Design, synthesis and biological evaluation of novel histone deacetylase inhibitors incorporating 4-aminoquinazolinyl systems as capping groups

Article abstract of DOI:10.1016/j.bmcl.2017.09.036

A series of hydroxamic acid-based HDACIs with 4-aminoquinazolinyl moieties as capping groups was profiled. Most compounds showed more potent HDACs inhibition activity than clinically used drug SAHA. Among them, compounds 5f and 5h selectively inhibited HDAC 1,2 over HDAC8, and showed strong activity in several cellular assays, not possessing significant toxicity to primary human cells and hERG inhibition. Strikingly, 5f possessed acceptable pharmacokinetic characteristics and exhibited significant antitumor activity in an A549 xenograft model study at well tolerated doses.

Full text of DOI:10.1016/j.bmcl.2017.09.036

Accepted Manuscript
Design, synthesis and biological evaluation of novel histone deacetylase inhib-
itors incorporating 4-aminoquinazolinyl systems as capping groups
Qingwei Zhang, Yang Li, Baoyin Zhang, Bingliu Lu, Jianqi Li
PII:
S0960-894X(17)30932-0
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.09.036
BMCL 25300
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
8 July 2017
13 September 2017
15 September 2017
Please cite this article as: Zhang, Q., Li, Y., Zhang, B., Lu, B., Li, J., Design, synthesis and biological evaluation
of novel histone deacetylase inhibitors incorporating 4-aminoquinazolinyl systems as capping groups, Bioorganic
& Medicinal Chemistry Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.09.036
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.

Design, synthesis and biological evaluation of novel histone
deacetylase inhibitors incorporating 4-aminoquinazolinyl
systems as capping groups
Qingwei Zhang*, Yang Li, Baoyin Zhang, Bingliu Lu, Jianqi Li*
Novel Technology Center of Pharmaceutical Chemistry, Shanghai Institute of Pharmaceutical Industry, China
State Institute of Pharmaceutical Industry, Shanghai 201203, P.R. China
*Corresponding author: Tel.: +86-021-20572000-5061. E-mail: sipiqingwei@163.com; lijianqb@126.com
Abstract:
A series of hydroxamic acid-based HDACIs with 4-aminoquinazolinyl moieties
as capping groups was profiled. Most compounds showed more potent HDACs
inhibition activity than clinically used drug SAHA. Among them, compounds 5f and
5h selectively inhibited HDAC 1,2 over HDAC8, and showed strong activity in
several cellular assays, not possessing significant toxicity to primary human cells and
hERG inhibition. Strikingly, 5f possessed acceptable pharmacokinetic characteristics
and exhibited significant antitumor activity in an A549 xenograft model study at well
tolerated doses.
Keywords:
Histone deacetylases
4-aminoquinazolinyl
selectivity
antiproliferative activity
synthesis
Histone acetylation/deacetylation is essential for chromatin remodeling and
epigenetic regulation of gene transcription, which is controlled by histone
acetytransferases (HATs) and histone deacetylases (HDACs).1, 2 Overexpression of
HDACs has been reported in tumor cells, and evidence has shown that development
of HDACs inhibitors (HDACIs) is a promising anticancer strategy. 3-5
To date, a number of HDACIs are currently undergoing different stages of
development as anticancer agents and have been shown to regulate a variety of
cellular responses including proliferation, differentiation, and apoptosis.6-8 Among
them, suberoylanilide hydroxamic acid (also known as SAHA or Vorinostat) has

emerged as the first HDACIs approved by the FDA in 2006 for the treatment of
advanced cutaneous T-cell lymphoma (CTCL).9, 10 However, although very powerful,
many HDACIs occasionally have been associated with problems such as poor
metabolic stability or a rapid excretion.11,12And thus, the search for novel, potent
HDACIs with better pharmaceutical and safety properties is still a goal to pursue. 13
Most HDACIs share the basic pharmacophore model, as depicted by SAHA (Fig.
1), which is composed of a zinc binding group (ZBG), a linker and a cap group.9 Our
initial SAR focused on structural transformation on the capping group of SAHA to
find compound 1, a reversed amide analogue of SAHA, exhibited a weak HDACs
inhibitory with an IC50 value of over 2μM. In our further study it was determined that
the amide residue could be further functionalized with nitrogen atom-containing
heterocyclic groups. Such as the benzimidazole derivatives Ia-e led to a significant
enhancement in HDACs inhibitory activity.14 Similar to other groups’ findings,15-17
our previous results confirmed the hypothesis that introduction of nitrogen
atom-containing heterocyclic substituents on the capping groups of HDACIs could
provide opportunities for discovering potent and possibly selective HDACIs.
As part of our efforts to search for novel HDACIs with improved potency and
better safety profiles, herein we investigated a series of potent hydroxamic acid-based
HDACIs containing 4-aminoquinazolinyl moieties as capping groups.
Figure 1. Functional domains of HDACIs

