Mild synthetic approach to novel indole-1-carbinols and preliminary evaluation of their cytotoxicity in hepatocarcinoma cells

Article abstract of DOI:10.1002/ardp.201100104

A mild and versatile method for the synthesis of some novel indole-1-carbinols has been developed via one-pot reaction of indoles and paraformaldehyde in the presence of an excess of CaO, MgO, ZnO or TiO ₂. The solvent-free reaction provided all the indole derivatives in moderate to good yields and short reaction times. Moreover, the effect of some selected indole-1-carbinols on cell proliferation of the hepatoma cell line FaO was evaluated.

Full text of DOI:10.1002/ardp.201100104

Arch. Pharm. Chem. Life Sci. 2012, 345, 195–202
195
Short Communication
Mild Synthetic Approach to Novel Indole-1-Carbinols
and Preliminary Evaluation of Their Cytotoxicity
in Hepatocarcinoma Cells
Gloria Zedda1, Gabriella Simbula2, Michela Begala1, Monica Pibiri2, Costantino Floris3,
Mariano Casu3, Laura Casu1, and Graziella Tocco1
1
`
Dipartimento Farmaco Chimico Tecnologico, Universita degli Studi di Cagliari, Cagliari, Italy
2 Dipartimento di Tossicologia, Sezione di Oncologia e Patologia Molecolare,
`
Universita degli Studi di Cagliari, Cagliari, Italy
Dipartimento di Scienze Chimiche, Universita degli Studi di Cagliari, Cittadella Universitaria di Monserrato,
3
`
Monserrato (CA), Italy
A mild and versatile method for the synthesis of some novel indole-1-carbinols has been developed via
one-pot reaction of indoles and paraformaldehyde in the presence of an excess of CaO, MgO, ZnO or
TiO2. The solvent-free reaction provided all the indole derivatives in moderate to good yields and short
reaction times. Moreover, the effect of some selected indole-1-carbinols on cell proliferation of the
hepatoma cell line FaO was evaluated.
Keywords: Apoptosis / Cytotoxicity / Hepatocarcinoma / Indole-1-carbinols / Solvent-free reaction
Received: March 24, 2011; Revised: June 1, 2011; Accepted: June 20, 2011
DOI 10.1002/ardp.201100104
Supporting information available online
:
Introduction
and active oligomers such as 3,30-diindolylmethanes (DIMs),
indolo[3,2b]-carbazole (ICZ) and linear as well as cyclic trimers
(LTr1, CTr). Consequently, indole-3-carbinol and DIMs are now
widely used as scaffolds in the development of new anti-
cancer agents [2, 3]. Recently, some remarkable attempts
to improve the stability and the metabolic profile of
indole-3-carbinol have been reported in the literature [4].
The best results were obtained with compound OSU-A9, a
novel I3C analogue, which is very stable in acidic conditions
and has a higher apoptosis-inducing potency, due to the
presence of a 3-chloro-2-nitrobenzene sulfonyl group linked
to the indole nitrogen [4] (Figure 2).
Indole is probably the most widespread heterocyclic frame-
work, present in a plethora of natural biologically active
molecules [1]. In particular, indole-3-carbinol (I3C), a dietary
component found exclusively in Brassica vegetables as a
metabolite of glucobrassicin, has emerged as an efficient
carcinogenesis modulator with a pleiotropic mode of action.
(Figure 1) [2].
However, the bioactivity of I3C may be plausibly related to
its conversion, in the acidic gastric milieu, to more stable
From a structural point of view, the characteristic acid
instability of I3C is associated to the high propensity of
the vinylogous hemiaminal moiety of the indole ring to
dehydrate. Preventing this mechanism through a substi-
tution could solve this drawback and may lead to an optim-
ization of the anticancer activity. Thus, in the present
research we report an extremely simple, mild and solvent-
free method for the synthesis of some novel indole-1-carbi-
nols, which are I3C structural isomers in which the N1-
position is functionalized with a hydroxymethyl group.
Correspondence: Dr. Graziella Tocco, Assistant Professor of Organic
`
Chemistry, Dipartimento Farmaco Chimico Tecnologico, Universita degli
Studi di Cagliari, via Ospedale 72, 09124 – Cagliari, Italy.
E-mail: toccog@unica.it
Fax: þ39 070 675 8553
Abbreviations: I3C, indole-3-carbinol; DIMs, 3,3’-diindolylmethanes; ICZ,
indolo[3,2b]-carbazole; LTr1; CTr, linear and cyclic trimers; PARP,
poly(ADP-ribose) polymerase; DMSO, dimethylsulfoxide; DMF,
dimethylformamide; I1C, indole-1-carbinol.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

196
G. Zedda et al.
OH
Arch. Pharm. Chem. Life Sci. 2012, 345, 195–202
H
HO
O
HO
S
OSO3
HO
OH
NCS
N
Myrosinase
N
H
N
H
N
H
Figure 1. Glucobrassicin as precursor of indole-
Glucobrassicin
Indole-3-methyl isothiocyanate
(unstable)
Indole-3-carbinol
(I3C)
3-carbinol (I3C).
