Senolytic activity of piperlongumine analogues: Synthesis and biological evaluation

Article abstract of DOI:10.1016/j.bmc.2018.06.013

Selective clearance of senescent cells (SCs) has emerged as a potential therapeutic approach for age-related diseases, as well as chemotherapy- and radiotherapy-induced adverse effects. Through a cell-based phenotypic screening approach, we recently identified piperlongumine (PL), a dietary natural product, as a novel senolytic agent, referring to small molecules that can selectively kill SCs over normal or non-senescent cells. In an effort to establish the structure-senolytic activity relationships of PL analogues, we performed a series of structural modifications on the trimethoxyphenyl and the α,β-unsaturated δ-valerolactam rings of PL. We show that modifications on the trimethoxyphenyl ring are well tolerated, while the Michael acceptor on the lactam ring is critical for the senolytic activity. Replacing the endocyclic C2–C3 olefin with an exocyclic methylene at C2 render PL analogues 47–49 with increased senolytic activity. These α-methylene containing analogues are also more potent than PL in inducing ROS production in WI-38 SCs. Similar to PL, 47–49 reduce the protein levels of oxidation resistance 1 (OXR1), an important oxidative stress response protein that regulates the expression of a variety of antioxidant enzymes, in cells. This study represents a useful starting point toward the discovery of senolytic agents for therapeutic uses.

Full text of DOI:10.1016/j.bmc.2018.06.013

Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Senolytic activity of piperlongumine analogues: Synthesis and biological
evaluation
Xingui Liua, Yingying Wanga, Xuan Zhanga, Zhengya Gaoa, Suping Zhanga, Peizhong Shia,
⁎
Xin Zhanga, Lin Songa, Howard Hendricksona, Daohong Zhoua,b, Guangrong Zhenga,c,
a
Department of Pharmaceutical Sciences, College of Pharmacy, University of Arkansas for Medical Sciences, Little Rock, AR 72205, United States
Department of Pharmacodynamics, College of Pharmacy, University of Florida, Gainesville, FL 32610, United States
b
c
Department of Medicinal Chemistry, College of Pharmacy, University of Florida, Gainesville, FL 32610, United States
A R T I C L E I N F O
A B S T R A C T
Keywords:
Selective clearance of senescent cells (SCs) has emerged as a potential therapeutic approach for age-related
diseases, as well as chemotherapy- and radiotherapy-induced adverse effects. Through a cell-based phenotypic
screening approach, we recently identified piperlongumine (PL), a dietary natural product, as a novel senolytic
agent, referring to small molecules that can selectively kill SCs over normal or non-senescent cells. In an effort to
establish the structure-senolytic activity relationships of PL analogues, we performed a series of structural
modifications on the trimethoxyphenyl and the α,β-unsaturated δ-valerolactam rings of PL. We show that
modifications on the trimethoxyphenyl ring are well tolerated, while the Michael acceptor on the lactam ring is
critical for the senolytic activity. Replacing the endocyclic C2–C3 olefin with an exocyclic methylene at C2
render PL analogues 47–49 with increased senolytic activity. These α-methylene containing analogues are also
more potent than PL in inducing ROS production in WI-38 SCs. Similar to PL, 47–49 reduce the protein levels of
oxidation resistance 1 (OXR1), an important oxidative stress response protein that regulates the expression of a
variety of antioxidant enzymes, in cells. This study represents a useful starting point toward the discovery of
senolytic agents for therapeutic uses.
Piperlongumine
Senescent cell
Senolytic agent
ROS
OXR1
1. Introduction
repair and remodeling. However, SCs can accumulate if their produc-
tion exceeds their elimination by the immune system or if the immune
Cellular senescence was termed by Hayflick and Moorhead on the
basis of their observation that normal cells have a limited proliferative
capacity and eventually enter a state of permanent growth arrest fol-
lowing extended culture.1,2 This phenomenon, also known as re-
plicative senescence, is the result of telomere shortening. Cells can also
undergo senescence in response to cellular stresses that are independent
of telomere length, such as exposure to genotoxic and oncogenic
stress.3,4 In this context, the senescence response acts as an effective
barrier to tumorigenesis by eliminating aged or damaged cells.3–9
However, senescent cells (SCs) are metabolically and transcriptionally
active. They secrete various pro-inflammatory cytokines, chemokines,
growth factors, and extracellular matrix proteases, a senescent pheno-
type termed the senescence-associated secretory phenotype (SASP).3
The SASP factors can recruit immune cells to clear damaged cells and
SCs, as well as stimulate the proliferation and differentiation of stem
and progenitor cells to repopulate the damaged tissue.3,10–12 Thus,
transient induction of senescence is considered beneficial in tissue
system is unable to efficiently clear these cells due to im-
munosenescence.13 Accumulation of SCs and the associated SASP over
time lead to deterioration in local tissue environment by inducing
chronic oxidative stress and inflammation, which plays a significant
role in aging and the development of age-related diseases, including
cancer.3,10,14–16
Recent studies have shown that genetic clearance of p16Ink4a-posi-
tive SCs via INK-ATTAC, which functions as a SC suicide transgene,
coupled with administration of a fusion protein dimerizer, AP20187,
prolonged lifespan of mice and delayed the onset of a number of age-
related diseases and disorders in both progeroid and naturally aged
mice.17,18 These findings indicate that pharmacological clearance of
SCs is a potential “anti-aging” approach to prolong healthspan, the
average numbers of years of living without disability or other impair-
ment. The great therapeutic potential of targeting SCs has encouraged
us and others to search “senolytic agents”, refer to small molecules that
can selectively kill SCs.19–26 We carried out a screening campaign
⁎
Corresponding author at: Department of Medicinal Chemistry, College of Pharmacy, University of Florida, Gainesville, FL 32610, United States.
E-mail address: zhengg@cop.ufl.edu (G. Zheng).
https://doi.org/10.1016/j.bmc.2018.06.013
Received 24 April 2018; Received in revised form 4 June 2018; Accepted 12 June 2018
0968-0896/©2018ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:Liu,X.,Bioorganic&MedicinalChemistry(2018),https://doi.org/10.1016/j.bmc.2018.06.013

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
against a library of structurally diverse, rationally selected small mo-
lecules that target pathways predicted to be important for SC survival
by titrating their cytotoxicity against normal or non-senescent human
WI-38 fibroblasts (WI-38 NCs) and ionizing radiation (IR)-induced se-
nescent WI-38 fibroblasts (WI-38 IR-SCs).20 As a result, we identified
two senolytic agents. The first one is ABT-263 (navitoclax), an inhibitor
of the anti-apoptotic Bcl-2 family proteins currently in Phase II clinical
trials for cancer treatment.27 We found that ABT-263 was highly spe-
cific in promoting apoptosis in SCs.20 Oral administration of ABT-263
effectively depletes IR-induced SCs in various tissues, including senes-
cent epithelial cells in the lungs and senescent hematopoietic stem cells
(HSCs) in bone marrow; and importantly, elimination of senescent
HSCs in bone marrow by ABT-263 functionally rejuvenates HSCs in
both sublethally irradiated (6 Gy) and naturally-aged mice.20 In addi-
tion, clearance of SCs in mice with ABT-263 reverses IR-induced pul-
monary fibrosis,28 attenuates doxorubicin caused heart dysfunction and
fatigue,29 and reduces atherogenesis onset.30 Thus, senolytic agents
have the potential to be developed into drugs that can extend health-
span or treat age-related diseases and cancer treatment-induced adverse
effects.
Through the targeted screen effort, we identified another senolytic
agent piperlongumine (PL, Fig. 1), a natural dietary product isolated
from Piper longum L. with a variety of bioactivities including antic-
ancer.31,32 PL has been proposed to exert its selective killing of cancer
cells by inhibiting oxidative stress response proteins which are im-
portant for cancer cell survival under elevated reactive oxygen species
(ROS) levels.33 The rationale of selecting PL for senolytic screening is
that compared with NCs, SCs are also under higher oxidative stress,34
thus may share common antioxidant survival pathways with cancer
cells. PL preferentially kill senescent WI-38 cells, induced by IR ex-
posure, oncogene transfection, or extensive replication, over WI-38
NCs.23 PL appears to be very safe, and its maximum tolerated oral dose
in mice is > 1 g/kg, while the effective therapeutic doses for cancer can
be as low as 2.4 mg/kg by oral dosing.33 Thus, PL was considered as a
novel lead for the development of senolytic agents.23 Although a large
variety of PL analogues have been reported as anticancer agents,35–50 it
is not clear how senolytic activity of PL analogues correlate with their
anticancer activity. Previously, we have synthesized several PL analo-
gues, including PL-7 (1), PL-FPh (2), and PL-DI (3) (Fig. 1), that have
Table 1
Senolytic activity of PL and analogues 1–9.
Compd
EC50 (µM)a
EC50 ratio (NCs/SCs)
NCsb
IR-SCsc
PLd
1d
2d
3d
4d
5d
6
20.3
13.0
5.9
8.0
8.9
1.1
0.7
208
46
2.7
9.7
4.9
10.5
2.5
1.5
5.4
2.1
0.8
2.7
3.8
3.7
0.8
1.5
1.5
170
126
10.2
35.7
4.1
7d
8
9
15.3
a
n ≥ 3.
b
c
WI-38 non-senescent cells (NCs).
Ionizing radiation induced WI-38 senescent cells (IR-SCs).
Data from Ref. 23.
d
been reported as potent anticancer agents.23 A good correlation of their
cytotoxicity against cancer cells and WI-38 IR-SCs was observed.
However, the selectivity between NCs and SCs, the more important
component of the senolytic activity, appeared to be unpredictable
(Table 1). Hence, in this report, we carried out an exploration of the
structure-senolytic activity relationships (SAR) of PL analogues. We
sought to investigate the effects of modifying the trimethoxyphenyl ring
and the α,β-unsaturated δ-valerolactam ring of PL.
2. Results and discussion
2.1. Senolytic activities of PL analogues
All PL analogues were evaluated for their senolytic activity against
WI-38 IR-SCs and NCs, by using a flow cytometry–based cell viability
assay as previously described.23 The results of these studies are sum-
marized in Tables 1–3. We first conducted an SAR study to understand
the basic structure required to mediate the senolytic activity of PL.
Previously, we found that compounds 4 and 5 (Fig. 1), in which the
C2–C3 or C7–C8 double bond of PL was replaced by a single bond,
Fig. 1. Structures of piperlongumine (PL) and analogues 1–9.
2

