1H NMR (400 MHz, DMSO-d6) d 8.05
(dd, J1 = 8.7, J2 = 5.1, 2H, ArH), 7.54 (dd, J1 = 8.5, J2 = 2.6, 2H, ArH),
7.48 (td J1 = 8.7, J2 = 2.6, 2H, ArH), 3.78 (t, J = 6.9, 4H, CH2), 1.99
(quint, J = 6.9, 2H, CH2). 13C NMR (100 MHz, DMSO-d6) d 165.53,
161.11 (d, JC–F = 242.9), 134.35, 129.60 (d, JC–F = 7.4), 127.94 (d, J
C–F = 9.5), 121.66 (d, JC–F = 21.1), 113.00 (d, JC–F = 23.3), 41.28,
30.96. MS (EI) m/z: 473.9 (M+). HPLC tR = 5.66 min (water/methanol
(20:80), purity 97.44%).
4.2.5.4. 1,2-(5,50-Dibromo(bis-1,2-benzisoselenazol-3(2H)-one))
ethane (8a).
Compound 3e (1.66 g, 5 mmol) and 1,2-diami-
noethane (0.16 g, 2.7 mmol) were used as reactants to give 1.17 g
of 8a as yellow solid (81%). 1H NMR (400 MHz, DMSO-d6) d 7.91
(d, J = 8.6, 2H, ArH), 7.87 (s, 2H, ArH), 7.73 (d, J = 8.6, 2H, ArH),
4.00 (s, 4H, CH2). 13C NMR (100 MHz, DMSO-d6) d 170.98, 146.42,
139.15, 137.32, 131.77, 127.23, 121.45, 42.40. MS (EI) m/z: 581.7
(M+). HPLC tR = 7.39 min (water/methanol (20:80), purity 96.83%).
4.2.4.9. 1,4-(5,50-Difluoro(bis-1,2-benzisoselenazol-3(2H)-one))-
4.2.5.5. 1,5-(5,50-Dibromo(bis-1,2-benzisoselenazol-3(2H)-one))
butane (6c).
Compound 3c (1.36 g, 5 mmol) and 1,4-diamin-
pentane (8b).
Compound 3e (1.66 g, 5 mmol) and 1,5-diam-
obutane (0.23 g, 2.6 mmol) were used as reactants to give 1.11 of
6c as yellow solid (91%). 1H NMR (400 MHz, DMSO-d6) d 8.04 (dd,
J1 = 8.8, J2 = 4.8, 2H, ArH), 7.52 (dd, J1 = 8.6, J2 = 2.6, 2H, ArH), 7.49
(td J1 = 8.8, J2 = 2.6, 2H, ArH), 3.75 (s, 4H, CH2), 1.65 (s, 4H, CH2).
13C NMR (100 MHz, DMSO-d6) d 165.29, 161.02 (d, JC–F = 236.7),
133.96, 129.51 (d, JC–F = 7.4), 127.73 (d, JC–F = 7.4), 119.32 (d, JC–
F = 23.6), 112.83 (d, JC–F = 22.8), 42.91, 26.89. MS (EI) m/z: 487.9
(M+). HPLC tR = 5.72 min (water/methanol (20:80), purity 97.74%).
inopentane (0.28 g, 2.7 mmol) were used as reactants to give
1.31 g of 8b as yellow solid (84%). 1H NMR (400 MHz, DMSO-d6)
d 7.97 (d, J = 8.5, 2H, ArH), 7.84 (s, 2H, ArH), 7.75 (d, J = 8.5, 2H,
ArH), 3.69 (t, J = 6.9, 4H, CH2), 1.65 (quint, J = 6.9, 4H, CH2), 1.31
(quint, J = 6.9, 2H, CH2). 13C NMR (100 MHz, DMSO-d6) d 170.66,
145.48, 138.92, 137.11, 131.57, 127.40, 121.79, 42.24, 29.39,
22.21. MS (EI) m/z: 621.8 (M+). HPLC tR = 11.28 min (water/metha-
nol (20:80), purity 97.50%).
4.2.5. 1,5-(5,50-Difluoro(bis-1,2-benzisoselenazol-3(2H)-
one))pentane (6d)
4.2.5.6. 1,2-(5,50-Dinitro(bis-1,2-benzisoselenazol-3(2H)-one))
ethane (9).