OH
It was also evidenced a new dimeric derivative (Scheme 1).
Mechanistically, we presume that a different reaction path-
way can be conceived for the formation of the last compound
(paper in preparation).
N
S
O
O
This observation diverted our attention toward the
N-hydroxyalkylation of indole. We began our study by inves-
tigating all the parameters involved in the reaction with
the aim of improving the yields, the regiochemistry and
the scope of the reaction with respect to indole-1-carbinol 2a.
We first explored the role of the CaO observing that the
reaction did not occur in its absence [7a]. Besides, the exper-
iment did not proceed at all at room temperature, both with
or without the alkaline oxide. Fixing the reaction tempera-
ture at 658C showed to be the best manner to obtain indole-1-
carbinol 2a with a high regioselectivity. In fact, higher
temperatures directed the reaction towards dimer
formation. The determining effect of the amount of CaO
on the reaction proceeding was also immediately apparent.
Indeed, as shown in Table 1, we observed that when
calcium oxide was used in large excess, the reaction seemed
to proceed smoothly, while the use of CaO in catalytic or
stoichiometric amounts did not seem to produce any effect.
Therefore, the alkaline earth metal oxide may act as a multi-
functional agent, showing a synergic effect together with the
temperature in the activation of the reaction, provoking a
rapid conversion of paraformaldehyde to free formaldehyde
O2N
Cl
Figure 2. OSU-A9.
Results and discussion
In contrast to benzotriazole [5], the N-alkylation and in
particular the N-hydroxyalkylation of indoles have not often
been investigated, probably due to the reduced nucleophi-
licity of their N–H function [6]. To date, apart from a few mild
procedures [7], many of the methods reported in the liter-
ature provide indole-1-carbinols in very low yields and long
reaction times [7b, 8]. Mostly, the reduction of 1-indolyl
carbamates [9] or strong basic conditions [10a–c] also involv-
ing organomethallic reagents are used [5c, 10d].
Consequently, the development of a simple and efficient
method to achieve the N-hydroxymethylation of indoles
under mild conditions, in very short reaction times and in
solvent-free conditions, represents an important challenge.
In the beginning of our study, with the aim of searching for
a new protocol to I3C dimers, indole 1a was allowed to react
with formaldehyde, derived from paraformaldehyde, under
solvent free conditions, at 1008C, in the presence of an excess
of CaO. Surprisingly, apart from the expected 3,30-diindolyl
methane, we observed the formation of indole-1-carbinol 2a.
Table 1. Base effect on reaction between indole 1a and
formaldehyde at 658C.
Substrate Oxidea/1a
Time
(min)
Product
Yield
(%)c
1a
1a
1a
1a
1a
–
120
120
120
120
30b
–
–
0.1/1
1/1
5/1
2a
2a
2a
2a
traces
5–7
25–55
25–65
O
CaO
+
+
I3C dimers
100 °C
H
H
N
N
H
17/1
OH
2a
1a
a
Best results with CaO and MgO.
No traces of dimers also for longer times.
Lit. Yield (%) 34 [7a], not reported [10a], 67 [20]
b
c
Scheme 1. Reaction between indole 1a and formaldehyde at
1008C.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2012, 345, 195–202
Indole-1-Carbinols as Cytotoxics in Hepatocarcinoma Cells
197
R1
R1
and neutralizing the traces of formic acid produced during
prolonged heating.
We assumed that CaO might induce a partial N–H proton
abstraction and, despite the modest ionic character of N–Ca
bond, formaldehyde could be so reactive to undergo a nucle-
ophilic attack, generating indole-1-carbinol 2a.
N
H
1b-j
N
N
OH
2b-j
O
oxide
65 °C
or
or
+
H
H
To test our hypothesis, we ran our experiments with MgO,
ZnO and TiO2. As expected, the oxides possessing strongly
electropositive cations, such as MgO and CaO, afforded the
best yields (Table 1).
R2
R2
N
H
OH
We also investigated the use of potassium hydroxide, but,
unfortunately, no traces of compound 2a could be detected. A
possible explanation is related to the thermal decomposition
under strong basic conditions, of indole-1-carbinol to regen-
erate indole 1a and HCHO [10a].
1k
2k
1b R1= -CH3; 1c R1= -2-OCH3Ph; 1d R1= -CHO; 1e R1= -COCH3;
1f R1= -COOCH3; 1g R1= -CN; 1h R1= -2-BrPh; 1i R1= -Br; 1j R1= -I;
1k R2= -CH3
No reaction with: 1l R2= -COOC2H5; 1m R2= -Ph; 1n R2= -4-ClPh;
Different solvents were studied in order to optimize the
reaction conditions further. Obviously, as the mechanism is a
typical nucleophilic addition, we focused our interest on
dimethylsulfoxide (DMSO), dimethylformamide (DMF), aceto-
nitrile and acetone, strongly dissociating polar aprotic sol-
vents which tend to favour attack at the indole nitrogen. As a
matter of fact, no reaction improvement was evidenced by
using tetrahydrofuran (THF) or apolar solvents such as
toluene or hexane. Besides, polar protic solvents have been
banned, in order to avoid a nucleophile competition or a
reduction of the strength of our indole nitrogen.
