The synthesis and biological evaluation of multifunctionalised derivatives of noscapine as cytotoxic agents

Article abstract of DOI:10.1002/cmdc.201300395

Noscapine, a phthalideisoquinoline alkaloid derived from Papaver somniferum, is a well-known antitussive drug that has a relatively safe in vitro toxicity profile. Noscapine is also known to possess weak anticancer efficacy, and since its discovery, effor

Full text of DOI:10.1002/cmdc.201300395

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201300395
The Synthesis and Biological Evaluation of
Multifunctionalised Derivatives of Noscapine as Cytotoxic
Agents
Aaron J. DeBono,[a] Sarah J. Mistry,[a] Jinhan Xie,[b] Divya Muthiah,[c] Jackson Phillips,[a]
Sabatino Ventura,[b] Richard Callaghan,[c] Colin W. Pouton,[b] Ben Capuano,*[a] and
Peter J. Scammells*[a]
Noscapine, a phthalideisoquinoline alkaloid derived from Papa-
ver somniferum, is a well-known antitussive drug that has a rela-
tively safe in vitro toxicity profile. Noscapine is also known to
possess weak anticancer efficacy, and since its discovery, efforts
have been made to design derivatives with improved potency.
Herein, the synthesis of a series of noscapine analogues, which
have been modified in the 6’, 9’, 1 and 7-positions, is de-
scribed. In a previous study, replacement of the naturally oc-
curring N-methyl group in the 6’-position with an N-ethylami-
nocarbonyl was shown to promote cell-cycle arrest and cyto-
toxicity against three cancer cell lines. Here, this modification
has been combined with other structural changes that have
previously been shown to improve anticancer activity, namely
halo substitution in the 9’-position, regioselective O-demethy-
lation to reveal a free phenol in the 7-position, and reduction
of the lactone to the corresponding cyclic ether in the 1-posi-
tion. The incorporation of new aryl substituents in the 9’-posi-
tion was also investigated. The study identified interesting new
compounds able to induce G2/M cell-cycle arrest and that pos-
sess cytotoxic activity against the human prostate carcinoma
cell line PC3, the human breast adenocarcinoma cell line MCF-
7, and the human pancreatic epithelioid carcinoma cell line
PANC-1. In particular, the ethyl urea cyclic ether noscapinoids
and a compound containing a 6’-ethylaminocarbonyl along
with 9’-chloro, 7-hydroxy and lactone moieties exhibited the
most promising biological activities, with EC50 values in the
low micromolar range against all three cancer cell lines, and
these derivatives warrant further investigation.
Introduction
Noscapine, a phthalideisoquinoline alkaloid derived from Papa-
ver somniferum, is a well-known antitussive drug that has a rela-
tively safe in vitro toxicity profile.[1,2] Although noscapine was
identified as a potential drug for the inhibition of cancer cell
growth decades ago, it was not until 1998, forty years on, that
a follow up on its relatively weak anticancer activity was con-
ducted.[3] Since then, Joshi et al. have continued to work with
noscapine and investigated the possible mode of action for its
anticancer activity. Research from Joshi and co-workers has il-
lustrated that noscapine can interfere with tubulin polymeri-
sation through an ability to disrupt tubulin dynamics. This
mechanism of action has been validated by three major classes
of drugs: taxanes, vinca alkaloids and colchicine; drugs that
affect tubulin dynamics by either depolymerising or over poly-
merising tubulin.[4] Current research efforts focussing on nosca-
pine and related analogues have principally been conducted
by the research groups of Joshi and Bennani. The initial series
of noscapine analogues by Aggarwal et al. was a small series
of aryl-substituted N-carbamoyl/N-thiocarbamoyl noscapine an-
alogues;[5] however, most of their subsequent chemical initia-
tive was focused on the synthesis of a series of 9’-halo-substi-
tuted noscapine analogues.[6]
The culmination of work at the 9’-position by Aneja et al. re-
sulted in the reduction of the lactone moiety and the synthesis
of the 9’-fluoro analogue, CEFNA (2), a cyclic ether fluorinated
noscapine analogue.[7] The 9’-halo noscapine analogues were
shown to possess promising biological activity as G2 mitotic ar-
restors in various cancer cell types: human breast adenocarci-
noma cell lines MCF-7 and MDA-MB-231, and human leukemic
lymphoblast cell line CEM. These 9’-halo noscapine analogues
showed improved activity over parent alkaloid noscapine
within in vitro models.[6] Biological evaluation of these halo-
genated noscapine analogues revealed lower EC50 values when
compared with noscapine. The 9’-bromo noscapine analogue
showed approximately 40-fold higher cytotoxic activity (EC50 =
1.0ꢀ0.2 mm) in MCF-7 cells compared with parent noscapine
(EC50 =39.6ꢀ2.2 mm).[6] The inhibition of cellular proliferation
was noted through the observation of numerous fragmented
[a] A. J. DeBono, Dr. S. J. Mistry, J. Phillips, Dr. B. Capuano, Prof. P. J. Scammells
Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences
Monash University (Parkville Campus)
381 Royal Parade, Parkville, Victoria, 3052 (Australia)
E-mail: ben.capuano@monash.edu
peter.scammells@monash.edu
[b] J. Xie, Dr. S. Ventura, Prof. C. W. Pouton
Drug Discovery Biology, Monash Institute of Pharmaceutical Sciences
Monash University (Parkville Campus)
381 Royal Parade, Parkville, Victoria, 3052 (Australia)
[c] D. Muthiah, Dr. R. Callaghan
Research School of Biology, Australian National University
Canberra ACT 0200 (Australia)
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 399

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
nuclei after 72 h of drug exposure.[6] Following the synthesis of
the 9’-halo analogues, Aneja et al. synthesised CEFNA (2), the
halogenated cyclic ether noscapine analogue, which showed
inhibition against estrogen- and progesterone-receptor-posi-
tive MCF-7 cells, and estrogen- and progesterone-receptor-
negative MDA-MB-231 cells, irrespective of their hormone sen-
sitivity status.[7] Research by Aneja et al. also indicated that
these compounds did not show any predictable correlation be-
tween intercellular activity, but did identify the cell-type specif-
icity of each analogue.[6]
showed the aniline to be more than twofold more potent than
noscapine, with no observable effect on nondividing Swiss 3T3
cells. The 7-triflate was also converted to the corresponding
thiophenol (7-SH) analogue that, interestingly enough, shared
similar activity to the 7-aniline, which arrested HEK293 cells in
the G2/M phase.[8] The activity was postulated to be due to
the presence of hydrogen bond donors at the 7-position, but
this hypothesis was discounted when the 7-H analogue also
demonstrated an ability to cause G2/M arrest.[9]
An extension of this work looked at the synthesis of the 7-
acetyl and 7-carbamate analogues.[10] Mishra et al. reported the
synthesis of 7-position benzofuranone analogues, which differ
in the steric bulk of the substituent. Acylation reactions were
used to generate two 7-acyl derivatives, namely the acetyl and
benzoyl compounds. The 7-hydroxy group was also reacted
with a number of isocyanate reagents to generate a small
number of 7-carbamate analogues: the ethyl, phenyl and
benzyl carbamates.[10] Variation in potency in a panel of cell
lines was evident within this series of analogues, which
showed strong mitotic arrest in some cell lines but weaker
arrest in others.[10] These 7-carbamate analogues showed simi-
lar activity to the 9’-halo noscapine analogues, which induced
a significant accumulation of MDA-MB-231 cells trapped in G2/
M arrest.[6,10]
Anderson et al. probed the structure–activity relationships
(SARs) at the 7-position, which identified phenol 3. The initial
procedure utilised sodium benzyloxide in benzyl alcohol and
Our ongoing chemical initiative focuses on a study of the
SAR of noscapine analogues in an attempt to improve their
therapeutic efficacy. We recently published the improved N-de-
methylation of noscapine and reduction of the lactone moiety,
to generate a cyclic ether analogue of N-nornoscapine in good
yield.[11] This compound served as a useful scaffold for further
derivatisation. Our initial series of N-substituted analogues fo-
cused on five main functionalities: alkyl, acyl, carbamoyl, thio-
carbamoyl, and carbamate. A small proportion of the com-
pounds within this series showed promising in vitro activity.
The main hurdle for the in vitro testing of these compounds
was their moderately weak antimitotic activity and their poor
aqueous solubility at concentrations that induce activity. Given
the moderately weak antimitotic activity of these compounds,
higher concentrations were required to potentiate in vitro ac-
tivity. Estimation of their potency was limited by the poor solu-
bility of the test compounds within media suitable for cell cul-
ture.[11] Manchukonda et al. subsequently prepared a series of
N-alkyl-N-nornoscapine derivatives and also identified com-
pounds with higher cell cytotoxicity to various cell lines than
noscapine itself.[12]
DMF at 1208C to give a 7-benzyloxy derivative. This reaction,
however, resulted in epimerisation of the phthalide stereocen-
tre.[8] The resultant mixture of diastereomers, however, showed
S-phase mitotic arrest in human embryonic kidney 293
(HEK293) cells. This was quite unexpected given that the cur-
rent literature for noscapine analogues suggests that cellular
arrest occurs in the G2/Mitosis (M) phase of the cell cycle.[6] In
an attempt to decrease epimerisation of the product, the pro-
cedure was modified using methylmagnesium bromide and
benzyl alcohol in a THF/toluene mixture to minimise epimerisa-
tion.[8] The outcome however for this reaction resulted in the
selective O-demethylation at the 7-position giving 3, with little
to no epimerisation.[8] The phenolic group was then functional-
ised by reacting 3 with a small number of aromatic alkyl hal-
ides, which widened the scope of the SAR at the 7-position.
These O-benzyl-substituted derivatives were generally ob-
served to be S-phase arrestors.[8] The phenol was also convert-
ed to its triflate, and Suzuki coupling reactions were used to
generate a small number of biaryl compounds, however, these
compounds did not show significant S-phase arrest in HEK293
cells.[8] Anderson et al. followed up this area of work by looking
at the conversion of the 7-hydroxy group to the corresponding
triflate, and a subsequent amination reaction using palladium
chemistry to generate the aniline (7-NH2) analogue.[9] Aniline
analogue 4 was observed to inhibit tubulin polymerisation in
vitro. Analysis within microtubules using Swiss 3T3 cells
Herein, we describe the further derivatisation of the N-ethyl-
aminocarbonyl noscapine analogue 5, from our previous publi-
cation, which showed promising activity in both the human
prostate carcinoma cell line PC3 and MCF-7 cells at 10 mm,
with EC50 values of 6.7 and 3.6 mm, respectively. Given that re-
cently published modifications at the 7- and 9’-position result-
ed in improved antiproliferative activity,[6,8] we attempted to in-
troduce these modifications on our lead scaffold. We attempt-
ed to introduce each of the literature halogen atoms at the 9’-
position along with a 9’-amino group, whilst retaining the N-
ethylaminocarbonyl and cyclic ether moiety of 5. We also in-
vestigated the introduction of a small series of substituted
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 400

