The design, synthesis and structure-activity relationships of novel isoindoline-based histone deacetylase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.06.015

The design, synthesis and biological evaluation of a novel series of isoindoline-based hydroxamates is described. Several analogs were shown to inhibit HDAC1 with IC₅₀ values in the low nanomolar range and inhibit cellular proliferation of HCT116 human colon cancer cells in the sub-micromolar range. The cellular potency of compound 17e was found to have greater in vitro anti-proliferative activity than several compounds in late stage clinical trials for the treatment of cancer. The in vitro safety profiles of selected compounds were assessed and shown to be suitable for further lead optimization.

Full text of DOI:10.1016/j.bmcl.2011.06.015

Bioorganic & Medicinal Chemistry Letters 21 (2011) 4909–4912
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The design, synthesis and structure–activity relationships of novel
isoindoline-based histone deacetylase inhibitors
Michael Shultz a, , Jianmei Fan a, Christine Chen a, Young Shin Cho a, Nicole Davis a, Sheri Bickford a,
⇑
Kristen Buteau a, Xueying Cao a, Mats Holmqvist a, Meier Hsu a, Lei Jiang a, Gang Liu a, Qiang Lu a,
Chetan Patel b, Joghee Raju Suresh b, Mannangatti Selvaraj b, Laszlo Urban a, Ping Wang a, Yan Yan-Neale a,
Lewis Whitehead a, Haiyan Zhang a, Liping Zhou a, Peter Atadja a
a Novartis Institutes for Biomedical Research, Inc., 250 Massachusetts Avenue, Cambridge, MA 02139, USA
b Biocon, Syngene, Bangalore 560 100, India
a r t i c l e i n f o
a b s t r a c t
Article history:
The design, synthesis and biological evaluation of a novel series of isoindoline-based hydroxamates is
described. Several analogs were shown to inhibit HDAC1 with IC50 values in the low nanomolar range
and inhibit cellular proliferation of HCT116 human colon cancer cells in the sub-micromolar range.
The cellular potency of compound 17e was found to have greater in vitro anti-proliferative activity than
several compounds in late stage clinical trials for the treatment of cancer. The in vitro safety profiles of
selected compounds were assessed and shown to be suitable for further lead optimization.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 30 March 2011
Revised 2 June 2011
Accepted 6 June 2011
Available online 11 July 2011
Keywords:
HDAC
hERG
Cancer
Histone deacetylase
Isoindoline
SAR
Histone acetyltransferases (HATs) catalyze the acetylation of ly-
sine residues on histone and non-histone proteins. The reverse
reaction is catalyzed by histone deacetylases (HDACs) thus
promoting a more closed chromatin structure where transcription
is repressed.1,2 HDAC inhibitors have found utility for the treatment
of human cancers; Zolinza (vorinostat, SAHA, (1), Merck) was ap-
proved by the FDA for the treatment of advanced cutaneous T-cell
lymphoma (CTCL). Several small molecule HDAC inhibitors such
as belinostat (PXD-101, (2), Curagen),3,4 entinostat (SNDX-275 (3),
Syndax),5,6 mocetinostat (MGCD-0103 (4), Methylgene)7–10
(Fig. 1) and dacinostat (LAQ824 (5), Novartis AG)11 (Fig. 2) have en-
tered clinical trials for the treatment of a variety hematological and
solid tumors.
Since the discovery of LAQ824 (5), further studies led us to iden-
tify both hydroxamate based (6)12 and non-hydroxamate-based
compounds ((7) and (8), Fig. 2)13 which demonstrate dissimilar
HDAC isoform selectivity profiles. In an effort to identify additional
HDAC inhibitors, isoindoline based scaffolds have been explored.
The synthesis and in vitro anti-proliferative activities of these
analogs will be discussed.
Our goal was to generate highly potent compounds that have an
acceptable solubility and safety profile for clinical development,
and have sufficient oral exposure in preclinical animal toxicology
models for rigorous evaluation. Building on earlier efforts that gen-
erated compound 6, we sought compounds that minimized the
number of rotatable bonds in order to increase oral exposure in
rodent models.14 To this end, a series of cyclic analogs via the lin-
ker nitrogen were explored.15 One such cyclization of the linker
nitrogen of 9 with the core phenyl ring results in the isoindoline
scaffold (10) (Fig. 3). Syntheses of isoindolines of this nature that
are amenable to rapid analog synthesis were lacking, therefore,
new synthetic approaches to key intermediates were required.
