Synthesis, toxicities and bio-activities of manganese complexes with CO and H2S dual donors

Article abstract of DOI:10.1016/j.ejmech.2018.10.004

A series of H₂S–CO dual-donors [Mn(CO)₄CS₂NR₁R₂] was synthesized, and evaluated from toxicity and bioactivity. The CO–H₂S measuring test showed all the complexes not only released CO, but released H₂S. The resulting data of cytotoxicity showed all the complexes had activities against the cell proliferation; among them, complexes 1, 2 and 7 displayed higher activities than the others, and their potencies were close to cis-platinum (DDP); whereas the precursors A₁-A₂₂ had almost no activities against all five tumor cell lines and W138 cell line proliferation. It is worth noting that complex 1 displayed the highest activity to MCF-7, complex 2 displayed the highest activity to HePG2, and complex 7 showed selectivity inhibition to both A549 and HeLa. The developmental toxicities of the complex were assessed using zebrafish embryos. The results showed complexes 1 and 2 had effect on the mortality and hatching rate of zebrafish embryos in dose-dependent manner. They caused zebrafish malformations when they were over 10 μM. Meanwhile, they displayed dose-dependent toxicities to larval zebrafish. In the test of bio-activities, complexes 1 and 2 had strong anti-inflammatory activities; they not only down-regulated the expression levels of iNOS and TNF-α up-regulated the expression of HO-1 and IL-10, but also up-regulated COX-2 levels. In contrast, the precursor compound (A₁ or A₂) displayed lower anti-inflammatory activity than the corresponding complex, which suggests both the CO and H₂S from the complex took synergistic effects in the process of anti-inflammation. In addition, the complex showed antihypertensive effect and myocardial protection. This effect also possibly resulted from this synergistic effect. All these suggest the complexes have potential to be candidate medicines.

Full text of DOI:10.1016/j.ejmech.2018.10.004

European Journal of Medicinal Chemistry 159 (2018) 339e356
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Synthesis, toxicities and bio-activities of manganese complexes with
CO and H2S dual donors
,
*
Zhongjie Bai a, Jinlong Zhang a, Qiuping Zhang a, Taofeng Zhang a, Jili Li a, Quanyi Zhao a
,
,
Zhen Wang a **, Dian He a, Jie Cheng b, Jingke Zhang b, Bin Liu c
a Institute of Medicinal Chemistry, School of Pharmacy of Lanzhou University, China
b GLP Lab Centre, School of Basic Medicine of Lanzhou University, China
c School of Stomatology of Lanzhou University; Lanzhou, 730000, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of H2SeCO dual-donors [Mn(CO)4CS2NR1R2] was synthesized, and evaluated from toxicity and
bioactivity. The COeH2S measuring test showed all the complexes not only released CO, but released H2S.
The resulting data of cytotoxicity showed all the complexes had activities against the cell proliferation;
among them, complexes 1, 2 and 7 displayed higher activities than the others, and their potencies were
close to cis-platinum (DDP); whereas the precursors A1-A22 had almost no activities against all five tumor
cell lines and W138 cell line proliferation. It is worth noting that complex 1 displayed the highest activity
to MCF-7, complex 2 displayed the highest activity to HePG2, and complex 7 showed selectivity inhi-
bition to both A549 and HeLa. The developmental toxicities of the complex were assessed using zebrafish
embryos. The results showed complexes 1 and 2 had effect on the mortality and hatching rate of
zebrafish embryos in dose-dependent manner. They caused zebrafish malformations when they were
Received 21 April 2018
Received in revised form
29 September 2018
Accepted 1 October 2018
Available online 4 October 2018
Keywords:
CO and H2S dual-donors
Manganese
Toxicity
over 10
activities, complexes 1 and 2 had strong anti-inflammatory activities; they not only down-regulated
the expression levels of iNOS and TNF- , up-regulated the expression of HO-1 and IL-10, but also up-
mM. Meanwhile, they displayed dose-dependent toxicities to larval zebrafish. In the test of bio-
Anti-inflammatory
Anti-hypertension
Myocardial protection
a
regulated COX-2 levels. In contrast, the precursor compound (A1 or A2) displayed lower anti-
inflammatory activity than the corresponding complex, which suggests both the CO and H2S from the
complex took synergistic effects in the process of anti-inflammation. In addition, the complex showed
antihypertensive effect and myocardial protection. This effect also possibly resulted from this synergistic
effect. All these suggest the complexes have potential to be candidate medicines.
© 2018 Elsevier Masson SAS. All rights reserved.
1. Introduction
H2S significantly reduced the accumulation of lipopolysaccharide
induced neutrophils in the tissues and organs (liver and lung),
Hydrogen sulfide (H2S) is a gas neurotransmitter like CO and
reduced tumor necrosis factor-
a (TNF-a) and plasma interleukin-1
mainly produced by the metabolism of sulfur-containing amino
(IL-1 ) levels, and increased plasma IL-10 levels [9].
b
acids, such as cysteine, catalyzed by cystathionine-
g
-lyase and
H2S is beneficial for regulation of physiological function, but it is
not used directly in clinic because of uncontrollable dose and high
toxicity. H2S-donor is considered as an effective way to take the
place of gaseous H2S to use. Presently, there have many potential
donors to be synthesized and evaluated preliminarily. Among
them, NSAID-modified H2S donors displayed good anti-
inflammatory activity and less gastrointestinal side effects [10].
NO and H2S dual donors which reported by Kashfi exhibited
stronger anti-inflammatory effects than aspirin [11]. Dual donors
possibly have more potential application in the future. However,
the studies about COeH2S dual donors are not reported until now,
there still has no data to comprehensively assess their possibly
cystathionine- -synthase [1e4]. H2S has an important role to relax
b
vascular smooth muscle [5]. H2S and its precursors were potent
inducers of the antioxidant and heme oxygenase-1 (HO-1) [6,7]. In
the inflammatory response, H2S exhibited the same beneficial ef-
fects as CO on the inflammatory response, which inhibited induc-
ible NO synthase and further reduced NO production [8]. Moreover,
* Corresponding author.
** Corresponding author.
E-mail addresses: zhaoqy@lzu.edu.cn (Q. Zhao), zhenw@lzu.edu.cn (Z. Wang).
https://doi.org/10.1016/j.ejmech.2018.10.004
0223-5234/© 2018 Elsevier Masson SAS. All rights reserved.

340
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
clinic application.
2.2. H2S and CO releasing tests
Therefore, based on our previous works, we synthesized a series
of H2S-COdual donors based on our previous works, and assessed
their H2S and CO releasing abilities, the cytotoxicities, as well as
their bioactivities, including anti-inflammatory activity, anti-
hypertension and myocardial protection activities. Meanwhile, we
tried to study the synergistic effects between the H2S and CO
released from the complex. By which, we hope to obtain some in-
formation about whether these dual donors fit for clinical
application.
The ability of CO-releasing was evaluated in a myoglobin (Mb)-
based aqueous assay as we did before [12,13]. As seen from Table 2
and Fig. 2, all the complexes were effective CO-releaser. Each
complex CO-releasing curve was similar to those complexes which
we reported before [12]. The CO-releasing half-lives of complex 2
and 6 were shorter than the others, only 5.6 min and 9.6 min when
they were 14 mM, respectively.
H2S-releasing ability of the complexes, the methylene blue
(MBþ) method was used to measure H2S amounts [14]. The MBþ
method is based on the reaction of zinc acetate with H2S to form
zinc sulfide, which reacting with N, N-dimethyl-1, 4-phenyl ene-
diamine sulfate to generate methylene blue in the presence of ferric
ions under acid medium condition. The amounts of H2S released
from the compounds were calculated according to the Na2S stan-
dard curve (SIeFig. 2). The testing results demonstrate all the
compounds were much slower H2S-releasing molecules, and even
some of them did not release H2S at all in PBS at room temperature.
However, in the presence of TCEP (a water-soluble phosphine
reducing agent, Hasegawa and van der Vlies, 2014), all the com-
pounds released H2S at 37.5 ꢁC [15].
2. Results and discussion
2.1. Syntheses and properties
The synthesis of complexes was completed in two steps, the first
was the synthesis of sodium dithiocarbamate (A1-22, precursor)
using amines and excess CS2, then sodium dithiocarbamate reacted
with Mn(CO)5Br to afford the target compounds (Schemes 1 and 2).
The final product was obtained by recrystallization three times
with THF and CYH. All the complexes showed a number of strong
terminal carbonyl stretches absorption bands at 2092e1928 cmꢀ1
in their IR spectra. The higher absorption bands were attributed to
terminal carbonyls coordinated to the Mn atom. In the 1H NMR
spectra of complexes 17e22, the benzene ring H signals appeared at
6.82e7.59 ppm.
In addition, the crystal structure of complex 3, complex 5 and
complex 16 were determined by single-crystal X-ray diffraction.
The molecular structures are shown in Fig. 1. The details of the data
collection and structure refinement are listed in Table 1.
All the complexes were yellow solid, and they were stable in air
for several days and had no effect to low intensity light. No
decomposition of the solid materials was observed as long as they
are stored in the dark. All the complexes were soluble in organic
solvents, like DMSO, MeOH and THF. Their sodium salts were sol-
uble and stable in water for several days.
H2S-release Cmax and H2S-release half-lives at 60 mM are shown
in Table 2, and the typical process of H2S-release in Fig. 2. Seen from
the data, all the precursors and the complexes were all H2S donors;
for most of the complexes, the Cmax was less than those from cor-
responding precursors under the same condition. Notably, for
complex 3 and A3, both Cmax had a big difference with 7.6
mM and
15.3 M, and both H2S-releasing half-lives were 6.8min and
m
21.3min, respectively. For other complexes, most H2S-release half-
lives were shorter than those of the corresponding precursors.
This suggests both sulfide atoms interaction with Mn atom made
H2S release more easily from complexes. However, a small number
of the complexes released H2S faster than the corresponding pre-
cursors, such as complexes 19e22, this maybe resulted from strong
interaction with TCEP, which was a promoter of H2S release.
Though methylene blue (MBþ) method to measure H2S has its
defects, and the resulting data possibly did not represent H2S levels
per se, it still reflected their H2S donor behavior.
To obtain sodium salt of complexes 1e2, Na2CO3 and NaHCO3
were added till the complexes were dissolved completely and the
pH of the solution was about 7. They were stable under acidic or
basic condition, which we had established [12].
In summary, all the complexes not only released CO, but also H2S
Scheme 1. Structures and synthetic route of complexes 1e6.

Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
341
Scheme 2. Structures and synthetic route of complexes 7e22.
Fig. 1. The structures of complexes: A for 3, B for 5, C for 16.

342
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
Table 1
Details of the data collection and structures refinement.
Complex
3
5
16
Formula
C9H8MnNO4S2
313.22
monoclinic
0.23 ꢂ 0.15 ꢂ 0.14
P21/n
C11H12MnNO4S2
341.28
monoclinic
0.21 ꢂ 0.15 ꢂ 0.14
C2/c
C13H10MnNO4S2
363.28
monoclinic
0.12 ꢂ 0.04 ꢂ 0.02
P21/c
Formula weight
Crystal system
Crystal size (mm3)
Space group
a(Å)
14.0410(11)
6.1761(4)
22.6646(14)
6.3236(3)
10.9053(10)
22.628(2)
b(Å)
c(Å)
14.3053(11)
1229.61(15)
4
20.8232(13)
2973.8(3)
8
6.1803(5)
1521.1(2)
4
Volume(Å3)
Z
rcalcg/cm3
1.692
1.413
1.525
1.175
1.586
1.154
m
(mmꢀ1
)
F(000)
range for data collection/ꢁ
Reflections collected
632.0
7.196 to 52.044
4639
1392.0
6.7 to 52.04
5831
736.0
6.574 to 52.044
6046
2
Q
Independent reflections
Data/restraints/parameters
2424 [Rint ¼ 0.0503]
2424/0/154
R1 ¼ 0.0500, wR2 ¼ 0.1209
R1 ¼ 0.0641, wR2 ¼ 0.1339
0.58/-0.80
2927 [Rint ¼ 0.0372]
2927/0/173
R1 ¼ 0.0441, wR2 ¼ 0.0688
R1 ¼ 0.0667, wR2 ¼ 0.0818
0.36/-0.29
2988 [Rint ¼ 0.0466]
2988/0/191
R1 ¼ 0.0451, wR2 ¼ 0.0707
R1 ¼ 0.0701, wR2 ¼ 0.0839
0.42/-0.36
Final R-indices [I ꢃ 2
s(I)]
Final R indices [all data]
Largest diff.peak and hole (e Åꢀ3
)
Table 2
CO and H2S releasing half-lives.
Complex
t1/2 (min)
Complex
Cmax of H2S(
m
M)
t'1/2(min)
Precursor compds
Cmax of H2S(
m
M)
t'1/2(min)
1
2
3
4
5
6
7
8
14.9
5.6
21.9
18.4
13.6
9.6
1
2
3
4
5
6
7
8
11.4
10.2
7.6
12.1
10.8
9.3
10.0
8.9
9.2
9.1
9.4
7.6
7.9
8.2
9.5
8.7
10.1
8.5
11.2
10.8
8.7
19.3
13.8
6.8
A1
A2
A3
A4
A5
A6
A7
A8
13.2
10.3
15.3
12.8
11.6
10.1
10.5
10.6
11.8
13.9
11.7
12.4
12.1
11.9
10.6
13.5
14.7
10.2
14.1
12.8
10.2
11.4
25.6
17.8
21.3
22.1
25.2
21.9
16.4
15.8
18.2
23.0
19.1
15.6
26.1
22.8
23.0
18.9
19.2
21.9
16.3
17.6
16.2
15.8
16.1
22.9
19.6
16.3
20.2
21.3
11.3
21.6
21.8
22.4
28.6
17.9
16.2
14.7
20.5
25.8
21.6
22.8
19.1
14.5
16.7
18.8
15.4
10.3
21.1
18.6
15.9
18.4
14.8
15.7
27.4
15.6
16.7
11.3
10.4
9
9
A9
10
11
12
13
14
15
16
17
18
19
20
21
22
10
11
12
13
14
15
16
17
18
19
20
21
22
A10
A11
A12
A13
A14
A15
A16
A17
A18
A19
A20
A21
A22
9.0
at the same time, they are CO and H2S dual releasers; however, H2S
and CO did not take effects at the same time because the half-lives
of H2S-release and CO-release were different. For CO and H2S,
which one releases first from the complex in vivo? We speculate CO
first to release through ligand exchange reaction, and not until the
fragments containing SeMn arrive at the place where pH is low or
GSH rich, H2S begins to release.
which the potencies were close to DDP. Complex 1 displayed the
higher inhibiting activity to HepG2 and MCF-7, which indicates
complex 1 had selectivity to inhibit proliferation of HepG2 and
MCF-7 cell lines. Similarly, complex 2 showed selectivity to HepG2,
and complex 7 to A549 and HeLa. Compared with
their se-
WI38,
lective index was up to 2.39. Meanwhile, we found complex 1
induced HeLa cell early apoptosis, and its apoptotic effect was dose-
dependent. Meanwhile, complex 1 showed a high activity to arrest
t
1/2: CO releasing half-lives when complex was at 14 mM; t’1/2:
H2S releasing half-lives when complex at 60
mM.
cell cycle, and 40 mM of which led to the G2/M fraction increasing
by 12.63% [12]. This indicates that the active complexes have po-
tential to be as drug candidates.
2.3. Cytotoxicities
The cytotoxicities of all the complexes were assayed by MTT
method using DDP as a positive control. The resulting data are
shown in Tables 3e4. Precursors A1-A22 almost displayed no any
toxicities to all five tumor cell lines and normal cell line (all
2.4. Toxicity of the complexes on zebrafish
2.4.1. Complex-induced mortality and hatchability of zebrafish
In order to evaluate the development toxicities, we chose
zebrafish as test subjects. In the presence of tested complexes,
zebrafish mortalities were measured at 24, 48, 72, 96 and 120 hpf
respectively. The data are shown in Fig. 3A-B. We can see when the
IC50 > 500
exhibited higher activity to inhibit tumor cell proliferation than the
others, and their IC50 values were in the range of 26.9e51.7 M,
mM); among the complexes, complexes 1, 2 and 7
m

Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
343
Fig. 2. CO and H2S-releasing curve of the complex vs time: a and e for complex 3; b and f for: complex 7; c and g for complex 10; d and h for complex 16; Each bar represents the
mean SD of four independent experiments.
complexes were 10
m
M below they had little effect on the survival
same conditions. This suggests complex 2 was lower toxicity than
complex 1.
rate of zebrafish at 24 hpf. But with the increase of incubation time,
the survival rates decreased; and with the increase of the concen-
tration, the mortalities increased. Distinctly, the effects of the
complexes on the mortalities of zebrafish were dose-dependent.
When complexes 1 and 2 were less than 10
toxic, and the survival rate of zebrafish was not less than 60% at 120
hpf (60%, 78% respectively). But when complex 1 was 20 M, the
A critical period of zebrafish embryonic development is hatch,
and the normal zebrafish embryos have a hatching period from 48
to 72 hpf. In the process of hatching, the complexes possibly harm
embryo development through inhibiting key enzymes which pro-
mote embryo to develop. Seen from Fig. 3C, when the embryos
were treated with complexes 1 and 2, the hatching rates of embryos
decreased with the concentration increasing compared with the
mM, both were less
m
mortality was 52% at 120 hpf, which was 6.5 times as high as that of
the control group (8.0%). As for complex 2, it was 38% under the
control group at 72 hpf; when they were at below 5.0 mM, the

