Leonurine-cysteine analog conjugates as a new class of multifunctional anti-myocardial ischemia agent

Article abstract of DOI:10.1016/j.ejmech.2011.05.073

The design, synthesis and biological evaluation of novel Leonurine-cysteine analog conjugates 3,5-dimethoxy-4-(2-amino-3-prop-2-ynylsulfanyl-propionyl)- benzoic acid 4-guanidino-butyl ester (1a), 3,5-dimethoxy-4-(2-animo-3- allysulfanyl-propionyl)-benzoic acid 4-guanidino-butyl ester (1b) and 3,5-dimethoxy-4-(3-(2-chlorocarbonyl-ethyldisulfanyl)-propionyl)-benzoic acid 4-guanidino-butyl ester (2) were reported in this paper. We tested their effects on hypoxia-induced neonatal rat ventricular myocytes. Our data showed that all of them had cardioprotective effects. Both of 1a and 1b were able to modulate hydrogen sulfide production, and 1a possessed higher biological activity than 1b and 2, which indicated that there was positive correlation between conjugates and their precursors. Furthermore we illuminated that the cardioprotective mechanism of 1a were related to increase SOD and CAT activity, decrease MDA and ROS level, protect some cell organs and regulate apoptosis-associated genes and proteins expression (bcl-2 and bax) via the caspase-3 pathway in molecular level. These results indicated that 1a had the potential to be a new class of multifunctional anti-myocardial ischemia agent. Most importantly, these results provided us important clues for the further design and modification of this type of Leonurine-cysteine analog conjugates in future.

Full text of DOI:10.1016/j.ejmech.2011.05.073

European Journal of Medicinal Chemistry 46 (2011) 3996e4009
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Leonurine-cysteine analog conjugates as a new class of multifunctional
anti-myocardial ischemia agent
,
a,
Chunhua Liu a, Wei Guo c, Xueru Shi a, M.A. Kaium c, Xianfeng Gu b , Yi Zhun Zhu
*
*
a Department of Pharmacology, School of Pharmacy and Institute of Biomedical Sciences, Fudan University, Shanghai 201203, China
b Department of Medicinal Chemistry, School of Pharmacy, Fudan University, Shanghai 201203, China
c Department of Pharmacology, School of Pharmacy, Fudan University, Shanghai 201203, China
a r t i c l e i n f o
a b s t r a c t
Article history:
The design, synthesis and biological evaluation of novel Leonurine-cysteine analog conjugates 3,5-
dimethoxy-4-(2-amino-3-prop-2-ynylsulfanyl-propionyl)-benzoic acid 4-guanidino-butyl ester (1a),
3,5-dimethoxy-4-(2-animo-3-allysulfanyl-propionyl)-benzoic acid 4-guanidino-butyl ester (1b) and 3,5-
dimethoxy-4-(3-(2-chlorocarbonyl-ethyldisulfanyl)-propionyl)-benzoic acid 4-guanidino-butyl ester (2)
were reported in this paper. We tested their effects on hypoxia-induced neonatal rat ventricular myo-
cytes. Our data showed that all of them had cardioprotective effects. Both of 1a and 1b were able to
modulate hydrogen sulfide production, and 1a possessed higher biological activity than 1b and 2, which
indicated that there was positive correlation between conjugates and their precursors. Furthermore we
illuminated that the cardioprotective mechanism of 1a were related to increase SOD and CAT activity,
decrease MDA and ROS level, protect some cell organs and regulate apoptosis-associated genes and
proteins expression (bcl-2 and bax) via the caspase-3 pathway in molecular level. These results indicated
that 1a had the potential to be a new class of multifunctional anti-myocardial ischemia agent. Most
importantly, these results provided us important clues for the further design and modification of this
type of Leonurine-cysteine analog conjugates in future.
Received 21 March 2011
Received in revised form
30 May 2011
Accepted 31 May 2011
Available online 14 June 2011
Keywords:
Leonurine-cysteine analog conjugates
Apoptosis
Cardioprotection
Crown Copyright Ó 2011 Published by Elsevier Masson SAS. All rights reserved.
1. Introduction
the activation of numerous intracellular signaling cascades and
causes apoptosis through accumulation of intracellular reactive
Ischemic heart disease (IHD), especially acute myocardial
ischemia (AMI), is a major cause of death and disability in both
developed and developing countries [1]. Sustained ischemia causes
several types of damage to cardiac tissues. Therefore, pharmaco-
logical strategies to prevent MI and thus reduce mortality rate are
of great importance in medicine. Recent evidence showed that
apoptosis of cardiac myocytes is a feature in several myocardial
disease states, including ischemic heart disease [2], which has
raised hopes that inhibition of myocyte apoptosis could prevent or
slow the loss of contractile cells and thus provide a new target for
the multimodal therapy of cardiac diseases [3]. Apoptosis is tightly
regulated by several mechanisms to prevent inadvertent cell loss.
Recent animals studies and clinical observations suggested that the
balance between anti-apoptotic Bcl-2 and pro-apoptotic Bax
protein plays a major role in regulating apoptosis [4]. In addition,
oxidative stress causes the formation of DNA adducts that leads to
oxygen species (ROS) [5]. ROS, possessing highly reactive and toxic
properties, generated as a result of ischemia, can affect significant
changes in membrane permeability, disrupt membrane lipid bila-
yers and mediate functional modifications of cellular proteins that
involved in cardiac tissue damage which exacerbate the degree of
myocardial damage sustained by the ischemic myocardium [4,6].
Therefore, preservation of cardiac tissues and cells from the dele-
terious effects of oxidative stress is another approach to treat
cardiac disease. In recent years, hydrogen sulfide (H2S) has been
discovered as the third gasotransmitter and gained much attention
due to its important biological effects [7]. The cardioprotection of
H2S is demonstrated in rat isolated ventricular myocytes by pyri-
doxal-5-phosphate-dependent enzyme cysthathionine-g-lyase
(CSE) pathway [8,9]. In addition, it has been reported that exoge-
nous H2S effectively protects myocytes and contractile activity, at
least by its direct scavenging of oxygen-free radicals and reducing
the accumulation of lipid peroxidations [6]. Thus, H2S can be a new
target to treat ischemic heart disease.
In view of the complex mechanisms of ischemic heart disease,
combination treatment is usually employed in clinical situations
* Corresponding authors. Tel.: þ86 21 51980112; fax: þ86 21 51980008.
E-mail addresses: xfgu@fudan.edu.cn (X. Gu), zhuyz@fudan.edu.cn (Y.Z. Zhu).
0223-5234/$ e see front matter Crown Copyright Ó 2011 Published by Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.05.073

C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
3997
[10]. In this regard, there is a great interest in the use of multitarget
drugs, also called polyvalent or multifunctional drugs, namely,
single products capable of interacting simultaneously with multiple
targets, directly or following metabolism [11,12]. The use of poly-
valent drugs shows some advantages over a cocktail of drugs,
which could indicate there was positive correlation between the
activities of conjugates and that of their precursors (SPRC, SAC and
3-(2-carboxy-ethyldisulfanyl)-propionic acid) [6], 1a was selected
to investigate the cardioprotective mechanism for this type of
conjugates related. Now we demonstrated that 1a, during hypoxia,
was able to increase CAT and SOD activity and decrease MDA and
ROS level so as to attenuate oxidative stress damage, protect some
cell organs in order to maintain their function, and regulate
apoptosis-associated genes and proteins expression (bcl-2 and
bax) via the caspase-3 pathway at lower molar concentration (10-
fold less than leonurine required and 100-fold less than SPRC
required) with some part of results had been published [14]. Most
importantly, these results provided us important clues for the
future design and modification of this type of Leonurine-cysteine
analog conjugates in future.
including
a lower risk of drugedrug interactions, improved
compliance by the patient, and a more predictable pharmacokinetic
profile [13].
“Medicinal chemical hybridization” (MCH) is one of the
approaches to design a polyvalent drug [11]. It can be carried out
by joining through an appropriate linker (conjugation) to two
drugs, the linker usually is susceptible to metabolic cleavage. An
ideal polyvalent drug can usually augment the potency of both
drugs or reduce the dosage [13,14]. Our group have reported that
Leonurine has cardioprotective effects both in vivo and in vitro
[15,16]. We also showed that several cysteine analogs, such as S-
allyl-
argyl-
L
-cysteine (SAC) (an effective component in garlic), S-prop-
2. Chemistry
L-cysteine (SPRC) (also called ZYZ802) which had been
synthesized in our laboratory and were novel endogenous
hydrogen sulfide (H2S) modulators, demonstrated potent protec-
tive effects on acute myocardial ischemia through regulating
endogenous H2S production by the CSE pathway [6,17]. Inspired by
the approach of MCH and pharmacological results about Leonur-
ine and cysteine analogs reported by our group, we designed and
synthesized Leonurine-cysteine analog conjugate through ester-
atic linkage at the phenolic hydroxyl group (Fig. 1). These novel
compounds were evaluated on neonatal rat ventricular myocytes
in vitro, we intended to investigate what are relations between the
activities of the cysteine analogue precursors and their corre-
sponding conjugates, which signal pathway and cell organs are
affected by this type of conjugate during hypoxia. Our data
showed that these novel compounds had potent cardioprotective
effect at lower molar concentration [14]. Both of 1a and 1b were
able to modulate hydrogen sulfide production. As 1a possessed
higher biological activity than 1b and 2 at same concentration,
Compound 5 is the key intermediate condensated with Boc-
protected cysteine analogs or their carboxylic chloride to build up
target compounds 1a, 1b and 2. To prepare key intermediate 5, the
condensation between 4-hydroxy-3,5-dimethoxy-benzoic acid 4-
amino-butyl ester 3 [18] and N,N0-Boc-methyl-isothiourea 4 in
DMF was tried. As compound 3 is insoluble in DMF under room
temperature, the reaction was performed at 60 ꢀC. However,
intermediate 5 was afforded with only 5% yield, and Boc depro-
tected byproduct was also isolated with 30% yield (Scheme 1).
Considering that the Boc protection groups are unstable under
higher temperature, another synthetic route for preparing key
intermediate 5 was explored. In this strategy, 4-hydroxyl butyl-
amine instead of 4-hydroxy-3,5-dimethoxy-benzoic acid 4-amino-
butyl ester 3 to condensate with N,N0-Boc-methyl-isothiourea 4 at
room temperature, and afford intermediate 6 with high yield (97%).
Subsequently, the resulting intermediate 6 condensated with
compound 7 in the presence of N,N0-Diisopro-pylcarbodiimide
Fig. 1. Leonurine, cysteine analogs and their conjugate compounds.