An efficient lead optimization campaign required compound 13
as a late stage intermediate that could be rapidly diversified
(Scheme 1). Conventional reaction conditions were explored to
convert 5-bromophthalamide (11) to 5-bromo-2,3-dihydro-1H-
isoindole (12), but only pretreatment with BF3OEt2 followed by
borane-THF reduction generated the free amine while leaving the
bromide intact. The resulting free amine was difficult to isolate
and was therefore converted to the tert-butyl carbamate 12 prior
to isolation and purification. A subsequent Heck reaction with
methyl acrylate followed by removal of the Boc protecting group
provided the desired intermediate 13. Conversion of the secondary
amine 13 to amides, sulfonamides or tertiary amines proceeded
⇑
Corresponding author. Tel.: +1 617 871 7551; fax: +1 617 871 4081.
E-mail address: michael.shultz@novartis.com (M. Shultz).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.015

4910
M. Shultz et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4909–4912
H
N
O
with relative ease. The desired hydroxamate analogs were
obtained in good yields by treating the methyl esters with hydrox-
ylamine in the presence of sodium methoxide.
O
O
H
N
OH
S
OH
N
H
N
H
O
O
Purified HDAC-1 was used to assess the biochemical activity of
each compound.11 The anti-proliferative activities of promising
compounds were determined in the HCT116 human colon cancer
cell line. Simple alkyl carbamate and amide derivatives
(compounds 15a–e, Table 1) demonstrated sub-micromolar HDAC
IC50 values, but these were generally poorly soluble with low activ-
ity in the cellular assay. Aromatic analogs did not improve the
potency or solubility (compounds 15f and 15g). Incorporation of
a methylene spacer (compound 15h) improved the cellular activity
vorinostat
belinostat
PXD-101 (2)
SAHA (1)
N2H
N2H
H
H
N
O
N
O
O
N
N
H
O
N
H
N
etinostat
SNDX-275 (3)
mocetinostat
MGCD0103 (4)
N
N
to a level comparable to vorinostat (HCT116 IC50 = 0.81 lM).
Fig. 1. Selection of HDAC inhibitors in clinical trials.
Replacement of the phenyl group with an indole (compound 15i)
resulted in a four-fold improvement in cellular potency and with
an acceptable hERG safety profile (hERG radioligand binding
OH
N
O
IC50 >30
lM), however thermodynamic solubility was low
OH
OH
N
H
(<20 M at pH 6.8).
l
We next investigated sulfonamides. In general, alkyl and cyclo-
alkyl sulfonamides (compounds 16a–16d, Table 2) were weakly ac-
tive. The phenyl sulfonamide derivative 16e had moderate activity
which was sensitive to steric and electronic effects on the phenyl
ring. Substitution at the ortho position (16l) was tolerated, while
improved potency was obtained with meta- and para-substituted
analogs (e.g., compounds 16j and 16k). Modifications that reduced
the aryl ring electron density (compounds 16g–16i, 16m and 16n)
generally resulted in lower HDAC-1 potency, while analogs with
greater electron density in the aromatic ring (compounds 16f,
16o, and 16p) had improved affinity in the enzymatic assay.
Several analogs were profiled to assess their solubility and
in vitro safety profile. The analogs from the amide and sulfonamide
series that were tested in the hERG patch clamp assay demon-
N
H
dacinostat
LAQ824
5
O
N
H
H
N
N
N
6
Cl
H2N
N
Cl
O
O
F
N
N
O
NH2
Cl
8
7
F
Fig. 2. Selection of HDAC inhibitors from previous internal efforts.
strated IC50 values greater than 30 lM, but all suffered from a
significantly lower than desired solubility (data not shown).