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Table 2
Structures and senolytic activity of PL analogues 10–45.
Compd
R
EC50 (µM)a
NCsb
EC50 ratiod
Compd
R
EC50 (µM)a
NCsb
EC50 ratiod
IR-SCsc
8.3
IR-SCsc
12.9
10
22.9
2.8
28
15.8
1.2
11
12
14.4
3.7
7.9
5.1
1.8
0.7
29
30
12.7
19.9
5.3
6.8
2.4
2.9
13
21.3
7.1
3.0
31
7.3
3.8
1.9
14
15
17.5
13.5
6.4
6.5
2.7
2.1
32
33
9.0
4.8
5.4
4.0
1.7
1.2
16
17
22.4
15.7
6.6
5.4
3.4
2.9
34
35
8.5
8.6
6.1
5.9
1.4
1.5
18
19
20
4.8
7.4
7.7
1.3
3.6
0.7
1.1
36
37
38
12.4
11.4
26.0
9.5
1.3
1.1
2.6
11.0
7.1
10.2
10.0
21
22
9.5
5.2
6.7
1.8
2.3
39
40
24.5
22.1
18.1
17.0
1.4
1.3
15.4
23
24
24.2
40.3
19.0
18.6
1.3
2.2
41
42
34.3
18.1
16.1
6.2
2.1
2.9
25
26
8.7
5.2
6.7
1.7
2.3
43
44
12.7
15.2
3.6
9.5
3.5
1.6
15.4
(continued on next page)
3

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Table 2 (continued)
Compd
R
EC50 (µM)a
NCsb
EC50 ratiod
Compd
R
EC50 (µM)a
NCsb
EC50 ratiod
IR-SCsc
19.0
IR-SCsc
8.5
27
24.2
1.3
45
19.7
2.3
a
n ≥ 3.
b
c
WI-38 non-senescent cells (NCs).
Ionizing radiation induced WI-38 senescent cells (IR-SCs).
EC50 ratio: NCs/IR-SCs.
d
respectively, had a substantial reduction in senolytic activities, in-
dicating that the integrity of the two-electrophile system is important
for the senolytic activity of PL.23 In addition, both compound 6, in
which the methoxy substituents of PL were removed, and 7,23 an ab-
breviated PL analogue with the trimethoxyphenyl ring removed, de-
monstrated retained or improved senolytic activity compared with PL,
indicating that modification on the trimethoxyphenyl group can be
tolerated. Further, retaining the lactam ring structure might be im-
portant for the selectivity of PL, as compounds 8 and 9 (Fig. 1) showed
diminished selectivity for WI-38 IR-SCs over NCs (Table 1).
Based on the initial SAR findings, a series of analogues were syn-
thesized by varying the substituents on the phenyl ring of PL or re-
placing the trimethoxyphenyl ring with a heteroaromatic ring while
holding the rest of the molecule constant (Table 2). Although well
tolerated, the SAR obtained from this part of modifications was sur-
prisingly shallow. Most of the analogues had comparable senolytic ac-
tivity (i.e., potency and selectivity) to PL but no apparent SAR con-
clusions can be drawn. Compound 18, in which the two meta-methoxy
groups on the phenyl ring of PL were removed, exhibited the best se-
nolytic activity in this series with an EC50 of 1.3 µM in killing WI-38 IR-
SCs and 3.6-fold selectivity over NCs. Replacing the 4-methoxy in 18
with another para-substitution group, including hydroxyl group (com-
pound 19), nitro group (compound 26), amino group (compounds
27–29), and halogens (compounds 30–32) resulted in reduced senolytic
activity. One important finding was that the para-position of the phenyl
ring is a suitable site to attach an affinity tag on PL. We synthesized
probes (compounds 42–45, Table 2) with a “clickable” tag, a terminal
alkyne or azide that can react with the corresponding immobilized
azides or alkynes, respectively, through click chemistry.51 One probe,
43, which has similar senolytic activity to PL, was selected for our
target-protein identification study.51
indirect evidence, increasing the steric hindrance of the Michael ac-
ceptor in compound 47 by introducing an isopropyl group (50 and 51,
Table 3) on the exocyclic methylene abolished senolytic activity. In
addition, introduction of a pyridinyl group (52) or a chlorophenyl
group (53 and 54) on the exocyclic methylene also resulted in complete
loss of senolytic activity, likely due to the steric hindrance or con-
jugation effect of these substitutions, which impair the electrophilicity
of the Michael acceptor.
Finally, we incorporated the exocyclic methylene containing lac-
tams into the best modification from Table 2, i.e., compound 18, with
the intent of further improving senolytic activity. However, replace-
ment of the trimethoxyphenyl in 47, 48, and 49 with a para-methox-
yphenyl ring (compounds 55, 56, and 57, respectively) resulted in a
decrease in both potency and selectivity in killing WI-38 IR-SCs, de-
monstrating the unpredictable nature of these types of molecules where
their bioactivities are likely derived from covalent interaction with
biological targets.
2.2. Induction of OXR1 degradation in senescent cells
Consistent with our hypothesis that PL targets the oxidative stress
response proteins in SCs, we recently identified oxidation resistance 1
(OXR1) as a senolytic target of PL.51 PL binds OXR1 directly and in-
duces its degradation through the ubiquitin-proteasome system.51
Compounds 47, 48, and 49 were selected to test their effects on OXR1.
WI-38 IR-SCs and NCs were treated with 5 µM of 47, 48, or 49 for 24 h
and cell lysates were analyzed by Western blotting. As expected, all
three analogues selectively reduced the levels of OXR1 in SCs, in-
dicating that they share a similar mechanism of action with PL (Fig. 2).
2.3. ROS production in senescent cells
As indicated by comparing the senolytic activity of 4 and 5
(Table 1), the C2–C3 double bond is more critical than the C7–C8
double bond for the senolytic activity of PL. It was also reported that
the lactam olefin was important for the cytotoxicity and ROS induction
in cancer cells.35 We therefore speculated that tuning the reactivity of
the Michael acceptor on the α,β-unsaturated δ-valerolactam ring of PL
would produce analogues with improved senolytic activity. In addition
to the ring-expanded ε-caprolactam analogue (1),23 we also synthesized
the ring-contracted γ-butyrolactam analogue (46) (Table 3).38 How-
ever, no improvement in terms of senolytic potency and selectivity was
observed with the change of the lactam ring size. Compounds 47, 48,
and 49 (Table 3), which contain lactams bearing an exocyclic double
bond at C2, were then synthesized. Compared to their endocyclic
double bond–containing counterparts with the same lactam size (i.e.,
PL, 46, and 1, respectively), 47–49 had 4–13-fold increases in senolytic
potency against WI-38 IR-SCs and had moderate increases in senolytic
selectivity for SCs over NCs. Compound 49 had the most potent seno-
lytic activity in this series with an EC50 of 0.67 µM in killing WI-38 IR-
SCs. The increased potency could be attributed to improved accessi-
bility of the exocyclic methylene unit by nucleophiles on the target
protein(s) and/or increased reactivity of the Michael acceptor. As
We evaluated compounds 47, 48, and 49 for their ability to induce
ROS in SCs using dihydrorhodamine 123 (DHR 123) as a probe, which can
passively diffuse across membranes where it is oxidized to green fluor-
escent rhodamine 123 in the presence of ROS. The fluorescence intensity
was measured by flow cytometry and was proportional to the cytosolic
ROS levels. The corresponding endocyclic double bond–containing ana-
logues, PL, 46, and 1, respectively, were also evaluated in this assay for
comparison. WI-38 IR-SCs were treated with various concentrations of
analogues for 1.5 h and the changes in ROS levels were assessed by flow
cytometry after treating with DHR 123 for 30 min. As shown in Fig. 3, PL
caused a moderate increase of the ROS levels in WI-38 IR-SCs. However,
compound 46 did not affect ROS levels in SCs while compound 1 slightly
inhibited ROS production at the same concentration range. Significant
increases in ROS production were observed when IR-SCs were treated with
compounds 47, 48, and 49. Although the ability of PL analogues to elevate
ROS did not correlate with their cell killing effects in SCs and cancer cells,
the largely elevated ROS levels caused by compounds 47–49 could be the
basis of their increased toxicity to SCs. However, further studies are
needed to determine whether induction of oxidative stress alone is suffi-
cient to kill SCs.
4

