Novel non-trimethoxylphenyl piperlongumine derivatives selectively kill cancer cells

Article abstract of DOI:10.1016/j.bmcl.2017.04.035

Piperlongumine (PL) is a natural alkaloid with broad biological activities. Twelve analogues have been designed and synthesized with non-substituted benzyl rings or heterocycles in this work. Most of the compounds showed better anticancer activities than the parent PL without apparent toxicity in normal cells. Elevation of cellular ROS levels was one of the main anticancer mechanisms of these compounds. Cell apoptosis and cell cycle arrest for the best compound ZM90 were evaluated and similar mechanism of action with PL was demonstrated. The SAR was also characterized, providing worthy directions for further optimization of PL compounds.

Full text of DOI:10.1016/j.bmcl.2017.04.035

Accepted Manuscript
Novel non-trimethoxylphenyl piperlongumine derivatives selectively kill can-
cer cells
Youjun Zhang, Hao Ma, Yuelin Wu, Zhongli Wu, Zhengguang Yao, Wannian
Zhang, Chunlin Zhuang, Zhenyuan Miao
PII:
S0960-894X(17)30411-0
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.04.035
BMCL 24888
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
26 February 2017
11 April 2017
12 April 2017
Please cite this article as: Zhang, Y., Ma, H., Wu, Y., Wu, Z., Yao, Z., Zhang, W., Zhuang, C., Miao, Z., Novel non-
trimethoxylphenyl piperlongumine derivatives selectively kill cancer cells, Bioorganic & Medicinal Chemistry
Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.04.035
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.

Novel non-trimethoxylphenyl piperlongumine derivatives selectively kill
cancer cells
a†
Youjun Zhang , Hao Ma , Yuelin Wu , Zhongli Wu , Zhengguang Yao , Wannian Zhang
a,b,†
c
a
b
a,b
,
a,*
Chunlin Zhuang , Zhenyuan Miao
a,*
a
Second Military Medical University, 800 Xiangyin Road, Shanghai 200433, China
Ningxia Medical University, Yinchuan 750004, China
b
c
Shanghai Institute of Technology, 100 Haiquan Road, Shanghai 201418, China
*Corresponding authors.
E-mail addresses: zclnathan@163.com (C.-L. Zhuang); miaozhenyuan@hotmail.com (Z.-Y.
Miao).
†
These authors contributed equally to this work.

Abstract
Piperlongumine (PL) is a natural alkaloid with broad biological activities. Twelve analogues have
been designed and synthesized with non-substituted benzyl rings or heterocycles in this work.
Most of the compounds showed better anticancer activities than the parent PL without apparent
toxicity in normal cells. Elevation of cellular ROS levels was one of the main anticancer
mechanisms of these compounds. Cell apoptosis and cell cycle arrest for the best compound
ZM90 were evaluated and similar mechanism of action with PL was demonstrated. The SAR was
also characterized, providing worthy directions for further optimization of PL compounds.
Keywords: Piperlongumine, Natural product, Drug design, Structure-activity relationships,
Anticancer

Piperlongumine (PL, Piplartine, 5,6-dihydro-1-[(2E)-1 -oxo-3-(3,4,5-trimethoxyphenyl)-2-
propenyl]-2(1H)-pyridinone, Scheme 1) is an active alkaloid isolated from the root of plant
species Piper longum L. [1] The therapeutic potentials include anti-cancer, [2-4] anti-diabetes,[5]
antiplatelet aggregation,[6] anti-inflammation,[7, 8] anti-fungal activity [9, 10] and
neurodegenerative diseases.[11] In 2011, Schreiber group reported the PL selectively killed
cancer cells probably due to elevation of ROS level.[3] However, the exact protein target(s) has
not well characterized. Further structure-activity relationship (SAR) study showed the PL
compounds had two key pharmacophores: C2-C3 and C7-C8 double bonds (Scheme 1,
highlighted in red) where contain two classical Michael acceptors.[12] Comparing these two
Michael acceptors, the C2-C3 double bond is more reactive with cysteines of the target proteins.
Removal of C7-C8 double bond led to a decrease in cytotoxicity without decreasing the ROS
level. In 2014, our group further extended the SARs by introducing halogen or morpholine
substituents at C2 and alkyl substituents at C7 position.[13] The 2-halogenated piperlongumines
(
e.g. ZM30, Scheme 1) showed potent in vitro and in vivo anti-cancer activities due to its higher
reactivity of C2-C3 Michael acceptor. The 2-morpholine substituted ones (IC50 >100 µM) were
deactivated probably due to steric hindrance. Compound ZM30 showed better activity than PL
against A549 cells. However, the selectivity for the cancer cells (cancer cells, A549, IC50 = 19.83
µM; normal cells, MRC-5, IC50 = 14.30 µM) was poor.

