Design, synthesis, and biological evaluation of quinazoline derivatives as dual HDAC1 and HDAC6 inhibitors for the treatment of cancer

Article abstract of DOI:10.1111/cbdd.13405

Fifty-eight quinazoline-based compounds were designed and synthesized based on the structural optimizations from the lead compound 23bb in an attempt to search for more potent dual HDAC1 and HDAC6 inhibitors. Among them, 32c (HDAC1, IC₅₀?=?31.1

Full text of DOI:10.1111/cbdd.13405

MISS LINHONG HE (Orcid ID : 0000-0002-7248-1259)
Article type
: Research Article
Design, Synthesis and Biological Evaluation of Quinazoline
Derivatives as Dual HDAC1 and HDAC6 Inhibitors for the
Treatment of Cancer
Jinying Chen b†, Zitai Sang c†, Youjun Jiang b, Chao Yang b, Linhong He a,#
a Pharmaceutical College, Guangxi Medical University, Nanning, Guangxi, China.
b SinoGrain Chengdu Storage Research Institute Co.Ltd
c State Key Laboratory of Biotherapy, West China Hospital, West China Medical
School, Sichuan University, Chengdu, Sichuan, PR China.
Running Title: Dual HDAC1 and HDAC6 inhibitors for the treatment of cancers
Correspondence
Linhong He, Pharmaceutical College, Guangxi Medical University, Nanning,
Guangxi, China. E-mail: HLhongedu@163.com.
This article has been accepted for publication and undergone full peer review but has
not been through the copyediting, typesetting, pagination and proofreading process,
which may lead to differences between this version and the Version of Record. Please
cite this article as doi: 10.1111/cbdd.13405
This article is protected by copyright. All rights reserved.

Keywords: dual inhibitors, HDAC, SAR study, cancer
Abstract
Fifty-eight quinazoline based compounds were designed and synthesized based
on the structural optimizations from the lead compound 23bb in an attempt to search
for more potent dual HDAC1 and HDAC6 inhibitors. Among them, 32c (HDAC1,
IC50 = 31.10 ± 0.37 nM; HDAC6, IC50 = 16.15 ± 0.62 nM) and 32d (HDAC1, IC50 =
37.00 ± 0.24 nM; HDAC6, IC50 = 35.00 ± 0.71 nM) were not only identified as potent
dual-acting HDAC1 and HDAC6 inhibitors with over 10-fold selectivity to the other
HDACs, but also displayed activities in tubulin acetylation and histone H3 acetylation
induction. Importantly, both of them displayed strong antiproliferative activities
against various tumor cell lines in vitro with IC50 values less than 40 nM, especially
for hematological tumors cells (U266 and RPMI8226, IC50 < 1 nM), which were even
better than 23bb and SAHA. Furthermore, 32c showed a significant tumor growth
inhibition (antitumor rate = 63.98%, p<0.05) in the resistant MCF-7/ADR xenograft
model without any obvious body weight changes and abnormal behaviors. Our
findings validate that 32c is a potent dual inhibitor of HDAC1/6 that can be an
efficacious treatment for breast cancer with adriamycin resistance.
Introduction
Histone deacetylases (HDACs) have been identified as attractive targets for
cancer therapeutics, which are responsible for deacetylation of lysine resides in
histone and non-histone substrates (p53 and tubulin), and play important roles in
regulating various tumor suppressor pathways (1-5). Until now, 18 human HDACs
have been categorized into four classes: class I (HDAC1, 2, 3 and 8), class II (class
IIa: HDAC4, 5, 7 and 9; class IIb: HDAC6 and 10), class III (sirtuins, SIRT1-7) and
class IV (HDAC11) (3). All the HDACs are zinc-dependent enzymes with a highly
This article is protected by copyright. All rights reserved.