344
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
Table 3
IC50 values of compounds A1-22 against cell proliferation.
Compound.
IC50
(
mM)
HeLa
A549
HT-29
HepG2
MCF-7
WI38
A1
A2
A3
A4
A5
A6
A7
A8
1234.94 2.9
1041.01 10.3
826.16 9.2
1260.46 8.4
857.13 2.8
1123.58 5.2
1026.72 7.2
783.72 2.9
642.06 1.9
746.13 4.5
599.48 9.4
676.43 3.3
842.57 3.6
573.46 2.4
731.68 6.2
1294.91 3.6
741.08 8.3
816.65 5.2
694.34 2.6
681.24 4.4
558.49 7.4
1106.36 8.9
920.89 8.2
901.16 3.4
715.54 2.5
978.21 3.7
743.84 6.4
881.31 4.3
1331.56 11.5
863.76 3.8
713.28 3.5
639.91 2.6
729.54 3.2
581.67 7.2
901.29 7.4
662.33 4.5
865.43 3.5
890.86 6.4
801.56 5.8
915.84 3.9
720.95 5.3
736.17 6.4
721.49 3.2
931.74 9.1
941.42 3.1
911.63 5.4
772.54 2.8
899.65 9.8
812.14 5.8
947.74 3.9
966.53 4.3
672.76 6.1
784.63 2.1
675.18 3.4
687.06 4.6
669.44 8.4
753.65 5.6
832.11 2.7
779.18 3.8
918.45 2.5
941.62 6.1
837.51 7.6
841.45 3.7
869.05 5.3
649.84 2.8
958.52 5.1
961.66 4.5
876.42 7.2
649.84 3.4
1011.67 6.3
699.42 9.1
912.48 8.6
865.29 6.6
725.61 4.4
819.20 6.7
582.33 7.2
783.64 6.2
734.52 2.9
698.32 5.2
629.15 7.4
804.56 2.6
1065.43 9.8
838.43 9.2
1219.83 4.5
665.82 2.8
615.22 2.6
691.75 4.6
892.48 4.2
1043.86 2.7
962.44 2.8
784.63 5.1
859.22 5.4
792.35 2.7
983.12 7.9
942.66 7.3
824.19 4.8
688.57 7.8
613.96 4.9
622.41 3.8
593.87 3.6
781.25 2.8
598.63 5.2
697.61 6.4
942.59 7.8
867.96 3.1
1004.16 5.2
734.57 3.6
798.21 3.1
598.22 5.7
904.16 6.1
881.67 6.2
856.13 3.6
685.84 7.4
782.84 3.5
665.18 4.1
876.73 6.3
902.38 8.4
647.82 3.9
652.56 5.2
568.42 2.7
668.39 4.9
522.51 6.8
645.37 4.2
523.44 4.5
684.27 2.9
901.65 6.4
810.23 2.8
785.86 7.2
720.07 6.5
634.67 3.9
526.85 2.4
685.54 3.1
A9
A10
A11
A12
A13
A14
A15
A16
A17
A18
A19
A20
A21
A22
Table 4
IC50 values of the complexes against cell proliferation.
Complex
IC50
(m
M)
HeLa
A549
HT-29
HepG2
MCF-7
W138
1
2
3
4
5
6
7
8
35.70 2.2
46.77 1.9
273.11 10.1
153.28 5.2
180.20 3.3
67.24 3.5
37.64 5.3
65.32 4.8
56.02 8.1
86.72 6.4
112.35 3.5
76.43 5.6
194.35 7.3
147.28 4.5
136.47 6.3
234.94 8.3
341.01 10.4
216.76 2.1
254.39 6.9
318.43 9.1
254.61 3.7
85.32 7.5
34.72 3.2
36.73 4.8
39.95 3.7
232.14 6.2
83.31 6.1
213.87 7.3
71.52 4.3
31.82 1.9
53.65 5.4
61.53 2.9
91.28 4.8
123.47 6.5
111.67 10.2
189.64 6.2
168.36 5.2
182.56 4.7
220.89 2.9
501.16 8.4
315.54 6.4
322.56 3.5
284.64 7.3
209.46 8.1
62.38 5.8
29.63 1.9
42.74 5.6
51.78 2.1
339.25 9.3
147.71 2.5
319.55 8.1
59.74 2.9
44.88 2.6
71.83 2.8
47.56 7.2
104.37 5.2
98.65 5.2
169.44 4.9
215.36 4.8
128.25 7.5
159.72 3.5
418.42 4.2
411.63 4.9
237.51 3.9
289.37 3.9
233.72 4.6
327.83 3.9
91.43 6.4
30.61 1.2
26.93 3.1
35.49 4.4
204.65 4.3
113.64 2.7
252.83 4.3
64.78 4.5
33.25 4.6
63.26 3.5
52.85 3.4
94.58 5.8
107.83 5.6
79.92 6.4
210.74 5.9
152.81 2.9
142.48 5.4
265.83 1.3
384.73 6.3
219.48 4.2
263.55 7.2
258.41 3.6
294.75 5.8
76.84 4.7
24.58 2.2
31.02 3.8
57.16 6.2
317.32 5.3
176.79 3.1
294.68 11.4
75.68 6.1
42.06 3.7
74.61 6.3
64.37 2.1
87.24 3.9
110.65 4.7
87.05 2.2
196.38 6.3
139.67 3.6
179.23 6.7
349.52 6.7
467.52 3.7
304.61 7.5
227.64 6.1
279.48 4.7
334.28 6.1
69.53
64.37 6.5
68.42 4.9
165.18 7.5
99.48 6.8
202.87 2.9
82.64 3.7
75.56 2.6
88.72 5.6
82.83 1.8
85.43 6.1
91.56 3.9
122.51 3.6
175.27 4.7
108.92 4.9
112.54 6.1
201.67 6.1
316.13 4.2
185.84 3.4
201.84 3.5
225.42 5.7
211.52 4.5
96.25 3.2
50.25 4.2
9
10
11
12
13
14
15
16
17
18
19
20
21
22
DDP
25.46 1.8
hatching rates of embryos were a little difference and the hatching
rates were more than 70%, but when they were at more than
rate of zebrafish depended largely on the dose of the complex
(Fig. 4). When the zebrafish treated with complexes 1 or 2 at 20 M,
m
5.0
m
M, the hatching rates were a big significant difference and the
several malformation patterns were observed, such as pericardial
edema, vitelline cyst, bent spine and bent tail (Fig. 5). Not only in
the hatched juvenile fish but also in unhatched embryos could also
be observed pericardial edema and vitelline cyst (Fig. 5D). These
observations showed pericardial edema and vitelline cyst were the
typical malformations in zebrafish after treatment with complexes
1 and 2.
hatching rates were below 50%. At the same concentration, the
embryos toxicity caused by complex 1 was higher than that of
complex 2, and it was concentration-dependent.
2.4.2. Complex-induced malformation of zebrafish
To make clear whether the complexes have any injury to the
zebrafish organs, we observed the zebrafish morphology at 96 hpf.
The results showed complexes 1 and 2 were able to induce zebra-
2.5. Anti-inflammatory properties
fish malformation. When the complexes were below 5.0 mM, the
malformation rates of zebrafish were less than 5%, and increased
2.5.1. Effects of the complexes on nitrite production
with the increase of concentration. That is to say, the malformation
Nitric oxide (NO) is one of inflammatory mediators.

Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
345
Fig. 3. Effects of the complexes on zebrafish mortality. A: for complex 1; B: for complex 2; C: for hatch rate of complexes 1 and 2 at 72hpf. *P＜0.05 compared to control. Data shown
represents three independent experiments. Values represent the mean SD of three replicates.
Overproduction of NO and following up-regulation of inducible NO
synthase (iNOS) is a critical step in the initiation and propagation of
the inflammatory response [16]. LPS activates iNOS to increase NO
level, but CO can inhibit the iNOS activity by soluble guanylate
cyclase pathway to decrease NO production. NO is oxidized easily to
nitrite by superoxide ions in vivo, and measurement of the level of
nitrite is a way to assess anti-inflammatory activity of CORMs.
Similarly, H2S not only can quench peroxynitrite [17] and interact
chemically with NO to yield a novel nitrosothiol [18], but also re-
the abilities of the complexes to release both CO and H2S. The
testing results demonstrate both complexes significantly reduced
nitrite levels in the presence of LPS. As shown in Fig. 6, both of them
attenuated LPS-induced nitrite production in a concentration-
dependent manner. However, complex 1 displayed the stronger
inhibition effect to nitrite production than complex 2. As we ex-
pected, compounds A1 and A2 also had the ability to inhibit nitrite
production (Fig. 6a and b); compared with complexes 1 and 2, they
displayed a distinctly lower inhibition activity (Fig. 6e). Complex 1
duces induction of iNOS and also inhibits the formation of IL-1
b
and
of 100 mM inhibited approximately 93% nitrite production in cells
TNF- by NF- B and AP-1 pathways [9]. Therefore, the level change
a
k
treated with LPS, and its potency was 1.65 times as high as that of
A1 under the same concentration. This indicates the anti-
inflammatory action of complex 1 did result from both CO and
H2S, not either of them. Possibly, they took a synergistic effect on
the anti-inflammatory effects by inhibiting inflammatory mediator
NO together.
of nitrite also can assess anti-inflammatory activity of H2S donors.
To evaluate the anti-inflammatory activities of the complexes,
we used lypopolysaccharide (LPS) activated macrophages, and
chose complexes 1 and 2 as tested molecules. In this testing, a
rational dose profile was made according to their IC50 values and
Fig. 4. Malformation rate of zebrafish exposed to complexes 1 and 2 (n ¼ 100). A: for complex 1. B: for complex 2. *P＜0.05 compared to control. Data shown represents three
independent experiments. Values represent the mean SD of three replicates.

346
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
Fig. 5. Malformation of zebrafish exposed to 20
mM complexes 1 and 2. A: normal larvae; BeC: abnormal larvae; D: abnormal embryos. PE, pericardial edema; VC, vitelline cyst; BS,
bent spine; BT, bent tail.
The testing assay will be meaningless if a major of macrophages
die in the process of testing, to further confirm testing results, the
macrophages viability was measured using MTT method in the
presence of tested complexes at appointed concentrations. The
results demonstrate cell viabilities were above 80% for each tested
compound (Fig. 7), which suggests that the tested concentration
range and anti-inflammatory results are acceptable and effective.
biliverdin/bilirubin, ferrous iron and carbon monoxide (CO). The
anti-inflammatory properties of HO-1 are related to inhibition of
adhesion molecule expression and reduction of oxidative stress.
The antioxidant properties of biliverdin/bilirubin and the signaling
action of CO could all contribute to suppression of inflammation. In
particular, HO-1 is an important system to counteract the cyto-
toxicity caused by excessive production of nitric oxide [19,20], and
induction of HO-1 in this context significantly reduces NO pro-
duction and potentially restores cellular homeostasis [21].
2.5.2. Effects of the complexes on HO-1
The enzyme heme oxygenase-1 (HO-1) is a cytoprotective and
anti-inflammatory protein that degrades heme to produce
As shown in Fig. 8, in the absence of LPS, the HO-1 levels in cells
treated with the tested compounds were almost unchanged with
Fig. 6. Effect of the complexes on nitrite level. Measurement in the media of murine RAW264.7 macrophages exposed to 10, 50 and 100 mM for 24 h a: for A1; b: for A2; c: for
complex 1; d: for complex 2; e: for nitrite levels compared with complexes 1, 2. Each bar represents the mean SD of four independent experiments. *Indicates P < 0.05 vs control;
yindicates P < 0.05 vs LPS alone.

Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
347
the control. But in the presence of LPS, all the tested compounds
enhanced the expression of HO-1 in cells, and this promotion effect
was concentration-dependent. Among all the tested compounds,
complex 1 was the strongest in enhancing the expression of HO-1
(Fig. 8e). Meanwhile, just as we expected, compounds A1 and A2
which only releasing H2S also enhanced the levels of HO-1, but this
effect was weaker than that resulting from the corresponding
complex which releasing both CO and H2S. This suggests CO and
H2S possibly played the synergistic effect in up-regulating the
expression level of HO-1.
contrast, compounds A1 down-regulated the expression level of
COX-2 in concentration dependent manner; A2 also up-regulated
COX-2 level, but it was not concentration dependent. Distinctly,
the complex and the precursor had opposite biological function to
COX-2. It suggests the complex exerted anti-inflammation by NO
synthase pathway but not COX-2 way.
2.6. Antihypertensive effect of the complexes in SHR
Many testing results showed H2S donors can modulate the
blood pressure. In order to prove this point, we chose SRH as animal
model to study the effect of the complex on blood pressure. The
results showed intraperitoneal injection of the complex (10, 25,
50mgkgꢀ1minꢀ1) produced dose-dependent reductions in systolic
blood pressure (SBP) and diastolic blood pressure (DBP) in SHR,
compared with the control (165 mmHg) (Fig. 12). The administra-
tion of complex 1 with a dose of 50 mg kgꢀ1 produced a marked
decrease in SBP (>e80 mmHg) and DBP (>-58 mmHg) of SHRs. As
we expected, precursor A1 (H2S donor) also decreased in SBP
(>e60 mmHg) and DBP (>-51 mmHg), but it exhibited the less
hypotensive efficacy than complex 1. As for complex 2 and A2, the
trend of hypotensive efficacy was similar to complex 1 and A1
(Fig. 13).
However, compared with captopril and nifedipine, complexes 1
and 2 were much less potent at the dose of 10 mg kgꢀ1, their effi-
cacy reached only about 0.5 times potent; moreover, their time of
duration was about 6 h, which was shorter than those of captopril
and nifedipine nifelat (12 h). Likewise, A1 and A2 exhibited the less
hypotensive efficacy. Following the administration of all doses of
each compound, no distinct differences were observed in the rate of
decrease of SBP/DBP and heart rates during the testing. In addition,
normotensive-control WKY rats also exhibited hypotensive re-
sponses to the complexes, but they were much less sensitive than
their hypertensive (SHR) counterparts.
2.5.3. Effects of the complexes on TNF-
TNF- is a pro-inflammatory cytokines stimulated by LPS in vivo
and in vitro. Seen from Fig. 9, in the absence or presence of LPS,
complexes 1e2 had an inhibitory effect on TNF- production (LPS)
in RAW264.7 macrophages, and this effect increased in
concentration-dependent manner. Complex 1 of 10 M reduced
TNF- level to 85% of the control, and 100 M made it decrease to
about 21%. As expected, compounds A1 and A2 also displayed in-
hibition activity to TNF- (Fig. 9aeb), but their effects were less
a and IL-10
a
a
m
a
m
a
than the corresponding complexes 1 and 2 (Fig. 9e). This is because
the complexes released CO and H2S while A1 and A2 only released
H2S. At the same time, we can see these four compounds all up-
regulated the expression of the anti-inflammatory cytokine IL-10,
which mediated by mitogen activated protein kinases (MAPKs)
[22]. Moreover, IL-10 levels increased with the concentration of the
compounds increasing. Among them, complex 1 displayed the
strongest bio-activity; and it increased IL-10 expression level 2.1-
fold as much as the control (Fig. 10c). This suggests the anti-
inflammatory effects of the complexes were taken by the
pathway of decreasing TNF-
10.
a and increasing the expression of IL-
2.5.4. Effects of the complexes on COX-2
During the process of testing, the complexes showed a stronger
anti-hypotensive effect than the corresponding precursor, this
suggests CO and H2S had synergistic effects on the hypotensive
effect.
Prostaglandin E (PGE) is also one of the inflammatory mediators,
and COX-2 (cyclooxygenase-2) is a key enzyme which catalyzes
arachidonic acid to transform into PGE. Traditional non-steroidal
anti-inflammatory drugs (NSAIDs) mainly inhibit the activity of
COX-2 to reduce prostaglandin E synthesis. COX-2 inhibitors are
used widely and very effective in clinic, and COX-2 is regarded as a
useful target for anti-inflammatory drugs. Whether the tested
complexes have effect on the expression of COX-2, it is a problem to
be deserved to care about.
2.7. Myocardial protective effects of the complexes
To confirm the protective effect of the complexes on myocardial
oxidative injury, H9c2 cell lines were exposed with 400 mM H2O2
In order to illustrate this more clearly, RAW264.7 macrophages
were treated with the complexes, and the levels of COX-2 were
measured using COX-2 ELISA Kit. The results showed that com-
plexes 1 and 2 up-regulated the expression levels of COX-2 in cells,
and this effect was concentration-dependent. Seen from Fig. 11,
complex 1 displayed the same activity as complex 2 on COX-2. In
for 1 h, 4 h and 8 h, respectively [23]. To get the authentic details
about the test, the H9c2 cell damage from the complex themselves
was first measured using MTT method.
Seen from Fig. 14, three precursors A1, A2 and A7 had no signif-
icant effect against H9c2 cell proliferation compared with control
group, indicating that three of them had almost no toxicity at below
Fig. 7. Effect of the complexes on cell viability. Measurement in the media of murine RAW264.7 macrophages exposed for 24 h to 10, 50 and 100
mM complex in the presence or
absence of 1
m
g mLꢀ1 LPS. a: for A1; b: for A2; c: for complex 1; d: for complex 2. Each bar represents the mean SD of four independent experiments.

348
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
Fig. 8. Effect of complexes on HO-1 production. a: for A1; b: for A2; c: for complex 1; d: for complex 2; e: for HO-1 levels compared with A1, A2. Each bar represents the mean SD of
four independent experiments. *Indicates P < 0.05 vs control; yindicates P < 0.05 vs LPS alone.
Fig. 9. Effect of complexes on TNF-
a production. a: for A1; b: for A2; c: for complex 1; d: for complex 2; e: for TNF-a levels compared with A1, A2. Each bar represents the mean SD
of four independent experiments. *Indicates P < 0.05 vs control; yindicates P < 0.05 vs LPS alone.
100
concentration-dependent toxicity. But when each was up to
100 M, the cell viability rates still were over 90%. This indicates our
concentration range (0e100 M) was reliable and effective.
Consequently, we did all testing at below 100 M for each
compound. As shown in Fig. 15, after H9c2 cells were treated with
m
M. For the tested complexes (1, 2 and 7), three all showed
H2O2 for 1 h, 4 h and 8 h, the cell survival rates were 81.3%, 76.3%
and 64.4%, respectively. After H9c2 cells treated with H2O2, then
cultured with precursor and complex, we found the cell viability
rates increased, and it was connected with the time of H2O2
treatment. When the cells treated with H2O2 for 1 h, precursor A1 of
m
m
m
100 mM, increased the survival rate by 12.2%; but the cells treated

Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
349
Fig. 10. Effect of complexes on IL-10 production. a: for A1; b: for A2; c: for complex 1; d: for complex 2; e: for IL-10 levels compared with A1, A2. Each bar represents the mean SD of
four independent experiments. *Indicates P < 0.05 vs control; yindicates P < 0.05 vs LPS alone.
Fig. 11. Effect of complexes on COX-2 production. a: for A1; b: for A2; c: for complex 1; d: for complex 2. Each bar represents the mean SD of four independent experiments.
*Indicates P < 0.05 vs control; yindicates P < 0.05 vs LPS alone.
for 4 h and 8 h, the increases of survival rate were lower and lower.
Interestingly, in the range of 10e100 M, complexes 1, 2 and 7 did
not show concentration-dependence. Complexes 1 and 2 signifi-
concentration was 50
complex 1 increased the cell survival rate by 14.6%. When com-
plexes 1, 2 and 7 were below at 50 M, their protective effects to
H9c2 cells were significantly better than those of precursor A1, A2
mM when the cells treated for 1e4 h; and
m
m
cantly increased H9c2 cells viability rates, and the optimal

350
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
Fig. 12. Effect of complex 1 on blood pressure in SHR. Each bar represents the mean SD of six independent experiments. *Indicates P < 0.05 vs control; yindicates P < 0.05 vs LPS
alone.
Fig. 13. Effect of complex 2 on blood pressure in SHR. Each bar represents the mean SD of six independent experiments. *Indicates P < 0.05 vs control; yindicates P < 0.05 vs LPS
alone.
and A7. This is probably due to the synergistic effect mediated by
H2S and CO, while the precursors only released H2S.
concentration was already beyond their margin of safety, which
was corresponding with the toxicity test.
However, when the tested complexes were increased to 100
their protective effects did not continue to increase, but began to
decline. Especially, after H9c2 cells treated with H2O2 for 8 h, both
m
M,
The protective effects of complex to H9c2 cells against oxidative
damage were also reflected by cell morphology. After treated with
complex 1 or A1, the cell which stained with Giemsa dye for 20 min
were observed under microscope.
complexes 2 and 7 at 100
mM made the cells survival rate decrease,
and they were even lower than H2O2 control group. Distinctly, this
As shown in Fig. 16, in the control group, the myocardial H9c2

Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
351
Fig. 14. Effects of complex on viability rate of H9c2 cell line. Each bar represents the mean SD of six independent experiments. *P < 0.05; **P < 0.01 vs the model group.
Fig. 15. The protective effects of complex to H9c2 cells against oxidative damage. Each bar represents the mean SD of six independent experiments. *P < 0.05; **P < 0.01 vs the
model group.
cells were a long shuttle type, firmly adherent and no abnormal;
but for tested group, the cells were an irregular shape and some
cells became round. With the increase of H2O2 oxidative damage,
the deformed cells were more and more and the injury became
more seriously (Fig. 16d). However, in the presence of A1 or com-
plex 1, part cells recovered their original state and the atrophic cells
were less and less. Meanwhile, we can see that cardioprotective
effect of the tested complex was stronger than that of the precursor
compound A (Fig. 16 h and m).
The damaged cells led to the decrease of intracellular SOD level,
but the declining tendency was inhibited by the treatment with
different concentrations of complexes 1, 2 and 7 (Fig. 17 AꢀC). In
addition, the treatment of H9c2 cells with H2O2 caused the increase
of MDA, but incubation of cells with the tested complexes markedly
decreased the level of MDA (Fig.17 DꢀF) [24,25]. This demonstrates
the antioxidant capacity of the complex is also concentration-