3998
C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
Scheme 1. The synthetic route of the leonurine and 5. Reaction conditions: (i) (Boc)2O NaHCO3,CHCl2, stir at room temperature for 7d; (ii) in DMF, 60 ꢀC, stir at room temperature
for 24 h.
(DIC) and 4-(dimethylamino)-pyridinium-4-toluene sulfonate
(DPTS) to afford 8 in 70% yields, wherein compound 7 was prepared
via the reaction of caryophyllic acid and acetyl chloride in the
presence of triethylamine. Finally, the acetyl group in compound 8
was hydrolyzed by catalytic amount of sodium methoxide at 4 ꢀC to
give the key intermediate 5 in 98% yields, and total yield from 4-
hydroxyl butylamine 3 is above 60% (Scheme 2).
To synthesize target compound 1a and 1b, cysteine analogs such
as S-propargyl cysteine 9a and S-allyl cysteine 9b were also
obtained from the reaction of cysteine with propargyl bromide or
allyl bromide based on literature method [17]. And the amino group
of 9a and 9b was protected by Boc anhydride under basic condition
to afford 10a and 10b in good yield [19] which subsequently
condensed with intermediate 5 in the presence of DPTS and DIC to
afford 11a and 11b. Finally the Boc groups in 11a and 11b was
removed by trifluoroacetic acid (TFA) to give target compound 1a
and 1b in good yield (above 70%) (Scheme 3). Target compound 2
was synthesized from 3-(2-carboxy-ethyldisulfanyl)-propionic acid
which firstly reacted with oxalyl chloride, and converted into its
carboxylic chloride. The resulting carboxylic chloride subsequently
condensated with key intermediate 5 in the presence of triethyl-
amine, and afforded 13. Finally the Boc groups in 13 was removed
by trifluoroacetic acid (TFA) to give target compound 2 in good yield
(above 60%) (Scheme 4).
3. Results and discussion
3.1. Protective effects of compounds 1a, 1b and 2 on hypoxia-
induced neonatal rat ventricular myocytes
The protective effects of compounds 1a, 1b and 2 were inves-
tigated by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT) assay which demonstrates cell viability [6]. It
showed that the hypoxic condition had an effect on neonatal rat
ventricular viability (P < 0.01). 1a, 1b and 2 significantly increased
cell viability at 0.1 mmol/L (P < 0.05) (Fig. 2A). Compared with the
model group, compound 1a, 1b and 2 remarkably inhibited LDH
leakage which is the marker of cell membrane integrity (Fig. 2B) [6],
and the LDH leakage of 1a and 2 were much lower than that of 1b
(P < 0.05). As shown in Fig. 2C, hypoxia potently inhibited H2S level
on neonatal rat ventricular myocytes, whereas 1a and 1b treated
groups significantly ameliorate the level of H2S (P < 0.05). Unfor-
tunately compound 2 treated group failed to improve the level of
H2S (P > 0.05). These data showed that 1a had better car-
dioprotective effects than those of 1b and 2 at the same concen-
tration, which indicated that there was positive correlation
between these conjugates with their precursors (SPRC, SAC and 3-
(2-carboxy-ethyldisulfanyl)-propionic acid) [6]. Recent studies
shows that hydrogen sulfide has cardioprotective effects. It is well
Scheme 2. The synthetic route of the key intermediate 3,5-dimethoxy-4-hydroxy-benzoic acid 4-(N,N0-Boc-guanidino)-butyl ester (5). Reaction conditions: (i) in DMF, stir at room
temperature for 3.5 h; (ii) Triethylamine, acetyl chloride, in dichloromethane, stir at room temperature for 2 h; (iii) DIC, DPTS stir at room temperature for 5 h; (iv) sodium
methoxide, 4 ꢀC, stir at room temperature for 1 h.

C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
3999
Scheme 3. The synthetic route of the 1a and 1b. Reaction conditions: (i) Di-tert butyl dicarbonate, in tetrahydrofuran and Na2CO3, stir at room temperature for 5 h; (ii) DIC, DPTS,
stir at room temperature for 5 h; (iii) TFA.
known that hydrogen sulfide could be produced from cysteine
catalyzed by CBS, CSE or 3MST/CAT in cells [7]. And methione, as
one of sulfur containing natural amino acid, is able to produce
homocysteine in biosystem via CeS bond cleavage reaction. With
the knowledge above in mind, SAC (Fig. 3), a cysteine analog
derived from garlic, was chosen as a mediator of hydrogen sulfide
production in our study. And we found that SAC has car-
dioprotective effect via hydrogen sulfide pathway [19]. To get more
potent hydrogen sulfide mediator, we further synthesized SPRC
(Fig. 3) in which a propargyl group instead of allyl group in SAC [17].
Chemically speaking, alkynyl is a stronger electron withdrawing
group than alkenyl because hybrid orbital of alkynyl is sp, whereas
that of alkenyl is sp2 (Fig. 3). We predicted that once the propargyl
group instead of allyl group, the density of electron cloud of
carbon between alkynyl and sulfur atom of SPRC become thinner
because of electron withdrawing effects of alkynyl group and surful
atom. And then the carbon is more apt to be attacked by nucle-
ophiles to produce cysteine which could be further catalyzed by
CBS, CSE or 3MST/CAT to produce hydrogen sulfide. As expected,
our studies showed that SPRC was more potent than SAC at the
same concentration [6,17] and also, compound 1a (SPRC-Leonurine
conjugate) was more potent than compound 1b (SAC-Leonurine
a
a
conjugate) at 0.1 mmol/L, which indicated that there was positive
correlation between these conjugates with their precursors.
Thenwe compared 1awith its precursors and data shown that 1a
possessed potent cardioprotective effects against hypoxia-induced
Scheme 4. The synthetic route of the 2. Reaction conditions: (i) Oxalyl Chloride, in dry DCM, stir at room temperature for 12 h; (ii) triethylamine, in dry DCM, stir at room
temperature for 5 h; (iii) TFA.

4000
C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
Fig. 2. (A) Effects of compounds 1a, 1b and 2 on cell viability in hypoxia-induced neonatal rat ventricular myocytes. The viability of control cell is presumed to be 100%. (B) Effects of
1a, 1b and 2 on LDH leakage in hypoxia-induced neonatal rat ventricular myocytes. (C) Effects of 1a, 1b and 2 on H2S content in hypoxia-induced neonatal rat ventricular myocytes.
1a, 1b and 2 were pretreated for 12 h before hypoxia. C denotes control group, M denotes model group, 1a denotes compound 1a-treated group, 1b denotes compound 1b-treated
#
group, 2 denotes compound 2-treated group. Values are expressed as means ꢁ S.E.M from three individual samples. *denotes P < 0.05 versus model group; denotes P < 0.05,
##denote P < 0.01 versus control group; $denote P < 0.05 versus 1a group.
neonatal rat ventricular myocytes damage at lower molar concen-
tration (10-fold less than leonurine required and 100-fold less than
SPRC required) (Fig. 4) [14]. We proposed the mechanism for
compound 1a as follows. First of all, Leonurine and SPRC, the
precursor of compound 1a, are zwitterion in aqueous system. The
phenolic hydroxyl group of leonurine and the carboxyl group of
SPRC are able to form negative charge by deprotonation, and the
guandine group of leonurine and amino group of SPRC are capable of
Fig. 3. Proposed mechanism of hydrogen sulfide formation mediated by 1a and 1b.

