Re-engineering and synthesis of cytotoxic 2,3:7,8-di(alkylenedioxy)-extended analogs of quaternary sanguinarine chloride

Article abstract of DOI:10.1080/10286020.2018.1520705

A method was developed to synthesize 2,3:7,8-di(alkylenedioxy)-extended analogs of quaternary sanguinarine chloride. 1-Bromo-2-bromomethyl-3,4-alkylenedioxy benzenes and 6,7-alkylenedioxynaphthalen-1-amines were synthesized first. Reactions to construct the target compounds with these two series of synthons involved alterations on a published method for synthesizing 2,3,7,8-tetraoxygenated derivatives of benzo[c]phenanthridinium, substituting benzyl bromides for benzoic aldehydes, prolonging the radical annulation time, and conducting N-methylation with formic acid and NaBH₄. All the target compounds showed the same or better in vitro growth inhibitory activities against cancer cell lines compared with the positive compound. The structure activity relationship relevant to cytotoxicity and lipophilicity of the target compounds was produced.

Full text of DOI:10.1080/10286020.2018.1520705

Journal of Asian Natural Products Research
ISSN: 1028-6020 (Print) 1477-2213 (Online) Journal homepage: http://www.tandfonline.com/loi/ganp20
Re-engineering and synthesis of cytotoxic 2,3:7,8-
di(alkylenedioxy)-extended analogs of quaternary
sanguinarine chloride
Qi-Lin Li, An-Jun Deng, Ming Ji, Zhi-Hong Li, Xiao-Guang Chen & Hai-Lin Qin
To cite this article: Qi-Lin Li, An-Jun Deng, Ming Ji, Zhi-Hong Li, Xiao-Guang Chen & Hai-
Lin Qin (2018): Re-engineering and synthesis of cytotoxic 2,3:7,8-di(alkylenedioxy)-extended
analogs of quaternary sanguinarine chloride, Journal of Asian Natural Products Research, DOI:
10.1080/10286020.2018.1520705
To link to this article: https://doi.org/10.1080/10286020.2018.1520705
Published online: 11 Nov 2018.
Submit your article to this journal
Article views: 7
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ganp20

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
https://doi.org/10.1080/10286020.2018.1520705
Re-engineering and synthesis of cytotoxic 2,3:7,8-
di(alkylenedioxy)-extended analogs of quaternary
sanguinarine chloride
Qi-Lin Li, An-Jun Deng, Ming Ji, Zhi-Hong Li, Xiao-Guang Chen and Hai-Lin Qin
State Key Laboratory of Bioactive Substance and Function of Natural Medicines, Institute of Materia
Medica, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing
100050, China
ABSTRACT
ARTICLE HISTORY
A method was developed to synthesize 2,3:7,8-di(alkylenedioxy)-
extended analogs of quaternary sanguinarine chloride. 1-Bromo-2-
bromomethyl-3,4-alkylenedioxy benzenes and 6,7-alkylenedioxynaph-
thalen-1-amines were synthesized first. Reactions to construct the
target compounds with these two series of synthons involved altera-
tions on a published method for synthesizing 2,3,7,8-tetraoxygenated
derivatives of benzo[c]phenanthridinium, substituting benzyl bro-
mides for benzoic aldehydes, prolonging the radical annulation time,
Received 19 June 2018
Accepted 4 September 2018
KEYWORDS
Aryl alkylenedioxy synthons;
benzo[c]phenanthridiniums;
quaternary sanguinarine
analogs; synthetic method;
growth inhibitory activity;
cancer cell lines
and conducting N-methylation with formic acid and NaBH . All the
4
target compounds showed the same or better in vitro growth inhibi-
tory activities against cancer cell lines compared with the positive
compound. The structure activity relationship relevant to cytotoxicity
and lipophilicity of the target compounds was produced.
1. Introduction
Bioactive natural compounds have been playing great roles as arsenals of clinically
applicable drugs or their precursors for human race to treat cancers and other diseases
in the history of modern medicine. At present, a shortcut has been set up by medicinal
chemists to use bioactive natural compounds to develop innovative drugs through exten-
sive medicinal and chemical studies, the investigated natural compounds being labelled
as lead compounds. Of the well investigated lead compounds, alkaloids are an especially
brilliant class of compounds getting clinical application, like the famous colchicine, vin-
blastine, and taxol, and so on. In the process of investigating active natural alkaloids,
CONTACT Xiao-Guang Chen
chxg@imm.ac.cn; Hai-Lin Qin
qinhailin@imm.ac.cn
ß 2018 Informa UK Limited, trading as Taylor & Francis Group

2
Q.-L. LI ET AL.
benzo[c]phenanthridines, a well-known class of alkaloids belonging to the isoquinoline
alkaloids, attracted much attention from medicinal chemists too. Biological activities,
such as antimicrobes [1,2], anti-inflammations [3], and anti-tumors [4], and so on, were
corroborated in some lab tests on benzo[c]phenanthridines from natural resources.
Quaternary benzo[c]phenanthridin-5-ium is a common existence form of benzo[c]phe-
nanthridines in natural world. Some of them, such as quarternary nitidine [5], quater-
nary fagaronine [6], and quaternary chelerythrine [7], were also reported to display
powerful growth inhibitive properties against human cancer cell lines. The well-known
example is that the synthetic benzo[c]phenanthridine derivative NK-109 with the struc-
ture of 7-hydroxy-8-methoxy-5-methyl-2,3-methylenedioxybenzo[c]phenanthridin-5-ium
hydrogensulfate dihydrate showed significant cytotoxicity against human cancer cell
lines. The mechanism was known as inhibiting DNA topoisomerase II through the sta-
bilization of the cleavable complex [8]. Clinical studies on NK-109 confirmed its signifi-
cant anticancer activities [9]. More importantly, it worked well on several drug-resistant
cancer cell lines [10] and was reported to show greater anticancer activity than any
known benzo[c]phenanthridiniums [11]. Although, somehow, the development of this
compound for innovative anticancer drug was interrupted and no further report was
found recently, there have still been many studies in this area and several investigators
have reported interesting chemical findings and biological effects [12].
Sanguinarine is a natural 2,3:7,8-di(methylenedioxy)-5-methylbenzo[c]phenanthridin-5-
ium chloride, i.e. quaternary sanguinarine chloride (QSC). This natural compound exhib-
ited both very strong growth inhibitive activities against several human cancer cell lines
and anticancer activity pharmacologically [13,14]. In order to seek a new benzo[c]phe-
nanthridin-5-ium salt with anticancer property, and because of the similar structure to
NK-109 and the documented significant bioactivities of QSC, our group recently engaged
in investigations on structural modification of QSC. In our study, the construction of
aromatic compounds with alkylenedioxy of middle ring afforded two series of synthons.
Reactions to construct the target compounds from these two series of synthons were
explored in depth, with the referenced condition for synthesizing similar compounds
being altered to a certain extent. Several quaternary 2,3:7,8-di(alkylenedioxy)-5-methyl-
benzo[c]phenanthridin-5-ium chlorides were successfully synthesized. Biological screening
of the synthetic compounds shed light on the impact of modifying one, or simultan-
eously two, of the methylenedioxy moieties into alkylenedioxy of middle ring on the
growth inhibitive activities against human cancer cell lines. In addition, it should be indi-
cated that the 2,3:7,8-di(alkylenedioxy)-extended analogs of QSC, and those only modify-
ing one alkylenedioxy moiety, have not yet been isolated from any natural source and
therefore can only be obtained by synthesis. This study reports on the applicable syn-
thetic method, the synthesis of involved compounds, and the biological screening of the
end compounds for the growth inhibitory activities against five human cancer cell lines.
2. Results and discussion
þ
The quaternary cation of 5,6-imine salt form of sanguinarine (SG ) was reported to
be one of two key determinants for this type of natural products to exhibit anti-pro-
liferative activity [15]. The other is, according to common sense along with published

