Synthesis and biological evaluation of 1-phenyl-1,2,3,4-dihydroisoquinoline compounds as tubulin polymerization inhibitors

Article abstract of DOI:10.1002/ardp.201100169

A series of 1-phenyl-3,4-dihydroisoquinoline derivatives and several 1-phenyl-1,2,3,4-tetrahydroisoquinoline, 1-phenyl-isoquinoline analogues were synthesized, and their cytotoxicity and tubulin polymerization inhibitory activity were evaluated. The 1-phenyl-3,4-dihydroisoquinoline compounds were found to be potential tubulin polymerization inhibitors. Compound 5n, bearing a 3′-OH and 4′-OCH₃ substituted 1-phenyl B-ring, was shown to confer optimal bioactivity. The single-crystal structure of 5n was further determined by X-ray diffraction, and the binding mode of 5n to tubulin was obtained by molecular docking, which can explain the structure-activity relationships. The studies presented here provide a new structural type for the development of novel antitumor agents.

Full text of DOI:10.1002/ardp.201100169

Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
1
Full Paper
Synthesis and Biological Evaluation of 1-Phenyl-1,2,3,4-
dihydroisoquinoline Compounds as Tubulin Polymerization
Inhibitors
Can-Hui Zheng1*, Jun Chen2*, Jia Liu1, Xiao-Tian Zhou1, Na Liu1, Duo Shi1, Jing-Jing Huang1,
Jia-Guo Lv1, Ju Zhu1, and You-Jun Zhou1
1 School of Pharmacy, Second Military Medical University, Shanghai, China
2 Chengdu General Hospital, Chengdu Military Command Area, Chengdu, China
A series of 1-phenyl-3,4-dihydroisoquinoline derivatives and several 1-phenyl-1,2,3,4-tetrahydro-
isoquinoline, 1-phenyl-isoquinoline analogues were synthesized, and their cytotoxicity and
tubulin polymerization inhibitory activity were evaluated. The 1-phenyl-3,4-dihydroisoquinoline
compounds were found to be potential tubulin polymerization inhibitors. Compound 5n, bearing
a 30–OH and 40–OCH3 substituted 1-phenyl B-ring, was shown to confer optimal bioactivity. The
single-crystal structure of 5n was further determined by X-ray diffraction, and the binding mode of
5n to tubulin was obtained by molecular docking, which can explain the structure–activity
relationships. The studies presented here provide a new structural type for the development of
novel antitumor agents.
Keywords: 1-Phenyl-1,2,3,4-dihydroisoquinoline / Antitumor agent / Tubulin polymerization inhibitor
Received: May 3, 2011; Revised: December 31, 2011; Accepted: January 13, 2012
DOI 10.1002/ardp.201100169
Introduction
a South African willow tree Combretum caffrum, binds to the
colchicine site and exerts potent cytotoxicity against a wide
variety of human cancer cell lines including MDR cancer cell
lines. However, CA-4 does not show efficacy in vivo due to its
low aqueous solubility and the isomerization of its cis-double
bond to the more thermally stable but inactive trans-isomer [4].
A more soluble prodrug, CA-4 phosphate disodium (CA-4P), has
displayed promising results in human cancer clinical trials [5],
fueling continued research in novel CSIs.
Microtubule arrangement and rearrangement affects many
aspects of cellular structure and function. Investigations into
the polymerization dynamics that regulate microtubule effi-
cacy and in turn cellular replication may prove useful in the
development of anticancer therapeutics [1]. Three small-
molecule binding sites of microtubules, including the vinca,
taxane, and colchicine sites, are well characterized. The col-
chicine site is located on the a/b-tubulin interface, and
inhibitors binding to this site disrupt the polymerization
of tubulins into microtubules [2].
CSIs include both natural and synthetic small molecules
[2, 4–7]. Though the molecular diversity among them is
enormous, all known CSIs have a common pharmacophore.
The most important pharmacophore points are two hydro-
phobic centers (H1 and H2), a single planar group (R1),
and two hydrogen bond acceptors (A1 and A2; Scheme 1).
Figure 1A employs colchicine to show the interaction of the
pharmacophore points with the tubulin structure. H2 is
often characterized by the trimethoxyphenyl (TMP) moiety,
and can foster Van der Waals interactions with the hydro-
phobic P1 pocket in the interface between a/b-tubulin. H1 is
typically represented by a methoxy carbon atom and can
The first colchicine site inhibitor (CSI; Scheme 1) was
extracted from the poisonous meadow saffron Colchicum autum-
nale L. and its high toxicity has limited its therapeutic appli-
cations [3]. Combretastatin A-4 (CA-4; Scheme 1), isolated from
Correspondence: You-Jun Zhou and Ju Zhu, School of Pharmacy,
Second Military Medical University, Shanghai 200433, China.
E-mail: zhouyoujun2006@yahoo.com.cn; zhuju@smmu.edu.cn
Fax: þ86 021 81871240
Abbreviations: colchicine site inhibitor (CSI); combretastatin A-4 (CA-4);
CA-4 phosphate disodium (CA-4P); trimethoxyphenyl (TMP).
* These authors contributed equally to this work.
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

2
C.-H. Zheng et al.
Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
Scheme 1. Two-dimensional structures of
four CSIs.
foster the Van der Waals interactions with the hydrophobic
P2 pocket together with R1. Accompanying these hydro-
phobic contacts are two potential hydrogen bonds: A2 to
the sulfur atom of Cys241 in the b subunit and A1 to the
amide nitrogen of Val181 in the a subunit. Furthermore, the
five pharmacophoric points can be partitioned among two
planes. Points H1, R1, and A1 lie in plane A, and points H2
and A2 lie in plane B. These two planes together form a tilt
and match the colchicine site [8, 9].
be expected to lead to activity on the colchicine site of tubulin
polymerization in the case of the similarly substituted 1,2,3,4-
dihydroisoquinoline. Moreover, derivatives of such a structure
may be convenient to synthesize.
