Synthesis and evaluation of 3-aroylindoles as anticancer agents: Metabolite approach

Article abstract of DOI:10.1021/jm900060s

BPR0L075 (2) is a potential anticancer drug candidate designed from Combretastatin A-4 (1) based on the bioisosterismprinciple.Metabolites of 2, proposed from in vitrohumanmicrosome studies,were synthesized, leading to the identification of metabolite-der

Full text of DOI:10.1021/jm900060s

J. Med. Chem. 2009, 52, 4941–4945 4941
DOI: 10.1021/jm900060s
Synthesis and Evaluation of 3-Aroylindoles as Anticancer Agents: Metabolite Approach
Yu-Shan Wu,†,‡,# Mohane Selvaraj Coumar,†,# Jang-Yang Chang,§,# Hsu-Yi Sun,† Fu-Ming Kuo,†
Ching-Chuan Kuo,§ Ying-Jun Chen,† Chi-Yen Chang,§ Chia-Ling Hsiao,† Jing-Ping Liou, Ching-Ping Chen,†
Hsien-Tsung Yao,† Yi-Kun Chiang,† Uan-Kang Tan,^ Chiung-Tong Chen,† Chang-Ying Chu,† Su-Ying Wu,†
Teng-Kuang Yeh,† Chin-Yu Lin,† and Hsing-Pang Hsieh*,†
†Division of Biotechnology and Pharmaceutical Research, National Health Research Institutes, 35 Keyan Road, Zhunan, Miaoli County 350,
Taiwan, Republic of China, ‡Department of Chemistry, Tunghai University, 181 Taichung Harbor Road Section 3, Taichung 407, Taiwan, ROC,
§National Institute of Cancer Research, National Health Research Institutes, Tainan 704, Taiwan, ROC, College of Pharmacy, Taipei Medical
University, Taipei 110, Taiwan, ROC, and ^Department of Chemical and Materials Engineering, Technology and Science Institute of Northern
Taiwan, Taipei 112, Taiwan, ROC. # These authors contributed equally to this work.
Received January 16, 2009
BPR0L075 (2) is a potential anticancer drug candidate designed from Combretastatin A-4 (1) based on the
bioisosterism principle. Metabolites of 2, proposed from in vitro human microsome studies, were synthesized,
leading to the identification of metabolite-derived analogue 10 with 40-350 pM potency against various
cancer cell lines. Insights gained from the major inactive metabolite of 2 led to the development of 29, with
better pharmacokinetics and improved potency in the tumor xenograft model than 2.
Introduction
cases in which inactive metabolites are formed, appropriate
structural modification could decrease the drug metabolism,
resulting in improved drug exposure.
The high costs, long development times, and high failure
rate typically associated with bringing drugs to the market
have prompted drug-discovery scientists to look more criti-
cally and at an earlier stage at one of the key factors for drug
candidate failure until a decade ago: drug metabolism and
pharmacokinetics (PKa).1,2 For example, in 1991, approxi-
mately 40% of thedrugattrition rate was attributedtoadverse
PK issues; in 2000, however, PK issues contributed to only
10% of drug attrition.2 This improvement, effected by a better
understanding of the PK properties of the candidate drugs,
helped to eliminate risky candidates at an early stage of drug
development. In view of this, drug metabolism, one of the
factors affecting the PK profile of a drug, and metabolite
identification have become integral parts of early drug dis-
covery programs.3 Drugs undergo a series of biotransforma-
tion reactions inside the body, in which more water-soluble
metabolites are formed so that the body can remove these
xenobiotic substances more easily. These metabolites could
result in pharmacological or toxicological activity or inactiv-
ity. Although in most cases the metabolism of drugs leads to
pharmacological inactivation, active metabolites have also