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
phenyl ring systems at the 9’-position. In summary, our aim
was to introduce onto our active lead scaffold 5 both 9’-posi-
tion functionality and a 7-hydroxy moiety in analogy to the
work of Joshi and Bennani et al.[7,8] This article describes our at-
tempts to introduce the favourable moieties determined by
other groups on to our own lead 5 to determine whether
these modifications would ameliorate the in vitro cellular activ-
ity of our compounds.
chloroform at ꢁ58C. This reaction was complete in 10 h, and
the pure 9’-chloro product was obtained following flash chro-
matography, albeit in moderate yield (54%). The formation of
a dichloro by-product was observed in this reaction, however,
this was minimised through the careful dropwise addition of
sulfuryl chloride in chloroform at low temperature to give pre-
dominantly desired monochloride product 9a. Compound 8
was successfully brominated using N-bromosuccinimide (NBS)
in acetic acid to afford desired 9’-bromo analogue 9b in 72%
yield. The iodination of 8 with N-iodosuccinimide (NIS) re-
quired the use of trifluoroacetic acid (TFA) as a solvent to acti-
vate the succinimide for aromatic substitution. This procedure
afforded desired product 9c in 68% yield. The synthesis of the
9-fluoro cyclic ether N-nornoscapine analogue was attempted
using the fluoride support resin Amberlyst A-26. Whilst litera-
ture precedence for this reaction exists,[7] our attempts to
apply this procedure to the fluorination of 8 were unsuccess-
ful. This possibly could be due to the inability to purchase the
identical fluoride support resin that had a higher fluoride load-
ing.
Results and Discussion
Synthesis
Noscapine (1) was initially converted to the corresponding N-
oxide (6) and then treated with iron sulfate heptahydrate in
methanol at ꢁ108C to afford N-nornoscapine (7) in good yield
(Scheme 1).[11] The lactone moiety of N-nornoscapine was sub-
sequently reduced to the corresponding cyclic ether (8) using
a modified literature procedure,[5] which employed boron tri-
fluoride diethyl etherate and sodium borohydride to generate
diborane in situ. The introduction of a 9’-halo group to the
cyclic ether analogue of N-nornoscapine required slight modifi-
cation from the procedures previously developed for the halo-
genation of noscapine.[5] The synthesis of 9’-chloro analogue
9a proved to be the most challenging amongst the halogen
series. The initial synthesis using N-chlorosuccinimide (NCS) in
acetic acid at room temperature did not yield the desired
product. The use of higher reaction temperatures afforded
a modest yield of the desired product; however, purification
problems resulted in the inability to separate the desired com-
pound from other products of the reaction. As a result, the
synthesis of 9a was conducted using sulfuryl chloride in
The next step of the synthesis following the aromatic halo-
genation was the reaction of the secondary amine scaffold
with ethyl isocyanate in acetonitrile to generate the desired
ethyl urea cyclic ether noscapinoids (10a–c). Although
a number of reaction pathways exist to synthesise compounds
10a–c, the pathway described in Scheme 1 gave the highest
yield and minimised the issues with problematic purification of
intermediate compounds.
Given the unsuccessful attempt of the fluorination reaction
using the Amberlyst A-26 support resin, an attempt to synthe-
sise the 9’-fluoro derivative using classical fluorination tech-
niques was attempted. This firstly required the synthesis of 9’-
aminonoscapine derivative 11. Literature precedence
for the generation of 9’-aminonoscapine exists
through the use of copper(I)-mediated, Ullman-type
coupling conditions for the generation of aniline
compounds from aromatic bromides, using copper(I)
iodide, sodium azide and an amino acid additive in
a basic medium (Scheme 2).[13,14] Compound 10b was
dissolved in anhydrous dimethyl sulfoxide under an
inert atmosphere. Copper(I) iodide, l-proline and
sodium azide were added, and the reaction was
heated at 1308C for 8 h. This reaction produced
a number of by-products that resulted in desired ani-
line 11 being obtained in modest yield (25%). At-
tempts to synthesise 9’-fluoro analogue 10d from
this aniline using Balz–Schiemann and Sandmeyer-
type reaction conditions resulted in the degradation
of 11.
We also explored the respective activity of intro-
ducing aromatic ring systems at the 9’-position using
Suzuki coupling reactions from aryl iodide 10c
(Scheme 3). The reactions were completed using
Suzuki reaction conditions whereby bis(triphenyl-
Scheme 1. Reagents and conditions: a) m-CPBA, CHCl3, ꢁ58C, HCl, 1 h, 100%;
b) FeSO4·7H2O, MeOH, ꢁ58C, 8 h, 78%; c) BF3·Et2O, NaBH4, THF, ꢁ58C!RT, 24 h, 78%;
d) For 9a: SO2Cl2, CHCl3, ꢁ58C!RT, 10 h, 54%; for 9b: NBS, CH3COOH, ꢁ58C!RT, 2 h,
72%; for 9c: NIS, TFAA, ꢁ58C!RT, 2 h, 65%; e) EtNCO, MeCN, ꢁ58C!RT, 2 h, 64%
(10a), 61% (10b), 68% (10c).
phosphine)palladium(II) dichloride was the catalyst of
choice, along with the use of sodium carbonate as
the base. A small series of 9’-aryl analogues were tar-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 401

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
hol.[8] The reaction proceeded in
good yield with only minor
quantities (~5%) of the epimer-
ized side-product observed in
the crude product by 1H NMR
analysis. The product was puri-
fied by flash chromatography
giving 64% of desired product
15. The lactone moiety of 15
was then reduced using the pre-
viously described modified pro-
cedure for lactone reduction[11]
Scheme 2. Reagents and conditions: a) NaN3, CuI, l-proline, DMSO, 1308C, 8 h, 25%; b) HBF4, NaNO2, NaBH4, THF,
reflux, 25%; c) NaNO2, HF, pyridine, ꢁ58C; d) HPF6, tBuONO, MeCN, [Emim]BF4, RT.
furnishing desired synthetic target 16. This reaction produced
a number of by-products, which resulted in 16 being obtained
in only modest yield following chromatographic purification.
Biological evaluation
Cell cycle assays
The novel noscapine derivatives were initially evaluated in
a cell-cycle assay to determine their ability to cause cell-cycle
arrest in the G2/M phase. In this assay, a human pancreatic epi-
thelioid carcinoma cell line PANC-1, along with PC3 and MCF-7
cell lines were treated with test compounds at 10 mm for 16 h.
The treated cells were subsequently stained with a cell-cycle
distribution dye (CellCycle Blue), and the cell-cycle phase distri-
bution was analysed by flow cytometry. The percentage in-
crease in arrested cells was calculated with reference to the
percentage of cells in G2/M phase following treatment with
vehicle (Table 1). Compounds that produced increases in the
number of cells in G2/M arrest below 20% were considered to
be practically inactive, while compounds causing greater than
Scheme 3. Reagents and conditions: a) For 12a: PhB(OH)2, PdCl2(PPh3)2, DME/
H2O (10:1), Cs2CO3, 1408C, m-wave (300 W), 30 min; for 12b–g: ArB(OH)2,
PdCl2(PPh3)2, THF, 1m Na2CO3, 1008C, 1–23 h; yield: 7–48%.
geted initially to probe the effect of aryl substitution and de-
termine if a more detailed investigation was warranted. 9’-
Phenyl analogue 12a along with the ortho, meta and para-me-
thoxyphenyl derivatives (12b–d) and similarly substituted
fluoro analogues (12e–g) were targeted to probe the effect of
9’-aryl substitution. These Suzuki reactions proceeded in rela-
tively modest yields ranging from 7 to 48%, but furnished suf-
ficient quantities of the desired products for pharmacological
evaluation.
We then focused our chemical synthesis on the introduction
of the 7-hydroxy functionality on 10a to determine whether
additive improvements in activity could be achieved through
this modification. The synthesis of the 7-hydroxy analogue was
achieved using the approach shown in Scheme 4. In this ap-
proach, the lactone moiety was retained until the last step of
the synthesis as its presence was required to effect a regiose-
lective O-demethylation at the 7-position. The synthesis began
with the halogenation of the N-nornoscapine scaffold. At-
tempts to chlorinate using sulfuryl chloride gave the desired
compound, however, purification of the product proved to be
problematic. As a result, chlorination was successfully carried
out using NCS, which gave a cleaner reaction profile and facili-
tated isolation of the target compound in respectable yield.
The subsequent reaction with ethyl isocyanate to generate N-
ethylcarbamoyl compound 14 was straightforward, with the
reaction of 13 with ethyl isocyanate at low temperature in ace-
tonitrile giving the desired urea (14) in excellent yield (86%).
The next step of the synthesis was the regioselective O-deme-
thylation using methylmagnesium bromide and benzyl alco-
Scheme 4. Reagents and conditions: a) NCS, TFAA, RT!reflux, 24 h, 34%;
b) EtNCO, MeCN, ꢁ58C!RT, 2 h, 86%; c) MeMgBr, BnOH, toluene, 1208C,
4 h, 64%; d) BF3·Et2O, NaBH4, THF, ꢁ58C!RT, 24 h, 23%.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 402

