Discovery of novel NO-releasing celastrol derivatives with Hsp90 inhibition and cytotoxic activities

Article abstract of DOI:10.1016/j.ejmech.2018.10.013

To develop multifunctional drugs, a series of celastrol/NO donor hybrids were designed, synthesized and evaluated. The detection of NO release amounts showed that the more NO of these hybrids released, the more tumor cells were inhibited. 11b, which released the highest level of NO in vitro, exhibited superior potency (IC₅₀ = 0.48 ± 0.06 μM) compared to the other compounds. Further pharmacological studies showed that 11b induced dysregulations of the Hsp90 clients (Akt and Cdk4), apoptosis, and cell cycle arrested at G₀/G₁ phase against A549 cells. These results suggested that inhibition of Hsp90 and release of NO was synergistic in cancer cells. Overall, the NO-releasing capacity and the inhibition of Hsp90 pathway signaling might explain the potent anti-proliferative activities of these compounds.

Full text of DOI:10.1016/j.ejmech.2018.10.013

Accepted Manuscript
Discovery of novel NO-releasing celastrol derivatives with Hsp90 inhibition and
cytotoxic activities
Na Li, Manyi Xu, Na Bao, Wei Shi, Qixing Li, Xiaowei Zhang, Jianbo Sun, Li Chen
PII:
S0223-5234(18)30875-4
10.1016/j.ejmech.2018.10.013
EJMECH 10799
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 14 August 2018
Revised Date: 1 October 2018
Accepted Date: 5 October 2018
Please cite this article as: N. Li, M. Xu, N. Bao, W. Shi, Q. Li, X. Zhang, J. Sun, L. Chen, Discovery of
novel NO-releasing celastrol derivatives with Hsp90 inhibition and cytotoxic activities, European Journal
of Medicinal Chemistry (2018), doi: https://doi.org/10.1016/j.ejmech.2018.10.013.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
A series of celastrol/NO donor hybrids were designed, synthesized and evaluated. 11b,
which released the highest level of NO in vitro, exhibited the best antiproliferative
effects. The results suggested that inhibition of Hsp90 and release of NO was
synergistic in cancer cells.

ACCEPTED MANUSCRIPT
Discovery of Novel NO-releasing Celastrol Derivatives with
Hsp90 Inhibition and Cytotoxic Activities
Na Li, Manyi Xu, Na Bao, Wei Shi, Qixing Li, Xiaowei Zhang, Jianbo Sun*, and Li
Chen*
State Key Laboratory of Natural Medicines, Department of Natural Medicinal Chemistry, China
Pharmaceutical University, 24 Tong Jia Xiang, Nanjing 210009, People's Republic of China
* Corresponding author: Tel: +86 83271447
E-mail address: sjbcpu@gmail.com (Jianbo Sun); chenli627@cpu.edu.cn (Li Chen)
Keywords: Celastrol, Nitric oxide, Hsp 90, Apoptosis, Cytotoxicity
ABSTRACT
To develop multifunctional drugs, a series of celastrol/NO donor hybrids were
designed, synthesized and evaluated. The detection of NO release amounts showed
that more NO of these hybrids released, the more tumor cells were inhibited. 11b,
which released the highest level of NO in vitro, exhibited superior potency (IC50 =
0.48 ± 0.06 µM) compared to the other compounds. Further pharmacological studies
showed that 11b induced dysregulations of the Hsp90 clients (Akt and Cdk4),
apoptosis, and cell cycle arrested at G0/G1 phase against A549 cells. These results
suggested that inhibition of Hsp90 and release of NO was synergistic in cancer cells.
Overall, the NO-releasing capacity and the inhibition of Hsp90 pathway signaling
might explain the potent anti-proliferative activities of these compounds.
1. Introduction
Celastrol (CEL) is a triterpenoid isolated from the root extracts of Thunder of God
Vine (Tripterygium wilfordii Hook F.). It has been widely investigated about its
anti-proliferative activity [1-3].
A
recent report demonstrated that the

ACCEPTED MANUSCRIPT
anti-proliferative activity of CEL results from targeting the Hsp90 (Heat shock protein
90) signaling pathways [4-5]. Hsp90 is a protein chaperone that helps its clients to
maintain their correct conformation [6-7]. Many of Hsp90 clients are demonstrated as
oncogenic proteins, such as protein kinase B (Akt), cyclin-dependent kinases 4 (Cdk4),
epidermal growth factor receptor (EGFR), Her-2, c-Met [8]. Thus, CEL can regulate
multiple signaling pathways simultaneously to inhibit the proliferation of cancer cells.
Nitric oxide (NO) is a key signaling and effector molecule in the development of
tumors [9]. Generally, high levels of NO can induce apoptosis in tumor cells while
lower concentrations of NO usually protect cells [10, 11]. It has been reported that
NO regulates several cancer-related signaling pathways, such as, Akt, extracellular
signal-regulated kinases (ERKs), mammalian target of rapamycin (mTOR), cyclin
D1/retinoblastoma (Rb) [12-13]. In addition, NO-donating hybrids can significantly
decrease the resistant rates of their parent drugs [14]. Thus, many chemical NO
donors have been designed to generate high levels of NO intracellular [15]. Among
them, furoxan, which is wildly used in the creation of drugs, can inhibit the
proliferation of cancer cells though generation of large amounts of NO [15]. In recent
decades, many hybrids from furoxan based NO donor displayed potent antitumor
activity [16-18]. For example, the IC50 value of β-elemene/furoxan hybrid (0.002 µM)
was much superior to that of β-elemene (197.65 µM). Moreover, this hybrid
significantly suppressed the tumor growth in vivo [19]. Hence, generation of furoxan
based hybrids is an effective strategy for the treatment of cancer.
Hybridization strategy, a well-known approach in drug design, allows a
combination of two or more pharmacophores with bioactive scaffolds to generate a
single molecular architecture. Generally, these hybrids possess improved affinity and
activity in comparison to their parent compounds. This strategy has been widely used
in the development of antibacterial, anticancer and antimalarial drugs [20-21]. In
order to develop novel anti-tumor drugs, hybrids based on natural products have been
extensively studied [22-24]. Recently, many studies reported that the hybrids of
triterpenoid and a NO donor showed enhanced anti-proliferative activities compared
to the parent compound [25-28]. However, the combination of CEL and NO donor