In this paper, a series of 1-phenyl-1,2,3,4-dihydroisoquino-
line derivatives and several 1-phenyl-1,2,3,4-tetrahydroisoqui-
noline, 1-phenyl-isoquinoline analogues were synthesized,
and their cytotoxicity and tubulin polymerization inhibitory
activity were evaluated. The binding mode of 1,2,3,4-dihydro-
isoquinoline derivatives to tubulin was also studied by
molecular docking.
Several derivatives of neoflavonoid, which is primarily iso-
lated from natural sources [10], were prepared by Bailly et al.
[11]. The neoflavonoid derivative 1 (Scheme 1) was found to
match the common pharmacophore of CSIs, avoiding inacti- Results and discussion
vation from cis-to-trans isomerization of the double bond of
combretastatin derivatives. It was proved to exert potent tubu-
lin polymerization inhibition and cytotoxicity against a CEM
leukemia cell line. The structural similarity between 1-phenyl-
1,2,3,4-dihydroisoquinoline (Scheme 1) and neoflavonoid can
Chemistry
The chemical synthesis of these compounds was outlined in
Scheme 2. As key intermediates of our designed compounds 2
and 3 were synthesized by the slight modification of the
Figure 1. Binding modes of colchicine (A) and 5n (B) to tubulin protein.
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
Tubulin Polymerization Inhibitors
3
Scheme 2. Synthesis of subject compounds.
known procedure [12, 13]. Treatment of 3,4,5-trimethoxybenz-
aldehyde with nitromethane, ammonium acetate, and acetic
acid yielded intermediate 2 in 73.8% yield. Then reduction of
2 by LiAlH4 in THF and treated with HCl gas provided inter-
mediate 3 in 77.7% yield. Then 3 was allowed to react with
different benzoyl chlorides 4 (a–m) to form different amides,
then POCl3 was added to provide the products of cyclization 5
(a–m) in medium yield (50–65%). Deacetylation of 5m by
NaOH in CH3CH2OH/H2O yielded 5n. The resulting 3,4-dihy-
droisoquinoline compounds 5n, 5d, and 5g were reduced in
excellent yield (85–95%) by NaBH4 in CH3OH to provide
1,2,3,4-tetrahydroisoquinoline compounds 6n, 6d, and 6g.
Compounds 5n and 5g were aromatized in good yield (80–
90%) by tetrahydronaphthalene with PdꢀC to provide isoqui-
noline compounds 7n and 7g. Reduction of 5l with Raney Ni
with NH2NH2ꢀH2O in ethanol yielded 5o.
hydroxy group on the 1-phenyl B-ring is critical to maintain-
ing cytotoxic activity. The potent compound 5n bears 30–OH
groups, while 30-unsubstituted, 30–OCH3 and 30–NO2 substi-
tuted compounds, such as 5b, 5c, and 5d, showed signifi-
cantly less activity. Another factor that was found to affect
cytotoxicity is the 40 position of the 1-phenyl B-ring. This was
evidenced by compound 5b, which has a 40–OCH3 substituted
1-phenyl B-ring and was found to be slightly more active than
5a, 5e, 5g, 5l, or 5o, which have 40-unsubstituted, 40–CH3, 40–F,
40–NO2, or 40–NH2 substituted 1-phenyl B-rings. The most
potent compound described in this paper, 5n, has a 30–OH
and 4–OCH3-substituted 1-phenyl B-ring, the same substi-
tution pattern as CA-4 and the neoflavonoid derivative 1.
The cytotoxicities of several compounds with halogen- and
nitro-substituted 1-phenyl B-rings were also evaluated.
Compound 5j, having a 20,50-dichloro-substituted 1-phenyl
B-ring, showed the best cytotoxic activity among these
compounds. However, the structure–activity relationship
of them was not clear, which should be with the help of
more derivatives synthesized in the future study. In addition,
the 3,4-dihydroisoquinoline scaffold of compounds 5n
was found to be critical to maintaining cytotoxic activity.
1,2,3,4-Tetrahydroisoquinoline analogues 6n and isoquino-
line analogues 7n resulted in a very significant loss of activity.
To test whether these compounds would bind to tubulin
and inhibit its polymerization, representative 3,4-dihydro-
isoquinoline compound 5n and the same substituted 1,2,3,4-
tetrahydroisoquinoline and isoquinoline compounds 6n and
7n were subjected to evaluation by tubulin assembly assay
(Table 2). CA-4 was also evaluated by this assay as positive
control. The extent of inhibition by these compounds
Pharmacology
The structures and cytotoxicity of target compounds are
reported in Table 1 [14]. All compounds showed stronger
cytotoxic activity against the CEM human leukemia cell line
than against the LOVO human colon carcinoma cell line by
MTT assay. All of the compounds except compound 5n, which
showed strong cytotoxicity, exhibited middle-to-low cytotoxic
activity levels against these two cell lines. The following dis-
cussion of the structure–activity relationships are based on the
activity in the CEM cell line, because these compounds were
more sensitive against this cell line than the LOVO cell line.
Among the 3,4-dihydroisoquinoline compounds, the
characteristics of 1-phenyl B-ring (Table 1) showed themselves
to be most significant for producing cytotoxic effects. A 30-
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

4
C.-H. Zheng et al.
Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
Cytotoxicity (IC50, mM)
Table 1. Cytotoxicity of subject compounds
Compound structure
No.