C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
4001
Fig. 4. (A) Effect of 1a, LEO and SPRC on hypoxia-induced neonatal rat ventricular myocytes; (B) Effects of 1a, LEO and SPRC on LDH leakage in hypoxia-induced neonatal rat
ventricular myocytes; (C) Effects of 1a, LEO and SPRC on H2S content in hypoxia-induced neonatal rat ventricular myocytes. C denotes control group, M denotes model group, 1a
denotes compound 1a-treated group, LEO denotes leonurine-treated group, SPRC denotes SPRC-treated group. Values are expressed as means ꢁ S.E.M from six individual samples.
*denotes P < 0.05 **denotes P < 0.01 versus vehicle group; #denotes P < 0.05, ##denote P < 0.01 versus control group.
producing positive charge by protonation. In our study, the phenolic
hydroxyl group of leonurine are esterified by SPRC to form more
lipid soluble and positive charge containing molecule-compound
1a, and this modification could help compound 1a transport into
cell much more easier than its precursor. Then compound 1a may be
easily hydrolyzed and readily release bioactive leonurine and SPRC
to show the synergism at the same time in cell. In this way, it would
have a lower risk of drugedrug interactions compared to the
mixture of leonurine and SPRC.
of anti-oxidant or free radical scavenging activity. In our study, we
observed an increase in catalase activity in 1a treated-group
(Fig. 5B) (P < 0.05). These results suggested 1a was able to attenuate
oxidative stress damage not only by remarkably degrading MDA
and ROS level, but also by significantly enhancing SOD and CAT
activity. As Leonurine and SPRC are potent substances in anti-
oxidative stress [6,22], the anti-oxidative effects of 1a could be
related to the release of Leonurine and SPRC, due to labile hydro-
lysis of its ester bond in body.
3.2. Effects of compound 1a on CAT, SOD, MDA and ROS in hypoxia-
induced neonatal rat ventricular myocytes
3.3. Ultrastructural and morphological changes of neonatal rat
ventricular myocytes
It is reported that oxidative stress is involved in ischemic heart
disease (IHD) [20]. And reactive oxygen species (ROS) is one of the
production of oxygen-derived species (ODS) during oxidative
stress. ROS have either unpaired electrons (e.g. Oꢂ2ꢃ, OHꢂ) or the
ability to attract electrons from other molecules (e.g. H2O2). Such
intermediates react with cellular DNA or macromolecules, either
damaging them directly or setting in motion a chain reaction
resulting in extensive damage to cellular structures, such as DNA
skeletons or cellular membranes [21]. In our study, we observed
a decrease in the ROS, and then followed by decreased MDA level in
1a treated-group (Fig. 5A) compared with model group, which
indicated that 1a have the ability to attenuate oxidative stress
damage during hypoxia. SOD and CAT are important anti-oxidative
enzymes and their activities are the marker for evaluating the effect
Apoptosis is a genetically well controlled programmed cells
death. Its chief morphologic features include cell shrinkage, chro-
matin condensation, formation of cytoplasmic blebs and apoptotic
bodies, preservation of membrane integrity and no inflammatory
response [2,23]. Liu et al. [15,24] and Fliss and Gattinger [25] had
reported that apoptosis can be induced by hypoxia or ischemia. And
we also demonstrated and confirmed that hypoxia indeed induced
apoptosis in neonatal rat ventricular myocytes, and treatment with
compound 1a resulted in less nuclei-shrunk and nuclear conden-
sation (Fig. 6A) [14]. In our present study, transmission electron
microscope was performed to observe the ultrastructural
morphological changes in the subcellular level. Ventricular myo-
cytes in control group showed little or no morphological changes,
nucleus were well preserved with finely dispersed chromatin and

4002
C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
Fig. 5. (A) Effect of 1a on MDA and ROS level on hypoxia-induced neonatal rat ventricular myocytes. (B) Effect of 1a on SOD and CAT activity in hypoxia-induced neonatal rat
ventricular myocytes. C denotes control group, M denotes model group, 1a denotes compound 1a-treated group. Values are expressed as means ꢁ S.E.M from three individual
samples. *denotes P < 0.05 versus model group; #denotes P < 0.05, ##denotes P < 0.01 versus control group.
prominent nucleolus. However, in model group, the nucleus
showed partial aggregation, the mitochondria displayed separation
or collapse of cristae and swelling of the matrix space, and the
myofilaments were disrupted and sarcomeres had disappeared. In
1a treated group, mitochondria displayed separation of cristae and
swelling of the matrix space and their arrangement were disor-
dered to some extent. However, when compared with the model
group, nucleus were less aggregated, myofilaments had preserved
their integrity and sarcomeres were distinct and visible (Fig. 6B).
Although we had reported anti-apoptosis of 1a with preventing
nuclei-shrunk which was preliminary morphological changes by
fluorescence microscope. Now our present work first exactly
pointed out that 1a was able to protect some cell organs, such as
nuclei, myofilaments and mitochondria, in order to maintain their
functions.
Then anti-apoptosis of 1a was further confirmed by AnnexinV-
FITC/PI with flow cytometry which was able to exactly distin-
guished the apoptotic cell and necrotic cell from total cells and
much more precise than hoechst staining (Fig. 7). The results
showed that 1a remarkably decreased the number of apoptotic cell
(P < 0.05), and 1a failed to inhibited the necrosis (P > 0.05) induced
by hypoxia.
suggested as a major controlling point in the pathway to apoptotic
cell death. The ratio of Bcl-2/Bax has been suggested as a marker for
determining survival or death after an apoptotic stimulus [26]. In
the present study, hypoxia significantly decreased the expression of
Bcl-2 and increased the expression of Bax, consistent with the
suggested roles of these genes in regulation of apoptotic cell death.
As shown in Fig. 8A, B, C and D, hypoxia severely inhibited bcl-2
expression (0.53-fold) and promoted bax expression (1.66-fold)
so as to reduce bcl-2/bax ratio to 0.36 in neonatal rat ventricular
myocytes (P < 0.01), whereas 1a remarkably reversed the effects of
hypoxia on bcl-2 (0.98-fold) and bax (0.89-fold) expressions to
increase bcl-2/bax ratio to 1.22 (P < 0.01).
A family of caspases is another key regulator of apoptotic
signaling pathway [27]. Generally, apoptotic caspases are catego-
rized into initiator (caspase-8, -9) and executioner caspases (cas-
pase-3, -6, -7) [28]. Inactive procaspases that exist as latent
zymogens under normal conditions are cleaved into their active
forms via other activated caspases in the apoptotic process. The
cleavage of procaspases, an irreversible post-translational modifi-
cation, has been used as an indicator of apoptosis. Pro-apoptotic
caspase-3 consists of two subunits, p12 and p17, that form a het-
erodimer to build the active protease. Active executioner caspase-3
can further cleave down-stream substrates involved in apoptotic
changes, such as poly-(ADP-ribose) polymerase [29]. As for gene
expression of the caspase-3 in Fig. 8E, its expression was relatively
weak in the control group. Hypoxia significantly increased caspase-
3 expression, whereas 1a effectively down-regulated its expression
(P < 0.05).
3.4. Effects of compound 1a on genes and proteins expression of
bcl-2, bax and caspase-3 in hypoxia-induced neonatal rat
ventricular myocytes
Although previous data showed 1a was able to prevent nuclei-
shrunk of hypoxia-induced neonatal rat ventricular myocytes, its
precise mechanism had not yet been eluminated. As known, many
genes have been reported to be associated with the regulation of
programmed cell death, especially the Bcl-2 family has been
In protein level, bcl-2 and bax showed consistent results with RT-
PCR (Fig. 9AeD). In addition, we observed that caspase-3 had been
cleaved and activated by hypoxia. At the same time,1a was also able
to remarkably reducing the cleaved-caspase-3/pro-caspase-3 ratio