ACCEPTED MANUSCRIPT
have not been reported. Thus, on the basis of the analysis above and the hybridization
strategy, the furoxan based NO donor was introduced at 29-position of CEL by
esterification and amidification of 20-carboxylic acid (9a-11b) to synthesize
celastrol/furoxan hybrids (Scheme 1). Furthermore, nitrate, another type of NO donor,
was also linked to CEL (13a-13d, Scheme 2). Their anti-proliferative activities were
assayed in vitro. Moreover, cell cycle arrest, apoptosis, NO-releasing ability and the
inhibition of Hsp90 clients of 11b were biologically evaluated to reveal the
mechanisms of action of these hybrids.
2. Results and discussion
2.1. Chemistry
The synthesis of hybrids 9a-e and 11a-11b was provided in Scheme 1. The
condensation of benzenethiol (2) and chloroacetic acid (3) under basic conditions
generated 2-(phenylthio) acetic acid (4) which was subjected to further oxidation in
30% hydrogen peroxide (H2O2) aqueous solution to provide 2-(phenylsulfonyl) acetic
acid (5) [29]. 5 was converted to diphenylsulfonyl-furoxan (6) by treatment with
fuming nitric acid (HNO3) [29]. 6 then reacted with corresponding diols or amines to
produce the indicated compounds (7a-7e, 10a-10b) [29]. 7a-7e was treated with
triphenylphosphine (PPh3) and carbon tetrabromide (CBr4) to give 8a-8e [30]. CEL (1)
further reacted with 8a-8e under basic conditions to provide 9a-9e [31]. Compounds
11a and 11b were obtained by coupling 1 with 10a-10b catalyzed by
benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate (PyBop) with
the presence of hunig’s base [31]. Then, piperazine and piperidine were used as the
linkers. However, the intermediates couldn’t react with CEL with the presence of
PyBop and hunig’s base or 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
hydrochloride (EDCI) and 1-Hydroxybenzotriazole (HOBT) (S-scheme 1). In order to
increase the reactivity of CEL, we try to convert the carboxylic acid of CEL to acyl
chloride with thionyl chloride (SOCl2) or oxalyl chloride [(COCl)2] (S-scheme 1). The
results showed that the structure of CEL was broken by the above reagents. Next,
2-methylaminoethanol was used to introduce furoxan (S-scheme 1). Unfortunately,
this reaction was failed again. These results indicated that the increasing steric

ACCEPTED MANUSCRIPT
hindrance of the intermediates might prevent the processes of the reactions. Thus, we
gave up introducing the other amines to CEL.
Scheme 1. Reagents and conditions (a) NaOH, reflux, 1 h; (b) H2O2, rt, 3 h; (c)
o
o
fuming HNO3, 100 C, 4 h; (d) corresponding diols, 25% NaOH, -15 C, 3 h; (e)
o
corresponding amines, DBU, DCM, -15 C, 3 h; (f) PPh3, CBr4, DCM, rt, 1 h; (g) 1,
NaHCO3, DMF, rt, 12 h; (h) 1, PyBop, Hunig’s base, DCM, rt, 10 h.
1 was conducted esterification with the indicated dibromides on the carboxyl to
provide 12a-12d [32]. Compounds (12a-12d) were further transformed to the
corresponding hybrids 13a-13d [33], respectively, with silver nitrate (AgNO3) in
acetonitrile (CH3CN) (Scheme 2).
Scheme 2. Reagents and conditions: (a) Br(CH2)nBr, NaHCO3, rt, 12 h; (b) AgNO3,
80 oC, CH3CN, 10 h.
2.2. In vitro anti-proliferative activity
We first detected the anti-proliferative effects of 1 and the derivatives against four
tumor cell lines (A549, HOS, MCF-7 and HepG2) by MTT method. CDDO-Me,
which was studied in phase III clinical trial, was used as positive control. As shown in
Table 1, generally celastrol/nitrate hybrids (13a-13d) displayed moderate
anti-proliferative activities compared to celastrol/furoxan hybrids (9a-9e and

ACCEPTED MANUSCRIPT
11a-11b). Furthermore, two celastrol/furoxan hybrids (11a and 11b), whose
carboxylic acid (C-20) was connected with amino, displayed more potent
anti-proliferative activities than the other hybrids. These results suggested that the
amide linkage might be more appropriated for the modification of CEL in this study.
Furthermore, introductions of different types of NO donors to the structure of CEL
brought different influences to the activity. Among all target compounds, 11b
displayed more potent activity (IC50 = 0.48 ± 0.06, 0.78 ± 0.08, 0.51 ± 0.09 and 1.41 ±
0.18 µM) in A549, MCF-7, HOS, and HepG2 cell lines. Thus, 11b was selected for
further investigation.
Table 1
Anti-proliferative activity of all the compounds against four cancer cell lines.
Comp.
Cytotoxicity IC50 (µM)a
A549
MCF-7
HOS
HepG2
2.28 ± 0.21
2.89 ± 0.33
2.46 ± 0.37
2.07 ± 0.23
1.11 ± 0.23
1.08 ± 0.15
0.48 ± 0.06
3.22 ± 0.23
4.21 ± 0.34
5.32 ± 0.11
3.67 ± 0.32
1.57 ± 0.15
0.45 ± 0.01
2.76 ± 0.32
2.12 ± 0.24
3.95 ± 0.31
2.02 ± 0.36
1.49 ± 0.15
1.52 ± 0.22
0.78 ± 0.08
2.54 ± 0.33
3.82 ± 0.37
3.01 ± 0.43
3.17 ± 0.32
1.56 ± 0.16
0.79 ± 0.21
1.69 ± 0.24
1.87 ± 0.21
2.47 ± 0 .11
3.03 ± 0.32
1.67 ± 0.13
1.47 ± 0.16
0.51 ± 0.09
1.37 ± 0.13
2.43 ± 0.22
2.90 ± 0.16
3.36 ± 0.21
0.93 ± 0.03
0.57 ± 0.12
4.75 ± 0.81
4.53 ± 0.22
3.53 ± 0.39
2.39 ± 0.31
2.39 ± 0.33
3.82 ± 0.42
1.41 ± 0.18
1.85 ± 0.17
6.08 ± 0.35
3.40 ± 0.56
4.71 ± 0.27
1.62 ± 0.19
0.33 ± 0.02
9a
9b
9c
9d
9e
11a
11b
13a
13b
13c
13d
1
CDDO-Meb
a
MTT methods, cells were incubated with corresponding compounds for 48 h. IC50
(µM) values (means ± SD, n = 3).
b Positive control.
2.3. NO generation measurement
To explore the association between NO release amounts of these hybrids and their
anti-proliferative activities, the NO release levels of these hybrids in vitro were
detected using a Nitrate/Nitrite assay according to the standard curve. As shown in
Fig. 1A, NO release amounts of the partial hybrids (4 µM of 9a, 9e, 11a-b and 13a-b)