been detected occasionally. Some of the active metabolites of
marketed drugs have in fact been developed into new drugs
such as acetaminophen and fexofenadine.3 The metabolism of
parent compounds leading to the formation of pharmacolo-
gically active metabolites may therefore be used as an advan-
tage during lead optimization in drug development. Also, in
Our research group has worked on the design and synthesis
of antimitotic agents.4-6 On the basis of bioisosteric replace-
ment of the olefinic linker and B ring in the Combretastatin
A-4 (CA4, 1) skeleton, 6-methoxy-3-(30,40,50-trimethoxy-
benzoyl)-1H-indole (2, BPR0L075) was designed (Figure 1).5
This compound, 2, not only exhibited potent in vitro activity
against a variety of human tumor cell lines, at iv doses of
50 mg/kg, it also showed activity against xenograft tumor
models of gastric carcinoma MKN-45 and human cervical
carcinoma KB in nude mice.7 We are planning a phase I
clinical trial to assess compound 2’s potential as an anticancer
agent. Recent studies using liquid chromatography tandem
mass spectrometry (LC/MS/MS) on the in vitro metabolic
fate of antitumor agent 2 indicated three metabolites were
derived from hydroxylation on the indole moiety and that
three other metabolites were derived from O-demethyla-
tion.8,9 The structural identity of five out of the six possible
metabolites were confirmed unequivocally as 4-8 by compar-
ison of the HPLC retention time and MS/MS data of the
metabolites with synthetic samples.8 Here we report the
synthesis of various possible phase I metabolites of 2 and
their biological evaluation. Identification of the active meta-
bolite of 2 led to the design and synthesis of analogue with
potent in vitro cytotoxicity in the picomolar range. In
addition, on the basis of the insight gained about the
major metabolic elimination pathways of 2, we have ident-
ified compound 29, which has better in vitro and in vivo
pharmacokinetic profiles and improved in vivo efficacy
in the tumor xenograft model than that of parent com-
pound 2.
*To whom correspondence should be addressed. Phone: þ886-37-
246-166 ext. 35708. Fax: þ886-37-586-456. E-mail, hphsieh@nhri.org.
tw.
a Abbreviations: AUC, area under curve; CA4, Combretastatin A-4;
CL, clearance; iv, intravenous; H460, human lung cancer cell line; HT-
29, colorectal carcinoma cell line; KB, human cervical carcinoma cell
line; LCMS, liquid chromatography coupled mass spectrometry; MKN-
45, gastric carcinoma cell line; NCS, N-chlorosuccinimide; NIS, N-
iodosuccinimide; PK, pharmacokinetics; SAR, structure-activity rela-
tionship; t1/2, plasma half-life.
Chemistry
Possible metabolites of 2, both demethylated and hydro-
xylated, are listed in Figure 1. We previously reported the
r
2009 American Chemical Society
Published on Web 07/08/2009
pubs.acs.org/jmc

4942 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 15
Wu et al.
Scheme 2a
a Reagents and conditions: (a) NO2BF4, CH3NO2, CH2Cl2, -40 ꢀC,
50%. (b) NaOAc, Ac2O, 60 ꢀC, 82%. (c) KN(TMS)2, CH3NO2, THF, -
78 ꢀC. (d) MsCl, CH2Cl2, 0 ꢀC, 36% over two steps. (e) Fe, SiO2, AcOH,
toluene 90 ꢀC, 31%. (f ) ZnCl2, EtMgBr, AlCl3, 3,4,5-trimethoxybenzoyl
chloride, CH2Cl2, 45%.
Scheme 3a
Figure 1. Structure of 2 and possible O-demethylated and hydro-
xylated metabolites of 2.
a Reagents and conditions: (a) (i) BnBr, K2CO3, EtOH, reflux; (ii)
LiOH, MeOH, 88%. (b) (i) SOCl2, CH2Cl2 or THF; (ii) ZnCl2, EtMgBr,
AlCl3, 6-methoxyindole, CH2Cl2, 25% for 7. (c) H2, Pd/C, ethyl acetate,
22% over two steps for 6.