Compound 3f (1.5 g, 5 mmol) and 1,2-diaminoe-
Compound 3c (1.36 g, 5 mmol) and 1,5-diaminopentane (0.28 g,
2.7 mmol) were used as reactants to give 1.15 g of 6d as yellow so-
lid (92%). 1H NMR (400 MHz, DMSO-d6) d 8.04 (dd, J1 = 8.8, J2 = 5.0,
thane (0.16 g, 2.6 mmol) were used as reactants to give 0.95 g of
9 as orange solid (74%). 1H NMR (400 MHz, DMSO-d6) d 8.43 (d,
J = 2.2, 2H, ArH), 8.36 (dd, J1 = 8.8, J2 = 2.2, 2H, ArH), 8.22 (d,

3826
J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
J = 8.8, 2H, ArH), 4.09 (s, 4H, CH2). 13C NMR (100 MHz, DMSO-d6) d
165.33, 147.92, 146.02, 128.49, 127.36, 125.11, 121.79, 43.10. MS
(EI) m/z: 514.1 (M+). HPLC tR = 7.19 min (water/methanol (40:60),
purity 95.25%).
NMR (100 MHz, DMSO-d6) d 169.29, 165.89, 148.76, 136.40,
128.67, 126.88, 125.58, 120.08, 43.18, 20.79. HRMS Calcd for
C
20H16N2O6Se2: 539.9339. Found 562.9232 (M++Na). HPLC
tR = 6.90 min (water/methanol (40:60), purity 96.29%).
4.2.5.7. 1,2-(4,40-Difluoro(bis-1,2-benzisoselenazol-3(2H)-one))
4.2.5.13.
one))ethane (14).
1,2-(5,50-Dihydroxy(bis-1,2-benzisoselenazol-3(2H)-
Compound 13 (2 g, 4.5 mmol) was dis-
ethane (10a).
Compound 3g (1.36 g, 5 mmol) and 1,2-diami-
noethane (0.16 g, 2.6 mmol) were used as reactants to give 0.92 g
of 10a as yellow solid (80%). 1H NMR (400 MHz, DMSO-d6) d 7.82
(dd, J1 = 8.5, J2 = 5.5, 2H, ArH), 7.74 (dd, J1 = 9.2, J2 = 2.3, 2H, ArH),
7.24 (td, J1 = 8.5, J2 = 2.3, 2H, ArH), 3.99 (s, 4H, CH2). 13C NMR
(100 MHz, DMSO-d6) d 166.65, 161.54 (d, JC–F = 224.5), 141.62 (d,
JC–F = 8.8), 129.33 (d, JC–F = 9.9), 124.37 (d, JC–F = 1.4), 113.88 (d, J
C–F = 23.4), 112.20 (d, JC–F = 26.3), 43.06. MS (EI) m/z: 460.0 (M+).
HPLC tR = 5.36 min (water/methanol (20:80), purity 98.74%).
solved in 160 ml DMSO containing 32 ml HCl. The mixture was
stirred for 2 h at 60 °C then poured into 900 ml icy water and fil-
trated under vacuum to give crude product which was recrystal-
lized by DMSO and water to give 1.35 g of 14. White solid (66%).
1H NMR (400 MHz, DMSO-d6) d 9.69 (s, 2H, OH), 7.72 (d, J = 8.6,
2H, ArH), 7.18 (d, J = 2.6, 2H, ArH), 7.04 (dd, J1 = 8.6, J2 = 2.6, 2H,
ArH), 3.93 (s, 4H, CH2). 13C NMR (100 MHz, DMSO-d6) d 166.40,
156.05, 128.67, 128.16, 126.60, 120.56, 112.46, 43.24. HRMS Calcd
for
C
16H12N2O4Se2: 455.9128. Found 456.9196 (M++H). HPLC
4.2.5.8. 1,5-(4,40-Difluoro(bis-1,2-benzisoselenazol-3(2H)-one))
tR = 4.78 min (water/methanol (40:60), purity 98.39%).
pentane (10b).
Compound 3g (1.36 g, 5 mmol) and 1,5-diam-
inopentane (0.27 g, 2.6 mmol) were used as reactants to give 0.97 g
of 10b as yellow solid (77%). 1H NMR (400 MHz, DMSO-d6) d 7.83
(dd, J1 = 8.5, J2 = 5.6, 2H, ArH), 7.69 (dd, J1 = 9.2, J2 = 2.2, 2H, ArH),
7.21 (td, J1 = 8.5, J2 = 2.2, 2H, ArH), 3.68 (t, J = 6.9, 4H, CH2), 1.65
(quint, J = 6.9, 4H, CH2), 1.32 (quint, J = 6.9, 2H, CH2). 13C NMR
(100 MHz, DMSO-d6) d 165.32, 163.87 (d, JC–F = 248.8), 141.11 (d,
JC–F = 11.1), 129.33 (d, JC–F = 10.2), 124.83, 114.01 (d, JC–F = 23.37),
112.32 (d, JC–F = 26.95), 43.07, 29.45, 23.20. MS (EI) m/z: 502.0
(M+). HPLC tR = 5.38 min (water/methanol (20:80), purity 97.00%).