C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
4003
Fig. 6. (A) Effects of 1a on morphological features of cultured neonatal rat ventricular myocytes. Fluorescence photomicrographs of cells stained with Hoechst 33258 (ꢄ400). Each
photograph is representative of three independent observations. (B) Effect of 1a on ultrastructure changes in hypoxia-induced neonatal rat ventricular myocytes with electron
microscopy (ꢄ11000). 1a was pretreated for 12 h before hypoxia. Nu denotes nucleus; Mit denotes mitochondria; mf denotes myofilament and ‘ ’ denotes sarcomere. Control
denotes control group, Model denotes model group, 1a denotes compound 1a-treated group.
(Fig. 9E) so as to inhibit the apoptotic process. Now our data first
demonstrated that 1a played the anti-apoptotic role by regulating
apoptosis-associated genes and proteins expression by caspase-3
pathway in molecular level.
collected on a HP5973 N analytical mass spectrometer. HRMS data
were determined on an IonSpec 4.7 T FTMS instrument. All
reagents of cell culture were purchased from Sigma (St. Louis, MO)
unless otherwise stated. The following substances Dulbecco’s
Modified Eagle medium (DMEM) and calf serum were purchased
from Invitrogen Inc. (MD, USA). All antibodies used for Western
blotting in this study were from Cell Signaling Technology, Inc. (MA,
US). The kit for catalase (CAT) was obtained from Jiancheng
Bioengineering Institute (Nanjing, China). The kit for ROS and
AnnexinV-FITC/PI were obtained from Beyotime Bioengineering
Institute (Lianyungang, China).
4. Conclusion
We have designed and successfully synthesized a series of novel
Leonurine-cysteine analog conjugate (e.g. 1a, 1b and 2). Pharma-
cological evaluation had shown their potent cardioprotective
effects. Both of 1a and 1b were able to modulate hydrogen sulfide
production. As shown in this paper, 1a had better cardioprotective
effects than those of 1b and 2 at the same concentration, which
indicated there was positive correlation between Leonurine-
cysteine analog conjugate and their precursors (SPRC, SAC and 3-
(2-carboxy-ethyldisulfanyl)-propionic acid) [6]. In our work we
detailed suggested that the cardioprotective mechanisms of 1a
were related to improve anti-oxidative enzyme activity, attenuate
ROS and MDA level, protect some cell organs and regulate
apoptosis-associated genes and proteins expression mainly by
caspase-3 pathway during hypoxia. Furthermore these results
provided us important clues for the future design and modification
of this type of Leonurine-cysteine analog conjugates in our next
study and supported 1a to be a new class of multifunctional anti-
myocardial ischemia agent.
5.1. N,N0-Boc-methyl-isothiourea (4)
The product of N,N0-Boc-methyl-isothiourea was obtained as
described previously in Ref. [30]. Briefly, S-methylisothiourea
(5.56 g, 20 mmol) was dissolved in CH2Cl2 (15 mL) and saturated
with NaHCO3 (15 mL) and (Boc)2O (17.44 g, 80 mmol) was added to
this solution. The solution was stirred for 7 days and diluted with
H2O (15 mL) and extracted with CH2Cl2 (3 ꢄ 20 mL). The combined
organic extracts were washed with brine and dried with Na2SO4.
The solvent was removed under vacuum and the crude material
was purified by silica gel chromatography to give a white solid
(5.51 g, 95.1%). 1H NMR(400 MHz, CDCl3)
2.39(s, 3H, eCH3), 11.60(s, 1H, eNH); 13C NMR(100 MHz, CDCl3)
14.36, 27.98, 171.43; MS(EI) (m/z): 291.1 (Mþ).
d 1.51(s, 18H, eBoc),
d
5. Materials and methods
5.2. N,N0-Boc-N00-(4-hydroxy-butyl)-guanidine (6)
Starting materials and reagents of chemical synthesis were
obtained from commercial suppliers and were used without puri-
fication. Nuclear magnetic resonance spectra were recorded on
a Brucker-DPX 400 MHz spectrometer, Mass spectral data was
4-amino-1-butanol (890 mg, 10 mmol) and N,N0-Boc-2-methyl-
isothiourea (2.9 g, 10 mmol) were dissolved and stirred in N,N0-
dimethyl formamide (20 mL) at room temperature for 3.5 h, The

4004
C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
concentrated in vacuo. The crude material was purified by silica
gel chromatography to give 7.96
(66.4%) of 7. 1H NMR
(400 MHz, CDCl3) 2.37(s, 3H,eOCOCH3), 3.90(s, 6H, 3,5-OCH3),
7.40(s, 2H, AreH), 11.67(br, 1H, COOH); 13C NMR(100 MHz, CDCl3)
g
d
d
20.4, 56.3, 106.8, 127.1, 133.3, 152.1, 168.2, 171.5; MS(ESI) (m/z)
239.2 [Mꢃ].
5.4. 3,5-Dimethoxy-4-acetoxy-benzoic acid 4-(N,N0-Boc-guanidino)-
butyl ester (8)
4-Acetoxy-3,5-dimethoxy-benzoic acid 7 (1.2 g, 5 mmol), N,N0-
tertiary butyl-oxycarbonyl-N00-(4-hydroxy-butyryl-)-guanidine(1.1 g,
3.3 mmol), N,N0-Diisopropyl-carbodiimide (DIC) (1.74 mL, 6.6 mmol)
and DPTS (2.06 g, 6.6 mol) were added in DCM (50 mL). The reaction
was allowed to stir at room temperature for 5 h. The reaction mixture
was diluted with DCM and washed with water three times. The
organic layer was separated, dried over Na2SO4, and solvent removed
by rotary evaporation. The product was dissolved in DCM and was
chromatographed on silica gel, eluting with 15% EtOAc/PE, to give
1.76 g of white solid (63.6%). 1H NMR(400 MHz, CDCl3)
d 1.48 (s, 9H,
BoceCH3), 1.49(s, 9H, BoceCH3), 1.68e1.73(m, 2H, eCH2CH2e),
1.80e1.87(m, 2H, eCH2CH2e), 2.35(s, 3H, eOCOCH3), 3.48e3.53(m,
2H, eCH2NHe), 3.87(s, 6H, eOCH3), 4.34(t, 2H, J ¼ 6.26 Hz,
eCOOCH2), 7.31(s, 2H, 2,AreH), 8.37(s, 1H, (CH2)4NHe), 11.50(s, 1H,
eCNHBoc); 13C NMR(100 MHz, CDCl3)
d 20.4, 25.8, 26.2, 28.0, 28.2,
40.4, 56.3, 64.8, 79.4, 83.2,106.2,128.2,132.5,152.0,153.3,156.1,165.8,
168.2; MS (ESI) (m/z): 554.0 [Mþ]; HRMS calculated mass for
C26H40N3O10 554.26355, found 554.27084.
5.5. 3,5-Dimethoxy-4-hydroxy-benzoic acid 4-(N,N0-Boc-guanidino)-
butyl ester (5)
3,5-dimethoxy-4-acetoxy-benzoic acid 4-(N,N0-Boc-guanidino)-
butyl ester 8 (5.9 g) was dissolved in DCM (50 mL) at 4 ꢀC, and 3
drops of sodium methoxide solution in MeOH was added and
stirred for 1 h. Dowex-50W acid cation-exchange resin was added
to adjust pH to 7 and then filtered, and solvent was removed under
reduced pressure to give 5.4 g of white solid 5 (98.9%). 1H
NMR(400 MHz, CDCl3)
d
1.48e1.49(d, 18H, J ¼ 5.48, Boc), 1.70e1.76
(m, 2H, eCH2CH2e), 1.79e1.87 (m, 2H, eCH2CH2e), 3.48e3.53 (m,
2H, eCH2NHe), 3.94(s, 6H, eOCH3), 4.32(t, 2H, J ¼ 6.26 Hz,
eCOOCH2), 7.30(s, 2H, AreH), 8.38(s, 1H, (CH2)4NHꢃ), 11.50(s, 1H,
Fig. 7. Anti-apoptotic effect of 1a in hypoxia-induced neonatal rat ventricular myo-
cytes confirmed by AnnexinV-FITC/PI. 1a was pretreated for 12 h before hypoxia. C
denotes control group, M denotes model group, 1a denotes compound 1a-treated
group. Values are expressed as means ꢁ S.E.M from three individual samples. *denotes
P < 0.05 versus model group; #denotes P < 0.05 ##denotes P < 0.01 versus control
group.
eCNHBoc); 13C NMR(100 MHz, CDCl3)
d 25.8, 26.2, 28.0, 28.2, 40.5,
56.4, 64.4, 79.4, 83.1, 106.5, 121.1, 139.1, 146.6, 153.2, 156.1, 166.3;
MS(ESI) (m/z) 512.0 [Mþ]; HRMS calcd mass for C24H38N3O9
512.25298; found 512.26048.
reaction mixture was then diluted with DCM and washed with
water 3 times. The organic layer was separated, dried over Na2SO4,
and solvent was removed by rotary evaporation. The product was
dissolved in DCM and was chromatographed on silica gel to give
3.22 g of white oleaginous solid (97.3%). 1H NMR(400 MHz, CDCl3)
5.6. 2-(N-Boc-amino)-3-prop-2-ynylsulfanyl-propionic acid (10a)
2-(N-Boc-amino)-3-prop-2-ynylsulfanyl-propionic acid 10a was
made as literature method [31]. Briefly 2-amino-3-prop-ynylsul-
fanyl-propionic acid (1.59 g, 10 mmol), Boc anhydride (2.18 g,
10 mmol) were both dissolved and stirred in tetrahydrofuran (THF)
(25 mL) and saturated with NaHCO3 (60 mL) for 5 h. The solution
was then diluted with H2O (15 mL) and EtOAc (15 mL), then pH was
adjusted to about 2e2.5, then separate the organic extracts which
were washed with brine and dried over Na2SO4. The solvent was
removed under vacuum and the crude material was purified by
silica gel chromatography to give 10a as oleaginous solid (1.823 g,
d
1.47(s, 9H, eBoc), 1.48(s, 9H, eBoc), 1.57e1.70(m, 4H, eCH2CH2e),
2.28(br, 1H, eOH), 3.41e3.46(m, 2H, NH2CH2), 3.66e3.69(m, 2H,
CH2OH), 8.38(s, 1H, eNH), 11.47(s, 1H, NHBoc); 13C NMR(100 MHz,
CDCl3)
d 25.5, 28.0, 28.2, 29.4, 40.3, 62.1, 79.3, 83.1, 153.2, 156.1,
163.4; MS(ESI) (m/z): 332.4 [Mþ].
5.3. 4-Acetoxy-3,5-dimethoxy-benzoic acid (7)
70.4%). 1H NMR(400 MHz, CDCl3)
d 1.46(s, 9H, Boc), 3.01e3.06(m,
Caryophyllic acid (9.9 g, 50 mmol), triethylamine (1.392 mL,
1 mmol) and acetyl chloride (3.56 mL, 55 mmol) were mixed
together in DCM (20 mL) and stirred at room temperature for 2 h.
The mixture was extracted with DCM and brine, and then
1H, eCCH), 3.09e3.15(m, 2H, CH2S), 3.25e3.35(m, 2H, SCH2),
4.55e4.59(m,1H, NHCH), 5.45(d,1H, J¼ 7.83 Hz, eNH),10.43(s,1H,eCOOH);
13C NMR(100 MHz, CDCl3)
d 20.0, 28.2, 33.5, 53.0, 72.0, 79.2, 80.6, 155.5,
174.9; MS(ESI) (m/z): 258.1 [Mꢃ].