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
observed changes in the percentage of G2/M-arrest-
ed cells compared with the vehicle control.
Table 1. The percentage increase in arrested cells trapped in the G2/Mitosis (M)
phase.
Compound 15 containing the 6’-ethylaminocar-
bonyl along with the 9’-chloro, 7-hydroxy and lactone
moieties also exhibited considerable activity in the
PC3 and MCF-7 cell lines, but lower activity against
PANC-1 cells compared with compounds 9a–c and
10a–c. Such variable results are common within no-
scapine analogues, illustrating their variable activity
when tested on different cell lines. The reduction of
the lactone moiety of 15 to give compound 16 re-
sulted in decreased activity against all three cell lines.
The activity for 16 was also significantly decreased
compared with 10a, which possesses similar func-
tionality but contains the 7-methoxy instead of the 7-
hydroxy functionality. The 7-hydroxy noscapine ana-
logue[8] exhibited potent activity within both PC3 and
MCF-7 cell lines, comparable with the activities deter-
mined for our compounds 9a–c, 10a–c, but showed
decreased activity in PANC-1 cells.
Compd
Structural Modifications
Increase in G2/M [%][a]
R1
R7
R6’
R9’
PC3
MCF-7
PANC-1
1[11]
5[11]
9a
9b
9c
10a
10b
10c
11
12a
12b
12c
12d
12e
12 f
12g
14
=O
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OH
Me
C(O)NHEt
H
H
H
Cl
Br
I
Cl
Br
I
NH2
2
133
142
154
117
162
159
146
85
2
172
215
209
180
229
233
227
31
nd[b]
nd[b]
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
=O
138
H
H
132
135
114
119
120
92
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
C(O)NHEt
Ph
3
11
18
17
14
9
14
21
ꢁ5
7
ꢁ12
5
ꢁ8
11
12
Ph(2-OMe)
Ph(3-OMe)
Ph(4-OMe)
Ph(2-F)
Ph(3-F)
Ph(4-F)
Cl
ꢁ21
ꢁ6
Cell viability assays
ꢁ15
ꢁ14
ꢁ17
ꢁ15
9.9
Compounds that showed activity in the cell-cycle
assay were evaluated further in cell viability assays to
determine their potency in inhibiting cancer cell pro-
liferation. Cells were treated for 5 days with a range
of concentrations (from 100 nm to 10 mm), and cell vi-
ability was determined using a CellTiter-blue assay.
EC50 values were determined from mean concentra-
tion–viability curves using GraphPad Prism v6. EC50
values were not determined in MCF-7 and PANC-
1 cells if the mean EC50 value of the test compound
in PC3 cells was greater than 3 mm. In contrast with
13
37
15
16
=O
H, H
Cl
Cl
155
99
238
80
98
93
OH
[a] The percentage increase in arrested cells was calculated with reference to the per-
centage of cells in G2/M in the vehicle control. The formula used: (percentage of cells
in G2/M treated with compoundsꢁpercentage of cells in G2/M treated with vehicle
control)/percentage of cells in G2/M treated with vehicle control ꢁ100%. [b] Cell cycle
analysis was not determined (nd) in PANC-1 cells.
a 100% increase were considered to be highly active com-
pounds.
the flow cytometry results, our cell viability results suggested
very minimal fluctuations in potency across the cell lines
chosen.
We investigated whether our second-generation noscapine
analogues could elicit advanced G2/M phase arrest through
the disruption of microtubule assembly compared with our
previously described noscapine derivatives.[11] The 9’-halogena-
tion of nornoscapine (compounds 9a,b) resulted in marginal
increases in activity compared to our lead compound from our
previous publication.[11] In each of the cell lines tested (PC3,
MCF-7 and PANC-1), these compounds presented on average
greater than 130% increase in the number of arrested cells
compared to the vehicle control (Table 1; Figure 1). 9’-Halogen-
ation was combined with the 6’-ethylaminocarbonyl moiety
giving compounds 10a–c, which resulted in a modest incre-
ment in activity. Treatment with compounds 10a–c gave rise
to an increased number of arrested cells in the G2/M phase
compared with the vehicle control in PC3 and MCF-7 cells, but
slightly lower activity in PANC-1 cells compared with corre-
sponding N-nor analogues 9a–c. Introduction of various sub-
stituted aryl functional groups at the 9’-position in compounds
12a–g effectively abolished activity against PC3, MCF-7 and
PANC-1 cell lines. None of our 9’-aryl noscapine analogues
demonstrated any discernible activity; that is, there were no
In our previous communication, we elucidated a number of
N-substituted noscapine analogues that possessed improved
G2/M activity compared with noscapine.[11] The most potent
noscapine derivative described in our previous study was com-
pound 5. This compound possesses a methylene at the 1-posi-
tion as part of a cyclic ether and a 6’-ethylaminecarbonyl sub-
stituent, rather than a lactone and N-methyl group, respective-
ly, which are present in the corresponding positions within no-
scapine. The EC50 values for compound 5 within PC3 and MCF-
7 cell lines were found to be 6.7 and 3.6 mm, respectively. The
addition of a halogen atom at the 9’-position afforded an im-
provement in activity in both the PC3 and MCF-7 cell lines
compared with 5. The EC50 values for these active compounds
were also determined in the PANC-1 cell line, which also
showed promising results. 9’-Chloro analogue 10a afforded
a 4.5-fold improvement in activity in the PC3 cell line and
a 2.1-fold improvement in the MCF-7 cell line, with EC50 values
of 1.5 and 1.7 mm, respectively. Although 9’-chloro analogue
10a showed on average lower EC50 values across each of the
cell lines tested, the differences are not statistically significant,
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 403

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
causes is the overexpression of
multidrug efflux pumps, such as
ABCB1 and ABCG2. Substrates
for these proteins are character-
ised by a considerable decrease
in cytotoxic potency in drug-re-
sistant cell lines.
The presence of ABCB1 in the
NCI-AdrRES cell line decreased the
cytotoxic potency of vinblastine
by 260-fold from 0.36ꢀ0.01 nm
in the parental cell line to 96ꢀ
15 nm. The presence of ABCG2
in MCF7FLV1000 cells decreased the
potency of mitoxantone by 17-
fold from 26ꢀ13 nm in drug-
sensitive MCF7 cells to 435ꢀ
123 nm. The cytotoxic drugs vin-
blastine and mitoxantrone are
substrates for ABCB1 and
ABCG2, respectively. Table 3
demonstrates the cytotoxic po-
tency of a series of noscapine
derivatives in the MCF7 cell line,
which does not express either
multidrug efflux pump, and in
the drug selected resistant cells.
There was no statistically signifi-
cant difference in the ability of
the noscapine derivatives 10a,
10b and 10c to produce cyto-
toxicity between the resistant
and sensitive cells. Compound 5
was characterised by an EC50
Figure 1. Effects of second-generation noscapine analogues on cell cycle. PC3 cells were treated with a) vehicle
(control), b) vincristine (100 nm) or test compounds (10 mm) c) 10a, d) 10b, e) 10c and f) 15 for 16 h. Vincristine
was used as a positive control.
and each of the 9’-halo compounds (10a–c) have very similar
in vitro activities (Table 2). The 9’-amino analogue was consid-
erably less potent than the 9’-halo compounds (10a–c) in all
of the three cell lines. Compound 11 was 22-, 15- and 14-fold
less active than 10a in the PC3, MCF-7 and PANC-1 cell lines,
respectively. Compound 14, which possesses a 9’-chloro group
and a 6’-ethylaminocarbonyl substituent with a carbonyl at the
1-position as part of a lactone (rather than a cyclic ether),
showed negligible activity (EC50 >50 mm) compared with 10a.
However, the conversion of the 7-methoxy group of 14 to the
corresponding phenol (7-OH) afforded 15, which demonstrated
significantly improved and consistent activity, with EC50 values
in PC3, MCF-7 and PANC-1 cells of 1.5, 1.2 and 1.4 mm, respec-
tively. The further reduction of the 1-lactone to the corre-
sponding cyclic ether decreased activity against all of the cell
lines tested by 1.5 to 4-fold depending on the cell line.
value of 29.8ꢀ7.8 mm in the ABCG2-expressing MCF7FLV1000
cells, which was 2.1-fold higher than observed in the control
Table 2. Cell cytotoxicity of noscapine derivatives in PC3, MCF-7 and
PANC-1 cells.[a]
Compd
EC50 [mm]
MCF-7[b]
PC3
PANC-1[b]
5[11]
9a
9b
9c
10a
10b
10c
11
6.7ꢀ1.2
4.4ꢀ0.3
2.9ꢀ0.2
7.1ꢀ1.3
1.5ꢀ0.4
1.6ꢀ0.1
1.4ꢀ0.1
28.5ꢀ0.2
3.6ꢀ0.4
nd
nd
nd
nd
nd
nd
nd
1.7ꢀ0.1
1.8ꢀ0.2
2.9ꢀ0.2
26.1ꢀ3.0
0.9ꢀ0.08
1.0ꢀ0.01
0.9ꢀ0.2
11.8ꢀ2.4
14
>50
>50
>50
15
16
1.5ꢀ0.1
5.9ꢀ0.9
1.2ꢀ0.1
2.1ꢀ0.1
1.3ꢀ0.1
2.0ꢀ0.3
[a] Cells were treated with various concentration of drug, and cell viability
was detected via a CellTiter-Blue assay. Fluorescence values were plotted
against log concentration of inhibitor, and EC50 values were calculated
using GraphPad Prism statistical software. Data represent the meanꢀ
SEM of n=3 experiments performed in triplicate. [b] EC50 was not deter-
mined (nd) in MCF-7 and PANC-1 cells if the mean EC50 value of the test
compound in PC3 cells was >3 mm.
The influence of multidrug efflux pumps ABCB1 and ABCG2
Noscapine and its derivatives have been demonstrated to
cause cytotoxicity in cultured cancer cells. A frequent, and un-
wanted, observation for cytotoxic agents is the presence or
emergence of drug resistance. One of the most common
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 404