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Table 3
Structures and senolytic activity of PL analogues 46–57.
Compd
Structure
EC50 (µM)a
NCsb
EC50 ratio (NCs/IR-SCs)
IR-SCsc
PL
20.3
8.0
8.8
8.9
2.5
1.5
1.5
46
12.9
13.0
1
47
12.1
2.1
5.8
48
49
5.6
2.0
2.0
2.8
3.0
0.67
50
> 40
> 40
–
51
52
53
> 40
> 40
> 40
> 40
> 40
> 40
–
–
–
54
> 40
> 40
–
(continued on next page)
5

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Table 3 (continued)
Compd
Structure
EC50 (µM)a
NCsb
EC50 ratio (NCs/IR-SCs)
IR-SCsc
55
56
57
15.8
9.0
3.9
4.3
1.8
1.9
1.8
7.4
7.6
a
n ≥ 3.
b
WI-38 non-senescent cells (NCs).
Ionizing radiation induced WI-38 senescent cells (IR-SCs).
c
2.4. Induction of apoptosis in senescent cells
2.5. Chemistry
PL induces apoptosis in cancer cells33 and SCs.23 To determine if
compound 49, the most potent PL analogue in this series, killed SCs by
inducing apoptosis, we used FITC labeled Annexin V and propidium io-
dide, which stain phosphatidylserine (PS) residues and DNA, respectively,
to detect apoptosis in WI-38 SCs by fluorescence-activated cell sorting
analysis. As shown in Fig. 4A, treatment with 1 μM of 49 increased the
number of Annexin-V-positive cells in SCs by 3.9-fold when compared to
the vehicle group. To further confirm that 49 killed SCs by apoptosis, we
treated WI-38 IR-SCs with the pan-caspase inhibitor Q-VD-OPh (QVD) to
inhibit apoptosis. As shown in Fig. 4B and C, QVD significantly reduced
apoptosis and partially rescued SCs from 49-induced death (Fig. 4B, C),
confirming the apoptotic cell-death mechanism.
PL and its analogues 6, 10–18, 20, 22, 25, 26, 28, and 30–40 with
modifications on the trimethoxyphenyl ring were synthesized via a one-
pot reaction by coupling α,β-unsaturated δ-lactam 61 with appropriate
cinnamic acid derivative 59, obtained commercially or prepared via
Knoevenagel condensation of aldehyde 58 with malonic acid. Acids 59
were pre-activated by forming mixed anhydrides 60 with pivaloyl
chloride (Scheme 1).36 Lactam 61 was prepared using a previously re-
ported procedure.36 Compounds 19 and 21 were synthesized by initial
treatment of 4- (62) and 2-hydroxycinnamic acid (63), respectively,
with Boc2O in the presence of DMAP and triethylamine, followed by
coupling with 61 and removal of the O-Boc group (Scheme 1). Reduc-
tion of the nitro group of 22 by treatment with Fe/NH4Cl in EtOH/H2O
Fig. 2. Compounds PL, 47, 49, and 48 abrogated the
upregulation of OXR1 in WI-38 senescent cells in-
duced by irradiation: (A) representative western blot
analysis of OXR1 and β-actin in WI-38 normal or non-
senescent cells (NCs) and ionizing radiation-induced
WI-38 senescent cells (IR-SCs) 24 h after incubation
with vehicle (Veh) or 5 µM of compound PL, 47, 49,
or 48; (B) quantitative analysis of OXR1 levels in WI-
38 NCs and IR-SCs 24 h after incubation with Veh or
5 µM of compound PL, 47, 49, or 48. Data in bar
graphs are presented as mean
SEM, n = 3,
*p < 0.05 indicates significant difference comparing
with IR-SCs with vehicle treatment.
6

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Fig. 3. Comparison of ROS production in IR-induced WI-38 senescent cells 1.5 h after incubation with compounds 47, 48, and 49 and their corresponding endocyclic
double bond–containing analogues PL, 46, and 1, respectively: (A) compound 47 vs. PL; (B) compound 48 vs 46; (C) compound 49 vs 1.; (D) Comparison of 47, PL,
49, 1, 48, and 46 in one graph. Data are presented as mean
senescent cells.
SEM, n = 3. Compounds concentration used corresponds to their EC50s against IR-induced WI-38
afforded amine 23, which was then N-acetylated to form compound 24
(Scheme 1). Similarly, compounds 27 and 29 were prepared from 26.
Compound 41 was synthesized by de-methylation of PL.35 Compounds
42–45 were prepared from 41 as we reported recently.51
group of 83 and 84 with NaBH4/CaCl2 gave the corresponding alcohols
85 and 86, which underwent dehydration in the presence of N,N-di-
cyclohexylcarbodiimide (DCC) and catalytic amount of CuI to furnish
the target lactams 72 and 73, respectively (Scheme 4B).53 For the
preparation of α-methylene-ε-caprolactam (74), we initially tried the
same method used for the synthesis of 72 and 73. To synthesize the
requisite starting material methyl 2-oxoazepane-3-carboxylate 88, me-
thyl carboxylate was first introduced to the α-position of N-Boc pro-
tected ε-caprolactam (80). However, the resulting N-Boc protected
methyl 2-oxoazepane-3-carboxylate 87 was unstable under various de-
Boc conditions; the carboxylate group was also removed to afford ε-
caprolactam (79) (Scheme 4C, method 1). An alternative route, which
started with reduction of 87 to alcohol 89, also failed at the de-Boc step
(Scheme 4C, method 2). Forming the N-Boc protected α-methylene-ε-
caprolactam 91 by dehydration of 89 with DCC/CuI or through forming
mesylate was also unsuccessful (Scheme 4C, method 3). Finally, 91 was
obtained by treating 87 with paraformaldehyde, 18-crown-6, and
K2CO3 (Scheme 4C, method 4).54 Removal of the Boc protection of 91
under acidic condition resulted in the desired lactam 74.
PL analogues 1 and 46–51 (Table 2) with modifications on the
lactam ring were synthesized in a similar manner to PL via coupling of
3,4,5-trimethoxycinnamic pivalic anhydride (68) with the appropriate
lactam (70–76) (Scheme 2). Similarly, analogues 55–57 were synthe-
sized by coupling 4-methoxycinnamic pivalic anhydride (69) with
lactam 72, 73, and 74, respectively (Scheme 2). The coupling reaction
between 68 and lactam 7138 resulted in a complex mixture, from which
we failed to isolate compound 46 with sufficient purity. Alternatively,
46 was prepared by Grubbs ring closing metathesis reaction of com-
pound 9. The synthesis of 9 was accomplished by a three-step process
started with conversion of 3,4,5-trimethoxycinnamic acid (77) to the
corresponding acid chloride, followed by coupling with allylamine and
then acylation with acryloyl chloride (Scheme 3).
Lactam 70 was synthesized according to previously published
method,52 with minor modifications (Scheme 4). Exocyclic double bond
containing δ- and γ-lactams (72 and 73) were synthesized with ethyl 2-
oxopiperidine-3-carboxylate (83) and ethyl 2-oxopyrrolidine-3-car-
boxylate (84), respectively, as starting materials. Reduction of the ester
Lactams 75 and 76 that bear an isopropyl substituent at the olefin
were prepared via a four-step transformation starting with N-Boc protec-
tion of δ-valerolactam (94) to form compound 95. Aldol condensation of
7

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Fig. 4. Compound 49 induced apoptosis in IR-induced WI-38 senescent cells (IR-SCs): (A) representative flow cytometric plots to measure apoptotic WI-38 IR-SCs at
48 h after treatment with vehicle (Veh), 49 (1 µM), Q-VD-Oph (QVD) (10 µM), or the combination of 49 (1 µM) and QVD (10 µM); (B) quantification of the percentage
of viable and apoptotic WI-38 IR-SCs after 48 h treatment with vehicle, QVD (10 µM), 49 (1 µM), or the combination of 49 (1 µM) and QVD (10 µM); (C) quanti-
fication of the percentage of viable WI-38 IR-SCs 72 h after treatment with vehicle with or without QVD (10 µM) or with 49 (1 µM) with or without QVD (10 µM).
Data in bar graphs are presented as mean
SD, n = 3, *p < 0.05 indicates significant difference.
95 and isobutyraldehyde afforded 96, a mixture of four possible diaster-
eomers. Using a standard dehydration method that involved the formation
of mesylate intermediates, this isomeric mixture was converted into a
mixture of Z- and E-isomers, 97 and 98 (1:5.2 ratio), which were separated
in pure form by silica gel chromatography. The Z/E configurations were
assigned based on NOE (NOE enhancement was observed for Hb when Ha
of compound 97 was irradiated, while little NOE enhancement was
observed for Hd when Hc of compound 98 was irradiated). Subsequently,
removal of the N-Boc group of 97 and 98 with TFA/CH2Cl2 led to the
corresponding lactam 75 and 76, respectively. Utilizing a similar proce-
dure for the preparation of 97/98, analogues 52–54 were synthesized
from compound 4. Under the same dehydration conditions, the aldol
condensation product 99, formed via reacting 4 with an appropriate
aromatic aldehyde, gave almost exclusively the E-isomer (Scheme 5).
Scheme 1. Reagents and conditions: (a) malonic acid, piperidine, pyridine, reflux; (b) trimethylacetyl chloride, Et3N, THF, 0 °C; (c) n-BuLi, THF, −78 °C; (d) Boc2O,
DMAP, Et3N, CH2Cl2, −15 °C; (e) Trifluoroacetic acid, CH2Cl2, rt; (f) Fe, NH4Cl, EtOH/H2O, 40–50 °C; (g) Ac2O, Et3N, CH2Cl2, rt.
8

