Novel Benzo[a]quinolizidine Analogs Induce Cancer Cell Death through Paraptosis and Apoptosis

Article abstract of DOI:10.1021/acs.jmedchem.6b00484

Paraptosis is nonapoptotic cell death characterized by massive endoplasmic reticulum (ER)- or mitochondria-derived vacuoles. Induction of paraptosis offers significant advantages for the treatment of chemotherapy-resistant tumors compared with anticancer drugs that rely on apoptosis. Because some natural alkaloids induce paraptotic cell death, a novel series of benzo[a]quinolizidine derivatives were synthesized, and their antiproliferative activity and ability to induce cytoplasmic vacuolation were analyzed. Structural optimization led to the identification of the potent compound 22b, which inhibited cancer cell proliferation in vitro and in vivo and profoundly facilitated paraptosis-like cell death and induced caspase-dependent apoptosis. Further investigation revealed that 22b-mediated vacuolation originated from persistent ER stress and upregulation of LC3B. Paraptosis induced by benzo[a]quinolizidine derivatives thus represents an alternative strategy for cancer chemotherapy.

Full text of DOI:10.1021/acs.jmedchem.6b00484

Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI
Article
Novel Benzo[a]quinolizidine Analogs Induce Cancer
Cell Death through Paraptosis and Apoptosis
Hongbo Zheng, Yiwen Dong, Lin Li, Bin Sun, Lei Liu, Huiqing Yuan, and Hongxiang Lou
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00484 • Publication Date (Web): 14 Apr 2016
Downloaded from http://pubs.acs.org on April 30, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
Novel Benzo[a]quinolizidine Analogs Induce Cancer Cell
Death through Paraptosis and Apoptosis
7
8
9
⊥
Hongbo Zheng ,†,§ Yiwen Dong,‡,§ Lin Li,† Bin Sun,†, Lei Liu,† Huiqing Yuan,*,‡ and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hongxiang Lou*,†
†Department of Natural Products Chemistry, Key Laboratory of Chemical Biology of
Ministry of Education, School of Pharmaceutical Sciences, Shandong University,
Jinan, China
‡Department of Biochemistry and Molecular Biology, School of Medicine, Shandong
University, Jinan, China
⊥National Glycoengineering Research Center, Shandong University, Jinan, China
Author Information
Corresponding Author
*L.H.:
louhongxiang@sdu.edu.cn.
*Y.H.: Tel: +86ꢀ531ꢀ8838ꢀ2346.
Tel:
+86ꢀ531ꢀ8838ꢀ2012.
Fax:
Fax:
+86ꢀ531ꢀ8838ꢀ2019.
+86ꢀ531ꢀ8838ꢀ2019.
Eꢀmail:
Eꢀmail:
lyuanhq@sdu.edu.cn.
Author Contributions
§Z. H. and D. Y. contributed equally to this work.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 36
1
2
3
4
5
6
7
8
9
ABSTRACT: Paraptosis is nonꢀapoptotic cell death characterized by massive
endoplasmic reticulum (ER)ꢀ or mitochondriaꢀderived vacuoles. Induction of
paraptosis offers significant advantages for the treatment of chemotherapyꢀresistant
tumors compared with anticancer drugs that rely on apoptosis. Because some natural
alkaloids induce paraptotic cell death, a novel series of benzo[a]quinolizidine
derivatives were synthesized, and their antiproliferative activity and ability to induce
cytoplasmic vacuolation were analyzed. Structural optimization led to the
identification of the potent compound 22b, which inhibited cancer cell proliferation in
vitro and in vivo and profoundly facilitated paraptosisꢀlike cell death and induced
caspaseꢀdependent apoptosis. Further investigation revealed that 22bꢀmediated
vacuolation originated from persistent ER stress and upregulation of LC3B. Paraptosis
induced by benzo[a]quinolizidine derivatives thus represents an alternative strategy for
cancer chemotherapy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS: Benzo[a]quinolizidines; antitumor activity; paraptosis; ER stress;
apoptosis.
ACS Paragon Plus Environment

Page 3 of 36
Journal of Medicinal Chemistry
1
2
3
4
INTRODUCTION
5
6
7
8
Most chemotherapeutic agents exert antitumor activity via caspaseꢀdependent
apoptosis.1 However, chemotherapy resistance in recurrent tumors is directly linked to
the limited success of cytotoxic drugs against cancers. Chemotherapy resistance has
also been attributed to the acquisition of antiꢀapoptotic mechanisms because cancer
cells often develop defects in apoptosis that provide a survival advantage and
contribute to chemotherapy resistance.2ꢀ4 Therefore, therapies based on the induction of
nonꢀapoptotic cell death, such as paraptosis, autophagic cell death, and necroptosis,
may provide an effective strategy for the treatment of malignant cancers that are
multiꢀdrug resistant and frequently resistant to apoptosis.5ꢀ6
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Paraptosis, a form of nonꢀapoptotic cell death characterized by extensive cytoplasmic
vacuolation resulting from swelling of the ER and/or mitochondria, is
caspaseꢀindependent and lacks the morphological hallmarks of apoptosis.1ꢀ2,7 Although
the mechanisms involved in triggering dilatation of the ER and/or mitochondria are
poorly defined, de novo protein synthesis and ER stress are likely essential because
inhibition of protein synthesis attenuates vacuolation.1,7 The unfolded protein response
(UPR), which can be initiated by ER stress, also contributes to cell death through
three ER stress sensors: protein kinase RNAꢀlike ER kinase (PERK),
inositolꢀrequiring enzyme 1α (IRE1α) or activating transcription factor 6 (ATF6; both
α and β isoforms). Extensive efforts have focused on the identification of ER
stress/UPR activators that specifically trigger these alternate cell death mechanisms as
novel cancer treatment options. For example, the cannabinoid receptor agonist
(R)ꢀ(+)ꢀ[2,3ꢀDihydroꢀ5ꢀmethylꢀ3ꢀ(4ꢀmorpholinylmethyl)pyrrolo[1,2,3ꢀde]ꢀ1,4ꢀbenzox
azinꢀ6ꢀyl]ꢀ1ꢀnapthalenylmethanone (WIN55, 212ꢀ2) selectively inhibits cell
proliferation via paraptosisꢀlike cell death after activating proꢀdeath UPR signaling.6
The immunological inhibitor celastrol also induces paraptosisꢀlike cytoplasmic
vacuolation in cancer cells, possibly by affecting a variety of pathways, including the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 36
1
2
3
4
5
6
7
8
9
proteasome, ER stress and Hsp90.8ꢀ9 Many UPR activators, such as radicicol and
tanespimycin, are also used in the treatment of various cancers.10 Activation of
nonꢀapoptotic cell death in parallel with apoptosis is thought to facilitate the tumor
response to the drug in anticancer treatments.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Many natural alkaloids are employed in clinical chemotherapy. Taxol, a diterpene
alkaloid, causes ER stress and paraptosis in a large range of human cancer cells in
addition to inducing apoptotic cell death as a major mechanism.11 Bromocriptine (Fig.
1) is another typical chemotherapeutic drug for the treatment of pituitary tumors
primarily via paraptosis.12ꢀ13 Benzo[a]quinolizidine derivatives, such as berberine,
emetine and deoxytubulosine (Fig. 1), are also important naturally occurring
antiproliferative agents, and their "privileged" structure always provides unique
activity.14ꢀ16 Thus, the development of
a
rapid chemical synthesis for
benzo[a]quinolizidines and further chemical modifications of these potential products
may facilitate the identification of more potent and selective derivatives.
Our ongoing research on antitumor compounds prompted us to synthesize a series of
benzo[a]quinolizidine derivatives for the discovery of novel agents with unique
activity.17ꢀ19 Based on a structureꢀactivity relationship (SAR) investigation, an
optimized compound, 22b, was identified that exhibited antitumor activity at low
concentration accompanied by massive cytoplasmic vacuoles in cellꢀbased assays. We
then examined the dependence of the cytoplasmic vacuolation induced by 22b on ER
stress, autophagy and protein translation. The results demonstrated that 22b triggers ER
vacuolation with specific characteristics of paraptosisꢀlike cell death in addition to
inducing apoptosis. The antitumor efficiency of 22b in PC3 tumorꢀbearing mice
indicated that 22b potently suppresses tumor growth without significant toxicity.
RESULTS AND DISCUSSION
Design. The novel benzo[a]quinolizidines were designed based on an analysis of
naturally occurring alkaloids that target subcellular organelles and induce
ACS Paragon Plus Environment