Scheme 1a
acylation of respective indoles 17 and 24 with 3,4,5-trimethox-
ybenzoyl chloride. 4-Benzyloxy-6-methoxy-1H-indole (17)
was synthesized in four steps from 2-hydroxy-4-methoxyben-
zyldehyde (13) as reported by Kita et al.10 Briefly, benzylation
of 2-hydroxy-4-methoxybenzaldehyde (13) followed by
Hemetsberger-Knittel indole synthesis to construct the in-
dole core afforded 15. Hydrolysis of ester 15 and subsequent
decarboxylation under thermal conditions gave the desired
indole, 17. To construct the indole ring of 7-acetoxy-6-meth-
oxy-1H-indole (24), the o,β-dinitrostyrene reductive cycliza-
tion method, as reported by Fukuyama et al., was applied
using isovanillin (19) as the starting material.11 Briefly, nitra-
tion of isovanillin (19) followed by acetylation yielded 21.
Condensation of 21 with nitromethane using KN(TMS)2,
followed by dehydration using MsCl, resulted in o,β-dinitros-
tyrene compound 23. Silica gel-assisted reduction of the nitro
group in 23 with Fe/AcOH, with concomitant cyclization,
afforded required indole 24.
a Reagents and conditions: (a) BnBr, K2CO3, EtOH, reflux, 86%. (b)
(i) N3CH2CO2Et, NaOEt, EtOH, -15 ꢀC; (ii) xylene, reflux, 74%. (c)
KOH, EtOH, reflux, quantitative. (d) Copper chromite, quinoline,
215 ꢀC, 73%. (e) ZnCl2, EtMgBr, AlCl3, 3,4,5-trimethoxybenzoyl chlo-
ride, CH2Cl2, 26%. (f ) H2, Pd/C, ethyl acetate, 89%.
synthesis of 5-hydroxy-6-methoxy-3-(30,40,50-trimethoxyben-
zoyl)-1H-indole (4) and 6-hydroxy-3-(30,40,50-trimethoxy-
benzoyl)-1H-indole (8).5,6 The synthesis of hydroxylated
metabolites 3 and 5 are depicted in Schemes 1 and 2.
4-Hydroxy-6-methoxy-3-(30,40,50-trimethoxybenzoyl)-1H-
indole (3) and 7-hydroxy-6-methoxy-3-(30,40,50-trimethoxy-
benzoyl)-1H-indole (5) were obtained by Friedel-Crafts
The synthesis of demethylated metabolites 6 and 7 are
shown in Scheme 3. Benzylation of 3-hydroxy-4,5-dimethox-
ybenzoic acid methyl ester (25) followed by LiOH hydrolysis
in methanol produced acid 26. Conversion of the acid to acid
chloride, followed by Friedel-Crafts acylation with 6-meth-
oxyindole, resulted in the formation of benzyl-protected 27,

Brief Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 15 4943
Scheme 4a
Table 1. Cytotoxicities (IC50) of Compounds 2-12 in Three Different
Cell Lines
IC50 (nM ( SD)a
compd
KB
H460
HT-29
2
3
4
5
6
7
8
9
1.78 ( 0.52
23500 ( 4900
3100 ( 300
2.2 ( 1.4
48.0 ( 1.0
73.7 ( 12.1
103.0 ( 13.0
9.0 ( 1.0
1.0 ( 0.3
38000 ( 9900
3100 ( 100
29.3 ( 5.1
220.7 ( 48.5
189.3 ( 15.0
218.5 ( 12.0
9.1 ( 0.6
1.1 ( 0.2
45000 ( 7100
2700 ( 200
98.9 ( 8.6
226 ( 21.2
276 ( 25.6
100.5 ( 7.8
2.5 ( 0.1
10
11
0.24 ( 0.05
2.9 ( 0.6
84.3 ( 11.0
11.0 ( 2.0
0.04 ( 0.01
4.1 ( 0.3
88.3 ( 5.6
0.35 ( 0.2
13.3 ( 7.6
75.7 ( 6.4
10.0 ( 2.0
a Reagents and conditions: (a) NCS, AcOH, 53%; (b) NIS, AcOH,
40%; (c) Fe(acac)3, NMP, RMgBr, THF, 0 ꢀC f room temperature,
37% for 11 and 28% for 12.