Scheme 1. Synthesis of compounds 5a-5l. Reagents and conditions: (a) SOCl2, reflux, 8 h; (b) IPA, TEA,
90℃，6h; (c)CH3OH, KOH, NH2OH. HCl, rt，0.5-1 h
Scheme 2. Synthesis of compound 5m. Reagents and conditions:(a) 10% Pd/C, THF, rt, 6h
The general synthetic pathway to the quinazolinyl-containing hydroxamates is
shown in Scheme 1. Firstly, chlorination of the 4-quinazolone analogues 2a-2j with
thionyl chloride gave the corresponding 4-chloroquinazoline analogues 3a-3j.
Reaction of the 4-chloroquinazoline intermediates with methyl 6-aminocaproate or
methyl 7-aminoheptoate in the presence of triethylamine provided the corresponding
ester intermediates 4a-4l. The obtained esters were further treated with hydroxylamine
in the presence of KOH in aqueous methanol affording target hydroxamic acids 5a-5l
at rt for a short time of around 30 min to 1h. Atmospheric hydrogenation of compound
5l using catalytic amount of palladium on carbon gave compound 5m. The structures
of all the synthesized intermediates and target compounds were confirmed by
1H-NMR and ESI-MS.
Table 1 HDACs inhibitory activities of compounds 5a-5m

HDACs
IC50 a (nM)
Compd
5a
Ar
n
1
33.76±7.09
13.06±2.87
20.17±5.22
41.19±6.14
1
1
0
5b
5c
5d
1
55.17±9.10
5e
1
0
1
1
1
3.08±0.19
10.54±2.08
8.17±1.15
5f
5g
5h
5i
151.90±17.13
20.68±4.14
5j

1
1
1
24.17±6.32
22.89±9.08
5k
5l
12,14±5.20
5m
42.96±8.17
SAHA
a Data represent mean values of at least three independent experiments and standard deviations are
calculated based on the results of these experiments.
The inhibitory effects of compounds 5a-5m on HDACs activity were determined
by using a fluorescence-based assay as described before. 18The data are presented as
IC50 values in Table 1. It was found that all these compounds showed promising
activities with IC50 values in the nanomolar range. They were almost more potent than
SAHA, which exhibited an IC50 value of 42.96±8.17 nM against HDACs in the same
assay. The potency of compounds 5a (IC50=33.76±7.09 nM) and 5b (IC50=13.06±2.87
nM) are comparable to that of their five carbon atom chain analogues 5d
(IC50=41.19±6.14 nM) and 5g (IC50=10.54±2.08 nM), respectively. This is consistent
with the SAR reported previously by others that changing the spacer length from five
to six methylene units made little difference in HDACs inhibitory activities. 15,18
Herein we selected a chain length of six as a basis for further work. The results in the
present manuscript represented that introduction of different substituents on the
terminal quinazoline ring could influence HDACs inhibitory activity. Attaching the
electronic donating groups such as –CH3 or –NH2 group on the quinazoline ring (5c
and 5l, IC50 values were 20.17±5.22 nM and 22.89±9.08 nM, respectively) did not
enhance the biding affinity to HDACs, in comparison with introduction of the halogen
group at the same position (5f and 5h, IC50 values were 3.08±0.19 nM and 8.17±1.15
nM, respectively). Among the regioisomers derived from moving the halogen linkage
to the different positions of the quinazoline ring, two 6- position substituents 5f and

5h were the most potent compounds (IC50 values were 3.08±0.19 nM and 8.17±1.15
nM, respectively). Compound 5f, in particular, was over 10-fold more potent than
SAHA.
In order to investigate the interaction between target compounds and HDACs, we
selected compound 5f to dock into the active site of HDLP (histone deacetylase-like
protein, PDB entry: 1C3S) using Syble/FlexX module.19 As the result shown in Figure
3, the hydroxamic acid group of 5f could form a chelated structure with the zinc ion in
the centre of active site and the side chain could occupy the long and narrow tunnel
which linked the zinc and the outside world. The 2D pattern showed that the
4-aminoquinazolinyl moiety was oriented in the correct direction to enable favor
interactions with neighbor residues (Tyr91, Glu92 and Gly140) on the entrance of
narrow lipophilic pocket. The above interactions might contribute to the affinity of 5f
to HDACs.