Page 5 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
paraptosis.20ꢀ21 Biologyꢀoriented modifications of the structures were performed to
target three potential active positions: (a) the aromatic substituent effect on
tetrahydroisoquinoline ring system; (b) functionalization and optimization of position 4;
and (c) modification and exploration of the important hydroxyl group by introducing
various chemical moieties while retaining the benzo[a]quinolizidine skeleton. Thus, a
novel series of benzo[a]quinolizidine derivatives were synthesized, and their
antiproliferative activity was evaluated in cellꢀbased assays.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chemistry. We developed a new method for constructing the benzo[a]quinolizidine
ring according to Gawley's strategy.22 The general procedure for benzo[a]quinolizidine
synthesis is shown in Scheme 1. Tetrahydroisoquinolines II were first prepared,
followed by the introduction of different Oꢀmodifications III and NꢀBoc protection IV.
NꢀBoc ethoxy derivatives V, the key intermediates, were synthesized by adding the
oxidant TEMPO and alcohol under conditions similar to those described previously.23
NꢀBoc ethoxy V was then coupled with various trimethylsilyl enol ethers VI prepared
from methyl vinyl ketones. We performed reactions in the presence of different Lewis
acids, such as TMSOTf, BF3·EtO2 and M(OTf)x, and Yb(OTf)3 was selected as the
most useful additive to facilitate the generation of 1ꢀsubstituted
tetrahydronisoquinolines VII under mild conditions. The protective group was then
removed by TFA, followed by intramolecular Michael addition in NH3·H2O/methanol
to afford the desired benzo[a]quinolizidines VIII as transꢀisomers with high
diastereoselectivity. The total yield of V from VIII (40–60%) was acceptable for
further modification. Reaction details can be obtained in Schemes S1ꢀS2 and Table
S1.
Ketone VIII was further modified via the Grignard reaction to afford IX. Reduction
of VIII using the WolffꢀKishnerꢀHuang reaction or other reductants (NaBH4 or
Lꢀselectride) led to the generation of products X and XI, respectively (Scheme 2).
All benzo[a]quinolizidine derivatives were prepared as racemates for biological
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 36
1
2
3
4
5
6
evaluation. The relative configuration was determined by analysis of NOESY spectral
data (Fig. S1).
7
8
9
In Vitro Antiproliferative Activity Screening and SAR Analysis. The effect of
these benzo[a]quinolizidine derivatives on cell proliferation was investigated in a panel
of human cancer cell lines by MTT assay. Nonꢀneoplastic prostate epithelial RWPE1
cells were included to test the cytotoxicity of these compounds. The bioactive
structures are shown in Chart 1, and their IC50 values are listed in Table 1. The
remaining derivatives and related antiproliferative activity are provided in Tables
S2ꢀS5. The preliminary screening of 9,10ꢀsubstituted benzo[a]quinolizidines revealed
that only compounds 1ꢀ3 exhibited an inhibitory effect on most of the tested tumor cell
lines. The basic ringꢀfused skeleton, particularly the type b configuration shown in
Chart 1 (cisꢀrelationship between Cꢀ2 and Cꢀ4), and the two benzyloxy groups at
position 9 and 10 were essential for high cytotoxicity.24ꢀ25 By contrast, all derivatives
with alkoxyl moieties instead of benzyloxy substituents at positions 9 and 10 were
completely inactive.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Next, we focused on modification of lead compound 1 in three positions. The
results of the MTT assay indicated that the following features contributed to optimal
activity. First, among the benzo[a]quinolizidines with various substituted phenyl rings
introduced at Cꢀ4, the order of antiꢀproliferation potency was ꢀNO2 (no cytotoxicity) <
3,4,5ꢀOMe < 4ꢀOMe < H < 3,4ꢀOMe < 4ꢀPh, indicating that electronꢀdonating groups
(phenyl or methoxyl moiety) were critical for cytotoxicity. In another series of alkyl
chain compounds 11ꢀ14, the paired diastereoisomers exhibited superior bioactivity.
The length of the alkyl chain modulated the potency, and the methyl group was the
optimal pharmacophore vector for the antiꢀproliferative effect. Second, reductive
product 15 was devoid of activity, implying that the 2ꢀhydroxyl is essential.
Additionally, introduction of a methyl group at Cꢀ2 increased cytotoxicity, whereas the
activity of analogues with a longer alkyl chain or phenyl ring decreased or was absent.
ACS Paragon Plus Environment