Overall, we noticed a sizeable improvement of the yields
when running the reaction in the presence of DMSO, but
after removal of the solvent, only a small amount of carbinol
2a could be isolated, due to its instability at high temperature
(Table 2).
Scheme 2. Synthesis of 2- and 3-substituted N-carbinols.
I
I
(a)
(b)
N
H
N
H
N
Ts
3
1a
1j
(c)
R3
R3
(d)
N
N
Ts
H
4
1c (R3 = -OCH3)
1h (R3 = -Br)
The hypothesis that the reaction, if successful with 1a,
would allow similar results when different indole derivatives
are used, led us to carry out some other exploratory reactions
with compounds 1b–n (Scheme 2).
Reagents and conditions: (a) I2, KOH, DMF; (b) TsCl, NaOH,
TBABr, CH2Cl2; (c) 2-methoxy or 2- bromo benzeneboronic acid,
Pd(PPh3)4,Na2CO3, EtOH, toluene, H2O; (d) NaOH 3 M, MeOH.
Almost all indole substrates were purchased from commer-
cial sources, apart from compounds 1i and 1j, which were
prepared according to literature [11], and indoles 1c and 1h,
Scheme 3. Synthesis of indole substrates 1c and 1h.
obtained via the Suzuki coupling synthetic approach [12]
(Scheme 3).
Table 2. Solvent effect on reaction between indole 1a and
formaldehyde at 658C.
The developed protocol showed to be valid for the sub-
strates 1b–k (Table 3). Unfortunately, no reaction was
observed by using substrate 1l. We presume a nucleophilicity
decrease of the correspondent indolyl anion due to its biden-
tate ligand behaviour. Similarly, the lack of reactivity of
indoles 1m and 1n might be related both to electronic and
steric substituent effects (Scheme 2).
Entry Solvent
Time (min)
Yield (%)
1
2
3
4
5
6
7
DMSO
DMF
CH3CN
Acetone
THF
30
60
90a
34a
22a
5b
–
120
120
120
120
120
Toluene
Hexane
–
–
In particular, as shown in Table 2, the reactions carried out
with compounds 1d–g, showed to be extremely regioselective
for the related carbinols; in fact, no traces of dimers were
noticed also for longer reaction times. Interestingly, products
a
GC yields.
Isolated yields
b
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

198
G. Zedda et al.
Arch. Pharm. Chem. Life Sci. 2012, 345, 195–202
Table 3. Reaction of indoles 1b–k with formaldehyde at 658C.
Substrate
R1, R2
Oxidea/1
Time (min)
Product
Yield (%)
Lit. Yield (%)
1b
1b
1c
1c
1d
1d
1e
1e
1f
R1 ¼ –CH3
–
17/1
–
17/1
–
17/1
–
17/1
–
17/1
–
17/1
17/1
17/1
17/1
–
120
30b
120
30b
120c
120c
120c
120c
120c
120c
120c
120c
60b
–
2b
–
2c
2d
2d
2e
2e
2f
–
43–55
–
50–70
15
65
10
60
15
70
8
67
65
20d
18d
–
10 [8], 80 [20]
R1 ¼ -2-OCH3Ph
R1 ¼ –CHO
–
33 [21]
R1 ¼ –COCH3
R1 ¼ –COOCH3
R1 ¼ –CN
–
–
1f
2f
1g
1g
1h
1i
1j
1k
1k
2g
2g
2h
2i
2j
2k
2k
49 [21]
R1 ¼ -2-BrPh
R1 ¼ -Br
–
–
–
–
–
30
30
120
30b
R1 ¼ -I
R2 ¼ –CH3
17/1
45–57
a
Lower yields with TiO2.
b
c
Longer times facilitate the formation of dimers.
No traces of dimers also for longer times.
Highly unstable products.
d
2d–g were also detected in very low yields without the use of
any alkaline oxide. This observed behavior is presumably due
to the presence of an electron-withdrawing substituent which
helps the removal of the N–H proton by reducing its pKa.
We applied the optimized procedure also to a selection of
aliphatic (propanal, butanal, hexanal, heptanal and dodecanal)
and aromatic aldehydes (benzaldehyde, 2-methoxybenzalde-
hyde, 3-methoxybenzaldehyde, 4-methoxybenzaldehyde, 2-
nitrobenzaldehyde, 3-nitrobenzaldehyde, 4-nitrobenzaldehyde,
piperonal, 2-thiophenaldehyde, 2-furaldehyde and 3-formyl
indole) but the results were unsatisfactory. In particular,
with the less reactive aromatic substrates, no N-carbinol
was never observed. As a consequence, no reaction also with
ketones (2- octanone, 2-hexanone, 40-methylacetophenone and
1-(p-methoxyphenyl)-2-propanone).
By employing aliphatic aldehydes with a-hydrogens, we
observed only the formation of 1-alkenyl-1-H-indoles, certainly
derived from in situ dehydration of the correspondent
N-carbinols.
