Exploring bis-(indolyl)methane moiety as an alternative and innovative CAP group in the design of histone deacetylase (HDAC) inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.03.101

In order to gather further knowledge about the structural requirements on histone deacetylase inhibitors (HDACi), starting from the schematic model of the common pharmacophore that characterizes this class of molecules (surface recognition CAP group-conne

Full text of DOI:10.1016/j.bmcl.2009.03.101

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2840–2843
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Exploring bis-(indolyl)methane moiety as an alternative and innovative
CAP group in the design of histone deacetylase (HDAC) inhibitors
*
Giuseppe Giannini , Mauro Marzi, Maria Di Marzo, Gianfranco Battistuzzi, Riccardo Pezzi, Tiziana Brunetti,
Walter Cabri, Loredana Vesci, Claudio Pisano
Sigma-Tau Research and Development, Via Pontina Km 30,400, 00040 Pomezia, Roma, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
In order to gather further knowledge about the structural requirements on histone deacetylase inhibitors
(HDACi), starting from the schematic model of the common pharmacophore that characterizes this class
of molecules (surface recognition CAP group—connection unit—linker region—Zinc Binding Group), we
designed and synthesized a series of hydroxamic acids containing a bis-(indolyl)methane moiety. HDAC
inhibition profile and antiproliferative activity were evaluated.
Received 13 February 2009
Revised 18 March 2009
Accepted 21 March 2009
Available online 26 March 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
HDAC inhibitors
Bis-indoles
Hydroxamic acids
Antitumor
Inhibition of histone deacetylases is emerging as a promising
new strategy in human cancer therapy.1 Histones are nuclear core
proteins accountable for the regulation of transcription and cell cy-
cle progression. These activities are dependent on the level of acet-
form-specificity. However, the exact mechanisms by which these
inhibitors lead to the observed biological effect are still not known.
The mode of action may differ from one inhibitor to another be-
cause of the chemical structure (leading to a particular modulation
of the various HDACs isoforms) or because of the pharmacokinetic
profile. Although, the requirement of isoform specific inhibition is
not yet unambiguously established, research in this area is mainly
oriented toward isoform-specific HDACi.4
According to the usual schematic segmentation of the common
pharmacophore (Fig. 2), HDACi are characterized by a surface rec-
ognition zone (CAP group, blue), a connection unit (or kink atom,
black), a linker region (usually hydrophobic, red) and a zinc bind-
ing group (ZBG, green).
ylation and deacetylation of specific lysine e-amino groups of the
proteins backbone. These processes are controlled by two families
of enzymes: histone acetyl transferases (HATs) and histone deacet-
ylases (HDACs), respectively. The 18 known human HDACs mem-
bers are classified into four categories. Class I (HDAC 1, 2, 3, 8),
class II (HDAC 4, 5, 6, 7, 9, 10), and class IV (HDAC 11) are Zn-
dependent enzyme, while class III (sirtuins) are NAD+-dependent
enzymes.2
The great potential of these epigenetic modulators was recog-
nized early on, but most of the early research aiming at finding
more potent HDAC inhibitors did not focus on the role of each iso-
form, giving rise to non selective inhibitors.
Over the past few years, a lot of efforts have been done in the
field of HDACi and more than a hundred patents claiming new
chemical series have emerged. A number of molecules targeting
HDACs are under clinical investigation as anticancer and the first
one (SAHA—vorinostat; ZolinzaÒ; Fig. 1) has been approved by
the FDA for the treatment of cutaneous T-cell lymphoma.3
Furthermore, few HDACi are also currently investigated as sin-
gle agent therapy or in combination with other active ingredients.
Most of them are pan-inhibitors, with only few exceptions of iso-
O
NHOH
N
H
H H
O
Figure 1. SAHA—vorinostat (ZolinzaÒ).
CAP
CU
LINKER
ZBG
* Corresponding author. Tel.: +39 0691393640; fax: +39 0691393638.
E-mail address: giuseppe.giannini@sigma-tau.it (G. Giannini).
Figure 2. HDACIs common pharmacophore schematic segmentation.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.101

G. Giannini et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2840–2843
2841
R
The hydrochloride salt of 5-morpholylmethyl derivative 4m was
prepared to enhance the solubility property.