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
3
Figure 1. Strategy for synthesizing the 2,3:7,8-di(alkylenedioxy)-extended analogs of QSC.
articles, the solubility [16]. Studies have found that, under the physiological alkaline
þ
conditions, there is an equilibrium between the efficacious SG form and the ineffica-
cious pseudobase form (SGOH) of 2,3,7,8-tetraoxygenated 5-methyl-5,6-dihydroben-
zo[c]phenanthridin-6-ol, with SGOH even being the major form [15]. This
equilibrium heavily restricts the in vivo anticancer efficacy. Nowadays, techniques to
manage this type of equilibrium were reported. The increasing of solubility was
thought of as an effective way to slash the SGOH form in a given physiological con-
þ
ditions from the equilibrium, which increases the SG form [16]. Thus, in this paper,
the target derivatives of QSC were designed according to two principles, i.e. both to
retain the efficacious quaternary iminium cation form and to increase the solubility,
albeit liposolubility.
To our knowledge, no research paper of squarely altering the 2,3- and 7,8-methyle-
nedixoy groups of QSC was published up to our investigation. Still, substituting dif-
ferent alkylenedioxy groups of structural diversities for one, or simultaneously two, of
the 2,3- and 7,8-methylenedioxy moieties in QSC would definitely cause the differen-
ces of molecular planarity among the different products. In some cases, based on the
pharmacochemical principles, even a little tortuosity of the molecular structure would
lead to great changes of bioactivities and other drug abilities. Thus, a series of QSC
analogs which substituted one, or simultaneously two, target alkylenedioxy groups for
the native 2,3- and 7,8-methylenedixoy groups of the lead compound were designed,
with successful syntheses being achieved in this study. For a systematic goal to screen
the growth inhibitory activity against the target cancer cell lines, the considered target
alkylenedioxy groups involved 1,2-dimethylenedioxy, 1,3-trimethylenedioxy, and 1,4-
tetramethylenedioxy. Apparently, the construction for two series of aryl alkylenedioxy
compounds of structural diversity, which was related to the two methylenedioxy
groups of QSC, as key synthons of synthesizing the designed compounds was two of
the crucial works for this study, although it was proved to be very comfortable
(Figure 1 and Schemes 1 and 2). The final route to synthesize the target compounds
on these two series of synthons was largely modeled after the reported synthetic route
for NK-109 [11], but some synthetic conditions were altered slightly or somewhat sig-
nificantly to improve the effectiveness and applicability of the method.
As indicated above, the syntheses of 1-bromo-2-bromomethyl-3,4-alkylenedioxy
benzenes and 6,7-alkylenedioxy naphthalen-1-amines as two crucial series of synthons
to synthesize the designed compounds involved only regular synthetic methods in
organic chemistry. The commercially obtainable 6-bromo-2,3-dihydroxybenzaldehyde
(1) was reacted with 1,x-dibromo-n-alkanes via a process of dioxy-de-dibromo

4
Q.-L. LI ET AL.
Scheme 1. Syntheses of 1-bromo-2-bromomethyl-3,4-alkylenedioxy benzenes (4a-d). Reagents and
ꢀ
ꢀ
conditions: (1) K CO , CuO, DMF, 110 C; (2) NaBH , MeOH, 0 C, HCl; (3) PBr , Py, DCM, rt.
2
3
4
3
Scheme 2. Syntheses of 6,7-alkylenedioxy naphthalen-1-amines (7a-d). Reagents and conditions:
ꢀ
(
1) K CO , CuO, DMF, 110 C; (2) H /Pd, MeOH, rt.
2
3
2
nucleophilic substitution under alkaline condition of K CO in DMF to afford the
2
3
6-bromo-2,3-alkylenedioxybenzaldehydes (2a-d) in yields between 51.0% and 83.0%.
The reduction of 6-bromo-2,3-alkylenedioxybenzaldehydes (2) using NaBH4 in
MeOH yielded 6-bromo-2,3-alkylenedioxybenzoic alcohol (3a-d) in yields between
7
8.7% and 86.3%. Substituting bromine for the hydroxyl group using PBr and pyri-
3
dine as agents in DCM solvent afforded the former series of synthons (4a-d) in yields
between 83.8% and 92.0% (Scheme 1). The commercially obtainable 5-nitronaphtha-
lene-2,3-diol (5) was reacted with 1,x-dibromo-n-alkanes via a process similar to the
aforementioned synthetic route for 2 to afford the 5-nitro-2,3-alkylenedioxynaphtha-
lenes (6a-d) in yields between 39.4% to 66.7%. In the synthetic process, it was noticed
that the yield of the desired 5-nitro-2,3-alkylenedioxynaphthalenes was influenced by
the relative ratio of 1,x-dibromo-n-alkanes. The open-and-shut example was the syn-
thesis of 8-nitro-2,3,4,5-tetrahydronaphtho[2,3-b][1,4]dioxocine (6d). When the equiv
of 1,4-dibromobutane was increased from 1.5 to 4.0 to promote the reaction balance,
an obvious impurity was spotted at the end of the reaction via thin-layer chromatog-
raphy (TLC) monitoring which was identified later as 6,7-bis(4-bromobutoxy)-1-
nitronaphthalene. Thus, 1.5 equiv of 1,4-dibromobutane was adopted to synthesize 6d
based on extensive exploration, with the desired compound being obtained as pure
compound conveniently. The reduction of 6 using H /Pd produced 6,7-alkylenedioxy
2
naphthalen-1-amines (7a-d) as the latter series of synthons in yields between 85.2%
to 92.0% (Scheme 2).
As phenanthridines make up a well-known class of molecules with established bio-
logical properties, there has been many protocols being reported to construct the ben-
zo[c]phenanthridine core, such as Bischler-Napieralski route by Ishii [17], Heck route
by Harayama [18], intramolecular homolytic aromatic substitution route by Bisai
[
19], Dual Palladium-catalyzed route by Maestri [12], Benzyne cyclization route by
Kessar [20], electron-demand [4 þ 2] aza-Diels-Alder cycloaddition of electron-rich

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
5
N-aryl imines with arynes route by Castillo [21], and Nakanishi’s route for synthesiz-
ing NK-109 [11], and so on. These routes were applicable to synthesizing native san-
guinarine and other explored natural benzo[c]phenanthridines, but none of them was
proved tolerant enough of great structural changes in the homologs and relevant tar-
get compounds, the aforementioned reports being limited to several kinds of substitu-
ent pattern. It is well-known that, sometimes, a little structural change of natural
organic molecules, even within homologs, would call for a new synthetic route or a
changed route to a considerable degree. The main goal of the current study was to
construct the aryl alkylenedioxy compounds of structural diversity and, based on the
construction of these synthons, to re-engineer the 2,3:7,8-di(alkylenedioxy) moieties
of active QSC in a total synthesis strategy to evaluate the SAR of certain bioactivity.
For the building of a synthesized library of compounds to screen bioactivities, toler-
ance of the synthetic route on structurally diverse homologous compounds or end
compounds is quite important. Thus, after an extensive and in-depth exploration on
the published routes, including careful test via practices as such, the radical annula-
tion route, which Nakanishi adopted to synthesize NK-109 and proved to be quite
tolerant of providing kilograms of NK-109 for clinical research, was chosen as a key
step to synthesize the end compounds of our study.
In the process of applying the published synthetic route to the goal of this study,
some conditions were re-optimized considering the findings in practice. First of all,
for the tertiary imine formation and reductive amination steps, the expected conver-
sion from the 6-bromo-2,3-alkylenedioxybenzaldehydes (2a-d) and 6,7-alkylenedioxy-
naphthalen-1-amines (7a-d) to the relevant compounds was affirmed to be erratic
when used the published method without any change. It was supposed that the elec-
tron-donating alkylenedioxy substituents on the benzoic aldehydes of 2a-d resulted in
a more drastic reaction situation as compared with the published synthesis using
6-bromo-2-hydroxy-3-methoxybenzaldehyde as substrate, and, as a general principle,
the produced unstable imines would take on a complex situation. Thus, the synthetic
route of our study was reengineered to substitute bromomethyl for the formyl moiety
on the starting materials to explore the applicability of nucleophilic substitution. In
the practice, this alteration was affirmed to yield the stable N-benzylnaphthalen-1-
amines directly (8a-f). As such, the reductive step of using expensive and smell-
unpleasant dimethylamine-borane as reagent in the published method was success-
fully avoided which significantly simplify the synthetic route and cut the cost.
Secondly, the radical annulation step of the reported route to synthesize NK-109 was
found to yield a mixture of two cyclization products in this specific study, i.e. the
spontaneously oxidated N-demethylated analogs and the N-demethylated 5,6-dihydro-
type analogs of the end compounds, although without using any oxidizing reagents.
This was an interesting phenomenon and, after a research of literatures, it was
noticed that this oxidation phenomenon had been observed when synthesizing NK-
109 with unprotected phenol. It was hypothesized that this kind of spontaneous oxi-
dation was caused by binding of trialkyltin moiety to hydroxyl oxygen with subse-
quent releasing of trialkyltin hydride in the process of synthesis, and the resonance
energy is probably the driving force of this spontaneous oxidation [22]. Although,
based on organic principle, when the hydroxyl and methoxy groups were replaced by

