Discovery of a novel HDAC3 selective inhibitor and its evaluation in lymphoma model

Article abstract of DOI:10.1002/bkcs.10619

Histone deacetylase (HDAC) inhibition is a potentially attractive approach to cancer therapy. A number of HDAC inhibitors are in clinical development stages for the treatment of cancer as well as immune and inflammatory disorders. Although there are several approved HDAC inhibitors by the US FDA, they show a broad inhibitory spectrum against HDAC subfamily. Herein, we synthesized a series of novel hydroxamate analogs, and evaluated them with lymphoma cancer cell. Conclusively, we identified an HDAC3 selective inhibitor which shows good anticancer activity for the lymphoma model, as well as a good drug metabolism and pharmacokinetics (DMPK) profile.

Full text of DOI:10.1002/bkcs.10619

Article
DOI: 10.1002/bkcs.10619
BULLETIN OF THE
C.-J. Choi et al.
KOREAN CHEMICAL SOCIETY
Discovery of a Novel HDAC3 Selective Inhibitor and its
Evaluation in Lymphoma Model
Chang-Ju Choi,†,‡ Mira Kim,†,‡ Sun Young Han,† Jiyoung Jeon,† Jae Ho Lee,† Jeong-In Oh,†
†,
Kwee Hyun Suh,† Dong-Churl Suh,‡, and Kwang-Ok Lee
*
*
†Department of Drug Discovery, Hanmi Research Center, Hwaseong-si 445-813, Korea.
*E-mail: medichem89@gmail.com
‡College of Pharmacy, Chung-Ang University, Seoul 156-756, Korea. *E-mail: dongsuh75@gmail.com
Received May 26, 2015, Accepted September 22, 2015, Published online December 17, 2015
Histone deacetylase (HDAC) inhibition is a potentially attractive approach to cancer therapy. A number of
HDAC inhibitors are in clinical development stages for the treatment of cancer as well as immune and inflam-
matory disorders. Although there are several approved HDAC inhibitors by the US FDA, they show a broad
inhibitory spectrum against HDAC subfamily. Herein, we synthesized a series of novel hydroxamate analogs,
and evaluated them with lymphoma cancer cell. Conclusively, we identified an HDAC3 selective inhibitor
which shows good anticancer activity for the lymphoma model, as well as a good drug metabolism and phar-
macokinetics (DMPK) profile.
Keywords: HDAC3 selective inhibitor, Lymphoma model
Human histone deacetylases (HDACs) play an important role
in regulating the expression and activity of numerous proteins
involved in both cancer initiation and cancer progression.1
More recently, HDACs inhibitors could have utility in the
treatment of immune and inflammatory disorders, including
rheumatoid arthritis, psoriasis, inflammatory bowel disease
(IBD), multiple sclerosis, and systemic lupus erythematosus
(SLE).2 To date, 18 HDAC isoforms have been identified in
humans, and they are subdivided into four classes (class I,
II, III, and IV).3
A number of small molecular HDAC inhibitors are in dif-
ferent stages of clinical development for the treatment of can-
cer as well as inflammatory disease. Currently, the US FDA
has approved three HDAC inhibitors for anticancer drugs
(Figure 1). Suberoylanilide hydroxamic acid (SAHA, vorino-
stat) 1 and romidepsin 2 were approved by the FDA for the
treatment of cutaneous T-cell lymphoma (CTCL).4,5
Recently, another HDAC inhibitor, belinostat 3 has been
approved for the treatment of peripheral T-cell lymphoma
(PTCL).6 These approved HDAC inhibitors showed a broad
inhibitory spectrum against HDAC subfamily, and their non-
selective activity for HDAC subfamily can cause an adverse
drugreactions. DespitetheprovenanticancereffectsofHDAC
inhibitors against certain cancers, many aspects of HDAC
enzymes are still not fully understood. Therefore, the discov-
ery of isoform-selective HDAC inhibitors will ultimately be
important for enhancing the safety profile, as well as under-
standing of the biological activities of each isoform HDAC
family.
regulationofoncoproteins inleukemiaandB-celllymphoma.7
There are many types of targeted therapies in clinical trials for
B-cell lymphoma patients, and two kinase inhibitors that
include a BTK inhibitor, ibrutinib, and a PI3K delta inhibitor,
idelalisib have recently been approved for the treatment of
chronic lymphocyte leukemia (CLL).8 However, anticancer
spectrum of above drugs have limited to partial population
of all lymphoma patients.9 Therefore, a selective HDAC3
inhibitor could be a potential candidate for use with blood can-
cer therapy, which currently is critically lacking in treatment
options. This paper will introduce the synthesis and optimiza-
tion process of a novel HDAC3 selective inhibitor and its bio-
logical evaluation in lymphoma model, as well as its drug
metabolism and pharmacokinetics (DMPK) study.
Design and Synthesis
From our in-house screening for the finding of active
compound in DOHH-2 lymphoma cell, we confirmed a hit
compound 4, showing GI50 as 449 nM (Figure 2). For enhan-
cing the in vitro cellular activity, we planned an optimization
approach to compound 4 as shown in Figure 2.
First, optimization of part B was done by introducing a
linker between phenyl moiety and hydroxamic acid of com-
pound 4 as shown in Scheme 1. Starting material 5 was
obtained by the reaction of commercially available cyanuric
acid and morpholine, in the presence of triethylamine in ace-
tone (not shown in Scheme 1). Carbon–carbon coupling reac-
tion using a palladium catalyst and 4-aminophenyl boronic
acidprovidedaminophenyltriazine 6, whichwascoupledwith
long chain acids to give intermediate 7. Finally, terminal
Recent research indicated, HDAC3 is one of the four mem-
bersofthehumanclassIHDAC, anditisalsooverexpressedin
various lymphoma cells, as well as play an integral role in
methyl ester of intermediate
7
was transformed to
Bull. Korean Chem. Soc. 2016, Vol. 37, 42–47
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Wiley Online Library
42