H
H
N
We also modified the nature of the chain, replacing the aliphatic
chain by an unsaturated variant (cinnamic system, 4o and 4p).
Finally, in order to evaluate an alternative ZBG, we prepared
compound 3q which is the o-aminobenzamide analogue of com-
pound 4d.
NR1R2
H
X
O
N
R
Condensation of two equivalents of (un)substituted indole
Figure 3. Bis-(indolyl)methane derivatives general structure.
starting material with one equivalent of
x-oxoaliphatic esters
[Scheme 1, step a, X = (CH2)n] or para-formyl trans-cinnamic acid
[Scheme 1, step a, X = Ph-CH2@CH2] using dysprosium triflate7 as
Lewis acid, led to the formation of the desired bis-indolyl system
in excellent yields.
Our project was oriented toward the exploration of a novel CAP
group, the bis-(indolyl)methane moiety, as a substitution of SAHA
and SAHA-like scaffolds.
The bis-(indolyl)methane derivatives show interesting biologi-
cal activities.5 Besides, bis-(indolyl)methane is a product obtained
under spontaneous dehydratation and condensation of the well-
known natural antitumoral agent, indole-3-carbinol.6
Geminal bis-(indolyl) moiety containing compounds were de-
signed and synthesized in order to gather information regarding
the steric and electronic requirements of HDACi. The compounds
thus obtained enabled to elaborate a SAR around the above men-
tioned four regions identified in Figure 2 (Fig. 3).
Besides Dy(OTf)3, used as a Lewis acid stoichiometrically, on
8
gram scale-up, we also used catalytic amount of I2 or of tri-
chloro-1,3,5-triazine (TCT) in CH3CN at rt.9 An alkaline hydrolysis
(Scheme 1, step b) was necessary to obtain the carboxylic acids
derivatives from the esters 1a–n. Sometimes, asymmetric 2,30
bis-indole derivatives (5) were isolated as side products. When
R = H and X = (CH2)5 this by-product was used, according to step
b (Scheme 1), to give intermediate 6 which was converted into
the corresponding hydroxamate derivative (7) in a two step proce-
dure (see Scheme 2).
We synthesized 2,20-bis-(indolyl)methane derivatives 4a–e—
where R = H and X = (CH2)n with n ranging from 2 to 6—to explore
the influence of methylene chain length on biological activity.
Experimental data showed that the optimal linker length consisted
of five methylenes (4d; ST2741).
The carboxylic acid intermediates 2a–p were condensed
(Scheme 2, step c) with O-benzyl-hydroxylamine hydrochloride
or with o-phenylenediamine to give the corresponding protected
hydroxamic acids intermediates 3a–p or o-aminobenzamide deriv-
ative 3q, which were easily purified by silica gel chromatography.
Subsequent condensation was performed using typical peptide
synthesis condensing agents (i.e., PyBOP or HATU) or, for gram
Based on these preliminary results, we synthesized pentyl
derivatives with various substituents on the indole ring (4f–n).
COOEt
COOEt
X
X
O
H
N
R
OR3
R
R
R
a
X
+
2
+
H
N
H
N
H
R
N
H
N
H
X = (CH2)n
R3 = Et
O
5
1a-1e X = (CH2)n; R = H; n = 2-6;
1f
X = (CH2)5; R = 4-F
1g
1h
1i
X = (CH2)5; R = 5-Me
X = (CH2)5; R = 7-MeO
X = (CH2)5; R = 7-Et
X = (C6H4)-CH2=CH2
R3 = H
X = (CH2)5
R= H
1l
1m
1n
X = (CH2)5; R = 2-Me
X = (CH2)5; R = 5-morpholylmethyl
X = (CH2)5; R = 5-NO2
a
b
b
COOH
COOH
X
COOH
X
X
H
N
R
R
R
R
N
H
N
H
N
H
N
H
N
H
6 X = (CH2)5
2o
2p
X = (C6H4)-CH2=CH2; R = H
X = (C6H4)-CH2=CH2; R = 2-Me
2a-2e X = (CH2)n; R = H; n = 2-6;
2f
X = (CH2)5; R = 4-F
2g
2h
2i
2l
2m
2n
X = (CH2)5; R = 5-Me
X = (CH2)5; R = 7-MeO
X = (CH2)5; R = 7-Et
X = (CH2)5; R = 2-Me
X = (CH2)5; R = 5-morpholylmethyl
X = (CH2)5; R = 5-NO2
Scheme 1. Bis-(indolyl)methane derivatives synthesis: condensation and hydrolysis steps. Reagents and conditions: (a) Dy(OTf)3, MeOH/H2O (3/1), rt–50 °C; (b) NaOH, THF/
MeOH/H2O (3:3:1), rt.