6
Q.-L. LI ET AL.
Scheme 3. Syntheses of 2,3:7,8-dialkylenedioxybenzo[c]phenanthridines (9a-f). Reagents and condi-
ꢀ
ꢀ
tions: (1) K CO , DMF, 110 C; (2) Bu SnH, toluene, N , AIBN, 110 C.
2
3
3
2
alkylenedioxy groups, the spontaneous oxidation effect would still exist in the synthetic
process due to the similar lone pair of electrons on their oxygen atom, the results of the
current study only indicated that it was relatively weaker than the case in the literature.
So the reaction time of radical annulation had to be elongated, to 16 h in our specific
experiment, to fully conduct the spontaneous oxidation, and the N-demethylated analogs
(9a-f) of the target compounds were successfully obtained as lone compounds in each
synthesis (Scheme 3). Thirdly, the N-methylation step in the published route, i.e. react-
ꢀ
ing at 110 C for 35 h under the presence of the custom-made methyl o-nitrobenzenesul-
fonate as reagents to yield the end compounds, made us daunted. Thus, the Ishii’s
versatile method to use formic acid and NaBH as methylation reagents was adopted to
4
complete the methylation reaction within 0.5 h [23], and the altered synthetic route was
affirmed to be a significant success (Scheme 4), leading to the syntheses of dihydro-type
intermediates (10a-f) of the target compounds. Finally, oxidative aromatization of these
dihydro-type intermediates using DDQ as reagent, and then salinization reaction using
HCl, successfully yielded the target compounds (11a-f).
1
All the synthesized end compounds were characterized using H NMR method
along with the consideration of reactions. Two sets of aromatic AB-type spin-spin
3
coupling signals in the resonance region of dH 8.01–8.82 with J values of around
9
.0 Hz assignable to CH(9)¼CH(10) and CH(11)¼CH(12), respectively, from the two
vic-tetrasubstituted benzene rings and three isolated aromatic singlets containing one
aromatic proton each in the resonance region of d 7.81–10.15 assignable to CH(1),
H
CH(4), and CH(6), respectively, were apparent, confirming the core structure of
2,3:7,8-tetrasubstituted benzo[c]phenanthridine core. One singlet of methyl group at
d 4.95–5.03 assignable to nitromethyl group linked to N-5 established the 5-methyl-
H
1
benzo[c]phenanthridin-5-ium structure. All the remaining signals in the H NMR
spectra of each end compounds were assigned to the modified aryl alkylenedioxy
moieties and were all compatible with the relevant structures (see supplementary data
13
for detailed spectra and data). In addition, the recorded C NMR spectra for typical
þ
representatives of the end compounds and the ESIMS data of all the synthesized tar-
get compounds conformed to their respective structures (see Experimental section).
The effect of substituting only one, or simultaneously two, alkylenedioxy groups of
structural diversities for the two 2,3- and 7,8-methylenedixoy groups of the lead com-
pound on the growth inhibitory activities against human cancer cell lines was eval-
uated using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide
(MTT) assay. QSC and the self-synthesized NK-109 were screened in the same batches

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
7
Scheme 4. Syntheses of quaternary 2,3:7,8-dialkylenedioxy-5-methylbenzo[c]phenanthridin-5-ium
ꢀ
chlorides (11a-f). Reagents and conditions: (1) HCOOH, NaBH , 40 C, NH Cl; (2) DDQ, NaOH, HCl.
4
4
of experiments as reference substances and positive controls. The explored human can-
cer cell lines included HCT-116, HepG2, BGC-823, NCI-H1650, and A2780. On evalu-
ation, all target cancer cells were treated successively using every tested compound for
96h, respectively. The MTT reduction assay procedure was modeled after our earlier
study [24]. Results of means of three replicates presented in Table 1 were expressed as
the concentration for compounds to inhibit the cell growth by 50%, i.e. the IC50 values.
All the synthesized 2,3:7,8-di(alkylenedioxy)-extended analogs of QSC (11b-11f) dis-
played significant activities against the five tested human cancer cell lines, with IC50 val-
ues ranging from <0.01 to 3.54 lM. And the activities of analog 11e, which contains
2,3:7,8-di(1,3-trimethylenedioxy) groups in its structure, were obviously stronger than
that of NK-109 as well as all the other target compounds, with its IC5 values for all the
0
7
ꢁ
five explored cancer cell lines reaching or exceeding the level of 10 M. Although the
IC50 values listed in Table 1 demonstrated that no significant SAR involved only in
structures was observed amongst the tested target compounds based on the limited alky-
lenedioxy groups in this study, the aforementioned findings demonstrated another point
that extending the size of the two 2,3- and 7,8-alkylenedixoy groups of the lead com-
pound improved or significantly improved the inhibitive activities against the explored
cancer cell lines. Thus, more quaternary 2,3:7,8-dialkylenedioxy-5-methylbenzo[c]phe-
nanthridin-5-ium chlorides, and other counteranion, will be synthesized in our subse-
quent planned study to explore the exact SAR as in-depth medicinal chemistry study.
Obviously, all the explored alkylenedioxy groups were known for the relatively similar
lipophilicity. Therefore, the finding about SAR concerning improving activity by increas-
ing lipophilicity was produced, i.e. the conclusion of the current study is in agreement
with previous findings about the influence of lipophilicity. It is worth mentioned that
the inhibitive activity of 11e against the human A2780 cervical cancer cell line reached
<
10 nM level, which provided us great confidence in our subsequent synthesis and
modification works.
To sum up, a versatile and convenient method for synthesizing the 2,3:7,8-di(alky-
lenedioxy)-extended analogs of cytotoxic QSC and, as expansion, the aryl alkylene-
dioxy compounds, were developed. 1-Bromo-2-bromomethyl-3,4-alkylenedioxy
benzenes and 6,7-alkylenedioxynaphthalen-1-amines as two crucial series of synthons
were synthesized using regular synthetic methods of organic chemistry. These two
series of synthons were condensed under the presence of K CO as base and DMF as
2
3
ꢀ
solvent at 110 C via a nucleophilic substitution reaction. Without purification step, the

8
Q.-L. LI ET AL.
Table 1. Cancer cells growth inhibitory activity of the target compounds.
IC50 (lM)
a
a
a
a
a
Cpds
R
1
R
2
C1
C2
C3
C4
C5
b
P
QSC
OMeþOH
CH
CH
2
3.94
2.98
2.66
3.54
1.27
0.66
2.38
1.19
2.62
3.01
3.11
1.14
0.32
1.95
1.84
2.16
3.39
2.74
1.75
0.98
3.40
1.75
2.49
2.69
2.72
1.57
0.94
1.94
0.90
2.38
2.00
2.13
0.67
<0.01
1.40
CH
CH
(CH
(CH
(CH
(CH
2
2
2
11b
11c
11d
11e
11f
a
(CH
(CH
(CH
(CH
(CH
2
2
2
2
2
)
3
)
4
)
2
)
3
)
2
2
2
2
2
)
)
)
)
2
3
3
4
C1: HCT-116; C2: HepG2; C3: BGC-823; C4: NCI-H1650; C5: A2780.
b
P: NK109.
crude intermediates in the form of reaction mixture were cyclized through radical annu-
lation reaction under the presence of AIBN and Bu SnH as radical initiator, with the
3
tertiary imine-type 2,3:7,8-di(alkylenedioxy)benzo[c]phenanthridines, i.e. the N-demethy-
lated compounds of the desired target products, being obtained in yields of between
23.7% and 38.5% from 6,7-alkylenedioxynaphthalen-1-amines. The end products of qua-
ternary 2,3:7,8-di(alkylenedioxy)-5-methylbenzo[c]phenanthridin-5-ium chlorides were
successfully synthesized through N-methylation step with formic acid in the presence of
NaBH4 as reductive reagents, aromatization step with alkaline DDQ as oxidative
reagents, and, finally, salinization reaction step using HCl, in yields ranging from 35.6%
to 65.1% from N-demethylated compounds. Aryl methylenedioxy group is a common
structural moiety in many active natural compounds, such as quaternary coptisine and
QSC. To our knowledge, the syntheses of various benzo[b][1,4]dioxacycloalkanes and
aminonaphtho[2,3-b][1,4]dioxacycloalkanes of beyond six-membered ring for the dioxa-
rings have not been reported up to this investigation. This investigated method of syn-
thesizing aryl alkylenedioxy compounds of structural diversity, and the 2,3:7,8-di(alkyle-
nedioxy)-extended analogs of QSC, is predicted to be applicable to the expeditious
synthesis of relevant compounds arsenals in the field of natural medicinal chemistry. All
the synthesized 2,3:7,8-di(alkylenedioxy)-extended analogs of QSC (11b-11f) displayed
significant activities against the five tested human cancer cell lines with IC50 values rang-
ing from <0.01 to 3.54 lM, the end compound 11e showing the most significant activity
against A2780 by the IC50 value at <10 nM level.
3. Experimental
3.1. General experimental procedures
Nuclear magnetic resonance (NMR) spectra were recorded on either a Varian
Mercury-300 NMR spectrometer or a Varian Mercury-400 NMR spectrometer or a
Bruker AV-III-500 NMR spectrometer and reported with tetramethylsilane (TMS) as
an internal standard and chloroform-d (CDCl ) (D, 99.8% þ 0.05% v/v TMS) or
3
dimethyl sulfoxide-d (DMSO-d ) (D, 99.9% þ 0.05% v/v TMS) (Cambridge Isotope
6
6
Laboratories, Inc., Andover, MA, USA) as solvents. Chemical shifts (d values) and
þ
coupling constants (J values) are given in ppm and Hz, respectively. ESIMS were
obtained using an Agilent 1100 series LC/MSD Trap SL mass spectrometer. All the
reagents and solvents were reagent grade or were purified by standard procedures