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
cell line. Although the difference was statistically significant,
the biological change was minor, as illustrated by the relative
difference in potency for the established ABCG2 substrate mi-
toxantrone. All of the compounds were considerably less
potent than the established cytotoxic drugs, vinblastine and
mitoxantrone, in MCF7 cells (see Table 3). However, each of the
noscapine derivatives was able to effect a 100% decrease in
cell growth, thereby indicating a high efficacy. The lack of dif-
to be substrates for efflux by the ABCB1 or ABCG2 transport-
ers.
Experimental Section
General Procedures: Chemicals and solvents were purchased from
standard suppliers and used without further purification. N-Norno-
scapine was prepared using methodology that has been previously
reported.[11] Davisil silica gel (40–63 mm), for flash column
chromatography, was supplied by Grace Davison Discov-
ery Sciences (Victoria, Australia) and deuterated solvents
Table 3. Cytotoxic potency of a series of selected noscapine derivatives and anticanc-
er drugs in drug-sensitive and -resistant breast cancer cell lines.[a]
were purchased from Cambridge Isotope Laboratories,
Inc. (USA, distributed by Novachem PTY. Ltd, Victoria,
Australia). Reactions were monitored by thin-layer chro-
matography (TLC) on commercially available precoated
aluminium-backed plates (Merck Kieselgel 60 F254), visual-
ised under UV light (254 and 366 nm), with general stain-
ing carried out using KMnO4 or phosphomolybdic acid.
A solution of ninhydrin in EtOH was used to visualize pri-
mary and secondary amines. All organic extracts collect-
ed after aqueous work-up procedures were dried over
anhydrous MgSO4 or Na2SO4 before gravity filtering and
evaporation to dryness. Organic solvents were evaporat-
ed in vacuo at ꢂ408C (water bath temperature). Purifica-
tion using preparative layer chromatography (PLC) was
carried out on Analtech preparative TLC plates
(200 mmꢁ200 mmꢁ2 mm). 1H NMR and 13C NMR spectra
were recorded on a Bruker Avance Nanobay III 400 MHz
Ultrashield Plus spectrometer at 400.13 MHz and
100.6 MHz, respectively. Spectra were obtained in CDCl3
unless otherwise stated, measured relative to the residu-
Compd
EC50 [mm]
MCF7WT
NCI-AdrRES
MCF7FLV1000
5
10a
10b
10c
13.6ꢀ2.7 (8)
1.87ꢀ0.67 (6)
1.24ꢀ0.31 (7)
4.07ꢀ0.72 (7)
0.36ꢀ0.01 (10)
26ꢀ13 (9)
12.4ꢀ2.8 (8)
2.07ꢀ0.47 (6)
1.93ꢀ0.52 (7)
2.44ꢀ0.18 (7)[b]
96ꢀ15 (10)[b]
–
29.8ꢀ7.77 (8)[b]
2.63ꢀ0.69 (6)
1.72ꢀ0.42 (7)
3.90ꢀ0.68 (7)
–
Vinblastine [nm]
Mitoxantrone [nm]
435ꢀ123 (9)[b]
[a] Drug cytotoxicity was measured in ABCB1-expressing NCI-AdrRES cells, ABCG2-ex-
pressing MCF7FLV1000 cells and the parental MCF7 cell line. Viable cell number was mea-
sured using an MTT assay, and all drugs were added in the concentration range 10ꢁ12
to 10ꢁ4 m. The cytotoxic potency (EC50) of each compound was estimated from nonlin-
ear regression of the dose–response relationship. All values indicate the mean ꢀSEM;
values stated in parenthesis are the number of independent observations. [b] Indi-
cates a statistically significant difference (P <0.05) in comparison to MCF7WT cells.
ference between the resistant and sensitive cell lines indicates
al solvent peak of deuterated solvent. Chemical shifts (d) are re-
ported in parts per million (ppm) with reference to the chemical
shift of the deuterated solvent. Coupling constants (J) and carbon–
fluorine coupling constants (JCF) are reported in Hertz (Hz), and the
significant multiplicities described by singlet (s), doublet (d), triplet
(t), quadruplet (q), broad (br), multiplet (m), doublet of doublets
(dd), doublet of triplets (dt). Spectra were assigned using appropri-
ate COSY, DEPT, HSQC and HMBC sequences.
that the compounds are unlikely to be substrates for transport
by either ABCB1 or ABCG2.
Conclusions
A series of second-generation noscapine analogues were syn-
thesized and evaluated for their anticancer potential, using
cell-cycle arrest and growth inhibition assays with three cancer
cell lines (PC3, MCF-7 and PANC-1). Most of our second-genera-
tion noscapine analogues showed improved anticancer activity
in the chosen cell lines compared with our lead compound 5
from our previous study.[11] The compounds that elicited G2/M
arrest within the PC3 cell arrest assay were also tested in the
MCF-7 and PANC-1 cell lines, and EC50 value were calculated
from the results. The EC50 values for some compounds were
not determined when weak cell cycle arrest was observed,
whilst for other compounds (10a,b and 15), we observed EC50
values of less than 2 mm across all three cell lines. More impor-
tantly, the target compounds for our second-generation ana-
logues, containing the desirable modifications identified by
other research groups, proved to be increasingly more active.
Our research identified four candidates for further study: 10a–
c and 15. These compounds showed marked improvements in
antiproliferative activity, with all four exhibiting a fourfold or
greater improvement in activity compared with lead com-
pound 5. Furthermore, our preliminary data from resistant and
sensitive cell lines suggests that these compounds are unlikely
Liquid chromatography–mass spectrometry (LCMS) were run to
verify reaction outcome and purity using an Agilent 6100 Series
Single Quad coupled to an Agilent 1200 Series HPLC (system A) or
an Agilent 1260 infinity 6120 Quadrupole LCMS (system B). For
system A, the following buffers were used: buffer A: 0.1% formic
acid in H2O; buffer B: 0.1% formic acid in MeCN. The following gra-
dient was used with a Phenomenex Luna 3 mm C8 15ꢁ4.6 mm
column, and a flow rate of 0.5 mLminꢁ1 and total run time of
12 min: 0–4 min 95% buffer A and 5% buffer B, 4–7 min 0% buffer
A and 100% buffer B, 7–12 min 95% buffer A and 5% buffer B.
Mass spectra were acquired in positive and negative ion mode
with a scan range of 0–1000 m/z at 5 V. UV detection was carried
out at 254 nm. For system B, the following buffers were used:
buffer A: 0.1% formic acid in H2O; buffer B: 0.1% formic acid in
MeCN. An Agilent poroshell 120 2.7 mm, EC-C18 3.0ꢁ50 mm
column, a 1 mLminꢁ1 flow rate and total run time of 5 min were
used. Mass spectra were acquired in positive and negative ion
mode. UV detection was carried out at 254 and 214 nm. All com-
pounds were of >95% purity unless otherwise stated.
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-4-me-
thoxy-5,6,7,8-tetrahydro[1,3]dioxolo[4,5-g]isoquinoline (8): A so-
lution of NaBH4 (0.22 g, 5.8 mmol) in THF (18 mL) was stirred at
ꢁ58C. N-nornoscapine (0.75 g, 1.88 mmol) dissolved in BF3·Et2O
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 405

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
(9 mL) was added dropwise at ꢁ58C. The reaction was allowed to
stir at ꢁ58C for 1 h, then allowed to warm to RT and stirred over-
night. The reaction was quenched with 10% aq HCl (10 mL) and al-
lowed to stir for 1 h before being extracted with CHCl3 (3ꢁ30 mL).
The organic layer was then washed with 2.2m aq NaOH (20 mL) to
yield the free-base form. The organic layer was dried over Na2SO4,
filtered, and the solvent was evaporated in vacuo. The crude prod-
uct was purified by flash chromatography (CHCl3/MeOH/NH4OH;
85:14:1) to afford the title compound as a white foam (0.540 g,
0.227 mmol) in TFA (2 mL) was added dropwise over a 30 min
period. The reaction was stirred at RT for 2 h, and then the solution
was basified to pH 9 with 2m aq NaOH. The reaction was then ex-
tracted with CHCl3 (2ꢁ10 mL), and the organic layers were collect-
ed, dried over anhydrous Na2SO4, filtered and evaporated in vacuo.
The crude product was purified by flash chromatography (CHCl3/
MeOH; 99:1) to give the title compound as a yellow oil, which was
then treated with 1m aq HCl to form the hydrochloride salt
(0.074 g, 0.135 mmol, 65%): 1H NMR: d=10.61–10.54 (m, 1H),
9.33–9.27 (m, 1H), 6.65 (d, J=8.3 Hz, 1H), 6.03 (dd, J=4.0, 1.4 Hz,
2H), 5.94 (d, J=8.3 Hz, 1H), 5.88–5.87 (m, 1H), 5.76–5.73 (m, 1H),
5.09–5.05 (m, 2H), 3.94 (s, 3H), 3.85 (s, 3H), 3.80 (s, 3H), 3.45–3.40
(m, 1H), 3.04–2.98 (m, 1H), 2.67–2.55 ppm (m, 2H); 13C NMR: d=
152.1, 151.4, 143.8, 140.4, 133.2, 132.2, 130.4, 129.3, 117.1, 114.1,
112.7, 100.8, 81.2, 69.3, 60.3, 59.7, 56.4, 53.1, 38.4, 30.02,
29.84 ppm; HRMS: m/z [M+H]+ calcd for C21H24ClINO6: 512.0565,
found: 512.0570.
1
1.39 mmol, 78%): Rf =0.25 (CHCl3/MeOH/NH4OH; 85:14:1); H NMR:
d=6.59 (d, J=8.2 Hz, 1H), 6.33 (s, 1H), 5.92 (q, J=1.5 Hz, 2H), 5.83
(dd, J=8.2, 0.9 Hz, 1H), 5.77 (td, J=2.6, 1.3 Hz, 1H), 5.36 (dd, J=
12.4, 2.8 Hz, 1H), 5.19–5.16 (m, 1H), 4.62 (d, J=4.0 Hz, 1H), 3.99 (s,
3H), 3.85 (s, 3H), 3.79 (s, 3H), 2.69–2.54 (m, 3H), 2.35–2.26 ppm (m,
2H); 13C NMR: d=151.3, 148.1, 143.0, 140.7, 134.4, 133.5, 132.7,
131.4, 118.7, 117.4, 112.2, 103.2, 100.7, 85.0, 71.8, 60.1, 59.5, 56.3,
55.3, 39.7, 29.9 ppm; HRMS: m/z [M+H]+ calcd for C21H24NO6:
386.1598, found: 386.1612.
(R)-9-Chloro-5-((S)-4,5-dimethoxy-1,3-dihydroisobenzofuran-1-
yl)-N-ethyl-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-g]isoquino-
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-9-
chloro-4-methoxy-5,6,7,8-tetrahydro-[1,3]dioxolo[4,5-g]isoquino-
line (9a): Compound 8 (0.124 g, 0.294 mmol) was dissolved in
CHCl3, and the solution was cooled to ꢁ58C. SO2Cl2 (0.119 g,
0.882 mmol) in CHCl3 (2 mL) was added dropwise to the reaction
over 1 h. The reaction mixture was allowed to warm to RT and
stirred for a further 10 h. The reaction was then poured into ice
and extracted with CHCl3 (2ꢁ10 mL). The organic layers were col-
lected, dried over anhydrous Na2SO4, filtered and evaporated in
vacuo. The crude product was purified by flash chromatography
(CH2Cl2/MeOH/NH4OH; 95:4:1) to give the title compound as
line-6(5H)-carboxamide
(10a):
Compound
9a
(0.050 g,
0.118 mmol) was dissolved in MeCN (5 mL), and solution was
cooled (ꢁ58C) under N2. Ethyl isocyanate (0.013 g, 0.177 mmol)
was added, and the solution was allowed to warm to RT. The reac-
tion was complete in 2 h, then quenched with H2O (1 mL) and ex-
tracted with CH2Cl2 (2ꢁ10 mL). The organic layers were collected,
dried over anhydrous Na2SO4, filtered and evaporated in vacuo.
The crude product was purified by flash chromatography (petrole-
um spirits/EtOAc; 2:1) to give the title compound as a pale yellow
foam (0.040 g, 0.080 mmol, 64%): Rf =0.18 (petroleum spirits/
1
1
a white foam (0.067 g, 0.158 mmol, 54%): H NMR: d=6.61 (d, J=
EtOAc, 2:1); H NMR: d=6.63 (d, J=8.2 Hz, 1H), 6.05–6.02 (m, 3H),
8.2 Hz, 1H), 6.02 (q, J=1.4 Hz, 2H), 5.88 (d, J=8.2 Hz, 1H), 5.71 (td,
J=2.7, 1.2 Hz, 1H), 5.35 (dd, J=12.1, 2.8 Hz, 1H), 5.16 (d, J=
12.1 Hz, 1H), 4.59 (d, J=4.2 Hz, 1H), 3.95 (s, 3H), 3.84 (s, 3H), 3.80
(s, 3H), 2.78–2.72 (m, 2H), 2.63–2.53 (m, 2H), 2.39–2.34 ppm (m,
1H); 13C NMR: d=151.4, 144.7, 143.1, 139.4, 134.9, 133.4, 132.5,
128.8, 120.4, 117.2, 112.3, 107.7, 101.5, 84.7, 71.9, 60.1, 59.6, 56.3,
55.1, 38.9, 26.4 ppm; HRMS: m/z [M+H]+ calcd for C21H23ClNO6:
420.1208, found: 420.1226.
5.89 (s, 1H), 5.80 (td, J=2.7, 0.7 Hz, 1H), 5.30 (d, J=3.6 Hz, 1H),
5.17–5.09 (m, 2H), 4.03 (s, 3H), 3.95–3.90 (m, 1H), 3.82 (s, 3H), 3.81
(s, 3H), 3.36–3.29 (m, 1H), 3.24–3.17 (m, 1H), 2.57–2.49 (m, 1H),
2.44–2.38 (m, 1H), 2.13–2.06 (m, 1H), 1.18 ppm (t, J=7.2 Hz, 3H);
13C NMR: d=159.7, 151.8, 145.2, 143.4, 138.6, 134.8, 132.1, 131.7,
128.8, 118.3, 117.4, 112.3, 107.5, 101.7, 85.9, 71.9, 60.2, 59.7, 56.4,
56.3, 37.3, 35.8, 24.9, 15.8 ppm; HRMS: m/z [M+H]+ calcd for
C24H28ClN2O7: 491.1580, found: 491.1586.
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-9-
(R)-9-Bromo-5-((S)-4,5-dimethoxy-1,3-dihydroisobenzofuran-1-
yl)-N-ethyl-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-g]isoquino-
bromo-4-methoxy-5,6,7,8-tetrahydro-[1,3]dioxolo[4,5-g]isoquino-
line hydrochloride (9b): Compound 8 (0.100 g, 0.237 mmol) was
dissolved in acetic acid, and solid N-bromosuccinimide (0.044 g,
0.249 mmol) was added portionwise. The reaction was allowed to
proceed for 2 h before being basified with aq NH4OH. The basic
aqueous solution was then extracted with CH2Cl2 (3ꢁ20 mL). The
organic extracts were collected, dried over MgSO4, filtered and
evaporated in vacuo. The crude was then dissolved in a solution of
4m HCl in dioxane (2 mL) to afford an oil, which was washed with
Et2O and decanted off to afford the title compound as a yellow
amorphous solid (0.085 g, 0.170 mmol, 72%): Rf =0.09 (CHCl2/
line-6(5H)-carboxamide
(10b):
Compound
9b
(0.032 g,
0.069 mmol) was dissolved in MeCN (5 mL), and the solution was
cooled to ꢁ58C under N2. Ethyl isocyanate (0.006 g, 0.084 mmol)
was added to the cold mixture, and the solution was allowed to
warm to RT. The reaction was complete in 2 h, and then quenched
with H2O (1 mL) before being extracted with CH2Cl2 (2ꢁ10 mL).
The organic layers were collected, dried over anhydrous Na2SO4, fil-
tered and evaporated in vacuo. The crude product was purified by
flash chromatography (toluene/EtOAc; 3:1) to give the title com-
pound as a white foam (0.022 g, 0.042 mmol, 61%): Rf =0.01 (tolu-
1
1
iPrOH; 95:5); H NMR: d=10.54 (brs, 1H), 9.34 (brs, 1H), 6.65 (d,
ene/EtOAc; 3:1); H NMR: d=6.64 (d, J=8.2 Hz, 1H), 6.05–6.02 (m,
J=8.3 Hz, 1H), 6.04 (dd, J=5.1, 1.3 Hz, 2H), 5.95 (d, J=8.3 Hz, 1H),
5.87 (s, 1H), 5.76–5.73 (m, 1H), 5.07 (d, J=12.9 Hz, 2H), 3.93 (s,
3H), 3.85 (s, 3H), 3.80 (s, 3H), 3.47–3.42 (m, 1H), 3.09–3.00 (m, 1H),
2.75–2.69 (m, 1H), 2.60–2.51 ppm (m, 1H); 13C NMR: d=152.0,
148.0, 143.8, 139.5, 134.5, 132.1, 130.4, 126.9, 117.0, 113.9, 112.7,
101.7, 96.0, 81.2, 72.2, 60.2, 59.7, 56.4, 53.0, 37.8, 25.1 ppm; HRMS:
m/z [M+H]+ calcd for C21H24BrClNO6: 464.0703, found: 464.0723.
3H), 5.91 (brs, 1H), 5.79 (td, J=2.6, 0.8 Hz, 1H), 5.31 (d, J=3.5 Hz,
1H), 5.17–5.09 (m, 2H), 4.04 (s, 3H), 3.94–3.86 (m, 1H), 3.82 (s, 3H),
3.81 (s, 3H), 3.33 (dt, J=13.3, 6.7 Hz, 1H), 3.20 (dd, J=13.3, 7.1 Hz,
1H), 2.57–2.49 (m, 1H), 2.42–2.35 (m, 1H), 2.16–2.08 (m, 1H),
1.18 ppm (t, J=7.2 Hz, 3H); 13C NMR: d=159.7, 151.8, 146.8, 143.4,
139.2, 134.5, 132.1, 131.7, 130.0, 118.7, 117.4, 112.3, 101.3, 96.3,
86.0, 72.0, 60.3, 59.7, 56.4, 56.3, 37.6, 35.8, 27.6, 15.8 ppm; HRMS:
m/z [M+H]+ calcd for C24H28BrN2O7: 535.1074, found: 535.1100.
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-9-iodo-
4-methoxy-5,6,7,8-tetrahydro-[1,3]dioxolo[4,5-g]isoquinoline hy-
drochloride (9c): Compound 8 (0.080 g, 0.207 mmol) was dis-
solved in TFA (3 mL), and a solution N-iodosuccinimide (0.051 g,
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-N-ethyl-
9-iodo-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-g]isoquinoline-
6(5H)-carboxamide (10c): Compound 9c (0.068 g, 0.133 mmol)
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 406

