Design, synthesis and biological screening of 2-aminobenzamides as selective HDAC3 inhibitors with promising anticancer effects

Article abstract of DOI:10.1016/j.ejps.2018.08.030

Histone deacetylases (HDACs) have been found as a potential target for anticancer therapy. A number of HDAC inhibitors have been used pre-clinically and clinically as anticancer agents. In the current study, we have designed and synthesized compound 12a by combining the scaffolds of CI-994 and BG45. Moreover, the structure of compound 12a was optimized and a series of 2-aminobenzamide derivatives were synthesized further. These compounds were tested for their HDAC inhibitory activity and found to be efficient HDAC inhibitors. Compound 26c showed 11.68-fold HDAC3 selectivity over pan HDACs, better than the prototype HDAC3 inhibitor BG45. Most of these compounds exhibited antiproliferative activity in both B16F10 and HeLa cell lines. Particularly, compound 26c exhibited better antitumor efficacy in the cell lines compared to the prototype inhibitors CI-994 and BG45. It was also found to promote apoptosis as well as induced significant cell growth arrest in the G2/M phase of cell cycle in B16F10 melanoma cells. This work may provide significant insight regarding structural information to design newer small molecule HDAC3 inhibitors to fight against the target specific malignancies in future.

Full text of DOI:10.1016/j.ejps.2018.08.030

Accepted Manuscript
Design, synthesis and biological screening of 2-aminobenzamides
as selective HDAC3 inhibitors with promising anticancer effects
Prakruti Trivedi, Nilanjan Adhikari, Sk. Abdul Amin, Tarun Jha,
Balaram Ghosh
PII:
S0928-0987(18)30398-1
doi:10.1016/j.ejps.2018.08.030
PHASCI 4667
DOI:
Reference:
To appear in:
European Journal of Pharmaceutical Sciences
Received date:
Revised date:
Accepted date:
5 June 2018
11 August 2018
23 August 2018
Please cite this article as: Prakruti Trivedi, Nilanjan Adhikari, Sk. Abdul Amin, Tarun
Jha, Balaram Ghosh , Design, synthesis and biological screening of 2-aminobenzamides as
selective HDAC3 inhibitors with promising anticancer effects. Phasci (2018), doi:10.1016/
j.ejps.2018.08.030
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Design, synthesis and biological screening of 2-aminobenzamides as selective HDAC3
inhibitors with promising anticancer effects
Prakruti Trivedi
Department of Pharmacy, BITS-Pilani, Hyderabad Campus, Shameerpet, Hyderabad, India,
00078
Natural Science Laboratory, Division of Medicinal and Pharmaceutical Chemistry, Department
of Pharmaceutical Technology, Jadavpur University, Kolkata, India 700032
a
, Nilanjan Adhikarib,#, Sk. Abdul Amin
b
, Tarun Jhab,*, Balaram Ghosha*
a
5
b
#
Present Address:
School of Pharmaceutical Technology, ADAMAS University, Barasat-Barrackpore Road, P. O.
Jagannathpur, Kolkata – 700126, West Bengal, India
*
corresponding author: (T. Jha) tjupharm@yahoo.com; (B. Ghosh)
balaram@hyderabad.bitspilani.ac.in
*
Corresponding author
Balaram Ghosh, PhD
Room. A016, Department of Pharmacy
Birla Institute of Technology & Sciences-Pilani,
Hyderabad Campus, Jawahar Nagar, Shameerpet,
Hyderabad- 500 078, India
E-mail: balaram@hyderabad.bits-pilani.ac.in
Phone# 91-406-630-3639 (Office)
9
1-986-694-1923 (Cell)
Fax# 91-406-630-3998
Abstract
Histone deacetylases (HDACs) have been found as a potential target for anticancer therapy. A
number of HDAC inhibitors have been used pre-clinically and clinically as anticancer agents. In
the current study, we have designed and synthesized compound 12a by combining the scaffolds
of CI-994 and BG45. Moreover, the structure of compound 12a was optimized and a series of 2-
aminobenzamide derivatives were synthesized further. These compounds were tested for their
HDAC inhibitory activity and found to be efficient HDAC inhibitors. Compound 26c showed
1