conserved catalytic site except class III that requires NAD+ as a cofactor (3,6).
Although the biological functions of many HDACs subtypes are still being defined,
there are compelling evidences that class I HDACs and HDAC6 overexpress in a
variety of cancers, including colorectal cancer, prostate cancer, breast cancer,
hepatocellular carcinoma (3,7,8). Besides, more and more HDAC inhibitors (Figure
1) have been proven to induce growth arrest, differentiation, and apoptosis in cancer
cells, as well as suppress in vivo tumor growth (3, 6). Therefore, HDAC inhibitors are
emerging as a new class of chemo-therapeutic agents.
Figure 1
Up to now, four HDAC inhibitors (Vorinostat, Romidepsin, Belinostat and
Panobinostat) approved by the US FDA have been proven to inhibit either all or at
least several HDAC members, which may trigger numerous side effects and low
potencies (6, 13). Meanwhile, single target drugs could not meet the clinical demand
due to their various drug resistance phenomena (9-10). To achieve curative effects
against tumors with fewer side effects, continuous efforts in this field has been
focused on the development of dual-acting (HDAC–Topo I) or triple (HDAC 1-3)
inhibitors (9-11). Since Dual-acting HDAC1 and HDAC6 inhibitors not only can
potently inhibit tumor growth, but also protect cortical neurons against
hypoxia-induced toxicity (3,7,8), developing dual inhibitors targeting HDAC1 and
HDAC6 is a promising and attractive strategy to broad their application in the
treatment of cancers, especially for solid tumors (8,11,12).
HDAC inhibitors generally consist of three canonical pharmacophoric
characteristics, a zinc binding group (ZBG) that chelates the active site Zn2+ ion, a
surface recognition cap group that interacts with the amino acid residues presenting at
the surface of the HDAC, and a linker unit between the former two parts (3).
Numerous investigations claimed that a hydroxamic acid group is the most common
ZBG forming one or more hydrogen bonds with histone deacetylase homologue
This article is protected by copyright. All rights reserved.

(Figure 1) (3,6). In our previous study, we have deeply investigated the structure
activity relationship (SAR) of the cap group that may contribute an important function
to the selective activity. A series of fragments were respectively docked into HDAC1
(PDB code 4BKX) and HDAC6 (homology model) (3). Among them,
2-methylquinazoline was picked out since it can comfortably occupy the surface
groove and form robust interaction with HDACs (3). Additionally, it indeed displayed
excellent inhibitory activity against HDAC1 and HDAC6 (3). Based on the initial
leading compound 23bb (HDAC1, IC50 = 422 nM; HDAC6 IC50 = 17 nM) with
favorable anti-tumor activity, we plan to get insights into its SAR and rationally
modify its cap and linker regions to seeking new dual HDAC1 and HDAC6 inhibitors.
In this paper, we report the synthesis and pharmacological evaluation of fifty-eight
compounds with antitumor activities, as well as inhibitory activities against HDAC1
and HDAC6, of which the promising compounds 32c and 32d were further conducted
in resistant MCF-7/ADR xenograft model in vivo to find the optimal candidate
compound.
Figure 2
1. Chemistry
A common route was applied for the synthesis of compounds 6a-6jj (Scheme 1).
Commercially available material, 2-methoxy-5-nitrophenol (1), was treated with ethyl
4-bromobutanoate in an alkaline condition to afford intermediate 2, which was
reduced to aniline derivative 3 with hydrogen and Palladium/C. Then 3 was reacted
with 4-chloro-2-methylquinazoline at room temperature (rt) to provide 4 in a good
yield. A variety of halogenated reagents were respectively introduced to 4 via
microwave heating to prepare 5a-5jj, which could improve the reaction speed and
yield. Target hydroxamic acid compounds 6a-6jj were directly converted through
reacting 5a-5jj with hydroxylamine.
This article is protected by copyright. All rights reserved.