C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
4005
Fig. 8. Effects of 1a on the gene expression of bcl-2, bax and caspase-3 in hypoxia-induced neonatal rat ventricular myocytes (A). Quantitative analysis of the gene of bcl-2 (B), bax
(C), and bcl-2/bax ratio(D) caspase-3 (E). 1a was pretreated for 12 h before hypoxia. C denotes control group, M denotes model group, 1a denotes compound 1a-treated group. The
value at each group represents the normalized mean ꢁ S.E.M from three individual samples. *denotes P < 0.05, **denotes P < 0.01 versus model group; ##denotes P < 0.01 versus
control group.
5.7. 2-(N-Boc-amino)-3-allysulfanyl-propionic acid (10b)
5.8. 3,5-Dimethoxy-4-(2-(N-Boc-amino)-3-prop-2-ynylsulfanyl-
propionyl)-benzoic acid 4-(N,N0-Boc-guanidino)-butyl ester (11a)
The product of 2-(N-Boc-amino)-3-Allysulfanyl-propionic acid
10b was made as described previously in Ref. [31]. Briefly 3-
allylsulfanyl-2-amino-propionic acid (1.61 g, 10 mmol), Boc
anhydride (2.18 g, 10 mmol) were dissolved and stirred in THF
(25 mL) and saturated with NaHCO3 (60 mL) for 5 h. The solution
was then diluted with H2O (15 mL) and EtOAc (15 mL), then pH
was adjusted to about 2e2.5, the organic extracts were then
separated and washed with brine and dried over Na2SO4. The
solvent was removed under vacuum and the crude material was
purified by silica gel chromatography to give 10b as oleaginous
3,5-dimethoxy-4-hydroxy-benzoic acid 4-(N,N0-Boc-guani-
dino)-butyl ester 5 (302 mg, 0.6 mmol), 2-(N-Boc-amino)-3-prop-
2-ynylsulfanyl-propionic acid 10a (155.4 mg, 0.6 mmol), N,N0-
Diisopro-pylcarbodiimide (75.5 g, 0.6 mmol) and DPTS (187 mg,
0.6 mmol) were dissolved and stirred in DCM (50 mL) at room
temperature for 5 h, The reaction mixture was diluted with DCM
and washed with water three times. The organic layer was sepa-
rated, dried over Na2SO4, and solvent removed by rotary evapo-
ration. The crude product was dissolved in DCM and was
chromatographed on silica gel, and eluted with 15% EtOAc/PE, to
give 312 mg of white oleaginous solid 11a (70.1%). 1H
solid (2.545 g, 97.5%). 1H NMR(400 MHz, CDCl3)
d 1.42(s, 9H, Boc),
2.84e2.97(m, 2H, CH2S), 3.11(d, 2H, J ¼ 7.05 Hz, SCH2), 4.49(dd,
1H, J ¼ 5.09, 12.52 Hz, CHNH), 5.08(s, 1H, CHCH2), 5.12(d, 1H,
J ¼ 5.87 Hz, CHCH2), 5.42(d, 1H, J ¼ 8.82 Hz, eNH), 5.67e5.77(m,
1H, CHCH2), 11.05(s, 1H, eCOOH); 13C NMR(100 MHz, CDCl3)
NMR(400 MHz, CDCl3)
d 1.45(s, 9H, Boc), 1.47(s, 9H, Boc), 1.48(s,
9H, Boc), 1.67e1.75(m, 2H, eCH2CH2e), 1.79e1.86(m, 2H,
eCH2CH2e), 2.27(t, 1H, J ¼ 2.35 Hz, eCCH), 3.19e3.45(m, 4H,
CH2SCH2e), 3.46(dd, 2H, J ¼ 7.04, 12.52 Hz, CH2NH), 3.85(s, 6H,
eOCH3), 4.32(t, 2H, J ¼ 6.26 Hz, eCOOCH2e), 4.91e4.94(m, 1H,
d
28.3, 32.5, 35.1, 53.1, 80.4, 118.0, 133.6, 155.5, 175.7; MS(ESI) (m/
z): 260.2 [Mꢃ].

4006
C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
Fig. 9. Effects of 1a on the protein expression of Bcl-2, Bax and Caspase-3 in hypoxia-induced neonatal rat ventricular Myocytes (A). Quantitative analysis of the protein of Bcl-2 (B),
Bax (C), Bcl-2/Bax ratio (D) and cleaved-caspase-3/pro-caspase-3 ratio (E). 1a was pretreated for 12 h before hypoxia. C denotes control group, M denotes model group, 1a denotes
compound 1a-treated group. The value at each group represents the normalized mean ꢁ S.E.M from three individual samples. *denotes P < 0.05, **denotes P < 0.01 versus model
group; ##denotes P < 0.01 versus control group.
OOCCHNHe), 5.43(d, 1H, J ¼ 6.22 Hz, NHCe), 7.28(s, 2H, AreH),
8.35e8.37(m, 1H, CNHBoce), 11.49(s, 1H, OOCCHNHe); 13C
NMR(100 MHz, CDCl3) d 20.2, 25.8, 26.1, 28.0, 28.2, 34.3, 40.4, 53.2,
were dissolved and stirred in DCM (50 mL) at room temperature for
5 h, The reaction mixture was diluted with DCM and washed with
water three times. The organic layer was separated, dried over
Na2SO4, and solvent was removed by rotary evaporation. The crude
product was dissolved in DCM and was chromatographed on silica
gel, and eluted with 15% EtOAc/PE, to give 356 mg of white oleag-
56.2, 64.8, 71.8, 79.3, 79.5, 80.2, 83.2, 106.1, 128.6, 132.0, 151.8,
153.3, 155.0, 156.2, 163.5, 165.7, 168.5; MS(ESI) (m/z) 753.4 [Mþ];
HRMS calculated mass for C35H53N4O12
753.33654.
S 753.33025, found
inous solid 11b (80.1%). 1H NMR(400 MHz CDCl3)
d 1.46(s, 9H, Boc),
1.47(s, 9H, Boc), 1.48(s, 9H, Boc), 1.69e1.75(m, 2H, eCH2CH2e),
1.79e1.86(m, 2H, eCH2CH2e), 3.01e3.17(m, 2H, CH2Se), 3.21(d, 2H,
J ¼ 7.44 Hz, SCH2e), 3.46(dd, 2H, J ¼ 6.26, 12.72 Hz, CH2NH), 3.84(s,
6H, OCH3), 4.32e4.35(m, 2H, eCOOCH2), 4.83e4.88(m, 1H, CNHe),
5.11e5.17(m, 2H, eCHCH2), 5.41(d, 1H, J ¼ 8.22 Hz, OOCCHNHe),
5.73e5.84(m, 1H, eCHCH2), 7.28(s, 2H, AreH), 8.36(s, 1H, CNHe),
5.9. 3,5-Dimethoxy-4-(2-(N-Boc-amide)-3-allysulfanyl-propionyl)-
benzoic acid 4-(N,N0-Boc-guanidino)-butyl ester (11b)
3,5-dimethoxy-4-hydroxy-benzoic acid 4-(N,N0-Boc-guani-
dino)-butyl ester (302 mg, 0.6 mmol), 2-(N-Boc-amino)
-3-allysulfanyl-propionic acid (156 mg, 0.6 mmol), N,N0-Diisopro-
pylcarbodiimide (75.5 mg, 0.6 mmol) and DPTS (187 mg, 0.6 mmol)
11.49(s, 1H, NHeBoc); 13C NMR(100 MHz, CDCl3)
28.2, 33.2, 35.4, 40.4, 56.2, 64.8, 79.3, 80.1, 83.1, 106.2, 117.9, 128.5,
d 25.8, 26.1, 28.0,