ACCEPTED MANUSCRIPT
1
1.68-fold HDAC3 selectivity over pan HDACs, better than the prototype HDAC3 inhibitor BG-
5. Most of these compounds exhibited antiproliferative activity in both B16F10 and HeLa cell
4
lines. Particularly, compound 26c exhibited better antitumor efficacy in the cell lines compared
to the prototype inhibitors CI-994 and BG45. It was also found to promote apoptosis as well as
induced significant cell growth arrest in the G2/M phase of cell cycle in B16F10 melanoma cells.
This work may provide significant insight regarding structural information to design newer small
molecule HDAC3 inhibitors to fight against the target specific malignancies in future.
Keywords: Anticancer agent; HDAC inhibitor; isoform-selective HDAC3 inhibitor; density
functional theory (DFT); 2-aminobenzamide; structure-activity relationship (SAR)
List of abbreviations
B16F10
CDCl
murine melanoma cells
3
deuterated chloroform
CTCL
DCM
cutaneous T-cell lymphoma
dichloromethane
DFT
density functional theory
4-Dimethylaminopyridine
Dulbecco’s modified Eagle’s media
dimethylformamide
DMAP
DMEM
DMF
DMSO
DMSO-d6
EDC
dimethyl sulphoxide
deuterated dimethyl sulphoxide
1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide
electron spray ionization
ESI
2

ACCEPTED MANUSCRIPT
EtOAc
FITC
ethyl acetate
fluorescein isothiocyanate
hydrochloric acid
HCl
HeLa
HOMO
HPLC
IC50
human cervical cancer cell line
highest energy occupied molecular orbital
high performance liquid chromatography
half maximal inhibitory concentration
liquid chromatography–mass spectrometry
lowest energy unoccupied molecular orbital
3-(4,5-dimethylthiazol-2-yl)-2,5-di-phenyltetrazolium bromide
sodium sulphate
LC-MS
LUMO
MTT
Na
2
SO
4
NaHCO
NCOR1
NMR
PBS
3
sodium bicarbonate
Nuclear receptor corepressor 1
nuclear magnetic resonance
phosphate buffered saline
PI
propidium iodide
ppm
parts per million
PTCL
Rt
peripheral T-cell lymphoma
retention time
RT
room temperature
SAHA
THF
TLC
suberoylanilide hydroxamic acid
tetrahydrofuran
thin layer chromatography
3

ACCEPTED MANUSCRIPT
USFDA
ZBG
US food and drug administrations
zinc binding group
4

ACCEPTED MANUSCRIPT
1
. Introduction
Cancer is one of the major causes of death worldwide. Even though progress has been made in
prevention and treatment of cancer, still cancer burden is increasing globally (1). In past few
years, deacetylation of histone has been considered as a promising target for anticancer therapy.
Histone deacetylases (HDACs) are a family of enzymes which deacetylate the lysine residue of
core histone and other cellular proteins and subsequently results in chromatin condensation and
transcriptional repression (2-4). These epigenetic alterations by HDACs are involved in
suppressing the gene transcription and finally lead to the silencing of tumor suppressor genes.
Abnormal expression of various HDACs has been reported in numerous cancer cell lines and
tumor tissues (5). HDAC inhibitors promote tumor cell death by inducing apoptosis, cell cycle
arrest, senescence, differentiation, autophagy, etc. Therefore, HDAC inhibition has emerged out
as a potential anticancer target (6-9).
Till date, a total of 18 HDACs have been identified and these HDACs are grouped into four
distinct classes depending on their cellular localization, size and the number of active sites along
with structural homology to yeast (10). Class I HDACs consists of four members namely HDAC
1
, 2, 3 and 8 whereas class IV contains only one HDAC namely HDAC11. However, class II is
subdivided into two groups, i.e., class IIa and class IIb. HDACs 4, 5, 7 and 9 belong to class IIa
subfamily whereas HDAC6 and 10 belong to class IIb subfamily. Class III HDACs are basically
sirtuins (sirtuins 1-7) which are completely different from the other class HDACs. These class I,
+
II and IV HDACs are zinc-dependent while sirtuins are NAD dependent metalloenzymes (11,
1
2).
Since past few years, a number of HDAC inhibitors have been studied pre-clinically and
clinically for various cancers (Figure 1).
5

ACCEPTED MANUSCRIPT
Figure 1. Potential HDAC inhibitors [clinically approved and in clinical trials]
Four of them have been approved by USFDA for clinical use. Vorinostat (SAHA) has been
approved for the treatment of cutaneous T-cell lymphoma (CTCL), whereas romidepsin is
6

ACCEPTED MANUSCRIPT
approved for both cutaneous T-cell lymphoma (CTCL) as well as peripheral T-cell lymphoma
(PTCL). Again, belinostat is utilized for peripheral T-cell lymphoma and panobinostat is
approved for multiple myeloma. Chidamide has been approved for peripheral T-cell lymphoma
in China (13). Many other promising HDAC inhibitors such as entinostat (MS-275),
mocetinostat (MGCD-0103), tacedinaline (CI-994) are now in clinical trials for the treatment of
different types of cancers (14-18). Generally, HDAC inhibitors have a common pharmacophore
model (Figure 2) which includes three regions, i.e., one cap group or surface recognition
domain, a linker function and a zinc binding group (ZBG) (19).
Figure 2.General pharmacophore model for HDAC inhibitors
The first generation HDAC inhibitors (such as vorinostat, romidepsin, belinostat and
panobinostat) containing hydroxamate function as ZBG having pan HDAC inhibitory effects are
found to be effective anticancer agents (Figure 1).
Among different HDACs, histone deacetylase 3 (HDAC3) has been found to play pivotal roles in
the cancer cell proliferation, migration and apoptosis (20) along with DNA repair, genome
stabilization and maintaining structural integrity of chromatin as well as histone deacetylation
mechanisms during cell cycle progression and apoptosis (21-23). Overexpression of HDAC3 has
been attributed in numerous cancers such as breast cancer (24, 25), lung cancer (26), liver cancer
(27), colon cancer (28), leukemia (29), multiple myeloma (30), lymphoma (31), melanoma (32)
7