2842
G. Giannini et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2840–2843
O
O
NR4R5
X
X
NHOH
f
R
R
R
R
N
H
N
N
H
N
H
H
4a-4e X = (CH2)n; R = H; n = 2-6;
3a-3e X = (CH2)n; R = H; n = 2-6
4f
X = (CH2)5; R = 4-F
X = (CH2)5; R = 5-Me
X = (CH2)5; R = 7-MeO
X = (CH2)5; R = 7-Et
3f
X = (CH2)5; R = 4-F
X = (CH2)5; R = 5-Me
X = (CH2)5; R = 7-MeO
X = (CH2)5; R = 7-Et
4g
4h
4i
3g
3h
3i
c
R4= OBz; R5= H
2a-2p
4l
X = (CH2)5; R = 2-Me
3l
X = (CH2)5; R = 2-Me
4m
4n
4o
4p
X = (CH2)5; R = 5-morpholylmethyl
X = (CH2)5; R = 5-NO2
X = (C6H4)-CH2=CH2; R = H
X = (C6H4)-CH2=CH2; R = 2-Me
3m
3n
3o
3p
3q
3r
X = (CH2)5; R = 5-morpholylmethyl
X = (CH2)5; R = 5-NO2
X = (C6H4)-CH2=CH2; R = H
X = (C6H4)-CH2=CH2; R = 2-Me
X = (CH2)5; R = H;
d
e
R4 = o-(C6H4)-NH2; R5=H
R4 = OH; R5=Me
2d
2d
X = (CH2)5; R = H;
O
(CH2)5
NHOH
c
f
6
H
N
N
H
7
Scheme 2. Bis-(indolyl)methane derivatives synthesis coupling and deprotection steps. Reagents and conditions: (c) PyBOP, NMM, O-Bz-hydroxylamine.HCl, DCM, rt–70 °C
(3a: 68%; 3b: 60%; 3c: 62%; 3d: 70%; 3e: 57%; 3f: 70%; 3g: 69%; 3h: 52%; 3i: 69%; 3l: 49%; 3m: 64%; 3n: 68%; 3o: 51%; 3p: 57%); (d) PyBOP, o-(NH2)2-C6H4, TEA, DCM, rt; (e)
ClCOOEt, TEA, CH3NHOH.HCl, THF, 0 °C–rt (3r: 60%). (f) Pd/C, H2, MeOH, rt (4a: 71%; 4b: 97%; 4c: 54%; 4d: 78%; 4e: 82%; 4f: 35%; 4g: 36%; 4h: 40%; 4i: 67%; 4l: 41%; 4m: 75%;
4n: 51%; 4o: 68%; 4p: 80%; 7: 8%).
scale-up, using the cheaper ClCOOEt (ethyl chloroformate). Finally,
benzyl removal under reductive conditions (Scheme 2, step f) affor-
ded the desired hydroxamic acids 4a–p.
In order to establish a SAR and investigate the chemical stability
of the inhibitors, we also focused our attention on N-atoms synthe-
sizing various N-methylated derivatives. Direct coupling of inter-
mediate 2d with N-methylhydroxylamine hydrochloride led to
compound 3r (Scheme 2, step e).
1. This intermediate was subjected to direct coupling with hydrox-
ylamine hydrochloride and with N-methylhydroxylamine hydro-
chloride to give compounds 9 and 10, respectively (Scheme 3).
To evaluate the effect of the monoindolyl group as CAP, we syn-
thesized an analogue of 4c starting from the commercially avail-
able 6-(1H-indol-3-yl)hexanoic acid 11 to obtain compound 12
(Scheme 4).