The effect of indole-3-carbinol and of some selected new
compounds on cell proliferation of the hepatoma cell line
FaO was also evaluated. We first investigated the possible role
of electron donor and electron-withdrawing substituents in
3-position of the indole ring, on the cytotoxic activity. With
this respect, compounds 2a, 2b, 2d and 2e have been the first
compounds of choice.
FaO cells were treated with an increasing concentration of
compounds 2a, 2d, 2e and 2b as well as indole-3-carbinol, for
24 hours. The results indicated a dose-dependent decrease of cell
viability as measured by NRU, compared to control (Figure 3).
Figure 3. Effect of increasing concentrations of I3C and derivatives 2a, 2d, 2e and 2b on FaO cell viability determined by NRU assay after
24 hours of treatment. Data are expressed as percentage of viable cells as assessed by Neutral Red uptake. Results are the means ꢀ SD
from three separate experiments.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2012, 345, 195–202
Indole-1-Carbinols as Cytotoxics in Hepatocarcinoma Cells
199
In particular, the derivatives 2a, 2d and 2e were two to four
fold more potent than I3C in suppressing the viability of FaO
cells showing IC50 values ranging from 100 mM to 200 mM
(Figure 3).
In view of their in vivo application via oral administration,
we also tested the stability in acidic conditions of compounds
2a, 2b, 2d, 2e, 2f and 2g, monitoring the experiments via NMR
analysis [4a]. We observed that after treatment with labeled
HCl, the most stable compounds were derivatives 2e and 2f,
while the others produced an acidic reaction mixture con-
sisting of dimers and other side products. All the other
derivatives are now under investigation in similar exper-
imental conditions.
Since preliminary evaluations of metabolic transform-
ations [13] did not evidence intracellular formation of formal-
dehyde or dimer derivatives, we suppose that the observed
cytotoxic activity is related to other interactions. Further
studies aimed at the synthesis and the evaluation of the
cytotoxicity of the most important I1Cs’ metabolites are
now in progress.
In conclusion, we have developed a simple and versatile
synthetic procedure to prepare several N-carbinols. This proc-
ess might represent a valuable alternative to the drastic
reaction conditions usually employed for the N-hydroxyme-
thylation of indoles. This makes our method a convenient
tool for the preparation of a wide variety of indole-1-carbi-
nols. Furthermore. all the tested derivatives showed an apop-
topic effect on FaO cells. Work continues on yield
improvement and on the determination of the antitumoral
effect of all the other novel derivatives, in order to determine
an accurate structure/activity relationship, which will be
discussed elsewhere.
In order to evaluate the possible apoptotic effect of I3C and
of our new derivatives, we determined the expression of the
poly(ADP-ribose) polymerase (PARP) protein by Western blot
analysis. The nuclear enzyme PARP transfers ADP-ribose moi-
eties from NADþ to various nuclear proteins, and poly(ADP)-
ribosylation has been reported to be responsible for changes
in nuclear structure and function. A pro-apoptotic function
of PARP has been proposed based on the finding that its
overexpression stimulates apoptosis [14] and that treatment
with a specific chemical inhibitor of PARP, 3-aminobenza-
mide, suppressed apoptosis [15]. Following caspase-3 acti-
vation, PARP is cleaved at Asp214 early in apoptosis; the
cleaved PARP loses enzymatic activity [16] and the resulting
N-terminal and C-terminal PARP fragments accelerate
apoptosis [17].
Experimental
Chemistry
Unless otherwise stated, all reagents were purchased from com-
mercial sources and used without further purification. Reaction
progress was monitored by TLC using Aldrich silica gel 60 F254
(0.25 mm) with detection by UV. Chromatography was per-
formed using Aldrich silica gel (60–120) mesh size with freshly
distilled solvents. 1H and 13C NMR were recorded by Varian 300
and 400 MHz using tetramethylsilane as internal standard.
Microanalysis for CHN were performed by a Carlo Erba 1106
Elemental Analyzer. GC-MS: Low resolution mass spectrometric
experiments were carried out on a Saturn 2000 ion-trap coupled
with a Varian 3800 gas chromatograph (Varian, Walnut Creek,
CA) operating under EI conditions (electron energy 70 eV, emis-
sion current 20 mA, ion-trap temperature 2008C, manifold
temperature 808C, automatic gain control (AGC) target 21.000)
with the ion trap operating in scan mode (scan range from
m/z 40–400 at a scan rate of 1 scan/s). Aliquots of 1 mL of solutions
1.0 ꢁ 10ꢂ5 M in chloroform have been introduced into the gas
chromatographer inlet. A CIP Sil-8 CB Lowbleed/MS capillary
column (30 m, 0.25 mm i.d., 0.25 mm film thickness) was used.
The oven temperature was programmed from 1508C (held for
2 min) to 3108C at 308C/min (held for 2 min). The temperature
was then ramped to 350 at 208C/min. The transfer line was
maintained at 2508C and the injector port (30:1 split) at
2308C. IR spectra were recorded in KBr or neat on a Perkin-
Elmer 1310 spectrophotometer.
The results shown in Figure 4 clearly indicate that, at the
highest tested concentration, I3C as well as derivatives 2a, 2d
and 2e caused an increase of the cleaved PARP protein in FaO
cells (Figure 4).