ACCEPTED MANUSCRIPT
and glioma (33). Therefore, inhibitors targeted to HDAC3 may be considered as a useful tool to
combat various malignancies.
In recent scenario to achieve increased selectivity and specificity, other ZBGs (such as
benzamides, thiols, sulphamides) have been taken into consideration (34). In general,
benzamides are found to inhibit class I HDACs (namely entinostat, mocetinostat and CI-994)
(35, 36). Modifications of the cap region having supremacy to target selective HDAC as it not
only binds to the enzyme surface but also forms complexes present nearby. These could help to
design the selective HDAC inhibitors (37, 38). Compounds possessing 2-aminobenzamide
moiety as zinc binding group (ZBG) has already been proven to execute selectivity towards
HDAC3 with promising anticancer properties (39-41). Depending on some crucial observations
obtained by our recent molecular modeling study (42) on selective HDAC3 inhibitors, newer
HDAC3 inhibitors are designed in the present work. These observations for designing newer
HDAC3 inhibitors are furnished below:
1
. The free amino group of benzamide moiety may serve as hydrogen bonding interaction.
Therefore, no substitution was made here.
2
. Hydrophobicity of the phenyl group is crucial for HDAC3 inhibition. Not only that, no
bulky steric and hydrophobic substitution at 5th position of the phenyl group of
benzamide moiety was considered for designing HDAC3 inhibitor as it was unfavorable
for HDAC3 inhibition.
3
. Smaller electron-withdrawing substituents (such as fluoro) at the para position of the
benzamide moiety was considered for designing newer HDAC3 inhibitors as it may favor
HDAC3 selectivity.
8

ACCEPTED MANUSCRIPT
4
. Carboxamide function attached to the benzamide scaffold was kept intact as the carbonyl
group is necessary for hydrogen bonding and amide group coordinated to the catalytic
2
+
Zn ion.
5
6
. Steric and hydrophobic aryl linker function was taken into consideration.
. Smaller aryl cap group was taken into account.
BG45, a benzamide-based class I HDAC inhibitor with preferential selectivity towards HDAC3
showed a significant inhibition of cellular proliferation in multiple myeloma in vitro. Moreover,
it was also found to be effective in vivo in arresting tumor growth as evidenced in the murine
xenograft model of multiple myeloma (43). In the present study, we have designed and
synthesized a series of novel benzamides by modifying the cap region of BG45 in order to
increase selectivity and efficacy towards HDAC3. These novel compounds were screened for
their potency and selectivity for HDAC3 enzyme in the in vitro enzyme inhibition assay and for
their anti-proliferative efficiency on different cancer cell lines.
2
. Results and Discussion
.1. Designing of benzamide-based HDAC3 inhibitors
2
Taking into account the density functional theory (DFT)-energy minimized electrostatic potential
maps of BG45 and CI-994 (both are prototype isoform-selective HDAC inhibitors), it is
observed that there is only a slight difference between them (Figure 3).
9

ACCEPTED MANUSCRIPT
Figure 3. Designing of the initial hit molecule compound 12a combining the structures of CI-
9
94 and BG45 [(A): Electrostatic potential map; (B): HOMO orbital electron density map; (C):
LUMO orbital electron density map].
Comparing activities of CI-994 and BG45, it is observed that both of these molecules exhibit
similar efficacy in HDAC3 inhibition. However, BG45 is smaller in size or shorter in length
compared to CI-994. Therefore, it may be inferred that as both of these molecules comprise the
carboxybenzamide moiety, the pyrazine scaffold obviously possesses a positive influence on
HDAC isoform selectivity. Comparing their electrostatic potential maps, it may be assumed that
due to the higher electron-rich region (yellow) imparted by both of the nitrogen atoms of the
pyrazine ring (Figure 3), BG45 produces more HDAC3 inhibition compared to CI-994.
Therefore, the phenyl ring of CI-994 has been replaced with that pyrazine scaffold in compound
1
0