ACCEPTED MANUSCRIPT
were increased continuously for 24 hours. However, the NO amounts declined slightly
with further incubation. Thus, we decided to evaluate the cellular levels of NO of all
celastrol/NO donor hybrids at 4 µM for 24 h (Fig. 1B). It was observed that the
hybrids designed in this study could produce different levels of NO intracellularly
which generally agreed with the anti-proliferative activities.
Fig. 1. Effects of NO produced by the target compounds on A549 cell proliferation.
(A) NO release amounts of partial hybrids producing vs. time. A549 cells were treated
with 4 µM of the selected compounds for different hours (mean ± SD, n = 3). (B) The
levels of NO produced by all of the compounds in A549 cells. A549 cells were treated
with 4 µM of all the hybrids for 24 hours (mean ± SD, n = 3).
In order to further demonstrate that the improved cytotoxicity of tumor cells was
due to the high NO release amounts, an NO scavenger (PTIO) was introduced in the
MTT assay [34]. Different concentrations of PTIO was added to the A549 cells for 2
h and then incubated with 11b (4 µM) for another 24 h. Next the anti-proliferative
activities were evaluated (Fig. 2). The results showed that the NO scavenger could
decrease the anti-proliferation of 11b dose-dependently in A549 cells. Thus, NO
release amounts were contributed to the enhanced activity of 11b in A549, which was
consistent with the design strategy in this study.

ACCEPTED MANUSCRIPT
Fig. 2. Effect of PTIO on the anti-proliferative activity of 11b. A549 cells were
pretreated with the indicated concentrations of PTIO for 2 h and then treated with 4
µM of 11b for another 24 h. Data were expressed as the mean ± SD (n=3). *P < 0.05,
**P < 0.01 vs. the group without PTIO.
2.4. Effect of 11b on Hsp90 client proteins
In the NO release assay, 11b still remained anti-proliferative against A549 cells in
the presence of high concentration of PTIO. We speculated that this activity might be
derived from the parent compound (CEL). CEL could inhibit the pathway of Hsp90 to
induce physiological apoptosis of the cells. We next evaluated whether the 11b also
inherited such ability. Cdk4 and p-Akt, as important clients of Hsp90, were measured
by western blotting. As shown in Fig. 3A & B, 11b significantly reduced the levels of
Cdk4 and p-Akt in A549 cells. However, 11b didn’t inhibit Hsp90 directly. These
results indicated that 11b might target the Hsp90 chaperoning machinery, which was
similar to CEL [35-37]. In this assay, we also observed that the Akt inhibitory activity
of 11b was stronger than that of CEL. The mechanism might be complex and was
worthy of further investigation.
Fig. 3. Effects of 11b on Hsp90 clients in A549 cells. A549 cells were treated with
varying concentrations of 11b for 24 h. (A) The expressions of Cdk-4 and Hsp90 were
determined by Western Blotting using specific antibodies. β-actin was used as internal
control; (B) The expressions of p-Akt and Akt were determined by Western Blotting

ACCEPTED MANUSCRIPT
using specific antibodies.
2.5. Synergism of inhibition of Hsp90 and release of NO in 11b
To evaluate whether inhibition of Hsp90 and release of NO of 11b was synergistic
in cancer cells, the two moieties of 11b (1 and 10b) and their mixture (1:1) were
examined for their anti-proliferative activities against A549 cells (MTT assay). As
shown in Fig. 4, the inhibitory activities of 1, 10b and their mixture were much less
potent compared to 11b, respectively. These results showed that inhibition of Hsp90
and release of NO of 11b was synergistic in this study.
Fig. 4. Anti-proliferative activity of 11b in comparison with that of 1, 10b and their
mixture in A549 cells.
2.6. The morphological apoptosis induced by 11b
As mentioned above, 11b not only inhibited Hsp90 clients (p-Akt and Cdk4) but
also released NO in vitro. Both of these two capabilities could induce the apoptosis of
cancer cells. Therefore, the morphological apoptosis induced by 11b was investigated
by Hoechst 3342 staining. As shown in Fig. 5, Hoechst 3342 stained the apoptotic
cells (bright blue) which appeared nuclear fragmentation and chromatin condensation
in A549 cells. All of the images were magnified for the observation of cell
morphology
S34-37.
and
the
original
photos
were
shown
in
Fig.

ACCEPTED MANUSCRIPT
Fig. 5. Fluorescence microscopy images of A549 cells stained by Hoechst 33342.
A549 cells were treated with the 11b at indicated concentrations for 48 h.
2.7. Effect of 11b on cell apoptosis
Then, the apoptosis induced by 11b was tested and compared with that of CEL.
Flow cytometry assay was performed to analyze the apoptotic cells mediated by 11b
and CEL in A549 cells. As shown in Fig. 6A, 11b induced apoptosis in a
dose-dependent manner in A549 cells. At the concentration of 1.2 µM (Fig. 6B), the
early and late apoptosis cells account for 12.9 % and 31.0 % respectively, which were
much higher than that of CEL (7.8 %; 11.2 %). Thus, introduction of NO donor to
CEL could promote the apoptosis of cancer cells.
Fig. 6. Effect of 11b on apoptosis of A549 cells. (A) A549 cells were treated with 11b
at the indicated concentrations for 48 h. Apoptotic effects were measured by flow
cytometry using Annexin V-FITC/PI staining protocol; (B) Representative histograms
for the numbers of cells (% of total) in the early and late stages of apoptosis for the
treatment groups.
2.8. Effect of 11b on cell cycle
From the results illustrated in Fig. 3, 11b could inhibit the expression of Cdk4
which is a regulator protein of cell cycle. In order to further investigate the