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
alumina and 4 Å molecular sieve, respectively. All other reagents and
solvents obtained from commercial sources were used without further
purification. If dry and air-free conditions were required, reactions
were performed in oven-dried glassware (130 °C) under a positive
pressure of argon. Flash chromatography was performed using silica gel
(230–400 mesh) as the stationary phase. Reaction progress was mon-
itored by thin layer chromatography (silica-coated glass plates) and
visualized by UV light, and by GC–MS or TLC-MS. NMR spectra were
recorded in CDCl3 at 400 MHz for 1H, and 100 MHz for 13C NMR.
Chemical shifts δ are given in ppm using tetramethylsilane as an in-
ternal standard. Multiplicities of NMR signals are designated as singlet
(s), broad singlet (br s), doublet (d), doublet of doublets (dd), triplet (t),
quartet (q), and multiplet (m). All final compounds for biological
testing were of ≥95% purity as analyzed by LC–MS, performed on a
Waters 2965 system using a Thermo, betabasic-18 column (3 μm,
150 × 4.6 mm) at ambient temperature. Gradient elution was used for
HPLC with a mobile phase of acetonitrile (ACN) and water containing
0.1% formic acid (0 min, ACN/0.1% formic acid in water = 40/60;
5 min, ACN/0.1% formic acid in water = 60/40; 10 min, ACN/0.1%
formic acid in water = 70/30; 13 min, ACN/0.1% formic acid in
water = 90/10; 15 min, ACN/0.1% formic acid in water = 40/60).
Scheme 2. Synthesis of Compounds 1, 46–51, and 55–57.
4.1.2. (E)-1-(3-(3,4,5-Trimethoxyphenyl)acryloyl)-1,5-dihydro-2H-
pyrrol-2-one (46)
Step 1: Preparation of 3,4,5-trimethoxycinnamoyl chloride. To a
solution of 3,4,5-trimethoxycinnamic acid (77) (5 g, 21 mmol) in
CH2Cl2 (100 mL) was added one drop of DMF and then oxalyl chloride
(7.2 mL, 84 mmol) dropwise at 0 °C. The resulting mixture was allowed
to stir at room temperature overnight and condensed under vacuum to
give the chloride as a yellow solid which was used in the next step
without further purification. Step 2: Preparation of (E)-N-allyl-3-(3,4,5-
trimethoxyphenyl) acrylamide (78). To a stirring solution of the above
chloride (0.50 g, 1.95 mmol) in Et2O (5 mL) was added allylamine
(1.10 g, 19.50 mmol) at 0 °C. The resulting mixture was stirred at 0 °C
for 10 min. The precipitated solid was filtered and washed with hexanes
to give 78 (0.38 g, 70%) as a white solid: 1H NMR (400 MHz, CDCl3) δ
7.53 (d, J = 14.6 Hz, 1H), 6.70 (s, 2H), 6.37 (d, J = 14.8 Hz, 1H), 6.05
(s, 1H), 5.87 (s, 1H), 5.30–5.00 (m, 2H), 4.00 (s, 2H), 3.83 (s, 9H); 13C
NMR (100 MHz, CDCl3) δ 165.89, 153.44, 141.19, 139.60, 134.17,
130.48, 120.02, 116.63, 105.01, 61.02, 56.18, 42.26. MS (EI) m/z
277.4 (M+). Step 3: Preparation of (E)-N-acryloyl-N-allyl-3-(3,4,5-tri-
methoxyphenyl)acrylamide (9). A solution of 78 (0.15 g, 0.54 mmol) in
THF (1 mL) was added dropwise to a stirred suspension of NaH (60% in
mineral oil, 43 mg, 1.10 mmol) in THF (3 mL) at room temperature. The
mixture was stirred for 1 h at room temperature, cooled to 0 °C, and
acrylyl chloride (53 μL, 0.65 mmol) was added slowly. The reaction was
allowed to stir at 40 °C for 2 h before quenched with aqueous NH4Cl
solution. The resulting mixture was extracted with EtOAc (10 mL × 3),
and the combined organic phases were washed with brine, dried over
Na2SO4, filtered, and concentrated under reduced pressure. The crude
product was purified by silica gel column chromatography (hexanes/
EtOAc 1:1) to afford 9 (94 mg, yield 51%) as a white solid: 1H NMR
(400 MHz, CDCl3) δ 7.69 (d, J = 15.4 Hz, 1H), 6.96 (d, J = 15.4 Hz,
1H), 6.79–6.68 (m, 3H), 6.48 (d, J = 16.5 Hz, 1H), 5.92 (ddd, J = 16.5,
10.3, 5.0 Hz, 1H), 5.83 (d, J = 10.3 Hz, 1H), 5.28–5.16 (m, 2H),
4.56–4.43 (m, 2H), 3.89 (m 9H); 13C NMR (100 MHz, CDCl3) δ 168.93,
168.79, 153.58, 145.35, 140.56, 133.09, 130.72, 130.24, 130.20,
119.66, 116.92, 105.68, 61.13, 56.34, 46.76. MS (EI) m/z 331.4 (M+).
Step 4: Ring-closing metathesis of 9. To a solution of 9 (15 mg,
0.045 mmol) in CH2Cl2 (8 mL) under N2 protection was added Grubbs
2nd generation catalyst (2 mg, 0.0022 mmol). The resulting mixture
was heated under reflux for 4 h. Upon completion, the reaction mixture
was cooled to room temperature, and solvent was removed under va-
cuum. The crude product was purified by silica gel column chromato-
graphy (hexanes/EtOAc 2:1) to afford 46 (8 mg, yield 58%) as a col-
orless oil: 1H NMR (400 MHz, CDCl3) δ 7.96 (d, J = 15.7 Hz, 1H), 7.83
Scheme 3. Reagents and conditions: (a) oxalyl chloride, DMF (cat.), CH2Cl2, rt;
(b) allylamine, Et2O, rt; (c) acryloyl chloride, NaH, THF, 0 °C; (d) Grubbs 2nd
generation catalyst, CH2Cl2, reflux.
3. Conclusion
In this report, we explored a series of structural modifications on the
trimethoxyphenyl and α,β-unsaturated δ-valerolactam moieties of PL.
These PL analogues were evaluated for their senolytic activity using WI-38
IR-SCs and NCs. One important finding from our initial SAR studies is that
structural modifications on the trimethoxyphenyl ring are well tolerated,
which allows us to design probe molecules with a “clickable” tag on the
phenyl ring to pull down target proteins in live cells, as well as to in-
troduce structural moieties on the phenyl ring to potentially optimize the
physiochemical properties of PL. Our SAR studies also reveal that the
Michael acceptor on the lactam ring is critical for the senolytic activity.
While changing the lactam ring size has minimal effect on the toxicity of
PL analogues to SCs, introducing an exocyclic methylene at C2 of the
lactam ring offers PL analogues, 47–49, with increased senolytic activity.
These α-methylene containing PL analogues strongly induce ROS pro-
duction in SCs, which could be partially attributed to their ability of
lowering the protein levels of OXR1, a potential target of PL and an im-
portant oxidative stress response protein that regulates the expression of a
variety of antioxidant enzymes. In addition, lead compound 49, which has
a 12-fold increase in potency when compared with PL, selectively induces
apoptosis in SCs. Further analogue development and in vitro and in vivo
characterization of novel analogues are in progress.
4. Experimental
4.1. Chemistry
4.1.1. General methods
THF, CH2Cl2, toluene, and DMF were obtained via a solvent pur-
ification system by filtering through two columns packed with activated
9

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
Scheme 4. Reagents and conditions: (a) (Boc)2O, Et3N, DMAP, CH2Cl2; (b) LiHMDS, PhSeCl, THF, -78 °C; (c) 30% H2O2, THF, 0 °C to rt; (d) TFA, CH2Cl2, 0 °C; (e)
NaBH4, CaCl2, EtOH, 0 °C to rt; (f) DCC, CuI, toluene, reflux; (g) 1.0 M LiHMDS, methyl chloroformate, THF, −78 °C; (h) 1) CH3SO2Cl, Et3N, toluene, 0 °C; 2) Et3N,
toluene, reflux; (i) paraformaldehyde, 18-crown-6, K2CO3, toluene, reflux; (j) LiHMDS, isobutyraldehyde, THF, −78 °C.
(d, J = 15.7 Hz, 1H), 7.36 (d, J = 6.1 Hz, 1H), 6.87 (s, 2H), 6.22 (d,
J = 6.1 Hz, 1H), 4.55 (s, 2H), 3.92 (s, 6H), 3.89 (s, 3H); 13C NMR
(100 MHz, CDCl3) δ 170.38, 165.32, 153.54, 147.01, 146.29, 140.50,
130.52, 127.99, 117.95, 105.89, 61.14, 56.35, 51.27. MS (ESI) m/z
304.2 [M+H]+
.
4.1.3. 3-(Hydroxymethyl)pyrrolidin-2-one (86)
To a suspension of NaBH4 (151 mg, 3.98 mmol) and CaCl2 (585 mg,
3.98 mmol) in EtOH (5 mL) was added ethyl 2-oxopyrrolidine-3-car-
boxylate (84) (250 mg, 1.59 mmol) at 0 °C. The resulting mixture was
stirred at room temperature overnight before quenched with HCl so-
lution (1.0 M). The resulting mixture was extracted with CH2Cl2/MeOH
(10/1) 10–20 times. The combined organic phases were washed with
brine, dried over Na2SO4, filtered, and concentrated under reduced
Scheme 5. Reagents and conditions: (a) LiHMDS, THF, −78 °C; (b) 1)
CH3SO2Cl, Et3N, toluene, 0 °C; 2) Et3N, toluene, reflux.
10