The third series that was investigated was the amine based iso-
indolines (Table 3). Attempts to synthesize a more diverse set of
isoindoline analogs via reductive amination and alkylation were
unsuccessful while identical approaches worked with scaffolds
cyclization
O
O
OH
OH
N
H
N
H
H
N
R N
R
9
10
Fig. 3. Design of isoindoline based HDAC inhibitors.
Table 1
Amide and carbamate series
O
Br
O
O
Br
a
N
OH
HN
O
N
H
O
N
12
R
O
11
15
HDAC-1a IC50
(
lM)
HCT116b IC50
(lM)
b, c
Compound
R
15a
15b
15c
15d
15e
(CH3)3CO
(CH3)3C
Cyclopropyl
Cyclopentyl
Cyclohexyl
0.52
0.42
0.42
0.35
0.39
nd
nd
nd
nd
nd
O
O
d, e, f, or g
O
O
HN
R
N
14
13
*
15f
0.20
5.82
F3C
h
15g
15h
3-pyridyl
benzyl
0.38
0.13
nd
0.83
O
OH
N
H
R
N
*
15i
0.020
0.19
10
HN
Scheme 1. Reagents and conditions: (a) i. BF3OEt2, ii. BH3 THF 40 °C, iii. Boc2O,
DMAP, THF (53% yield over two steps); (b) methyl acrylate, Pd(OAc)2, P(o-tol)3,
Et3N, DMF (42% yield); (c) i. TFA/CH2Cl2; (d) RCOCl, NMM, CH2Cl2 ii. HCl, dioxane
(97% yield); (e) ROCOCO2R, Et3N, CH2Cl2; (f) RSO2Cl, Et3N, CH2Cl2; (f) RCH2Br, Et3N,
CH2Cl2; (g) RCHO, NaBH3CN, AcOH; (h) NH2OH, NaOMe, MeOH.
nd – not determined
a
Inhibition of recombinant HDAC1.
b
Anti-proliferative activity in human colon cancer HCT116 cell line.

M. Shultz et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4909–4912
4911
Table 2
Table 3
Sulfonamide series
Amine series
O
O
OH
OH
O
S
N
H
N
H
N
R
O
N
R
16
17
Compound
R
HDAC-1a IC50
(
lM)
HCT116b IC50
(
lM)
Compound
R
HDAC-1a
HCT116b
hERG patch
clampc IC50
(lM)
IC50
(lM)
IC50
(lM)
16a
16b
16c
16d
16e
16f
Methyl
Butyl
Cyclopropyl
Benzyl
Phenyl
2-Thiophenyl
4-Pyridyl
0.63
0.15
0.79
0.12
0.19
0.089
0.48
0.15
nd
3.21
nd
1.12
3.42
0.80
*
17a
17b
0.20
1.04
27.5
22
*
0.10
0.90
0.22
16g
16h
3-Pyridyl
4.26
8.84
*
N
N
16i
16j
16k
0.14
17c
17d
*
*
*
0.016
0.078
>30
5.2
HN
*
*
0.047
0.062
1.58
1.42
0.53
N
N
*
16l
0.13
0.16
2.23
2.43
17e
17f
0.026
0.17
0.042
2.2
4.8
HN
*
16m
F3C
F3C
*
16n
16o
16p
0.16
2.05
0.93
1.13
>30
*
O
N
*
0.035
0.092
O
O
a
b
c
Inhibition of recombinant HDAC1.
Anti-proliferative activity in human colon cancer HCT116 cell line.
Activity in automated patch clamp assay for hERG activity.
*
nd – not determined
a
Inhibition of recombinant HDAC1.
Anti-proliferative activity in human colon cancer HCT116 cell line.