Scheme 1
To study the SAR of linker with different length alkyl chains, novel compounds
14a-14f were prepared according to Scheme 2. The phenolic hydroxyl group (7) was
firstly protected by chloromethyl methyl ether (MOMCl) to give 8, which was
reduced with hydrogen and coupled with 4-chloro-2-methylquinazoline in sequence to
give intermediate 10. The further N-methylated for 10 was carried out in a strong
alkaline condition with iodomethane to offer 11, whose protection group was removed
using saturated hydrogen chloride in ethyl acetate, giving the key intermediate 12.
The subsequent reactions were the same as Scheme1.
Scheme 2
Another kind of linker, different amide alkyl chains, was also explored to search
more potent HDAC inhibitors. Final compounds 20a-20e were prepared according to
Scheme 3, and the reaction procedures were the same as Scheme 1.
Scheme 3
Next, we also focused on the linker position on phenyl group. As shown in
Scheme 4, N-methyl-4-nitroaniline (21) and 4-chloro-2-methylquinazoline were used
as the starting materials, and the following reactions were similar as in Scheme 1.
Eventually, target compounds 25a-25e were obtained.
Scheme 4
Based on the SAR analysis of those new compounds, 32a-32f were also
synthesized according to Scheme 5. 4-Fluoro-3-nitroaniline (26) was reacted with
sodium methoxide to give 27, which was treated with corresponding reactants in the
same procedures mentioned above to offer the resulting compounds 32a-32f.
Scheme 5
This article is protected by copyright. All rights reserved.

2. Result and discussion
2.1 Antiproliferative activity tests for synthesized compounds
Given that the lead compound 23bb could significantly inhibit HCT116 cells to
proliferation (3), all synthesized derivatives were conducted to assess their
antiproliferative activities against HCT116 cells using MTT assay. The primary tested
concentrations for compounds were set as 5 μM, 500 nM, 50 nM and 5 nM to find out
candidate compounds as many as possible. As summarized in Table 1, 23bb with a
methyl on R1 position displayed a good potency. Compounds replacing the methyl by
a substituted group containing branched chains or distinct benzyl group, such as 6c-g,
6k-o, 6aa-6jj, completely lost their activities, which may result from steric hindrance
effect and long side chains contributed to steric effects in a certain degree.
Interestingly, 6j was more potent than 6a, despite the fact that the volume of allyl
group is larger than ethyl. Similar trend was observed between 6h and 6g, which may
be attributed to their good conjugation. When linked with electron-donor and
electron-acceptor, the charge transfer is obvious, thus showed a good potency. Taking
into account the SAR known from 23bb’s discovery program and our first
modifications on R1 position, it was very likely that
N-substituted-N,2-dimethylquinazoline fragment was the optimal cap.
Table 1
Results from the comparisons of 23bb and 14b, or 20c and 25c, demonstrated
that compounds moving the linker from 3’-phenyl to 4’-phenyl led to different
degrees of loss in activities. Moreover, compounds upon introducing a methoxy at
4’-phenyl (23bb, 32c and 32d) significantly increased potencies, whose IC50 values
were less than 50 nM, indicating that this methoxy was necessary to improve activity.
Variation of the linker length verified the optimal distance between phenyl and
hydroxamic acid was 4-atom, while 5-atom slightly reduced the potency, see
compounds 20c and 20d, or 32c and 32d. Instead of ether chain at 3’-phenyl, the
This article is protected by copyright. All rights reserved.

amide hydrocarbon chain owned better activity by comparison 23bb and 32c. In brief,
the importance of the changes on 23bb’s cap and linker regions was summarized in
Figure 3.
Figure 3
2.2 Inhibitory activities against HDAC1 and HDAC6
In an attempt to confirm their potential inhibitory activities against HDAC1 and
HDAC6 isoforms, compounds (6a, 6h, 6j, 14a, 14f, 20c, 32a, 32b, 32c and 32d)
potently inhibiting HCT116 cells to growth were further evaluated. Interestingly, the
inhibitory activities against HDAC6 not HDAC1 for all tested compounds were in
line with the SAR mentioned above (Table 2). However, apart from 14f, all the
compounds increased the potencies of inhibiting HDAC1 in contrast to 23bb.
Although slight potencies decrease versus the positive control SAHA, both 32c and
32d significantly inhibited the activities of HDAC1 and HDAC6, whose inhibitory
activities against HDAC1 were much better than that of 23bb. To highlight the
inhibition of 32c and 32d towards HDAC1 and HDAC6, automated docking studies
were carried out using the Discovery studio 2.1 software. As expected, 32c and 32d
matched perfectly with the configuration of the binding sites of HDAC1 and HDAC6,
whose results were consist with their enzyme potencies (Supplementary Material
Figure S1).
Table 2
2.3 Antiproliferative activity tests for selected compounds
Since HDAC1 and HDAC6 have been verified to play important roles in various
types of cancers, whose inhibitors have demonstrated potencies as novel therapeutics
in preclinical and clinical studies of both solid and haematological cancers,
compounds 32a-32d were selected to investigate their antiproliferative activities
against another eight tumor cell lines on the basis of their satisfied potencies in vitro,
including solid tumors (liver cancer HepG2 cells, breast cancer M-M-231and MCF-7
This article is protected by copyright. All rights reserved.