Article
BULLETIN OF THE
ISSN (Print) 0253-2964 | (Online) 1229-5949
KOREAN CHEMICAL SOCIETY
hydroxamate 8 (n = 5), 9 (n = 6), and 10 (n = 7) by the reac-
tion with aqueous hydroxylamine. Also, alkylation of amino-
phenyl triazine 6 with alkyl bromide afforded methyl ester 11,
which was sequentially converted to hydroxamate 12 as same
method of previous hydroxamates. Another carbon–carbon
coupling of starting material 5 and 4-hyroxyphenyl boronic
acid gave the 4-hydroxy phenyl triazine 13, which was
alkylated to provide methyl ester 14, and was transformed
to hydroxamate 15.
Lastly, modification of part A was achieved by changing
phenyl ring into imidazole or pyrazole that is shown in
Scheme 2. For the pyrozole ring, alkylation of pyrazole boro-
nate 16 gave boronate 17 then, a palladium mediated coupling
with starting material 5 which provided ester 18. Finally, the
ester group was converted into hydroxamate using the same
method as in Scheme 1 to afford the objective compound 19
Figure 2. Optimizationof compound 4toenhance the invitro activity
Figure 1. HDAC inhibitors approved by the US FDA.
in lymphoma cell.
ꢀ
Scheme 1. Reagents and conditions: (a) 4-aminophenyl boronic acid, Pd(PPh3)4, 2 M Na2CO3 (aq.), 1,4-dioxane/H2O, 90 C; (b) monomethyl
pimelate (n = 5) ꢀor monomethyl suberate (n = 6) or monomethyl azelate (n = 7), EDCI, HOBt, DIPEA, DMF, rt; (c) 50% aq. NH2OH, KCN,
ꢀ
THF/MeOH, 60 C; (d) ethyl 7-bromoheptanoate, NaH, DMF, 0 C to rt; (e) 4-hydroxyphenyl boronic acid, Pd(PPh3)4, 2 M Na2CO3 (aq.), 1,4-
ꢀ
ꢀ
dioxane/H2O, 90 C; (f ) ethyl 7-bromoheptanoate, K2CO3, DMF, 90 C.
Bull. Korean Chem. Soc. 2016, Vol. 37, 42–47
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de
43