R1
R2
R3
R4
R5
CEM
LOVO
5a
5b
5c
5d
5e
5f
5g
5h
5i
5j
5k
5l
5n
5o
H
H
H
H
H
H
H
F
H
Cl
H
H
H
H
H
H
OCH3
NO2
H
CH3
H
H
Cl
H
NO2
H
OH
H
H
OCH3
OCH3
OCH3
CH3
H
F
F
Cl
H
H
H
H
H
H
H
H
H
H
H
NO2
H
H
H
H
H
H
H
H
H
H
H
H
Cl
H
H
H
H
44.00
33.17
53.36
41.33
43.32
53.22
95.48
39.15
28.10
11.74
51.07
48.28
2.73
>100
93.78
>100
52.64
100.0
33.08
>100
51.93
13.73
10.68
45.64
92.36
13.37
>100
O
O
A
N
O
R5
R4
R1
R2
B
H
R3
NO2
OCH3
NH2
45.01
O
O
6n
6d
6g
H
H
H
OH
NO2
H
OCH3
OCH3
F
H
H
H
H
H
H
31.07
77.11
80.45
>100
>100
87.85
A
NH
R1
O
R5
R4
B
R2
R3
O
O
A
N
7n
7g
H
H
OH
H
OCH3
F
H
H
H
H
31.81
79.92
50.04
97.50
O
R5
R4
R1
R2
B
R3
Colchicine
/
/
/
/
/
/
0.017
0.10
IC50 values are the means of at least three independent determinations (SD <10%).
increased monotonically with their concentration. The IC50
of 5n in 11.4 mM, and the inhibition activities of compounds
6n and 7n were found to be significantly lower than that of
5n. Considering that compound 5n, which showed strong
cytotoxicity, clearly inhibited tubulin assembly, and that 6n
and 7n, which showed little cytotoxicity, showed very little
effect, it seems that tubulin represents a potential target for
3,4-dihydroisoquinoline compounds.
Because the inhibition of tubulin polymerization has been
implicated in G2/M phase cell cycle arrest in various cancer
cell lines, the effects of compound 5n on the cell cycle
progression on LOVO cells was determined (Fig. 2). The results
of flow cytometry analysis indicate 7n induces an evident
accumulation of cells in the G2/M phase (65% compared to
14% for untreated cells) at 50 mM (Fig. 2B).
Binding mode of 3,4-dihydroisoquinoline derivatives
To validate observed structure–activity relationships, mol-
ecular docking studies of the most potent 3,4-dihydroisoqui-
noline compound, 5n, to tubulin were performed. One
conformation of the X-ray crystal structure of 5n (Fig. 3A,
Table 2. In vitro antitubulin effect of target compounds
Compounds
5n
6n
7n
CA-4
IC50 (mM)
11.4
72.2
74.2
1.77
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
Tubulin Polymerization Inhibitors
5
tubulin protein, this may be the reason why the 3,4-dihydro-
isoquinoline scaffold of compounds 5n is critical to main-
taining cytotoxic activity.
Conclusion
This paper describes the synthesis of a series of 3,4-dihydroi-
soquinoline derivatives and several 1,2,3,4-tetrahydroisoquino-
line, isoquinoline analogues, and evaluates their cytotoxicity
and tubulin polymerization inhibitory activity levels. The 3,4-
dihydroisoquinoline compounds seem to be potential tubulin
polymerization inhibitors. Compound 5n in particular,
bearing a 30–OH and 40–OCH3 substituted 1-phenyl B-ring,
confers optimal bioactivity. The single-crystal structure of 5n
was further determined by X-ray diffraction, and the binding
mode of 5n to tubulin protein was obtained by molecular
docking, which explains the structure–activity relationships.
The studies presented here provide a new structural type for
the development of novel antitumor agents.
Figure 2. Cell cycle distribution determined by flow cytometry on
LOVO cells treated for 24 h without any compounds (A) or with
compound 5n at 50 mM (B).
Number CCDC 747709) was selected for molecular docking.
The X-ray crystal structure of the DAMA–colchicine–tubulin
complex (PDB code 1SA0) was used as the tubulin protein
template. The predictive binding mode of 5n (Fig. 1B) to
tubulin protein was found to be similar to that of colchicine
derived from the X-ray crystal structure 1SA0 (Fig. 1A). Based
on its binding mode, 5n also matched the common pharma-
cophore of known CSIs (Scheme 1). The TMP moiety of 5n
(corresponding to pharmacophoric point H2) can foster Van
der Waals interactions with the hydrophobic P1 pocket in the
interface between a/b-tubulin. The –OCH3 substituted on it
(corresponding to A2) can form hydrogen bonds with Cys241
in b subunits. Its 1-phenyl B-ring of 5n and 40–OCH3 (corre-
sponding to R1 and H1) can foster Van der Waals interactions
with the hydrophobic P2 pocket. The 30–OH substituted on
1-phenyl B-ring (corresponding to A1) can form hydrogen
bonds with Val181 in a subunits, which explains its import-
ant function in maintaining cytotoxic activity. In addition,
the two planes in which the TMP moiety and 1-phenyl B-ring
of 5n lie, respectively, have a similar tilt with that of colchi-
cine in the binding mode. There are differences in the tilt
between 3,4-dihydroisoquinoline compounds and 1,2,3,4-tet-
rahydroisoquinoline, isoquinoline analogues, which can be
inferred from the comparison of the X-ray crystal structures
of 5n and 7n (Fig. 3B, Number CCDC 747708). Considering
that the central ring does not directly interact with the
Experimental
Chemistry
Melting points were determined on an electrically heated RK-Z
melting point apparatus and are uncorrected. Mass spectra (MS)
were measured on a Micromass Qtof-Micro LC-MS instrument.
1HNMR spectra were recorded at 300 MHz on a Bruker AC-300P
spectrometer with Me4Si as the internal standard. Chemical
shifts are given in ppm (d), and the spectral data are consistent
with the assigned structures. Infrared (IR) spectra were recorded
on a PE Spectrum One FI-IR spectrometer as KBr pellets. The MS
were recorded on a Micromass Qtof-Micro LC-MS instrument.
silica gel column chromatography was performed with silica
gel 60 G (Qindao Haiyang Chemical, China). Commercial solvents
and reagents were of reagent grade and, when necessary, were
purified and dried by standard protocols. Organic solutions were
dried over anhydrous sodium sulfate. The purity of final com-
pounds was assessed on the basis of analytical HPLC, and the
results were greater than 95%.
Figure 3. X-ray crystal structures of compound 5n (A) and compound 7n (B).
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

6
C.-H. Zheng et al.
Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
6.83–6.86 (d, 1H, J ¼ 8.4 Hz, ar-H), 6.62 (s, 1H, ar-H), 3.94 (s, 3H,
–OCH3), 3.91 (s, 3H, –OCH3), 3.91 (s, 3H, –OCH3), 3.81 (s, 3H,
–OCH3), 3.66–3.70 (m, 2H, –CH2–), 3.43 (s, 3H, –OCH3),
2.62–6.66 (t, 2H, –CH2–); ESI-MS m/z 358.38 (Mþ1)þ; HRMS calcd
for C20H23NO5 357.1576, found 357.1587.
1,2,3-Trimethoxy-5-(2-nitrovinyl)benzene (2)
Compound (2) was obtained by the slight modification of the
known procedure as follows [12, 13]: A mixture of 3,4,5-trimethoxy-
benzaldehyde (66.7 g, 0.34 mol), nitromethane (54 mL, 1 mol),
ammonium acetate (25 g), and acetic acid (150 mL) was refluxed
3 h. Then the reaction mixture was poured into ice water and
filtered. The filter cake was washed with water three times and
recrystallized from isopropanol to yield 2 (6g, 73.8%) as yellow
needles: mp 120–1238C (lit. mp 125.5–126.58C) [13].
6,7,8-Trimethoxy-1-(4-methoxy-3-nitrophenyl)-3,4-
dihydroisoquinoline (5d)
Light yellow crystal, 62.6%, mp 176–1778C; 1H NMR (CDCl3,
300 MHz, d): 8.45–8.48 (m, 1H, ar-H), 7.92–7.93 (d, 1H,
J ¼ 1.8 Hz, ar-H), 7.35–7.38 (m, 1H, ar-H), 6.71 (s, 1H, ar-H),
4.08 (s, 3H, –OCH3), 4.03 (s, 3H, –OCH3), 3.96 (s, 2H, –CH2–),
3.82 (s, 3H, –OCH3), 3.61 (s, 3H, –OCH3), 3.02–3.06 (t, 2H,
–CH2–); ESI-MS m/z 373.38 (Mþ1)þ; HRMS calcd for C19H20N2O6
372.1321, found 372.1330.
2-(3,4,5-Trimethoxyphenyl)ethanamine (3) hydrochloride
Compound (3) was obtained by the slight modification of the
known procedure as follows [12, 13]: LiAlH4 (10 g) in tetrahydro-
furan was added dropwise to a solution of 2 (5.0 g) in dry
tetrahydrofuran in ice bath, and the mixture was refluxed for
5 h. After cooling at room temperature, 10% aqueous NaOH was
added dropwise to the mixture to quench the reaction. After
filtration and evaporation of solvent, the resulting oil was dis-
solved in ether (50 mL) and treated with HCl gas. The resulting
solid was filtered and recrystallized from ethanol and acetone to
afford 3 hydrochloride (3.8 g, 77.7%) as white needles: mp 183–
1848C (lit. mp 1828C) [12].
6,7,8-Trimethoxy-1-p-tolyl-3,4-dihydroisoquinoline (5e)
hydrochloride
Light yellow crystal, 59.5%, mp 143–1458C; 1H NMR (CDCl3,
300 MHz, d): 7.60–7.63 (dd, 2H, ar-H), 7.20–7.23 (dd, 2H, ar-H),
6.44 (s, 1H, ar-H), 3.91 (s, 3H, –OCH3), 3.78 (s, 3H, –OCH3),
3.67–3.73 (dd, 2H, –CH2–), 3.37 (s, 3H, –OCH3), 2.70–2.74 (t, 2H,
–CH2–), 2.39 (s, 3H, –CH3–); ESI-MS m/z 312.19 (Mþ1)þ; HRMS
calcd for C19H21NO3 311.1521, found 311.1532.
General procedure for the preparation of 5 (a–l)
A mixture of 3 hydrochloride (2.9 g, 0.012 mol), triethylamine
(5 mL), and dichloromethane (25 mL) was stirred at room tempera-
ture for 30 min. Then a solution of 4 (a–l; 0.012 mol) in dichloro-
methane was added dropwise and the mixture was stirred at room
temperature for 4 h. After evaporation of the solvent, the residue
was dissolved in dry acetonitrile (50 mL). Then POCl3 (10 mL) was
added and the mixture was refluxed for a further 4 h. After evap-
oration of the solvent, the residue was dissolved in dichlorome-
thane (50 mL), washed with saturated solution of Na2CO3
(3 ꢁ 10 mL) and water (3 ꢁ 10 mL) and dried over Na2SO4.
Distillation of the solvent under reduced pressure and recrystalli-
zation of the resulting solid yielded compounds 5 (a–l; 50–65%).
6,7,8-Trimethoxy-1-m-tolyl-3,4-dihydroisoquinoline (5f)
hydrochloride
Light yellow crystal, 65.9%, mp 132–1358C; 1H NMR (CDCl3,
300 MHz, d): 7.33 (s, 1H, ar-H), 7.23–7.27 (m, 2H, ar-H),
7.16–7.19 (m, 1H, ar-H), 6.61 (s, 1H, ar-H), 3.94 (s, 3H, –OCH3),
3.80 (s, 3H, –OCH3), 3.68–3.73 (m, 2H, –CH2–), 3.39 (s, 3H, –OCH3),
2.63–2.68 (t, 2H, –CH2–), 2.37 (s, 3H, –CH3–); ESI-MS m/z 312.17
(Mþ1)þ; HRMS calcd for C19H21NO3 311.1521, found 311.1531.
1-(4-Fluorophenyl)-6,7,8-trimethoxy-3,4-
dihydroisoquinoline (5g)
Light yellow crystal, 53.8%, mp 126–1278C; 1H NMR (CDCl3,
300 MHz, d): 7.45–7.50 (m, 2H, ar-H), 7.01–7.07 (m, 2H, ar-H),
6.61 (s, 1H, ar-H), 3.93 (s, 3H, –OCH3), 3.81 (s, 3H, –OCH3),
3.67–3.72 (m, 2H, –CH2–), 3.39 (s, 3H –OCH3), 2.62–2.67 (dd,
2H, –CH2–); ESI-MS m/z: 316.69 (Mþ1)þ; HRMS calcd
for C18H18FNO3 315.1271, found 315.1278.