ACCEPTED MANUSCRIPT
1
2a to judge whether the activity will increase or not. Interestingly, it displays both the features
of BG45 and CI-994 (Figure 4).
Figure 4. Designing of BG45-derived compounds
Nevertheless, the electrostatic potential map reveals that compound 12a consists of more
electronegative regions (yellow) compared to both BG45 and CI-994 (Figure 3). Moreover, the
HOMO-LUMO electron density maps of both BG45 and compound 12a suggest that both of the
HOMO and LUMO orbitals are located terminally and the LUMO orbitals of both these
compounds comprise the overlapping feature of both the nucleophilic (blue) and electrophilic
(red) regions at the pyrazine ring system (Figure 3). It clearly implies the ability of the pyrazine
ring to involve charge transfer mechanisms or π-π interaction. Interestingly, it has been found
1
1

ACCEPTED MANUSCRIPT
that compound 12a results in better efficacy regarding both HDAC3 inhibition as well as cellular
toxicity which strongly implicated the importance of pyrazine scaffold in the molecular structure
for exerting better efficacy.
Considering compound 12a as an initial hit molecule, newer molecules have been designed and
synthesized by substitution at the para position of the benzamide moiety with small electron-
withdrawing substituents (namely fluorine) as well as by substitution with alkyl carboxamide /
aryl carboxamide/ phenyl/ aryl amine functions at the pyrazine scaffold (Figure 4).
1
2
As CI-994 contains -NHCOCH
3
group, initially it was tried in both these positions (R and R ).
Then, the effect of NHCOPh was taken into consideration whether bulkier substitution was better
1
2
or not. Finally, phenyl, aminophenyl and aminobenzyl substitutions were tried at these R and R
positions to judge or compare their efficacy with respect to the earlier ones. As there is no
compounds of such type reported earlier combining scaffolds of these two prototype compounds
(CI-994 and BG-45), these groups (alkyl carboxamide / aryl carboxamide/ phenyl/ aryl amine
1
2
functions) were chosen particularly at these R and R positions to get an initial idea that which
type of substituents was preferable at these positions to synthesize newer better active molecules.
2
.2.
Chemistry
Scheme 1 outlines the synthesis of an intermediate; tert-butyl (2-aminophenyl) carbamate (2).
Here o-phenylenediamine (1) was protected at one amino group using di-tert-butyl dicarbonate.
Scheme 1. Reagent and conditions: (a) di-tert-butyl dicarbonate, triethylamine, ethanol, RT, 5h.
1
2

ACCEPTED MANUSCRIPT
Scheme 2 represents the synthesis of an intermediate tert-butyl (2-amino-5-fluorophenyl)
carbamate (6). The reaction started with boc protection of 5-fluro-2-nitroaniline (3) which
yielded the compound (4) with diprotection at the amino group. This compound (4) was
selectively deprotected with controlled condition using trifluoroacetic acid gave monoprotected
amino group (5) (44). Compound 5 upon catalytic hydrogenation with Pd/C produced compound
6
which was further utilized as an intermediate for synthesis of the final compounds.
Scheme 2. Reagent and conditions: (a) di-tert-butyl-dicarbonate, DMAP, 2-methyl THF, RT,
2h (b) trifluoroacetic acid(3%), DCM, RT, 2 h (c) Pd/C, H balloon, Methanol, RT, 4h
1
2
Scheme 3 illustrates the synthetic route of final molecules (compound 12a, 12b, 12c and12d).
Here, methyl 5-(chloropyrazine)-2-carboxylate (7) was first converted to methyl 5-
(aminopyrazine)-2-carboxylate (8) using sodium azide and triphenylphosphene (45). Here,
substituted chloropyrazine got converted to substituted azidopyrazine using sodium azide which
also isomerizes to its tetrazolo form in situ. The azidopyrazine forms iminophosphorane in the
presence of triphenylphosphene which gets hydrolyzed to amine under acidic condition, a typical
Staudinger reaction (46). N-acetylation of methyl 5-(aminopyrazine)-2-carboxylate (8) yielded
compound 9a while N-benzoylation produced compound 9b. These intermediates were
hydrolyzed to their acid derivatives compound 10a and compound 10b, and were coupled with 2
and 6 to get 11a, 11b, 11c and 11d. Deprotection of the compounds afforded final compounds
1
2a, 12b, 12c, 12d.
1
3

ACCEPTED MANUSCRIPT
Scheme 3. Reagent and conditions: (a) i. sodium azide, triphenylphosphene, dimethyl
o
o
o
sulphoxide, 120 C, 4h ii. 1N HCl, 120 C, 2h iii. NaHCO
3
(b) i. acetic anhydride, 90 C, 2h ii.
benzoyl chloride, sodium hydride, DMF, 0°C, 5 h (c) lithium hydroxide, THF:water (3:1), RT,
3
h (d) compd. 2 or compd. 6, EDC, DMAP, DCM:pyridine (1:1), RT, 12h (e) 4M HCl in
o
dioxane, DCM, 0 C, 2h.
1
4