Further studies are needed in order to better characterize
the I3C and 2a, 2d, 2e proapoptotic effect in FaO cells.
Moreover, similar investigations are now in progress with
the other indole-1-carbinol derivatives 2c, 2f, 2g, 2h and 2k.
Nuclear magnetic resonance analysis of acid stability.
Indole-1-carbinols 2a, 2b, 2d, 2e and 2f were dissolved in
0.75 mL of CD3OD. To each solution, 100 mL of deuterium-
labeled HCl (35 wt% in D2O, 99 atom% D) was added and nuclear
magnetic resonance (NMR) spectra were recorded in a 300 MHz
NMR spectrometer at room temperature at different time
intervals.
Figure 4. Western blot analysis of PARP in FaO cells treated with
increasing concentrations of indole-3-carbinol (I3C) and derivatives
2a, 2d and 2e for 24 hours. Control cells (CO) were treated with an
equivalent amount of the solvent alone. Actin was used as loading
control.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

200
G. Zedda et al.
Arch. Pharm. Chem. Life Sci. 2012, 345, 195–202
General synthetic procedure for the synthesis of indole-1-
carbinols 2a–k
Methyl 1-(hydroxymethyl)-1H-indole-3-carboxylate 2f
White solid (m.p. 111–1138C). H NMR (300 MHz, CDCl3): d 7.96
1
(d, J ¼ 7.5 Hz, 1H), 7.62 (s, 1H), 7.38 (d, J ¼ 7.5 Hz, 1H), 7.20–7.14
(m, 2H), 5.43 (s, 2H), 3.73 (s, 3H) ppm. 13C NMR (300 MHz, DMSO):
d 169.81, 137.02, 126.65, 126.41, 123.00, 122.47, 121.43, 116.12,
109.67, 69.99, 52.84 ppm. EI-MS m/z (%): 205 (23), 175 (74), 161 (5),
144 (100), 116 (27), 89 (22). IR (KBr): 3270, 1750, 1650, 1260,
1160 cmꢂ1. Anal. Calcd for C11H11NO3: C 64.39, H 5.36, N 6.83.
Found: C 64.41, H 5.34, N 6.82.
A
mixture of indole
1
(8.54 mmol), paraformaldehyde
(10.25 mmol) and oxide (145.18 mmol) was stirred for the period
indicated (TLC) at 658C. After reaction, the crude mixture was
washed with acetone and filtered through celite. The concen-
trated filtrate was flash chromatographed (hexane/ethyl acetate
3:1, 5:1 or 1:3) on silica gel, obtaining the desired product.
(1H-Indol-1-yl)methanol 2a
1-(Hydroxymethyl)-1H-indole-3-carbonitrile 2g
Sticky oil. 1H NMR (300 MHz, CDCl3): 7.62 (d, J ¼ 7.8 Hz, 1H), 7.45
(d, J ¼ 7.8 Hz, 1 H), 7.25 (t, 1H), 7.16 (m, 2H), 6.53 (d, J ¼ 3.3, 1 H),
5.61 (d, J ¼ 4.8 Hz, 2 H) ppm. 13C NMR (300 MHz, CDCl3):
White solid (m. p. 105–1078C). 1H NMR (300 MHz, DMSO): d 8.45
(s, 1H), 7.84 (d, J ¼ 8.5 Hz, 1H), 7.75 (d, J ¼ 8.5 Hz, 1H), 7.50–7.38
(m, 2H), 6.93 (t, J ¼ 7.2 Hz, 1H), 5.72 (d, J ¼ 7.2 Hz, 2H) ppm.
13C NMR (300 MHz, DMSO): d 141.87, 139.93, 132.55, 128.72,
127.44, 123.88, 121.07, 117.29, 89.26, 74.81 ppm. EI-MS m/z (%):
172 (45), 142 (68), 116 (100). IR (KBr): 3300, 2860, 2250, 1600,
1480, 1260 cmꢂ1. Anal. Calcd for C10H8N2O: C 69.77, H 4.65,
N 16.28. Found: C 69.76, H 4.65, N 16.29.
d
135.93, 128.80, 125.89, 121.24, 119.00, 118.42, 110.12,
109.06, 68.70 ppm. EI-MS m/z (%): 147 (55), 130 (4), 117 (100),
103 (1), 90 (40), 89 (45). IR (neat): 3250, 2980, 1625, 1470,
1260 cmꢂ1. Anal. Calcd for C9H9NO: C 73.47, H 6.12 N 9.52.
Found: C 73.40, H 6.03, N 9.49.