ACCEPTED MANUSCRIPT
mechanism of its anti-proliferative activity, the effect of 11b on the cell cycle
progression was evaluated. As shown in Fig. 7, 11b arrested cell cycle at G0/G1 phase
in A549 cells, which was another pathway to induce apoptosis.
Fig. 7. Cell cycle analysis of 11b in A549 cells. Cells were treated with 11b at 0.2, 0.4
and 0.8 µM for 24 h, harvested, stained with PI, and then analyzed by flow cytometry.
3. Conclusion
Based on the strategy of creating multifunctional drugs, 11 new NO-releasing
celastrol hybrids were designed and synthesized. Their anti-proliferative activities
were evaluated against A549, HOS, MCF-7 and HepG2 cells. The intracellular NO
productions of all the hybrids were tested in vitro. The NO release amounts detection
indicated that there was a positive correlation between their anti-proliferative
activities and the NO release. Moreover, introductions of different types of NO donors
to the structure of CEL brought different influences to the activity. In this study, 11b
showed the most potent activity in A549 cells. Further mechanism studies showed that
11b induced apoptosis through inhibiting the activity of Hsp90, releasing high level of
NO. These results showed that inhibition of Hsp90 and release of NO was synergistic
in cancer cells. In conclusion, the current finding may provide a new insight for the
design of celastrol derivatives to enhance the efficacy of candidates.
4. Experimental
4.1. Chemistry
4.1.1. General

ACCEPTED MANUSCRIPT
1
All reagents were purchased from chemical and biological company. H NMR and
13C NMR spectra were recorded on a Bruker AVANCE instrument at 25°C. The
molecular weights were detected on HP 1100LC/MSD spectrometer. Intermediates
7a-e, 8a-8e, 10a-b and 12a-12d were synthesized according to the previous methods.
4.1.2. General procedure for synthesis of 9a-9e
8a-e (0.11 mM, 1.1 eq) and 1 (0.10 mM, 1 eq) were dissolved in DMF (5.0 mL).
o
Then NaHCO3 (0.50 mM) was added. The solution was stirred for 12 h at 25 C. The
reaction mixture was then diluted with CH2Cl2, washed with water, and dried over
anhydrous Na2SO4, and concentrated. The residue was purified by flash column
chromatography using PE/EA (6:1, v/v) as an eluent to afford 9a-9e.
4.1.2.1. 9a
Red solid, yielded 49%. mp 155.5-157.2 °C. 1H NMR (300 MHz, CDCl3) δ: 8.04
(2H, d, J = 7.3 Hz), 7.75 (1H, t, J = 7.3 Hz), 7.59 (2H, t, J = 7.4 Hz), 6.89 (1H, d, J =
6.7 Hz), 6.53 (1H, s), 6.34 (1H, d, J = 7.0 Hz), 4.60 (2H, m), 3.95-4.14 (2H, m), 2.18
13
(3H, s), 1.45 (3H, s), 1.27 (3H, s), 1.23 (3H, s), 1.12 (3H, s), 0.60 (3H, s). C NMR
(75 MHz, CDCl3) δ: 178.3, 178.2, 170.0, 164.8, 158.7, 146.0, 138.0, 135.7 (C×2),
134.2, 129.7 (C×2), 128.4 (C×2), 127.3, 119.6, 118.2, 117.1, 68.7, 61.4, 45.0, 44.2,
42.9, 40.6, 39.4, 38.3, 36.3, 36.4, 33.6, 32.8, 31.6, 31.0, 30.6, 29.7 (C×2), 28.6, 21.6,
18.7, 10.3. HRMS (ESI) calculated for C39H47N2O9S [M + H]+ 719.2999, found
719.3002.
4.1.2.2. 9b
Red solid, yielded 56%. mp 186.0-187.4 °C. 1H NMR (300 MHz, CDCl3) δ: 7.98
(2H, d, J = 7.7 Hz), 7.70 (1H, d, J = 7.0 Hz), 7.60 (2H, t, J = 7.4 Hz), 7.03 (1H, d, J =
6.9 Hz), 6.51 (1H, s), 6.35 (1H, d, J = 6.9 Hz), 4.49 (2H, d, J = 7.2 Hz), 4.04-4.13 (2H,
m), 2.21 (3H, s), 1.45 (3H, s), 1.29 (3H, s), 1.19 (3H, s), 1.11 (3H, s), 0.54 (3H, s). 13C
NMR (75 MHz, CDCl3) δ: 178.3, 178.2, 169.7, 164.6, 158.8, 146.0, 137.9, 135.6,
134.2, 129.7 (C×2), 128.4 (C×2), 127.4 119.4, 118.2, 117.2, 67.9, 60.2, 45.0, 44.2,
42.9, 40.5, 39.5, 38.2, 36.3, 34.9, 33.6, 32.9, 31.6, 31.5, 30.9, 29.7 (C×2), 28.6, 27.7,
21.6, 18.5, 10.3. HRMS (ESI) calculated for C40H49N2O9S [M + H]+ 733.3159, found
733.3157.