ACCEPTED MANUSCRIPT
Scheme 4 discusses the synthesis of target molecules18a and 18b which mimics scheme 3 in
mechanisms with difference of methyl 6-(chloropyrazine)-2-carboxylate (13) instead of methyl
5
-(chloropyrazine)-2-carboxylate (7).
Scheme 4. Reagent and conditions: (a) i. sodium azide, triphenylphosphene, dimethyl
o
o
o
sulphoxide, 120 C, 4h ii. 1N HCl, 120 C, 2h, iii. NaHCO (b) acetic anhydride, 90 C, 2h (c)
3
lithium hydroxide, THF:water (3:1), RT, 3h (d) compounds 2 or 6, EDC, DMAP, DCM:pyridine
o
(
1:1),RT, 12h (e) 4M HCl in dioxane, DCM, 0 C, 2h
Scheme 5 shows the synthesis of target molecules 22a and 22b. The first reaction describes
Suzuki coupling where methyl 5-(chloropyrazine)-2-carboxylate (7) was reacted with phenyl
1
5

ACCEPTED MANUSCRIPT
boronic acid in presence of bis(triphenylphosphine)palladium(II)dichloride as a catalyst and
potassium carbonate using 1,4-dioxane solvent to get compound 19. This compound (19) on
alkaline hydrolysis formed its acid derivative 20. This was coupled with intermediates 2 and 6 by
EDC catalyzed acid-amine coupling to get 21a and 21b which upon boc deprotection by acid
made final products 22a and 22b respectively.
Scheme 5. Reagent and conditions: (a) phenyl boronic acid, bis (triphenylphosphine)
o
palladium(II) dichloride, potassium carbonate, 1,4-dioxane, 100 C, 12h (b) lithium hydroxide,
THF: Water (3:1), RT, 3h (c) 2 or 6, EDC, DMAP, DCM:pyridine (1:1), RT, 12h (d) 4M HCl in
o
dioxane, DCM, 0 C, 2h.
Scheme 6 depicts synthesis of target molecules 26a, 26b, 26c and 26d. Methyl 5-(chloro
pyrazine)-2-carboxylate (7) was reacted with aniline or benzylamine in presence of p-toluene
sulphonic acid and respective amines (compound 23a, 23b) were prepared. These compounds on
1
6

ACCEPTED MANUSCRIPT
alkaline hydrolysis formed respective acids (compound 24a, 24b). These were coupled with
intermediate compounds 2 and 6 by EDC catalyzed acid-amine coupling to get compounds 25a,
2
5b, 25c, 25d which upon deprotection by acid made final products 26a, 26b, 26c and 26d.
Scheme 6. Reagent and conditions (a) aniline (n=0) or benzylamine (n=1), p-toluenesulphonic
o
acid, 1,4-dioxane, 100 C,12h (b) lithium hydroxide, THF:water (3:1), RT, 3h (c) compound 2 or
o
compound 6, EDC, DMAP, DCM:pyridine (1:1), RT, 12h (d) 4M HCl in dioxane, DCM, 0 C,
2
h.
1
13
Purity of these final compounds was evaluated using H NMR, C NMR and LC-MS analysis.
Spectral data of these final compounds are provided in the supplementary materials (Text S1).
2
.3.
Pan HDAC inhibition assay using HeLa nuclear extract
The target compounds were screened for their HDAC inhibitory activity using color de lys®
HDAC assay kit (BML-AK501, ENZO life sciences). CI-994 and BG45 are used as positive
1
7

ACCEPTED MANUSCRIPT
controls for the assay. HeLa nuclear extract (expresses HDAC1 and HDAC2 dominantly) (47)
was used as a source of HDAC activity. The results are summarized in Table 1.
Table 1. HDAC inhibitory data of target compounds
%
inhibition in
% inhibition in
HDAC3 assay
(at 1 µM)
43.54
Cpda
R1
R2
R3
HeLa nuclear
extract (at 5 µM)
59.6
1
1
1
1
1
1
2
2
2
2
2
2
2a
2b
2c
2d
8a
8b
2a
2b
6a
6b
6c
6d
H
H
H
H
CONHCH
CONHCH
NHCOPh
NHCOPh
H
3
H
F
H
F
H
F
H
F
H
F
H
F
--
H
3
64.1
57.1
57.6
48.6
39.04
44.70
28.60
29.27
38.70
32.11
22.24
40.24
67.48
62.24
59.04
53.65
CONHCH
CONHCH
3
3
H
Ph
Ph
50.4
H
H
H
H
H
H
--
H
35.47
29.27
63.9
71.5
65.6
65.0
53.8
55.4
NHPh
NHPh
NHBnz
NHBnz
--
CI-994
BG45
H
57.97
a
Compound number
All compounds were found to exhibit HDAC enzyme inhibition at 5µM in duplicate. Compounds
6b, 26c and 26d demonstrated an increased HDAC inhibition compared to CI-994 and BG45.
These compounds were screened for determination of IC50 of enzyme inhibition along with CI-
2
9
94 and BG45. Compound 26b, 26c and 26d (IC50= 1.972±2.18 µM, 2.869±1.36 µM and
3
.072±1.56 µM, respectively) showed more promising HDAC inhibitory efficacy compared to
CI-994 (IC50= 5.369±0.78 µM) and BG-45 (IC50= 5.506±1.21 µM) (Table 2) and (Figure-5 A
and B).
1
8