(3-Methyl-1H-indol-1-yl)methanol 2b
(3-(2-Bromophenyl)-1H-indol-1-yl)methanol 2h
Sticky oil. 1H NMR (300 MHz, CDCl3): d 7.59 (d, J ¼ 7.5 Hz, 1H),
7.39 (d, J ¼ 7.5 Hz, 1H), 7.27 (t, 1 H), 7.19 (t, 1 H), 6.91 (s, 1H),
5.50 (s, 2H), 2.34 (s, 3H) ppm. 13C NMR (300 MHz, DMSO): d 135.83,
128.79, 125.96, 121.21, 118.84, 118.53, 110.12, 109.56, 68.35,
9.67 ppm. EI-MS m/z (%): 161 (77), 144 (8), 139 (100), 103 (10),
77 (12). IR (neat): 3300, 2860, 1600, 1480, 1260 cmꢂ1. Anal. Calcd
for C10H11NO: C 74.53, H 6.83, N 8.69. Found: C 74.49, H 6.82,
N 8.70.
Gummy. 1H NMR (300 MHz, DMSO): d 7.93 (d, J ¼ 8.2 Hz, 1H),
7.86 (s, 1H), 7.77 (d, J ¼ 8.2 Hz, 1H), 7.73 (d, J ¼ 7.8 Hz, 1H),
7.56 (t, J ¼ 7.8 Hz, 2H), 7.39–7.24 (m, 3H), 6.61 (t, J ¼ 7.2 Hz,
1H), 5.71 (d, J ¼ 7.2 Hz, 2H) ppm. 13C NMR (300 MHz, DMSO):
d 136.30, 134.95, 128.69, 127.43, 127.07, 126.11, 124.69, 122.61,
120.78, 120.13, 118.51, 109.68, 69.93 ppm. EI-MS m/z (%): 303
[Mþ2] (24), 301 (22), 271 (45), 194 (100), 116 (67), 77 (40).
IR (KBr): 3310, 1640, 1420, 1260, 1130 cmꢂ1. Anal. Calcd
for C15H12 BrNO: C 59.80, H 3.99, N 4.65. Found: C 59.77,
H 4.00, N 4.62.
(3-(2-Methoxyphenyl)-1H-indol-1-yl)methanol 2c
Gummy. 1H NMR (300 MHz, CDCl3): d 7.78 (d, J ¼ 7.8 Hz, 1H),
7.62 (d, J ¼ 7.8 Hz, 1H), 7.53 (d, J ¼ 8.1 Hz, 1H), 7.28 (m, 2H),
7.19 (t, 1H), 7.04 (q, 2H), 5.67 (s, 2H), 5.30 (s, 1H), 3.85 (s, 3H) ppm.
13C NMR (300 MHz, DMSO): d 156.33, 144.01, 135.66, 129.71,
128.13, 127.21, 123.70, 121.35, 120.60, 119.97, 119.76, 111.70,
110.60, 68.78, 54.36 ppm. EI-MS m/z (%): 253 (25), 223 (100), 208
(42), 180 (26), 152 (20), 130 (8). IR (KBr): 3410, 1650, 1420, 1260,
1130 cmꢂ1. Anal. Calcd for C16 H15NO2: C 75.89, H 6.38, N 5.53.
Found: C 75.90, H 6.33, N.5.52.
(3-Bromo-1H-indol-1-yl)methanol 2i
White solid (decomposes). EI-MS m/z (%): 227 [Mþ2] (100), 225 (98),
197 (25), 195 (23), 116 (50), 89 (20).
(3-Iodo-1H-indol-1-yl)methanol 2j
White solid (decomposes). EI-MS m/z (%): 273 (100), 243 (88), 116
(65), 89 (23).
(2-Methyl-1H-indol-1-yl)methanol 2k
1-(Hydroxymethyl)-1H-indole-3-carbaldehyde 2d
1
Sticky oil . H NMR (300 MHz, CDCl3): d 7.54 (d, J ¼ 8.4 Hz, 1H),
White solid (m.p. 101–1038C). 1H NMR (300 MHz, DMSO): d 10.06
(s, 1H), 8.47 (s, 1H), 8.21 (d, J ¼ 8.1 Hz, 1H), 7.79 (d, J ¼ 8.1 Hz,
1H), 7.47–7.36 (m, 2H), 6.92 (t, 1H), 5.73 (d, 2H) ppm. 13C NMR
(300 MHz, DMSO): d 190.22, 145.64, 141.71, 130.09, 128.77,
12786, 126.20, 122.65, 116.77, 74.90 ppm. EI-MS m/z (%): 175
(23), 147 (20), 146 (55), 116 (100). IR (KBr): 3350, 2980, 2850,
1725, 1660, 1280 cmꢂ1. Anal. Calcd for C10H9NO2: C 68.57,
H 5.14, N 8.00. Found: C 68.55, H 5.16, N 7.87.
7.33 (d, J ¼ 8.4 Hz, 1H), 7.21 (t, J ¼ 8.4 Hz, 1H), 7.13 (t, J ¼ 8.4 Hz,
1H), 6.81 (s, 1H), 5.38 (s, 2H), 2.28 (s, 3H) ppm. 13C NMR (300 MHz,
DMSO): d 141.59, 133.02, 130.96, 125.41, 124.48, 114.83, 105.56,
105.43, 64.95, 19.04 ppm. EI-MS m/z (%): 161 (50), 144 (16),
130 (100), 117 (5), 103 (20). IR (neat): 3310, 2840, 1610, 1480,
1260 cmꢂ1. Anal. Calcd for C10 H11NO: C 74.53, H 6.83, N 8.69.