b
Table 4
In vitro drug–drug interaction profile of select compounds
containing larger ring sizes than the isoindoline.16 Synthetic issues
in this series included highly variable yields and multiple by-
products that hampered purification. Even though a complete
SAR trend in the amine series remains frustratingly out of reach,
several key analogs provide valuable information. The benzyl
analog 17a was moderately potent against HDAC-1 enzyme and
Compound
HCT116a IC50
M)
CYP3A4 IC50
M)
CYP3A4/HCT116
ratio
(
l
(l
SAHA (1)
Belinostat (2) 0.16
SNDX-275 (3) 0.67
MGCD0103
(4)
0.81
>10
>10
0.73
0.57
>12
>62
1.1
0.31
1.8
had improved solubility (106 lM at pH 6.8). Potency was further
LAQ824 (5)
15h
15i
16f
16o
17b
17c
17d
17e
0.019
0.83
0.19
0.80
0.93
0.90
0.22
0.53
0.042
14
>50
nd
>50
>50
>50
>10
>50
>10
736
>60
improved with the cyclohexylethyl moiety (17b) however, hERG
inhibition appeared. Potency was further enhanced with the incor-
poration of the indole derivative 17c with a concomitant decrease
in hERG activity. Incorporation of a methyl group at the 2-position
of the indole (17e) resulted in a five-fold increase in cellular
potency, a similar increase in the hERG patch clamp activity was
also observed. Compound 17f was found to be less active than
other analogs in this series.
To determine if a full lead optimization campaign with this
series was warranted, in vitro profiling in comparison to several
clinical compounds was performed (Tables 4–6). Cancer patients
often take multiple medications and therefore drug–drug interac-
tions are a major concern,17 especially with the potential to use
an HDAC inhibitor as part of a combination therapy regime.18
CYP3A4 is generally regarded as the most relevant CYP450 impli-
cated in drug–drug interactions and thus was the focus of our
profiling.19 All compounds tested had CYP3A4 IC50 values greater
>63
>54
>56
>45
>94
>238
nd – not determined
a
Anti-proliferative activity in human colon cancer HCT116 cell line.
Additional profiling included the hERG automated patch clamp
(QPatch) assay (Table 5). To allow for rapid decision making on
cardiac safety, we utilized the in vitro cardiac safety index (iCSI).12
The iCSI is defined as a ratio of the hERG and the cellular anti-
proliferative IC50 values. All amides and sulfonamides tested were
inactive and relatively few amines (compounds 17b, 17d and 17e)
demonstrated activity in the patch clamp assay. While the late
stage clinical HDAC inhibitors disclosed below are inactive in
the patch clamp assay, their moderate cellular activity makes it dif-
ficult to precisely determine iCSI values. Compound 17e is the most
than 10
lM and with some having IC50 values greater than
50 M. The potency of CYP3A4 inhibition of all compounds tested
l
in this series appears significantly less than benzamide class of
HDAC inhibitors.

4912
M. Shultz et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4909–4912
Table 5
In summary, a novel class of hydroxamate based HDAC inhibi-
In vitro cardiac safety index of select compounds
tors with the general structure 10 has been identified. Synthetic
routes that enable late stage diversification, including a previously
unreported method of efficient phthalamide carbonyl reduction in
the presence of an aryl bromide functionality have been developed.
A systematic approach to improve solubility, potency and meta-
bolic stability resulted in several compounds with potency supe-
rior to several HDAC inhibitors in late stage clinical trials.
Compound 17e was identified as a candidate for lead optimization
with promising drug-like properties. This compound possesses low
nanomolar in vitro anti-proliferative activity, good solubility and
an acceptable CYP3A4 profile. The low micromolar activity in the
hERG patch clamp assay would have to be addressed by either
improving HDAC potency, reducing hERG affinity or a combination
thereof. Efforts toward this end will be reported in due course.
Compound
HDAC1 IC50
M)
HCT116a IC50
M)
hERG patch clampb
IC50 (lM)
iCSI
(l
(
l
SAHA (1)
Belinostat
(2)
SNDX-275
(3)
MGCD0103 0.152
(4)
LAQ824 (5) 0.015
0.077
0.015
0.81
0.16
>30
>30
>37
>187
0.485
0.67
0.31
>30
>30
>44
>97
0.019
0.83
0.19
0.90
0.22
0.53
0.042
2.1
10.3
>30
>30
22.2
>30
3.9
542
>36
>158
25
>136
7.4
15h
15i
0.13
0.020
17b
17c
17d
17e
17f
4.8
>30
114
>15
References and notes
a
Anti-proliferative activity in human colon cancer HCT116 cell line.