ACCEPTED MANUSCRIPT
Table 2. IC50 values of promising compounds and their selectivity profile
IC50 (µM)
HeLa nuclear extract
Pan HDAC inhibition)
Selectivity for HDAC3
(Pan HDAC/HDAC3)
Cpda
HDAC3
(
2
2
2
6b
6c
6d
1.972
2.869
3.072
5.369
5.506
0.335
0.245
0.540
0.917
0.558
5.87
11.68
5.68
5.85
9.87
CI-994
BG45
Compound number
a
Figure 5. IC50 determination of HDAC enzyme inhibition using HeLa nuclear extract for
compound 26b, 26c and 26d [A] and BG-45 and CI-994 [B]. The data represents mean ± SD
(n=2).
1
9

ACCEPTED MANUSCRIPT
2
.4.
HDAC3 inhibition assay using recombinant HDAC3/NCOR1
Target compounds were evaluated for their inhibition potency towards HDAC3 enzyme using
HDAC3/NCOR1 fluorometric drug discovery kit (BML-AK531, ENZO life sciences). All these
compounds showed more inhibition of HDAC3 activity compared to nuclear extract (Table 1).
For better understanding, the most potent HDAC3 inhibitor compounds 26b, 26c and 26d
(67.48%, 62.24% and 59.04% respectively with 1µM compound concentration in duplicate) were
further evaluated for their IC50 values along with CI-994 and BG45 as positive control (Table 2,
Figure 6 A and B). Compound 26b, 26c, 26d, BG-45 and CI-994 were evaluated for IC50
determination. It was observed that the selected compounds were more effective than CI-994 and
BG-45. Compound 26c appeared to be as a lead in the series having the highest HDAC3
inhibition (IC50 = 0.245 µM) with ~12 fold HDAC3 selectivity over Pan HDACs.
2
0

ACCEPTED MANUSCRIPT
Figure 6. IC50 determination of HDAC3 enzyme inhibition using recombinant HDAC3/NCOR1
complex for compound 26b, 26c and 26d and BG-45 [A] and CI-994 [B]. The data represents
mean ± SD (n=2).
Some interesting relationship between the structures of the synthesized molecules and their
biological activity came out from the study. Incorporating fluoro substitution at the para position
of the benzamide moiety at compound 12a resulted in lower Hela nuclear extract inhibition than
compound 12b. However, compound 12a showed slightly lower HDAC3 inhibition (39.04%)
compared to compound 12a (43.54%). Again, the corresponding methyl carboxamide
1
2
substitution at R position instead of R position led to a decreased Hela nuclear extract
2
inhibition that strongly suggested that R substituents (such as CONHCH
3
) were not preferable
1
than the corresponding R substituents (compound 12a vs compound 18a; compound 12b vs
2
1

ACCEPTED MANUSCRIPT
compound 18b) as far as the HeLa nuclear extract inhibition was concerned. However,
2
CONHCH
3
substitution at R position yielded better HDAC3 inhibitors (compound 12a vs
compound 18a; compound 12b vs compound 18b). Hence, further structural modification was
2
performed at R position to judge the activity trend. Incorporation of phenyl group at the terminal
end of amino carbonyl moiety did not improve HeLa nuclear extract inhibition rather showed
decrease activity (compound 12a vs compound 12c; compound 12b vs compound 12d).
However, these compounds also displayed lower (compound 12b vs compound 12d) or
comparable (compound 12a vs compound 12c) HDAC3 inhibitory activity. Therefore, the
carbonyl function has been withdrawn from the next design. Incorporation of bulky phenyl ring
2
at R decreased both the HeLa nuclear extract inhibitory and HDAC3 inhibitory properties
(compound 22a vs compound 12a). The structure was further modified with the introduction of
2
aryl amine function at R position which yielded compounds with comparable HDAC3 inhibitory
efficacy (compound 26a vs compound 12a). However, both higher HeLa nuclear inhibitory and
HDAC3 inhibitory activities were observed for the corresponding p-fluoro analogs (compound
2
6b vs compound 12b). Additionally, incorporation of a methylene spacer between the aryl
(phenyl) and amine nitrogen resulted in both comparable HeLa nuclear inhibition and HDAC3
inhibition (compound 26c and 26d). Moreover, it was noticed that for both HeLa nuclear
inhibitory activity and HDAC3 inhibitory activity, arylamine groups (NHPh and NHBnz) at R2
position were better effective than the corresponding phenyl groups (compounds 26a-26d vs
compounds 22a-22b). Therefore, it may be assumed that not only the steric or hydrophobic
2
effect at R position (phenyl or benzyl) is necessary but also the presence of associated amino
function is crucial for imparting higher inhibitory effects. The overall approach of derivatization
of BG-45 and CI-994 resulted in the identification of compound 26c with 11.68-fold HDAC3
2
2