Page 7 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Third, changes in substituents at the phenyl ring of the benzyloxy group at Cꢀ9 and
Cꢀ10 affected activity. The IC50 values of 20-27 in Table 1 revealed that derivatives
with the πꢀrich aryl ring were significantly active, and these substituents enhanced
activity in the order 3,5ꢀCF3 > 3,5ꢀMe > 4ꢀtBu > 4ꢀBr > 3,5ꢀF > 3ꢀF > H. Similarly,
derivatives 28ꢀ34 featuring a methyl group (type IX analogues) displayed a similar
pattern, with the exception of the 3ꢀfluorobenzyl derivative 31, which exhibited
superior antitumor activity. Cellꢀbased activity assays indicated that most type VIII
derivatives were inactive, whereas several 4ꢀmethyl analogues were moderately
cytotoxic. Notably, compound 35 (3ꢀF benzyl) displayed strong inhibition of
MDAꢀMBꢀ231 cells (IC50 = 1.3 ꢁM, Table S4). Octahydroazocino[2,1ꢀa]isoquinolines
with specific 8ꢀmembered rings (37 and 38), which were obtained unexpectedly, also
exhibited moderate cytotoxicity (IC50 = 8.3–13.0 ꢁM).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The SAR analysis (Fig. 3) indicated that 9,10ꢀbisbenzoxyl groups, 4ꢀmethyl and
2ꢀhydroxyl, were very important for activity. Prostate cancer (PCa) and breast cancer
cells appeared to be more sensitive to the novel benzo[a]quinolizidine derivatives.
PC3, an androgenꢀindependent tumor cell line, exhibits elevated expression of
antiꢀapoptotic proteins, enabling apoptosis resistance26; therefore, we investigated the
potent compounds 20b, 22aꢀ22b, 24aꢀ24b, 27b, 28, 30ꢀ31 in PC3 cells. These
compounds profoundly inhibited cell proliferation in association with the induction of
massive cytoplasmic vacuoles (Table 1 and Fig. 4), and the electronꢀrich benzoxyl
groups were essential for vacuolationꢀassociated cell death. 3,5ꢀDimethylbenzyl
derivatives (e.g., 20b, 22b; Fig. 4) induced vacuole production more potently than
3,5ꢀbis(triꢀfluoromethyl)benzyl (e.g., 30) and other electronꢀrich benzoxyl derivatives
(or electronꢀdeficient benzoxyl derivatives such as 26b). However, the positive control,
docetaxel, did not induce vacuolation. Notably, among these compounds, 22b
exhibited smaller inhibitory effects against RWPE1 cells with a higher IC50 (Table 1).
Compounds 22bꢀ1 and 22b-2, optical isomers (Fig. S2), displayed similar potency in
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 36
1
2
3
4
5
6
vacuoleꢀassociated cell death, and 22b was selected for subsequent analysis in PC3
cells.
7
8
9
22b Induces Cytoplasmic Vacuolation and Apoptosis. Given the inhibitory effect
of compound 22b on PCa cells, we analyzed changes in cell cycle distribution and
apoptosis in response to 22b. Flow cytometry analysis of PC3 cells treated with 22b
(0.5, 1, and 2 ꢁM) for 24 h revealed that compared with the control (62.5% of cells in
G0/G1 phase), 22b induced cell cycle arrest at G0/G1 phase (Fig. 5A). The percentage
of cells in G0/G1 phase was 68.1, 72.4, and 81.4% at 0.5, 1, and 2 ꢁM, respectively.
The results in Fig. 5B indicate that 22b induced apoptosis in PC3 cells in a
doseꢀdependent manner, with apoptosis levels of 3.7%, 8.5%, and 56.9% at
concentrations of 0.5, 1.0 and 2.0 ꢁM, respectively. To determine the role of apoptosis
in 22bꢀmediated cell death, PC3 cells were pretreated with the panꢀcaspase inhibitor
ZꢀVADꢀfmk to block caspaseꢀdependent apoptosis. As shown in Fig. 5D, ZꢀVADꢀfmk
partially rescued 22bꢀinduced cell death, indicating that caspaseꢀindependent
mechanisms also partially contributed to 22bꢀinduced cell death. The cells underwent
cytoplasmic vacuolation in the presence of 22b, and thus we determined if the
characteristics of the vacuoles were correlated with cell death. Cellular vacuoles were
readily observed by phaseꢀcontrast microscopy as shown in Fig. 5C. Small vacuoles
appeared in 22bꢀtreated cells after approximately 4 h and increased in size and number
in a timeꢀdependent manner. Robust vacuolation persisted up to 24 h of treatment; at
this time point, nonviable cells were evident, and most of the cells had detached from
the dish. Timeꢀcourse studies were also performed to examine the change in cleavage
of poly (ADPꢀribose) polymerase (PARP), an apoptosis marker, by activated caspase 3.
As shown in Fig. 5D, only modest cleavage of PARP was observed after 24 h of
treatment, implying a critical role of cytoplasmic vacuolation in 22bꢀinduced cell death
because caspaseꢀdependent apoptosis occurred after vacuole production. Thus,
alternative mechanisms are involved in 22bꢀinduced cell death because apoptosis
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 9 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
occurred as a late event during treatment. More importantly, ZꢀVADꢀfmk partially
restored 22bꢀinduced cell death but did not significantly affect cytoplasmic vacuolation.
Cells were partially rescued from 22bꢀinduced cell death by another panꢀcaspase
inhibitor, AcꢀDEVDꢀCHO. (Fig. 5D). Thus, the formation of vacuoles plays an
important role in 22bꢀinduced caspaseꢀindependent cell death.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22b-Mediated Paraptosis Depends on ER Stress and Protein Synthesis. Vacuoles
were observed at the perinuclear region and in the cytosol; therefore, we utilized
ERꢀTracker Red staining to determine if the vacuoles were derived from the ER. As
shown in Fig. 6A, untreated cells displayed a smear pattern upon ER staining, whereas
fluorescence intensity notably increased in cells treated with 22b for 4 h. Furthermore,
vacuoles were readily observed after 8 h of exposure and were surrounded by a positive
ERꢀspecific marker, suggesting that the vacuoles were of ER origin. The generation of
ER vacuoles is usually associated with persistent ER stress; therefore, we reasoned that
22b-induced vacuolation may be preceded by the induction of ER stress. The results in
Fig. 6B demonstrate that 22b upregulated glucoseꢀregulated protein GRP78, a hallmark
of the ER stress response. Expression of GRP78 significantly increased after 4 h of
exposure to 22b, and vacuoles started to appear, reaching a peak at approximately 8–12
h and sustained at high levels for up to 24 h. Activation of pꢀPERK and
PERKꢀmediated phosphorylation of eukaryotic initiation factor (pꢀeIF2α) were also
observed (Fig. 6B), indicating activation of the PERK branch of UPR signaling.27ꢀ28
Persistent UPR ultimately initiates apoptosis if these adaptive responses are
insufficient to protect cells from ER stress.29 We also observed a change in the
expression of CHOP, a protein involved in the ERꢀmediated apoptosis. Upregulation of
CHOP was observed at 8 h of treatment and sustained up to 24 h, confirming the ability
of 22b to trigger ER stress, which may also contribute to caspaseꢀdependent apoptosis.
To investigate the requirement for ER stress in the formation of vacuoles, we used
4ꢀphenylbutyric acid (4ꢀPBA), an ER stress inhibitor, to suppress 22bꢀmediated ER
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 36
1
2
3
4
5
6
7
8
9
stress. As shown in Fig. 6C, enhancement of GRP78 by 22b was abolished in the
presence of 4ꢀPBA. Meanwhile, 22bꢀmediated activation of XBP1 mRNA splicing
was also attenuated when combined with 4ꢀPBA. More importantly, 4ꢀPBA alone did
not cause any detectable change in cellular morphology and cytotoxicity, whereas
4ꢀPBA markedly alleviated vacuole production and cell death induced by 22b,
highlighting the importance of ER stress in the regulation of vacuole formation, which
is closely correlated with 22bꢀmediated cell death. Given the importance of ER stress
in 22bꢀinduced vacuole formation and cell death, we determined if proteasome
inhibitors that trigger ER stress facilitate 22bꢀinduced vacuolation and cell death.
CompuSyn software was used to analyze the effect of 22b combined with 39 (MG132)
or bortezomib, proteasome inhibitors that trigger ER stress.30ꢀ32 The results in Fig. 6D
and Table S6 demonstrate that the combination of 22b at IC30 and 39 at various
concentrations yielded combination index (CI) values of 0.1 to 0.3, indicating a strong
synergistic effect. For example, 22b at IC30 resulted in 54% inhibition of PC3 cell
growth when combined with 39 at its IC10 concentration. Bortezomib also displayed
synergistic effects (Table S7). Protein overload in the ER lumen may induce the stress
response; therefore, we hypothesized that abrogating protein synthesis with
cycloheximide (CHX) would promote cell survival in the presence of 22b. As shown in
Fig. 6E, halting protein synthesis via the addition of CHX significantly reduced
22bꢀinduced cytoplasmic vacuolation and cell death. We also demonstrated that the
transcriptional inhibitor actinomycinꢀD (ActꢀD) dramatically reduced cellular
vacuolation as well as cell death, revealing an important role of de novo gene
expression in cytoplasmic vacuolation and the cell death response. Thus, the
characteristics of cell death induced by 22b included cytoplasmic vacuolation,
induction of ER stress, the requirement for active gene expression, and the absence of
inhibition by caspase inhibitors, fulfilling the criteria for paraptosis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22b-Induced Cytoplasmic Vacuolation Requires Elevated LC3. Vacuolation is
ACS Paragon Plus Environment