Figure 3. ( A ) 3D model of the interaction between compound 5f and the active site of HDLP. (B)
2D molecular docking modeling of compound 5f into the binding site of HDLP is analyzed using
the ligplot4.22. The hydrogen bonds are displayed as green dotted lines.
Two of most potent target compounds, 5f and 5h, were selected for the HDACs
isoforms selective inhibition experiments. We conducted enzymatic inhibitory assays
against HDAC1 (Class I), HDAC8 (Class I) and HDAC6 (Class IIb). In contrast to
SAHA, two compounds exhibited more potent inhibitory activity toward HDAC1 and
HDAC6 while showing only micromolar activity against HDAC8 (IC50 values against
HDAC8 were 1.97±0.20 μM and 1.46±0.18 μM, respectively), which represented that
the quinazolinyl-based HDACIs could selectively inhibit HDAC 1,6 over the HDAC8
isoform. Next, we determined the antiproliferative activities of compounds 5f and 5h
against a panel of cancer cell lines using MTT assay, 20 which consisted of
MDA-MB-231, PC-3, HCT-116, K562, and A549 cell lines. Data from Table 2
demonstrated that two selected compounds exhibited broad antiproliferative activities
and outperformed SAHA in most of the cell lines we examined. Notably, compound
5f potently inhibited the human breast cancer MDA-MB-231 line with an IC50 =
0.17±0.03 μM, and had an IC50 of 0.080±0.009 μM in the non-small cell lung cancer
A549 line, a 10-fold improvement over SAHA (IC50=0.86±0.11 μM).
On the basis of the above results, we thought worthwhile to examine the

ancillary safety profiles of the representative derivatives as anticancer agent,
including growth inhibitory activities towards human normal cell lines test and hERG
automated patch clamp (QPatch) assay. As is apparent from the data in Table 2, all the
tested compounds including SAHA exhibited weak inhibitory activity against the
human fetal lung fibroblast (MRC-5) normal cell lines (IC50>50 μM) and were
inactive in our hERG binding assay (IC50>30 μM).
The pharmacokinetic study of 5f was carried out using female Sprague Dawley
rats dosed iv at 2 mg/kg and orally at 20mg/kg（Table 3）.The terminal phase half-life
of 5f after iv dosing was 2.87 h. In contrast,SAHA after iv administration
demonstrated a short half-life (t1/2 =12± 1 min). 22 5f was quickly absorbed after oral
dosing, with a Tmax of 0.25 h. The oral bioavailability (F) of 5f was 16.8%, which is
higher than SAHA (F =11%).22
Given these promising pharmacokinetic parameters, the most promising
compound 5f was further evaluated in the non-small cell lung cancer A549 xenograft
mousemodel. 5f displayed 53.12% and 76.39% tumor growth inhibition, respectively,
when intraperitoneal (ip) administration dosed once every other day at 20 and 40
mg/kg for a 21-day efficacy study, compared with vehicle (Fig. 3). No adverse effects
or signs of toxicity were observed.

Table 2. The HDACs isoform inhibitory activities, anti-proliferative activities and hERG binding assay of compounds 5f ,5h and SAHA.
IC50a (μM)
IC50 b(μM)
Compd
hERG
HDAC1
HDAC6
HDAC8
MDA-MB-231
PC-3
HCT-116
K562
A549
MRC-5
>30
>30
>30
5f
5h
0.011±0.003
0.007±0.001
1.97±0.20
0.17±0.03
0.76±0.11
2.11±0.09
1.21±0.18
0.080±0.009
54.9±5.1
0.013±0.002
0.008±0.002
1.46±0.18
0.49±0.07
1.02±0.24
3.45±0.33
0.86±0.21
0.147±0.03
76.5±10.4
SAHA
0.045±0.008
0.019±0.004
0.78±0.11
1.98±0.34
2.92±0.05
2.35±0.21
0.98±0.15
0.86±0.11
68.7±6.8
a Data represent mean values of at least three independent experiments and standard deviations are calculated based on the results of these
experiments.
b hERG Patch clamp screen as described in Ref. 21. IC50 values represent the concentration to inhibit 50% of hERG current (IKr). Numbers
represent IC50 values generated from 3-point concentration response relationships in duplicate.
Table 3. Pharmacokinetic evaluation of compound 5f a
T1/2
iv (h)
2.87
CL
iv (L/h/kg)
3.18
Tmax
po (h)
0.25
Cmax
po( ng/mL)
145.7
F
Compound
(%)
16.8
5f
a For pharmacokinetic study, blood was collected from rats at various time points up to 24 h, and plasma samples were analyzed using an Agilent
1200 HPLC system coupled with Agilent 6410 B triple quadruple mass spectrometer. A solution of 0.05 N HCl in saline was used as the vehicle
for both iv and po dosing.