Scheme 1. The design strategy of non-trimethoxylphenyl piperlongumine derivatives.
Based on above information and SAR study, the importance of the trimethoxylphenyl group has
not been evaluated, which appears in lots of natural anticancer agents, such as colchicine and
combretastatin A4. Herein, non-substituted benzyl rings and heterocycles were designed to
replace the trimethoxylphenyl group (Scheme 1A). We also proposed to replace the C7-C8 double
bond by the bioisosteric groups, such as cyclopropyl group (Scheme 1B) and triple bond (Scheme
1
C). Removal of the C7-C8 olefin (Scheme 1D) is another aspect for evaluation.

Scheme 2. Conditions and reagents: (a). anhydrous DCM, oxalyl chloride, one drop of DMF, 3 h;
(b). anhydrous THF, TEA, room temperature, overnight; (c). For chlorination: PCl
, CHCl ; room
5 3
o
, DCM, 0 C, then ZnI, Br
temperature, crude product for next step; For bromination, PCl
5
2
, room
o
, LiCl, anhydrous DMF, 130 C, 7 h; (e). anhydrous DCM, oxalyl
temperature; (d). Li CO
2
3
chloride, one drop of DMF, 3 h; (f). anhydrous THF, TEA, room temperature, overnight.
Twelve analogues (Scheme 2) were obtained to evaluate the in vitro cytotoxicity against six
cancer cell lines (human lung carcinoma A549, human colorectal carcinoma HCT116, human
breast carcinoma MDA-MB-231, human Hepatic adenocarcinoma SK-Hep-1, human
osteosarcoma U2OS and Saos-2) and a normal cell line (human lung fibroblast cell lines WI38)
by MTT assay. PL was used as a reference compound. The IC50 values for each compound are
summarized in Table 1.

Table 1. In vitro anticancer activity of the derivatives
a
IC50 (µM)
Compounds
HCT-
16
MDA-MB-
231
SK-
Saos- WI38 (Normal
A549
U2OS
1
Hep-1
13.3
72.9
1.19
>100
13.6
4.83
3.52
75.2
0.71
1.90
1.27
2.78
3.96
2
cells)
>100
>100
>100
>100
>100
31.0
PL
6.84
85.1
7.93
>100
18.6
8.64
10.6
>100
4.12
2.97
8.30
11.0
8.60
7.34
64.5
0.92
>100
5.89
3.84
10.2
>100
1.67
2.36
4.49
4.37
4.20
10.6
9.49
68.3
1.12
>100
8.60
5.67
6.63
97.5
1.17
1.67
1.47
2.26
7.29
7.31
23.5
0.42
89.9
3.86
2.55
2.96
26.7
0.011
0.53
0.28
1.15
0.82
ZM83
ZM84
ZM85
ZM86
ZM87
ZM88
ZM89
ZM90
ZM91
ZM92
ZM93
65.1
1.73
>100
19.2
4.83
8.68
>100
>100
84.6
>100
1.67
2.97
47.0
5.52
>100
>100
89.5
4.97
ZM94
5.04
a
Values were measured with MTT method.
Overall, most of the novel PL analogues showed better cytotoxicity than the parent compound
PL, except ZM83, ZM85 and ZM89. These compounds were sensitive to human osteosarcoma
cells with the IC50 values in nanomolar or low micromolar range. To our delight, all the
compounds show no apparent cytotoxicity against normal cells, recovering the selectivity of PL.
The SAR study demonstrated the following results: (1) The trimethoxyl group was removed from
ZM30 to obtain ZM90. This compound showed excellent cytotoxicity against all the six cell lines,
especially the SK-Hep-1 (IC50 = 710 nM) and Saos-2 (IC50 = 11 nM), with only 84.6 µM of the