For all synthesized compounds, we first evaluated the in vitro
anti-proliferative activity, using H460 and HCT116 tumor cell lines
(Table 1) with SAHA10 as a reference compound.
Starting from 1-methylindole, we obtained bis-indolyl carbox-
ylic acid derivative 8, according to step a and b of previous Scheme
Biological data suggested product 4d as possessing an optimal
linker length as
derivatives.
a hit compound among the un-substituted
Then, the following compounds that were representative of the
various structural modifications, although not always showing a
good anti-proliferative activity, were selected for further biological
investigation: 7-methoxy- (4h; ST3043) and the 5-nitro-derivative
O
NHOH
(CH2)5
g
O
OH
N
N
Me
Me
(CH2)5
9
OH
OH
NHOH
O
N
O
N
N
Me
O
Me
Me
(CH2)5
h
8
i
N
N
Me
N
N
Me
H
H
10
11
12
Scheme 3. N-methylated derivatives synthesis. Reagents and conditions: (g)
ClCOOEt, TEA, NH2OH.HCl, THF, 0°–rt (9: 53%); (h) (i) ClCOOEt, TEA, NH2OBz.HCl,
THF, 0°–rt (67%), (ii) CH3I, NaH, THF, rt, (iii) Pd/C, EtOH, H2 (10: 52% 2 steps).
Scheme 4. Monoindol-3-yl derivate synthesis. Reagents and conditions: (i) HATU,
DIPEA, NH2OH.HCl, DMF, rt, 2 h (60%).

G. Giannini et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2840–2843
2843
Table 1
In vitro cytotoxic activity of novel bis-indolyl derivativesa
Compound
SAHA
X
R
R1
R2
H460 IC50
(lM)
HCT116 IC50 (lM)
b
b
3.4
1.2
4a
4b
4c
4d
4e
4f
4g
4h
4i
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-30
3-20
(CH2)2
(CH2)3
(CH2)4
(CH2)5
(CH2)6
(CH2)5
(CH2)5
(CH2)5
(CH2)5
(CH2)5
(CH2)5
(CH2)5
(C6H4)–CH2@CH2
(C6H4)–CH2@CH2
(CH2)5
(CH2)5
(CH2)5
H
H
H
H
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
H
Me
H
>20
>20
5.2
1.2
2.7
>20
>20
5.8
0.6
3.5
H
4-F
5-Me
7-MeO
7-CH3CH2
2-Me
5-Morpholylmethyl
5-NO2
H
2-Me
H
1-Me
1-Me
H
H
0.7
0.4
0.51
0.55
4.8
4.3
1.1
0.27
0.41
3.5
3.2
0.96
1.2
4l
4m
4n
4o
4p
4q
10
9
3r
7
12
3.2
2.6
2.6
>20
>20
>20
8.8
>20
0.82
22
>20
16.5
>20
6.6
>20
0.84
15
o-(C6H4)–NH2
OH
OH
OH
OH
(CH2)5
(CH2)5
Monoindol-3yl
a
Growth inhibition was measured by SRB (sulphorodamine B) assay after 24 h of treatment and 48 h of recovery.
Values are means of three experiments.
b
Table 2
Compounds 4d, 4h, 4n, 4o and 4q HDAC isoform selectivity profilea,11
Compounds
IC50(nM)
HDAC6
HDAC1
HDAC2
HDAC3
HDAC4
HDAC5
HDAC7
HDAC8
HDAC9
HDAC10
HDAC11
4d
4h
4n
4o
4q
730
83
140
4800
21,000
2000
440
590
3900
1100
350
23
18
440
260
1300
220
220
4800
150,000
590
311
140
4900
28,000
6
530
420
220
4200
/
240
180
600
730
74,000
640
340
150
3300
50,000
870
240
250
4700
73,000
430
/
/
410
5800
2.6
20
4900
/
a
Assay condition was started with a 50 lM solution, 10 doses with 1:3 dilution. Trichostatin A (TSA) was used as reference compound.