12
colchicine
18.0 ( 7.0
a Values are expressed as the mean of at least three independent
experiments.
which was then deprotected by hydrogenation to give the
desired 30-OH metabolite 6. For the synthesis of 40-OH
metabolite 7, syringic acid (28) was treated with thionyl
chloride to give the acid chloride, which was immediately
converted to the desired product, 7, through Friedel-Crafts
acylation with 6-methoxyindole.
The preparation of metabolite-derived analogues 9-12 are
shown in Scheme 4. 7-Chloro and 7-iodo analogues 9 and 10
were synthesized by treating parent compound 2 with NCS
and NIS, respectively. 7-Iodo analogue 10 was then subse-
quently converted to 7-alkyl analogues 11 and 12 by iron-
catalyzed cross-coupling reaction with the required Grignard
reagents.
inhibition against cancer cell lines. Thus, when an alkyl
substituent was introduced at the seventh position of indole
of parent compound 2, the activity was retained for methyl
analogue 11, while ethyl analogue 12 had lower cytotoxicity.
Introduction of chloro and iodo functions at the seventh
position, however, led to potent cytotoxicity in 9 and 10,
respectively. Of particular interest is the iodo analogue 10,
which showed potency in the picomolar range (40-350 pM)
for all three cell lines tested. It is noteworthy that the IC50 of
iodo analogue 10 was as low as 40 pM in the H460 cell line.
It is established that the growth inhibition of cancer cells by
compound 2 is associated with its binding to the colchicine-
binding site of tubulin.7 To reason for the differences in
cytotoxicity of compounds 3, 4, and 5, which differ at the
position ofthe hydroxyl group, the ability of these compounds
to inhibit colchicine binding to the colchicine-binding site of
tubulin and their ability to inhibit tubulin polymerization was
determined; the results are shown in Table 1s of Supporting
Information. Among the three compounds, 5, which has
potent cytotoxicity, demonstrated higher binding affinity to-
ward the colchicine-binding site of tubulin as it replaced >95%
of [3H] colchicine at concentrations as low as 1 μM and also
inhibited tubulin polymerization in vitro. Additionally, com-
pound 10, the iodo analogue of 5, showed potent inhibition of
tubulin polymerization and colchicine site binding; com-
pounds 3, 4, and 12, which have lower cytotoxicity, showed
less binding affinity toward the colchicine-binding site and had
poor ability to inhibit tubulin polymerization. These results
suggest that the greater ability of compounds 5 and 10 to bind
to the colchicine-binding site and inhibit tubulin polymeriza-
tion leads to potent cytotoxicity for these analogues.