normal WI38 cells. This result indicated that the trimethoxyl group is not a pharmacophore for
the PL compounds. (2) ZM84 with triple bond showed excellent cytotoxicity against five cell
lines (IC50 = 420~1730 nM) except A549. (3) The heterocycle-substituted PL compounds
(ZM91-94) exhibited good activities. ZM91 with 3-pyridinyl group was potent against the six
cell lines (IC50 = 530~2970 nM). The activities of ZM92 with 2-furyl, ZM93 with 3-furyl and
ZM94 with double 2-halo-lactams were comparable with that of PL towards A549 cells. For the
other five cell lines, they showed 2~9-fold better activities than PL. (4) Removal of the C7-C8
olefin to directly connect the chlorides to the lactam was unfavorable to the cytotoxicity.
Compared with ZM91 and ZM87, the C7-C8 olefin was removed, leading to 2 ~ 5-fold decrease
in the cytotoxicity. Compounds ZM85 and ZM87~89 showed moderate cytotoxicity.
To summarize the SAR, C7-C8 olefin is of great importance to the cytotoxicity. The alkynyl was
compatible to replace the double bond but not for the cyclopropyl group. The enhancement of the
molecular flexibility or disruption of the conjugated system of α,β-unsaturated system were
proposed to be the main reasons. Moreover, the trimethoxyl group is not a critical pharmacophore
to the PL analogues.
It is reported that the antitumor mechanism of PL is to increase the level of ROS and apoptotic
cell death.[3] We further evaluated the cellular ROS levels of the most potent compounds (ZM84,
ZM90, and ZM91) using two sensitive cancer cell lines (HCT-116 and Saos-2) by fluorescence
microscopy (Figure 1). The HCT-116 cellular ROS levels were obviously increased after 1 h
treatment with 10 μM of ZM90 or ZM91, which was better than the PL. However, the ZM84
without C7-C8 double bond showed no apparent change. In Saos-2 cells, the compounds
exhibited two-fold change in ROS level, but less than that of PL.

Figure 1. Piperlongumine derivatives induced ROS elevation. HCT-116 and Saos-2 cells were
treated with compounds (10 μM) or DMSO (control) for 1 h.
Further dose- and time-response manners of the active compound ZM90 were then evaluated.
The compound showed a significant time-dependent manner in both cell lines. For the dose, the
compound increased the cellular ROS levels from 1 to 5 μM. And at 10 μM, the ROS levels in
two cell lines were slightly decreased (Figure 2).
Figure 2. Piperlongumine derivatives induced ROS elevation. HCT-116 and Saos-2 cells were
treated with compound ZM90 in dose (1, 5, 10 μM)- and time (1 h and 3 h)-response manner.
Cell morphology and fluorescence-activated cell sorting was used to evaluate the effect of
compound ZM90 on induction of apoptosis (Figure 3). Compared with the DMSO control group,
compound PL and ZM90 can induce the apoptosis of both cell lines in a dose-response manner.
As shown in Figure 3A, the percentage of HCT116 apoptotic cells after ZM90 treatment for 24 h

was 16% and 17% at 1 and 5 μM, which was higher than that of PL. For Saos-2 cells, the
percentage of apoptotic cells after PL treatment was 26% and 33%, where ZM90 has 12.7% and
2
1.5% induction (Figure 3B). The result suggested that the piperlongunime derivatives showed
remarkable apoptotic effect in cancer cells.
Figure 3. Cell apoptosis induced by ZM90 after 24 h treatment on HCT-116 (A) and Saos-2 (B)
cell lines.
In addition, flow cytometric analysis was also performed to determine the cell cycle arrest of