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
9
before using. The reaction progress was monitored using TLC on glass plates pre-
coated with silica gel GF254 (Qindao Haiyang Chemical, Qingdao, China). The spots
were visualized under UV light. Column chromatography (CC) was carried out over
silica gel (200–300 mesh size; Qingdao Haiyang Chemical, Qingdao, China). The con-
centration of solution after reactions involved the use of a rotary evaporator operated
at a reduced pressure of ca. 9.0 mbar.
3
.2. Synthetic procedures
3
3
.2.1. Synthesis of 6-bromo-2,3-alkylenedioxybenzaldehydes (2a-d)
.2.1.1. Synthesis of 5-bromobenzo[d][1,3]dioxole-4-carbaldehyde (2a). To a mix-
ture containing 6-bromo-2,3-dihydroxybenzaldehyde (1) (1.85 g, 8.5 mmol), K CO3
2
(
(
2.35 g, 17.0 mmol), and CuO (0.1 g) in DMF (30 ml) was added dibromomethane
1.19 ml, 17.0 mmol). The reaction mixture was stirred at 110 C, with the monitoring
ꢀ
being carried out using TLC inspection until it was apparent that the reaction was
known for completion. The reaction mixture was cooled to rt. and then filtered to
remove K CO and CuO. After the filtrate was poured into the solvent of H O
2
3
2
(150 ml), a black precipitate appeared which was collected via filtrating. The filter
cake was recrystallized using i-PrOH as a crystalline solvent, with yellow powder (2a,
1
1
1
.0 g) being obtained in a yield of 51% from 1. H NMR (DMSO-d , 300 MHz) d:
6
þ
0.11 (s, 1H), 7.18 (d, J ¼ 8.2 Hz, 1H), 7.09 (d, J ¼ 8.2 Hz, 1H), 6.21 (s, 2H); ESIMS
þ
m/z 228.9 [M þ H] .
3
.2.1.2. Synthesis of 6-bromo-2,3-dihydrobenzo[b][1,4]dioxine-5-carbaldehyde (2b).
Compound 2b was obtained as yellow amorphous powder from 1 (1.85 g, 8.5 mmol),
K CO (2.35 g, 17.0 mmol), CuO (0.1 g), and 1,2-dibromoethane (1.46 ml, 17 mmol)
2
3
1
using a similar procedure to that of 2a in a yield of 53.1% from 1. H NMR (DMSO-
d , 400 MHz) d: 10.22 (s, 1H), 7.17 (d, J ¼ 8.7 Hz, 1H), 7.06 (d, J ¼ 8.7 Hz, 1H),
6
þ
þ
4
.39–4.28 (m, 4H); ESIMS m/z 244.1 [M þ H] .
3
.2.1.3. Synthesis of 7-bromo-3,4-dihydro-2H-benzo[b][1,4]dioxepine-6-carbalde-
hyde (2c). Compound 2c was obtained as yellow amorphous powder from 1 (1.85 g,
8
1
.5 mmol), K CO (2.35 g, 17.0 mmol), CuO (0.1 g), and 1,3-dibromopropane (1.73 ml,
2 3
7 mmol) using a similar procedure to that of 2a in a yield of 83.0% from 1. H
1
NMR (CDCl , 400 MHz) d: 10.31 (s, 1H), 7.17 (d, J ¼ 8.6 Hz, 1H), 6.99 (d, J ¼ 8.6 Hz,
3
þ
1
2
H), 4.33 (t, J ¼ 5.7 Hz, 2H), 4.25 (t, J ¼ 5.7 Hz, 2H), 2.31–2.19 (m, 2H); ESIMS m/z
þ
58.1 [M þ H] .
3
.2.1.4. Synthesis of 8-bromo-2,3,4,5-tetrahydrobenzo[b][1,4]dioxocine-7-carbalde-
hyde (2d). Compound 2d was obtained as yellow amorphous powder from 1 (1.85 g,
8
1
.5 mmol), K CO (2.35 g, 17.0 mmol), CuO (0.1 g), and 1,4-dibromobutane (2.03 ml,
2 3
7 mmol) using a similar procedure to that of 2a in a yield of 60% from 1. H NMR
1
(
4
2
CDCl , 400 MHz) d: 10.35 (s, 1H), 7.32 (d, J ¼ 8.8 Hz, 1H), 6.93 (d, J ¼ 8.8 Hz, 1H),
3
.13 (t, J ¼ 7.5 Hz, 2H), 4.05 (t, J ¼ 5.3 Hz, 2H), 2.25–2.12 (m, 2H), 2.02–1.88 (m,
þ þ
H); ESIMS m/z 272.1 [M þ H] .

1
0
Q.-L. LI ET AL.
3
3
.2.2. Syntheses of 6-bromo-2,3-alkylenedioxybenzoic alcohols (3a-d)
.2.2.1. Synthesis of (5-bromobenzo[d][1,3]dioxol-4-yl)methanol (3a). To a solution
ꢀ
of 2a (5.0 g, 21.8 mmol) in MeOH (80 ml) at 0 C was added NaBH (0.83 g,
4
ꢀ
21.8 mmol) batchwise. The reaction mixture was stirred at 0 C for 0.5 h, with the
completion of the reductive reaction being indicated by TLC inspection. The reaction
was quenched via adding 10% HCl aq. (300 ml) into the reaction mixture. The reac-
tion mixture was extracted three times with EtOAc (100 ml, each time). The organic
layer was combined and washed using, first pure H O (100 ml) and then brine
2
(
100 ml), and dried over anhydrous MgSO and filtered. The filtrate was concentrated
4
under reduced pressure to yield a crude product, which was purified using CC, eluted
using a 10% EtOAc in PE (v/v) isocratic elution to yield a white amorphous powder
1
(
6
3a, 4 g) in a yield of 78.7%. H NMR (CDCl , 500 MHz) d: 7.04 (d, J ¼ 8.3 Hz, 1H),
3
.80 (d, J ¼ 8.3 Hz, 1H), 6.06 (s, 2H), 5.07 (t, J ¼ 4.4 Hz, 1H), 4.45 (d, J ¼ 4.4 Hz, 2H).
3
.2.2.2. Synthesis of (6-bromo-2,3-dihydrobenzo[b][1,4]dioxin-5-yl)methanol (3b).
Compound 3b was obtained as white amorphous powder from 2b (1.5 g, 6.2 mmol)
and NaBH (0.23 g, 6.2 mmol) using a similar procedure to that of 3a in a yield of
4
1
8
1
6.7%. H NMR (CDCl , 500 MHz) d: 7.04 (d, J ¼ 8.7 Hz, 1H), 6.73 (d, J ¼ 8.7 Hz,
3
H), 4.82 (s, 2H), 4.35–4.30 (m, 2H), 4.27–4.24 (m, 2H).
3
(
1
.2.2.3. Synthesis of (7-bromo-3,4-dihydro-2H-benzo[b][1,4]dioxepin-6-yl)methanol
3c). Compound 3c was obtained as white amorphous powder from 2c (5.0 g,
9.4 mmol) and NaBH (0.74 g, 19.4 mmol) using a similar procedure to that of 3a in
4
1
a yield of 85.1%. H NMR (CDCl , 400 MHz) d: 7.17 (d, J ¼ 8.7 Hz, 1H), 6.87 (d,
3
J ¼ 8.7 Hz, 1H), 4.83 (s, 2H), 4.31 (t, J ¼ 5.6 Hz, 2H), 4.20 (t, J ¼ 5.5 Hz, 2H),
2.27–2.18 (m, 2H).
3
.2.2.4. Synthesis of (8-bromo-2,3,4,5-tetrahydrobenzo[b][1,4]dioxocin-7-yl)metha-
nol (3d). Compound 3d was obtained as white amorphous powder from 2d (7.3 g,
6.9 mmol) and NaBH (1.0 g, 26.9 mmol) using a similar procedure to that of 3a in a
2
4
1
yield of 86.3%. H NMR (CDCl , 500 MHz) d: 7.15 (d, J ¼ 8.7 Hz, 1H), 6.84 (d,
3
J ¼ 8.7 Hz, 1H), 4.80 (s, 2H), 4.46 (t, J ¼ 5.2 Hz, 2H), 4.29 (t, J ¼ 4.8 Hz, 2H),
1.97–1.87 (m, 4H).
3
3
.2.3. Syntheses of 1-bromo-2-bromomethyl-3,4-alkylenedioxy benzenes (4a-d)
.2.3.1. Synthesis of 5-bromo-4-bromomethylbenzo[d][1,3]dioxole (4a). To a mix-
ture containing 3a (1.50 g, 6.5 mmol) and pyridine (439 ll, 5.5 mmol) in DCM (20 ml)
ꢀ
at 0 C was added PBr (514 ll, 5.4 mmol) dropwise. The reaction mixture was stirred
3
at rt. for 1 h, the reaction being known for completion by TLC monitoring. After
extra DCM (20 ml) was added, the reaction mixture was washed two times using
brine (100 ml, each time) and then dried over anhydrous MgSO and filtered. The fil-
4
trate was concentrated under reduced pressure to yield a crude product which was
purified via CC, eluted using a 10% EtOAc in PE (v/v) isocratic elution, to yield
1
white amorphous powder (4a, 1.6 g) in a yield of 83.8% from 3a. H NMR (CDCl ,
3