352
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
Fig. 16. H9c2 cells morphology under all kind conditions: for blank group (a, e and j); for H2O2 group (b, c and d); for H2O2-A (50
mM) group (f, g and h); for H2O2-complex 1
1
(50
mM) group (k, l and m). The cells were cultured in 6-well plates and the tested complex was given for 24 h. From light to right, H2O2 treated for 1 h, 4 h and 8 h respectively.
Fig. 17. Effect of the complex on myocardial oxidative injury. A, B and C: for SOD level; D, E and F: for MDA level. Data shown are mean SD, n ¼ 6. *P < 0.05, **P < 0.01.
dependence. Furthermore, the tested complexes showed better
antioxidant capacity than the precursors. These data suggests that
antioxidant capacity of the complex was attributed to dual donor,
and the increase in antioxidant capacity may be an important factor
of inhibiting apoptotic cell death (Figs. 15 and 17).
In summary, the tested complexes showed better myocardial
protection in a range of effective concentrations. However, the
protective effect on severe injury was not obvious enough, which
may be due to the high degree of damage that leads to the cells
being hard to repair.

Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
353
3. Conclusions
4.2. Preparation of all the complexes
H2S and CO are signal conducting molecules, and both are the
products of metabolism in vivo. They have the same biological ac-
tivities; they probably have synergetic effect in some aspects. Based
on this theory, we synthesized these H2SeCO dual donors. They not
only released CO, but H2S. Theoretically, each complex releases four
CO and one H2S; but in fact, every complex releases CO and H2S in
different speeds and maxium amount under the same condition.
Among these complexes, complexes 1, 2 and 7 displayed stronger
anti-proliferative activities to the tested cell lines. Complex 1
showed the highest inhibiting activity to MCF-7, complex 2 to
HePG2, and complex 7 to A549 and Hella cells. After zebrafish
embryos were exposed to the complexes, their survival rates and
hatchabilities reduced with the concentration increasing; more-
over, complexes 1 and 2 caused malformation of zebrafish, and its
severity index was a dose-dependent.
All compounds were prepared according our previous works
[12]. The general procedure: Amines (1 mmol) was dissolved in
20 mL of MeOH. Equal molar NaOH aqueous solution was added to
the methanol solution. Excess CS2 was slowly dropped into the
reaction flask in an ice bath. The reaction was completed at room
temperature for 8 h. The solution was then evaporated and the
residue was precipitated three times with diethyl ether and filtered
and dried to give a white powder product. All sodium dithiocar-
bamate (A1-22) are white solid. A1-22 and equal molar Mn (CO) 5Br
were dissolved in 15 mL of methanol. The solution immediately
became yellow and was stirred at room temperature for 3 h. The
solution was then evaporated and the yellow residue was dissolved
in Ether. The solution was placed in 2e4 ꢁC for 2 h and the super-
natant was evaporated to give a yellow solid. After recrystallization
three times with DCM and CYH, the final product was obtained.
In the test of bio-activities, complexes 1 and 2 had strong anti-
inflammatory activities; as we expected, precursors A1 and A2
displayed lower activities than them. This possibly contributed to
the synergistic effect mediated by the CO and H2S. In the process of
anti-inflammation, the complexes down-regulated the expression
Complex 1 Yield: 46%. 1H NMR (DMSO‑d6):
4.40 (s, 1H, eOH), 3.71 (m, 2H, eNCH2), 3.17 (s, 1H, eCHOH), 2.25
d 4.62 (m,1H, eNCH),
(m, 2H, eCHCH2). 13C NMR (DMSO‑d6, ppm):
d 224.3, 220.7, 212.8,
173.7, 59.8, 50.2, 20.5, 23.7. IR (KBr, cmꢀ1): 2084 (vs),1997 (vs),1952
(vs), 1940 (vs), 1715 (m), 1470 (m), 1452 (m). ESI-HRMS (m/z): Calcd.
for C10H8O6NS2MnNa [MþNa]þ: 379.9071; found 379.9062.
levels of iNOS and TNF-a, and up-regulated the expression of HO-1
and IL-10. Thereby, they maintain the homeostasis of the intracel-
lular environment. We know the mechanism of action of NSAIDs is
connected with COX-2, we found the tested complexes had a
certain effect on COX-2 and slightly up-regulated COX-2 levels, this
suggests this anti-inflammatory effect of the complexes may
mediated by many pathways, including many inflammatory me-
diators activation [26].
Besides anti-inflammation, the complexes displayed antihy-
pertensive effect and myocardial protection. Under the same mass
concentration condition, complexes 1 and 2 had already displayed
better activities than the precursors A1 and A2, this also maybe was
resulting from the vasodilation and against oxidative damage
mediated by both H2S and CO, and these phenomena contribute to
synergistic effect of CO and H2S on the bio-responses. All these
suggest the complexes displayed the same bioactivities and func-
tion as endogenous CO and H2S, and have a potential application in
clinic. However, which disease may be used or what kind patient
will be administrated with it? To answer these questions, it needs
lots of work to do.
Complex 2 Yield: 48%. 1H NMR (DMSO‑d6):
d 4.62 (m, 1H), 4.40
(s, 1H), 3.71 (m, 2H), 3.17 (s, 1H), 2,25 (m, 2H). 13C NMR (DMSO‑d6,
ppm):
d 218.1, 214.7, 208.3, 170.2, 69.1, 64.3, 57.5, 33.6. IR (KBr,
cmꢀ1): 2092 (vs), 2014 (vs), 1991 (vs), 1948 (vs), 1720 (m), 1441 (m).
ESI-HRMS (m/z): Calcd. for C10H8O7NS2MnNa [MþNa]þ: 395.9020;
found 395.9016.
Complex 3 Yield: 52%. 1H NMR (CDCl3):
d
3.57 (d, J ¼ 6.7 Hz, 2H),
3.42 (s, 2H), 1.97 (d, J ¼ 6.0 Hz, 4H). 13C NMR (CDCl3, ppm):
d 216.4
(MneCO), 213.5 (MneCO), 211.3 (CS2), 54.4 (CH2), 26.2 (CH2). IR
(KBr, cmꢀ1): 2028 (vs), 2005 (vs), 1987 (vs), 1944 (vs), 1452 (m),
1444 (m). ESI-HRMS (m/z): Calcd. for C9H8O4NS2MnNa [MþNa]þ:
335.9675; found 335.9652.
Complex 4 Yield: 42%. 1H NMR (CDCl3):
d 2.97e2.87 (m, 4H),1.79
(s, 1H), 1.22 (td, J ¼ 13.9, 7.0 Hz, 4H), 1.00 (d, J ¼ 6.3 Hz, 3H). 13C NMR
(CDCl3, ppm):
d
210.6, 202.4, 49.1, 39.2, 23.7. IR (KBr, cmꢀ1): 2077
(s), 2001 (vs), 1978 (vs), 1946 (vs), 1454 (m), 1431 (m). ESI-HRMS
(m/z): Calcd. for
C
11H13O4NS2Mn [MþH]þ: 341.9866; found
341.9810.
Complex 5 Yield: 49%. 1H NMR (CDCl3):
d
4.45 (d, J ¼ 17.1 Hz, 2H),
2.93 (s, 1H), 2.62 (t, J ¼ 11.8 Hz, 1H), 1.85e1.44 (m, 4H), 1.18 (s, 1H),
0.90 (d, J ¼ 6.7 Hz, 3H). 13C NMR (CDCl3, ppm):
d 211.7, 196.3, 54.6,
4. Experimental
48.7, 31.2, 17.3. IR (KBr, cmꢀ1): 2083 (s), 1995 (vs), 1977(vs), 1928
(vs), 1455 (m), 1432 (m). ESI-HRMS (m/z): Calcd. for C11H12O4N-
S2MnNa [MþNa]þ: 363.9186; found 363.9177.
4.1. Reagents and instruments
Complex 6 C12H14MnNO5S2 Yield: 42%. 1H NMR (CDCl3):
d 4.76
Myoglobin (AR, Sigma), BCA protein quantification kit (Sigma),
ELISA kit (Sigma), MnBr(CO)5 (AR, Sigma). Captopril and nifedipine
nifelat were purchased from Beijing bailingwei Technology Co. Ltd.
Murine RAW264.7 monocyte macrophages and zebrafish were
purchased from the cell resources Center for Shanghai Life Science
Institute of Chinese Academy of Sciences (China). Spontaneously
hypertensive rats (SHR) rats and WKY (Wistar-Kyoto) rats were
provided from the Animal Center of GLP, Lanzhou University.
A Lambda 25 UVevisible spectrophotometer and a Maxis-4G
TOF Mass spectrometer (ESI) were used. Blood press was
measured using BP-300A Automatic non-invasive blood pressure
detector (Chengdu, taimeng, China, Software Co. Ltd.). NMR spectra
were recorded on a Bruker AM-400 MHz instrument at room
temperature. 1H and 13C NMR spectra are referenced to solvent, and
chemical shifts are expressed in ppm (parts per million). Infrared
(s, 1H), 3.58 (m, 2H), 2.91e3.01 (m, 4H), 1.26e1.51 (m, 7H). 13C NMR
(CDCl3, ppm):
d
212.4, 195.4, 57.4, 48.5, 31.8, 13.6. IR (KBr, cmꢀ1):
2086 (s), 1987 (vs), 1946 (vs), 1928 (vs), 1452 (m), 1234 (m). ESI-
HRMS (m/z): Calcd. for C12H15O5NS2Mn [MþH]þ: 371.9672; found
371.9621.
Complex 7 Yield: 45%. 1H NMR (CDCl3):
d 5.02 (s, 2H), 4.01 (s,
4H), 3.49 (d, J ¼ 5.5 Hz, 4H). 13C NMR (CDCl3, ppm):
d 210.6, 202.5,
195.7, 59.3, 57.2. IR (KBr, cmꢀ1): 2074 (s), 1995 (vs), 1967 (vs), 1932
(vs), 1208 (m). ESI-HRMS (m/z): Calcd. for C9H11O6NS2Mn [MþH]þ:
347.9108; found 347.9121.
Complex 8 Yield: 42%. 1H NMR (CDCl3):
d 5.02 (s, 1H), 3.94 (m,
2H), 3.67 (m, 2H), 3.45 (m, 2H), 1.43 (t, J ¼ 7.6 Hz, 3H). 13C NMR
(CDCl3, ppm):
d
213.2, 207.1, 201.8, 62.3, 57.2, 12.8. IR (KBr, cmꢀ1):
2084 (s), 1979 (vs), 1952 (vs), 1945 (vs), 1478 (m), 1232 (m). ESI-
HRMS (m/z): Calcd. for C9H10O5NS2MnNa [MþNa]þ: 353.9879;
found 353.9853.
spectra were obtained with
spectrophotometer.
a
Nicolet NEXUS 360 FTIR
Complex 9 Yield: 39%. 1H NMR (CDCl3):
d 5.01 (m, 1H), 3.79 (t,