6,7,8-Trimethoxy-1-phenyl-3,4-dihydroisoquinoline (5a)
hydrochloride
Light yellow crystal, 52.6%, mp 170–1728C; 1H NMR (CDCl3,
300 MHz, d): 7.45–7.49 (m, 2H, ar-H), 7.35–7.38 (m, 3H, ar-H),
6.61 (s, 1H, ar-H), 3.94 (s, 3H, –OCH3), 3.80 (s, 3H, –OCH3),
3.69–3.74 (m, 2H, –CH2–), 3.37 (s, 3H, –OCH3), 2.63–2.68 (dd,
2H, –CH2–); ESI-MS m/z 298.71 (Mþ1)þ; HRMS calcd
for C18H19NO3 297.1365, found 297.1376.
1-(2,4-Difluorophenyl)-6,7,8-trimethoxy-3,4-
dihydroisoquinoline (5h) hydrochloride
Light yellow crystal, 55.4%, mp 190–1928C; 1H NMR
(CDCl3, 300 MHz, d): 7.84–7.92 (m, 1H, ar-H), 7.12–7.15
(m, 1H, ar-H), 6.85–6.92 (m, 1H, ar-H), 6.66 (s, 1H, ar-H), 4.89
(s, 2H, –CH2–), 4.01 (s, 3H, –OCH3), 3.79 (s, 3H, –OCH3), 3.67–
3.72 (m, 2H, –CH2–), 3.55 (s, 3H –OCH3), 3.02 (s, 2H, –CH2–); ESI-MS
m/z: 334.38 (Mþ1)þ; HRMS calcd for C18H17F2NO3 333.1176,
found 333.1185.
6,7,8-Trimethoxy-1-(4-methoxyphenyl)-3,4-
dihydroisoquinoline (5b) hydrochloride
Brown crystal, 59.1%, mp 157–1588C; 1H NMR (CDCl3, 300 MHz,
d): 7.75 (s, 2H, ar-H), 7.02 (s, 2H, ar-H), 6.70 (s, 1H, ar-H), 4.01 (s, 3H,
–OCH3), 3.93 (s, 3H, –OCH3), 3.89 (s, 2H, –CH2–), 3.82 (s, 3H,
–OCH3), 3.52 (s, 3H, –OCH3), 2.99 (s, 2H, –CH2–); ESI-MS
m/z 328.43 (Mþ1)þ; HRMS calcd for C19H21NO4 327.1471, found
327.1483.
1-(3,4-Dichlorophenyl)-6,7,8-trimethoxy-3,4-
1-(3,4-Dimethoxyphenyl)-6,7,8-trimethoxy-3,4-
dihydroisoquinoline (5c)
White crystal, 51.2%, mp 142–1438C; H NMR (CDCl3, 300 MHz,
d): 7.14–7.15 (d, 1H, J ¼ 1.8 Hz, ar-H), 7.01–7.05 (dd, 1H, ar-H),
dihydroisoquinoline (5i)
Light yellow crystal, 57.6%, mp 110–1128C; 1H NMR (CDCl3,
300 MHz, d): 7.59–7.60 (d, 1H, J ¼ 2.1, ar-H), 7.44–7.46 (d, 1H,
J ¼ 8.1, ar-H), 7.32–7.35 (dd, 1H, ar-H), 6.62 (s, 1H, ar-H), 3.95
1
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
Tubulin Polymerization Inhibitors
7
(s, 3H, –OCH3), 3.81 (s, 3H, –OCH3), 3.70–3.75 (dd, 2H, –CH2–),
3.46 (s, 3H, –OCH3), 2.64–2.69 (t, 2H, –CH2–); ESI-MS m/z: 367.41
(Mþ1)þ; HRMS calcd for C18H17Cl2NO3 365.0585, found
365.0592.
3059, 3015, 2986, 2959, 2939, 2895, 2834, 2717, 1591, 1556,
1529; ESI-MS m/z 344.59 (Mþ1)þ; HRMS calcd for C19H21NO5
343.142, found 343.1430.
4-(6,7,8-Trimethoxy-3,4-dihydroisoquinolin-1-yl)aniline
(5o) hydrochloride
1-(2,6-Dichlorophenyl)-6,7,8-trimethoxy-3,4-
dihydroisoquinoline (5j)
Catalytic amount Raney Ni was added slowly into a solution of 4l
(0.006 mol) in methanol (10 mL) and 85% NH2NH2 H2O (2.5 mL),
and the reaction mixture was stirred at room temperature for
30 min. After evaporation of solvent, the residue was dissolved in
5% aqueous HCI solution and washed with ethyl acetate
(3 ꢁ 5 mL). Then the pH value of the solution was adjusted to
13–14, extracted with dichloromethane and dried over Na2SO4.
After filtration, the filtrate was treated with HCl gas. The resulting
solid was filtered and recrystallized from ethanol and acetone to
provide 6o hydrochloride (85%) as orange crystal: mp 2158C, dec.;
1H NMR (CDCl3, 300 MHz, d): 7.37–7.38 (d, 1H, J ¼ 1.8 Hz, ar-H),
7.35 (d, 1H, J ¼ 1.8 Hz, ar-H), 6.67–6.68 (d, 1H, J ¼ 1.8 Hz, ar-H),
6.65–6.66 (d, 1H, J ¼ 1.8 Hz, ar-H), 6.60 (s, 1H, ar-H), 3.93 (s, 3H,
–OCH3), 3.82 (s, 3H, –OCH3), 3.64–3.68 (m, 2H, –CH2–), 3.43 (s, 3H,
–OCH3), 2.60–2.65 (t, 2H, –CH2–); ESI-MS m/z: 312.40 (Mþ1)þ; HRMS
calcd for C18H20N2O3 312.1474, found 312.1482.