Page 11 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
also a feature of autophagy, and CHX was unable to completely abolish the formation
of vacuoles induced by 22b (Fig. 6E); we therefore examined whether 22b activates
autophagy in PC3 cells. As shown in Fig. 7A, the expression of LC3, an
autophagosome marker, was slightly upregulated in treated cells compared with the
vehicle control, whereas conversion of LC3BꢀI to lipidated LC3BꢀII (a PEꢀconjugated
form) was observed in cells after 4 h of treatment and markedly robust at up to 24 h of
treatment with 22b, suggesting activation of autophagic flux. However, the expression
level of p62/SQSTM1, a protein preferentially degraded by the autophagic pathway,
remained unaffected following 4 h of treatment and was elevated in cells exposed to
22b for prolonged treatment despite the increased LC3ꢀII.33 The lack of decrease in
LC3BꢀII and p62 levels over time during the treatments indicates that autophagosomes
may not be efficiently fused with lysosomes. To determine if 22b interrupts the fusion
of autophagosomes with lysosomes and causes defective autophagy, a pHꢀsensitive
LC3ꢀreporter plasmid containing both GFP and RFP fusion tags (mRFPꢀGFPꢀLC3)
was transfected into cells prior to exposure to 22b.34 The green GFP signal in the
LC3ꢀreporter is quenched in acidic lysosomes; therefore, autophagosomes appear
yellow due to overlapped red and green fluorescence at less acidic pH, and
autolysosomes appear red. As shown in Fig. 7B, rapamycin, a wellꢀknown inducer of
autophagy, activated an obvious autophagy flux from autophagosomes to
autolysosomes, as evidenced by the strong red punctate dots. However, green/yellow
fluorescence of LC3 puncta was clearly observed in 22bꢀtreated cells, indicating that
22b caused impaired autophagosomeꢀlysosome fusion, in agreement with the
observations that 22b treatment resulted in the accumulation of lipidꢀconjugated LC3
and p62 (Fig. 7A). Notably, increased granulation of LC3 was observed during
rapamycinꢀinduced autophagy, whereas the LC3 in 22bꢀtreated cells displayed a
perivacuolar distribution, although several LC3ꢀpositive vacuoles were clearly
observed. These results prompted us to explore the role of autophagy in 22bꢀinduced
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 36
1
2
3
4
5
6
7
8
9
vacuole formation and cell death. As shown in Fig. 7C, in the presence of
3ꢀmethyladenine (3ꢀMA), which is a PI3K inhibitor to inactivate autophagy,
22bꢀinduced cell viability was partially reversed but had no detectable impact on
cytoplasmic vacuolation. The cell response to Eꢀ64d, an inhibitor of lysosomal
enzymes, was similar to the response to 3ꢀMA. The persistent upregulation of LC3ꢀII
and p62 and the lack of protective effect of 3ꢀMA and Eꢀ64d demonstrated that
22bꢀinduced vacuolation was independent of classic autophagy and lysosomal
degradation. However, elevated LC3ꢀII was critical in vacuole formation induced by
22b, and the depletion of LC3 by LC3 siRNA nearly completely abolished the
22bꢀinduced vacuole level (Fig. 7D). This finding is consistent with reports that
increased LC3ꢀII is not specific for autophagy but plays a nonꢀautophagic role during
cytoplasmic vacuolation.35ꢀ36 Thus, 22bꢀinduced vacuolation associated with the
formation of LC3ꢀIIꢀpositive nonꢀautophagic vacuoles contributes to the cytotoxicity
of 22b.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Vivo Anticancer Activity of 22b. The inhibitory efficiency of 22b against the
growth of PCa cells was evaluated in a PC3MꢀlucꢀC6 (modified PC3M cells expressing
luciferase) xenograft model. The mice were administered 22b (15 mg/kg)
intraperitoneally every other day for 18 consecutive days and were compared with
placeboꢀtreated mice. A group of mice injected with docetaxel (15 mg/kg) served as a
positive control. As shown in Fig. 8A and 8B, 22b significantly reduced the growth of
tumors by 40.5% compared with the placebo control, as indicated by the reduction in
volume (410.2±180.3 for treatment versus 845.6±84.3 for vehicle) and weight
(0.42±0.08 for treatment versus 0.71±0.06 for vehicle). Although docetaxel exerted
higher antitumor efficacy than 22b (percent growth inhibition was 92% on day 18),
docetaxel exhibited higher toxicity as indicated by the significant body weight loss
compared with 22b treatment, which did not affect mouse body weight or other
morbidity (Fig. 8C, 13.3 g vs. 16.3 g). Additionally, one mouse died in the
ACS Paragon Plus Environment

Page 13 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
docetaxelꢀtreated group, whereas no mice were lost during 22b treatment, supporting
the conclusion that 22b was less toxic in animal trials at the same dosage.
7
8
9
Bioluminescence measurements of tumors in each mouse on days 0, 9, and 18 of the
treatments revealed that all tumorꢀbearing mice exhibited ~108 photons/s/cm2 in
photon intensity prior to treatment. Mice treated with 22b exhibited 1.8 × 109
photons/s/cm2 on day 18, and the placebo group exhibited 2.5 × 1010 photons/s/cm2 on
day 18 (Fig. 8E), indicating that 22b caused a significant decrease in tumor burden
with diminished cell activity compared with the placebo group. Immunohistochemical
staining supported the notion that 22b inhibited tumor cell proliferation because a
significant decrease in Kiꢀ67ꢀpositive cells was observed in xenograft tumors treated
with 22b compared with the placebo group (Fig. 8F), consistent with the
bioluminescence data. Importantly, H&E staining clearly showed extensive vacuoles
in mice treated with 22b, providing evidence that 22b induces paraptosis in vivo (Fig.
8F). In addition, consistent with the observations in cell culture, the increased LC3,
p62, and GRP78 levels in tumors (Fig. 8G) revealed that 22b activated an immature
autophagic response and induced remarkable ER stress in vivo. Together, our data
clearly demonstrate that 22b exhibits antiꢀtumor effects by eliciting paraptosisꢀlike
cell death with low toxicity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSION
In summary, a new series of novel benzo[a]quinolizidines were synthesized based on
SAR analysis. Screening assays of these derivatives identified the lead compound 22b
as a potential antitumor agent. The novel compound 22b, which features
3,5ꢀdimethylbenzyloxy groups at positions 9 and 10, is particularly attractive for future
studies based on the following features: (i) inhibition of cell proliferation with a low
IC50 via the induction of both apoptosis and paraptosis by a combination of
mechanisms, which will be beneficial for chemotherapyꢀresistant cancers; (ii) the
impairment of autophagy maturation by 22b contributes to its antitumor efficacy
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 36
1
2
3
4
5
6
7
8
9
because the activation of autophagy is generally cytoprotective; and (iii) the low
toxicity in in vivo experiments supports potential human applications. The activation of
more than one type of cell death by 22b highlights its potential therapeutic applications
in cancer treatment.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
General Materials and Methods. Melting points were determined with an Xꢀ6
melting point apparatus (Beijing TECHInstrument Co. Ltd.) and are uncorrected. IR
spectra were recorded on a Nicolet iN 10 Micro FTIR spectrometer. NMR spectra
were obtained using a Bruker Avance AVIIIꢀ400 spectrometer operating at 400 (1H) or
101 (13C) MHz. HRꢀESIꢀMS was conducted with an LTQ Orbitrap XL instrument.
HPLC was performed with an Agilent 1200 system equipped with a G1311A isopump,
a G1322A degasser, and a G1315D DAD detector using an Eclipse XDBꢀC18 (150
mm × 4.6 mm × 5 µm) and Chiralpak ASꢀH (250 mm × 4.6 mm × 5 µm). Column
chromatography was conducted on silica gel or alumina (200–300 mesh). All solvents
used were of analytical grade. Precoated silica gel GF254 plates (Qingdao Haiyang
Chemical Co., Ltd.) were used for thinꢀlayer chromatography. All reactions were
conducted under a nitrogen atmosphere with dry solvents under anhydrous conditions.
The purities of all biologically evaluated compounds were established as ≥ 95% by
HPLC analysis with UV detection.
Cell Culture and Treatments. The human prostate carcinoma cell lines PC3 and
DU145, the breast carcinoma cell line MDAꢀMB231, MCFꢀ7 cells, HeLa cells, and the
myelogenous leukemia cell line K562 (The American Type Culture Collection
(ATCC), Rockville, MD) were cultured in RPMI 1640 medium (HyClone, Logan, UT)
supplemented with 10% fetal bovine serum (FBS) (Hyclone). The human prostate
epithelial cell line RWPEꢀ1 was maintained in Keratinocyte1 medium (KꢀSFM)
supplemented with 50 mg/L bovine pituitary extract and 5 ꢁg/L epidermal growth
factor (Gibco, Grand Island, NY). Luciferaseꢀexpressing PC3M cells (PC3MꢀLucꢀC6)
ACS Paragon Plus Environment