Activity in automated patch clamp assay for hERG activity.
b
1. Lehrmann, H.; Pritchard, L. L.; Harel-Bellan, A. Adv. Cancer Res. 2002, 86, 41.
2. Mai, A.; Massa, S.; Rotili, D.; Cerbara, I.; Valente, S.; Pezzi, R.; Simeoni, S.; Ragno,
R. Med. Res. Rev. 2005, 25, 261.
3. Gimsing, P.; Hansen, M.; Knudsen, L. M.; Knoblauch, P.; Christensen, I. J.; Ooi, C.
E.; Buhl-Jensen, P. Eur. J. Haematol. 2008, 81, 170.
Table 6
4. Qian, X.; LaRochelle, W. J.; Ara, G.; Wu, F.; Petersen, K. D.; Thougaard, A.;
Sehested, M.; Lichenstein, H. S.; Jeffers, M. Mol. Cancer Ther. 2006, 5, 2086.
5. Gore, L.; Rothenberg, M. L.; O’Bryant, C. L.; Schultz, M. K.; Sandler, A. B.; Coffin,
D.; McCoy, C.; Schott, A.; Scholz, C.; Eckhardt, S. G. Clin. Cancer Res. 2008, 14,
4517.
Solubility and in vitro microsomal stability profile
Compound
Rat t1/2
(min)
Human t1/2
(min)
Thermodynamic
solubility at pH 6.8
M)
(
l
6. Suzuki, T.; Ando, T.; Tsuchiya, K.; Fukazawa, N.; Saito, A.; Mariko, Y.; Yamashita,
T.; Nakanishi, O. J. Med. Chem. 1999, 42, 3001.
SAHA (1)
Belinostat (2)
SNDX-275 (3) 405
MGCD0103
(4)
LAQ824 (5)
15a
15i
16f
16j
16k
16o
17a
17c
17d
17e
17f
114
18
405
405
405
37.8
19
7. Zhou, N.; Moradei, O.; Raeppel, S.; Leit, S.; Frechette, S.; Gaudette, F.; Paquin, I.;
Bernstein, N.; Bouchain, G.; Vaisburg, A.; Jin, Z.; Gillespie, J.; Wang, J.; Fournel,
M.; Yan, P. T.; Trachy-Bourget, M. C.; Kalita, A.; Lu, A.; Rahil, J.; MacLeod, A. R.;
Li, Z.; Besterman, J. M.; Delorme, D. J. Med. Chem. 2008, 51, 4072.
8. Garcia-Manero, G.; Assouline, S.; Cortes, J.; Estrov, Z.; Kantarjian, H.; Yang, H.;
Newsome, W. M.; Miller, W. H., Jr.; Rousseau, C.; Kalita, A.; Bonfils, C.; Dubay,
M.; Patterson, T. A.; Li, Z.; Besterman, J. M.; Reid, G.; Laille, E.; Martell, R. E.;
Minden, M. Blood 2008, 112, 981.
9. Fournel, M.; Bonfils, C.; Hou, Y.; Yan, P. T.; Trachy-Bourget, M. C.; Kalita, A.; Liu,
J.; Lu, A. H.; Zhou, N. Z.; Robert, M. F.; Gillespie, J.; Wang, J. J.; Ste-Croix, H.;
Rahil, J.; Lefebvre, S.; Moradei, O.; Delorme, D.; MacLeod, A. R.; Besterman, J. M.;
Li, Z. Mol. Cancer Ther. 2008, 7, 759.
734
401
33
31.3
2.3
14
15
nd
4.1
7
8.5
29
5
12
97
99
20
34
20
20
<5
517
490
765
263
490
131
173
nd
11
15
19
10. Kell, J. Curr. Opin. Investig. Drugs 2007, 8, 485.
50
11. Remiszewski, S. W.; Sambucetti, L. C.; Bair, K. W.; Bontempo, J.; Cesarz, D.;
Chandramouli, N.; Chen, R.; Cheung, M.; Cornell-Kennon, S.; Dean, K.;
Diamantidis, G.; France, D.; Green, M. A.; Howell, K. L.; Kashi, R.; Kwon, P.;
Lassota, P.; Martin, M. S.; Mou, Y.; Perez, L. B.; Sharma, S.; Smith, T.; Sorensen,
E.; Taplin, F.; Trogani, N.; Versace, R.; Walker, H.; Weltchek-Engler, S.; Wood,
A.; Wu, A.; Atadja, P. J. Med. Chem. 2003, 46, 4609.