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
pressure. The crude product was purified by silica gel column chro-
matography (EtOAc/MeOH 10:1) to afford 86 (90 mg, yield 50%) as a
white solid: 1H NMR (400 MHz, CDCl3) δ 6.05 (br s, 1H), 3.94–3.84 (m,
1H), 3.80–3.70 (m, 1H), 3.43–3.35 (m, 2H), 2.68–2.55 (m, 1H),
4.1.8. (Z)-(97) and (E)-t-Butyl 3-(2-methylpropylidene)-2-oxopiperidine-
1-carboxylate (98)
Step 1: To a solution of t-butyl 2-oxopiperidine-1-carboxylate (95)
(800 mg, 4 mmol) in THF (15 mL) was added LiHDMS (1.60 M, 2.76 mL,
4.4 mmol) dropwise at −78 °C. After stirring at the same temperature
for 1 h, isobutyraldehyde (0.44 mL, 4.8 mmol) was added dropwise. The
resulting mixture was allowed to stir for an additional hour before
quenched by slowly adding saturated aqueous NH4Cl solution. The
mixture was extracted with EtOAc. The organic phase was washed with
brine, and dried over anhydrous Na2SO4, filtered, and condensed under
reduced pressure. The crude product was purified by silica gel column
chromatography (hexanes/EtOAc 3:1) to afford t-butyl 3-(1-hydroxy-2-
methylpropyl)-2-oxopiperidine-1-carboxylate (96) as a mixture of four
diasteromers, which was used directly in the next step. Step 2: To a
solution of 96 (470 mg, 1.73 mmol) and Et3N (0.725 mL, 5.19 mmol) in
toluene (10 mL) was added methanesulfonyl chloride (0.2 mL,
2.60 mmol) dropwise at 0 °C. After stirred at 0 °C for 1 h, additional
Et3N (0.725 mL) was added. The resulting mixture was stirred under
reflux for 6 h. Upon cooling to room temperature, the reaction was
quenched with water, and extracted with EtOAc (10 mL × 3). The
combined organic phases were washed with brine, dried over anhy-
drous Na2SO4, filtered, and condensed under reduced pressure. The
crude product was purified by silica gel column chromatography
(hexanes/EtOAc 10:1) to afford 97 (Z-isomer, 25 mg, 5.7%) and 98 (E-
isomer, 130 mg, 30%). Compound 97: 1H NMR (400 MHz, CDCl3) δ
5.63 (dt, J = 9.7, 1.5 Hz, 1H), 3.69–3.58 (m, 2H), 3.49–3.30 (m, 1H),
2.41 (td, J = 6.8, 1.5 Hz, 2H), 1.84 (td, J = 12.4, 6.6 Hz, 2H), 1.53 (s,
9H), 0.99 (s, 3H), 0.98 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 166.41,
153.02, 151.06, 127.65, 82.71, 45.60, 30.62, 28.22, 28.07, 22.97,
22.83. MS (APCI) m/z 254.2 [M+H]+. Compound 98: 1H NMR
(400 MHz, CDCl3) δ 6.72 (dt, J = 10.0, 2.0 Hz, 1H), 3.67–3.57 (m, 2H),
2.56–2.46 (m, 1H), 2.44–2.39 (m, 2H), 1.84–1.76 (m, 2H), 1.48 (s, 9H),
0.96 (s, 3H), 0.94 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 165.67, 153.31,
149.05, 127.70, 82.67, 45.86, 28.05, 27.56, 24.12, 22.35, 21.75. MS
2.30–2.19 (m, 1H), 2.02–1.88 (m, 1H). MS (APCI) m/z 116 [M+H]+
.
4.1.4. 3-Methylenepyrrolidin-2-one (73)
To a solution of 86 (16 mg, 0.139 mmol) in toluene was added CuI
(2.6 mg, 0.0139 mmol) and DCC (36 mg, 0.174 mmol). The resulting
mixture was stirred under reflux for 3 h. After cooling down to room
temperature, water was added, and the mixture was extracted with
CH2Cl2 (×20). The combined organic phases were washed with brine,
dried over Na2SO4, and concentrated under reduced pressure. The
crude product was purified by silica gel column chromatography
(CH2Cl2/MeOH 20:1) to afford 73 (11 mg, yield 82%): 1H NMR
(400 MHz, CDCl3) δ 6.20 (br s, 1H), 6.01 (s, 1H), 5.38 (s, 1H), 3.44 (t,
J = 6.4, 2H), 2.91 (m, 2H). MS (APCI) m/z 98.2 [M+H]+
.
4.1.5. 1-(t-Butyl) 3-methyl 2-oxoazepane-1,3-dicarboxylate (87)
To a solution of t-butyl 2-oxoazepane-1-carboxylate (80) (2.45 g,
11.5 mmol) in THF (50 mL) was added LiHMDS (1.0 M in THF, 13.2 mL,
13.2 mmol) dropwise at −78 °C. The resulting mixture was stirred at
the same temperature for 45 min, and methyl chloroformate (1.0 mL,
12.6 mmol) was added. After stirring for an additional hour, the reac-
tion was quenched with saturated aqueous NH4Cl solution, extracted
with EtOAc (50 mL × 3). The combined organic phases were washed
with brine, dried over anhydrous Na2SO4, filtered, and condensed
under reduced pressure. The crude product was purified by silica gel
column chromatography (hexanes/EtOAc 3:1) to afford 87 (2.7 g, yield
87%): 1H NMR (400 MHz, CDCl3): 1H NMR (400 MHz, CDCl3) δ 5.36 (s,
1H), 3.81 (s, 3H), 3.58 (s, 2H), 2.12 (dd, J = 11.4, 6.6 Hz, 2H), 1.75 (dt,
J = 11.3, 5.8 Hz, 2H), 1.60–1.49 (m, 2H), 1.46 (s, 9H); 13C NMR
(100 MHz, CDCl3) δ 154.06, 153.14, 145.54, 111.34, 81.05, 77.48,
77.16, 76.84, 55.10, 47.13, 29.39, 28.23, 24.52, 24.02. MS (APCI) m/z
(APCI) m/z 254.2 [M+H]+
.
272.1 [M+H]+
.
4.1.9. (Z)-3-(2-Methylpropylidene)piperidin-2-one (75)
To a solution of 97 (130 mg, 0.514 mmol) in CH2Cl2 (8 mL) at 0 °C
was added TFA (0.79 mL, 10.3 mmol). The resulting mixture was stirred
at 0 °C for 2 h. Afterwards, the reaction was quenched with saturated
aqueous NaHCO3 and extracted with CH2Cl2 (×3). The combined or-
ganic phases were washed with brine, dried over anhydrous Na2SO4,
filtered, and condensed under reduced pressure. The crude product was
purified by silica gel column chromatography (hexanes/EtOAc 3:1) to
afford 75 (70 mg, yield 90%): 1H NMR (400 MHz, CDCl3) δ 5.66 (s, 1H),
5.57 (dd, J = 9.6, 0.7 Hz, 1H), 3.73 (ddt, J = 13.3, 9.6, 6.6 Hz, 1H),
3.34–3.19 (m, 2H), 2.51–2.32 (m, 2H), 1.89–1.78 (m, 2H), 0.98 (s, 3H),
0.97 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 167.03, 149.93, 125.65,
4.1.6. t-Butyl 3-methylene-2-oxoazepane-1-carboxylate (91)
To a suspension of 87 (90 mg, 0.33 mmol), 18-crown-6 (26 mg,
0.1 mmol), and K2CO3 (137 mg, 1 mmol) in toluene (15 mL) was added
paraformaldehyde (348 mg, 11.55 mmol). The resulting mixture was
stirred under reflux for 5 h. After cooling down to room temperature,
water was added, and the mixture was extracted with EtOAc
(15 mL × 3). The combined organic phases were washed with brine,
dried over anhydrous Na2SO4, filtered and condensed under reduced
pressure. The crude product was purified by silica gel column chro-
matography (hexanes/EtOAc 5:1) to afford 91 (30 mg, yield 41%) as a
white solid: 1H NMR (400 MHz, CDCl3) δ 5.76 (s, 1H), 5.38 (s, 1H),
3.70–3.63 (m, 2H), 2.48–2.36 (m, 2H), 1.84–1.65 (m, 4H), 1.52 (s, 9H);
13C NMR (100 MHz, CDCl3) δ 172.64, 153.05, 147.29, 123.00, 82.74,
42.51, 32.40, 27.63, 23.61, 22.96. MS (APCI) m/z 154.3 [M+H]+
.
4.1.10. (E)-3-(2-Methylpropylidene)piperidin-2-one (76)
46.07, 32.59, 29.23, 28.22, 28.17; MS (APCI) m/z 226 [M+H]+
.
By utilizing the same procedure to that described for the prepara-
tion of compound 75, compound 76 (13.5 mg, yield 90%) was prepared
from 98: 1H NMR (400 MHz, CDCl3) δ 7.03 (s, 1H), 6.64 (dt, J = 9.9,
1.8 Hz, 1H), 3.45–3.21 (m, 2H), 2.64–2.51 (m, 1H), 2.50–2.42 (m, 2H),
1.87–1.75 (m, 2H), 1.00 (s, 3H), 0.98 (s, 3H); 13C NMR (100 MHz,
CDCl3) δ 167.32, 145.49, 126.48, 42.01, 27.07, 24.39, 22.92, 22.09. MS
4.1.7. 3-Methyleneazepan-2-one (74)
To a solution of 91 (30 mg, 0.133 mmol) in CH2Cl2 (2 mL) at 0 °C
was added TFA (0.2 mL, 2.66 mmol). The resulting mixture was stirred
at 0 °C for 2 h. Saturated aqueous NaHCO3 was added to adjust the pH
to > 7, and the mixture was extracted with CH2Cl2 (10 mL × 3). The
combined organic phases were washed with brine, dried over anhy-
drous Na2SO4, filtered, and condensed under reduced pressure. The
crude product was purified by silica gel column chromatography
(hexanes/EtOAc 2:1) to afford 74 (11 mg, yield 66%) as a white solid:
1H NMR (400 MHz, CDCl3) δ 6.60 (s, 1H), 5.58 (s, 1H), 5.26 (s, 1H),
3.17 (dd, J = 9.9, 5.2 Hz, 2H), 2.42–2.25 (m, 2H), 1.81–1.58 (m, 4H);
13C NMR (100 MHz, CDCl3) δ 176.03, 146.42, 120.37, 42.82, 32.77,
(APCI) m/z 254.3 [M+H]+
.
4.1.11. General procedure for the preparation of 47–51 and 55–57
Step 1: Preparation of t-butyl-carbonic (E)-3-(3,4,5-trimethox-
yphenyl)acrylic anhydride (68) and t-butyl-carbonic (E)-3-(4-methox-
yphenyl)acrylic anhydride (69). To a solution of 3,4,5-trimethox-
ycinnamic acid or 4-methoxycinnamic acid (1.0 eq.) in THF (∼0.3 M)
was added Et3N (1.3 eq.) and then trimethyl acetic chloride (1.2 eq.)
dropwise at 0 °C. The resulting mixture was stirred at 0 °C for 30 min
29.69, 28.98. MS (APCI) m/z 126.2 [M+H]+
.
11