ACCEPTED MANUSCRIPT
selectivity over pan-HDAC (BG-45; 9.9-fold and CI-994; 5.85-fold). In a nutshell, it may be
2
inferred that aminobenzyl or aminophenyl group at R position is better effective than the
phenyl, methylcarboxamido (NHCOMe) or methylaminocarbonyl (CONHMe) functions at the
same position.
2
.5.
Antiproliferative assay using cancer cell lines
Cytotoxicity studies were performed on two different cell lines, i.e., B16F10 (murine melanoma
cell line) and HeLa (Human cervical cancer cell line) by MTT assay method. All the synthesized
compounds were evaluated at two different concentrations (100 µM and 10 µM) in triplicate. CI-
9
94 and BG45 were taken as the standard reference compounds in the assay (Supplementary
Figure S1). From the assay results, it was observed that majority of the compounds displayed
good cytotoxicity on both the cell lines. Compounds 12a, 22a, 26a, 26c and 26d were found to
be more promising and these were further subjected to evaluate for their 50% inhibitory
concentration (IC50) with a wider of concentrations. The cell viability after 72 hours of treatment
was measured by MTT assay (Figure 7).
2
3

ACCEPTED MANUSCRIPT
Figure 7. Dose response curve and IC50 values of promising compounds 12a, 22a, 26a, 26c and
2
6d along with BG-45 and CI-994. All compounds were explored in larger range of nine
different concentrations on B16F10 (A) and HeLa (B) cells. Cells were treated with compounds
for 72 hours and cell viability was measured by in vitro MTT assay method. Data represents
mean ± SD (n=2) and plotted in dose response format. IC50 was calculated using nonlinear
regression analysis method using Graph Pad Prism5.
Most of the selected compounds exhibited better cytotoxicity compared to CI-994 and BG45.
Interestingly compound 12a containing the molecular structural information of both BG45 and
2
4

ACCEPTED MANUSCRIPT
CI-994 exhibited similar efficacy compared to CI-994 but more efficacious than BG45 in both
the B16F10 and HeLa cell lines (Figure 7). Compound 26c (IC50 of 5.33 µM in B16F10 cells
and 3.99 µM in HeLa cells) and compound 26d (IC50 of 6.00 µM in B16F10 cells and 5.80 µM
in HeLa cells) were found to be the most potent among these compounds and therefore, these
compounds (compounds 26c and 26d) were further carried over for other biological assays.
2
.6.
Apoptosis assay
The extent of apoptosis induced by novel molecules (compounds 26c and 26d) was determined
by using Annexin V FITC/PI dead cell detection kit using B16F10 cells. HDAC inhibitors
induce apoptosis which results in cell death by altering gene expression (48). The percentage
apoptosis induced by target compounds 26c and 26d was investigated in the experiment along
with CI-994 and BG45 (Figure 8).
2
5

ACCEPTED MANUSCRIPT
Figure 8. Induction of apoptosis in B16F10 cells quantified by Annexin V/PI assay using flow
cytometry. (A) Control (B) CI-994 (C) BG45 (D) 26c (E) 26d. [Q2 quadrant depicts late
apoptosis and Q4 quadrant shows early apoptosis. X-axis: Annexin-V intensity, Y-axis:
propidium iodide intensity.
Results displayed (Figure 8) that 10.7±1.07% and 12.8±1.07% cells were present in late
apoptotic phase (Q2) for BG45 and CI-994 respectively. For compound 26d, 26.1±1.94% cells
were present in Q2 quadrant demonstrated more effective while the highest apoptosis was
exhibited by compound 26c showing 34.7±1.26% cells in the late apoptotic phase [Q2]. The
experiment proved an enhanced efficacy of the novel synthesized lead derivatives compared to
CI-994 and BG45.
2
.7.
We further investigated the effect of 26c and 26d on cell cycle progression in B16F10 cells
Figure 9). From the result, it was found that compound 26c and 26d induced growth arrest at
Cell cycle analysis
(
G2/M phase of cell growth. Compound 26c induced significant growth arrest around 43.7% in
G2/M phase compared to control showing 9.32%, while G0/G1 phase decreased to 11.6% from
7
4.6%. The similar pattern of growth arrest was found for 26d and BG-45 also. While CI-994
induced cell arrest in G0/G1 phase of cell cycle was 82.3%. These results indicated that
compound 26c might produce anticancer activity with the growth arrests at G2/M phase.
2
6