Figure 3. Antitumor activity of 5f in mouse A549 xenograft model
In summary, a series of hydroxamic acid-based HDACIs with
4-aminoquinazolinyl moieties as capping groups have been designed and synthesized.
Compared with SAHA, these inhibitors have displayed more potent HDACs
inhibition activity than clinically used drug SAHA. Among them, compound 5f was
identified as a candidate for lead optimization with promising drug-like properties.
This compound displayed significant antiproliferative effect in vitro, not possessing
significant toxicity to primary human cells and hERG inhibition. Meanwhile, 5f
showed selective inhibition of HDAC 1,6 over the HDAC8 isoform and possessed
acceptable pharmacokinetic characteristics. In an A549 non-small cell lung cancer
mouse xenograft model study in vivo, 5f showed significant tumor growth inhibition
without signs of adverse effects. Further evaluation of compound 5f is ongoing and
will be reported in due time.
Acknowledgement
This work was financially supported by the National Science and Technology
Major Project (No. 2017ZX09101004-002-009), the National Natural Science
Foundation of China (No. 81703358) and Science and Technology Commission of
Shanghai Municipality (Nos. 14YF1412800, 17431903900 ,15QB1404400).

References
1. Wang ZB, Zang CZ, Cui KR, et al. Cell. 2009:138:1019.
2. Turner BM. Bioessays. 2000:22:836.
3. Johnstone RW. Nat Rev Drug Discov. 2002:1:287.
4. Dokmanovic M, Marks PA. J Cell Biochem. 2005:96:293.
5. Bolden JE, Peart MJ, Johnstone RW, Nat Rev Drug Discov. 2006:5:769.
6. Secrist JP, Zhou X, Richon VM. Curr Opin Invest Dr. 2003:4:1422.
7. Nguyen T, Dai Y, Attkisson E, et al. Clin Cancer Res. 2011:17:3219.
8. Frew AJ, Johnstone RW, Bolden JE. Cancer Lett. 2009:280:125.
9. Finnin MS, Donigian JR, Cohen A, et al. Nature. 1999:401:188.
10. Zhou J, Wu R, Luo HB. Phys Chem Chem Phys. 2015:17:29483.
11. Sandhu P, Andrews PA, Baker MP, et al. Drug Metab Lett. 2007:1:153.
12. Kelly WK, Richon VM, O’Connor O, et al. Clin Cancer Res. 2003:9:3578.
13. Guha M. Nat Rev Drug Discov. 2015:14:225.
14. Zhang QW, Feng J, Li JQ. B Kor Chem Soc. 2014:35:129.
15. Dai Y, Guo Y, Guo J, et al. Bioorg Med Chem Lett. 2003:34:1897.
16. Dai Y, Guo Y, Curtin ML, et al. Bioorg Med Chem Lett. 2003:13:3817.
17. Zhang Y, Feng J, Jia Y, et al. Eur J Med Chem. 2011:46:5387.
18. Chen PC, Patil V, Guerrant W, et al. Bioorg Med Chem Lett. 2008:16:4839.
19. Zhang J, Li XY, Jiang YQ, et al. Bioorg Med Chem.2014:22:3055.
20. Wang L, Hou X, Fu H, et al. Bioorgan Med Chem. 2015:23:4364.
21. Dubin AE, Nasser N, Rohrbacher J, et al. J Biomol Screen. 2005:10:168.
22. Sandhu P, Andrews PA, Baker MP, et al. Drug Metab Lett. 2007:1:153.

Highlights
 A series of 4-aminoquinazolinyl-containing HDACIs were
synthesized.
 Target compounds were screened for their HDACs inhibitory
activity.
 Compound 5f showed good profiles in vitro and had low affinities
for hERG channels.
 5f possessed acceptable pharmacokinetic characteristics.
 5f exhibited significant antitumor activity in vivo.

Graphical abstract