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
and filtered. The filtrate was used for the next step without further
purification.
55%) as a δ 7.71 (d,
white solid: 1H NMR (400 MHz, CDCl3)
J = 15.6 Hz, 1H), 7.54 (d, J = 8.5 Hz, 2H), 7.46 (d, J = 15.6 Hz, 1H),
6.89 (d, J = 8.5 Hz, 2H), 6.40 (s, 1H), 5.51 (s, 1H), 3.97–3.76 (m, 5H),
2.65 (t, J = 6.3 Hz, 2H), 1.97–1.84 (m, 2H); 13C NMR (100 MHz,
CDCl3) δ 170.36, 166.98, 161.36, 143.53, 138.86, 130.15, 128.06,
125.73, 119.91, 114.32, 55.50, 45.18, 29.46, 22.62; MS (EI) m/z 271.1
(M+); TR = 5.34 min.
Step 2: To a solution of lactam 72, 73, 74, 75, or 76 (1.0 eq.) in THF
(∼0.3 M) was added n-BuLi (1.6 M in THF, 1.2 eq.) dropwise at −78 °C.
After stirred at −78 °C for 45 min, the filtrate from last step (1.0 eq.)
was added. The resulting mixture was allowed to stir at −78 °C until
completion of the reaction (monitored by TLC). The reaction was then
quenched with saturated aqueous NH4Cl and extracted with EtOAc
(3×). The combined organic phases were washed with brine, dried over
anhydrous Na2SO4, filtered, and concentrated under reduced pressure.
The crude product was purified by silica gel column chromatography.
4.1.11.7. (E)-1-(3-(4-Methoxyphenyl)acryloyl)-3-methylenepyrrolidin-2-
one (56). Purified with hexanes/EtOAc (2:1) to afford 56 (47 mg, yield
68%) as
a δ 7.95 (d,
white solid: 1H NMR (400 MHz, CDCl3)
J = 15.7 Hz, 1H), 7.83 (d, J = 15.7 Hz, 1H), 7.59 (d, J = 8.7 Hz, 2H),
6.90 (d, J = 8.7 Hz, 2H), 6.25 (t, J = 2.6 Hz, 1H), 5.57 (t, J = 2.1 Hz,
4.1.11.1. (E)-3-Methylene-1-(3-(3,4,5-trimethoxyphenyl)acryloyl)
piperidin-2-one (47). Purified with hexanes/EtOAc (2.5:1) to afford 47
(53 mg, yield 35.8%) as a white solid: 1H NMR (400 MHz, CDCl3) δ 7.63
(d, J = 15.5 Hz, 1H), 7.45 (d, J = 15.5 Hz, 1H), 6.79 (s, 2H), 6.39 (d,
J = 1.6 Hz, 1H), 5.51 (dd, J = 3.3, 1.6 Hz, 1H), 3.88 (s, 6H), 3.86 (s,
3H), 3.84 (d, J = 6.0 Hz, 2H), 2.71–2.56 (m, 2H), 1.93 (dt, J = 15.4,
6.2 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 170.05, 166.96, 153.44,
143.64, 140.05, 138.78, 130.83, 125.86, 121.58, 105.61, 61.07, 56.31,
45.15, 29.36, 22.55; MS (ESI) m/z 332.1 [M+H]+; TR = 4.40 min.
1H), 3.89 (t, J = 7.3 Hz, 2H), 3.84 (s, 3H), 2.81 (t, J = 7.3 Hz, 2H); 13
C
NMR (100 MHz, CDCl3) δ 167.74, 167.24, 161.67, 145.53, 140.28,
130.39, 127.81, 120.69, 116.69, 114.37, 55.49, 42.47, 23.00; MS (ESI)
[M+H]+ m/z 258; TR = 4.54 min.
4.1.11.8. (E)-1-(3-(4-Methoxyphenyl)acryloyl)-3-methyleneazepan-2-one
(57). Purified with hexanes/EtOAc (3:1) to afford 57 (49 mg, yield
72%) as
a δ 7.75 (d,
white solid: 1H NMR (400 MHz, CDCl3)
J = 15.5 Hz, 1H), 7.52 (d, J = 8.6 Hz, 2H), 7.30–7.21 (m, 1H), 6.89
(d, J = 8.7 Hz, 2H), 5.84 (s, 1H), 5.48 (s, 1H), 3.97–3.87 (m, 2H), 3.83
(s, 3H), 2.55–2.39 (m, 2H), 1.91–1.70 (m, 4H); 13C NMR (100 MHz,
CDCl3) δ 174.83, 167.96, 161.36, 147.12, 143.99, 130.09, 127.94,
123.78, 118.23, 114.33, 55.48, 44.09, 32.79, 29.39, 28.09; MS (EI) m/z
285.1 (M+); TR = 6.02 min.
4.1.11.2. (E)-3-Methylene-1-(3-(3,4,5-trimethoxyphenyl)acryloyl)
pyrrolidin-2-one (48). Purified with hexanes/EtOAc (2:1) to afford 48
(20 mg, yield 44%) as a white solid: 1H NMR (400 MHz, CDCl3) δ 7.97
(d, J = 15.7 Hz, 1H), 7.79 (d, J = 15.7 Hz, 1H), 6.85 (s, 2H), 6.27 (t,
J = 2.8 Hz, 1H), 5.59 (t, J = 2.4 Hz, 1H), 3.85–3.95 (m, 11H),
2.92–2.67 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 167.89, 167.00,
153.53, 145.92, 140.46, 140.22, 130.56, 120.97, 118.34, 105.82,
4.2. Senolytic activity assay on normal and senescent WI-38 cells
61.13, 56.34, 42.56, 23.08; MS (ESI) m/z 318.5 [M+H]+
;
TR = 3.95 min.
Human WI-38 fibroblasts (WI-38, catalog No. CCL-75, American
Type Culture Collection, Manassas, VA) were cultured in a complete
cell culture medium (CM) (Dulbecco’s Modified Eagle Medium) sup-
plemented with 10% Fetal Bovine Serum (FBS, catalog No. 16000044,
Thermo Fisher Scientific, Waltham, MA), 100 U/mL penicillin and
100 µg/mL streptomycin (purchased from Atlanta Biologicals, Norcross,
GA) in a 37 °C, humidified incubator with 5% CO2. Low passaged WI-38
(< 25 passages) cells were used as controls or for the induction of se-
nescence. Senescent WI-38 cells were prepared from normal WI-38 cells
by exposing low-passage WI-38 cells to 15 Gy of IR at a dose rate of
1.080 Gy/min in a J.L. Shepherd Model Mark I 137Cesium γ-irradiator
(J.L. Shepherd, Glendale, CA). Cell viability was measured by flow
cytometry as previously described.20,23 Dose-response curves were
generated, and the half-maximal effective concentrations (EC50 values)
were calculated with GraphPad Prism 6 software.
4.1.11.3. (E)-3-Methylene-1-(3-(3,4,5-trimethoxyphenyl)acryloyl)azepan-
2-one (49). Purified with hexanes/EtOAc (3:1) to afford 49 (22 mg,
yield 50%) as a white solid: 1H NMR (400 MHz, CDCl3) δ 7.70 (d,
J = 15.5 Hz, 1H), 7.28 (d, J = 15.5 Hz, 1H), 6.79 (s, 2H), 5.85 (d,
J = 1.1 Hz, 1H), 5.50 (d, J = 1.1 Hz, 1H), 3.92–3.88 (m, 11H),
2.56–2.44 (m, 2H), 1.89–1.71 (m, 4H); 13C NMR (100 MHz, CDCl3) δ
174.94, 167.68, 153.50, 147.13, 144.33, 140.10, 130.71, 123.91,
119.84, 105.56, 61.11, 56.33, 44.18, 32.83, 29.44, 28.12; MS (EI) m/
z 345.2 (M+); TR = 5.00 min.
4.1.11.4. (Z)-3-(2-Methylpropylidene)-1-((E)-3-(3,4,5-trimethoxyphenyl)
acryloyl)piperidin-2-one (50). Purified with hexanes/EtOAc (5:1) to
afford 50 (13 mg, yield 53%) as a colorless oil: 1H NMR (400 MHz,
CDCl3) δ 7.64 (d, J = 15.5 Hz, 1H), 7.39 (d, J = 15.5 Hz, 1H), 6.80 (s,
2H), 5.78 (dd, J = 9.7, 1.4 Hz, 1H), 3.90 (s, 6H), 3.88 (s, 3H),
3.83–3.77 (m, 2H), 3.42 (ddt, J = 13.2, 9.7, 6.7 Hz, 1H), 2.58–2.44
(m, 2H), 1.99–1.85 (m, 2H), 1.04 (d, J = 6.6 Hz, 6H); 13C NMR
(100 MHz, CDCl3) δ 169.62, 168.76, 153.45, 152.29, 143.31, 140.02,
130.96, 127.62, 121.67, 105.68, 61.12, 56.35, 43.53, 30.37, 28.27,
22.84, 22.62; MS (EI) m/z 373.3 (M+); TR = 8.24 min.
4.3. ROS assay
Both normal WI-38 cells and senescent WI-38 cells were cultured
overnight in a 37 °C, humidified incubator with 5% CO2. Cells were
then treated with various concentrations of PL or PL analogues for
1.5 h. The medium was then aspirated and replaced with pre-warmed
new culture medium containing 1 µM dihydrorhodamine 123 (DHR
123, catalog No. D632, Thermo Fisher Scientific). After incubating at
37 °C for 30 min, the culture medium was removed and the cells were
washed with PBS and lifted with trypsin. Culture medium was then
added to stop the reaction and cell suspensions were transferred to
plastic tubes. Mean fluorescence intensity (MFI) of DHR 123 was de-
termined with a BD LSR II flow cytometer (BD Biosciences, San Jose,
CA).
4.1.11.5. (E)-3-(2-Methylpropylidene)-1-((E)-3-(3,4,5-trimethoxyphenyl)
acryloyl)piperidin-2-one (51). Purified with hexanes/EtOAc (5:1) to
afford 51 (12 mg, yield 50%) as a colorless oil: 1H NMR (400 MHz,
CDCl3) δ 7.63 (d, J = 15.5 Hz, 1H), 7.50 (d, J = 15.5 Hz, 1H), 6.86 (d,
J = 10.0 Hz, 1H), 6.80 (s, 2H), 3.90 (s, 6H), 3.88 (s, 3H), 3.86–3.81 (m,
2H), 2.70–2.59 (m, 1H), 2.55 (dd, J = 9.1, 4.0 Hz, 2H), 1.98–1.82 (m,
2H), 1.07 (d, J = 6.6 Hz, 6H); 13C NMR (100 MHz, CDCl3) δ 170.05,
168.01, 153.46, 150.42, 143.36, 139.98, 131.01, 127.97, 121.91,
105.60, 61.11, 56.35, 44.22, 27.96, 24.30, 22.39, 21.88; MS (EI) m/z
373.2 (M+); TR = 7.79 min.
4.4. Apoptosis assay
WI-38 cells were first treated with 10 µM QVD-Oph (QVD, catalog
No. A1901 APExBIO, Houston, TX) or vehicle for 4 h, and then incubate
with 1 μM of compound 49 for 48 h. The cells were harvested and
4.1.11.6. (E)-1-(3-(4-Methoxyphenyl)acryloyl)-3-methylenepiperidin-2-
one (55). Purified with hexanes/EtOAc (2:1) to afford 55 (41 mg, yield
12