C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
4007
132.0, 133.8, 151.9, 153.3, 155.0, 156.2, 163.5, 165.8, 168.7; MS(ESI)
5.13. 3,5-Dimethoxy-4-(3-(2-chlorocarbonyl-ethyldisulfanyl)-
propionyl)-benzoic acid 4-(N,N0-Boc-guanidino)-butyl ester (13)
(m/z): 755.3 [Mþ]; HRMS calculated mass for C35H55N4O12
S
755.34590, found 755.35377.
3-(2-carboxy-ethyldisulfanyl)-propionic acid (420 mg, 0.2 mmol)
and oxalyl chloride (1 mL) were dissolved and stirred in dry DCM
(10 mL) at room temperature for 12 h, the solvent removed by rotary
evaporation, to give 12 489 mg (99.0%). 3,5-dimethoxy-4-hydroxy-
5.10. 3,5-Dimethoxy-4-(2-amino-3-prop-2-ynylsulfanyl-propionyl)
-benzoic acid 4-guanidino-butyl ester (1a)
benzoic acid 4-(N,N0-Boc-guanidino)-butyl ester
5
(511 mg,
0.1 mmol), 3-(2-Chlorocarbonyl-ethyldisulfanyl)-propionyl chloride
12 (123.5 mg, 50 mol) and triethylamine (252 mg, 250 mmol) were
3,5-dimethoxy-4-(2-(N-Boc-amino)-3-prop-2-ynylsulfanyl-
m
propionyl)-benzoic acid 4-(N,N0-Boc-guanidino)-butyl ester (11a)
(75.2 mg, 10 mmol) and trifluoroacetic acid (TFA, 1 mL) were
dissolved and stirred in DCM (2 mL) at room temperature for 2 h,
the solvent removed by rotary evaporation, and then transferred
the remaining solvent to 50 mL EP tube, 40 mL of ether at 4 ꢀC
was added to the solvent and centrifugated at 4000 rpm for
10 min, supernatant was removed to give 25 mg of flaxen solid
1a (65.4%). The HPLC purified condition of 1a as follows: Mobile
phase: Methanol þ 0.05% TFA: H2O þ 0.05% TFA 45:55; RF: 4 mL/
dissolvedandstirredindryDCM (50 mL) atroomtemperature for 5h,
the solvent was removed by rotary evaporation. The crude product
was dissolved in DCM and was chromatographed on silica gel, eluted
with 25% EtOAc/PE, to give 598 mg of white solid 13 (50%). 1H
NMR(400 MHz, CDCl3)d 1.48(s, 18H, Boc), 1.49(s, 18H, Boc),
1.68e1.76(m, 4H, eCH2CH2e), 1.80e1.86(m, 4H, eCH2CH2e), 3.09(s,
8H, OOCCH2CH2S), 3.48(dd, 4H, J ¼ 7.04, 12.52 Hz, CH2NH), 3.87(s,
12H, eOCH3), 4.33e4.36(m, 4H, eCOOCH2), 7.30(s, 4H, AreH), 8.38(s,
2H, eCH2NH), 11.50(s, 2H, eCNH); 13C NMR(100 MHz CDCl3)
d 25.8,
min; Chromatographic column: SunfireÔ Prep C18 5.0
mm,
26.1, 28.0, 28.2, 29.7, 33.1, 33.8, 40.4, 56.3, 64.8, 79.4, 83.2,106.2,128.3,
132.3, 151.9, 153.3, 156.2, 165.8, 169.0; MS(ESI) (m/z): 1197.6 (Mþ),
559.1 (M/2eþ); HRMS calculated mass forC54H81N6O20S2 1197.48688,
found 1197.49246.
10 mm ꢄ 150 mm; Detect wavelength: 254 nm. The purity of 1a
was 99%. 1H NMR (400 MHz, DMSO þ D2O)
d 1.64e1.71 (m, 2H,
eCH2CH2e), 1.79e1.86(m, 2H, eCH2CH2e), 3.19e3.30(m, 4H,
CH2Se, CH2NHe), 3.47 (dd, 1H, J ¼ 4.69, 14.87 Hz, CHCe),
3.53e3.70(m, 2H, SCH2e), 3.92(s, 6H, eOCH3), 4.26e4.40(m, 2H,
eCOOCH2), 4.83(dd, 1H, J ¼ 4.70, 9.12 Hz, CHNH2), 7.39(s, 2H,
5.14. 3,5-Dimethoxy-4-(3-(2-chlorocarbonyl-ethyldisulfanyl)-
propionyl)-benzoic acid 4-guanidino-butyl ester (2)
AreH),. 13C NMR(100 MHz, DMSO þ D2O)
d 19.0, 25.1, 25.4, 31.4,
40.2, 51.3, 56.3, 64.7, 74.4, 79.8, 105.9, 128.8, 130.8, 151.5, 157.0,
165.0, 166.3; MS(ESI) (m/z): 453.3 [Mþ], 227(M/2eþ); HRMS
calculated mass for C20H29N4O6S 453.17296, found 453.18015.
3,5-dimethoxy-4-(3-(2-chlorocarbonyl-ethyldisulfanyl)-pro-
pionyl)-benzoic acid 4-(N,N0-Boc-guanidino)-butyl ester (119.6 mg,
10 mmol) and trifluoroacetic acid (TFA, 1 mL) were dissolved and
stirred in DCM (2 mL) at room temperature for 2 h, the solvent
removed by rotary evaporation, and the remaining solvent was
transferred to 50 mL EP tube, 40 mL of ether at 4 ꢀC was added to
the solvent and centrifugated at 4000 rpm for 10 min, supernatant
was removed to give 51.8 mg of white solid 2 (65.1%). The HPLC
purified condition of 2 as follows: Mobile phase: Methanol þ 0.05%
TFA: H2O þ 0.05% TFA 55:45; RF: 4 mL/min; Chromatographic
5.11. 3,5-Dimethoxy-4-(2-animo-3-allysulfanyl-propionyl)-benzoic
acid 4-guanidinobutyl ester (1b)
3,5-dimethoxy-4-(2-(N-Boc-amino)-3-prop-2-ynylsulfanyl-
propionyl)-benzoic acid 4-(N,N0-Boc- guanidino)-butyl ester (11b)
(75.4 mg, 10 mmol) and trifluoroacetic acid (TFA, 1 mL) were dis-
solved and stirred in DCM (2 mL) at room temperature for 2 h, the
solvent removed by rotary evaporation, and then the remaining
solvent was transferred to 50 mL EP tube, 40 mL of ether at 4 ꢀC was
added to the solvent and centrifugated 4000 rpm for 10 min,
supernatant was removed to give 27.2 mg of flaxen solid 1b (60.2%).
The HPLC purified condition of 1b as follows: Mobile phase:
Methanol þ 0.05% TFA: H2O þ 0.05% TFA 45:55; RF: 4 mL/min;
column: SunfireÔ Prep C18 5.0
wavelength: 254 nm. The purity of 2 was 98%. 1H NMR (400 MHz,
1.55e1.62(m, 4H, eCH2CH2e), 1.70e1.77 (m, 4H,
mm, 10 mm ꢄ 150 mm; Detect
DMSO þ D2O)
d
eCH2CH2e), 3.03(s, 8H, CH2CH2Se), 3.10e3.16(m, 4H, eCH2NH),
3.80(s, 12H, eOCH3), 4.28(t, 4H, J ¼ 6.42 Hz, eCOOCH2), 7.27(s, 4H,
AreH); 13C NMR(100 MHz, DMSO þ D2O)
d 25.3, 25.6, 32.9, 33.1,
40.3, 56.3, 64.7, 106.0, 128.2, 132.1, 152.0, 156.7, 165.2, 168.9. MS(ESI)
(m/z): 797.3 (Mþ), 399.2 (M/2eþ); HRMS calculated mass for
C34H49N6O12S2 797.27716, found 797.28400.
Chromatographic column: SunfireÔ Prep C18 5.0
10 mm ꢄ 150 mm; Detect wavelength: 254 nm. The purity of 1b
was 99%. 1H NMR(400 MHz, DMSO-d6)
1.55e1.62(m, 2H,
mm,
d
5.15. Primary cardiomyocyte culture
eCH2CH2e), 1.68e1.77(m, 2H, eCH2CH2e), 2.99e3.31(m, 6H,
CH2SCH2e, CH2NH), 3.82(s, 6H, eOCH3), 4.20e4.31(m, 2H,
eCOOCH2), 4.69(dd, 1H, J ¼ 4.70, 7.04 Hz, CHNH2), 5.11e5.23(m, 2H,
eCHCH2), 5.73e5.83(m, 1H, eCHCH2), 7.30(s, 2H, AreH),
7.92e7.94(m, 1H, eNH), 8.83(s, 2H, eNH2); 13C NMR(100 MHz
DMSO-d6) dC 25.1, 25.4, 30.5, 34.0, 40.4, 51.5, 56.3, 64.7, 105.9, 118.2,
128.8, 130.8, 133.8, 147.5, 156.9, 158.6, 165.0, 166.4. MS(ESI) (m/z)
455.4 [Mþ], 228(M/2eþ); HRMS calculated mass for C20H31N4O6S
455.18861, found 455.19791.
Primary cultures of cardiomyocytes were obtained from the
ventricles of newborn SpragueeDawley rats (1e3 days) according
to the method described by Wang et al. [17]. Isolated car-
diomyocytes were seeded at a density of 1 ꢄ105 cells/mL in DMEM
supplemented with 10% FBS, 100 U/mL penicillin, 100 mg/mL
streptomycin, and 100 mM 5-bromodeoxyuridine (Sigma). Cultures
were maintained at 37 ꢀC in a humidified incubator with 95% air
and 5% CO2 for 72 h before being used. Cardiomyocytes were then
divided into several groups: normoxia (control group), hypoxia
(model group), hypoxia þ different compounds pretreatment for
12 h before hypoxia. Then hypoxia was induced based on the
technique described by Rakhit et al. [32]. All of groups, excluding
the control group, were placed in an hypoxic solution (composition
(in mM): NaCl 116; KCl 50; CaCl2 1.8; MgCl2$6H2O 2; NaHCO3 26;
NaH2PO4$2H2O 1) in an anaerobic chamber (BD Diagnostics
System, 261215, USA) and maintained at 37 ꢀC with a humidified
5.12. 3-(2-Chlorocarbonyl-ethyldisulfanyl)-propionyl chloride (12)
3-(2-carboxy-ethyldisulfanyl)-propionic
acid
(420
mg,
0.2 mmol) and oxalyl chloride (1 mL) were dissolved and stirred in
dry DCM (10 mL) at room temperature for 12 h, the solvent
removed by rotary evaporation, to give 489 mg (99.0%) of solid