As the colchicine binding results for compounds 3-5 have
shown good correlation between cytotoxicity and the com-
pounds’ abilitytoinhibitthe colchicinebinding, weundertook
molecular modeling studies to investigate probable reasons
for the different levels of binding of the compounds 3-5 to the
colchicine-binding site of tubulin. This could help reveal the
role of the position of hydroxyl groups in determining the
binding efficacy to the colchicine-binding site. The most
probable orientation of 1-5 in the colchicine-binding site
(PDB ID: 1SA0) was determined by molecular docking using
Gold (Figure 1s of Supporting Information). The docking of
compound 2 coincides very well with 1, and a hydrogen bond
was observed between the 3-OH group of 1 and the backbone
NH of Val-181 residue. The docking results of compounds 3
and 4 revealed that the 4- or 5-hydroxyl group hindered the
Biological Results and Discussion
Figure 1 shows all the O-demethylated and hydroxylated
analogues of 2 that were synthesized. Compounds 2-8 were
subjected to antiproliferation evaluation against KB (human
cervical carcinoma), H460 (human lung cancer), and HT-29
(colorectal carcinoma) cell lines; the results are shown in
Table 1. Compared to compound 2, the O-demethylated
metabolites 6-8 showed decreased cytotoxicity, confirming
the importance of all the methoxy moieties as reported pre-
viously.12 In contrast to the O-demethylated metabolites, the
hydroxylated metabolite 4 displayed a loss of cytotoxicity
more than 1000-fold, revealing that hydroxylation at the fifth
position of 2 is not tolerated. Similarly, hydroxylation at the
fourth position of the parent compound also led to loss of
activity by more than 3 orders of magnitude in all three cell
lines tested for compound 3. In contrast to compound 3 and
metabolite 4, the seventh position hydroxylated metabolite 5
exhibited potent cytotoxicity in a range similar to that of the
parent compound 2 for the KB cell line. Thus, out of the five
metabolites synthesized for 2, the minor metabolite 5 showed
potent cytotoxicity similar to parent compound 2, while the
major metabolite 8 was at least 100-fold less active than the
parent compound 2. A similar finding, that hydroxylation at
the seventh position (5) of the parent compound 2 is well
tolerated, while the fifth position hydroxylated analogue 4
losses activity was recently reported by another group, further
substantiating our results.13 In addition, it was found that
metabolite 5 was sensitive to air/light oxidation, as reported by
Ty et al.;13 this precluded further development of this com-
pound, so we streamlined our efforts to identify analogues of 5.
Hence, we began to investigate the effect of introducing
different substituents at this position of indole on growth

4944 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 15
Table 2. Comparative in Vitro and in Vivo Profile of 2, 10, and 29
Wu et al.
in vitro metabolic ratea
(pmol/min/mg protein)
in vivo pharmacokineticsa
tubulin
polymerization
IC50( μM)a
KB
compd IC50a(nM)
colchicine
bindinga,b
mouse
rat
dog
human
CL
AUC
t1/2
2
1.78 ( 0.52
0.24 ( 0.05 96.04 ( 0.12
1.92 ( 0.68 104.4 ( 4.8
98.2 ( 2.9
3.3 ( 0.5
3.5 ( 0.4
2.2 ( 0.7
71.0 ( 1.9 57.4 ( 5.6 38.0 ( 10.2 43.5 ( 4.4
29.8 ( 9.1 2739 ( 760
5.8 ( 1.7
10
29
56.3 ( 0.9 39.3 ( 8.5 28.3 ( 9.7
66.0 ( 5.4 46.8 ( 7.6 36.8 ( 3.0
25.1 ( 11.3
20.0 ( 6.2
21.6 ( 4.4 4585 ( 803 11.5 ( 3.1
a Values are expressed as the mean of at least three independent experiments. b Percentage inhibition of [3H] colchicine binding; [3H] colchicine and test
compound concentration was at 5 μM.
compounds from binding in the same orientation as that of 1
due to steric clash with Met-259, which resulted in binding of
these compounds in a different orientation than that of 1. In
contrast, the hydroxyl group from the seventh position of
indole moiety appeared to pose less steric hindrance to the
binding of 5 in an orientation similar to that of 1. Moreover
the seventh-position OH group is within hydrogen-bonding
distance of the backbone NH of Val-181, which could be
the reason for the potent colchicine binding affinity and
cytotoxicity of 5.
of iodo group next to the 7-methoxy group in the indole ring
had improved the metabolic stability of 10 in all the species
tested. This could be due to the steric hindrance offered by the
bulky iodo group next to the 7- methoxy group. Becuse the
iodo analogue 10, compared to 29, has poor drug like proper-
ties, such as higher molecular weight and higher clogP, we
selected 29 for further development.