ACCEPTED MANUSCRIPT
4.1.2.3. 9c
Red solid, yielded 58%. mp 156.1-157.8 °C. 1H NMR (300 MHz, CDCl3) δ: 8.05
(2H, d, J = 7.4 Hz), 7.76 (1H, d, J = 7.4 Hz), 7.62 (2H, t, J = 7.0 Hz), 7.02 (1H, d, J =
7.0 Hz), 6.54 (1H, s), 6.36 (1H, d, J = 7.1 Hz), 4.46 (2H, t, J = 6.2 Hz), 3.92-4.14 (2H,
m), 2.21 (3H, s), 1.46 (3H, s), 1.28 (3H, s), 1.21 (3H, s), 1.12 (3H, s), 0.59 (3H, s). 13C
NMR (75 MHz, CDCl3) δ: 178.3 (C×2), 169.97, 164.7, 158.9, 146.0, 135.6, 134.2,
129.7 (C×3), 127.4 (C×3), 119.5, 118.2, 117.1, 70.9, 63.6, 45.1, 44.2, 42.9, 40.5, 39.4,
38.2, 36.3, 34.8, 33.5, 32.9, 31.6, 30.9, 30.6, 29.8, 29.7, 25.4, 24.9, 21.6, 18.6, 14.1,
10.3. HRMS (ESI) calculated for C41H51N2O9S [M + H]+ 747.3315, found 747.3316.
4.1.2.4. 9d
Red solid, yielded 55%. mp 164.3-165.3 °C. 1H NMR (300 MHz, CDCl3) δ: 8.05
(2H, d, J = 7.8 Hz), 7.75 (1H, d, J = 7.4 Hz), 7.63 (2H, t, J = 7.9 Hz), 7.02 (1H, d, J =
7.1 Hz), 6.54 (1H, s), 6.35 (1H, d, J = 7.1 Hz), 4.75 (1H, t, J = 3.3 Hz), 3.36-4.48 (2H,
m), 2.22 (3H, s), 1.47 (3H, s), 1.46 (3H, d, J = 6.3 Hz), 1.29 (3H, s), 1.22 (3H, s), 1.13
(3H, s), 0.63 (3H, s). 13C NMR (75 MHz, CDCl3) δ: 179.6, 178.7, 169.7, 164.6, 158.8,
146.0, 137.6, 135.6, 134.2, 129.7 (C×2), 128.5 (C×2), 127.4, 124.4, 119.4, 118.2, 72.4,
67.6, 45.0, 42.9, 40.6, 38.2, 36.3, 34.9, 33.2, 31.5, 30.7, 30.6, 29.6 (C×2), 28.7, 21.7,
21.6, 19.0, 18.8, 16.0, 10.2. HRMS (ESI) calculated for C40H49N2O9S [M + H]+
733.3159, found 733.3166.
4.1.2.5. 9e
Red solid, yielded 50%. mp 173.8-175.0 °C. 1H NMR (300 MHz, CDCl3) δ: 8.00
(2H, d, J = 7.1 Hz), 7.74 (1H, d, J = 7.3 Hz), 7.59 (2H, t, J = 8.0 Hz), 7.03 (1H, d, J =
7.1 Hz), 6.50 (1H, s), 6.35 (1H, d, J = 7.2 Hz), 4.35 (2H, t, J = 5.3 Hz), 3.82-4.12 (2H,
m), 2.22 (3H, s), 1.44 (3H, s), 1.29 (3H, d, J = 7.3 Hz), 1.28 (3H, s), 1.19 (3H, s), 1.12
(3H, s), 0.54 (3H, s). 13C NMR (75 MHz, CDCl3) δ: 178.2, 178.1, 169.9, 164.7, 146.0,
138.0, 135.6, 134.4, 129.7 (C×3), 128.5 (C×3), 127.4, 119.5, 118.2, 72.4, 65.3, 45.0,
44.3, 42.9, 40.6, 39.5, 38.2, 36.3, 34.9, 33.6, 32.8, 32.4, 31.6, 30.57, 29.7, 28.6, 21.6,
18.5, 13.8, 13.7, 10.2. HRMS (ESI) calculated for C41H51N2O9S [M + H]+ 747.3315,
found 747.3315.
4.1.3. General procedure for synthesis of 11a-11b

ACCEPTED MANUSCRIPT
10a-b (0.11 mmol, 1.1 eq) were respectively added to a solution of 1 (0.10 mmol, 1
eq), PyBop (0.20 mmol) and hung’s base (0.25 mmol) in CH2Cl2 (5.0 mL). The
suspension was stirred at room temperature (10 h). The filtrate was evaporated under
reduced pressure. The resulting crude product was purified by column
chromatography (PE/EA = 3:1 v/v) to generate 11a-11b.
4.1.3.1. 11a
Red solid, yielded 46%. mp 151.1-152.6 °C. 1H NMR (300 MHz, CDCl3) δ: 8.03
(2H, d, J = 7.5 Hz), 7.78 (1H, d, J = 7.3 Hz), 7.62 (2H, t, J = 7.9 Hz), 7.02 (1H, d, J =
6.5 Hz), 6.50 (1H, s), 6.33 (1H, d, J = 7.2 Hz), 4.44 (2H, m), 3.52-3.78 (2H, m), 2.23
13
(3H, s), 1.45 (3H, s), 1.28 (3H, s), 1.21 (3H, s), 1.15 (3H, s), 0.64 (3H, s). C NMR
(75 MHz, CDCl3) δ: 178.8, 178.3, 170.6, 165.0, 146.0, 139.2, 135.8, 134.5, 129.8
(C×3), 128.3 (C×3), 121.9, 119.4, 118.2, 70.0, 45.0, 44.3, 43.0, 40.6, 39.4, 38.6, 38.2,
36.3, 34.8, 33.8, 33.6, 31.6, 31.0, 30.9, 30.0, 29.4, 28.7, 21.6, 18.4, 10.3. HRMS (ESI)
calculated for C39H48N3O8S [M + H]+ 718.3162, found 718.3157.
4.1.3.2. 11b
Red solid, yielded 42%. mp 146.2-147.8 °C. 1H NMR (300 MHz, CDCl3) δ: 8.03
(2H, d, J = 7.9 Hz), 7.77 (1H, d, J = 7.5 Hz), 7.63 (2H, t, J = 7.9 Hz), 7.02 (1H, d, J =
7.1 Hz), 6.53 (1H, s), 6.53 (1H, d, J = 7.5 Hz), 4.45 (2H, t, J = 5.9 Hz), 3.36-3.39 (2H,
m), 2.06 (3H, s), 1.35 (3H, s), 1.27 (3H, s), 1.19 (3H, s), 1.14 (3H, s), 0.66 (3H, s). 13C
NMR (75 MHz, CDCl3) δ: 178.3, 178.2, 169.8, 164.7, 146.0, 139.0, 134.1, 131.7,
129.3 (C×3), 128.8 (C×3), 127.0, 119.5, 118.2, 117.1, 70.3, 45.1, 44.3, 43.0, 40.5,
39.4, 38.2, 36.3 (C×3), 34.8, 33.4, 32.8, 31.6, 30.8, 30.5, 29.7, 28.6, 21.7, 18.6, 14.1,
10.3. HRMS (ESI) calculated for C40H50N3O8S [M + H]+ 732.3319, found 732.3320.
4.1.4. General procedure for synthesis of 13a-13d
To a solution of 12a-d (0.10 mmol, 1eq) in CH3CN (5.0 mL) was added AgNO3
(0.15 mmol, 1.5 eq). After 10 h stirring at 80 °C in the darkness, the filtrate was
evaporated under reduced pressure. The crude product was purified by column
chromatography (PE/EA = 10:1 v/v) to give 13a-13d.
4.1.4.1. 13a
1
Red solid, yielded 50%. mp 165.3-166.6 °C. H NMR (300 MHz, CDCl3) δ:

ACCEPTED MANUSCRIPT
7.03(1H, d, J = 7.2 Hz), 6.55 (1H, s), 6.36 (1H, d, J = 7.1 Hz), 4.54 (2H, m),
3.95-4.14 (2H, m), 2.23 (3H, s), 1.44 (3H, s), 1.30 (3H, s), 1.27 (3H, s), 1.12 (3H, s),
0.56 (3H, s). 13C NMR (75 MHz, CDCl3) δ: 178.3, 178.2, 170.2, 164.8, 146.0, 134.2,
128.4, 1194, 118.2, 117.3, 71.6, 64.8, 50.2, 44.3, 40.4, 39.4,38.2, 36.3, 34.7, 33.6,
32.7, 31.6, 30.7, 30.5, 29.8, 29.6, 28.6, 28.2, 26.2, 257, 21.6, 18.4, 10.2. HRMS (ESI)
calculated for C32H44NO7 [M + H]+ 544.3118, found 544.3120.
4.1.4.2 13b
Red solid, yielded 59%. mp 166.1-167.7 °C. 1H NMR (300 MHz, CDCl3) δ: 7.04
(1H, d, J = 6.6 Hz), 6.56 (1H, s), 6.36 (1H, d, J = 6.8 Hz), 4.50 (2H, t, J = 6.41 Hz),
3.88-4.04 (2H, m), 2.23 (3H, s), 1.47 (3H, s), 1.28 (3H, s), 1.27 (3H, s), 1.12 (3H, s),
13
0.56 (3H, s) C NMR (75 MHz, CDCl3) δ: 178.2 (C×2), 173.3, 165.4, 147.1, 134.1,
127.6, 119.9 (C×2), 118.2, 72.5, 63.5, 45.0, 44.2, 42.9, 40.4, 38.3, 36.3, 34.7, 33.5,
32.8, 31.6, 30.8, 30.5, 30.2, 29.8, 29.7 (C×2), 28.6, 24.8, 23.8, 21.6, 18.5, 10.3.
HRMS (ESI) calculated for C33H46NO7 [M + H]+ 568.3274, found 568.3273.
4.1.4.3. 13c
Red solid, yielded 46%. mp 170.4-171.8 °C. 1H NMR (300 MHz, CDCl3) δ: 7.04
(1H, d, J = 6.96 Hz), 6.55 (1H, s), 6.36 (1H, d, J = 6.84 Hz), 4.47 (2H, t, J = 6.45 Hz),
3.85-4.04 (2H, m), 2.23 (3H, s), 1.44 (3H, s), 1.30 (3H, s), 1.27 (3H, s), 1.12 (3H, s),
13
0.56 (3H, s) C NMR (75 MHz, CDCl3) δ: 178.6, 178.3, 167.1, 165.0, 146.3, 134.4,
128.4, 119.9, 118.5, 118.0, 71.4, 61.3, 45.4, 44.6, 43.3, 40.8, 39.8, 38.5, 36.7, 35.2,
33.7, 33.1, 31.9,31.8, 31.1, 30.8, 30.5, 30.1, 30.0,28.9, 28.2, 22.0, 18.9, 10.6. HRMS
(ESI) calculated for C34H48NO7 [M + H]+ 582.3431, found 582.3436.
4.1.4.4. 13d
Red solid, yielded 41%. mp 169.3-171.1 °C. 1H NMR (300 MHz, CDCl3) δ: 6.99
(1H, d, J = 7.11 Hz), 6.32 (1H, s), 5.09 (1H, d, J = 8.01 Hz), 4.34 (2H, t, J = 6.51 Hz),
3.71-4.03 (2H, m), 1.95 (3H, s), 1.35 (3H, s), 1.20 (3H, s), 1.12 (3H, s), 0.99 (3H, s),
0.61 (3H, s). 13C NMR (75 MHz, CDCl3) δ: 178.9, 178.7, 172.5, 150.9, 142.8, 130.0,
126.2, 120.0, 119.9, 73.1, 64.2, 46.4, 45.8, 41.4, 40.4, 39.2, 37.4, 36.6, 32.9, 31.6,
31.4, 31.2, 30.6, 29.7, 29.6, 28.2, 26.8, 26.7, 25.7, 25.1, 23.3, 19.4, 18.9, 10.9. HRMS
(ESI) calculated for C35H50NO7 [M + H]+ 596.3587, found 596.3588.