Page 15 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(Shanghai Genomics, China) were cultured in MEM medium (HyClone) supplemented
with 1% nonessential amino acids (SigmaꢀAldrich), 1% Lꢀglutamine (ICN), and 1 mM
sodium pyruvate (Sigma–Aldrich). The cells were incubated with 100 units/mL
penicillin and 100 ꢁg/mL streptomycin and were maintained in 5% CO2 at 37 °C until
they reached approximately 50−70% confluence and were then treated with various
concentrations of compounds. Dimethyl sulfoxide (DMSO) was used as the control
vehicle.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Antiproliferation Assay. The effects of the compounds on the proliferation of the
human
cell
lines
were
evaluated
by
MTT
[3ꢀ(4,5ꢀdimethylthiazolꢀ2ꢀyl)ꢀ2,5ꢀdiphenylꢀ2Hꢀtetrazolium bromide, Sigma, St. Louis,
MO] assay. Briefly, cells were seeded into 96ꢀwell plates for 24 h and then treated with
vehicle, the desired compounds alone, or docetaxel as a positive control for 24 h. After
incubation with 10 ꢁL of MTT (5 mg/mL) for 4 h, absorbance of the soluble MTT
product was measured at 570 nm. The antiproliferation assay was performed in
triplicate. The RealTime Glo assay was also performed for individual experiments.
Cell Cycle and Apoptosis Assay. For cell cycle analysis, after treatment for 24 h, the
PC3 cells were fixed in 70% cold ethanol overnight at 4 °C, incubated for 30 min in
PBS containing 100 ꢁg/mL RNase (Invitrogen) and then incubated with 50 ꢁg/mL PI
(Propidium iodide, SigmaꢀAldrich) at 4 °C for 30 min in the dark. Data were processed
using ModFit LT Software. Apoptotic cells were detected using an Annexin
V−fluorescein isothiocyanate (FITC)/propidium iodide (PI) apoptosis detection kit
(BD, San Jose, CA, USA). Flow cytometry was performed with a FACScan cytometry
instrument (FACSCalibur, Becton Dickinson, USA). Data were analyzed using
MODFIT or CELLQUEST software (Verity Software House, Topsham, Maine, USA).
In the apoptotic analysis, necrotic (Annexin V−/PI+), late apoptotic (doubleꢀstained),
normal (Annexin V−/PI−) and early apoptotic (Annexin V+/PI−) cells were labeled as
B1ꢀB4.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 36
1
2
3
4
5
6
7
8
9
Microscopy. PC3 cells in various stages after treatment were incubated at 37 °C for
15 min with ERꢀTracker (1:3000, Beyotime, China), LysoꢀTracker (1:20000,
Beyotime, China) or MitoTracker (200 nM, Beyotime, China) and 10 ꢁg/mL of
Hoechst 33342 (Sigma–Aldrich). The cells were then washed with PBS for confocal
microscopy (Carl Zeiss). We used ERꢀTracker, LysoꢀTracker, MitoꢀTracker or Hoechst
33342 for endoplasmic reticulum, lysosomal, mitochondrial and nuclear integrity
assays, respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Western Blotting. The cells were lysed in RIPA lysis buffer. A total of 40 ꢁg of
protein was quantified and loaded onto an SDSꢀPAGE gel. After electrophoresis, the
proteins in the gel were transferred to a nitrocellulose membrane and incubated with
primary antibodies at 4 °C overnight. The membrane was washed and incubated with
HRPꢀconjugated secondary antibodies for 45 min. The immunoblot bands were
detected by an ECL system, and the membranes were exposed to Xꢀray films.
Drug Combination Analysis. Three replicates were averaged, and repeats of these
data sets were entered into CalcuSyn (Biosoft, Ferguson, MO) for analysis. This
program uses the median effect analysis algorithm, which produces the combination
index (CI) value as a quantitative indicator of the degree of synergy or antagonism.
Using this analysis method, CI < 1, CI = 1, and CI > 1 indicate synergism, an additive
effect, and antagonism, respectively.
Autophagy Assays. To evaluate the formation of autophagosomes and
autolysosomes, the mRFPꢀGFPꢀLC3 reporter plasmid was transfected into PC3 cells
and incubated with 22b for 8 h. Images were acquired using confocal microscopy.
siRNA Transfection. PC3 cells were plated into sixꢀwell plates at 20–30%
confluency, and 24 h later, the cells were transfected with 50 nM siRNA using
Lipofectamine 2000 (Invitrogen). The siRNAs targeting LC3 and the scramble control
siRNA were designed, modified and synthesized by Invitrogen.
Inhibition of Tumor Growth In Vivo. Male athymic (BALB/cꢀnu) mice (6 weeks
ACS Paragon Plus Environment