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
11
5
00 MHz) d: 7.09 (d, J ¼ 8.3 Hz, 1H), 6.70 (d, J ¼ 8.3 Hz, 1H), 6.11 (s, 2H), 4.61
(s, 2H).
3
.2.3.2. Synthesis of 6-bromo-5-bromomethyl-2,3-dihydrobenzo[b][1,4]dioxine (4b).
The synthon 4b was obtained as white amorphous powder from 3b (1.30 g,
.3 mmol), pyridine (359 ll, 4.5 mmol) , and PBr (420 ll, 4.4 mmol) using a similar
5
3
1
procedure to that of 4a in a yield of 92.0%. H NMR (DMSO-d , 300 MHz) d: 7.09
6
(
(
d, J ¼ 8.8 Hz, 1H), 6.84 (d, J ¼ 8.8 Hz, 1H), 4.61 (s, 2H), 4.44–4.30 (m, 2H), 4.30–4.22
m, 2H).
3
.2.3.3. Synthesis of 7-bromo-6-bromomethyl-3,4-dihydro-2H-benzo[b][1,4]dioxe-
pine (4c). The synthon 4c was obtained as white amorphous powder from 3c
14.10 g, 54.4 mmol), pyridine (3.7 ml, 46.3 mmol), and PBr (4.3 ml, 45.3 mmol) using
(
3
1
a similar procedure to that of 4a in a yield of 87.2%. H NMR (CDCl , 400 MHz) d:
3
7
4
.14 (d, J ¼ 8.7 Hz, 1H), 6.85 (d, J ¼ 8.7 Hz, 1H), 4.71 (s, 2H), 4.31 (t, J ¼ 5.6 Hz, 3H),
.20 (t, J ¼ 5.5 Hz, 3H), 2.23 (tt, J ¼ 5.6 Hz, 5.5 Hz, 2H).
3
.2.3.4. Synthesis of 8-bromo-7-bromomethyl-2,3,4,5-tetrahydrobenzo[b][1,4]dioxo-
cine (4d). The synthon 4d was obtained as white amorphous powder from 3d (7.50 g,
7.5 mmol), pyridine (1.9 ml, 23.4 mmol), and PBr (2.2 ml, 22.9 mmol) using a simi-
2
3
1
lar procedure to that of 4a in a yield of 87.2%. H NMR (DMSO-d , 300 MHz) d:
6
7
4
.22 (d, J ¼ 8.7 Hz, 1H), 6.94 (d, J ¼ 8.7 Hz, 1H), 4.66 (s, 2H), 4.38 (t, J ¼ 5.1 Hz, 2H),
.23 (t, J ¼ 5.1 Hz, 2H), 2.04–1.65 (m, 4H).
3
3
.2.4. Syntheses of 5-nitro-2,3-alkylenedioxynaphthalenes (6a-d)
.2.4.1. Synthesis of 5-nitronaphtho[2,3-d][1,3]dioxole (6a). To a mixture containing
5
-nitronaphthalene-2,3-diol (5) (2.40 g, 11.7 mmol), K CO (3.23 g, 23.4 mmol), and
2 3
CuO (0.2 g) in DMF (30 ml) was added dibromomethane (1.6 ml, 23.4 mmol) drop-
ꢀ
wise. The reaction mixture was stirred at 110 C for 3 h until the reaction was known
via TLC monitoring for completion. The reaction mixture was cooled to rt. and then
filtered to remove K CO and CuO. After the filtrate was poured into the solvent of
2
3
H O (150 ml), a black precipitate appeared which was collected via filtrating. The fil-
2
ter cake was recrystallized using i-PrOH as a crystalline solvent, with pale yellow
1
amorphous powder (6a, 1.4 g) being obtained in a yield of 56.0%. H NMR (CDCl ,
3
4
00 MHz) d: 8.11 (d, J ¼ 7.8 Hz, 1H), 7.97 (s, 1H), 7.90 (d, J ¼ 8.1 Hz, 1H), 7.37 (dd,
þ þ
J ¼ 7.8, 8.1 Hz, 1H), 7.19 (s, 1H), 6.13 (s, 2H); ESIMS m/z 218.2 [M þ H] .
3
.2.4.2. Synthesis of 6-nitro-2,3-dihydronaphtho[2,3-b][1,4]dioxine (6b). Compound
6
b was obtained as yellow amorphous powder from 5 (2.40 g, 11.7 mmol), K CO3
2
(
3.23 g, 23.4 mmol), CuO (0.2 g), and dibromoethane (2.0 ml, 23.4 mmol) using a
1
similar procedure to that of 6a in a yield of 66.7%. H NMR (CDCl , 500 MHz) d:
3
8
4
.25–8.15 (m, 2H), 7.96 (d, J ¼ 8.1 Hz, 1H), 7.36 (s, 1H), 7.33 (d, J ¼ 7.9 Hz, 1H),
þ þ
.52–4.37 (m, 4H); ESIMS m/z 232.2 [M þ H] .