4.3. Biology
TrxR of rat liver was prepared and purified according to the
method of Luthman and Holmgren.27 Trx of Escherichia coli, GR
and DTNB were purchased from Sigma–Aldrich. BIAM was pur-
chased from Invitrogen.
4.3.1. Inhibition of TrxR in vitro
DTNB assay was used to determine the TrxR inhibitory activity
of the compounds tested. All assays were conducted at room tem-
perature in 96-well microplates with a total volume of 80
Twenty microliter of TrxR (10 g) was added to 50 l assay solu-
tion containing 40 l of 100 mM potassium phosphate, 2 mM
EDTA, pH 7.4 and 10 l of 100 M NADPH. Then 10 l of inhibitors
at various concentrations were added to each well followed by a
five minute pre-incubation period. The same amounts of DMSO
were added to the control experiments. The reaction was initiated
4.2.5.9. 1,2-(4,40-Dichloro(bis-1,2-benzisoselenazol-3(2H)-one))
ll.
ethane (11a).
Compound 3h (1.44 g, 5 mmol) and 1,2-diami-
l
l
noethane (0.16 g, 2.6 mmol) were used as reactants to give 0.92 g
of 11a as yellow solid (75%). 1H NMR (400 MHz, DMSO-d6) d 8.06
(d, J = 1.8, 2H, ArH), 7.75 (d, J = 8.3, 2H, ArH), 7.44 (dd, J1 = 8.3,
J2 = 1.8, 2H, ArH), 3.69 (s, 4H, CH2). 13C NMR (100 MHz, DMSO-
d6) d 170.62, 150.30, 135.11, 133.90, 130.88, 127.74, 121.59,
44.01. MS (EI) m/z: 492.0 (M+). HPLC tR = 7.83 min (water/methanol
(20:80), purity 96.52%).
l
l
l
l
by the addition of 100 ll DTNB and TrxR activities were measured
by the increase of absorbance at 412 nm over the initial 120 s.
4.2.5.10.
one))pentane (11b).
1,5-(4,40-Dichloro(bis-1,2-benzisoselenazol-3(2H)-
Compound 3h (1.44 g, 5 mmol) and
4.3.2. Cell culture and growth inhibition assay
Human cancer cell lines U87 MG (glioma), MIA PaCa-2 (pancre-
atic carcinoma), A549 (lung cancer) were cultured in DMEM med-
ium and LoVo (colon cancer) was cultured in RPMI 1640 medium
supplemented with 10% fetal bovine serum, 100 U/ml penicillin G
and 100U/ml streptomycin in a humidified 5% CO2 atmosphere at
37 °C. Exponentially growing cells were then detached from the
cell culture to form a single cell suspension. Cells were then plated
into 96-well plates with 4000–5000 cells per well. Compounds dis-
1,5-diaminopentane (0.27 g, 2.6 mmol) were used as reactants to
give 1.03 g of 11b as yellow solid (77%). 1H NMR (400 MHz,
DMSO-d6) d 8.07 (s, 2H ArH), 7.75 (d, J = 7.8, 2H, ArH), 7.44 (d,
J = 7.8, 2H, ArH), 3.68 (t, J = 6.8, 4H, CH2), 1.65 (quint, J = 6.8, 4H,
CH2), 1.31 (quint, J = 6.8, 2H, CH2). 13C NMR (100 MHz, DMSO-d6)
d 171.33, 150.87, 135.55, 133.51, 131.23, 126.91, 120.32, 43.56,
33.20, 26.51. MS (EI) m/z: 533.9 (M+). HPLC tR = 8.05 min (water/
methanol (20:80), purity 97.11%).
solved in DMSO at different concentrations (0–50 lM) were added
and incubated for 48 h. The effects of selected compounds on the
growth of human cancer cell lines were determined by MTT assay
as described.23
4.2.5.11.
one))ethane (12).