Page 17 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
old) were obtained from the Animal Center of the Chinese Academy of Medical
Sciences (Beijing, China). Our experimental protocol was reviewed and approved by
the Institutional Animal Care Committee of Shandong University. PC3MꢀlucꢀC6 cells
(5 × 106 cells used for each injection) were injected subcutaneously into the right
anterior flanks of the mice. When the tumor volume reached approximately 0.1 cm3, the
mice were sorted into three groups (n = 5), and administration was started. Fifteen
mg/kg of 22b suspension was injected via intraperitoneal injection, and docetaxel (15
mg/kg) was injected as a positive control. Administration was performed every 2 days
for three consecutive weeks. The percentages of growth inhibition were defined as the
ratio of tumor weight to that in the vehicle control. Tumor dimensions were determined
using calipers, and the tumor volume (mm3) was calculated using the following the
formula: volume = length × (width)2 × 0.5. Tumor size was assessed in each mouse via
in vivo bioluminescence measurement using the IVIS Imaging System (Caliper Life
Sciences, USA). The photometry of the tumor was calculated using Living Image 3.1.0
(Caliper Life Sciences) software, and tumor growth curve was calculated.
Bioluminescent PC3MꢀlucꢀC6 cells constitutively expressing luciferase were
implanted under the skin of mice for primary tumor assessment. For the luciferase
detection imaging, luciferin (Shanghai Genomics, China) was dissolved in PBS and
injected (i.p., 150 mg/kg, 200 ꢁL) before imaging. During in vivo imaging, the mice
were immobilized using the anesthetic isoflurane (1.5–2.5%).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Immunocytochemistry. H&E staining was performed using routine methods. The
frozen tissue sections were rewarmed at 60 °C for 30 min, washed with PBS and
incubated overnight at 4 °C with primary antibody (antiꢀKi67 (1:500, mouse; Santa
Cruz Biotechnologies), antiꢀLC3 (1:800, rabbit; Nonvus), antiꢀp62 (1:500, mouse;
Santa Cruz Biotechnologies) and antiꢀGRP78 (1:500, rabbit; Stressgen)) diluted in
PBS. After staining with IgGꢀconjugated HRP and DAB (Vector Laboratories), the
samples were counterstained with hematoxylin, and images were captured by
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 36
1
2
3
4
microscopy (Nikon).
5
6
7
8
Statistical Analysis. Data are presented as the mean ± SD for triplicate experiments.
Statistically significant differences between treated groups and controls were analyzed
using Student’s tꢀtest, and p < 0.05 was considered statistically significant.
ASSOCIATED CONTENT
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information.
1
1
HRꢀESIꢀMS, H and 13C NMR data for all benzo[a]quinolizidines; the key H−1H
COSY, HSQC, HMBC, and NOESY correlations of compounds 1, 4, 11b, 16a, 18; IR
spectra of compounds 22aꢀ22b, 28; the whole MTT results of all the derivatives;
Optimization for the ZꢀVADꢀfmk assays; westernblot analysis of decrease of IkBꢀa and
pꢀP65 reponse to CHX and ActD and CI values. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*L.H.:
louhongxiang@sdu.edu.cn.
*Y.H.: Tel: +86ꢀ531ꢀ8838ꢀ2346.
Tel:
+86ꢀ531ꢀ8838ꢀ2012.
Fax:
Fax:
+86ꢀ531ꢀ8838ꢀ2019.
+86ꢀ531ꢀ8838ꢀ2019.
Eꢀmail:
Eꢀmail:
lyuanhq@sdu.edu.cn.
Author Contributions
The manuscript was written through contributions of all authors.
§Z. H. and D. Y. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
This work was financially supported by the National Natural Science Foundation of
China (81172956 and 81473107), and Program for Changjiang Scholars and
Innovative Research Team in University (IRT13028).
ACS Paragon Plus Environment

Page 19 of 36
Journal of Medicinal Chemistry
1
2
3
4
ABBREVIATIONS
5
6
7
8
ER, endoplasmic reticulum; LC3, microtubuleꢀassociated protein 1 light chain 3;
Hsp90, Heat shock protein 90; TEMPO, 2,2,6,6ꢀTetramethylꢀ1ꢀpiperidinyloxy; SAR,
9
structureꢀactivity
3ꢀ(4,5ꢀdimethylthiazolꢀ2ꢀyl)ꢀ2,5ꢀdiphenyltetrazolium
(ADPꢀribose) polymerase;
Relationship;
MTT,
poly
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
bromide;
PARP,
ZꢀVADꢀfmk,
ZꢀValꢀAlaꢀDLꢀAsp(OMe)ꢀfluoromethylketone; GRP78, gastric inhibition polypeptide
78; UPR, unfolded protein response; CHOP, C/EBPꢀhomologous protein; PERK,
PKRꢀlike endoplasmic Reticulum Kinase; eIF2α, PERKꢀmediated phosphorylation of
eukaryotic initiation factor 2α; 4ꢀPBA, 4ꢀphenylbutyricacid; 3ꢀMA, 3ꢀmethyladenine;
CHX, cycloheximide; ActꢀD, actinomycinꢀD; mRFP, monomeric red fluorescence
protein; GFP, green fluorescence protein; PC3MꢀlucꢀC6, modified PC3M cells that
express luciferase.
REFERENCES
(1) Sperandio, S.; de Belle, I.; Bredesen, D. E. An alternative, nonapoptotic form of
programmed cell death. Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 14376−14381.
(2) Bröker, L. E.; Kruyt, F. A.; Giaccone, G. Cell death independent of caspases: A
review. Clin. Cancer Res. 2005, 11, 3155−3162.
(3) Isoldi, M. C.; Visconti, M. A.; Castrucci, A. M. Antiꢀcancer drugs: Molecular
mechanisms of action. Mini. Rev. Med. Chem. 2005, 5, 685–695.
(4) Alexander, K.; Véronique, M.; Rastogi, S. K.; Florence, L.; Robert, K.
Therapeutic agents triggering nonapoptotic cancer cell death. J. Med. Chem. 2013, 56,
4823−4839.
(5) Kroemer, G. L.; Galluzzi, P.; Vandenabeele, P.; Abram, J.; Alnemri, E. S.;
Baehrecke, E. H.; Blagosklonny, M. V.; ElꢀDeiry, W. S.; Golstein, P.; Green, D. R.;
Hengartner, M.; Knight, R. A.; Kumar, S.; Lipton, S. A.; Malorni, W.; Nunez, G.; Peter,
M. E.; Tschopp, J.; Yuan, J.; Piacentini, M.; Zhivotovsky, B.; Melino, G. Classification
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 36
1
2
3
4
of cell death. Cell Death Differ. 2009, 16, 3–11.
5
6
7
8
(6) Wasik, A. M.; Almestrand, S.; Wang, X.; Hultenby, K.; Dackland, Å. L.;
Andersson, P.; Kimby, E.; Christensson, B.; Sander, B. WIN55,212ꢀ2 induces
cytoplasmic vacuolation in apoptosisꢀresistant MCL cells. Cell Death Dis. 2011, 2,
e225.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(7) Sperandio, S.; Poksay, K.; de Belle, I.; Lafuente, M. J.; Liu, B.; Nasir, J.;
Bredesen, D. E. Paraptosis: Mediation by MAP kinases and inhibition by AIPꢀ1/Alix.
Cell Death Differ. 2004, 11, 1066−1075.
(8) Yoon, M. J.; Lee, A. R.; Jeong, S. A.; Kim, Y. S.; Kim, J. Y.; Kwon, Y. J.; Choi, K.
S. Release of Ca2+ from the endoplasmic reticulum and its subsequent influx into
mitochondria trigger celastrolꢀinduced paraptosis in cancer cells. Oncotarget 2014, 5,
6816−6831.
(9) Wang, W. B.; Feng, L. X.; Yue, Q. X.; Wu, W. Y.; Guan, S. H.; Jiang, B. H.; Yang,
M.; Liu, X.; Guo, D. A. Paraptosis accompanied by autophagy and apoptosis was
induced by celastrol, a natural compound with influence on proteasome, ER stress and
Hsp90. J. Cell. Physiol. 2012, 227, 2196−2206.
(10) Claudio, H.; Eric, C.; Heather, P. H. Targeting the unfolded protein response in
disease. Nat. Rev. Drug. Discov. 2013, 12, 703–719.
(11) Sun, Q.; Chen, T.; Wang, X.; Wei, X. Taxol induces paraptosis independent of
both protein synthesis and MAPK pathway. J. Cell. Physiol. 2010, 222, 421−432.
(12) Alexander, K.; Véronique, M.; Rastogi, S. K.; Florence, L.; Robert, K.
Therapeutic agents triggering nonapoptotic cancer cell death. J. Med. Chem. 2013, 56,
4823–4839.
(13) Palmeri, C. M.; Petiti, J. P.; Sosa, L. D. V.; Gutiérrez, S.; Paul, A. L. D.; Mukdsi,
J. H.; Torres, A. L. Bromocriptine induces parapoptosis as the main type of cell death
responsible for experimental pituitary tumor shrinkage. Toxicol. Appl. Pharm. 2009,
240, 55–65.
ACS Paragon Plus Environment