ZM90 (Figure 4 and Figure S1). After 24 h treatment of compounds in HCT-116 cells, it was
obvious that PL resulted in an increase in the percentage of cells blocked in the S and G0/G1
phase. Similar effect was observed for ZM90 and more cell cycle populations in G0/G1 phase
were arrested in a dose-response manner (Figure 4A). In Saos-2 cells, no significant change was
demonstrated in G2/M and S phases. Slight decrease in G0/G1 phase was shown (Figure 4B). The
above results indicated ZM90 and PL had a similar mechanism of action for the cancer treatment.
Figure 4. The effect of PL or ZM90 treatment on HCT-116 and Saos-2 cell cycle distributions for
2
4 h.
In conclusion, twelve PL analogues were designed and synthesized with non-substituted benzyl
rings or heterocycles. Most of the compounds exhibited better anticancer activities than the parent
PL. Different from the initial halogenated PL, these compounds had a higher selectivity towards
cancer cells and normal cells. ROS elevation was one of the main anticancer mechanisms of these
compounds. Cell apoptosis and cell cycle arrest for compound ZM90 were evaluated and similar
mechanism of action with PL was demonstrated. The SAR has been also characterized, providing
worthy directions for further optimization of PL compounds.
Acknowledgments
This research was funded by the grants from the National Natural Science Foundation of China
(81502978 to C.Z. and 81673352 to Z.M.), the Bio-Pharmaceutical Project of Science and
Technology of Shanghai (15431901700 to Z.M.), and Shanghai ‘‘ChenGuang’’ Project (16CG42

to C.Z.).
Supplementary data
Supplementary data associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.bmcl.2017.###.
References and notes
[
1] A. Chatterjee, C.P. Dutta, The structure of piperlongumine, a new alkaloid isolated from the roots of
Piper longum., Sci. Cult., 29 (1963) 568-570.
2] D.P. Bezerra, G.C. Militao, F.O. de Castro, C. Pessoa, M.O. de Moraes, E.R. Silveira, M.A. Lima, F.J.
[
Elmiro, L.V. Costa-Lotufo, Piplartine induces inhibition of leukemia cell proliferation triggering both
apoptosis and necrosis pathways, Toxicol. In Vitro, 21 (2007) 1-8.
[
3] L. Raj, T. Ide, A.U. Gurkar, M. Foley, M. Schenone, X. Li, N.J. Tolliday, T.R. Golub, S.A. Carr, A.F.
Shamji, A.M. Stern, A. Mandinova, S.L. Schreiber, S.W. Lee, Selective killing of cancer cells by a small
molecule targeting the stress response to ROS, Nature, 475 (2011) 231-234.
[
4] P. Zou, Y. Xia, J. Ji, W. Chen, J. Zhang, X. Chen, V. Rajamanickam, G. Chen, Z. Wang, L. Chen, Y.
Wang, S. Yang, G. Liang, Piperlongumine as a direct TrxR1 inhibitor with suppressive activity against
gastric cancer, Cancer Lett., 375 (2016) 114-126.
[
5] V.R. Rao, P. Muthenna, G. Shankaraiah, C. Akileshwari, K.H. Babu, G. Suresh, K.S. Babu, R.S.
Chandra Kumar, K.R. Prasad, P.A. Yadav, J.M. Petrash, G.B. Reddy, J.M. Rao, Synthesis and biological
evaluation of new piplartine analogues as potent aldose reductase inhibitors (ARIs), Eur. J. Med. Chem., 57
(
[
2012) 344-361.
6] B.S. Park, D.J. Son, W.S. Choi, G.R. Takeoka, S.O. Han, T.W. Kim, S.E. Lee, Antiplatelet activities of
newly synthesized derivatives of piperlongumine, Phytother. Res., 22 (2008) 1195-1199.
7] S.K. Ku, J.A. Kim, J.S. Bae, Vascular barrier protective effects of piperlonguminine in vitro and in vivo,
Inflamm Res, 63 (2014) 369-379.
8] Y.H. Seo, J.K. Kim, J.G. Jun, Synthesis and biological evaluation of piperlongumine derivatives as
potent anti-inflammatory agents, Bioorg. Med. Chem. Lett., 24 (2014) 5727-5730.
9] H.M. Navickiene, A.C. Alecio, M.J. Kato, V.D. Bolzani, M.C. Young, A.J. Cavalheiro, M. Furlan,
Antifungal amides from Piper hispidum and Piper tuberculatum, Phytochemistry, 55 (2000) 621-626.
10] R. Vasques da Silva, H.M. Navickiene, M.J. Kato, S. Bolzani Vda, C.I. Meda, M.C. Young, M. Furlan,
Antifungal amides from Piper arboreum and Piper tuberculatum, Phytochemistry, 59 (2002) 521-527.
11] S. Peng, B. Zhang, X. Meng, J. Yao, J. Fang, Synthesis of piperlongumine analogues and discovery of
[
[
[
[
[
nuclear factor erythroid 2-related factor 2 (Nrf2) activators as potential neuroprotective agents, J. Med.
Chem., 58 (2015) 5242-5255.
[
12] D.J. Adams, M. Dai, G. Pellegrino, B.K. Wagner, A.M. Stern, A.F. Shamji, S.L. Schreiber, Synthesis,
cellular evaluation, and mechanism of action of piperlongumine analogs, Proc. Natl. Acad. Sci. U. S. A.,
09 (2012) 15115-15120.
13] Y. Wu, X. Min, C. Zhuang, J. Li, Z. Yu, G. Dong, J. Yao, S. Wang, Y. Liu, S. Wu, S. Zhu, C. Sheng, Y.
1
[
Wei, H. Zhang, W. Zhang, Z. Miao, Design, synthesis and biological activity of piperlongumine derivatives
as selective anticancer agents, Eur. J. Med. Chem., 82 (2014) 545-551.