1
2
Q.-L. LI ET AL.
3.2.4.3. Synthesis of 7-nitro-3,4-dihydro-2H-naphtho[2,3-b][1,4]dioxepine (6c).
Compound 6c was obtained as yellow amorphous powder from 5 (2.40 g, 11.7 mmol),
K CO (3.23 g, 23.4 mmol), CuO (0.2 g), and dibromopropane (2.4 ml, 23.4 mmol)
2
3
1
using a similar procedure to that of 6a in a yield of 59.2%. H NMR (DMSO-d ,
6
4
1
2
00 MHz) d: 8.25 (d, J ¼ 8.0 Hz, 1H), 8.22 (d, J ¼ 8.0 Hz, 1H), 7.99 (s, 1H), 7.75 (s,
H), 7.55 (dd, J ¼ 8.0, 8.0 Hz, 1H), 4.32 (t, J ¼ 5.6, Hz, 2H), 4.30 (t, J ¼ 5.6, Hz, 2H),
þ
þ
.22 (tt, J ¼ 5.6, 5.6 Hz, 2H); ESIMS m/z 246.2 [M þ H] .
3
.2.4.4. Synthesis of 8-nitro-2,3,4,5-tetrahydronaphtho[2,3-b][1,4]dioxocine (6d).
Compound 6d was obtained as yellow amorphous powder from 5 (2.40 g, 11.7 mmol),
K CO (3.23 g, 23.4 mmol), CuO (0.2 g), and dibromobutane (2.1 ml, 17.5 mmol)
2
3
1
using a similar procedure to that of 6a in a yield of 39.4%. H NMR (DMSO-d ,
6
4
1
1
00 MHz) d: 8.21 (d, J ¼ 8.0 Hz, 1H), 8.18 (d, J ¼ 8.0 Hz, 1H), 7.98 (s, 1H), 7.74 (s,
H), 7.51 (dd, J ¼ 8.0, 8.0 Hz, 1H), 4.42 (t, J ¼ 4.8, Hz, 2H), 4.37 (t, J ¼ 5.2, Hz, 2H),
þ
þ
.95–1.78 (m, 4H); ESIMS m/z 260.3 [M þ H] .
3
3
.2.5. Syntheses of 6,7-alkylenedioxy naphthalen-1-amines (7a-d)
.2.5.1. Synthesis of naphtho[2,3-d][1,3]dioxol-5-amine (7a). To a solution of 6a
(1.0 g, 4.6 mmol) in MeOH (10 ml) at rt. was added 10% Pd (10% on carbon, 0.1 g).
The reaction mixture was stirred under the existence of H (1 atm) at rt. for 2 h until
2
the reaction was known for completion via TLC inspection. The reaction mixture was
filtered through diatomite to remove Pd and the filtrate was evaporated under vac-
uum to yield a crude product, which was purified using CC, eluted using a 10%
EtOAc in PE (v/v) isocratic elution, to yield brown powder (7a, 0.73 g) in a yield of
1
8
7
5
5.0%. H NMR (DMSO-d , 300 MHz) d: 7.46 (s, 1H), 7.12 (s, 1H), 7.01 (dd, J ¼ 7.9,
6
.4 Hz, 1H), 6.92 (br d, J ¼ 7.9 Hz, 1H), 6.54 (dd, J ¼ 7.4, 1.1 Hz, 1H), 6.06 (s, 2H),
þ
þ
.43 (s, 2H); ESIMS m/z 188.2 [M þ H] .
3
.2.5.2. Synthesis of 2,3-dihydronaphtho[2,3-b][1,4]dioxin-6-amine (7b). The syn-
thon 7b was obtained as yellow amorphous powder from 6b (1.0 g, 4.3 mmol) and
1
1
0% Pd (0.1 g) using a similar procedure to that of 7a in a yield of 92.0%. H NMR
(
DMSO-d , 400 MHz) d: 7.47 (s, 1H), 7.10 (s, 1H), 6.96 (dd, J ¼ 8.0, 7.6 Hz, 1H), 6.86
6
(d, J ¼ 8.0 Hz, 1H), 6.43 (dd, J ¼ 7.6, 1.2 Hz, 1H), 5.41 (s, 2H), 4.38–4.20 (m, 4H);
þ
þ
ESIMS m/z 202.2 [M þ H] .
3
.2.5.3. Synthesis of 3,4-dihydro-2H-naphtho[2,3-b][1,4]dioxepin-7-amine (7c). The
synthon 7c was obtained as yellow amorphous powder from 6c (1.0 g, 4.1 mmol) and
1
1
0% Pd (0.1 g) using a similar procedure to that of 7a in a yield of 84.4%. H NMR
(
DMSO-d , 400 MHz) d: 7.66 (s, 1H), 7.27 (s, 1H), 7.04 (dd, J ¼ 8.1, 7.8 Hz, 1H), 6.91
6
(br d, J ¼ 8.1 Hz, 1H), 6.51 (dd, J ¼ 7.8, 1.0 Hz, 1H), 5.52 (s, 2H), 4.16–4.13 (m, 4H),
þ
þ
2
.11 (tt, J ¼ 5.6, 5.6 Hz, 2H); ESIMS m/z 216.3 [M þ H] .
3
.2.5.4. Synthesis of 2,3,4,5-tetrahydronaphtho[2,3-b][1,4]dioxocin-8-amine (7d).
The synthon 7d was obtained as yellow amorphous powder from 6d (1.0 g, 3.9 mmol)
1
and 10% Pd (0.1 g) using a similar procedure to that of 7a in a yield of 85.2%. H

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
13
NMR (DMSO-d , 500 MHz) d: 7.66 (s, 1H), 7.27 (s, 1H), 7.03 (dd, J ¼ 8.1, 7.6 Hz,
6
1
1
H), 6.91 (d, J ¼ 8.1 Hz, 1H), 6.50 (d, J ¼ 7.6 Hz, 1H), 5.51 (s, 2H), 4.34–3.14 (m, 4H),
þ þ
.92–1.72 (m, 4H); ESIMS m/z 230.3 [M þ H] .
3
3
.2.6. Syntheses of 2,3:7,8-dialkylenedioxybenzo[c]phenanthridines (9a-f)
.2.6.1. Synthesis of 2,3:7,8-di(methylenedioxy)benzo[c]phenanthridine (9a). To a
mixture containing 4a (744 mg, 2.5 mmol) and 7a (425 mg, 2.3 mmol) in DMF
10 ml) at rt. was added K CO (470 mg, 3.4 mmol) batchwise. The reaction mixture
(
2
3
ꢀ
was heated to 110 C and stirred successively until the substitutive reaction was
known for completion on TLC monitoring. The reaction mixture was cooled to rt.
and then poured into a solvent of H O (25 ml) and filtered. The filter cake was
2
resolved in DCM and dried over anhydrous Na SO and filtered. Evaporating of the
2
4
filtrate under vacuum yielded a white amorphous powder (436 mg). This white
amorphous powder (1.1 mmol), along with Bu SnH (885 ll, 3.3 mmol), was added
3
ꢀ
into a solvent of toluene (48 ml). The solution was heated to 90 C under the N2
atmosphere and, then, AIBN (361 mg, 2.2 mmol) was added in, and the reaction mix-
ꢀ
ture was stirred at 110 C for 16 h until the reaction was known for completion via
TLC monitoring. The reaction mixture was cooled to rt. and then poured into a solv-
ent of H O (50 ml). The solution was extracted two times with EtOAc (25 ml, each
2
time). The organic solution was combined and washed using brine (50 ml), and then
dried over anhydrous MgSO4 and filtered. The filtrate was concentrated under
reduced pressure to yield a crude product which was purified via CC, eluted using a
1
0% EtOAc in PE (v/v) isocratic elution, to yield yellow amorphous powder of pre-
0
0
cursor
[1,4]dioxolo[4 ,5 :4,5]benzo[1,2-c][1,3]dioxolo[4,5-i]phenanthridine
(9a,
1
255 mg) in a yield of 35.5%. H NMR (DMSO-d , 400 MHz) d: 9.40 (s, 1H), 8.55 (d,
6
J ¼ 9.0 Hz, 1H), 8.50 (s, 1H), 8.40 (d, J ¼ 8.7 Hz, 1H), 7.99 (d, J ¼ 8.9 Hz, 1H), 7.69 (d,
J ¼ 8.7 Hz, 1H), 7.52 (s, 1H), 6.38 (s, 2H), 6.21 (s, 2H).
3
.2.6.2. Synthesis of 2,3-(1,3-trimethylenedioxy)-7,8-methylenedioxybenzo[c]phe-
nanthridine (9b). Precursor 9b was obtained as yellow amorphous powder from 4a
744 mg, 2.5 mmol) and 7c (502 mg, 2.3 mmol) using a similar procedure to that of
(
1
9
a in a yield of 23.7%. H NMR (DMSO-d , 500 MHz) d: 9.67 (s, 1H), 8.64 (s, 1H),
6
8
.59 (d, J ¼ 9.0 Hz, 1H), 8.47 (d, J ¼ 9.0 Hz, 1H), 8.04 (d, J ¼ 8.9 Hz, 1H), 7.68 (d,
J ¼ 8.9 Hz, 1H), 7.56 (s, 1H), 6.25 (s, 2H), 4.51 (t, J ¼ 5.3 Hz, 2H), 4.36 (t, J ¼ 5.3 Hz,
1
3
2
1
1
H), 2.36–2.26 (m, 2H); C NMR (DMSO-d , 100 MHz) d: 149.0, 148.9, 148.6, 147.2,
6
45.7, 130.3, 130.0, 128.7, 128.6, 128.4, 126.9, 126.8, 120.7, 119.3, 118.2, 105.2, 102.3,
01.3, 72.1, 71.3, 31.6.
3
.2.6.3. Synthesis of 2,3-(1,4-tetramethylenedioxy)-7,8-(1,2-dimethylenedioxy)ben-
zo[c]phenanthridine (9c). Precursor 9c was obtained as yellow amorphous powder
from 4b (2.13 g, 6.9 mmol) and 7d (1.45 g, 6.3 mmol) using a similar procedure to
1
that of 9a in a yield of 38.5%. H NMR (DMSO-d , 500 MHz) d: 9.59 (s, 1H), 8.76
6
(
(
4
s, 1H), 8.58 (d, J ¼ 9.0 Hz, 1H), 8.38 (d, J ¼ 9.0 Hz, 1H), 8.01 (d, J ¼ 8.9 Hz, 1H), 7.69
s, 1H), 7.61 (d, J ¼ 8.9 Hz, 1H), 4.60–4.56 (m, 2H), 4.50–4.46 (m, 2H), 4.44–4.40 (m,
H), 1.93–1.85 (m, 4H).