Page 21 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
(14) Eo, S. H.; Kim, J. H.; Kim, S. J. Induction of G2/M arrest by berberine via
activation of PI3K/Akt and p38 in human chondrosarcoma cell line. Oncol. Res. 2015,
22, 147–157.
9
(15) Kim, J. H.; Cho, E. B.; Lee, J.; Jung, O.; Ryu, B. J.; Kim, S. H.; Cho, J. Y.; Ryou,
C.; Lee, S. Y. Emetine inhibits migration and invasion of human nonꢀsmallꢀcell lung
cancer cells via regulation of ERK and p38 signaling pathways. Chem. Biol. Interact.
[Online early access]. DOI: 10.1016/j.cbi.2015.08.014. Published Online: Aug 30,
2015.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(16) Rao, K. N.; Venkatachalam, S. R. Dihydrofolate reductase and cell growth
activity inhibition by the βꢀcarbolineꢀbenzoquinolizidine plant alkaloid
deoxytubulosine from Alangium lamarckii: Its potential as an antimicrobial and
anticancer agent. Bioorg. Med. Chem. 1999, 7, 1105–1110.
(17) Lin, Z. M.; Guo, Y. X.; Gao, Y. H.; Wang, S. Q.; Wang, X. N.; Xie, Z. Y.; Niu, H.
M.; Chang, W. Q.; Liu, L.; Yuan, H. Q.; Lou, H. X. entꢀKaurane diterpenoids from
chinese liverworts and their antitumor activities through Michael addition as in situ
detected by a florescence probe. J. Med. Chem. 2015, 58, 3944–3456.
(18) Xie, Z.; Liu, L.; Chen, W.; Zheng, H.; Xu, Q.; Yuan, H.; Lou, H. Practical
metalꢀfree C(sp3)ꢀH functionalization: Construction of structurally diverse
αꢀsubstituted Nꢀbenzyl and Nꢀallyl carbamates. Angew. Chem. In. Ed. 2014, 53,
3985−3989.
(19) Wang, Y.; Wang, L.; Hu, Z.; Yuan, J.; Lin, Z.; Yuan, H.; Ji, M.; Lou, H. A novel
derivative of riccardin D induces cell death through lysosomal rupture in vitro and
inhibits tumor growth in vivo. Cancer Lett. 2013, 28, 207−216.
(20) Overmeyer, J. H.; Young, A. M.; Bhanot, H.; William, A. M. A chalconeꢀrelated
small molecule that induces methuosis, a novel form of nonꢀapoptotic cell death, in
glioblastoma cells. Mol. Cancer 2011, 10, 1−17.
(21) Giordani, R. B.; Vieira P. B.; Weizenmmann, M.; Rosemberg, D. B.; Souza, A.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 36
1
2
3
4
5
6
7
8
P.; Bonorino, C.; De Carli, G. A.; Bogo, M. R.; Zuanazzi, J. A.; Tasca, T.
Candimineꢀinduced cell death of the amitochondriate parasite Trichomonas vaginalis..
J. Nat. Prod. 2010, 73, 2019−2023.
9
(22) Beng, T. K.; Gawley, R. E. Highly enantioselective catalytic dynamic resolution
of NꢀBocꢀ2ꢀlithiopiperidine: Synthesis of (R)ꢀ(+)ꢀNꢀBocꢀpipecolic acid,
(S)ꢀ(−)ꢀconiine, (S)ꢀ(+)ꢀpelletierine, (+)ꢀβꢀconhydrine, and (S)ꢀ(−)ꢀropivacaine and
formal synthesis of (−)ꢀlasubine II and (+)ꢀcermizine C. J. Am. Chem. Soc. 2010, 132,
12216–12217.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(23) Sun, S.; Li, C.; Floreancig, P. E.; Lou, H.; Liu, L. Highly enantioselective
catalytic crossꢀdehydrogenative coupling of Nꢀcarbamoyl tetrahydroisoquinolines and
terminal alkynes. Org. Lett. 2015, 17, 1684−1687.
(24) Moreno, L.; Párraga, J.; Galán, A.; Cabedo, N.; Primo, J.; Cortes, D. Synthesis
of new antimicrobial pyrrolo[2,1ꢀa]isoquinolinꢀ3ꢀones. Bioorg. Med. Chem. 2012, 20,
6589−6597.
(25) Iwasa, K.; Moriyasu, M.; Tachibana, Y.; Kim, H. S.; Wataya, Y.; Wiegrebe, W.;
Bastow, K. F.; Cosentino, L. M.; Kozuka, M.; Lee, K. H. Simple isoquinoline and
benzylisoquinoline alkaloids as potential antimicrobial, antimalarial, cytotoxic, and
antiꢀHIV agents. Bioorg. Med. Chem. 2001, 9, 2871−2884.
(26) Ameri, K.; Harris, A. L. Activating transcription factor 4. Int. J. Biochem. Cell B.
2008, 40, 14−21.
(27) Tardito, S.; Isella, C.; Medico, E.; Marchiò, L.; Bevilacqua, E.; Hatzoglou, M.;
Bussolati, O.; FranchiꢀGazzola, R. The thioxotriazole copper (II) complex A0 induces
endoplasmic reticulum stress and paraptotic death in human cancer cells. J. Biol. Chem.
2009, 284, 24306−24319.
(28) Zhang, S.; Zhau, H. E.; Osunkoya, A. O.; Iqbal, S.; Yang, X.; Fan, S.; Chen, Z.;
Wang, R.; Marshall, F. F.; Chung, L. W. K.; Wu, D. Vascular endothelial growth factor
regulates myeloid cell leukemiaꢀ1 expression through neuropilinꢀ1ꢀdependent
ACS Paragon Plus Environment