354
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
J ¼ 5.5 Hz, 2H), 3.67e3.62 (m, 2H), 3.17 (d, J ¼ 5.1 Hz, 3H). 13C NMR
4.76 (s, 2H), 4.74 (s, 1H), 3.71 (m, 2H), 3.43 (m, 2H). 13C NMR (CDCl3,
ppm):
215.2, 204.8, 139.5, 132.7, 130.3, 58.1. IR (KBr, cmꢀ1):
(CDCl3, ppm):
d
212.5, 209.2, 198.8, 62.3, 57.8, 41.4. IR (KBr, cmꢀ1):
d
2077 (s), 2001 (vs), 1978 (s), 1944 (s), 1198 (m). ESI-HRMS (m/z):
2084(s), 1998(vs), 1952(s), 1944(vs), 1248 (m), 745 (m). ESI-HRMS
(m/z): Calcd. for C14H12O5NS2MnNa [MþNa]þ: 415.9435; found
415.9462.
Calcd. for C8H8O5NS2MnNa [MþNa]þ: 339.9522; found 339.9536.
Complex 10 Yield: 44%. 1H NMR (CDCl3):
d
5.71 (s, 1H), 5.24 (s,
207.6, 199.2,
2H), 4.25 (s, 2H), 3.13 (s, 3H). 13C NMR (CDCl3, ppm):
d
120.3, 58.7, 43.2. IR (KBr, cmꢀ1): 2078 (s), 1995 (vs), 1977 (vs), 1940
(vs), 1657 (m). ESI-HRMS (m/z): Calcd. for C9H8O4NS2MnNa
[MþNa]þ: 335.9773; found 335.9712.
4.3. CO-releasing tests with myoglobin assay
The release of CO from the metal carbonyl complexes was
studied spectrophotometrically by measuring the conversion of
deoxy-myoglobin (deoxy-Mb) to carbonyl myoglobin (Mb-CO). The
amount of Mb-CO formed was quantified by measuring the
absorbance at 540 nm. A stock solution of myoglobin (lyophilised
Complex 11 Yield: 46%. 1H NMR (CDCl3):
d 6.04 (m, 2H), 5.34 (m,
4H), 3.87 (d, J ¼ 6.0 Hz, 4H). 13C NMR (CDCl3, ppm):
d 212.5, 209.2,
205.6, 141.2, 57.2. IR (KBr, cmꢀ1): 2084(m), 1995(vs), 1974(s),
1929(vs), 1652(m). ESI-HRMS (m/z): Calcd. for C11H10O4NS2MnNa
[MþNa]þ: 361.9029; found 361.9062.
horse heart) (60 mM final concentration) was prepared fresh by
Complex 12 Yield: 38%. 1H NMR (CDCl3):
d
3.71 (q, J ¼ 7.1 Hz, 4H),
dissolving the protein in phosphate buffered saline (PBS, 0.1 M,
pH ¼ 7.4). Sodium dithionite (0.1%) was added to convert the
myoglobin stock to deoxy-Mb. A 2 mL quantity of this solution was
assayed to obtain a deoxy-Mb spectrum and then bubbled with CO
to get an Mb-CO spectrum. Each complex was dissolved in water
and added to deoxy-Mb in the cuvette, mixed using a pipette and
1.26 (t, J ¼ 7.2 Hz, 6H). 13C NMR (CDCl3, ppm):
d
208.7, 201.4, 53.2,
14.1. IR (KBr, cmꢀ1): 2084(s), 1987(vs), 1946(vs), 1938(vs), 1450 (m).
ESI-HRMS (m/z): Calcd. for C9H10O4NS2MnNa [MþNa]þ: 337.9429;
found 337.9483.
Complex 13 Yield: 48%. 1H NMR (CDCl3):
d
3.47 (s, 4H), 2.11 (m,
2H), 0.94 (m,12H). 13C NMR (CDCl3, ppm):
d
211.8, 63.4, 24.1. IR (KBr,
then overlaid with 500 mL light mineral oil to prevent CO escaping
cmꢀ1): 2086(s), 1996(vs), 1952(s), 1945(vs), 1382(m). ESI-HRMS (m/
or the myoglobin being oxygenated. This is the standard procedure;
other experiments have been undertaken using different concen-
trations of myoglobin.
In these experiments, the absorption peak of deoxy-Mb at
560 nm was replaced by the two peaks of Mb-CO at 540 and
578 nm. The concentration of myoglobin in the stock solution was
calculated from the absorption maximum of the Mb-CO solution at
540 nm.
z): Calcd. for C13H19O4NS2Mn [MþH]þ: 372.0736; found 372.0763.
Complex 14 Yield: 37%. 1H NMR (CDCl3):
d
3.25 (t, J ¼ 6.7 Hz, 4H),
1.63 (m, 4H), 1.46 (m, 4H), 1.04 (t, J ¼ 7.6 Hz, 6H). 13C NMR (CDCl3,
ppm):
d
212.5, 199.6, 55.3, 31.6, 21.4, 15.2. IR (KBr, cmꢀ1): 2084 (s),
2001(vs), 1947(s), 1928(s), 1464(m), 1448(m). ESI-HRMS (m/z):
Calcd. for C13H19O4NS2Mn [MþH]þ: 372.0836; found 372.0810.
Complex 15 Yield: 46%. 1H NMR (CDCl3):
d 3.71 (m, 2H), 3.28 (s,
3H), 3.54 (t, J ¼ 6.8 Hz, 2H), 3.11 (s, 2H), 1.02 (t, J ¼ 7.4 Hz, 3H). 13
C
NMR (CDCl3, ppm):
d
215.5, 198.1, 78.3, 55.6, 53.4, 15.7. IR (KBr,
4.4. H2S measurement
cmꢀ1): 2078(s), 1978(vs), 1952(s), 1942(vs), 1446(m), 739 (s). ESI-
HRMS (m/z): Calcd. for C10H13O5NS2MnH [MþH]þ: 345.9816;
found 345.9810.
A 5 mM solution of Na2S in sodium phosphate buffer (20 mM,
pH 7.4) was prepared (Na2S.9H2O, 120.20 mg in 100 mL volumetric
flask) and used as the stock solution. Aliquots of 50, 100, 200, 400,
Complex 16 Yield: 41%. 1H NMR (CDCl3):
d
7.14e7.37 (m, 5H),
201.9, 197.4, 132.7,
4.85 (s, 2H), 3.09 (s, 3H). 13C NMR (CDCl3, ppm):
d
600, 800,1000,1500
50 mL volumetric flask and dissolved in sodium phosphate buffer to
obtain the standard solutions in 5, 10, 20, 40, 60, 80, 100, 150 M,
respectively. 1 mL aliquot of the respective solution was reacted
with the methylene blue (MBþ) cocktail: 30 mM FeCl3 (200
L) in
1.2 M HCl, 20 mM of N,N-dimethyl-1,4- phenylenediamine sulfate
(200 L) in 7.2 M HCl, 1%w/v of Zn(OAc)2 (100 L) in H2O at room
mL of the Na2S stock solution were added into a
131.6, 61.4, 46.7. IR (KBr, cmꢀ1): 2086(s), 1995(vs), 1977(s), 1929(vs),
1450(m), 729 (m). ESI-HRMS (m/z): Calcd. for C13H11O4NS2Mn
[MþH]þ: 363.0010; found 363.0057.
m
Complex 17 Yield: 43%. 1H NMR (CDCl3):
d
7.22e7.44 (m, 10H),
m
4.83 (s, 4H). 13C NMR (CDCl3, ppm):
d 213.4, 141.7, 138.6, 135.4, 56.2.
IR (KBr, cmꢀ1): 2083(s), 2005(vs), 1987(s), 1944(vs), 1452(m), 750
(m). ESI-HRMS (m/z): Calcd. for C19H14O4NS2MnNa [MþNa]þ:
461.9442; found 461.9472.
m
m
temperature for at least 15 min (each reaction was performed in
triplicate). The absorbance of methylene blue was measured at
lmax ¼ 670 nm in UVeVis spectrophotometer (Lambda 25). The
Na2S calibration curve was obtained.
Complex 18 Yield: 51%. 1H NMR (CDCl3):
d
7.18e7.44 (m, 5H),
3.51 (s, 1H). 13C NMR (CDCl3, ppm):
d 209.3, 144.3, 142.7, 137.8, 46.1.
IR (KBr, cmꢀ1): 2084(s), 1996(vs), 1978(s), 1946(vs), 1450(m),
753(m). ESI-HRMS (m/z): Calcd. for C12H8O4NS2MnK [MþK]þ:
387.0212; found 387.0213.
The reaction was initiated by adding 15
mL of stock solution of
the complex or ligand (60 M) into pH 7.4 phosphate buffer (30 mL)
m
containing accelerator for TECP (1.0 mM). Then 2.0 mL of reaction
aliquots were periodically taken and transferred to colorimetric
Complex 19 Yield: 41%. 1H NMR (CDCl3):
d
7.11e7.59 (m, 4H),
3.46 (s, 3H). 13C NMR (CDCl3, ppm):
d
221.6, 204.5, 151.7, 138.2,
cuvette containing zinc acetate (1% w/v, 200
1,4-phenylenediamine sulfate (20 mM, 400
ferric chloride (30 mM, 400 L) in 1.2 M HCl. The absorbance
mL) and N, N-dimethyl-
133.6, 45.4. IR (KBr, cmꢀ1): 2078(s), 1979(vs), 1947(s), 1940(vs),
1093 (m). ESI-HRMS (m/z): Calcd. for C12H8O4ClNS2Mn [MþH]þ:
383.8964; found 383.8979.
mL) in 7.2 M HCl and
m
(670 nm) of the resulted solution was determined 15 min thereafter
using an UVeVis spectrometer (Lambda 25). The H2S concentration
of each sample was calculated against a calibration curve of Na2S.
H2S amount released from the compounds was measured every
3 min, then described the curve about H2S releasing amount vs
time.
Complex 20 Yield: 36%. 1H NMR (CDCl3):
d
7.28e7.56 (m, 4H),
218.3, 212.4, 144.6, 135.2,
133.8, 47.2. IR (KBr, cmꢀ1): 2082(s), 1998(vs), 1949(vs), 1943(vs),
3.43 (s, 3H). 13C NMR (CDCl3, ppm):
d
1091 (m). ESI-HRMS (m/z): Calcd. for
[MþNa]þ: 405.8783; found 405.8792.
Complex 21 Yield: 41%. 1H NMR (CDCl3):
C
12H7O4ClNS2MnNa
d
6.82e7.16 (m, 4H),
4.78 (s, 2H), 3.44 (s, 3H), 3.05 (s, 3H). 13C NMR (CDCl3, ppm):
d
213.6,
4.5. Cytotoxicity assays
204.7, 138.1, 126.3,133.6. IR (KBr, cmꢀ1): 2078(s), 1999(vs), 1967(vs),
1946(vs), 1252(m), 792(m). ESI-HRMS (m/z): Calcd. for
All cell lines were purchased from cell resources Center for
Shanghai Life Science Institute of Chinese Academy of Sciences
(China). The cells were cultivated at 37 ꢁC, 10% CO2, 100% humidity
C
14H12O5NS2MnNa [MþNa]þ: 415.9435; found 415.9421.
Complex 22 Yield: 42%. 1H NMR (CDCl3):
d 7.13e7.28 (m, 5H),

Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
355
in RPMI1640 medium, enriched with glucose and supplemented
with 10% fetal bovine serum, non-essential amino acids, antibiotics
(penicillin/streptomycin) and antifungals.
Typically, 8 mmol of the complex was dissolved in 1.0 mL of
culture medium with vigorous stirring. Growth inhibitory effect
toward the cell line was determined by means of MTT colorimetric
atmosphere and experiments were conducted on cells at approxi-
mately 80e90% confluence.
Macrophages were exposed for 24 h to LPS (1
presence or absence of complexes 1, 2 or ligands A1, A2 (10, 50 and
100 M) and nitrite levels was determined at the end of the incu-
m
g mLꢀ1) in the
m
bation. Nitrite levels were determined using the Griess method as
previously described [30]. The measurement of this parameter is
widely accepted as indicative of NO production. Briefly, the me-
dium from treated cells was removed and placed into a 96-well
assay. The cells (100
m
L, 1 ꢂ 105 cells mLꢀ1) were seeded into 96-
well plates and left to adhere for 24 h. The media was removed
from the wells and replaced with fresh media containing the
complexes with different concentrations (25, 50, 100, 200, 400 and
plate (50 mL per well). The Griess reagent was added to each well
800
another 24 h before the incubation media were replaced with the
complete medium and MTT (10
L, 5 mg mLꢀ1 in phosphate buffer
m
mol Lꢀ1), respectively. The cells were then incubated for
to begin the reaction, the plate was shaken for 10 min and the
absorbance read at 570 nm on a microplate reader. The nitrite level
in each sample was calculated from a standard curve generated
m
solution, PBS) was added to each well of the plate. The cells were
further incubated for 4 h before the media were replaced with
with sodium nitrite (0e100 mM in cell culture medium).
DMSO (100
was recorded with a microplate reader. In the MTT assay, DMSO
(100 L) in a well was used as blank and cells in the well without
the addition of any complexes were taken as a control (100% in cell
viability).
m
L). Absorbance at 490 nm for each well of the plates
4.8. Bicinchoninic acid (BCA) protein assay
m
In order to exclude the influence of the total amount of protein
on the determination of the following cytokine levels, protein
quantification of each protein was carried out using BCA protein
Assay. Briefly, samples were removed and placed into a 96-well
4.6. Zebrafish toxicity test
plate (20
mL per well). BCA reagent A and reagent B were formu-
lated as working solutions, which was added to each well (200
m
L
Adult zebrafish of the wild-type strain (AB) were raised and
maintained at 28 1 ꢁC with a 14 h light/10 h dark photoperiod
(lights on at 8:00) in a recirculation system. The fish water supplied
to the system was filtered by reverse osmosis (pH6.5e7.5), and
instant ocean salt was added to the water, to raise the conductivity
per well) and left to react for 0.5 h in 37 ꢁC. And then the plate was
shaken for 10 min and the absorbance read at 562 nm on a
microplate reader. The concentration of protein in each sample was
calculated from a standard curve generated with standard solution.
to 450e500
ms/cm. The zebrafish were fed twice daily with
4.9. Determination of TNF-a, HO-1, IL-10, and COX-2 levels
decapsulated, freshly hatched brine shrimps (Brine Shrimp Direct,
USA) according to the description of Zhou [27] The zebrafish em-
bryos were obtained from spawning adults in tanks overnight with
a sex ratio of 1:1. The embryos were collected within 1 h after the
light was switched on and washed using standard zebrafish E3
culture medium. The zebrafish use and handling protocol con-
formed to the Guide for the Care and Use of Laboratory Animals and
was approved by the Institutional Animal Use and Care Committee
(IAUCC) of the Lanzhou University. At 4 h post-fertilization (hpf),
the embryos were examined under a dissecting light microscope,
and the specimens that had developed normally were selected for
the further experiments according to the description of Kimmel
[28].
The levels of cytokine and COX-2 present in each sample were
determined using a commercially available kit from Shanghai
MLBIO Biotechnology Co.Ltd. The assay was performed according to
the manufacturers’ instructions. Briefly, cell were collected from 6-
well plate immediately after lysed by RIPA Lysis Buffer and spun at
12,000 ꢂ g for 10 min to remove any particulates. Each protein so-
lution was measured by BCA Protein Assay Kit. The protein solution
for concentration of 0.1 mg mlꢀ1 was added to a 96-well plate
precoated with affinity-purified polyclonal antibodies specific for
the mouse of our cytokines. An enzyme-linked polyclonal antibody
specific for the mouse of our cytokines was added to the wells and
left to react for 0.5 h followed by a final wash to remove any un-
bound antibody-enzyme reagent. The intensity of the color detec-
ted at 450 nm was measured after addition of chromogen solution
A, B and stop solution and was proportional to the amount of four
cytokines produced. The four cytokines levels which in each sample
was calculated from a standard curve generated with standard
solution.
The zebrafish embryos were exposed to complex for 24, 48, 72,
96 and 120 hpf and then assessed for toxicity. The toxicological end
points were determined based on previous reports in the literature
[29]. Each group consisted of 100 embryos randomly divided into
three replicate groups. The embryos were kept in sterile 96-well
plates with one embryo per well containing 200 mL of the solu-
tion. The plates were covered with sealing film to prevent evapo-
ration and the mortality of the zebrafish was recorded at 24, 48, 72,
96 and 120 hpf.
4.10. Anti-hypertensive effects of the complexes
Normal embryos were exposed to the control vehicle and
complex from 4 to 72 hpf, the hatch rate was measured at 72 hpf.
The normal embryos (4 hpf) were exposed to the control vehicle
complex for 96 hpf, and then the malformation of the zebrafish
were observed using a stereoscopic dissecting microscope.
A total of 40 male clean spontaneously hypertensive rats (SHR, 8
weeks age; weight, 300 25 g) were divided into normal control
group, captopril group (10 mg kgꢀ1), nifedipine nifelat group
(10 mg kgꢀ1) and compound group (25, 50, 100 mg kgꢀ1) by ran-
domized block design. Male WKY rats were used as the normal
control group.
4.7. Anti-inflammatory actions assays
All the drugs and compounds were administered by intraperi-
toneal injection as a 2 mL solution of H2O per kilogram. The control
group and WKY group were injected mixed solution without con-
taining compound or drug. Tail blood pressure and heart rate were
measured before administration. After administration, blood pres-
sure and heart rate were measured every 30min, lasting for 6 h. The
arterial systolic pressure, diastolic arterial pressure (DBP) and heart
rate were recorded before and after administration.
Murine RAW264.7 monocyte macrophages were purchased
from the cell resources Center for Shanghai Life Science Institute of
Chinese Academy of Sciences (China), and cultured in medium
(DMEM) supplemented with 10% fetal bovine serum, 2 mM L-
glutamine, 100 units mlꢀ1 penicillin and 0.1 mg mLꢀ1 streptomycin.
Cultures were maintained at 37 ꢁC in a 5% CO2 humidified