Light yellow crystal, 59.5%, mp 147–1498C; 1H NMR (CDCl3,
300 MHz, d): 7.35–7.36 (d, 1H, J ¼ 0.9 Hz, ar-H), 7.30 (d, 1H,
J ¼ 0.9 Hz, ar-H), 7.17–7.20 (m, 1H, ar-H), 6.59 (s, 1H, ar-H),
3.93 (s, 3H, –OCH3), 3.84–3.89 (m, 2H, –CH2–), 3.77 (s, 3H,
–OCH3), 3.35 (s, 3H, –OCH3), 2.76–2.81 (t, 2H, –CH2–); ESI-MS:
m/z: 366.81 (Mþ1)þ; HRMS calcd for C18H17Cl2NO3 365.0585,
found 365.0593.
1-(3,5-Dinitrophenyl)-6,7,8-trimethoxy-3,4-
dihydroisoquinoline (5k) hydrochloride
Brown crystal, 64.3%, mp 211–2138C; 1H NMR (CDCl3, 300 MHz,
d): 9.19 (s, 1H, ar-H), 8.90 (s, 2H, ar-H), 6.77 (s, 1H, ar-H), 4.15 (s, 2H,
–CH2–), 4.06 (s, 3H, –OCH3), 3.81 (s, 3H, –OCH3), 3.51 (s, 3H,
–OCH3), 3.15 (s, 2H, –CH2–); ESI-MS m/z: 388.85 (Mþ1)þ; HRMS
calcd for C18H17N3O7 387.1066, found 387.1073.
General procedure for the preparation of 6n, 6d and 6g
0.24 g (0.0064 mol) NaBH4 was added slowly into a solution of
5n, 5d, or 5g (0.0021 mol) in methanol (5 mL) in an ice bath, and
the mixture was refluxed for 2 h. After evaporation of the
solvent, the residue was dissolved in water (5 mL), extracted with
ether (3 ꢁ 10 mL), and dried over Na2SO4. Distillation of the
solvent under reduced pressure and recrystallization of the
residue yielded compounds 6n, 6d, or 6g (85–95%).
6,7,8-Trimethoxy-1-(4-nitrophenyl)-3,4-
dihydroisoquinoline (5l) hydrochloride
Brown crystal, 60.2%, mp 180–1828C; 1H NMR (CDCl3, 300 MHz,
d): 8.30–8.37 (m, 2H, ar-H), 7.79–7.83 (m, 2H, ar-H), 6.72 (s, 1H,
ar-H), 4.03 (s, 3H, –OCH3), 3.96–3.99 (m, 2H, –CH2–), 3.80 (s, 3H,
–OCH3), 3.47 (s, 3H, –OCH3), 2.99–3.04 (t, 2H, –CH2–); ESI-MS
m/z: 343.57 (Mþ1)þ; HRMS calcd for C18H18N2O5 342.1216, found
342.1224.
2-Methoxy-5-(6,7,8-trimethoxy-1,2,3,4-
2-Methoxy-5-(6,7,8-trimethoxy-3,4-dihydroisoquinolin-1-
yl) phenol (5n)
A mixture of 3 hydrochloride (2.9 g, 0.012 mol), triethylamine
(5 mL), and dichloromethane (25 mL) was stirred at room
tetrahydroisoquinolin-1-yl) phenol (6n)
Light yellow crystal, 88.6%, mp 174–1768C; 1H NMR (CDCl3,
300 MHz, d): 6.76–6.79 (d, 1H, J ¼ 8.1 Hz, ar-H), 6.71–6.72
(d, 1H, J ¼ 2.1 Hz, ar-H), 6.64–6.67 (m, 1H, ar-H), 6.47 (s, 1H, ar-
H), 5.19 (s, 1H, >CH–), 3.87 (s, 3H, –OCH3), 3.86 (s, 3H, –OCH3), 3.79
(s, 3H, –OCH3), 3.40 (s, 3H, –OCH3), 2.68–3.04 (m, 4H, –CH2–CH2–);
13C NMR (CDCl3, 75 MHz, d): 152.4, 150.6, 145.5, 145.2, 140.3,
138.7, 130.9, 123.3, 119.7, 114.6, 110.2, 107.2, 60.6, 59.9, 55.9,
55.0, 38.0, 29.1; IR (KBr): 3447, 3315, 3002, 2956, 2936, 2915,
2838, 1582, 1505; ESI-MS: m/z: 346.2 (Mþ1)þ; HRMS calcd for
C19H23NO5 345.1576, found 345.1586.
temperature for 30 min. Then
a solution of 4m (2.7 g,
0.012 mol) in dichloromethane was added dropwise and the
mixture was stirred at room temperature for 4 h. After evapor-
ation of the solvent, the residue was dissolved in dry acetonitrile
(50 mL). Then POCl3 (10 mL) was added and the mixture was
refluxed for a further 4 h. After evaporation of the solvent, the
residue was dissolved in dichloromethane (50 mL), washed with
saturated solution of Na2CO3 (3 ꢁ 10 mL) and water (3 ꢁ 10 mL)
and dried over Na2SO4. The solvent was distilled under reduced
pressure and the residue was added to ethanol (25 mL) and water
(8 mL) with NaOH (1.6 g). The reaction mixture was stirred at
room temperature for 2 h. Then the pH value of the solution was
adjusted to 4–5. After evaporation of the solvent, the resulting
solid was dissolved in ethyl acetate and dried over Na2SO4.
Distillation of the solvent under reduced pressure and recrystal-
lization of the resulting solid from petroleum ether and ethyl
acetate yielded 5n (2.3 g, 57%) as light yellow crystal: mp 132–
6,7,8-Trimethoxy-1-(4-methoxy-3-nitrophenyl)-1,2,3,4-
tetrahydroisoquinoline (6d)
Light yellow crystal, 92.1%, mp 2118C, dec.; 1H NMR (CDCl3,
300 MHz, d): 7.86 (b, 1H, ar-H), 7.27 (s, 1H, ar-H), 7.15–7.17
(d, 1H, J ¼ 6 Hz, ar-H), 6.51 (s, 1H, ar-H), 5.78 (s, 1H, >CH–),
3.96 (s, 3H, –OCH3), 3.89 (s, 3H, –OCH3), 3.78 (s, 3H, –OCH3),
3.50 (s, 3H, –OCH3), 3.14–3.33 (m, 2H, –CH2–), 3.00–3.05
(m, 2H, –CH2–); ESI-MS: m/z: 375.2 (Mþ1)þ; HRMS calcd for
C19H22N2O6 374.1478, found 374.1483.