1,2-(5,50-Dimethoxy(bis-1,2-benzisoselenazol-3(2H)-
Compound 3j (1.42 g, 5 mmol) and 1,2-
diaminoethane (0.16 g, 2.6 mmol) were used as reactants to give
0.88 g of 12 as yellow solid (73%). 1H NMR (400 MHz, DMSO-d6)
d 7.82 (d, J = 8.8, 2H, ArH), 7.29 (d, J = 2.7, 2H, ArH), 7.20 (dd,
J1 = 8.8, J2 = 2.7, 2H, ArH), 3.98 (s, 4H, CH2), 3.80 (s, 6H, OCH3).
13C NMR (100 MHz, DMSO-d6) d 166.36, 158.14, 130.49, 128.66,
127.00, 120.49, 109.77, 55.50, 43.51. HRMS Calcd for
4.3.3. BIAM assay
The BIAM assay was conducted according to the literature8
using
(1.1
arately at room temperature for 2 h. The control was incubated
with the same amount of DMSO. Next, 2 l of the reaction mixtures
were taken out and added to new tubes containing 100 M BIAM
(pH 6.5 and 8.5) following incubation at 37 °C for 30 min. 40
of BIAM-modified enzymes were mixed with 40 l loading buffer,
then 40 l of the samples were subjected to SDS-PAGE on a 8.0%
a
minor modification. Generally, NADPH-reduced TrxR
l
M) and 12 with different concentrations were incubated sep-
C
18H16N2O4Se2: 483.9441. Found 484.9558 (M++H). HPLC
tR = 4.81 min (water/methanol (20:80), purity 96.98%).
l
l
ll
4.2.5.12.
one))ethane (13).
1,2-(5,50-Diacetoxy(bis-1,2-benzisoselenazol-3(2H)-
Compound 3k (1.56 g, 5 mmol) and 1,2-
l
l
diaminoethane (0.16 g, 2.7 mmol) were used as reactants to give
1.19 g of 13 as white solid (88%). 1H NMR (400 MHz, DMSO-d6) d
7.97 (d, J = 8.6, 2H, ArH), 7.55 (d, J = 2.5, 2H, ArH), 7.35 (dd,
J1 = 8.6, J2 = 2.5, 2H, ArH), 4.01 (s, 4H, CH2), 2.28 (s, 6H, CH3). 13C
gel, and the separated proteins were transferred to nitro cellulose
membrane. Proteins labeled with BIAM were detected with horse-
radish peroxidase-conjugated streptavidin and enhanced chemilu-
minescence detection.

J. He et al. / Bioorg. Med. Chem. 20 (2012) 3816–3827
3827
4.3.4. Cellular TrxR inhibition assay
with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.bmc.2012.04.033.
U87 MG cells were incubated with different concentrations of
selected compounds for 12 h. The same amount of DMSO (1%, v/v)
was added to the control group. Then cells were washed with phos-
phate-buffered saline and lysed with cell lysis buffer (1% Triton X-
100, 1% sodium deoxycholate, 10 mM EDTA, 0.1% SDS and 1 mM
phenylmethylsulfonyl fluoride) and centrifuged at 13000g for
30 min at 4 °C. Protein concentration was determined by BCA meth-
od.38 To determine the cellular TrxR activity, insulin-reducing assay
was employed as shown in our previous study.21 with minor mod-
References and notes
1. Arnér, E. S. J.; Holmgren, A. Eur. J. Biochem. 2000, 267, 6102.
2. Nakamura, H. Antioxid. Redox Signal. 2005, 7, 823.
3. Sandalova, T.; Zhong, L.; Lindqvist, Y.; Holmgren, A.; Schneider, G. Proc. Natl.
Acad. Sci. U.S.A. 2001, 98, 9533.
4. Zhong, L.; Arnér, E. S. J.; Holmgren, A. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 5854.
5. Lillig, C. H.; Holmgren, A. Antioxid. Redox. Sign. 2007, 9, 25.
6. Grogan, T. M.; Fenoglio-Prieser, C.; Zeheb, R.; Bellamy, W.; Frutiger, Y.; Vela, E.;
Stemmerman, G.; Macdonald, J.; Richter, L.; Gallegos, A.; Powis, G. Hum. Pathol.
2000, 31, 475.
7. Smart, D. K.; Ortiz, K. L.; Mattson, D.; Bradbury, C. M.; Bisht, K. S.; Sieck, L. K.;
Brechbiel, M. W.; Gius, D. Cancer Res. 2004, 64, 6716.