(4n; ST5732), the cinnamic derivative (4o; ST2887) and the o-ami-
nobenzamide derivative (4q; ST3071).
Supplementary data
For these selected molecules, we evaluated the in vitro inhibi-
tory activity against the 11 HDACs isoforms. In Table 2, we show
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.03.101.
IC50 (lM) data obtained using human HDAC enzymes and a fluoro-
genic peptide (p53 379–382 RHKKAc residues) at 50
lM, as the
References and notes
substrate.11
Compound 4d together with the derivatives 4h and 4n showed
an interesting HDACi profile ranging from low to high nM. These
data were consistent with their anti-proliferative activity. 4h was
found particularly promising with a selective inhibition activity,
against the isoforms HDAC1 and HDAC3 besides HDAC6, in the
low nM range.
In summary, we have demonstrated that the bis-(indolyl)meth-
ane moiety can be used as a valid CAP in the design of HDAC
inhibitors.
A Structure–Activity-Relationship (SAR) analysis has delineated
the importance of the linker length in HDAC inhibition (the most
potent derivatives present a penta-methylene linker) and the rele-
vance of the bis-(indolyl)methane system (3-indolyl derivative 12
is less potent than bis-(indolyl) analogue 4d).
The o-aminobenzamide derivative (4q), although less active
than all the other examples of HDACi having as ZBG an o-amino-
benzamide moiety, showed an unusual selectivity towards the
HDAC3 isoform. However, the data confirm the importance of the
hydroxamic acid as the ZBG of choice and show that the nature
of the substituent on the indole ring, when not too hydrophilic,
only slightly influences the in vitro activity.
1. Moradei, O.; Maroun, C. R.; Paquin, I.; Vaisburg, A. Curr. Med. Chem. Anti-Cancer
Agents 2005, 5, 529.
2. Paris, M.; Porcelloni, M.; Binaschi, M.; Fattori, D. J. Med. Chem. 2008, 51, 1505.
3. Kelly, W. K.; O’Connor, O. A.; Krug, L. M.; Chiao, J. H.; Heaney, M.; Curley, T.;
MacGregore-Cortelli, B.; Tong, W.; Secrist, J. P.; Schwartz, L.; Richardson, S.;
Chu, E.; Olgac, S.; Marks, P. A.; Scher, H.; Richon, V. M. J. Clin. Oncol. 2005, 23,
3923.
4. Bieliauskas, A. V.; Pflum, M. K. H. Chem. Soc. Rev. 2008, 37, 1402.
5. Su, Y.; Vanderlaag, K.; Ireland1, C.; Ortiz1, J.; Grage, H.; Safe, S.; Frankel, A. E.
Breast Cancer Research 2007, 9, 1.
6. Weng, J. R.; Tsai, C. H.; Kulp, S. K.; Wang, D.; Lin, C. H.; Yang, H. C.; Ma, Y.;
Sargeant, A.; Chiu, C. F.; Tsai, M. H.; Chen, C. S. Cancer Res. 2008, 67, 7815.
7. Giannini, G.; Marzi, M.; Moretti, G. P.; Penco, S.; Tinti, M. O.; Pesci, S.; Lazzaro,
F.; De Angelis, F. Eur. J. Org. Chem. 2004, 2411; Chen, D. et al Tetrahedron Lett.
1996, 37, 4467.
8. Bandgar, B. P.; Shaikh, K. A. Tetrahedron Lett. 2003, 44, 1959.
9. Sharma, G. V. M.; Janardhan Reddy, J.; Sree Lakshmi, P.; Radha Krishna, P.
Tetrahedron Lett. 2004, 45, 7729.
10. SAHA was in-house synthesized.
11. The substrate, a fluorogenic moiety bound to specific p53 fragment—residues
379–392: Arg-His-Lys-Lys(Ac)—which comprises an
e-acetylated lysine side
chain, was incubated with the 11 single HDAC purified enzymes. Upon
deacetylation of the substrate, the fluorophore was released given rise to
fluorescence emission. The latter was detected by a fluorimeter, and the IC50
values of the compounds were determined by analyzing dose-response
inhibition curves. Trichostatin
A
(TSA) was used as reference
compound.Screening was performed by Reaction Biology Corp.