cells, lung cancer H1975 and H460 cells, and cervical cancer Hela cells) and
hematological tumors (myeloma U266 and RPMI8226 cells). As listed in Table 3, all
of them could significantly suppress the proliferation for tested cell lines (IC50 < 400
nM). As expected, myeloma U266 and RPMI8226 cells were presented the best
inhibition, in accordance with the fact that most of the current HDAC inhibitors were
used for the treatment of hematologic malignancies, but a lack of visible efficacy
against solid tumors (3). Among them, 32c and 32d showed significant
antipoliferative potentials with IC50 values ranging from 0.01 to 40 nM in these tumor
cell lines, which were much better than 23bb and SAHA. It turned out that both 32c
and 32d could not only be used in the therapy of hematologic malignancies, but also
solid tumors.
Table 3
2.4 HDAC isoform selectivity for 32c and 32d
Since both 32c and 32d exhibited potent bioactivities in vitro, they were further
determined for the activities on the other HDAC isoforms. As summarized in Table 4,
compound 32c was potent at 1000 nM against class one HDACs and HDAC6, but at
more than one micromolar level against the other HDACs. Similar trend was observed
by 32d. What’s more, both of them showed over 10-fold selectivity of HDAC1 or
HDAC6 relative to the inhibition of HDAC2, HDAC3 and HDAC8, indicating that
32c and 32d were dual-acting HDAC1 and HDAC6 inhibitors.
Table 4
2.5 Tubulin acetylation and H3 acetylation test
Cytoblot experiments were performed to evaluate the histone H3 (a known
substrate for HDACs 1, 2 and 3) acetylation (Ac-H3) and tubulin (a known substrate
for HDAC6) acetylation (Ac-Tub) for compounds 32c and 32d in Hela cells. As
expected, both of them displayed better Ac-Tub activities than Ac-H3 (Table 5),
which were in consistent with the in vitro enzymatic data, indirectly demonstrating
This article is protected by copyright. All rights reserved.

that 32c and 32d owned a slightly better inhibition to HDAC6 than HDAC1.
Additionally, 32c showed better activity than 23bb and SAHA in both situations.
Table 5
2.6 Immunofluorescence staining
Since recent reports have highlighted HDAC involvement in the deacetylation of
not only histone regulatory proteins, but also important non-histone ones, such as
tubulin (2, 3, 9), compounds 32c and 32d were further investigated their potencies of
inhibiting tubulin polymerization or de-polymerization by immunofluorescence
staining. Paclitaxel (PTX) was selected as microtubules polymerizer and MPC6827
was an anti-tubulin agent. As shown in Figure 4, the morphology of microtubules and
spindles in H460 cells treated with 32c and 32d for 12 h were significantly suppressed
the microtubule polymerization and spindle formation, which were similar to those
incubated with MPC6827, but not PTX, demonstrating that 32c and 32d possessed the
potencies of preventing tubulins to form microtubules. Those results confirmed that
the favorable anti-tumor activities of these dual-acting HDAC1 and HDAC6
inhibitors may be attributed from the inhibition of tubulins to form microtubules.
Figure 4
2.7 Anti-tumor activity in MCF-7 xenograft model
To determine the anti-tumor potencies of compounds 32c and 32d in drug
resistant tumor xenograft model in vivo, human breast cancer MCF-7/ADR model
were established, and a broad spectrum anticancer drug Adriamycin (ADM) was set
as positive control. 30 Balb/C mice were randomly assigned to four groups when the
tumor volumes reached about a size of 100 mm3. Treatment groups respectively
received tail intravenously ADM (4 mg/kg, every 7 days), 32c (2.5 kg/kg, every 2
days) and 32d (5 mg/kg, every 2 days) for 34 days, and the vehicle (control) group
mice were given equal amount of physiological saline every 2 days in the same way.
This article is protected by copyright. All rights reserved.