4008
C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
atmosphere of 5% CO2, 10% H2 and 85% N2. Chambers were sealed
before incubation at 37 ꢀC for 5 h.
CGGGAGAACAGGGTATGA-30, the reverse primer sequence used
was 50-CAGGCTGGAAGGAGAAGAT-30; the forward bax primer
sequence used was 50-GCAGGGAGGATGGCTGGGGAGA-30, the
reverse primer sequence used was 50-T CCAGACAAGCAGCCG
CTCACG-30; the forward caspase-3 primer sequence used was 50-CT
GGACTGCGGTATTGAG-30; the reverse primer sequence used was
5.16. Analysis of cell viability
Cell viability was measured using the 3-(4,5-dimethylthiazol-2-
yl)-2,5-diphenyl tetrazolium (MTT) assay [33]. Briefly, cells were
lysed in dimethyl sulfoxide (DMSO) after incubation with MTT for
4 h and the amount of MTT formazan was quantified by deter-
mining the absorbance at 570 nm, using a microplate reader. Cell
viability is expressed as a percentage of the control. The viability of
control cell is presumed as 100%.
50- GGGTGCGGTAGAGTAAGC-30; the forward
b-actin primer
sequence used was 50-ATCCGTAAAGACCTCTATGC-30, the reverse
primer sequence used was 50-AACGCAGCTCAGTAACAGTC-30.
actin was used as the internal control.
b-
5.21. Western blot
5.17. Measurement of H2S Concentration
Western blotting was performed as previously described in
Ref. [35]. Briefly, at the end of hypoxia, cardiac myocytes were
scraped off in cell lysis buffer (Beyotime Biotechnology, Haimen,
China) and centrifugated at 12,000 rpm for 10 min. The supernatant
was collected, and protein concentrations were quantified using
the enhanced BCA Protein Assay Kit (Beyotime Biotechnology,
Haimen, China). Fifty micrograms of total proteins were separated
on SDS-PAGE gels as previously described. Separated proteins were
The concentration of H2S in medium was determined as
described previously in Ref. [34]. Briefly, 0.5 mL of supernatant
were added to a 1.5 mL microtube containing 0.25 mL of 1% zinc
acetate. H2S is chemiabsorbed by zinc acetate and transformed into
stable zinc sulfide. The sulfide is then recovered by extraction with
water. Next, 20 mM of NNDPD in 7.2 M HCl and 30 mM FeCl3 in
1.2 M HCl were added to the microtube. The solution was then
incubated at room temperature for 10 min. In contact with FeCl3 in
a strongly acid solution, the extracted sulfide reacts with N,N-
dimethyl-p- phenylendiammonium ion to yield methylene blue.
Next, 0.25 mL of 10% TCA were added to remove the proteins
present. The microtubes were briefly shaken after each addition.
This final solution was then centrifuged at 12,000 g for 10 min at
4 ꢀC. The optical absorbance of the resulting solution was measured
at 670 nm using a 96-well microplate reader (Tecan Systems). Each
sample was assayed in duplicate, and H2S concentration was
transferred onto 0.45-mm polyvinylidene difluoride (PVDF)
membrane (Milli pore Corporation). PVDF membrane was blocked
in Tris-buffered saline (TBS) containing 5% skim milk for 2 h at room
temperature and probed with rabbit anti-rat bcl-2, bax and
caspase-3 polyclonal antibody (Cell Signaling Technology, Beverly,
MA, USA) diluted in TBST for 1.5 h at room temperature, Anti-rabbit
IgG diluted in TBST was used as secondary antibody. Protein band
intensities were quantified by using a Western blotting detection
system (Alpha Innotech, USA).
calculated against
a
standard calibration curve of NaHS
5.22. Statistic analysis
(3.125e250 M) that was performed on the same day as the sample
m
measurement.
Data were presented as means ꢁ S.E.M and analyzed by SPSS
software. Student’s test (two-tailed) was used to test the signifi-
cance of differences between mean values. Mean values are given
with standard errors of the mean. Pictures were processed with
Photoshop software. Differences at P < 0.05 were considered
statistically significant.
5.18. Measurements of catalase (CAT) activity, ROS level and
apoptosis
The levels of CAT, ROS and the apoptosis of cell with AnnexinV-
FITC/PI were measured using their kits according to the manufac-
ture’s protocol.
Contribution of the authors
5.19. Morphological assessment and quantification of apoptotic
myocytes
Chunhua Liu is the main person conducted the study, Wei Guo,
Xueru Shi, Kaium M. A. and Xianfeng Gu assisted the experiments,
both of Yizhun Zhu and Xianfeng Gu are corresponding author for
this paper, Yizhun Zhu is the PI for this project.
Neonatal rat ventricular myocytes were collected and then
routinely processed for transmission electron microscopy. Briefly
the cells were fixed for 2 h in PBS containing 2.5% glutaraldehyde,
post-fixed for 1 h in PBS containing 1% osmium tetroxide, and then
cells were subsequently dehydrated in a graded series of ethanol
and embedded in Epon 812. After polymerization, ultrathin (60 nm)
sections were cut on a LKB-V ultramicrotome (Nova, Sweden)
equipped with a diamond knife. All sections were stained with
uranyl acetate and lead citrate before viewing under a Philips, CM
120 electron microscope.
Acknowledgments
This study was supported by National Basic Research Program of
China (973 Program) (No. 2010CB912600), National Natural Science
Foundation of China (No. 30772565, No. 30888002) and New
Teacher Foundation from Ministry of Education in China (No.
20090071120).
5.20. RT-PCR
References
Total RNA from cultured cardiomyocytes was isolated using
Trizol reagent (Invitrogen, Carlsbad, CA). RNA was quantified
spectrophotometrically by measuring the optical density of
samples at 260/280 nm. Two micrograms of RNA were reverse-
transcribed to cDNA with the PrimeScriptTM 1st Strand cDNA
Synthesis Kit. The primer sequences of genes of interest were as
follows: the forward bcl-2 primer sequence used was 50-
[1] Y.Z. Zhu, C.L. Chong, S.C. Chuah, S.H. Huang, H.S. Nai, H.T. Tong, M. Whiteman,
P.K. Moore, Cardioprotective effects of nitroparacetamol and paracetamol in
acute phase of myocardial infarction in experimental rats, Am. J. Physiol-Heart
C 290 (2006) H517eH524.
[2] A. Haunstetter, S. Izumo, Apoptosis e basic mechanisms and implications for
cardiovascular disease, Circ. Res. 82 (1998) 1111e1129.
[3] A. Haunstetter, S. Izumo, Toward antiapoptosis as a new treatment modality,
Circ. Res. 86 (2000) 371e376.