Found: C 74.51, H 6.79, N 8.66.
Biological general methods
Cell line and culture
1-(1-(Hydroxymethyl)-1H-indol-3-yl)ethanone 2e
White solid (m.p. 128–1308C). H NMR (300 MHz, DMSO): d 8.51
1
(s, 1H), 8.32 (d, J ¼ 8.7 Hz, 1H), 7.75 (d, J ¼ 8.7 Hz, 1H), 7.37
(m, 2H), 6.84 (t, J ¼ 7.5 Hz, 1H), 5.70 (d, J ¼ 7.5 Hz, 2H), 2.57
(s, 3H) ppm. 13C NMR (300 MHz, DMSO): d 218.51, 136.42, 127.65,
126.58, 122.93, 122.27, 121.56, 116.20, 109.99, 69.61, 28.14 ppm.
EI-MS m/z (%): 189 (27), 174 (12), 146 (45), 116 (100). IR (KBr): 3400,
2850, 1715, 1660, 1420, 1300 cmꢂ1. Anal. Calcd for C11H11NO2:
C 69.84, H 5.82, N 7.41. Found: C 69.82, H 5.79, N 7.40.
Rat FaO cell line was supplied by Interlab Cell Line Collection
(Servizio Biotecnologie, IST, Genova, Italy). FaO cells were main-
tained in Dulbecco’s medium (DMEM plus Glutamax I,
Invitrogen S.r.l., Milano, Italy) supplemented with penicillin,
streptomycin and 10% heat-inactivated fetal calf-serum (FCS)
(Invitrogen) in a humidified atmosphere of 5% CO2/95% air, at
378C. Indole-3-carbinol (I3C) and compounds 2a, 2b, 2d and 2e
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2012, 345, 195–202
Indole-1-Carbinols as Cytotoxics in Hepatocarcinoma Cells
201
were dissolved in DMSO and were added to the culture media to
the final concentrations specified in the text. Control cells were
treated with an equivalent amount of the solvent alone.
This work was supported by Ministero dell’Universita`, dell’Istruzione e
della Ricerca MIUR- PRIN 2008 (Processi e metodologie per la preparazione
di molecole eterocicliche biologicamente attive). We thank Professor
Gabriele Cruciani for helpful use of the MetaSite (Molecular Discovery
LtD) computational procedure for metabolism prediction.
Cell viability
Cell viability was determined by the uptake of Neutral Red (NRU)
by the lysosomes of viable cells.
The authors have declared no conflict of interest.
Determination of viability of the adherent cells by NRU assay
was performed according to Borefreund and Puerner [18]. The
value obtained for treated cells was expressed as percentage of
the value obtained in control cells. All experiments were per-
formed in triplicate and were repeated at least three times.
References
[1] R. J. Sundberg, In The Chemistry of Indoles; (Ed. Academic Press,
New York), 1996, p 113.
Western blotting analysis
[2] J.-R. Weng, C.-H. Tsai, S. K. Kulp, C. S. Chen, Cancer Lett. 2008,
262, 153–163.
Cell lysates, obtained by homogenization with lysis buffer
(1ꢁ PBS, 1% IGEPAL, 0.5% sodium deoxycholate, 0.1% SDS with
100 mg/mL phenylmethylsulfonylfluoride (PMSF), 300 mg/mL
aprotinin and 100 mg/mL sodium orthovanadate, all purchased
from Sigma), were incubated 30 min on ice and centrifuged at
12.000 rpm at 48C, and the supernatants were recovered. Protein
concentration was determined according to the method of
Bradford [19] using BSA as standard (DC Protein Assay, Bio Rad
Laboratories, Hercules, CA). For immunoblotting analysis, equal
amounts of proteins were electrophoresed on SDS 8% polyacyl-
amide gels. After gel electrotransfer onto nitrocellulose mem-
branes, at 800 mA for 4 h at 48C, to ensure equivalent protein
loading and transfer in all lanes, the membranes and gels were
stained with 0.5% (w/v) Ponceau S red (ICN Biomedicals) in 1%
acetic acid, and with Coomassie blue (ICN Biomedicals) in 10%
acetic acid, respectively. Before staining, gels were fixed in 25%
(v/v) isopropanol and 10% acetic acid (Sigma). After blocking in
TBS containing 0.05% Tween 20 (Sigma) (TBS-T) and 5% non-fat
dry milk, overnight at 48C, membranes were washed in TBS-T and
incubated with the appropriate primary antibodies diluted in
blocking buffer. The same membrane was used to detect the
expression of PARP and actin. Depending on the origin of
primary antibody, filters were incubated with anti-mouse, or
anti-rabbit horseradish peroxidase-conjugated IgGs (Santa Cruz
Biotechnology). Immunoreactive bands were identified with
chemiluminescence detection system, as described by the man-
ufacturer (Supersignal Substrate, Pierce, Rockford, IL). For immu-
noblotting experiments, the rabbit polyclonal anti-poly(ADP-
ribose) polymerase (PARP) (Cell Signaling Technologies,
Beverly, MA) and mouse monoclonal anti-actin (clone AC-40)
(Sigma-Aldrich, Milano, Italy) were used.