As shown in Figure 5B-D, 32c caused a remarkable reduction in tumor growth
(63.98%, p<0.05) as compared with administration of vehicle only, which was better
than 32d (48.72%, p<0.05) and the positive control ADM (16.1%). As expected, the
tumor growth inhibition of reference drug ADM was low since it had developed
resistance to MCF-7/ADM cell line. Moreover, ADM-treated mice were observed
obvious loss of body weight. However, the optimal compound 32c treatment group
exhibited no significant body weight changes and abnormal behaviors, which were a
little better than that of 32d-treated mice. These results proved that 32c was an
effective compound for the treatment of ADM resistant breast cancer without
significant toxicity.
Figure 5
3. Conclusion
Fifty-eight quinazoline based compounds were synthesized, and their
bioactivities in vitro revealed a relatively clear SAR. Compounds 32a-d displayed
strong antiproliferative activities against various tumor cell lines, including solid
tumors (HCT116, HepG2, M-M-231, MCF-7, H460, H1975, H460 and Hela cells)
and hematological tumors (U266 and RPMI8226 cells). Among them, 32c and 32d
were not only identified as good dual-acting HDAC1 and HDAC6 inhibitors with
over 10-fold selectivity to the other isoforms, but also exhibited good activities in
tubulin acetylation and histone H3 acetylation induction, of which 32c was even better
than 23bb and SAHA. Additionally, both of them could effectively suppress tubulins
to form microtubule that was supposed to potently enhance their anti-tumor activities.
In the resistant MCF-7/ADR xenograft animal model, 32c significantly inhibited
tumor growth without significant weight changes and abnormal behaviors, which was
better than 32d. Hence, further investigation of compound 32c is ongoing and will be
reported in due time.
This article is protected by copyright. All rights reserved.

Acknowledgments
We are grateful to the National Key Programs of China during the 13th Five-Year
Plan Period.
Conflict of Interest
The authors have declared no conflict of interest.
Supplementary Material
Docking studies: 32c and 32d were docked into HDAC1 and HDAC6,
respectively.
Experimental section: Synthetic details, characterization data for all compounds,
and biological procedure reported in this manuscript.
NMR Spectrum: 1H- and 13C-NMR spectrum of all final new compounds.
References and notes
1. Minucci S., Pelicci P.G. (2006) Histone deacetylase inhibitors and the promise of
epigenetic (and more) treatments for cancer. Nat. Rev.Cancer. 6: 38-51.
2. Hubbert C., Guardiola A., Shao R., Kawaguchi Y., Ito A., Nixon A., Yoshida M.,
Wang X. F., Yao T.P. (2002) HDAC6 Is a Microtubule-Associated Deacetylase.
Nature. 417: 455-458.
3. Yang Z., Wang T., Wang F., Niu T., Liu Z., Chen X., Long C., Tang M., Cao D.,
Wang X., Xiang W., Yi Y., Ma L., You J., Chen L. (2016) Discovery of Selective
This article is protected by copyright. All rights reserved.

