Indole-3-carbinol (I3C), a naturally occurring phytochemical
found in cruciferous vegetables, has received much attention due
to its translational potential in cancer prevention and therapy.1-4
Under acidic conditions in vitro, and following ingestion, I3C
readily selfcondenses to form a mixture of oligomeric products
including cyclic tetraindole (CTet).5,6 I3C and its oligomeric
products are under study as cytostatic and tumor-suppressive
agents, in particular, CTet.4 Development of CTet as a drug faces
some problems; poor solubility, its synthetic procedures are low-
yielding that requires HPLC purification of product from the
complex acid reaction mixture of I3C. Inspired by CTet structure
and activity, we discovered novel tetraindoles as anticancers. In
particular compound SK228 (Figure 1) showed promising
activity against breast adenocarcinoma, various human lung and
esophageal cancer cell lines.7 In that study, we have investigated
the structural features of tetraindoles.7-10 Several indole
derivatives were synthesized and tested.7,10 5-Hydroxyindole
moiety showed the highest activity. The core structure was also
found to affect the activity (phenyl showed much better activity
than 2-thienyl).7 In continue of our interest in tetraindoles; this
study aims to explore-in more detail- the structural features of the
core structure. Thereby development of tetraindoles which are
expected to act by I3C-like mechanisms against several cancers
and to offer substantially improved potency and activity that
would make them applicable to cancer therapy.
trioxide in presence of acetic anhydride, acetic acid and sulfuric
acid.20 Compound 13 was used for the synthesis of p-terephenyl-
4,3’,6’,4’’-tetracarb-aldehyde 32 via the Suzuki coupling reaction
with 4-formylphenylboronic acid (Scheme 2). On the other hand,
2,3,5,6-tetramethylterephthalaldehyde 12 was synthesized from
1,2,3,5-tetramethylbenzene
as
repoted.21
Finally,
tetrachloroterephthalaldehyde 14 and tetrafluoroterephthalalde-
hyde 15 were prepared from commercially available 2,3,5,6-
tetrachloro-1,4-dicyanobenzene as reported.23
All the synthesized compounds were adequately
characterized by spectroscopic methods including 1H and 13C
NMR, ESI-HRMS. The purity was established as being >95% by
HPLC. The target compounds (16−31 and 33) were evaluated for
their in vitro anti-proliferative activity against human breast
cancer cell line (MDA-MB-231) for the determination of IC50.
The IC50 values (μM) are presented in Table 1. Based on these
results, some preliminary structure–activity relationship aspects
were deduced.
Initially our study focused on the effects of the rigidity of the
core structure on the anticancer activity. Compound 16 is 20
times less potent than SK228 (IC50 value of 0.45 μM against
breast adenocarcinoma MDA-MB-231 cells). This result refers to
the importance of rigidity and planarity of the core structure.
Biphenyl ring is less rigid and non-planar. Increasing the distance
between the diindolylmethinyl and the phenylene group of the
linker by incorporating additional methylene group, compound
17, markedly decreased the activity and this confirms the
importance of rigidity and extended conjugation of the core
structure. It is clear that the optimum spacer between the indole
rings and phenylene moiety is a single methinyl group.
Interestingly, 18 showed better activity than 17 due to the
presence of double bond, which is more rigid and conjugated
with the phenyl ring. Although 31 contain more indole ring but it
showed much lower activity (IC50 > 20 μM) because of presence
of saturated carbon that makes it flexible. Compounds 19 to 22
have rigid and conjugated core structure but they exhibited lower
activity. These results suggested that indole rings interact in some
way or a particular topology might be required for high activity.
The order of activity among these compounds support this view,
22 has the best activity. On the other hand, the anthracene
nucleus is present in many compounds that have activity against
tumor cells. These compounds typically are DNA-intercalating
agents, based on the ability of the large, flat anthracene
chromophore to bind strongly between the base pairs. Bisantrene
is one example of a 9,10-disubstituted anthracene antitumor
agent. The very important anthracyclines such as doxorubicin and
daunorubicin also may be considered as anthracene derivatives.
9,10-Disubstituted anthracenes showed much more potent than
its 1,4-isomers.24 Inspired by anthracene as a core structure,
compound 23 was synthesized. To our delight, this compound
showed much improved activity, most likely due to their stronger
π-stacking ability as compared to the results with compounds
having naphthalene.25 On the other hand, using 2,9-
phenanthroline as a core structure (compound 24) resulted in
much lower activity. NMR spectrum of this compound showed
that the indole rings are non-equivalent and possibly non-planar.
Based on these results compound 25 was synthesized. As we
expected this compound showed lower activity in comparison
with 23. The saturated ring of tetrahydroanthracene exists in a
half-chair form,26 this makes the structure unsymmetrical and
non-planar.
We speculated that the rigidity and conjugation of the core
structure of tetraindoles might play important roles in their
anticancer effects. Moreover, core structures containing
napthalene, anthracene and phenanthroline may act as DNA
intercalating-crosslinkers. In order to identify the key structural
features of phenylenebis(methylene)-linked tetraindoles, a series
of compounds were prepared in which the core structure was
systematically varied. Accordingly, tetraindoles 16–30 were
generated by the reaction of 5-hydroxyindole with various
dialdehydes 1–15 in acetonitrile and presence of catalytic
amounts of molecular iodine at room temperature (Scheme 1).11,12
The crude products were subsequently purified by column then
recrystallization to give the pure compounds in yields ranging
from moderate to very good yield. Dialdehydes were synthesized
by standard methods.13-21 Pentaindole 31 (Figure 2) was obtained
as a side product during synthesis of tetraindole 18, due to
Micheal addition of indole to α,β-unsaturated aldehyde.22 While
octaindole 33 was obtained from the tetraaldehyde 32 (Scheme
2). Aromatic dicarboxaldehydes were generally synthesized
either by double oxidation of benzylic alcohols, double reduction
of dinitriles or directed dilithiation strategies followed by
electrophilic quenching with DMF. Accordingly, 1,4-benzene-
diacetaldehyde 2 was synthesized from the commercially
available 1,4-benzenediacetic acid by esterification, reduction to
the corresponding alcohol then selective oxidation.13 In analogy,
2,7-naphthalenedicarbox-aldehyde
4
and 2,6-naphthalenedi-
carboxaldehyde 5 were obtained from their corresponding
dimethyl esters.14 1,4-Benzenediacryaldehyde 3 was prepared by
an aldol condensation between terephthalaldehyde and
acetalaldehyde.15 1,5-naphthalene-dicarboxaldehyde 6 obtained
from 1,5-diaminona-phthalene by Sandmyer reaction to get
diiodo which is converted into dicyano that reduced with
DIBAL-H.16 Similarly, dialdehydes 7, 8 and 10 were generated
by reduction of the corresponding dicyano derivative.17 1,10-
Phenanthroline-2,9-dicarboxaldehde 9 was prepared by oxidation
of the commercially available neocuproine by selenium dioxide
in 4% water in dioxane.18 2,5-Dimethylbenzene-1,4-
dicarboxaldehyde 11 was obtained from 1,4-dibromo-p-xylene
by reacting with n-BuLi and DMF.19 The same starting material
was used for the synthesis of 2,5-dibromobenzene-1,4-
dicarboxaldehyde 13 by controlled oxidation via chromium
The higher activity of SK228 prompted us to study the stereo-
electronic effects of its substituted phenyl ring on anticancer
activity. Accordingly, compounds 26−30 were synthesized and
tested. Compound 26 having 2,5-dimethyl exhibited lower
activity in comparison with 27 which is more symmetrical. But
3