1
1348C; H NMR (CDCl3, 300 MHz, d) 7.27 (s, 1H, ar-H), 7.11–7.12
(d, 1H, J ¼ 1.8 Hz, ar-H), 7.00–7.03 (dd, 1H, ar-H), 6.82–6.85 (d, 1H,
J ¼ 1.8 Hz, ar-H), 6.60 (s, 1H, ar-H), 3.94 (s, 3H, –OCH3), 3.91 (s, 3H,
–OCH3), 3.81 (s, 3H, –OCH3), 3.67–3.70 (dd, 2H, –CH2–), 3.45 (s, 3H,
–OCH3), 2.61–2.66 (t, 2H, –CH2–); 13C NMR (CDCl3, 75 MHz, d):
165.3, 155.5, 152.4, 147.4, 145.2, 140.9, 137.3, 134.8, 119.2, 115.8,
114.1, 109.8, 105.9, 61.4, 61.0, 56.0, 55.9, 47.0, 27.8; IR (KBr): 3283,
1-(4-Fluorophenyl)-6,7,8-trimethoxy-1,2,3,4-
tetrahydroisoquinoline (6g)
White crystal, 92.3%, mp 171–1738C; H NMR (CDCl3, 300 MHz,
d): 9.88 (s, 1H, N–H), 7.31–7.36 (m, 2H, ar-H), 7.03–7.09 (t, 2H, ar-H),
1
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

8
C.-H. Zheng et al.
Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
6.49 (s, 1H, ar-H), 5.71 (s, 1H, >CH–), 3.88 (s, 3H, –OCH3), 3.78
(s, 3H, –OCH3), 3.45–3.50 (m, 2H, –CH2–), 3.38 (s, 3H, –OCH3), 3.16
(s, 2H, –CH2–); ESI-MS: m/z: 318.78 (Mþ1)þ; HRMS calcd for
C18H20FNO3 317.1427, found 317.1439.
warmed 96-well plates at 378C in the presence of tested
compounds at varying concentrations. The reaction was
started by warming the samples at 378C. The mass of polymer
formed was monitored by turbidimetry at 340 nm every
1 min for 45 min with a BioTek’s Synergy 4 multifunction
microplate reader.
General procedure for the preparation of 7n and 7g
A catalytic amount of Pd/C was added into a solution of 5n or 5c
(0.0032 mol) in tetrahydronaphthalene (20 mL), and the mixture
was refluxed for 6 h. After evaporation of solvent, the residue was
purified by column chromatography and eluent petroleum
ether/ethyl acetate (8:2) to afford compounds 7n or 7g (80–90%).
X-ray crystal structure determination
For each of compounds 5n and 7n, a single crystal of suitable
dimensions was selected for X-ray diffraction determination.
The data were collected at room temperature on a Bruker
Smart APEX diffractometer equipped with a normal focus,
2.4 kW sealed tube X-ray source (Mo Ka radiation,
2-Methoxy-5-(6,7,8-trimethoxyisoquinolin-1-yl) phenol
(7n)
1
White crystal, 87.3%, mp 155–1578C; H NMR (CDCl3, 300 MHz,
˚
l ¼ 0.71073 A) operating at 50 kV and 30 mA. A hemisphere
d): 8.41–8.43 (d, 1H, J ¼ 5.7 Hz, ar-H), 7.45–7.47 (d, 1H, J ¼ 5.4 Hz,
ar-H), 7.07 (d, 1H, J ¼ 1.8 Hz, ar-H), 6.96 (m, 2H, ar-H), 6.88–6.91
(d, 1H, J ¼ 8.4 Hz, ar-H), 4.02 (s, 3H, –OCH3), 3.93 (s, 3H, –OCH3),
3.91 (s, 3H, –OCH3), 3.41 (s, 3H, –OCH3); 13C NMR (CDCl3, 75 MHz,
d): 157.5, 156.4, 150.0, 146.3, 144.5, 142.7, 141.3, 136.8, 135.7,
120.6, 118.7, 118.1, 115.6, 109.7, 101.5, 61.2, 61.1, 56.0; IR (KBr):
3447, 3422, 3079, 3039, 2995, 2930, 2901, 2833, 2787, 2688,
2555, 1610, 1577, 1548, 1510; ESI-MS: m/z: 342.2 (Mþ1)þ;
HRMS calcd for C19H19NO5 341.1263, found 341.1265.
of intensity data was collected with a scan width of 0.608 in v
and exposure time of 10 s per frame. An empirical absorption
correction based on the SADABS program was applied. The
structure was solved by direct methods followed by succes-
sive difference Fourier methods. All calculations were per-
formed using SHELXS and SHELXTL-97 program [17]. All non-
H atoms were refined anisotropically. The hydrogen atoms
were located by difference Fourier synthesis and refined geo-
metrically. The molecular graphics were created by using
SHELXTL. Crystallographic data for the structure in this
paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication
numbers CCDC 747709 (compound 5n) and CCDC 747708
(compound 7n).
1-(4-Fluorophenyl)-6,7,8-trimethoxyisoquinoline (7g)
Light yellow crystal, 81.7%, mp 127–1298C; 1H NMR (CDCl3,
300 MHz, d): 8.44–8.47 (m, 1H, ar-H), 7.46–7.51 (m, 3H, ar-H),
7.41–7.43 (m, 1H, ar-H), 7.09–7.15 (m, 1H, ar-H), 6.99 (s, 1H,
ar-H), 4.04 (s, 3H, –OCH3), 3.92–3.93 (m, 3H, –OCH3), 3.33–3.35
(m, 3H, –OCH3); ESI-MS: m/z: 314.64 (Mþ1)þ; HRMS calcd for
C18H16FNO3 313.1114, found 313.1120.
Biology
Molecular docking
All calculations were performed on an Origin 300 Server. One
conformation of the X-ray crystal structure of 5n (Fig. 3,
Number CCDC 747709) was selected for molecular docking.