8. Fang, J.; Lu, J.; Holmgren, A. J. Biol. Chem. 2005, 280, 25284.
9. Arnér, E. S. J. Biochim. Biophys. Acta. 2009, 1790, 495.
10. Krishnamurthy, D.; Karver, M. R.; Fiorillo, E.; Stanford, S. M.; Bottini, N.; Barrios,
A. M. J. Med. Chem. 2008, 51, 4790.
11. Rubbiani, R.; Kitanovic, I.; Alborzinia, H.; Can, S.; Kitanovic, A.; Onambele, L. A.;
Stefanopoulou, M.; Geldmacher, Y.; Sheldrick, W. S.; Wolber, G.; Prokop, A.;
Ott, I. J. Med. Chem. 2010, 53, 8608.
ification. Briefly, 40
with reaction buffer (0.2 M HEPES, pH 7.6, 5 mM EDTA, 1 mM
NADPH, 1 M E. coli Trx and 2.5 mg/ml bovine insulin) for half an
hour at 37 °C. Reactions were then quenched by the sequential
addition of 120 l of 6 M guanidine-HCl, 50 mM Tris (pH 8.0), and
lg of protein in the cell extracts were incubated
l
l
0.6 mg/ml DTNB. The amount of free thiols generated from insulin
reduction was measured by recording the absorbance at 412 nm.
For each test sample, the corresponding control was measured by
the same assay but without the addition of E. coli Trx. TrxR activity
was calculated after subtracting the control value.
12. Lo, Y. C.; Ko, T. P.; Su, W. C.; Su, T. L.; Wang, A. H. J. J. Inorg. Biochem. 2009, 103,
1082.
13. Casini, A.; Gabbiani, C.; Sorrentino, F.; Rigobello, M. P.; Bindoli, A.; Geldbach, T.
J.; Marrone, A.; Re, N.; Hartinger, C. G.; Dyson, P. J.; Messori, L. J. Med. Chem.
2008, 51, 6773.
4.4. Docking Study
14. Qiu, X.; Liu, Z.; Shao, W. Y.; Liu, X.; Jing, D. P.; Yu, Y. J.; An, L. K.; Huang, S. L.; Bu,
X. Z.; Huang, Z. S.; Gu, L. Q. Bioorg. Med. Chem. 2008, 16, 8035.
15. Zhou, C.; Boggess, J. F.; Bae-Jump, V.; Gehrig, P. A. Gynecol. Oncol. 2007, 105,
218.
16. Cenas, N.; Nivinskas, H.; Anusevicius, Z.; Sarlauskas, J.; Lederer, F.; Arnér, E. S. J.
J. Biol. Chem. 2004, 279, 2583.
17. Mehta, M. P.; Shapiro, W. R.; Phan, S. C.; Gervais, R.; Carrie, C.; Chabot, P.;
Patchell, R. A.; Glantz, M. J.; Recht, L.; Langer, C.; Sur, R. K.; Roa, W. H.; Mahe, M.
A.; Fortin, A.; Nieder, C.; Meyers, C. A.; Smith, J. A.; Miller, R. A.; Renschler, M. F.
Int. J. Radiat. Oncol. 2009, 73, 1069.
18. Ramanathan, R.; Abbruzzese, J.; Dragovich, T.; Kirkpatrick, L.; Guillen, J.; Baker,
A.; Pestano, L.; Green, S.; Von Hoff, D. Cancer Chemother. Pharmacol. 2011, 67,
503.
19. Hashemy, S. I.; Ungerstedt, J. S.; Avval, F. Z.; Holmgren, A. J. Biol. Chem. 2006,
281, 10691.
20. Baker, A. F.; Dragovich, T.; Tate, W. R.; Ramanathan, R. K.; Roe, D.; Hsu, C.-H.;
Kirkpatrick, D. L.; Powis, G. J. Lab. Clin. Med. 2006, 147, 83.
21. Zhao, F.; Yan, J.; Deng, S.; Lan, L.; He, F.; Kuang, B.; Zeng, H. Cancer Lett. 2006,
236, 46.
22. Xing, F.; Li, S.; Ge, X.; Wang, C.; Zeng, H.; Li, D.; Dong, L. Oral Oncol. 2008, 44,
963.
23. Yin, H.; Li, J.; Xiong, K.; Wang, L.; Wang, T.; Tan, Q.; Fu, J.; Ren, X.; Zeng, H.
Differentiation 2011, 81, 49.