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
was dissolved in MeCN (5 mL), and the solution was cooled to
ꢁ58C under N2. Ethyl isocyanate (0.014 g, 0.197 mmol) was then
added to the cold mixture, and the solution was allowed to warm
to RT. The reaction was complete in 2 h, and then quenched with
H2O (1 mL) before being extracted with CH2Cl2 (2ꢁ10 mL). The or-
ganic layers were collected, dried over anhydrous Na2SO4, filtered
and evaporated in vacuo. The crude product was purified by flash
chromatography (petroleum spirits/EtOAc; 1:1) to give the title
compound as a pale yellow foam (0.049 g, 0.084 mmol, 68%): Rf =
39.3, 35.6, 26.1, 15.7 ppm; HRMS: m/z [M+H]+ calcd for
C30H33N2O7: 533.2282, found: 533.2288.
(5R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-N-
ethyl-4-methoxy-9-(2-methoxyphenyl)-7,8-dihydro-[1,3]dioxolo-
[4,5-g]isoquinoline-6(5H)-carboxamide (12b): A degassed solu-
tion of 1m aq Na2CO3 (0.33 mL) and THF (0.66 mL) in a microwave
vial was treated with 2-methoxyphenylboronic acid (0.016 g,
0.108 mmol), 10c (0.057 g, 0.098 mmol) and PdCl2(PPh3)2 (0.007 g,
0.009 mmol). The vial was sealed under N2 and heated to 1008C
for 1.25 h. The reaction was cooled to RT, and the mixture was par-
titioned between EtOAc (10 mL) and water (10 mL). The organic
layer was washed with water (2ꢁ10 mL) and dried over Na2SO4.
The solvent was evaporated in vacuo, and the crude product was
purified by PLC (Et2O/CH2Cl2; 9:1) to afford the title compound as
a pale yellow foam (0.020 g, 0.036 mmol, 37%): Rf =0.32 (Et2O/
1
0.23 (EtOAc/petroleum spirits; 2:1); H NMR: d=6.64 (d, J=8.2 Hz,
1H), 6.04–6.01 (m, 3H), 5.88 (s, 1H), 5.78 (td, J=2.6, 0.8 Hz, 1H),
5.31 (d, J=3.6 Hz, 1H), 5.17–5.09 (m, 2H), 4.04 (s, 3H), 3.90–3.83
(m, 1H), 3.83 (s, 3H), 3.82 (s, 3H), 3.36–3.29 (m, 1H), 3.24–3.16 (m,
1H), 2.53–2.46 (m, 1H), 2.35–2.29 (m, 1H), 2.20–2.12 (m, 1H),
1.18 ppm (t, J=7.2 Hz, 3H); 13C NMR: d=159.7, 151.8, 150.1, 143.4,
140.1, 133.3, 132.5, 132.1, 131.7, 119.0, 117.5, 112.3, 100.5, 86.0,
71.9, 70.0, 60.3, 59.7, 56.5, 56.4, 38.3, 35.8, 32.6, 15.8 ppm; HRMS:
m/z [M+H]+ calcd for C24H28IN2O7: 583.0936, found: 583.0928.
1
CH2Cl2; 9:1); H NMR (CD3CN): d=7.38 (m, 1H), 7.09 (m, 3H), 6.85
(m, 1H), 6.54 (m, 1H), 5.86 (m, 2H), 5.47 (m, 3H), 4.96 (m, 2H),
3.70–3.91 (m, 12H), 3.06–3.31 (m, 3H), 2.78–2.98 (m, 1H), 2.52–2.64
(m, 1H), 2.21–2.34 (m, 1H), 0.98–1.12 ppm (m, 3H); 13C NMR
(CD3CN): d=159.3, 158.2, (152.7, 152.6)*, 152.2, 151.8, (150.6,
150.5)*, 141.5, (140.4, 140.3)*, (133.4, 133.3)*, (132.8, 132.6)*, 131.3,
130.9, 130.5, (124.2, 124.2)*, (121.5, 121.4)*, 119.65, (119.0, 118.9)*,
113.5, (112.3, 112.3)*, (102.0, 101.9)*, (87.5, 87.2)*, (71.9, 71.7)*, 60.4,
(60.1, 60.0)*, 57.0, (56.3, 56.1)*, 40.6, (36.3, 36.1)*, (26.2, 26.1)*,
16.1 ppm (* denotes peaks attributed to conformational isomers);
HRMS: m/z [M+H]+ calcd for C31H34N2O8: 563.2388, found:
563.2390.
(R)-9-Amino-5-((S)-4,5-dimethoxy-1,3-dihydroisobenzofuran-1-
yl)-N-ethyl-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-g]isoquino-
line-6(5H)-carboxamide
(11):
Compound
10b
(0.146 g,
0.273 mmol) was dissolved in anhydrous DMSO (0.73 mL) under N2.
CuI (0.052 g, 0.272 mmol), l-proline (0.125 g, 1.088 mmol), and
NaN3 (0.035 g, 0.544 mmol) were sequentially added under an inert
atmosphere, and the reaction was heated in a pressure vessel at
1308C for 8 h. The reaction was quenched with 10% aq NH4Cl, and
the mixture was filtered through Celite. The reaction mixture was
then diluted with water (4 mL) and extracted with CH2Cl2 (3ꢁ
15 mL). The organic layer was washed with water (2ꢁ10 mL) and
brine (2ꢁ10 mL), before being dried over anhydrous Na2SO4, fil-
tered and evaporated in vacuo. The crude product was purified by
flash chromatography (toluene/MeOH; 95:5) to give the title com-
pound as a yellow amorphous solid (0.032 g, 0.068 mmol, 25%):
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-N-ethyl-
4-methoxy-9-(3-methoxyphenyl)-7,8-dihydro-[1,3]dioxolo[4,5-
g]isoquinoline-6(5H)-carboxamide (12c): A degassed solution of
1m aq Na2CO3 (0.33 mL) and THF (0.66 mL) in a microwave vial
was treated with 3-methoxyphenylboronic acid (0.022 g,
0.14 mmol), 10c (0.075 g, 0.13 mmol), and PdCl2(PPh3)2 (0.009 g,
0.013 mmol). The vial was sealed under N2 and heated at 1008C for
3 h. The reaction was cooled to RT, and the mixture was parti-
tioned between EtOAc (10 mL) and water (10 mL). The organic
layer was washed with water (2ꢁ10 mL) and brine (10 mL), and
then dried over MgSO4. The solvent was evaporated in vacuo, and
the crude product was purified by PLC (Et2O/CH2Cl2; 9:1) to afford
the title compound as a white foam (0.015 g, 0.027 mmol, 21%):
1
Rf =0.12 (toluene/MeOH; 95:5); H NMR: d=6.60 (d, J=8.2 Hz, 1H),
6.04 (d, J=8.2 Hz, 1H), 5.95 (s, 3H), 5.90–5.88 (m, 1H), 5.26 (d, J=
3.7 Hz, 1H), 5.22–5.11 (m, 2H), 4.07–4.01 (m, 1H), 3.97 (s, 3H), 3.83
(s, 3H), 3.80 (s, 3H), 3.37–3.27 (m, 3H), 3.26–3.15 (m, 1H), 2.39–2.31
(m, 1H), 2.10–2.04 (m, 1H), 2.01–1.93 (m, 1H), 1.18 ppm (t, J=
7.2 Hz, 3H); 13C NMR: d=159.9, 151.7, 143.4, 135.5, 135.3, 133.1,
132.1, 132.0, 122.6, 117.6, 117.3, 117.0, 112.3, 101.2, 86.0, 71.9, 60.3,
59.9, 56.7, 56.4, 36.7, 35.8, 22.4, 15.9 ppm; HRMS: m/z [M+H]+
calcd for C24H30N3O7: 472.2078, found: 472.2086.
1
Rf =0.25 (Et2O/CH2Cl2; 9:1); H NMR (CD3CN): d=7.35 (t, J=7.9 Hz,
1H), 6.81–6.95 (m, 4H), 6.48 (d, J=7.9 Hz, 1H), 5.89 (d, J=14.6 Hz,
2H), 5.53–5.62 (m, 2H), 5.40–5.46 (m, 1H), 4.92–5.04 (m, 2H), 3.89
(s, 3H), 3.80 (s, 6H), 3.75 (s, 3H), 3.05–3.25 (m, 3H), 2.89–3.01 (m,
1H), 2.35–2.55 (m, 2H), 1.07 ppm (t, J=7.2 Hz, 3H); 13C NMR
(CD3CN): d=161.2, 160.3, 159.8, 157.5, 156.8, 156.4, 153.3, 147.8,
141.1, 137.6, 134.5, 133.9, 130.9, 130.9, 124.1, 119.5, 117.6, 114.3,
114.1, 102.7, 87.9, 72.5, 61.0, 60.8, 57.6, 55.5, 41.6, 36.8, 27.2,
16.6 ppm; HRMS: m/z [M+H]+ calcd for C31H34N2O8: 563.2388,
found: 563.2390.
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-N-ethyl-
4-methoxy-9-phenyl-7,8-dihydro-[1,3]dioxolo[4,5-g]isoquinoline-
6(5H)-carboxamide (12a): Compound 10c (0.094 g, 0.16 mmol),
PdCl2(PPh3)2 (10 mol%) and Cs2CO3 (0.173 g, 0.532 mmol) were dis-
solved in DME (2 mL) and H2O (0.2 mL). Phenylboronic acid
(25.9 mg, 0.212 mmol) was added, and the mixture was heated in
a microwave reactor at 1408C, 300 W for 30 min. The reaction was
then diluted with EtOAc, and the solution was filtered through
a pad of Celite. The filtrate was concentrated in vacuo to give
a crude product that was purified by flash chromatography
(CH2Cl2/MeOH; 99:1) to yield the title compound as an off-white
foam (0.016 g, 0.031 mmol, 19%): 1H NMR: d=7.44–7.40 (m, 2H),
7.37–7.34 (m, 1H), 7.27–7.27 (m, 2H), 6.69 (d, J=8.0 Hz, 1H), 6.15
(d, J=7.6 Hz, 1H), 6.08 (brs, 1H), 5.94 (dd, J=2.8, 1.2 Hz, 2H),
5.73–5.71 (m, 1H), 5.47 (d, J=4.0 Hz, 1H), 5.19–5.10 (m, 2H), 4.07
(s, 3H), 3.85 (s, 3H), 3.84 (s, 3H), 3.52–3.46 (m, 1H), 3.42–3.33 (m,
1H), 3.28–3.19 (m, 1H), 2.52–2.40 (m, 2H), 2.07–2.05 (m, 1H),
1.21 ppm (t, J=7.2 Hz, 3H); 13C NMR: d=160.2, 151.6, 146.2, 143.2,
139.0, 134.3, 133.9, 132.3, 131.8, 130.0 (2C), 129.6, 128.3 (2C), 127.5,
117.8, 117.3, 117.0, 112.0, 100.8, 86.5, 71.7, 60.1, 59.5, 56.4, 56.3,
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-N-ethyl-
4-methoxy-9-(4-methoxyphenyl)-7,8-dihydro-[1,3]dioxolo[4,5-
g]isoquinoline-6(5H)-carboxamide (12d): A degassed solution of
1m aq Na2CO3 (0.33 mL) and THF (0.66 mL) in a microwave vial
was treated with 4-methoxyphenylboronic acid (0.021 g,
0.14 mmol), 10c (0.074 g, 0.13 mmol) and PdCl2(PPh3)2 (0.009 mg,
0.013 mmol). The vial was sealed under N2 and heated to 1008C
for 1 h. The reaction was cooled to RT, and the mixture was parti-
tioned between EtOAc (10 mL) and water (10 mL). The organic
layer was washed with water (2ꢁ10 mL) and brine (10 mL), and
then dried over MgSO4. The solvent was evaporated in vacuo, and
the crude product was purified by PLC (Et2O/petroleum spirits; 9:1)
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 407