The X-ray crystal structure of the DAMA–colchicine–tubulin
complex (PDB code 1SA0) was used as the tubulin protein
template. The AutoDock program was used [18]. A grid of
45 ꢁ 40 ꢁ 40 was used to ensure that the area probed was
adequate for the ligands to explore all possible binding
modes. Docking runs were performed using the
Larmarckian genetic algorithm (LGA) with some modifi-
cations of the docking parameters. Each docking experiment
was derived from 50 different runs that were set to terminate
after a maximum of 10 000 000 energy evaluations or
370 000 generations, yielding 50 docked conformations. At
the end of any docking job with multiple runs, AutoDock
performed cluster analysis. Docking solutions with ligand all-
In vitro cytotoxicity assay
Cytotoxic effects were examined in the CEM human leukemia
and the LOVO human colon carcinoma cell lines. Cells in
logarithmic phase were diluted to a density of 40 000–
50 000 cells/mL in culture medium based on growth charac-
teristics. For each well of a 96-well microplate, 100 mL of cell
dilution was seeded, allowed to attach overnight, and then
exposed to varying concentrations (10ꢂ4–10ꢂ9 M) of com-
pounds for 72 h (378C, 5% CO2 atmosphere). The number
of living cells was estimated by MTT assay. Absorbance at
570 nm was recorded on a Wellscan MK-2 multifunction
microplate reader (Labsystems Inc.). Compounds were tested
in triplicate in at least three independent assays. The IC50
values were determined by a nonlinear regression analysis.
Average values were reported.
˚
atom RMSDs within 0.5 A of each other were clustered
In vitro tubulin polymerization assay [11, 15, 16]
Bovine brain tubulin was purchased from Cytoskeleton.
Tubulin (>99% pure, 2 mg/mL) in 100 mL of general tubulin
buffer (80 mM Na-Pipes, pH 6.9, 1 mM EGTA, 1 mM MgCl2,
1 mM GTP, and 5% glycerol) at 08C was placed in a pre-
together and ranked by the lowest energy representative.
Low-energy binding modes and the lowest energy confor-
mation from clusters of five or more occurrences were
selected for further refinement. These were then refined by
SYBYL [19], with the binding mode that had the lowest com-
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2012, 000, 1–9
Tubulin Polymerization Inhibitors
9
[8] T. L. Nguyen, C. McGrath, A. R. Hermone, J. C. Burnett, D. W.
Zaharevitz, B. W. Day, P. Wipf, E. Hamel, R. Gussio, J. Med.
Chem. 2005, 48, 6107–6116.
plex energy after minimization being selected as the final
binding mode.
[9] Y. W. Li, Y. J. Zhou, J. Zhu, C. H. Zheng, J. Chen, C. Q. Sheng,
J. G. Lv, H. Tang, Y. N. Li, J. Zhang, J. Acta Chim. Sin. 2008, 66,
1735.
The work was supported by the Key Program for Basic Research of
Shanghai ‘‘Technology Innovation Action Plan’’ (Grant No.
09JC1417500), National Natural Science Foundation of China (Grant
No. 21172260), Shanghai Natural Science Foundation (Grant No.
09ZR1438800), and National Natural Science Foundation of China
(Grant No. 30901859).
[10] G. D. Monache, B. Botta, A. S. Netoa, R. A. Limaa,
Phytochemistry 1983, 22, 1657–1658.
[11] C. Bailly, C. Bal, P. Barbier, S. Combes, J. P. Finet, M. P.
Hildebrand, V. Peyrot, N. Wattez, J. Med. Chem. 2003, 46,
5437–5444.
The authors have declared no conflict of interest.
[12] T. Ohshita, H. Ando, Jpn. J. Toxicol. Environ. Health 1992, 38,
571–580.
[13] D. A. Dauzonne, Chem. Pharm. Bull. 1986, 34, 1628–
1633.
References
[1] M. A. Jordan, L. Wilson, Nat. Rev. Cancer 2004, 4, 253–265.
[2] N. H. Nam, Curr. Med. Chem. 2003, 10, 1697–1722.
[14] Y. J. Zhou, J. Zhu, J. Chen, W. W. Zhu, Y. W. Li, C. H. Zheng, H.
Tang, J. G. Lv, Y. N. Li, (The Second Military Medical
University). CN101607938. 2010.
[3] G. M. Tozer, C. Kanthou, B. C. Baguley, Nat. Rev. Cancer 2005,
5, 423–435.
[15] E. Hamel, Cell Biochem. Biophys. 2003, 38, 1–22.
[16] Q. Zhang, Y. Peng, X. I. Wang, S. M. Keenan, S. Arora, W. J.
Welsh, J. Med. Chem. 2007, 50, 749–754.
[4] A. Chaudhary, S. N. Pandeya, P. Kumar, P. P. Sharma, S.
Gupta, N. Soni, K. K. Verma, G. Bhardwaj, Mini-Rev. Med. Chem.
2007, 7, 1186–1205.
[17] SHELXS 97, Program for X-ray Crystal Structure Solution
and Refinement, University of Go¨ttingen, Germany.
[5] G. C. Tron, T. Pirali, G. Sorba, F. Pagliai, S. Busacca, A. A.
Genazzani, J. Med. Chem. 2006, 49, 3033–3044.
[18] G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey, W. E.
Hart, R. K. Belew, A. J. Olson, J. Comp. Chem. 1998, 19, 1639–
1662.
[6] V. Srivastava, A. S. Negi, J. K. Kumar, M. M. Gupta, S. P. S.
Khanuja, Bioorg. Med. Chem. 2005, 13, 5892–5908.
[19] SYBYL, release version 6.9, Tripos Associates, Inc., St. Louis,
Missouri.
[7] Y. W. Li, J. Liu, N. Liu, D. Shi, X. T. Zhou, J. G. Lv, J. Zhu, C. H.
Zheng, Y. J. Zhou, Bioorg. Med. Chem. 2011, 19, 3579–3584.
ß 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com