24. Wang, L.; Fu, J.; Wang, J.; Jin, C.; Ren, X.; Tan, Q.; Li, J.; Yin, H.; Xiong, K.; Wang,
T.; Liu, X.; Zeng, H. Anti-Cancer Drugs 2011, 22, 732.
25. Ninomiya, M.; Garud, D. R.; Koketsu, M. Coordin. Chem., Rev. 2011, 255, 2968.
26. Wang, L.; Yang, Z.; Fu, J.; Yin, H.; Xiong, K.; Tan, Q.; Jin, H.; Li, J.; Wang, T.; Tang,
W.; Yin, J.; Cai, G.; Liu, M.; Kehr, S.; Becker, K.; Zeng, H. Free Radical Biol. Med.
2012, 52, 898.
27. Luthman, M.; Holmgren, A. Biochemistry 1982, 21, 6628.
28. Sakurai, T.; Kanayama, M.; Shibata, T.; Itoh, K.; Kobayashi, A.; Yamamoto, M.;
Uchida, K. Chem. Res. Toxicol. 2006, 19, 1196.
29. Schöneich, C.; Narayanaswami, V.; Asmus, K. D.; Sies, H. Arch. Biochem. Biophys.
1990, 282, 18.
Structure of human thioredoxin reductase (PDB code: 2J3N) ob-
tained from the Brookhaven Protein Database (www.rcsb.org/pdb)
was applied as receptor model for the docking procedure in the
Discovery Studio 2.5 software (Accelrys). Co-crystallized water
was removed and hydrogen atoms were added. The receptor pro-
tein was then typed by applying the CHARMm forcefield before fol-
lowing a three-step minimization protocol (steepest descent,
conjugate gradient and adopted basis Newton–Rapheson), while
the ligand to be docked was first drawn by ChemDraw software
and then was prepared for the docking simulation by employing
the Prepare Ligands application in Discovery Studio 2.5. The pre-
pared ligand was docked into the active site of TrxR by means of
GOLD 3.0.1.39,40 The site sphere was defined as 15 Å with selection
of Cys 497, Sec 498, Cys 59 and Cys 64 (redox-active centers in C-
terminal and N-terminal respectively). In an attempt to stimulate
the formation of a covalent bond between two selenium atoms
of 12 and Cys497 and Sec498, respectively, two distance con-
straints of the two pairs of atoms were made: one was applied as
2.11 Å (minimum) to 2.57 Å (maximum) around an average Se–
Se distance35 of 2.34 Å between one selenium atom in the inhibitor
and the selenium atom of Sec498; another was applied as 2.70 Å
(minimum) to 3.30 Å (maximum) between the other selenium
atom in the inhibitor and the sulfur atom of Cys497 to allow their
interactions.3 Parameters were kept as default values and 10 ge-
netic algorithm runs were applied in terms of GoldScore fitness
function.
30. Kim, J. R.; Yoon, H. W.; Kwon, K. S.; Lee, S. R.; Rhee, S. G. Anal. Biochem. 2000,
283, 214.
31. Stadtman, T. C. Ann. Rev. Biochem. 1996, 65, 83.
32. Besse, D.; Siedler, F.; Diercks, T.; Kessler, H.; Moroder, L. Angew. Chem., Int. Ed.
1997, 36, 883.
Acknowledgments
We thank the comprehensive center for drug discovery and
development at Peking University Health Science Center for
technical assistance in molecular docking studies. This work is
supported by China Postdoctoral Science Foundation (No.
2011M500526).
33. Fang, J.; Holmgren, A. J. Am. Chem. Soc. 1879, 2006, 128.
34. Fritz-Wolf, K.; Urig, S.; Becker, K. J. Mol. Biol. 2007, 370, 116.
35. Allen, F. H.; Kennard, O.; Watson, D. G. J. Chem. Soc., Perkin. Trans. II 1987, S1.
36. Gromer, S.; Wessjohann, L. A.; Eubel, J.; Brandt, W. ChemBioChem 2006, 7,
1649.
37. Cheng, Q.; Sandalova, T.; Lindqvist, Y.; Arnér, E. S. J. J. Biol. Chem. 2009, 284,
3998.
38. Morton, R. E.; Evans, T. A. Anal. Biochem. 1992, 204, 332.
39. Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. J. Mol. Biol. 1997, 267,
727.
Supplementary data
40. Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D. Proteins
Struct., Funct., Bioinf. 2003, 52, 609.
Supplementary data (synthetic procedures and analytical data
for 15 and 16; inhibitory effects of 12, 4g, 13 on GR.) associated