Histone Deacetylase 6 Inhibitors Using the Quinazoline as the Cap for the
Treatment of Cancer. J. Med. Chem. 59:1455-1470.
4. Zhou Y., Dun Y., Fu H., Wang L., Pan X., Yang X., Fang H. (2017) Design,
synthesis, and preliminary bioactivity evaluation of N-benzylpyrimidin-2-amine
derivatives as novel histone deacetylase inhibitor. Chem. Biol. Drug Des. 90:
936-942.
5. Zhang S., Huang W., Li X., Yang Z., Feng B. (2015) Synthesis, Biological
Evaluation, and Computer-Aided Drug Designing of New Derivatives of
Hyperactive Suberoylanilide Hydroxamic Acid Histone Deacetylase Inhibitors.
Chem. Biol. Drug Des. 86: 795-804.
6. Cheng J., Qin J., Guo S., Qiu H., Zhong Y. (2014) Design, synthesis and
evaluation of novel HDAC inhibitors as potential antitumor agents. Bioorg. Med.
Chem. Lett. 24: 4768-4772.
7. Sun Q., Yao Y., Liu C., Li H., Yao H., Xue X., Liu J., Tu Z., Jiang S. (2013)
Design, synthesis, and biological evaluation of novel histone deacetylase 1
inhibitors through click chemistry. Bioorg. Med. Chem. Lett. 23: 3295-3299.
8. Tsai H., Wu J., Kao M., Chen J., Sun G.Y., Ong W., Lin T. (2016) Clinacanthus
nutans Protects Cortical Neurons Against Hypoxia-Induced Toxicity by
Downregulating HDAC1/6. NeuroMol. Med. 18: 274.
9. Guerrant W., Patil V., Canzoneri J.C., Yao L., Hood R., Oyelere D.K. (2013)
Dual-acting histone deacetylase-topoisomerase I inhibitors. Bioorg. Med. Chem.
Lett. 23: 3283-3287.
10. Jadhavar P.S., Ramachandran S.A., Riquelme E., Gupta A., Quinn K.P.,
Shivakumar D., Ray S., Zende D., Nayak A.K., Miglani S.K., Sathe B.D., Raja
M., Farias O., Alfaro I., Belmar S., Guerrero J., Bernales S., Chakravarty S.,
Hung D.T., Lindquist J.N., Rai R. (2016) Targeting prostate cancer with
This article is protected by copyright. All rights reserved.

compounds possessing dual activity as androgen receptor antagonists and
HDAC6 inhibitors. Bioorg. Med. Chem. Lett. 26: 5222-5228.
11. Ma X., Lv X., Zhang J. (2018) Exploiting polypharmacology for improving
therapeutic outcome of kinase inhibitors (KIs): An update of recent medicinal
chemistry efforts. Euro. J. Med. Chem. 143: 449-463.
12. Wen J., Niu Q., Liu J., Bao Y., Yang J., Luan S., Fan Y., Liu D., Zhao L. (2016)
Novel thiol-based histone deacetylase inhibitors bearing
3-phenyl-1H-pyrazole-5-carboxamide scaffold as surface recognition motif:
Design, synthesis and SAR study. Bioorg. Med. Chem. Lett. 26: 375-379.
13. Zhang Q., Lu B., Li J. (2017) Design, synthesis and biological evaluation of
4-piperazinyl-containing Chidamide derivatives as HDACs inhibitors. Bioorg.
Med. Chem. Lett. 27: 3162-3166.
Caption
Figure 1. Examples of HDAC inhibitors.
Figure 2. The idea to explore the SAR of compound 23bb.
Figure 3. Graphical representation of SAR for quinazoline based compounds with
antiproliferative activities against HCT116.
Figure 4. Effect of compounds PTX, MPC6828, 32c and 32d on the microtubule in
H460 cells. Treated cells were respectively incubated with compounds PTX,
MPC6828, 32c and 32d at concentrations of 1 nM and 10 nM for 12 h, then the
untreated (control) and treated cells were fixed in methanol and stained with α-tubulin
and counterstained with 4,6-diamidino-2-phenylindole (DAPI). Microtubules and
unassembled tubulin are shown in green. DNA, stained with DAPI, is shown in blue.
This article is protected by copyright. All rights reserved.