On the basis of the in vitro metabolic study, to confirm if 29
indeed can have better exposure than 2, we carried out in vivo
pharmacokinetic evaluation of compound 29 and compared
the data to that for 2. Rats were dosed iv at 5 mg/kg and the
PK parameters measured; the results are shown in Table 2.
Compared to 6-methoxycompound 2, 6-ethoxy compound 29
exhibited lower clearance, higher AUC, and longer t1/2. This
indicates thatthe in vivo “drug-elimination time” for 6-ethoxy
analogue 29 is longer than for parent compound 2 and has
higher drug exposure.
Through the synthesis of metabolites 4-8 and the testing of
them for cytotoxicity, we found that the O-demethylation of
parent compound 2 at the sixth position of indole led to the
formation of major metabolite 8 in vitro. The major metabo-
lite 8 is at least 100-fold less potent in cytotoxicity assay than
parent compound 2. In addition, metabolite 8 failed to show
antitumor activity in the human KB tumor xenograft animal
model at 50 mg/kg (data not shown), while parent compound
2 was active at 50 mg/kg.7 On the basis of these findings, we
reasoned that blocking the major inactivating O-demethylat-
ing metabolic route through appropriate structural modifica-
tion of 2 could improve the in vivo exposure time of the active
compound and hence improve the therapeutic efficacy. To
realize this goal, we selected 6-ethoxy analogue 29, which was
previously synthesized and showed similar levels of in vitro
cytotoxicity in a variety of cancer cell lines as that of 6-
methoxy analogue 2.5 We also observed that both compounds
2 and 29 possessed similar levels of KB cell line cytotoxicity
and colchicine binding (Table 2). To investigate the possibility
that the presence of bulky ethoxy group instead of methoxy
group at the sixth position of indole could result in lower
O-dealkylation reaction rate due to more steric hindrance
offered by the ethyl group than the methyl group, the in vitro
metabolic stability for compounds 2 and 29, using liver
microsome preparation from mouse, rat, dog, and human,
were determined by measuring the amount of compound
remaining unchanged after 30 min of incubation with the
microsome; the results are shown as metabolic rate in Table 2.
In all four species tested, the in vitro metabolic rates for 2 were
higher than for 29; the difference is particularly prominent
with human enzyme, indicating that compound 29 might have
slower metabolic clearance and hence higher residence time
than 2 in vivo.
On the basis of the positive findings in in vivo PK studies,
we set out to evaluate the in vivo antitumor activity of both
In addition, we have also carried out the in vitro metabolic
stability of the most potent analogue identified in this study,
the iodo analogue 10 (Table 2). It was found that introduction
Figure 2. Inhibition of human KB xenografts’ growth in vivo in
nude mice by compounds 2 and 29. Compound 29 at 5 mg/kg
showed significant (p < 0.05) tumor growth inhibition.

Brief Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 15 4945
compounds 2 and 29 in the KB tumor xenograft nude mouse
model. Compound 2 was dosed at 10 mg/kg, and compound
29 at 1 and 5 mg/kg for five days per week for three
consecutive weeks. At the dose of 1 mg/kg of compound 29,
the tumor progressed at the same rate as the untreated mice,
indicating that compound 29 is inactive at this dosage
(Figure 2). When the mice were treated with 5 mg/kg of 29,
KB tumor size reduced (43%) significantly (p<0.05). Sig-
nificant reduction in KB tumor size, however, was not ob-
served in the mice treated with 10 mg/kg of compound 2,
showing 29 to be effective even at half of the dose. However, it
should be noted that compound 29 may also be more toxic
than 2, as indicated by the percentage body weight gain after
drug treatment. Compound 29 at 5 mg/kg showed body
weight loss of ∼5%, while 2 at 10 mg/kg showed body weight
gain ∼10%. Considering the fact that the in vitro cytotoxi-
cities of compounds 2 and 29 are similar in the KB and other
cell lines, compound 29 has better in vivo antitumor efficacy
than 2 because of the improved PK profiles of 29.