BULLETIN OF THE
Article
Discovery of an HDAC3 Selective Inhibitor and its Evaluation
KOREAN CHEMICAL SOCIETY
ꢀ
Scheme 2. Reagents and conditions:ꢀ(a) ethyl 4-bromobutyrate (n = 3) or ethyl 5-bromovalerate (n = 4), NaH, DMF, 0 C to rt; (b) Pd(PPh3)4,
ꢀ
2 M Na2CO3 (aq.), 1,4-dioxane, 80 C; (c) 50% aq. NH2OH, KCN, THF/MeOH, 60 C; (d) methyl 4-imidazole carboxylate, K2CO3, CH3CN,
ꢀ
60 C; (e) LiOH, THF/MeOH/H2O, rt; (f ) ethyl 4-aminobutyrate (n = 3) or ethyl 5-aminopentanoate (n = 4), EDCI, HOBt, DIPEA, DMF, rt.
(n = 3) and 20 (n = 4). Tointroduce an imidazole ring, methyl
4-imidazole carboxylate was coupled with the starting mate-
rial by general substitution reaction in the presence of
K2CO3. Then, hydrolysis using lithium hydroxide was done
to afford acid 21. Amidation with ethyl 4-aminobutyrate
(n = 3) or ethyl 5-aminopentanoate (n = 4) and following
reaction with hydroxyl amine afforded hydroxamate 22
and 23.
zinc binding group for HDAC inhibition profiles. Finally, a
pyrazole ring was introduced to part A which led to the discov-
ery of analog 20 which shows the most potent in vitro
activity.10
As shown in Table 2, further in vitro screening for analog 20
was performed in a variety of lymphoma cancer cells such as
follicular lymphoma, activated B-cell type (ABC) diffuse
large B-cell lymphoma (DLBCL) and germinal center B-cell
type (GCB) DLBCL.9 Reference compounds including ibru-
tinib as BTK inhibitor, and idelalisib as PI3K-delta inhibitor,
were selected from targeted agents in current clinical stages of
lymphoma. These reference compounds demonstrated a nar-
row spectrum of activity only against TMD-8 (CD79 mutant
ABC DLBCL cell line) and DOHH-2 (follicular lymphoma
cell line). On the contrary, the analog 20 exhibited a broad
spectrum for all lymphoma cancer cells with high in vitro
potency. For the identification of the role of hydroxamate in
the analog 20, we have synthesized a carboxylic acid substit-
uent to hydroxamate and evaluated in vitro activities for
DOHH-2 lymphoma cells (data are not shown in this paper).
Then we found that the introduction of a carboxylic acid sub-
stitution loses the anticancer activity.
Anticancer and DMPK Evaluation
In this study, nine newly synthesized analogs were screened
for the use of the follicular lymphoma cell lines, DOHH-2,
and the results summarized in Table 1. Most analogs showed
good in vitro activities except analog 22 and 23 which have an
imidazole group between triazine and hydroxamate. Among
analogs 8, 9, 10, 12, and 15 that have a benzene ring in the
region of part A as seen in Figure 2, analog 9, 12, and 15 exhib-
ited sub-micromolar cell growth inhibition activity. Although
further modification of the linker is needed for the explanation
of structure–activity relationships, it seemed that the length of
the carbon chain is an important factor to increase in vitro cel-
lular activity. Introducing imidazole-4-carboxamide to part A
(analogs 22 and 23) resulted in decreased potency or loss of
activity. The result of changed length of aliphatic chain (ana-
log 19) indicated the decreased potency. It seemed that the
length of the carbon chain is an important factor for the activ-
ity. Thelengthofaliphaticchainbetweenhydroxamicacidand
core is a critical variable in optimizing the positioning of the
Finding the mechanism of action of analog 20, we screened
eleven different HDAC enzymes which shown in Table 3. As
most of the HDAC inhibitors have a hydroxamate group as a
key interaction of moiety with zinc in the HDAC pocket, we
assumed that our analog bind directly to the HDAC enzymes.
As expected, analog 20 showed inhibitory activities for eleven
HDAC subfamilies. Notably, it showed a potent inhibition
Bull. Korean Chem. Soc. 2016, Vol. 37, 42–47
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de
44