1
4
Q.-L. LI ET AL.
3.2.6.4. Synthesis of 2,3-(1,2-dimethylenedioxy)-7,8-(1,3-trimethylenedioxy)ben-
zo[c]phenanthridine (9d). Precursor 9d was obtained as yellow amorphous powder
from 4c (1.06 g, 3.3 mmol) and 7b (604 mg, 3.0 mmol) using a similar procedure to
1
that of 9a in a yield of 34.6%. H NMR (DMSO-d , 500 MHz) d: 9.71 (s, 1H), 8.70 (s,
6
1
H), 8.58 (d, J ¼ 9.0 Hz, 1H), 8.51 (d, J ¼ 9.0 Hz, 1H), 8.04 (d, J ¼ 8.9 Hz, 1H), 7.72
d, J ¼ 8.9 Hz, 1H), 7.61 (s, 1H), 4.53 (t, J ¼ 5.2 Hz, 2H), 4.50–4.43 (m, 4H), 4.39 (t,
J ¼ 5.2 Hz, 2H), 2.40–2.31 (m, 2H).
(
3
.2.6.5. Synthesis of 2,3:7,8-di(1,3-trimethylenedioxy)benzo[c]phenanthridine (9e).
Precursor 9e was obtained as yellow amorphous powder from 4c (612 mg, 1.9 mmol)
and 7c (372 mg, 1.7 mmol) using a similar procedure to that of 9a in a yield of
1
2
4.2%. H NMR (DMSO-d , 400 MHz) d: 9.78 (s, 1H), 8.93 (s, 1H), 8.29 (d,
6
J ¼ 8.9 Hz, 1H), 8.14 (d, J ¼ 9.0 Hz, 1H), 7.80 (d, J ¼ 9.0 Hz, 1H), 7.51 (s, 1H), 7.48 (d,
J ¼ 8.9 Hz, 1H), 4.51 (t, J ¼ 5.5 Hz, 2H), 4.40–4.32 (m, 6H), 2.42–2.33 (m, 2H),
2.33–2.25 (m, 2H).
3.2.6.6. Synthesis of 2,3-(1,2-dimethylenedioxy)-7,8-(1,4-tetramethylenedioxy)ben-
zo[c]phenanthridine (9f). Precursor 9f was obtained as yellow amorphous powder
from 4d (906 mg, 2.7 mmol) and 7b (493 mg, 2.4 mmol) using a similar procedure to
1
that of 9a in a yield of 28.7%. H NMR (DMSO-d , 500 MHz) d: 9.67 (s, 1H), 8.67 (s,
6
1
(
4
H), 8.57 (d, J ¼ 9.0 Hz, 1H), 8.52 (d, J ¼ 9.0 Hz, 1H), 8.02 (d, J ¼ 8.8 Hz, 1H), 7.68
d, J ¼ 8.8 Hz, 1H), 7.60 (s, 1H), 4.67–4.59 (m, 2H), 4.59–4.53 (m, 2H), 4.49–4.43 (m,
H), 2.08–1.95 (m, 4H).
3
.2.7. Syntheses of 2,3:7,8-dialkylenedioxy-5-methylbenzo[c]phenanthridin-5-ium
chloride (11a-f)
.2.7.1. Synthesis of QSC (11a). To a solution of 9a (200 mg, 0.6 mmol) in formic
3
acid (20 ml) at rt. was added NaBH (3.6 g, 95 mmol) batchwise. The reaction mixture
4
ꢀ
was stirred at 40 C until the reaction was known for completion via TLC monitor-
ing. After adding aq. 10% NH Cl (50 ml) into the reaction mixture, the mixture was
4
ꢀ
stirred at 25 C for 10 min, and then extracted two times using EtOAc (20 ml, each
time). The organic phase was combined and washed with brine (50 ml), and then
dried over anhydrous Na SO4 and filtered. The filtrate was concentrated under
2
reduced pressure to yield a crude residue. This crude residue was added into a solv-
ent of toluene (3 ml) containing aq. 5% NaOH (1.5 ml) and DDQ (286 mg, 1.2 mmol).
After stirred at rt. for 1 h, the reaction mixture was concentrated to dryness, the resi-
ꢀ
due was dissolved in CHCl (10 ml), the solution was cooled to 0 C. With aq. 10%
3
HCl (1ml) being added in dropwise, a red precipitation was crushed out, which was
collected via filtration and recrystallized using 1,4-dioxane as solvent to yield QSC
1
(
11a) as red powder in a yield of 80.4%. H NMR (DMSO-d , 500 MHz) d: 10.13 (s,
6
1
1
3
1
H), 8.78 (d, J ¼ 9.0 Hz, 1H), 8.64 (d, J ¼ 8.7 Hz, 1H), 8.30 (d, J ¼ 9.0 Hz, 1H), 8.27 (s,
H), 8.12 (d, J ¼ 8.7 Hz, 1H), 7.77 (s, 1H), 6.60 (br s, 2H), 6.34 (br s, 2H), 4.91 (s,
1
3
H); C NMR (DMSO-d , 100 MHz) d: 150.0, 148.8 (ꢂ2), 147.6, 146.3, 132.2, 131.5,
6
31.4, 127.2, 125.7, 120.3, 120.0, 118.9, 117.4, 109.5, 105.8, 105.0, 104.3, 102.9, 52.2,
þ
þ
ESIMS m/z 332.1 [M-Cl] .

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
15
3.2.7.2. Synthesis of 2,3-(1,3-trimethylenedioxy)-7,8-methylenedioxy-5-methyl-ben-
zo[c]phenanthridin-5-ium chloride (11b). The target compound 11b was obtained as
yellow amorphous powder from 9b (120 mg, 0.35 mmol) using a similar procedure to
1
that of 11a in a yield of 32.3%. H NMR (DMSO-d , 300 MHz) d: 10.15 (s, 1H), 8.78
6
(
d, J ¼ 9.0 Hz, 1H), 8.66 (d, J ¼ 8.9 Hz, 1H), 8.41 (s, 1H), 8.30 (d, J ¼ 8.9 Hz, 1H), 8.13
(
d, J ¼ 8.8 Hz, 1H), 7.88 (s, 1H), 6.60 (s, 2H), 4.95 (s, 3H), 4.38 (t, J ¼ 6.0 Hz, 2H),
1
3
4
7
1
.36 (t, J ¼ 6.0 Hz, 2H), 2.27 (tt, J ¼ 6.0 Hz, 6.0 Hz, 2H); C NMR (DMSO-d ,
6
5 MHz) d: 152.6, 151.6, 149.8, 148.1, 146.6, 132.1, 131.6, 131.4, 127.6, 126.7, 120.8,
20.5, 119.8 (ꢂ2), 118.8, 118.1, 110.2, 105.4, 71.2, 71.1, 52.5, 31.0; ESIMS m/z 360.4
þ
þ
[
M-Cl] .
3.2.7.3. Synthesis of 2,3-(1,4-tetramethylenedioxy)-7,8-(1,2-dimethylenedioxy)-5-
methyl-benzo[c]phenanthridin-5-ium chloride (11c). The target compound 11c was
obtained as yellow amorphous powder from 9c (350 mg, 0.94 mmol) using a similar
1
procedure to that of 11a in a yield of 32.3%. H NMR (DMSO-d , 400 MHz) d: 10.06
6
(
s, 1H), 8.80 (d, J ¼ 9.0 Hz, 1H), 8.62 (d, J ¼ 9.0 Hz, 1H), 8.50 (s, 1H), 8.30 (d,
J ¼ 9.0 Hz, 1H), 8.00 (d, J ¼ 9.0 Hz, 1H), 7.89 (s, 1H), 5.02 (s, 3H), 4.70–4.65 (m, 2H),
þ
4
.60–4.51 (m, 4H), 4.48–4.41 (m, 2H), 1.98–1.84 (m, 4H); ESIMS m/z 388.1
þ
[
M-Cl] .
3.2.7.4. Synthesis of 2,3-(1,2-dimethylenedioxy)-7,8-(1,3-trimethylenedioxy)-5-
methyl-benzo[c]phenanthridin-5-ium chloride (11d). The target compound 11d was
obtained as yellow amorphous powder from 9d (120 mg, 0.33 mmol) using a similar
1
procedure to that of 11a in a yield of 35.3%. H NMR (DMSO-d , 300 MHz) d: 10.05
6
(
s, 1H), 8.79 (d, J ¼ 9.1 Hz, 1H), 8.61 (d, J ¼ 9.1 Hz, 1H), 8.45 (s, 1H), 8.31 (d,
J ¼ 9.0 Hz, 1H), 8.01 (d, J ¼ 9.0 Hz, 1H), 7.90 (s, 1H), 5.00 (s, 3H), 4.79–4.62 (m, 2H),
þ
4
3
.62–4.52 (m, 2H), 4.48–4.27 (m, 4H), 2.27 (tt, J ¼ 5.4, 5.4 Hz, 2H); ESIMS m/z
þ
74.1 [M-Cl] .
3
.2.7.5. Synthesis of 2,3:7,8-di(1,3-trimethylenedioxy)-5-methyl-benzo[c]phenanthri-
din-5-ium chloride (11e). The target compound 11e was obtained as yellow amorph-
ous powder from 9e (151 mg, 0.40 mmol) using a similar procedure to that of 11a in
1
a yield of 32.3%. H NMR (DMSO-d , 500 MHz) d: 10.14 (s, 1H), 8.82 (d, J ¼ 9.0 Hz,
6
1
H), 8.71 (d, J ¼ 9.0 Hz, 1H), 8.48 (s, 1H), 8.33 (d, J ¼ 9.0 Hz, 1H), 8.04 (d, J ¼ 9.0 Hz,
H), 7.92 (s, 1H), 5.03 (s, 3H), 4.59 (t, J ¼ 5.3 Hz, 2H), 4.45 (t, J ¼ 5.3 Hz, 2H), 4.40
1
(
t, J ¼ 5.3 Hz, 2H), 4.38 (t, J ¼ 5.3 Hz, 2H), 2.38 (tt, J ¼ 5.3, 5.3 Hz, 2H), 2.29 (tt,
þ þ
J ¼ 5.3, 5.3 Hz, 2H); ESIMS m/z 388.1 [M-Cl] .
3
.2.7.6. Synthesis of 2,3-(1,2-dimethylenedioxy)-7,8-(1,4-tetramethylenedioxy)-5-
methyl-benzo[c]phenanthridin-5-ium chloride (11f). The target compound 11f was
obtained as yellow amorphous powder from 9f (250 mg, 0.67 mmol) using a similar
1
procedure to that of 11a in a yield of 65.8%. H NMR (DMSO-d , 400 MHz) d: 10.05
6
(
s, 1H), 8.74 (d, J ¼ 9.2 Hz, 1H), 8.69 (d, J ¼ 9.2 Hz, 1H), 8.35 (s, 1H), 8.28 (d,
J ¼ 9.2 Hz, 1H), 8.01 (d, J ¼ 9.2 Hz, 1H), 7.81 (s, 1H), 5.00 (s, 3H), 4.74–4.69 (m, 2H),
13
4
.56–4.51 (m, 2H), 4.51–4.44 (m, 4H), 2.03–1.93 (m, 4H); C NMR (DMSO-d ,
6