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
washed twice with Annexin V binding buffer and then stained with
Alexa Fluor 647-Annexin V (1:50, catalog No. 640912, BioLegend, San
Diego, CA) and propidium iodide (PI, 10 µg/ml, catalog No. P4170,
Sigma-Aldrich) following the manufacturer’s instructions (Biotium,
Hayward, CA). All of the stained cells were analyzed with the BD LSR II
flow cytometer.
disease: from mechanisms to therapy. Nat Med. 2015;21:1424–1435.
12. Munoz-Espin D, Serrano M. Cellular senescence: from physiology to pathology. Nat
Rev Mol Cell Biol. 2014;15:482–496.
13. Augert A, Bernard D. Wang Z, Inuzuka H, eds. Immunosenescence and Senescence
Immunosurveillance: One of the Possible Links Explaining the Cancer Incidence in Ageing
Population, Senescence and Senescence-related Disorders. InTech; 2013:87–111.
14. Rodier F, Campisi J. Four faces of cellular senescence. J Cell Biol. 2011;192:547–556.
15. Tchkonia T, Zhu Y, Deursen JV, et al. Cellular senescence and the senescent secretory
phenotype: therapeutic opportunities. J Clin Invest. 2013;123:966–972.
16. Salama R, Sadaie M, Hoare M, et al. Cellular senescence and its effector programs.
Genes Dev. 2014;28:99–114.
4.5. OXR1 degradation assay using western blot
17. Baker DJ, Wijshake T, Tchkonia T, et al. Clearance of p16Ink4a-positive senescent
cells delays ageing-associated disorders. Nature. 2011;479:232–236.
18. Baker DJ, Childs BG, Durik M, et al. Naturally occurring p16(Ink4a)-positive cells
shorten healthy lifespan. Nature. 2016;530:184–189.
Cells were treated with 5 μM PL or PL analogues for 24 h and then
lysed in RIPA buffer with EDTA and EGTA for protein extraction (cat-
alog No. BP-115DG, Boston BioProducts, Ashland, MA), supplemented
with 1% Phosphatase Inhibitor Cocktail 3 (catalog No. P0044, Sigma-
Aldrich) and 1% Protease Inhibitor Cocktail (catalog No. P8340, Sigma-
Aldrich). Proteins were quantified with BCA assay kit (catalog No.
23225, pierce, Thermofisher Scientific, Waltham, MA) and an equal
amount of protein (30 µg/lane) from each cell extract was resolved on a
4–20% Mini-PROTEAN® TGX™ Precast Protein Gels (Bio-rad, Hercules,
CA). Proteins were transferred to a NOVEX PVDF membrane (catalog
No. LC2002, Life Technologies). The membranes were blocked with 5%
nonfat milk in TBST blocking buffer and probed with primary anti-
bodies at a predetermined optimal concentration overnight at 4 °C, in-
cluded OXR1 (catalog No. A302-035A, Bethyl Laboratories Inc,
Montgomery, TX), β-actin (catalog No. SC-1615, Santa Cruz
Biotechnology, Dallas, TX). The secondary antibodies (anti-rabbit or
anti-goat) (CST, USA) were conjugated to horseradish peroxidase.
Signals were detected using the ECL system (catalog No. WBKLS0100,
EMD Millipore, Newmarket, Suffolk, UK) and recorded with auto-
radiography (Pierce Biotech, Rockford, IL, USA).
19. Zhu Y, Tchkonia T, Pirtskhalava T, et al. The Achilles' heel of senescent cells: from
transcriptome to senolytic drugs. Aging Cell. 2015;4:644–658.
20. Chang J, Wang Y, Shao L, et al. Clearance of senescent cells by ABT263 rejuvenates
aged hematopoietic stem cells in mice. Nat Med. 2016;22:78–83.
21. Zhu Y, Tchkonia T, Fuhrmann-Stroissnigg H, et al. Identification of a novel senolytic
agent, navitoclax, targeting the Bcl-2 family of anti-apoptotic factors. Aging Cell.
2016;15:428–435.
22. Yosef R, Pilpel N, Tokarsky-Amiel R, et al. Directed elimination of senescent cells by
inhibition of BCL-W and BCL-XL. Nat Commun. 2016;7:11190.
23. Wang Y, Chang J, Liu X, et al. Discovery of piperlongumine as a potential novel lead
for the development of senolytic agents. Aging. 2016;8:2915–2926.
24. Baar MP, Brandt RM, Putavet DA, et al. Targeted apoptosis of senescent cells restores
tissue homeostasis in response to chemotoxicity and aging. Cell. 2017;169:132–147.
25. Zhu Y, Doornebal EJ, Pirtskhalava T, et al. New agents that target senescent cells: the
flavone, fisetin, and the BCL-XL inhibitors, A1331852 and A1155463. Aging.
2017;9:955–963.
26. Fuhrmann-Stroissnigg H, Ling YY, Zhao J, et al. Identification of HSP90 inhibitors as
a novel class of senolytics. Nat Commun. 2017;8:422.
27. Rudin CM, Hann CL, Garon EB, et al. Phase II study of single-agent navitoclax (ABT-
263) and biomarker correlates in patients with relapsed small cell lung cancer. Clin
Cancer Res. 2012;18:3163–3169.
28. Pan J, Li D, Xu Y, et al. Inhibition of Bcl-2/xl with ABT-263 selectively kills senescent
yype II pneumocytes and reverses persistent pulmonary fibrosis induced by ionizing
radiation in mice. Int J Radiat Oncol Biol Phys. 2017;99:353–361.
29. Demaria M, O'Leary MN, Chang J, et al. Cellular senescence promotes adverse effects
of chemotherapy and cancer relapse. Cancer Discov. 2017;7:165–176.
30. Childs BG, Baker DJ, Wijshake T, et al. Senescent intimal foam cells are deleterious at
all stages of atherosclerosis. Science. 2016;354:472–477.
31. Bezerra DP, Pessoa C, de Moraes MO, et al. Overview of the therapeutic potential of
piplartine (piperlongumine). Eur J Pharm Sci. 2013;48:453–463.
32. Prasad S, Tyagi AK. Historical spice as a future drug: therapeutic potential of pi-
perlongumine. Curr Pharm Des. 2016;22:4151–4159.
Acknowledgements
This work was supported in part by grants from the U.S. National
Institutes of Health (R01CA122023 and R01CA211963 to D.Z. and
P20GM109005 and R56AG056372 to D.Z. and G.Z.) and a sponsored
research agreement between UNITY Biotechnology and the University
of Arkansas for Medical Sciences. Y.W., J.C., X.L., G.Z., and D.Z. filed a
patent application for the use of PL and PL analogues as anti-aging
agents. A potential royalty stream to Y.W., J.C., X.L., G.Z., and D.Z. may
occur consistent with University of Arkansas for Medical Sciences
policy. G.Z. is a consultant and D.Z. is a co-founder and advisor of
UNITY Biotechnology that develops senolytic drugs.
33. Raj L, Ide T, Gurkar AU, et al. Selective killing of cancer cells by a small molecule
targeting the stress response to ROS. Nature. 2011;475:231–234.
34. Chandrasekaran A, Idelchik M, Melendez JA. Redox control of senescence and age-
related disease. Redox Biol. 2017;11:91–102.
35. Adams DJ, Dai MJ, Pellegrino G, et al. Synthesis, cellular evaluation, and mechanism
of action of piperlongumine analogs. Proc Natl Acad Sci USA.
2012;109:15115–15120.
36. Rao VR, Muthenna P, Shankaraiah G, et al. Synthesis and biological evaluation of
new piplartine analogues as potent aldose reductase inhibitors (ARIs). Eur J Med
Chem. 2012;57:344–361.
37. Boskovic ZV, Hussain MM, Adams DJ, et al. Synthesis of piperlogs and analysis of
their effects on cells. Tetrahedron. 2013;69:7559–7567.
A. Supplementary data
38. Seo YH, Kim JK, Jun JG. Synthesis and biological evaluation of piperlongumine
derivatives as potent anti-inflammatory agents. Bioorg Med Chem Lett.
2014;24:5727–5730.
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.bmc.2018.06.013.
39. Wu YL, Min X, Zhuang CL, et al. Design, synthesis and biological activity of pi-
perlongumine derivatives as selective anticancer agents. Eur J Med Chem.
2014;82:545–551.
References
40. Peng SJ, Zhang BX, Meng XK, et al. Synthesis of piperlongumine analogues and
discovery of nuclear factor erythroid 2-related factor 2 (Nrf2) activators as potential
neuroprotective agents. J Med Chem. 2015;58:5242–5255.
41. Sun LD, Wang F, Dai F, et al. Development and mechanism investigation of a new
piperlongumine derivative as a potent anti-inflammatory agent. Biochem Pharmacol.
2015;95:156–169.
42. Liao Y, Niu X, Chen B, et al. Synthesis and antileukemic activities of piperlongumine
and HDAC inhibitor hybrids against acute myeloid leukemia cells. J Med Chem.
2016;59:7974–7990.
43. Wang Y, Wang J, Li J, et al. Design, synthesis and pharmacological evaluation of
novel piperlongumine derivatives as potential antiplatelet aggregation candidate.
Chem Biol Drug Des. 2016;87:833–840.
44. Yan WJ, Wang Q, Yuan CH, et al. Designing piperlongumine-directed anticancer
agents by an electrophilicity-based prooxidant strategy: a mechanistic investigation.
Free Radical Biol Med. 2016;97:109–123.
45. Lad NP, Kulkarni S, Sharma R, et al. Synthesis and in vitro exploration for therapeutic
potential against HeLa cancer cell lines. Eur J Med Chem. 2017;126:870–878.
46. Meegan MJ, Nathwani S, Twamley B, et al. Piperlongumine (piplartine) and analo-
gues: antiproliferative microtubule-destabilising agents. Eur J Med Chem.
2017;125:453–463.
1. Hayflick L, Moorhead PS. The serial cultivation of human diploid cell strains. Exp Cell
Res. 1961;25:585–621.
2. Hayflick L. The limited in vitro lifetime of human diploid cell strains. Exp Cell Res.
1965;37:614–636.
3. Campisi J. Aging, cellular senescence, and cancer. Annu Rev Physiol.
2013;75:685–705.
4. Aravinthan A. Cellular senescence: a Hitchhiker's guide. Hum Cell. 2015;28:51–64.
5. Braig M, Lee S, Loddenkemper C, et al. Oncogene-induced senescence as an initial
barrier in lymphoma development. Nature. 2005;436:660–665.
6. Chen Z, Trotman LC, Shaffer D, et al. Crucial role of p53-dependent cellular senes-
cence in suppression of Pten-deficient tumorigenesis. Nature. 2005;436:725–730.
7. Campisi J, Fagagna F. Cellular senescence: when bad things happen to good cells. Nat
Rev Mol Cell Biol. 2007;8:729–740.
8. Prieur A, Peeper DS. Cellular senescence in vivo: a barrier to tumorigenesis. Curr Opin
Cell Biol. 2008;20:150–155.
9. Kuilman T, Michaloglou C, Mooi WJ, et al. The essence of senescence. Genes Dev.
2010;24:2463–2479.
10. Ohtani N, Takahashi A, Mann D, et al. Cellular senescence: a double-edged sword in
the fight against cancer. Exp Dermatol. 2012;21:1–4.
11. Childs BG, Durik M, Baker DJ, et al. Cellular senescence in aging and age-related
47. Wang HB, Jin XL, Zheng JF, et al. Developing piperlongumine-directed glutathione S-
13