Figure 5. Anti-tumor activities of 32c, 32d and ADM on the MCF-7ADR xenograft
models. Body weight (A) and tumor volume (B) of mice were measured every other
day during the treatment. Solid tumors were got by sacrificing those mice after 34
days’ treatment, then measured (C) and photographed (D). *p < 0.05, significantly
different compared with control.
Scheme 1. Synthesis of 6a-6jj. Reagents and conditions: (a) Br(CH2)3COOEt,
Cs2CO3, MeCN, reflux; (b) H2, Pd/C (5%), MeOH, rt; (c)
4-Chloro-2-methylquinazoline, Me2CHOH, rt; (d) R1Cl(R1Br), K2CO3, KI, DMF,
MW (80 W,100 ºC, 1 h) . (e) 50% NH2OH (aq), NaOH, DCM/MeOH (v/v, 1/2), rt.
Scheme 2. Synthesis of 14a -14f. Reagents and conditions：(a) NaH, MOMCl, DMF,
0 °C to rt; (b) H2, Pd/C (5%), MeOH, rt; (c) 4-Chloro-2-methylquinazoline,
Me2CHOH, rt; (d) NaH, CH3I, DMF, 0 °C to rt; (e) EtOAc, HCl; (f) Br(CH2)nCOOEt,
Cs2CO3, MeCN; (g) 50% NH2OH (aq), NaOH, DCM/MeOH (v/v, 1/2), rt.
Scheme 3. Synthesis of 20a-20e. Reagents and conditions：(a)
4-Chloro-2-methylquinazoline, Me2CHOH, rt; (b) NaH, CH3I, DMF, 0 °C to rt; (c)
H2, Pd/C (5%), MeOH, rt; (d) ClCO(CH2)nCOOEt, Et3N, DCM, rt; (e) 50% NH2OH
(aq), NaOH (K2CO3), DCM/MeOH (v/v, 1/2), rt.
Scheme 4. Synthesis of 25a-25e. Reagents and conditions: (a)
4-Chloro-2-methylquinazoline, Me2CHOH, rt; (b) H2, Pd/C (5%), MeOH, rt; (c)
ClCO(CH2)nCOOEt, Et3N, DCM, rt; (d) 50% NH2OH (aq), NaOH (K2CO3),
DCM/MeOH (v/v, 1/2), rt.
Scheme 5. Synthesis of 32a-32f. Reagents and conditions: (a) CH3ONa, MeOH, rt;
(b) 4-Chloro-2-methylquinazoline, Me2CHOH, rt; (c) NaH, CH3I, DMF, 0 °C to rt;
(d) H2, Pd/C (5%), MeOH, rt; (e) ClCO(CH2)nCOOEt, Et3N, DCM, rt; (f) 50%
NH2OH (aq), NaOH (K2CO3), DCM/MeOH (v/v, 1/2), rt.
This article is protected by copyright. All rights reserved.

Table 1. Antiproliferative activities of all the synthesized compounds against
HCT116 cells a
Comp. potency
Comp. potency
Comp. potency
Comp.
20d
20e
potency
6a
6b
6c
6d
6e
6f
++
+
-
6p
6q
6r
+
+
-
6ee
6ff
-
+
-
+
6gg
6hh
6ii
-
25a
25b
25c
++
+
-
6s
-
-
-
6t
+
-
-
+
-
6u
6v
6jj
-
25d
25e
+
6g
6h
6i
-
+
-
14a
14b
14c
14d
14e
14f
20a
20b
20c
++
+
+
+
-
-
++
-
6w
6x
32a
32b
32c
++
++
++++
+++
+
+
+
+
-
6j
++
-
6y
6k
6l
6z
32d
32e
-
6aa
6bb
6cc
6dd
++
+
+
++
6m
6n
6o
-
-
32f
+
-
-
23bb
+++
-
-
a The test of antiproliferative activities against HCT116 cells for all synthesized
compounds were conducted with four concentrations: 5 μM, 500 nM, 50 nM and 5
nM. Thus, “-” means inhibition rate less than 50% at 5 μM, “+” means inhibition rate
over 50% at 5 μM, “++” means inhibition rate over 50% at 500 nM, “+++” means
inhibition rate over 50% at 50 nM, “++++” means inhibition rate over 50% at 5 nM.
Data are expressed as the mean ± SD from at least three independent experiments.
This article is protected by copyright. All rights reserved.