356
Z. Bai et al. / European Journal of Medicinal Chemistry 159 (2018) 339e356
[4] S.P.H. Alexander, A. Mathie, J.A. Peters, Guide to receptors and channels
In a quiet state of consciousness, the rats were fixed to a special
(GRAC), Br. J. Pharmacol. 153 (2008) S2.
fixture. The rat tail was covered until the roots by pressure cuff,
which strapped pulse transducer to contact closely to the ventral
tail artery. When the rat pulse wave stabilized, it began to measure.
The pressure cuff was inflated and pressurized until the blocking
pressure was 250 mmHg. (The pulse wave disappeared when the
pressure reached the systolic pressure level.) After the pressure
maintains 6 s, it slowly decompresses. When the first pulse waves
appear, the corresponding blood pressure is the arterial systolic
pressure, the peak of the corresponding curve is the arterial dia-
stolic pressure, and the system automatically displays the heart
rate. Each group was continuously measured 6 times. All data are
expressed as the mean standard error (SE); n refers to number of
animals studied in each group. Significance was accepted at
P < 0.05.
[5] Y. Zhao, H. Wang, M. Xian, Cysteine activated hydrogen sulfide (H2S) donors,
J. Am. Chem. Soc. 133 (2011) 15e17.
[6] Z. Qingyou, D. Junbao, Z. Weijin, et al., Impact of hydrogen sulfide on carbon
monoxide/heme oxygenase pathway in the pathogenesis of hypoxic pulmo-
nary hypertension, Biochem. Biophys. Res. Commun. 317 (2004) 30e37.
[7] K. Erdmann, B.W.Y. Cheung, S. Immenschuh, et al., Heme oxygenase-1 is a
novel target and antioxidant mediator of S-adenosylmethionine, Biochem.
Biophys. Res. Commun. 368 (2008) 937e941.
[8] R. Motterlini, B. Haas, R. Foresti, Emerging concepts on the anti-inflammatory
actions of carbon monoxide-releasing molecules (CO-RMs), Med. Gas Res. 2
(2012) 28.
[9] L. Li, G. Rossoni, A. Sparatore, et al., Anti-inflammatory and gastrointestinal
effects of a novel diclofenac derivative, Free Radic. Biol. Med. 42 (2007)
706e719.
[10] J.L. Wallace, G. Caliendo, V. Santagada, G. Cirino, Markedly reduced toxicity of
a hydrogen sulphide-releasing derivative of naproxen (ATB-346), Br. J. Phar-
macol. 159 (2010) 1236e1246.
[11] R. Kodela, M. Chattopadhyay, K. Kashfi, NOSH-Aspirin:
a novel nitric
oxideehydrogen sulfide-releasing hybrid: a new class of anti-inflammatory
pharmaceuticals, ACS Med. Chem. Lett. 3 (2012) 257e262.
[12] Zhang Taofeng, Gong Yaguo, Li Meng, Xi Na, Zheng Yawen, Zhao Quanyi,
Chen Yonglin, Liu Bin, Syntheses, properties and bio-activities of water-
soluble CO-releasing molecule based on manganese, J. Biol. Inorg. Chem. 21
(2016) 807e824.
[13] Pengpeng Wang, Huapeng Liu, Quanyi Zhao, Yonglin Chen, Bin Liu,
Baoping Zhang, Qian Zheng, Syntheses and evaluation of drug-like properties
of CO-releasing molecules containing ruthenium and group 6 metal, Eur. J.
Med. Chem. 74 (2014) 199e215.
[14] N.O. Devarie-Baez, P.E. Bagdon, B. Peng, et al., Light-induced hydrogen sulfide
release from “caged” gem-dithiols, Org. Lett. 15 (2013) 2786e2789.
[15] U. Hasegawa, A.J. van der Vlies, Design and synthesis of polymeric hydrogen
sulfide donors, Bioconjug. Chem. 25 (2014) 1290e1300.
4.11. Myocardial protective effects of the complexes
The rat cardiac H9c2 cells were purchased from the Cell Bank of
the Chinese Academy of Sciences (Shanghai, China) and cultured in
DMEM supplemented with 10% FBS,100 UꢄmL-1 of penicillin, and
100
of 5% CO2.
m
gꢄmL-1 of streptomycin at 37 ꢁC in a humidified atmosphere
Cardiomyocyte survival rate The H9c2 cells were cultured in 96 -
well plates at a density of 5 ꢂ 104 cells/mL and cultured for 8 h at
37 ꢁC and 5% CO2 incubator for 8 h. The tests were divided into
[16] N.E.J. West, H.S. Qian, T.J. Guzik, et al., Nitric oxide synthase (nNOS) gene
transfer modifies venous bypass graft remodeling effects on vascular smooth
muscle cell differentiation and Superoxide production, Circulation 104 (2001)
1526e1532.
[17] M. Whiteman, J.S. Armstrong, S.H. Chu, et al., The novel neuromodulator
hydrogen sulfide: an endogenous peroxynitrite ‘scavenger’? J. Neurochem. 90
(2004) 765e768.
[18] M. Whiteman, L. Li, I. Kostetski, et al., Evidence for the formation of a novel
nitrosothiol from the gaseous mediators nitric oxide and hydrogen sulphide,
Biochem. Biophys. Res. Commun. 343 (2006) 303e310.
[19] R. Foresti, R. Motterlini, The heme oxygenase pathway and its interaction with
nitric oxide in the control of cellular homeostasis, Free Radic. Res. 31 (1999)
459e475.
[20] H.Y. Lin, S.H. Juan, S.C. Shen, et al., Inhibition of lipopolysaccharide-induced
nitric oxide production by flavonoids in RAW264. 7 macrophages involves
heme oxygenase-1, Biochem. Pharmacol. 66 (2003) 1821e1832.
[21] W.W. Wang, D.L.H. Smith, S.D. Zucker, Bilirubin inhibits iNOS expression and
NO production in response to endotoxin in rats, Hepatology 40 (2004)
424e433.
blank group, H2O2 group, complex(10
mM、50 mM、100 mM)-H2O2
group. H2O2 was 400 M, and the control group only added cells
m
without drugs. The cells survival rate was detected after 24 h. After
cells were treated with H2O2 for 1 h, 4 h and 8 h, respectively, the
culture medium was discarded and the H2O2 group was supple-
mented with fresh medium. Then the cell viability was measured
after 24 h of incubation.
Assays for enzyme activities The H9c2 cells were cultured in 6-
well plates overnight and then treated with H2O2 (400
another 1 h. Then A1-3 and complex 1e3 (10, 50, and 100
mM) for
mM) was
added and the cells were cultured for additional 24 h. The activities
of SOD and the MDA content were measured using commercial kits.
Conflicts of interest
[22] L.E. Otterbein, F.H. Bach, J. Alam, et al., Carbon monoxide has anti-
inflammatory effects involving the mitogen-activated protein kinase
pathway, Nat. Med. 6 (2000) 422e428.
[23] Yu Zhao, Chuntao Yang, et al., Design, synthesis, and cardioprotective effects
of N-Mercapto-Based hydrogen sulfide donors, J. Med. Chem. 58 (2015)
7501e7511.
The authors confirm that this article content has no conflict of
interest.
Acknowledgements
[24] Shakila Banu Ansari, Gino A. Kurian, Hydrogen sulfide modulates sub-cellular
susceptibility to oxidative stress induced by myocardial ischemic reperfusion
injury, Chem. Biol. Interact. 252 (2016) 28e35.
[25] Yanyan Li, Chao Ga, et al., Carbon monoxide alleviates ethanol-induced
oxidative damage and inflammatory stress through activating p38 MAPK
pathway, Toxicol. Appl. Pharmacol. 273 (2013) 53e58.
This work was financially supported by the National Natural
Science Foundations of China (no. 21171079) and the Recruitment
Program of Global Experts (1000 Talents Plan).
[26] L.C. Lin, F.M. Ho, S.J. Yen, et al., Carbon monoxide induces cyclooxygenase-2
expression through MAPKs and PKG in phagocytes, Int. Immunopharm. 10
(2010) 1520e1525.
Appendix A. Supplementary data
[27] R. Zhou, J. Si, H. Zhang, Z. Wang, J. Li, X. Zhou, Y. Liu, The effects of x-ray ra-
diation on the eye development of zebrafish, Hum. Exp. Toxicol. 33 (2014)
1040e1050.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2018.10.004.
[28] W. Zhang, K. Lin, Y. Miao, Q. Dong, C. Huang, H. Wang, X. Cui, Toxicity
assessment of zebrafish following exposure to CdTe QDs, J. Hazard Mater. 213
(2012) 413e420.
References
[29] Yaguo Gong, Taofeng Zhang, Meng Li, Na Li, Yawen Zheng, Quanyi Zhao,
Yonglin Chen, Bin Liu, Toxicity, bio-distribution and metabolism of CO-
releasing molecules based on cobalt, Free Radic. Biol. Med. 97 (2016) 362~374.
[30] R. Foresti, J.E. Clark, C.J. Green, et al., Thiol compounds interact with nitric
oxide in regulating heme oxygenase-1 induction in endothelial cells
Involvement of superoxide and peroxynitrite anions, J. Biol. Chem. 272 (1997)
18411e18417.
[1] C. Zoccali, C. Catalano, S. Rastelli, Blood Pressure Control: Hydrogen Sulfide, a
New Gasotransmitter, Takes Stage, Nephrology Dialysis Transplantation, 2009,
p. gfp053.
[2] O. Kabil, R. Banerjee, Redox biochemistry of hydrogen sulfide, J. Biol. Chem.
285 (2010) 21903e21907.
[3] N. Shibuya, Y. Mikami, Y. Kimura, et al., Vascular endothelium expresses 3-
mercaptopyruvate sulfurtransferase and produces hydrogen sulfide,
J. Biochem. 146 (2009) 623e626.