nd
112
30
9
2
6
17
12. Shultz, M.; Cao, X.; Chen, C.; Cho, Y. S.; Davis, N.; Eckman, J.; Fan, J.; Fekete, A.;
Firestone, B.; Flynn, J.; Green, J.; Growney, J.; Holmqvist, M.; Hsu, M.; Jansson,
D.; Jiang, L.; Kwon, P.; Liu, G.; Lombardo, F.; Lu, Q.; Majumdar, D.; Meta, C.;
Perez, L.; Pu, M.; Ramsey, T.; Remiszewski, S.; Skolnik, S.; Traebert, M.; Urban,
L.; Uttamsingh, V.; Wang, P.; Whitebread, S.; Whitebread, L.; Yan-Neale, Y.;
Yao, T. M.; Zhou, L.; Atadja, P. J. Med. Chem., in press. doi:10.1021/jm200388e.
13. Whitehead, L.; Dobler, M. R.; Radetich, B.; Zhu, Y.; Atadja, P. W.; Claiborne, T.;
Grob, J. E.; McRiner, A.; Pancost, M. R.; Patnaik, A.; Shao, W.; Shultz, M.;
Tichkule, R.; Tommasi, R. A.; Vash, B.; Wang, P.; Stams, T. Bioorg. Med. Chem., in
press. doi:10.1016/j.bmc.2011.06.030.
potent HDAC inhibitor identified in this series and has an iCSI of
114. The hERG SAR in this series appears to track with HDAC activ-
ity and additional efforts to improve this safety margin would be
warranted during a lead optimization campaign.
hERG IC50
hERG IC50
iCSI ¼
¼
cellular IC50 HCT 116 IC50
14. Veber, D. F.; Johnson, S. R.; Cheng, H. Y.; Smith, B. R.; Ward, K. W.; Kopple, K. D.
J. Med. Chem. 2002, 45, 2615.
15. Cho, Y. S.; Whitehead, L.; Li, J.; Chen, C. H.; Jiang, L.; Vogtle, M.; Francotte, E.;
Richert, P.; Wagner, T.; Traebert, M.; Lu, Q.; Cao, X.; Dumotier, B.; Fejzo, J.;
Rajan, S.; Wang, P.; Yan-Neale, Y.; Shao, W.; Atadja, P.; Shultz, M. J. Med. Chem.
2010, 53, 2952.
While the hydroxamate class of clinical HDAC inhibitors
appears to be more potent and less susceptible to CYP450 inhibi-
tion than the benzamide class, hydroxamates generally have poor
PK properties in rodent species.20 This creates an additional chal-
lenge in the development of anti-cancer agents, which generally
rely on rodent based xenograft models. The microsomal stability
and solubility of select compounds were evaluated and are
reported below (Table 6). All the analogs in the isoindoline series
have poor microsomal stability in rat liver microsomes, whereas
several analogs show enhanced stability in human liver micro-
somes. Compounds 17d and 17e have a reasonable combination
of potency, solubility and human microsomal to justify further
investigation.
16. Manuscript in preparation.
17. Egorin, M. J. Cancer Chemother. Pharmacol. 1998, 42, S22.
18. Schneider, B. J.; Kalemkerian, G. P.; Bradley, D.; Smith, D. C.; Egorin, M. J.;
Daignault, S.; Dunn, R.; Hussain, M. Invest. New Drugs 2010. doi:10.1007/
s10637-010-9503-6.
19. Williams, J. A.; Bauman, J.; Cai, H.; Conlon, K.; Hansel, S.; Hurst, S.; Sadagopan,
N.; Tugnait, M.; Zhang, L.; Sahi, J. Curr. Opin. Drug Discov. Devel. 2005, 8, 78.
20. Flipo, M.; Charton, J.; Hocine, A.; Dassonneville, S.; Deprez, B.; Deprez-Poulain,
R. J. Med. Chem. 2009, 52, 6790.