Acknowledgment. The study was financially supported by
the National Health Research Institutes and National Science
Council (grant no. NSC-95-2113-M-400-001-MY3), Taiwan,
ROC. We thank Mark Swofford for helping with the English
editing.
Supporting Information Available: General synthetic meth-
ods, synthesis, and spectral data for 3, 5, 6, 7, 9, 11, and 12,
HPLC purity determinations protocol, in vitro and in vivo
biological evaluation protocols, colchicine and tubulin inhi-
bition data for 3-5, 10, 12, molecular docking studies of 3-5
in the colchicine-binding site of tubulin, mice body weight
data for in vivo antitumor evaluation of 2 and 29. This mate-
rial is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Nassar, A. E.; Kamel, A. M.; Clarimont, C. Improving the
decision-making process in the structural modification of drug
candidates: enhancing metabolic stability. Drug Discovery Today
2004, 9, 1020–1028.
(2) Kola, I.; Landis, J. Can the pharmaceutical industry reduce attri-
tion rates?. Nat. Rev. Drug Discovery 2004, 3, 711–715.
(3) Fura, A. Role of pharmacologically active metabolites in drug
discovery and development. Drug Discovery Today 2006, 11, 133–
142.
(4) Liou, J. P.; Chang, J. Y.; Chang, C. W.; Chang, C. Y.; Mahindroo,
N.; Kuo, F. M.; Hsieh, H. P. Synthesis and structure-activity
relationships of 3-aminobenzophenones as antimitotic agents. J.
Med. Chem. 2004, 47, 2897–2905.
(5) Liou, J. P.; Chang, Y. L.; Kuo, F. M.; Chang, C. W.; Tseng, H. Y.;
Wang, C. C.; Yang, Y. N.; Chang, J. Y.; Lee, S. J.; Hsieh, H. P.
Concise synthesis and structure-activity relationships of combre-
tastatin A-4 analogues, 1-aroylindoles and 3-aroylindoles, as novel
classes of potent antitubulin agents. J. Med. Chem. 2004, 47, 4247–
4257.
(6) Liou, J. P.; Mahindroo, N.; Chang, C. W.; Guo, F. M.; Lee, S. W.;
Tan, U. K.; Yeh, T. K.; Kuo, C. C.; Chang, Y. W.; Lu, P. H.; Tung,
Y. S.; Lin, K. T.; Chang, J. Y.; Hsieh, H. P. Structure-activity
relationship studies of 3-aroylindoles as potent antimitotic agents.
ChemMedChem 2006, 1, 1106–1118.
(7) Kuo, C. C.; Hsieh, H. P.; Pan, W. Y.; Chen, C. P.; Liou, J. P.; Lee,
S. J.; Chang, Y. L.; Chen, L. T.; Chen, C. T.; Chang, J. Y.
BPR0L075, a novel synthetic indole compound with antimitotic
activity in human cancer cells, exerts effective antitumoral activity
in vivo. Cancer Res. 2004, 64, 4621–4628.
(8) Yao, H. T.; Wu, Y. S.; Chang, Y. W.; Hsieh, H. P.; Chen, W. C.;
Lan, S. J.; Chen, C. T.; Chao, Y. S.; Chang, L.; Sun, H. Y.; Yeh, T.
K. Biotransformation of 6-methoxy-3-(30,40,50-trimethoxy-ben-
zoyl)-1H-indole (BPR0L075), a novel antimicrotubule agent, by
mouse, rat, dog, and human liver microsomes. Drug Metab. Dispos.
2007, 35, 1042–1049.
(9) Chang, Y. W.; Chen, W. C.; Lin, K. T.; Chang, L.; Yao, H. T.;
Hsieh, H. P.; Lan, S. J.; Chen, C. T.; Chao, Y. S.; Yeh, T. K.
Development and validation of a liquid chromatography-tandem
mass spectrometry for the determination of BPR0L075, a novel
antimicrotuble agent, in rat plasma: application to a pharmacoki-
netic study. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci.