BULLETIN OF THE
Article
Discovery of an HDAC3 Selective Inhibitor and its Evaluation
KOREAN CHEMICAL SOCIETY
Table 1. In vitro activities of triazine analogs in DOHH-2 follicular lymphoma cell.
No
A
B
DOHH-2a (GI50, μM)
8
2.15
9
0.18
10
12
1.76
0.85
15
0.30
19
20
22
23
a
2.44
0.020
>10
>10
The experiment was performed in duplicate trial.
Table 2. In vitro activities of analog 20 in lymphoma cell.
In vitro cell growth inhibitiona (GI50, nM)
ABC DLBCL
Follicular lymphoma
DOHH-2
GCB DLBCL
SU-DHL-4
Compound
TMD-8b
HBL-1b
U2932c
RC-K8d
Ibrutinibe
Idelalisibf
Analog 20
a
319
3298
20
1
245
3
>10 000
>10 000
30
9831
>10 000
148
>10 000
>10 000
94
1474
1623
62
The experiment was performed in duplicate trial.
CD79 mutant cell.
b
c
d
e
f
TAK1 mutant cell.
A20 mutant cell.
Ibrutinib: Bruton’s tyrosine kinase (BTK) inhibitor.
Idelalisib: PI3K delta kinase inhibitor.
Bull. Korean Chem. Soc. 2016, Vol. 37, 42–47
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de
45

BULLETIN OF THE
Article
Discovery of an HDAC3 Selective Inhibitor and its Evaluation
KOREAN CHEMICAL SOCIETY
activity for theHDAC3enzymewith atleast a10-fold increase
of selective activity compared with other HDAC enzymes.
We checked drug metabolism and pharmacokinetics
(DMPK) of analog 20 before the in vivo evaluation which
include CYP isozyme (3A4 and 2C9) inhibition, microsomal
stability, and plasma stability as shown in Table 4. Analog 20
showed no activities of two CYP isozymes such as 3A4 and
2C9. This result indicated good metabolic stability in human,
dog, and rat of liver microsomes and moderate stability in
mouse of liver microsome after 1-h incubation. Also, they
showed good plasma stability in human, dog, and rat of plas-
mas, and moderate stability in mouse of plasma after 6-h incu-
bation. As shown in Table 5, a pharmacokinetic study with
analog 20 in mice has an oral bioavailability of 19.7%. Also,
based on the results of DMPK studies, we concluded that ana-
log 20 is a viable candidate for in vivo evaluation as an oral
agent.
Finally, in vivo evaluation of analog 20 in TMD-8 lym-
phoma model11 was done. The in vivo efficacy of analog 20
demonstrated a tumor growth inhibition (TGI) rate as 81%
(Figure 3). During the period of treatment, any adverse reac-
tions were not observed which include body weight loss.
Conclusion
In this paper, we synthesized some novel hydroxamate ana-
logs and evaluated their anticancer activity in lymphoma can-
cer cells. Among the synthesized analogs, analog 20 showed
the most potent in vitro activities, as well as good in vivo effi-
cacyandDMPK profile. Biologicalassay ofanalog 20 showed
as an HDAC3 selective inhibitor. The establishment of
structure–activity relationships is currently undergoing for
further modifications. The results will be updated as the
research progressed.
Table 3. In vitro inhibition of HDAC subfamily of analog 20.
HDAC isoforms
IC50 (nM)a
1
2
87
133
3
7
4
5
6
>10 000
>10 000
155
7
8
9
10
11
a
>10 000
2441
>10 000
272
1600
Control
1400
Analogue 20, 30 mg/kg
70
1200
HDAC assay was performed at Reaction Biology (Malvern, PA, USA).
1000
800
600
400
200
0
Table 4. In vitro drug metabolism of analog 20.
CYP inhibition
for 3A4 and
2C9a (IC50,
μM)
Microsomal stability
for four speciesb
(mouse/rat/dog/
human)
Plasma stability for
four speciesc
(mouse/rat/dog/
human)
>20/>20
66/95/95/94
52/92/88/82
a
Data were analyzed by rhCYP-fluorescence method.
Remaining percentage of metabolism by incubation of the parent
molecule (5 μM) with liver microsomes of mouse, rat, dog, and human
for 60 min.
Remaining percentage was measured by incubating the plasma and
0.05 M potassium phosphate buffer (pH 7.4) with parent molecule
(0.5 μM, final 0.2% DMSO) for 6 h at 37 ꢀC (400 rpm, n = 3).
b
c
1
3
5
7
9
11 13 15 17 19
Days
Figure 3. In vivo xenograft study of analog 20 in TMD-8 model.
Table 5. Pharmacokinetics of analog 20 in mice.
Pharmacokinetics in mice (n = 3 per time point)a
iv (dose = 5 mg/kg)
po (dose = 30 mg/kg)
F (%)
C0 (ng/mL)
AUCINF (ng Á h/mL)
t1/2 (h)
2.0
Cmax (ng/mL)
AUCINF (ng Á h/mL)
t1/2 (h)
1.1
2287.5
363.4
339.7
431.3
19.7
a
Compound was dissolved in a vehicle as a solution of 0.3 M HCl in DMSO, Tween 80, PEG 400, and distilled water (1, 0.5, 1, and 7.5).
Pharmacokinetic parameters are calculated from plasma concentration-time data by a noncompartmental method using Phoenix™ WinNonlin® 6.1
(Pharsight, Princeton, NJ, USA).
Bull. Korean Chem. Soc. 2016, Vol. 37, 42–47
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de
46