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
to afford the title compound as an off-white foam (0.023 g,
0.041 mmol, 31%): Rf =0.14 (Et2O/petroleum spirits; 9:1); 1H NMR
(CD3CN): d=7.22 (d, J=8.4 Hz, 2H), 6.99 (d, J=8.4 Hz, 2H), 6.85 (d,
J=8.2 Hz, 1H), 6.49 (d, J=8.0 Hz, 1H), 5.88 (d, J=15.5 Hz, 2H),
5.53–5.62 (m, 2H), 5.42–5.46 (m, 1H), 4.90–5.04 (m, 2H), 3.87 (s,
3H), 3.82 (s, 3H), 3.80 (s, 3H), 3.75 (s, 3H), 3.06–3.22 (m, 3H), 2.88–
3.00 (m, 1H), 2.36–2.54 (m, 2H), 1.07 ppm (t, J=7.2 Hz, 3H);
13C NMR (CD3CN): d=160.0, 159.7, 152.7, 147.3, 143.9, 135.3, 133.9,
133.3, 132.3, 130.3, 127.6, 120.2, 118.9, 117.7, 114.9, 113.5, 102.0,
87.4, 71.9, 60.4, 60.2, 56.9, 56.0, 54.9, 41.0, 36.3, 26.7, 16.1 ppm;
HRMS: m/z [M+H]+ calcd for C31H34N2O8: 563.2388, found:
563.2390.
117.6, 117.2 (d, J=21.4 Hz), 115.8, 114.6 (d, J=21.0 Hz), 101.1, 86.6,
71.9, 60.3, 59.7, 56.7, 56.5, 39.6, 35.8, 26.3, 15.8 ppm; HRMS: m/z
[M+H]+ calcd for C30H31FN2O7: 551.2188, found: 551.2191.
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-N-ethyl-
9-(4-fluorophenyl)-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-g]iso-
quinoline-6(5H)-carboxamide (12g): A degassed solution of 1m
aq Na2CO3 (0.33 mL) and THF (0.66 mL) in a microwave vial was
treated with 4-fluorophenylboronic acid (0.026 g, 0.18 mmol), 10c
(0.097 g, 0.17 mmol), and PdCl2(PPh3)2 (0.012 g, 0.017 mmol). The
vial was sealed under N2 and heated to 1008C for 23 h. The reac-
tion was cooled to RT, and the mixture was partitioned between
EtOAc (10 mL) and water (10 mL). The organic layer was washed
with water (2ꢁ10 mL) and brine (10 mL), and then dried over
MgSO4. The solvent was evaporated in vacuo, and the crude prod-
uct was purified by PLC (Et2O/petroleum spirits; 7:3) to afford the
title compound as a white foam (0.007 g, 0.012 mmol, 7%): Rf =
(5R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-N-
ethyl-9-(2-fluorophenyl)-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-
g]isoquinoline-6(5H)-carboxamide (12e): A degassed solution of
1m aq Na2CO3 (0.33 mL) and THF (0.066 mL) in a microwave vial
was treated with 2-fluorophenylboronic acid (0.042 g, 0.302 mmol),
10c (0.117 g, 0.202 mmol) and PdCl2(PPh3)2 (0.014 g, 0.02 mmol).
The vial was sealed under N2 and heated to 1008C for 3.5 h. The
reaction was cooled to RT, and the mixture was partitioned be-
tween EtOAc (10 mL) and water (10 mL). The organic layer was
washed with water (2ꢁ10 mL) and brine (10 mL), and then dried
over MgSO4. The solvent was evaporated in vacuo, and the crude
product was purified by PLC (Et2O/petroleum ether; 9:1) to afford
the title compound as an off-white foam (0.036 g, 0.066 mmol,
33%): Rf =0.26 (EtOAc/hexanes; 1:1); 1H NMR: d=7.31–7.39 (m,
1H), 7.10–7.25 (m, 3H), 6.68 (d, J=8.2 Hz, 1H), 6.17 (d, J=8.2 Hz,
1H), 5.91–5.96 (m, 2H), 5.71–5.76 (m, 1H), 5.43–5.48 (m, 1H), 5.08–
5.19 (m, 2H), 4.04–4.11 (m, 3H), 3.83 (s, 3H), 3.81–3.82 (m, 3H),
3.52–3.62 (m, 1H), 3.30–3.40 (m, 1H), 3.18–3.28 (m, 1H), 2.24–2.48
(m, 2H), 1.96–2.06 (m, 1H), 1.18–1.22 ppm (m, 3H); 13C NMR
(CD3CN): d=(160.8 (d, J=281.7 Hz), 160.7 (d, J=281.9 Hz))*, 152.7,
(147.7, 147.6)*, (146.6, 146.5)*, (145.3, 145.3)*, 145.1, 143.9, 143.0,
140.9, 135.5, 135.0, 133.8, (133.7 (d, J=3.3 Hz), 133.4 (d, J=
3.1 Hz))*, 131.1 (d, J=8.2 Hz), 130.7 (125.31 (d, J=3.3 Hz), 119.5
(118.9, 118.9)* (116.6 (d, J=22.4 Hz), 116.5 (d, J=22.2 Hz))*, (113.6,
113.6)*, (102.3, 102.2)*, (87.4, 87.4)*, (72.0, 71.9)*, 60.4, (60.1, 60.1)*,
56.9, 55.0, 40.6, 36.3, 26.3, 16.0 ppm; HRMS: m/z [M+H]+ calcd for
C30H32FN2O7: 551.2188, found: 551.2191.
1
0.21 (Et2O/petroleum spirits; 7:3); H NMR (CD3CN): d=7.31 (dd, J=
8.9, 5.6 Hz, 2H), 7.18 (dd, J=8.9, 8.9 Hz, 2H), 6.85 (d, J=8.2 Hz,
1H), 6.50 (d, J=8.1 Hz, 1H), 5.86–5.93 (m, 2H), 5.52–5.64 (m 2H),
5.42–5.47 (m, 1H), 4.90–5.05 (m, 2H), 3.88 (s, 3H), 3.80 (s, 3H), 3.75
(s, 3H), 3.05–3.23 (m, 3H), 2.89–3.00 (m, 1H), 2.39–2.51 (m, 2H),
1.07 ppm (t, J=7.2 Hz, 3H); 13C NMR (CD3CN): d=164.3, 160.7 (d,
J=222.3 Hz), 152.7, 147.3, 143.9, 140.5, 135.1, 133.8, 133.3, 133.1
(d, J=8.2 Hz), 131.7, 130.2, 120.3, 118.9, 116.1 (d, J=21.6 Hz), 113.5,
102.1, 87.4, 71.9, 60.4, 60.1, 56.9, 54.9, 40.9, 36.3, 26.6, 16.1 ppm;
HRMS: m/z [M+H]+ calcd for C30H31FN2O7: 551.2188, found:
551.2191.
(S)-3-((R)-9-Chloro-4-methoxy-5,6,7,8-tetrahydro-[1,3]dioxolo[4,5-
g]isoquinolin-5-yl)-6,7-dimethoxyisobenzofuran-1(3H)-one (13):
Compound 7 (0.60 g, 1.50 mmol) was dissolved in trifluoroacetic
acid (TFA) (14 mL), and the solution was stirred. NCS (0.220 g,
1.65 mmol) was slowly added portionwise over 1 h, and the reac-
tion was heated at reflux overnight. The reaction was allowed to
cool and then diluted with water (20 mL) and basified with 2.5m
aq NaOH. Extraction with CH2Cl2 (3ꢁ20 mL) afforded the crude
product that was purified by flash chromatography (EtOAc/petrole-
um spirits; 5:1) to give the title compound as a pale yellow foam
(0.221 g, 0.509 mmol, 34%): Rf =0.38 (EtOAc/petroleum spirits;
5:1); 1H NMR: d=6.93 (d, J=8.3 Hz, 1H), 6.01 (t, J=4.1 Hz, 3H),
5.82 (d, J=4.0 Hz, 1H), 4.75 (d, J=4.0 Hz, 1H), 4.04 (s, 3H), 3.99 (s,
3H), 3.81 (s, 3H), 3.69–2.63 (m, 1H), 2.52–2.45 (m, 1H), 2.30–
2.12 ppm (m, 3H); 13C NMR: d=168.4, 152.3, 148.0, 145.1, 140.9,
139.2, 134.7, 129.4, 119.3, 118.8, 118.6, 117.4, 107.6, 101.6, 80.3,
62.3, 59.7, 56.7, 52.9, 38.9, 26.2 ppm; HRMS: m/z [M+H]+ calcd for
C21H21ClNO7: 434.1001, found: 434.1003.
(R)-5-((S)-4,5-Dimethoxy-1,3-dihydroisobenzofuran-1-yl)-N-ethyl-
9-(3-fluorophenyl)-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-g]iso-
quinoline-6(5H)-carboxamide (12 f): A degassed solution of 1m
aq Na2CO3 (0.330 mL) and THF (0.660 mL) in a microwave vial was
treated with 3-fluorophenylboronic acid (0.036 g, 0.26 mmol), 10c
(0.100 g, 0.172 mmol) and PdCl2(PPh3)2 (0.012 g, 0.026 mmol). The
vial was sealed under N2 and heated to 1008C for 2.75 h. The reac-
tion was cooled to RT, and the mixture was partitioned between
EtOAc (10 mL) and water (10 mL). The organic layer was washed
with water (2ꢁ10 mL) and then brine (10 mL). The organic layer
was collected and dried over MgSO4. The solvent was evaporated
in vacuo to afford a brown solid. The crude product was purified
by flash column chromatography (petroleum spirits/EtOAc; 3:1) to
afford the title compound as a white foam (0.045 g, 0.082 mmol,
48%): Rf =0.28 (petroleum spirits/EtOAc; 3:1); 1H NMR: d=7.33–
7.40 (m, 1H), 7.00–7.07 (m, 2H), 6.94–6.99 (m, 1H), 6.66–6.71 (m,
1H), 6.12 (d, J=7.9 Hz, 1H), 5.91–5.96 (m, 2H), 5.67–5.71 (m, 1H),
5.46 (d, J=3.8 Hz, 1H), 5.08–5.20 (m, 2H), 4.06 (s, 3H), 3.84 (s, 3H),
3.83 (s, 3H), 3.40–3.50 (m, 1H), 3.31–3.40 (m, 1H), 3.17–3.28 (m,
1H), 2.48–2.58 (m, 1H), 2.36–2.46 (m, 1H), 1.93–2.02 (m, 1H),
1.20 ppm (t, J=7.2 Hz, 3H); 13C NMR: d=162.7 (d, J=246.4 Hz),
161.6, 160.4, 151.9, 146.3, 143.4, 139.5, 136.6 (d, J=8.0 Hz), 134.0,
132.5, 131.8, 129.9 (d, J=8.6 Hz), 129.7, 125.9 (d, J=2.9 Hz), 117.8,
(R)-9-Chloro-5-((S)-4,5-dimethoxy-3-oxo-1,3-dihydroisobenzofur-
an-1-yl)-N-ethyl-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-g]isoqui-
noline-6(5H)-carboxamide (14): Compound 13 (0.200 g,
0.461 mmol) was dissolved in MeCN (5 mL), and the solution was
cooled to ꢁ58C under N2. Ethyl isocyanate (0.039 g, 0.553 mmol)
was then added, and the solution was allowed to warm to RT. The
reaction was complete in 2 h, and then quenched with H2O (1 mL)
before being extracted with CH2Cl2 (2ꢁ10 mL). The organic layers
were collected, dried over anhydrous Na2SO4, filtered and evaporat-
ed in vacuo. The crude product was purified by flash chromatogra-
phy (EtOAc/petroleum spirits; 3:1) to give the title compound as
a pale yellow foam (0.200 g, 0.396 mmol, 86%): Rf =0.16 (EtOAc/
1
petroleum spirits; 3:1); H NMR: d=7.06 (d, J=8.3 Hz, 1H), 6.69 (d,
J=8.3 Hz, 1H), 5.96 (dd, J=5.3, 1.2 Hz, 2H), 5.88 (d, J=3.2 Hz, 1H),
5.67 (d, J=3.2 Hz, 1H), 4.99 (brs, 1H), 4.00 (s, 3H), 3.84 (s, 3H),
3.82 (s, 3H), 3.67–3.61 (m, 1H), 3.34–3.16 (m, 2H), 2.71 (t, J=6.2 Hz,
2H), 2.46–2.43 (m, 1H), 1.15 ppm (t, J=7.2 Hz, 3H); 13C NMR: d=
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 408