2007, 846, 162–168.
Conclusions
We have successfully synthesized and evaluated hydroxy-
lated and O-demethylated phase I metabolites of potent
antitumor agent 2. Four of five metabolites 5-8 were found
active against various cancer cell lines in the nanomolar
concentration range. The iodo derivative 10 of the most
potent 7-hydroxy metabolite 5, exhibited extremely potent
anticancer activity against cancer cell lines, with the IC50
reaching picomolar potency in the KB, H460, and HT-29 cell
lines. Further structure-activity relationship studies at the
seventh positions of the indole may provide new insights into
combretastatin analogue design in the future.
As the major O-demethylated metabolite 8 was at least
100-fold less active than parent compound 2, a strategy to
block the metabolic inactivation was worked out by identify-
ing compound 29 with 6-ethoxy substituent, which could offer
increased steric hindrance to O-demethylating metabolic de-
activation reaction. Evaluation of in vitro metabolic stabi-
lity and in vivo PK studies clearly demonstrated the super-
iority of 29 over parent compound 2 in terms of decreased in
vitro metabolic rate and improved drug exposure in vivo.
Further in vivo studies in the KB xenograft model showed
that the antitumor effect of 29 is better than that of the parent
compound 2, validating our metabolite approach to develop
potent anticancer compounds as backup candidates for novel
antmitotic agent 2. Thus, insights obtained from the synthesis
and testing of in vitro metabolites of 2 were used successfully
to design and develop better anticancer agents.
(10) Kita, Y.; Tohma, H.; Inagaki, M.; Hatanaka, K.; Yakura, T. Total
Synthesis of Discorhabdin-C;A General Aza Spiro Dienone
Formation from O-Silylated Phenol Derivatives Using a Hyperva-
lent Iodine Reagent. J. Am. Chem. Soc. 1992, 114, 2175–2180.
(11) Fukuyama, Y.; Iwatsuki, C.; Kodama, M.; Ochi, M.; Kataoka, K.
Antimicrobial indolequinones from the mid-intestinal gland of the
muricid gastropod Drupella fragum. Tetrahedron 1998, 54, 10007–
10016.
(12) Hsieh, H. P.; Liou, J. P.; Mahindroo, N. Pharmaceutical design of
antimitotic agents based on combretastatins. Curr. Pharm. Des.
2005, 11, 1655–1677.
(13) Ty, N.; Dupeyre, G.; Chabot, G. G.; Seguin, J.; Tillequin, F.;
Scherman, D.; Michel, S.; Cachet, X. Synthesis and biological
evaluation of new disubstituted analogues of 6-methoxy-3-(30,40,50-
trimethoxybenzoyl)-1H-indole (BPR0L075), as potential antivas-
cular agents. Bioorg. Med. Chem. 2008, 16, 7494–7503.
Experimental Section
(7-Iodo-6-methoxy-1H-indol-3-yl)-(3,4,5-trimethoxyphenyl)-
methanone (10). To a solution of NIS (0.05 g, 0.22 mmol) in
acetic acid (1 mL), a solution of (6-methoxy-1H-indol-3-yl)-
(3,4,5-trimethoxyphenyl)methanone5 (2) (0.07 g, 0.20 mmol) in
acetic acid (5 mL) was added at room temperature. After the
mixture was stirred at room temperature for 16 h, the solvent
was evaporated under reduced pressure; the residue was purified
by flash column chromatography over silica gel (EtOAc:
n-hexane=1:2) to give 10 (37 mg, 40%). 1H NMR (300 MHz,
CDCl3) δ 3.89 (s, 6H), 3.94 (s, 3H), 3.99 (s, 3H), 6.95 (d, J=8.7
Hz, 1H), 7.11 (s, 2H), 7.69 (d, J=3.0 Hz, 1H), 8.27 (d, J=8.7 Hz,
1H), 8.70 (s, br, 1H). LCMS (M þ H)þ 468.0.