X. Liu et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxx–xxx
transferase inhibitors by an electrophilicity-based strategy. Eur J Med Chem.
Cell. 2018:e12780.
2017;126:517–525.
52. Alvarez S, Dominguez G, Gradillas A, et al. Unusual skeletal rearrangement of un-
saturated seven-membered lactams into fused pyrrolidinolactones. Eur J Org Chem.
2013:3094–3102.
53. Zachariassen ZG, Thiele S, Berg EA, et al. Design, synthesis, and biological evaluation
of scaffold-based tripeptidomimetic antagonists for CXC chemokine receptor 4
(CXCR4). Bioorg Med Chem. 2014;22:4759–4769.
54. Riofski MV, John JP, Zheng MM, et al. Exploiting the facile release of trifluoroacetate
for the α-methylenation of the sterically hindered carbonyl groups on (+)-sclar-
eolide and (−)-eburnamonine. J Org Chem. 2011;76:3676–3683.
48. Xu X, Fang X, Wang J, Zhu H. Identification of novel ROS inducer by merging the
fragments of piperlongumine and dicoumarol. Bioorg Med Chem Lett.
2017;27:1325–1328.
49. Zhang Y, Ma H, Wu Y, et al. Novel non-trimethoxylphenyl piperlongumine deriva-
tives selectively kill cancer cells. Bioorg Med Chem Lett. 2017;27:2308–2312.
50. Zou Y, Yan C, Zhang H, et al. Synthesis and evaluation of N-heteroaromatic ring-
based analogs of piperlongumine as potent anticancer agents. Eur J Med Chem.
2017;138:313–319.
51. Zhang X, Zhang S, Liu X, et al. Oxidation resistance 1 is a novel senolytic target. Aging
14