C. Liu et al. / European Journal of Medicinal Chemistry 46 (2011) 3996e4009
4009
[4] S.K. Wattanapitayakul, J.A. Bauer, Oxidative pathways in cardiovascular
disease e roles, mechanisms, and therapeutic implications, Pharmacol. Ther-
apeut 89 (2001) 187e206.
[20] J. Sun, B.K.H. Tan, S.H. Huang, M. Whiteman, Y.Z. Zhu, Effects of natural
products on ischemic heart diseases and cardiovascular system, Acta Phar-
macologica Sinica 23 (2002) 1142e1151.
[5] L.J.S. Greenlund, T.L. Deckwerth, E.M. Johnson, Superoxide-dismutase delays
neuronal apoptosis e a role for reactive oxygen species in Programmed
Neuronal Death, Neuron 14 (1995) 303e315.
[21] Y.Z. Zhu, S.H. Huang, B.K.H. Tan, J. Sun, M. Whiteman, Y.C. Zhu, Antioxidants in
Chinese herbal medicines: a biochemical perspective, Nat. Prod. Rep. 21
(2004) 478e489.
[6] Q. Wang, X.L. Wang, H.R. Liu, P. Rose, Y.Z. Zhu, Protective effects of cysteine
analogues on acute myocardial ischemia: novel modulators of endogenous
H2S production, Antioxid. Redox Signaling 12 (2010) 1155e1165.
[7] L. Li, A. Hsu, P.K. Moore, Actions and interactions of nitric oxide, carbon
monoxide and hydrogen sulphide in the cardiovascular system and in
inflammation e a tale of three gases!, Pharmacol. Therapeut 123 (2009)
386e400.
[8] B. O’Rourke, Pathophysiological and protective roles of mitochondrial ion
channels, J. Physiology-London 529 (2000) 23e36.
[9] T.T. Pan, Z.N. Feng, S.W. Lee, P.K. Moore, J.S. Bian, Endogenous hydrogen
sulfide contributes to the cardioprotection by metabolic inhibition pre-
conditioning in the rat ventricular myocytes, J. Mol. Cell. Cardiol. 40 (2006)
119e130.
[22] X.H. Liu, L.L. Pan, P.F. Chen, Y.Z. Zhu, Leonurine improves ischemia-induced
myocardial injury through antioxidative activity, Phytomedicine 17 (2010)
753e759.
[23] B. Freude, T.N. Masters, S. Kostin, F. Robicsek, J. Schaper, Cardiomyocyte
apoptosis in acute and chronic conditions, Basic Res. Cardiol. 93 (1998) 85e89.
[24] X. Liu, L. Pan, Q. Gong, Y. Zhu, Leonurine (SCM-198) improves cardiac recovery
in rat during chronic infarction, Eur. J. Pharmacol. 649 (2010) 236e241.
[25] J. Kajstura, W. Cheng, K. Reiss, W.A. Clark, E.H. Sonnenblick, S. Krajewski,
J.C. Reed, G. Olivetti, P. Anversa, Apoptotic and necrotic myocyte cell deaths
are independent contributing variables of infarct size in rats, Lab. Invest. 74
(1996) 86e107.
[26] Z.Q. Zhao, M. Nakamura, N.P. Wang, J.N. Wilcox, S. Shearer, R.S. Ronson,
R.A. Guyton, J. Vinten-Johansen, Reperfusion induces myocardial apoptotic
cell death, Cardiovasc. Res. 45 (2000) 651e660.
[27] E.S. Alnemri, D.J. Livingston, D.W. Nicholson, G. Salvesen, N.A. Thornberry,
W.W. Wong,J. Yuan, HumanICE/CED-3proteasenomenclature,Cell87(1996)171.
[28] N.A. Thornberry, Y. Lazebnik, Caspases: enemies within, Science 281 (1998)
1312e1316.
[10] K. Hamano, Y. Ikeda, T. Kobayashi, T.S. Li, H. Ito, A. Mikamo, CABG in combi-
nation with neovascularization or surgical ventricular restoration for the
treatment of severe ischemic heart disease, J. Card. Fail. 11 (2005) S255.
[11] R. Morphy, C. Kay, Z. Rankovic, From magic bullets to designed multiple
ligands, Drug Discov. Today 9 (2004) 641e651.
[12] R. Morphy, Z. Rankovic, Designed multiple ligands. An emerging drug
discovery paradigm, J. Med. Chem. 48 (2005) 6523e6543.
[13] A. Gasco, D. Boschi, K. Chegaev, C. Cena, A. Di Stilo, R. Fruttero, L. Lazzarato,
B. Rolando, P. Tosco, Multitarget drugs: focus on the NO-donor hybrid drugs,
Pure Appl. Chem. 80 (2008) 1693e1701.
[29] S.H. Kaufmann, S. Desnoyers, Y. Ottaviano, N.E. Davidson, G.G. Poirier, Specific
proteolytic cleavage of poly(ADP-ribose) polymerase: an early marker of
chemotherapy-induced apoptosis, Cancer Res. 53 (1993) 3976e3985.
[30] S. Han, R.A. Moore, R.E. Viola, Synthesis and evaluation of alternative
substrates for arginase, Bioorg. Chem. 30 (2002) 81e94.
[14] C. Liu, X. Gu, Y.Z. Zhu, Synthesis and biological evaluation of novel leonurine-
SPRC conjugate as cardioprotective agents, Bioorg. Med. Chem. Lett. (2010).
[15] X.H. Liu, H. Xin, A.J. Hou, Y.Z. Zhu, Protective effects of leonurine in neonatal
rat hypoxic cardiomyocytes and rat infarcted heart, Clin. Exp. Pharmacol. P 36
(2009) 696e703.
[16] K.P. Loh, J. Qi, B.K. Tan, X.H. Liu, B.G. Wei, Y.Z. Zhu, Leonurine protects middle
cerebral artery occluded rats through antioxidant effect and regulation of
mitochondrial function, Stroke 41 (2010) 2661e2668.
[31] H. Yanagisawa, S. Ishihara, A. Ando, T. Kanazaki, S. Miyamoto, H. Koike,
Y. Iijima, K. Oizumi, Y. Matsushita, T. Hata, Angiotensin-Converting enzyme-
Inhibitors e Perhydro-1,4-Thiazepin-5-One Derivatives, J. Med. Chem. 30
(1987) 1984e1991.
[32] R.D. Rakhit, R.J. Edwards, J.W. Mockridge, A.R. Baydoun, A.W. Wyatt,
G.E. Mann, M.S. Marber, Nitric oxide-induced cardioprotection in cultured rat
ventricular myocytes, Am. J. Physiol. Heart Circ. Physiol. 278 (2000)
H1211eH1217.
[17] Q. Wang, H.R. Liu, Q. Mu, P. Rose, Y.Z. Zhu, S-propargyl-cysteine protects both
adult rat hearts and neonatal cardiomyocytes from ischemia/hypoxia injury:
the contribution of the hydrogen sulfide-mediated pathway, J. Cardiovasc.
Pharm. 54 (2009) 139e146.
[18] K.F. Cheng, C.S. Yip, H.W. Yeung, Y.C. Kong, Leonurine, an improved synthesis,
Experientia 35 (1979) 571e572.
[33] P.F. Li, R. Dietz, R. von Harsdorf, Differential effect of hydrogen peroxide and
superoxide anion on apoptosis and proliferation of vascular smooth muscle
cells, Circulation 96 (1997) 3602e3609.
[34] L. Li, G. Rossoni, A. Sparatore, L.C. Lee, P. Del Soldato, P.K. Moore, Anti-
inflammatory and gastrointestinal effects of a novel diclofenac derivative, Free
Radic. Biol. Med. 42 (2007) 706e719.
[19] S.C. Chuah, P.K. Moore, Y.Z. Zhu, S-allylcysteine mediates cardioprotection in
an acute myocardial infarction rat model via a hydrogen sulfide-mediated
pathway, Am. J. Physiol-Heart C 293 (2007) H2693eH2701.
[35] T. Kakita, K. Hasegawa, T. Morimoto, S. Kaburagi, H. Wada, S. Sasayama, p300
protein as a coactivator of GATA-5 in the transcription of cardiac-restricted
atrial natriuretic factor gene, J. Biol. Chem. 274 (1999) 34096e34102.