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
167.1, 158.6, 152.6, 148.1, 145.6, 139.9, 138.5, 134.5, 128.9, 118.8,
118.4, 118.2, 116.8, 107.1, 101.7, 81.3, 62.4, 59.4, 56.9, 53.5, 38.4,
36.0, 24.6, 15.6 ppm; HRMS: m/z [M+H]+ calcd for C24H26ClN2O8:
505.1372, found: 505.1372.
cancer cell line PANC-1 (CRL-1469) were purchased directly from
American Type Culture Collection (ATCC) and cultured in Ham’s F-
12K (Kaighn’s) medium (Invitrogen 21127-022), minimum essential
medium alpha (MEMa) (Invitrogen 12561-056) and Dulbecco’s
modified eagle medium (DMEM) (Invitrogen 10566–016), respec-
tively. In addition, each culture medium was supplemented with
10% foetal bovine serum (FBS) (Invitrogen 10100-147) containing
100 U of penicillin–streptomycin (Invitrogen 15140-122). Cells were
maintained at 378C in a humidified atmosphere containing 5%
CO2. Before use, cell lines were tested for the presence of myco-
plasma by using a MycoAlert mycoplasma detection kit (Lonza
LT07-318) according to the manufacturer’s instructions; cells em-
ployed in the assays tested negative for mycoplasma.
(R)-9-Chloro-N-ethyl-5-((S)-4-hydroxy-5-methoxy-3-oxo-1,3-dihy-
droisobenzofuran-1-yl)-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-
g]isoquinoline-6(5H)-carboxamide (15): A solution of MeMgBr in
THF/toluene (0.318 mL, 1.4m, 0.445 mmol) was slowly added to
a stirred solution of anhydrous BnOH (0.182 mL, 1.749 mmol) in an-
hydrous toluene (0.5 mL). The mixture was vigorously stirred for
5 min at RT, and then heated to 1208C. Compound 14 (0.160 g,
0.318 mmol) was added portionwise, and the resulting mixture
was stirred at 1208C for 4 h before being allowed to cool to RT.
The reaction mixture was acidified using a saturated citric acid so-
lution. The solution was then diluted with water (2 mL), followed
by the addition of EtOAc/hexane (1:1, 20 mL) to remove any excess
BnOH. The aqueous layer was basified using saturated aq Na2CO3
and then extracted with CH2Cl2 (3ꢁ10 mL). The organic layers were
collected, dried over anhydrous Na2SO4, filtered and evaporated in
vacuo. The crude material was purified by flash chromatography
(EtOAc/petroleum spirits; 9:1) to give the desired title compound
as a yellow oil (0.100 g, 0.204 mmol, 64%): Rf =0.14 (EtOAc/petrole-
Cell-cycle assay: Cells were detached by 0.05% trypsin–EDTA (Invi-
trogen 15400054) and seeded at 104 cells per well in 24-well plates
48 h prior to treatment. Cells were then treated with experimental
analogues at 10 mm, vincristine at 100 nm (reference agent), or ve-
hicle control for 16 h before detachment by 0.5% trypsin–EDTA.
Cells were then fixed with 4% paraformaldehyde (PFA) for 15 min
before being washed with phosphate-buffered saline (PBS). Cells
were permeablised with 0.5% Triton X-100 in PBS for 30 min. Cells
were then washed again with PBS and stained with CellCycle 405-
blue (Invitrogen A10034) for 30 min. The different stages of the cell
proliferation cycle were detected by a FACSCanto II fluorescence-
activated cell sorting (FACS) analyser (BD Biosciences, Australia)
using the pacific blue channel (450 nm).
1
um spirits; 9:1); H NMR: d=7.52–7.31 (m, 1H), 7.00 (d, J=8.1 Hz,
1H), 6.47 (d, J=8.1 Hz, 1H), 5.97–5.96 (m, 3H), 5.71 (d, J=3.4 Hz,
1H), 5.01–4.86 (m, 1H), 3.87 (s, 3H), 3.85 (s, 3H), 3.68–3.62 (m, 1H),
3.34–3.17 (m, 2H), 2.72 (t, J=6.1 Hz, 2H), 2.51–2.47 (m, 1H),
1.15 ppm (t, J=7.2 Hz, 3H); 13C NMR: d=170.2, 158.5, 147.3,
145.69, 145.50, 138.6, 138.3, 134.6, 128.9, 118.4, 116.8, 114.7, 112.2,
107.2, 101.7, 82.9, 59.4, 56.9, 53.1, 38.5, 36.0, 24.6, 15.6 ppm; HRMS:
m/z [M+H]+ calcd for C23H24ClN2O8: 491.1216, found: 491.1212.
EC50 determination: EC50 values were determined for all compounds
that were observed to cause arrest of cells in the G2/M phase. PC3,
PANC-1 and MCF-7 cells were seeded at 2000 cells per well in 96-
well plates 48 h before treatment. Cells were then treated with var-
ious concentrations of the test analogues. After 5 days of treat-
ment, cell viability was detected by CellTiter-Blue assay (Promega
G8080) according to the manufacturer’s protocol. Briefly, a fluores-
cent product was produced by viable cells and detected using an
Envision microplate reader (PerkinElmer) at 560 nm excitation/
590 nm emission. Increased fluorescence indicated the presence of
larger numbers of viable cells. Fluorescence values were plotted
against log concentration of inhibitor, and EC50 values were calcu-
lated using GraphPad Prism statistical software (version 6).
(R)-9-Chloro-N-ethyl-5-((S)-4-hydroxy-5-methoxy-1,3-dihydroiso-
benzofuran-1-yl)-4-methoxy-7,8-dihydro-[1,3]dioxolo[4,5-g]iso-
quinoline-6(5H)-carboxamide (16): A solution of NaBH4 (0.02 g,
0.529 mmol) in dry THF (3 mL) was treated dropwise with a solution
of 15 (0.088 g, 0.179 mmol) in BF3·Et2O (0.88 mL) at ꢁ58C. The mix-
ture was stirred at ꢁ58C for 1 h and then allowed to warm to RT
and stirred overnight. The reaction was poured onto ice, quenched
with cold 10% aq HCl (4 mL), and allowed to stir for a further 1 h.
The reaction mixture was then basified to pH 8 with 2.5m aq
NaOH and extracted with CHCl3 (3ꢁ10 mL). The organic phase was
dried over anhydrous Na2SO4 and filtered. Evaporation of solvent in
vacuo afforded the crude product, which was purified by flash
chromatography (CH2Cl2/MeOH/CH3COOH; 95:4:1) to afford the
title compound as yellow oil (0.020 g, 0.042 mmol, 23%): Rf =0.15
Cell cytotoxicity assay: MCF7 breast cancer cells were obtained
from the ATCC and routinely grown in DMEM supplemented with
10% foetal calf serum (FCS) and penicillin/streptomycin at 378C
with 5% CO2. The ABCG2-expressing derivative cell line MCF7FLV1000
was grown under identical conditions with every fourth passage
supplemented with 0.5 nm flavopiridol to maintain selection pres-
sure. The ABCB1-expressing derivative cell line was grown as de-
scribed for MCF7 cells, with every third passage supplemented
with 3 mm doxorubicin to maintain selection pressure. All cells
were maintained up to a maximum of 20 passages.
1
(CH2Cl2/MeOH/CH3COOH; 95:5:1%); H NMR: d=6.69 (d, J=8.5 Hz,
1H), 6.57 (d, J=8.5 Hz, 1H), 6.50–6.40 (m, 1H), 6.01 (dd, J=13.1,
1.4 Hz, 2H), 5.86 (d, J=3.9 Hz, 1H), 5.02 (d, J=3.9 Hz, 1H), 4.96–
4.85 (m, 1H), 4.54 (d, J=12.5 Hz, 1H), 4.34 (d, J=12.6 Hz, 1H), 4.05
(s, 3H), 3.86 (s, 3H), 3.32–3.20 (m, 3H), 3.10–3.00 (m, 1H), 2.71–2.65
(m, 1H), 1.45–1.32 (m, 1H), 1.13 ppm (t, J=7.2 Hz, 3H); 13C NMR:
d=160.2, 146.4, 145.2, 144.7, 138.9, 134.9, 131.8, 127.8, 125.5,
120.5, 120.3, 109.8, 106.1, 101.8, 59.9, 56.8, 56.2, 2ꢁ56.1, 41.3, 36.2,
23.4, 15.6 ppm; HRMS: m/z [M+H]+ calcd for C23H26ClN2O7:
477.1423, found: 477.1413.
Cells were seeded in flat-bottomed 96-well plates at a density of
1000 cells per 100 mL per well (determined by counting with a hae-
mocytometer) and allowed to attach to the surface for 24 h at
378C. At this time, test compounds and reference drugs were
added from a 2ꢁ stock solution in 100 mL medium in the concen-
tration range 10ꢁ12 to 10ꢁ4 m. All compounds evaluated in this
assay were obtained from a stock solution in DMSO, and the final
solvent concentration in the wells was 0.2% (v/v). Test compounds
and cells were left for a period of 6 days at 378C at which point
the control wells reached confluence. The number of viable cells
was determined using a standard MTT assay for cell cytotoxicity.
Briefly, 20 mL of a 5 mgmLꢁ1 solution was added to each well, and
Pharmacology
Compounds preparation: For each compound, a 10ꢁ2 m stock solu-
tion was prepared by dissolving the compound (~1 mg) in DMSO
(Sigma D2650), and the solution was stored at ꢁ208C until use.
Cell culture and reagents: The human prostate cancer cell line PC3
(CRL-1435), breast cancer cell line MCF-7 (HTB-22) and pancreatic
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 409