BULLETIN OF THE
Article
Discovery of an HDAC3 Selective Inhibitor and its Evaluation
KOREAN CHEMICAL SOCIETY
7. M. Gupta, J. J. Han, M. Stenson, L. Wellik, T. E. Witzig, Leuke-
mia 2012, 26, 1356.
References
8. (a) T. Robak, Ann. Hematol. Oncol. 2014, 1(1), 1001;
(b) A. Markham, Drugs 2014, 74, 1701.
1. (a) M. A. Glozak, E. Seto, Oncogene 2007, 26, 5420;
(b) J. E. Bolden, M. J. Peart, R. W. Johnstone, Nat. Rev. Drug
Discov. 2006, 5, 769; (c) H.-J. Kim, S.-C. Bae, Am. J. Transl.
Res. 2011, 3, 166; (d) A. Mai, S. Massa, D. Rotili, I. Cerbara,
S. Valente, R. Pezzi, S. Simeoni, R. Raqno, Med. Res. Rev.
2005, 25, 261.
2. (a) K. J. Falkenberg, R. W. Johnstone, Nat. Rev. Drug Discov.
2014, 13, 673; (b) C. Balagué, S. L. Kunkel, N. Godessart, Drug
Discov. Today 2009, 14, 926; (c) S. J. Shuttleworth, S. G. Bailey,
P. A. Townsend, Curr. Drug Targets 2010, 11, 1430.
3. (a) I. V. Gregoretti, Y.-M. Lee, H. V. Goodson, J. Mol. Biol.
2004, 338, 17; (b) A. J. M. De Ruijter, A. H. Van Gennip,
H. N. Caron, S. Kemp, A. B. P. Van Kulenburg, Biochem. J.
2003, 370, 737.
9. (a) A. L. Shaffer 3rd. , R. M. Young, L. M. Staudt, Annu. Rev.
Immunol. 2012, 30, 565; (b) M. Roschewski, L. M. Staudt,
W. H. Wilson, Nat. Rev. Clin. Oncol. 2014, 11, 12.
10. Spectral data of analog 20: 1H-NMR (300 MHz, DMSO-d6)
δ 10.35 (s, 1H), 8.68 (s, 1H), 8.34 (s, 1H), 7.97 (s, 1H),
4.13 (t, J = 6.8 Hz, 2H), 3.75 (m, 8H), 3.63 (m, 8H), 1.96
(t, 2H), 1.74 (m, 2H), 1.43 (m, 2H). MS (ESI+): m/z = 433.2
[M+H]+.
11. TMD-8 tumor cells were implanted subcutaneously into female
SCID mice. Treatment of tested compounds was initiated
when the mean tumor volumes reached 220 mm3. Analog 20
were administered orally, once a day at doses of 30 mg/kg,
respectively. Tumor growth inhibition (TGI) % is defined as:
([TVvehichle Day x – TVvehichle Initial] – [TVtreatment Day x –
TVtreatment Initial]/[TVvehichle Day x – TVvehichle Initial]) ×
100%. TV means tumor volume.
4. P. A. Marks, R. Breslow, Nat. Biotechnol. 2007, 25, 84.
5. K. M. VanderMolen, W. McCulloch, C. J. Pearce,
N. H. Oberlies, J. Antibiot. 2011, 64, 525.
6. R. M. Poole, Drugs 2014, 74, 1543.
Bull. Korean Chem. Soc. 2016, Vol. 37, 42–47
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de
47