ACCEPTED MANUSCRIPT
4.2. Biological experiments
4.2.1. Cytotoxic assay in vitro
All hybrids were evaluated for their inhibitory activities against A549, MCF-7,
HOS and HepG2 by MTT method. Briefly, 100 µL of each cell (5.0 × 104 cells/ml)
were seeded into 96-well plates and incubated for 24 h. Different concentrations of
the drug were added into the 96-well plates. Then these plates were incubated for 48 h.
MTT solution (5 mg/ml) was added, and cultured (4 h). Supernatant was abandoned
before adding 100 ml DMSO to each well. The OD at 570 nm was measured using a
microplate reader (POLARstar Omega, Offenburg, Germany). Every assay was
performed in triplicate. Data are presented as the mean ± SD (n = 3).
4.2.2. NO release amounts detection
NO release amounts in vitro were calculated by Nitrate/Nitrite Assay Kit
(KeyGEN, China). Firstly, A549 cells (3×100 µL, 5.0×104 cells/ml) were incubated
in 96 well-plates for 24 h at 37 oC and then treated with the indicated hybrids (4 µM)
for different time. Secondly, the amounts of NO in vitro were measured according to
the manufacturer’s instructions. The OD at 550 nm was measured using a microplate
reader (POLARstar Omega, Offenburg, Germany). Data were mean of three
independent experiments.
4.2.3. Western blotting analysis
A549 cells were seeded (5×105 cells) in 6 cm dishes and incubated with 24 h. 11b
was added to the dishes with indicated concentration and then these dishes were
incubated for another 24 h. The cells were washed and lysed using lysis buffer. The
lysates were centrifuged at 10,000 g for 20 min at 4 °C. Then the proteins were
diluted to 3 mg/ml (BCA method). Each sample (10 µL) was analyzed by
SDS-PAGE (10 % gel). Then the proteins were detected by the conventional method.
The monoclonal antibodies were purchased (Abcam, Cambridge, UK).
4.2.4. Cell cycle analysis
A549 cells (4×105 cells) were incubated in 6 cm dishes for 24 h. Then 11b and 1
were added to the dishes with different concentrations. After incubation of 24 h, the
cells were harvested and washed twice with cold PBS. Then the cell was fixed in

ACCEPTED MANUSCRIPT
ethanol (75 %) at - 20 °C for at least 12 h. The cells were washed with buffer A to
remove the ethanol. The cells were followed suspended in buffer A (250 µL)
containing 250 mg/ml RNase A and cultured at 37 °C for 30 min. Finally, 2.5 µL
propidium iodide (PI) (KeyGEN, China) was added and incubated at room
temperature in dark for 15 min. The cells were analyzed by flow cytometry (BD
Accuri C6 flow cytometer, Franklin Lakes, NJ).
4.2.5. Cell apoptosis analysis
A549 cells (4×105 cells) were incubated in 6 cm dish for 24 h. Then 11b and 1 were
added to the cells with indicated concentrations. After another incubation of 48 h, the
cells were harvested softly and washed twice with cold PBS (2000 g for 5 min). Then
the cells were added to 250 µL binding buffer containing 2.5 µL PI and 2.5 µL
Annexin V (KeyGEN, China). The mixture was incubated at room temperature for 15
min in the darkness and analyzed by flow cytometry.
4.2.6. Hoechst-33342-staining
A549 cells (4×105 cells) were incubated in 6 cm dish for 24 h. Then A549 cells
were treated with the 11b at indicated concentrations for 48 h. Cells were washed
twice by PBS. Then cells were incubated with Hoechst-33342 (10 µg/ml) for 5 min at
room temperature in the darkness. After incubation, stained cells were observed under
a fluorescent microscope (CORPORATION, Japan).
Acknowledgements
This work was financially supported by the Academic Innovation Project of Jiangsu
Department of Education (No. 1131740036).
Appendix A. Supplementary data
Supplementary data related to this article can be found at.
References
[1] A. Salminen, M. Lehtonen, T. Paimela, K. Kaarniranta. Biochem. Biophys. Res.
Commun. 394 (2010) 439-442.
[2] S.A.C. Figueiredo, J.A.R. Salvador, R. Cortés, M. Cascante. Eur. J. Med. Chem.
139 (2017) 836-848.

ACCEPTED MANUSCRIPT
[3] W. Tang, J. Wang, X. Tong, J. Shi, X. Liu, J. Li. Eur. J. Med. Chem. 95 (2015)
166-173.
[4] T. Zhang, Y. Li, Y. Yu, P. Zou, Y. Jiang, D. Sun. J. Biol. Chem. 51 (2009)
35381-35389.
[5] T. Zhang, A. Hamza, X. Cao, B. Wang, S. Yu, C. Zhan, D. Sun. Mol. Cancer Ther.
7 (2008) 162-170.
[6] S.M. Roe, M.M. Ali, P. Meyer, C.K. Vaughan, B. Panaretou, P.W. Piper, C.
Prodromou, L.H. Pearl. Cell 116 (2004) 87-98.
[7] L. Whitesell, S.L. Lindquist. Nat. Rev. Cancer 5 (2005) 761-772.
[8] L. H. Pearl, Curr. Opin. Genet. Dev. 15 (2005) 55-61.
[9] B. Bonavida, S. Khineche, S. Huerta-Yepez, H. Garban. Updates 9 (2006)
157-173.
[10] B. Bonavida, S. Baritaki, S. Huerta-Yepez, M.I. Vega, D. Chatterjee, K. Yeung.
Nitric Oxide 19 (2008) 152-157.
[11] Z. Huang, J. Fu, Y. Zhang. J. Med. Chem. 60 (2017) 7617-7635.
[12] B. Bonavida, S. Khineche, S. Huerta-Yepez, H. Garban. Drug Resist. Updates 9
(2006) 157-173.
[13] R. Alexandrova, M. Mileva, E. Zvetkova. Exp. Pathol. Parasitol 4 (2001) 13-18.
[14] X. Gu, Z. Huang, Z. Ren, X. Tang, R. Xue, X. Luo, S. Peng, Hui Peng, B. Lu, J.
Tian, Y. Zhang. J. Med. Chem. (60) 2017, 928-940.
[15] G. Aguirre, M. Boiani, H. Cerecetto, M. Fernández, M. González, E. León, C.
Pintos, S. Raymondo, C. Arredondo, J.P. Pacheco, M.A. Basombrío. Pharmazie 61
(2006) 54-59.
[16] M.A.E. Mourad, M. Abdel-Aziz, G.E.D.A.A. Abuo-Rahma, H.H. Farag. Eur. J.
Med. Chem. 54 (2012) 907-913.
[17] Y. Ai, F.H. Kang, Z.J. Huang, X.W. Xue, Y.S. Lai, S.X. Peng, J.D. Tian, Y.H.
Zhang. J. Med. Chem. 58 (2015) 2452-2464.
[18] N. Zhao, K.T. Tian, K.G. Cheng, T. Han, X. Hu, D.H. Li, Z.L. Li, H.M. Hua.
Bioorg. Med. Chem. 24 (2016) 2971-2978.
[19] J. Chen, T. Wang, S. Xu, P. Zhang, A. Lin, L. Wu, H. Yao, W. Xie, Z. Zhu, J. Xu.