ACCEPTED MANUSCRIPT
2
7

ACCEPTED MANUSCRIPT
Figure 9. Cell cycle arrest induced in B16F10 cells by (A) control, (B) CI-994, (C) BG-45, (D)
2
6c, (E) 26d.
3
. Conclusion
Here, we have designed and synthesized a series of 2-aminobenzamide derivatives as selective
HDAC3 inhibitors. All target compounds were evaluated for their HDAC enzyme inhibition
efficiency with HeLa nuclear extract and recombinant HDAC3 enzyme and these compounds
were found to be effective HDAC inhibitors with preferential selectivity for HDAC3. These
compounds were also tested for their antiproliferative activity in two different cell lines (B16F10
and HeLa). Compound 26c exhibited potent cell growth inhibition compared to both CI-994 and
BG45. One of the lead compound (compound 26c) showed appreciable selectivity (11.68-fold)
in HDAC3 inhibition over pan HDACs which is quite significant than CI-994 (5.85-fold) and
BG-45 (9.9-fold). Compound 26c induced significant cell growth arrest in G2/M phase and
increased apoptosis level indicating improved anticancer potential compared to CI-994 and
BG45. Overall effort to make potent and selective HDAC inhibitors with different structural
medication on the overlapping backbone of BG45 and CI-994 resulted in compound 26c as a
potential HDAC3 inhibitor. Additionally, all these biological evaluation studies indicated
compound 26c can be further explored as a promising compound for anticancer therapy.
4
. Experimental
.1. Chemistry
4
All starting materials and reagents were commercially available and used without further
purification. All reactions were monitored by thin layer chromatography (TLC) using precoated
2
8

ACCEPTED MANUSCRIPT
1
13
plates with silica gel F254 from Merck Millipore Co., USA. H and C NMR spectrum were
recorded in DMSO-d and CDCl using Bruker-400 MHz and chemical shifts reported in ppm
6
3
using tetramethylsilane as internal standard. Mass spectroscopy was performed in LCMS-2020,
Schimadzu using ESI mode. The Purity of the compounds were analyzed by Analytical LC/MS.
Analysis was performed on a LCMS-8040 equipped with a photodiode array detector using a
Shiseido C18 4.6×150 mm, 5µ column at a flow rate of 1 mL/min with isocratic flow (50% A:
5
3
0% B for compound 12a, 12b, 18a and 18b while for rest of the compounds it was 65%A:
5%B; Solvent A = water, solvent B = methanol).
4
.1.1. Preparation of tert-butyl (2-aminophenyl)carbamate (2)
o-Phenylenediamine (5g, 0.0462mmol) was dissolved in 20 mL ethanol and triethylamine
7.01g, 0.0693mmol) was added. The mixture was stirred in ice bath for 20 minutes. Di-tert-
(
butyl dicarbonate (10.09g, 0.04623mmol) was mixed with 10 mL of ethanol and added slowly to
stirring reaction mixture and stirred for 5 hours at ambient temperature. The solvent was
evaporated and the residue was fractioned between ethyl acetate and water. The ethyl acetate
layer was evaporated and crude mixture was purified using column chromatography
1
(
EtOAc:Hexane;1:9) to give 6.5 g (yield 67.7 %) of 2 as white solid. H NMR (400 MHz,
DMSO-d
6
) δ 8.29 (s, 1H), 7.17 (d, J = 7.6 Hz, 1H), 6.83 (t, J = 7.6 Hz, 1H), 6.67 (dd, J= 7.9, 1.3
[M]: 208.26; MS (ESI)
Hz, 1H), 6.54 – 6.49 (m, 1H), 4.82 (s, 2H), 1.45 (s, 10H). C11
H
16
N
2
O
2
+
+
m/z: [M+H] : 209.10 [M-H] : 207.10
.1.2. Di boc protection of 5-fluoro 2-nitro aniline as compound 4
-fluoro-2-nitroaniline (3) (4g, 0.0256 mmol) was dissolved in 2-methyl THF (40 mL) and 4-
dimethyl amino pyridine (62mg, 0.0005mmol) was added. The reaction mixture was cooled at
°C, di-tert-butyl dicarbonate (11.18g, 0.05124 mmol) was added drop wise and stirred at room
4
5
0
2
9