FIGURE LEGEND
Scheme 1. The design strategy of non-trimethoxylphenyl piperlongumine derivatives.
Scheme 2. Conditions and reagents: (a). anhydrous DCM, oxalyl chloride, one drop of DMF, 3 h;
(b). anhydrous THF, TEA, room temperature, overnight; (c). For chlorination: PCl
, CHCl ; room
5 3
o
, DCM, 0 C, then ZnI, Br
temperature, crude product for next step; For bromination, PCl
5
2
, room
o
, LiCl, anhydrous DMF, 130 C, 7 h; (e). anhydrous DCM, oxalyl
temperature; (d). Li CO
2
3
chloride, one drop of DMF, 3 h; (f). anhydrous THF, TEA, room temperature, overnight.
Table 1. In vitro anticancer activity of the derivatives
Figure 1. Piperlongumine derivatives induced ROS elevation. HCT-116 and Saos-2 cells were
treated with compounds (10 μM) or DMSO (control) for 1 h.
Figure 2. Piperlongumine derivatives induced ROS elevation. HCT-116 and Saos-2 cells were
treated with compound ZM90 in dose (1, 5, 10 μM)- and time (1 h and 3 h)-response manner.
Figure 3. Cell apoptosis induced by ZM90 after 24 h treatment on HCT-116 (A) and Saos-2 (B)
cell lines.
Figure 4. The effect of PL or ZM90 treatment on HCT-116 and Saos-2 cell cycle distributions for
24 h.

Graphical abstract
Twelve analogues have been designed and synthesized with non-substituted benzyl rings or
heterocycles in this work.

Highlights
1
2
3
. Non-substitution on benzyl rings or heterocycles-containing piperlongumine were designed.
. Most of the compounds selectively killed cancer cells.
. ZM90 had a similar mechanism of action with PL.