Page 23 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
activation of cꢀMET signaling in human prostate cancer cells. Mol. Cancer 2010, 9, 9.
(29) Eva, S.; Susan, E. L.; Adrienne, M. G.; Afshin S. Mediators of endoplasmic
reticulum stressꢀinduced apoptosis. EMBO rep. 2006, 7, 880–885.
(30) Chen, H. Y.; Ren, X. Y.; Wang, W. H.; Zhang, Y. X.; Chen, S. F.; Zhang, B.;
Wang, L. X. Upregulated ROS production induced by the proteasome inhibitor
MGꢀ132 on XBP1 gene expression and cell apoptosis in Tcaꢀ8113 cells. Biomed.
Pharmacother. 2014, 68, 709–713.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(31) Fribley, A.; Zeng, Q.; Wang, C. Y. Proteasome inhibitor PSꢀ341 induces
apoptosis through induction of endoplasmic reticulum stressꢀreactive oxygen species
in head and neck squamous cell carcinoma cells. Mol. Cell. Biol. 2004, 24, 9695–
9704.
(32) Steffan, T. N.; Jennifer, S. C.; Kenneth, D. Jr.; Lawrence, H. B.; Paul, J. C.; Peng
H.; James, L. A.; David, J. M. Bortezomib inhibits PKRꢀLike endoplasmic reticulum
(ER) kinase and induces apoptosis via ER stress in human pancreatic cancer cells.
Cancer Res., 2005, 65, 11510–11519.
(33) Mizushima, N.; Yoshimori, T.; Levine, B. Methods in mammalian autophagy
research. Cell 2010, 140, 313–326.
(34) Kimura, S.; Noda, T.; Yoshimori, T. Dissection of the autophagosome
maturation process by a novel reporter protein, tandem fluorescentꢀtagged LC3.
Autophagy 2007, 3, 452–460.
(35) Kar, R.; Singha, P. K.; Venkatachalam, M. A. A novel role for map1 LC3 in
nonautophagic cytoplasmic vacuolation death of cancer cells. Oncogene 2009, 28,
2556−2568..
(36) Subramani, S.; Malhotra, V. Nonꢀautophagic roles of autophagyꢀrelated proteins.
EMBO Rep. 2013, 14, 143−151.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 36
Table 1. In vitro antiproliferative activity of benzo[a]quinolizidine derivatives against various human cancer cell lines.
1
IC50 (ꢁM)a
2
3
4
5
Compoundb
MDAꢀMBꢀ231
11.8±0.63
Hela
K562
PC3
DU145
MCFꢀ7
RWPEꢀ1
11.2±0.29
6.96±0.25
8.37±0.06
12.0±0.73
20.1±0.39
>40 (86.8)
1
6
7
8
9
Various 4ꢀsubstituted derivatives:
5.01 ±0.13
7.22±0.52
3.46±0.15
3.42±.025
1.06±0.18
2.66±0.08
5.27±0.28
2.43±0.43
5.49 ±0.28
2.38±0.26
5.44±0.21
5.42±0.53
3.24±0.22
3.26±0.37
1.52±0.34
3.81±0.47
1.41±0.07
3.88±0.36
2.32±0.02
NTc
NT
NT
NT
2.08±0.02
4.42±0.28
NT
5
1.15±0.22
3.28±0.36
7.22±0.40
1.15±0.02
10.2±0.01
2.88±0.54
6.72±0.14
2.57±0.31
1.51±0.12
1.16±0.07
7.67±0.45
2.81±0.54
3.52±0.24
2.15±0.36
NT
1.59±0.10
1.36±0.27
1.95±0.13
2.18±0.35
3.37±0.21
1.68±0.06
NT
2.00±0.18
0.28±0.01
2.50±0.25
2.20±0.09
3.02±0.28
2.79±0.44
NT
8b
10b
11a
11b
12a
12b
13a
13b
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1.84±0.03
NT
2.20±0.15
NT
NT
NT
NT
NT
Modification of Cꢀ2 position, addition and reduction:
5.48±0.52
6.47±0.59
2.37±0.19
6.57±0.81
5.21±0.36
6.34±0.58
1.98±0.16
NT
NT
2.60±0.28
NT
11.6±0.46
2.11±0.11
NT
9.2±0.33
1.39±0.16
NT
2.02±0.14
NT
16b
18
6.81±0.79
NT
19
Various 9,10ꢀbenzyloxy groups:
0.97±0.11
1.30±0.20
1.33±0.35
0.82±0.20
3.16±0.57
2.11±0.24
0.56±0.14
0.51±0.16
1.46±0.15
3.25±0.63
2.25±0.24
0.88±0.13
1.23±0.21
0.58±0.03
2.35±0.34
2.81±0.42
0.53±0.12
0.59±0.13
1.20±0.25
0.98±0.14
2.52±0.24
3.05±0.09
4.67±0.38
3.24±0.58
1.65±0.27
3.04±0.53
1.96±0.46
2.84±0.09
2.01±0.05
2.56±0.07
2.06±0.28
2.88±0.53
1.16±0.22
1.31±0.16
1.73±0.25
1.25±0.10
1.29±0.19
1.21±0.11
0.96±0.09
1.88±0.04
0.91±0.10
2.99±0.41
3.95±0.17
3.92±0.41
2.35±0.23
1.44±0.18
2.29±0.27
1.89±0.33
1.15±0.07
1.35±0.06
2.88±0.21
2.04±0.32
3.15±0.61
1.22±0.10
1.08±0.14
1.42±0.17
1.46±0.15
1.13±0.14
1.84±0.19
1.01±0.18
1.79±0.06
1.35±0.16
1.15±0.04
1.71±0.26
0.72±0.06
0.66±0.07
NT
4.53±0.57
3.96±0.22
1.51±0.14
1.96±0.23
NT
1.32±0.12
3.52±0.37
0.65±0.08
0.51±0.06
NT
3.04±0.18
2.24±0.04
1.35±0.04
5.06±0.29
NT
20b
21b
22a
22b
23a
23b
24a
24b
25a
25b
26a
26b
27a
27b
28
1.35±0.27
0.68±0.11
0.45±0.07
NT
1.88±0.25
1.87±0.36
1.74±0.04
NT
1.71±0.18
1.87±0.28
0.52±0.06
NT
4.42±0.19
2.32±0.31
1.15±0.10
NT
3.79±0.63
NT
2.22±0.17
NT
0.86±0.05
NT
1.91±0.16
2.89±0.30
NT
1.24±0.16
1.36±0.09
0.82±0.10
0.89±0.04
1.29±0.17
0.61±0.04
0.41±0.06
0.79±0.05
1.11±0.19
2.27±0.37
2.58±0.12
3.05±0.42
3.85±0.27
1.95±0.33
1.33±0.23
2.24±0.20
2.14±0.22
1.97±0.02
1.46±0.28
4.25±0.11
1.29±0.19
0.62±0.08
0.49±0.02
0.50±0. 03
1.98±0.06
0.81±0.07
0.47±0.03
0.84±0.15
2.61±0.44
2.23±0.39
4.45±0.21
1.49±0.14
4.77±0.22
4.45±0.38
2.57±0.47
1.86±0.18
2.95±0.61
4.03±0.16
NT
29
30
31
32
33
d
—
34
Octahydroazocino[2,1ꢀa]isoquinoline derivatives:
2.72±0.17
3.43±0.63
5.61±0.26
NT
NT
NT
NT
38b
Docetaxele
0.032±0.003
0.061±0.015 0.039±0.007 0.013±0.001 0.026±0.005 0.022±0.003
1.73±0.08
aResults are expressed as the mean IC50 of three experiments. The standard deviations were < 20% of the mean. bOther compounds
were inactive (IC50 > 20 ꢁM) or are shown in the SI. Not tested, Not active (IC50 > 20 ꢁM). Docetaxel was used as the positive
control.
c
d
e
ACS Paragon Plus Environment

Page 25 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Figure 1. Examples of bioactive compounds with specific quinolizidine or benzo[a]quinolizidine skeletons.11ꢀ15
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 36
1
2
3
4
5
6
7
Figure 2. Design and analysis of the potential bioactive positions of benzo[a]quinolizidine.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 27 of 36
Journal of Medicinal Chemistry
Scheme 1. General procedure for benzo[a]quinolizidine synthesis.a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aReagents and conditions: (a) paraformaldehyde, formic acid, 50 °C, 12 h. (b) diꢀtertꢀbutyl dicarbonate, Et3N, DMF or THF, r.t., 2–10
h. (c) HBr, reflux, 4 h. (d) benzyl bromide, K2CO3, DMF, r.t., 3 h. (e) TEMPO, EtOH, CH2Cl2, r.t., 24 h. (f) Yb(OTf)3, CH2Cl2, r.t., 2–
3 h. (g) TFA, CH2Cl2, 0 °C, 4 h. (h) NH3ꢂH2O/MeOH, 0 °C, 2 h.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Scheme 2. Synthesis of benzo[a]quinolizidine derivatives.a
Page 28 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aReagents and conditions: (a) Grignard reagent, THF, 0 °C, 1 h. (b) N2H4ꢂH2O, KOH, 80 °C to 180 °C, 5 h. (c) NaBH4, MeOH, 0 °C,
1 h. (d) Lꢀselectride, THF, ꢀ78 °C, 2 h.
ACS Paragon Plus Environment

Page 29 of 36
Journal of Medicinal Chemistry
Chart 1. Bioactive benzo[a]quinolizidine derivatives.a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aOnly the type b configuration was obtained for these compounds.
ACS Paragon Plus Environment