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[4] M. A. Jordan, Curr. Med. Chem. Anti-Cancer Agents 2002, 2, 1–17.
[5] S. Aggarwal, N. N. Ghosh, R. Aneja, H. C. Joshi, R. Chandra, Helv. Chim.
Acta 2002, 85, 2458–2462.
[6] R. Aneja, S. N. Vangapandu, M. Lopus, V. G. Viswesarappa, N. Dhiman, A.
Verma, R. Chandra, D. Panda, H. C. Joshi, Biochem. Pharmacol. 2006, 72,
415–426.
the plates were incubated for 4 h at 378C. Medium and MTT were
then aspirated, wells were washed with PBS, and the formazan
crystals were dissolved in 150 mL DMSO. The absorbance in each
well (l=570 nm) was measured using an iMARK 96-well plate
reader. The absorbance in the control wells was assigned a value
of 100%, and all absorbances were normalised to this value. The
proportion of viable cells was plotted as a function of added com-
pound concentration, and the EC50 value to effect cell death was
estimated from the dose–response curve.
[7] R. Aneja, S. N. Vangapandu, H. C. Joshi, Bioorg. Med. Chem. 2006, 14,
8352–8358.
[8] J. T. Anderson, A. E. Ting, S. Boozer, K. R. Brunden, J. Danzig, T. Dent, J. J.
Harrington, S. M. Murphy, R. Perry, A. Raber, S. E. Rundlett, J. Wang, N.
Wang, Y. L. Bennani, J. Med. Chem. 2005, 48, 2756–2758.
[9] J. T. Anderson, A. E. Ting, S. Boozer, K. R. Brunden, C. Crumrine, J.
Danzig, T. Dent, L. Faga, J. J. Harrington, W. F. Hodnick, S. M. Murphy, G.
Pawlowski, R. Perry, A. Raber, S. E. Rundlett, A. Stricker-Krongrad, J.
Wang, Y. L. Bennani, J. Med. Chem. 2005, 48, 7096–7098.
[10] R. C. Mishra, P. Karna, S. R. Gundala, V. Pannu, R. A. Stanton, K. K. Gupta,
M. H. Robinson, M. Lopus, L. Wilson, M. Henary, R. Aneja, Biochem. Phar-
macol. 2011, 82, 110–121.
Acknowledgements
The authors thank the Australian Research Council and GlaxoS-
mithKline Australia for financial support. The contributions of
R.C. and D.M. were funded by an Association for International
Cancer Research project grant number 12-0008.
[11] A. J. DeBono, J. H. Xie, S. Ventura, C. W. Pouton, B. Capuano, P. J. Scam-
mells, ChemMedChem 2012, 7, 2122–2133.
[12] N. K. Manchukonda, P. K. Naik, S. Santoshi, M. Lopus, S. Joseph, B. Srid-
har, S. Kantevari, PLoS One 2013, 8, e77970.
[13] J. T. Markiewicz, O. Wiest, P. Helquist, J. Org. Chem. 2010, 75, 4887–
4890.
[14] N. K. Manchukonda, B. Sridhar, P. K. Naik, H. C. Joshi, S. Kantevari, Bioorg.
Med. Chem. Lett. 2012, 22, 2983–2987.
Keywords: antitumor agents · cytotoxic activity · natural
products
·
noscapine analogues
·
structure–activity
relationships
[1] A. Mooraki, A. Jenabi, M. Jabbari, M. I. Zolfaghari, S. Z. Javanmardi, M.
Mahmoudian, B. Bastani, Nephrology 2005, 10, 348–350.
[2] M. S. Segal, M. M. Goldstein, E. O. Attinger, Dis. Chest. 1957, 32, 305–
309.
Received: October 8, 2013
[3] K. Ye, Y. Ke, N. Keshava, J. Shanks, J. A. Kapp, R. R. Tekmal, J. Petros, H. C.
Joshi, Proc. Natl. Acad. Sci. USA 1998, 95, 1601–1606.
Published online on December 11, 2013
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 399 – 410 410