ACCEPTED MANUSCRIPT
Eur. J. Med. Chem. 135 (2017) 414-423.
[20] N. Kerru, P. Singh, N. Koorbanally. R. Raj, V. Kumar. Eur. J. Med. Chem. 142
(2017) 179-212.
[21] P. Klahn, M. Brönstrup. Nat. Prod. Rep. 34 (2017) 832-885.
[22] K. Singh, V. Verma, K. Yadav, V. Sreekanth, D. Kumar, A. Bajaj, V. Kumar. Eur.
J. Med. Chem. 87 (2014) 150-158.
[23] Y. Garazd, M. Garazd, R. Lesyk Saudi. Pharm. J. 25 (2017) 214-223.
[24] Z. Huang, J. Wu, Y. Zou, H. Yuan, Y. Zhang, Y. Fei, A. Bhardwaj, J. Kaur, E.E.
Knaus, Y. Zhang. J. Med. Chem. 61 (2018) 1833-1844.
[25] L. Chen, Y. Zhang, X. Kong, E. Lan, Z. Huang, S. Peng, D.L. Kaufman, J. Tian.
J. Med. Chem. 51 (2008) 4834-4838.
[26] Z. Huang, Y. Zhang, L. Zhao, Y. Jing, Y. Lai, L. Zhang, Q. Guo, S. Yuan, J.
Zhang, L. Chen, S. Peng, J. Tian. Org. Biomol. Chem. 8 (2010), 632-639.
[27] Y. Ai, F.H. Kang, Z.J. Huang, X.W. Xue, Y.S. Lai, S.X. Peng, J.D. Tian, Y. H.
Zhang. J. Med. Chem. 58 (2015) 2452-2464.
[28] Y. Lai, L.Shen, Z. Zhang, W. Liu, Y. Zhang, H. Ji, J. Tian. Bioorg. Med. Chem.
Lett. 20 (2010) 6416-6420.
[29] N. Bao, J. Ou, M. Xu, F. Guan, W. Shi, J. Sun, L. Chen. Eur. J. Med. Chem.
137 (2017) 88-95.
[30] K.M. Miranda, M.G. Espey, D.A. Wink. Nitric Oxide 5 (2001) 62-71.
[31] F. Jiang, H.J. Wang, Q.C. Bao, L. Wang, Y.H. Jin, Q. Zhang, D. Jiang, Q.D. You,
X.L. Xu. Bioorg. Med. Chem. 24 (2016) 5431-5439.
[32] P. Tongkate, W. Pluempanupat, W. Chavasiri. Tetrahedron Letters 49 (2008)
1146-1148.
[33] L. Chen, Y. Zhang, X. Kong, S. Peng, J. Tian. Bioorg. Med. Chem. Lett. 17
(2007) 2979-2982.
[34] F. Amano, T. Noda. FEBS Lett. 368 (1995) 425-428.
[35] A. Chadli, S.J. Felts, Q. Wang, W.P. Sullivan, M. Victoria Botuyan, A. Fauq, M.
Ramirez-Alvarado, G. Mer. J. Biol. Chem. 6 (2010) 4224-4231.
[36] S. Sreeramulu, H.R.A. Jonker, T. Langer, C. Richter, C. Roy D. Lancaste, H.

ACCEPTED MANUSCRIPT
Schwalbe. J. Biol. Chem. 6 (2006) 3885-3896.
[37] S. Sreeramulu, S.L. Gande, M. Göbel, H. Schwalbe. Angewandte Chemie 48
(2009) 5853-5855.

ACCEPTED MANUSCRIPT
Figure captions
Fig. 1. Effects of NO produced by the target compounds on A549 cell proliferation.
(A) NO release amounts of partial hybrids producing vs. time. A549 cells were treated
with 4 µM of the selected compounds for different hours (mean ± SD, n = 3). (B) The
levels of NO produced by all of the compounds in A549 cells. A549 cells were treated
with 4 µM of all the hybrids for 24 hours (mean ± SD, n = 3).
Fig. 2. Effect of PTIO on the anti-proliferative activity of 11b. A549 cells were
pretreated with the indicated concentrations of PTIO for 2 h and then treated with 4
µM of 11b for another 24 h. Data were expressed as the mean ± SD (n=3). *P < 0.05,
**P < 0.01 vs. the group without PTIO.
Fig. 3. Effects of 11b on Hsp90 clients in A549 cells. A549 cells were treated with
varying concentrations of 11b for 24 h. (A) The expressions of Cdk-4 and Hsp90 were
determined by Western Blotting using specific antibodies. β-actin was used as internal
control; (B) The expressions of p-Akt and Akt were determined by Western Blotting
using specific antibodies.
Fig. 4. Anti-proliferative activity of 11b in comparison with that of 1, 10b and their
mixture in A549 cells.
Fig. 5. Fluorescence microscopy images of A549 cells stained by Hoechst 33342.
A549 cells were treated with the 11b at indicated concentrations for 48 h.
Fig. 6. Effect of 11b on apoptosis of A549 cells. (A) A549 cells were treated with 11b
at the indicated concentrations for 48 h. Apoptotic effects were measured by flow
cytometry using Annexin V-FITC/PI staining protocol; (B) Representative histograms
for the numbers of cells (% of total) in the early and late stages of apoptosis for the
treatment groups.
Fig. 7. Cell cycle analysis of 11b in A549 cells. Cells were treated with 11b at 0.2, 0.4
and 0.8 µM for 24 h, harvested, stained with PI, and then analyzed by flow cytometry.

ACCEPTED MANUSCRIPT
ò A series of celastrol/NO donor hybrids were designed, synthesized and evaluated.
ò 11b exhibited superior potency compared to the other compounds.
ò 11b induced degradation of the Hsp90 clients (Akt and Cdk4), apoptosis and the
cell cycle arrested at the G0/G1 phase in A549 cells.