[3] W. R. Chao, D. Yean, K. Amin, C. Green, L. Jong, J. Med. Chem.
2007, 50, 3412–3415.
[4] (a) J.-R. Weng, C.-H. Tsai, S. K. Kulp, D. Wang, C.-H. Lin, H.-C.
Yang, Y. Ma, A. Sargeant, C.-F. Chiu, M.-H. Tsai, C.-S. Chen,
Cancer Res. 2007, 67, 7815–7824; (b) H. A. Omar, A. M.
Sargeant, J.-R. Weng, D. Wang, S. K. Kulp, T. Patel, C.-S.
Chen, Mol. Pharm. 2009, 76, 957–968.
[5] (a) A. R. Katritzky, S. Rachwal, B. J. Rachwal, Chem. Soc. Perkin
Trans. 1, 1987, 799–804; (b) A. R. Katritzky, K. Manju, S. K.
Singh, N. K. Meher, Tetrahedron 2005, 62, 2555–2581; (c) G.
Deguest, L. Bischoff, C. Fruit, F. Marsais, Org. Lett. 2007, 9,
1165–1167.
[6] L. Hong, W. Sun, C. Liu, L. Wang, R. Wang, Chem. Eur. J. 2010,
16, 440–444.
[7] (a) Y. Sui, L. Liu, J.-L. Zhao, D. Wang, Y.-J. Chen, Tetrahedron
Lett. 2007, 48, 3779–3782; (b) Q. Wang, X. Cui, S. Liu, L. Sun, C.
Li, Synlett 2008, 11, 1729–1733.
[8] J. M. McFarland, N. S. Joshi, M. B. Francis, J. Am. Chem. Soc.
2008, 130, 7639–7644.
[9] H.-C. Hsu, D.-R. Hou, Tetrahedron Lett. 2009, 50, 7169–
7171.
[10] (a) D. Dhanak, C. B. Reese, J. Chem. Soc. Perkin Trans 1, 1986,
2181–2186; (b) J.-P. Lio, C.-Y. Wu, H.-P. Hsieh, C.-Y. Chang,
C.-M. Che, C.-C. Kuo, J.-Y. Chang, J. Med. Chem. 2007, 50, 4548–
4552; (c) W. Kemnitzer, J. Drewe, S. Jiang, H. Zhang, C.
Crogan-Grundy, D. Labreque, M. Bubenick, G. Attardo, R.
Denid, S. Lamothe, H. Gourdeau, B. Tseng, S. Kasibhatla,
Xiong S. Cai, J. Med. Chem. 2008, 51, 417–423; (d) M. S. Scott,
A. C. Lucas, C. A. Luckhurst, J. C. Prodger, D. J. Dixon, Org
Biomol. Chem. 2006, 4, 1313–1327.
Statistical analysis
[11] S. Roy, G. W. Gribble, Synth. Commun. 2007, 37, 829–
837.
Instat software (GraphPad Prism5, San Diego, CA) was used to
analyze the data. One-way analysis of variance (ANOVA) with
post hoc analysis using Tukey’s multiple comparison test was
used for parametric data. The results of multiple observations
were presented as the means ꢀ S.D of at least 3 separate exper-
[12] Y. Cui, F. Li, Z.-H. Lu, S. Wang, Dalton Trans. 2007, 2634–
2643.
[13] (a) Predictions on metabolic transformations have been
performed by using the MetaSite 3.0.4 (Molecular Discovery
LTd) computational procedure; (b) G. Cruciani, E. Carosati,
B. De Boeck, K. Ethirajulu, C. Mackie, T. Howe, R. Vianello,
J. Med. Chem. 2005, 48, 6970–6979.
iments.
A p value of <0.05 was considered statistically
significant.
Supporting information
[14] L. Van Gool, R. Meyer, E. Tobiasch, C. Cziepluch, J. C.
Jauniaux, A. Mincheva, P. Lichter, G. G. Poirier, A. Burkle,
J. H. Kupper, Eur. J. Biochem. 1997, 244, 15–20.
1H and 13C NMR spectra.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

202
G. Zedda et al.
Arch. Pharm. Chem. Life Sci. 2012, 345, 195–202
[15] C. Nosseri, S. Coppola, L. Ghibelli, Exp. Cell. Res. 1994, 212,
367–373.
[18] E. Borenfreund, J. A. Puerner, Toxicology 1986, 39, 121.
[19] M. M. Bradford, Anal. Biochem. 1976, 72, 248.
[16] Y. A. Lazebnik, S. H. Kauffmann, S. Desnoyers, G. G. Poirier,
W. C. Earnshaw, Nature 1994, 371, 346–347.
[20] H.-C. Hsu, D.-R. Hou, Tetrahedron Lett. 2009, 50, 7169–7171.
[21] H. Mitsudera (Sumitomo Chemical Company). EP 1710234
A1. 2006.
[17] J. W. Kim, K. Kim, K. Kang, C. O. Joe, J. Biol. Chem. 2000, 275,
8121–8125.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com