1
6
Q.-L. LI ET AL.
1
1
00 MHz) d: 149.3, 147.5, 145.5, 145.0, 144.3, 134.3, 131.9, 130.845, 130.837, 130.8,
25.3, 119.82, 119.80, 118.8, 118.5, 114.5, 113.8, 74.3, 72.6, 64.6, 64.4, 52.2, 26.3, 25.8;
þ
þ
ESIMS m/z 388.1 [M-Cl] .
3
.3. Growth inhibitory activity assay against human cancer cell lines
The growth inhibitory activity of all the target compounds against the human HCT-
16, HepG2, BGC-823, NCI-H1650, and A2780 cell lines were examined using a pub-
1
lished method [24].
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This work was financially supported by grant from CAMS Initiation Fund for Innovative
Medicine [grant number 2016-I2M-1-010].
References
[
[
[
[
[
1] G.Y. Zuo, F.Y. Meng, X.Y. Hao, Y.L. Zhang, G.C. Wang, and G. L. Xu, J. Pharm.
Pharm. Sci. 11, 90 (2008).
2] F. Meng, G. Zuo, X. Hao, G. Wang, H. Xiao, J. Zhang, and G. Xu, J. Ethnopharmacol.
1
25, 494 (2009).
3] K. P ꢀe n ꢀc ꢁı kov ꢁa , P. Koll ꢁa r, V.M. Z ꢁa valov ꢁa , E. T ꢁa borsk ꢁa , J. Urbanov ꢁa , and J. Ho ꢀs ek,
Phytomedicine 19, 890 (2012).
4] J. Hammerov ꢁa , S. Uldrijan, E. T ꢁa borsk ꢁa , and I. Slaninov ꢁa , J. Dermatol. Sci. 62, 22
(
2011).
5] X. Sun, L. Lin, Y. Chen, T. Liu, R. Liu, Z. Wang, K. Mou, J. Xu, B. Li, and H. Song,
Mol. Med. Rep. 13, 3161 (2016).
[
[
[
6] L.K. Wang, R.K. Johnson, and S.M. Hecht, Chem. Res. Toxicol. 6, 813 (1993).
7] W. Lin, J. Huang, Z. Yuan, S. Feng, Y. Xie, and W. Ma, Sci. Rep. 7, 2022 (2017).
8] M. Fukuda, M. Inomata, K. Nishio, K. Fukuoka, F. Kanzawa, H. Arioka, T. Ishida, H.
Fukumoto, H. Kurokawa, M. Oka, and N. Saijo, Jpn. J. Cancer Res. 87, 1086 (1996).
9] K. Morohashi, A. Yoshino, A. Yoshimori, S. Saito, S. Tanuma, K. Sakaguchi, and F.
Sugawara, Biochem. Pharmacol. 70, 37 (2005).
[
[
[
[
[
10] F. Kanzawa, K. Nishio, T. Ishida, M. Fukuda, H. Kurokawa, H. Fukumoto, Y. Nomoto,
K. Fukuoka, K. Bojanowski, and N. Saijo, Br. J. Cancer 76, 571 (1997).
11] T. Nakanishi, M. Suzuki, A. Mashiba, K. Ishikawa, and T. Yokotsuka, J. Org. Chem. 63,
4
235 (1998).
ꢁ
12] G. Maestri, M.H. Larraufie, E. Derat, C. Ollivier, L. Fensterbank, E. Lac o^ te, and M.
Malacria, Org. Lett. 12, 5692 (2010).
13] M. Sun, C. Liu, N. Nadiminty, W. Lou, Y. Zhu, J. Yang, C.P. Evans, Q. Zhou, and A.C.
Gao, Prostate 72, 82 (2012).
[
[
14] H. Ahsan, S. Reagan-Shaw, J. Breur, and N. Ahmad, Cancer Lett. 249, 198 (2007).
15] M. Janovsk ꢁa , M. Kubala, V. Sim ꢁa nek, and J. Ulrichov ꢁa , Anal. Bioanal. Chem. 395, 235
ꢀ
(
2009).
[
16] G. Ping, Y. Wang, L. Shen, Y. Wang, X. Hu, J. Chen, B. Hu, L. Cui, Q. Meng, and C.
Li, Chem. Commun. (Camb.) 53, 7381 (2017).

JOURNAL OF ASIAN NATURAL PRODUCTS RESEARCH
17
[
17] H. Ishii, T. Ishikawa, T. Watanabe, Y.I. Ichikawa, and E. Kawanabe, J. Chem. Soc.
Perkin Trans. I, 2283 (1984).
[
[
[
18] T. Harayama, Heterocycles 65, 697 (2005).
19] S. De, S. Mishra, B.N. Kakde, D. Dey, and A. Bisai, J. Org. Chem. 78, 7823 (2013).
20] S.V. Kessar, Y.P. Gupta, P. Balakrishnan, K.K. Sawal, T. Mohammad, and M. Dutt,
J. Org. Chem. 53, 1708 (1988).
[
[
[
[
21] J.C. Castillo, J. Quiroga, R. Abonia, J. Rodriguez, and Y. Coquerel, Org. Lett. 17, 3374
(
2015).
22] J. Styskala, P. Cankar, M. Soural, J. Hlavac, P. Hradil, J. Vicar, and V. Simanek,
Heterocycles 73, 769 (2007).
23] H. Ishii, T. Ishikawa, Y.I. Ichikawa, M. Sakamoto, M. Ishikawa, and T. Takahashi,
Chem. Pharm. Bull. 32, 2984 (1984).
24] Q. Li, A.J. Deng, L. Li, L.Q. Wu, M. Ji, H.J. Zhang, Z.H. Li, L. Ma, Z.H. Zhang, X.G.
Chen, and H.L. Qin, J. Nat. Prod. 80, 2189 (2017).