Table 2. The inhibitory activities of selected compounds against HDAC1 and
HDAC6 a
IC50 (nM)
HDAC1
IC50 (nM)
HDAC1
Comp.
Comp.
HDAC6
HDAC6
272.10±3.44 76.23±0.71
591.40±2.99 120.30±1.00
364.60±4.70 45.28±1.29
395.23±3.01
108.20±1.33
87.46±1.03
31.10±0.37
37.00±0.24
11.40±0.39
82.04±1.23
59.27±2.08
50.16±1.00
16.15±0.62
35.00±0.71
16.01±1.37
6a
20c
6h
32a
6j
32b
32c
31.06±0.83
85.37±2.00
14a
14f
23bb
578.26±1.96 322.94±2.02
422.00±0.93 19.03±1.80
32d
SAHA
a Compounds were tested in the 8-dose IC50 mode in duplicate with 5-fold serial
dilutions starting at 5 μM. The IC50 values are the mean of at least two experiments.
This article is protected by copyright. All rights reserved.

Table 3. Activities of compounds 32a-32d and 23bb against various tumor cell Lines a
IC50 (nM)
Comp.
H460
RPMI822
6
HepG2
M-M-231
MCF-7
H1975
Hela
U266
40.94±2.0 121.03±1.0 376.34±19.2 59.28±0.4
46.73±1.20 38.51±1.52
1.06±0.37 2.91±0.61
1.54±0.04 2.98±0.30
32a
32b
7
6
0
9
12.51±0.9
2
230.63±2.80 13.86±1.2
3
4.77±0.72
14.29±1.03
31.71±0.77
3.50±0.48
4.51±0.33
2.65±0.71
7.41±0.81
2.55±0.40 2.88±0.69
1.05±0.02
2.41±0.59 ＜0.01
＜0.01
3.16±0.15 0.15±0.01 0.37±0.21
50.02±0.7 13.83±1.0 52.21±0.6
32c
32d
8.62±1.01 39.51±1.00 37.59±2.62
55.13±4.72
39.07±1.11
>1000
-
-
-
23bb
4
1
6
>1000
720±13.0 569±21.0
>1000
>1000
>1000
420±9.37
SAHA
2
0
a IC50 = compound concentration required to inhibit tumor cell proliferation by 50%. Data are expressed as the mean ± SD from the
dose-response curves of at least three
independent experiments.
This article is protected by copyright. All rights reserved.

Table 4. HDAC inhibition activities of compounds 32c and 32d a
IC50 / nM
isoform
32c
32d
23bb
SAHA
HDAC1
HDAC2
HDAC3
HDAC8
HDAC4
HDAC5
HDAC7
HDAC9
HDAC6
HDAC10
HDAC11
31.10±0.37
349±12.05
401±18.99
934.17±42.01
>1000
37.00±0.24
512.08±21.49
510.66±9.43
>1000
422.00±0.93
390.01±5.41
410.29±10.21
>1000
11.40±0.39
34.98±3.06
32.10±2.77
170.00±19.02
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
16.15±0.62
>1000
35.00±0.71
>1000
19.03±1.80
>1000
16.01±1.37
170.00±15.32
>1000
>1000
>1000
>1000
a Compounds were tested in the 8-dose IC50 mode in duplicate with 5-fold serial
dilutions starting at 10 μM. The IC50 values are the mean of at least two experiments.
This article is protected by copyright. All rights reserved.

Table 5. H3 acetylation and tubulin acetylation induction in Hela cell of compounds
32c and 32d a
EC50/nM
32c
32d
217.20±11.93 1739.03±83.91 377.02±9.03
75.33±6.00 59.00±10.28 401.11±13.97
23bb
SAHA
Ac-H3
170±9.22
43.92±6.07
Ac-Tub
a EC50 values of tubulin acetylation and Ac-H3 are based on ELISA experiments run in
duplicate in Hela cell. Data are expressed as the mean ± SD from the dose-response
curves of at least three independent experiments.
This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.