Development of Purine-Based Hydroxamic Acid Derivatives: Potent Histone Deacetylase Inhibitors with Marked in Vitro and in Vivo Antitumor Activities

Article abstract of DOI:10.1021/acs.jmedchem.6b00579

In the present study, a series of novel histone deacetylase (HDAC) inhibitors using the morpholinopurine as the capping group were designed and synthesized. Several compounds demonstrated significant HDAC inhibitory activities and antiproliferative effects against diverse human tumor cell lines. Among them, compound 10o was identified as a potent class I and class IIb HDAC inhibitor with good pharmaceutical profile and druglike properties. Western blot analysis further confirmed that 10o more effectively increased acetylated histone H3 than panobinostat (LBH-589) and vorinostat (SAHA) at the same concentration in vitro. In in vivo efficacy evaluations of HCT116, MV4-11, Ramos, and MM1S xenograft models, 10o showed higher efficacy than SAHA or LBH-589 without causing significant loss of body weight and toxicity. All the results indicated that 10o could be a suitable candidate for treatment of both solid and hematological cancer.

Full text of DOI:10.1021/acs.jmedchem.6b00579

Subscriber access provided by University of Sussex Library
Article
Development of Purine-Based Hydroxamic Acid Derivatives: Potent Histone
Deacetylase Inhibitors with Marked in Vitro and in Vivo Antitumor Activities
Yong Chen, Xiaoyan Wang, Wei Xiang, Lin He, Minghai Tang, Fang Wang, Taijin Wang, Zhuang
Yang, Yuyao Yi, Hairong Wang, Ting Niu, Li Zheng, Lei Lei, Xiaobin Li, Hang Song, and LiJuan Chen
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00579 • Publication Date (Web): 17 May 2016
Downloaded from http://pubs.acs.org on May 18, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
University
Song, Hang; Sichuan University, School of Chemical Engineering
Chen, LiJuan; State Key Laboratory of Biotherapy, West China Medical
School
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
Development of PurineꢀBased Hydroxamic Acid Derivatives: Potent Histone
Deacetylase Inhibitors with Marked in Vitro and in Vivo Antitumor Activities
a,b,§
a,§
a,§
a,§
a
a
Yong Chen , Xiaoyan Wang , Wei Xiang , Lin He , Minghai Tang , Fang Wang ,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
a
c
a
c
a
Taijin Wang , Zhuang Yang , Yuyao Yi , Hairong Wang , Ting Niu , Li Zheng , Lei
a
a
b
a,
Lei , Xiaobin Li , Hang Song and Lijuan Chen *
a
State Key Laboratory of Biotherapy/Collaborative Innovation Center of Biotherapy
and Cancer Center, West China Hospital of Sichuan University; Chengdu, China.
b
School of Chemical Engineering, Sichuan University, Chengdu, China.
c
Department of Hematology and Research Laboratory of Hematology, West China
1
Hospital of Sichuan University, Chengdu, 610041, China.
§
11
These authors contributed equally and should be considered as coꢀfirst authors
ACS Paragon Plus Environment

Page 3 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
2
2
2
3
4
5
6
7
8
9
0
1
ABSTRACT:
In the present study, a series of novel histone deacetylase (HDAC) inhibitors
using the morpholinopurine as the capping group were designed and synthesized.
Several compounds demonstrated significantly HDAC inhibitory activities and
antiꢀproliferative effects against diverse human tumor cell lines. Among them,
compound 10o was identified as a potent class I and IIb HDAC inhibitor with good
pharmaceutical profile and drugꢀlike properties. Western blot analysis further
confirmed that 10o more effectively increased acetylated histone H3 than panobinostat
(LBHꢀ589) and vorinostat (SAHA) at the same concentration in vitro. In in vivo
efficacy evaluations of HCT116, MV4ꢀ11, Ramos and MM1S xenograft models, 10o
showed higher efficacy than SAHA or LBHꢀ589 without causing significant loss of
body weight and toxicity. All the results indicated that 10o could be a suitable
candidate for treatment of both solid and hematological cancer.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22
2
3
24
2
5
INTRODUCTION
26
The histone proteins play important roles in the control of gene expression via
modification of chemical reactions such as acetylation, phosphorylation, and
2
7
1
, 2
28
methylation. In most cases, histone acetylation enhances transcription while histone
3, 4
2
9
deacetylation represses transcription. Inhibition of histone deacetylases (HDACs)
30
causes the accumulation of acetylated histones, bringing about a variety of cell type
dependent responses such as apoptosis, differentiation, cell survival, and inhibition of
3
1
5
32
proliferation. There are 18 HDAC isozymes and they are divided into four classes:
3
3
class I (HDAC 1ꢀ3, and 8), class II (HDAC 4, 5, 6, 7, 9, 10) and class IV (HDAC 11)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
3
3
3
3
3
4
4
4
4
4
5
6
7
8
9
0
1
2
3
are all zincꢀdependent deacetylases, and class III isozymes including Sirt1 to Sirt7 are
+
6,7
NAD ꢀdependent isozymes.
Most HDAC inhibitors fit
a
threeꢀmotif
pharmacophoric model consisting of a zinc binding group (ZBG), a linker, and a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
ꢀ10
surface recognition cap region.
Hydroxamic acid is by far the most common ZBG
moiety in HDAC inhibitors owing to its ability to reliably chelate activeꢀsite zinc
11, 12
13, 14
ions.
Several HDAC inhibitors such as vorinostat (SAHA)
, panobinostat
1
5
16
(LBHꢀ589)
,
belinostat (PXDꢀ101) were approved by FDA for the treatment of
cutaneous T cell lymphoma (CTCL), T cell lymphoma or multiple myeloma. Among
them, LBHꢀ589, as a panꢀHDAC inhibitor, was approved by FDA for the treatment of
multiple myeloma in 2015. A number of HDAC inhibitors are currently at various
stages of clinical trials either as single agents or in combination treatment with other
chemotherapy drugs for the treatment of solid and hematologic malignancies such as
44
4
5
17
18
19, 20
46
pracinostat (SB939) , entinostat (MS275) and rocilinostat (ACYꢀ1215)
.
4
7
The class I phosphatidylinositol 3ꢀkinases (PI3Ks) are also attractive targets for
21ꢀ23
48
the treatment of cancer.
The class I PI3Ks are classed into the class IA and class
4
9
IB including phosphoinositide 3ꢀkinase α (PI3Kα), phosphoinositide 3ꢀkinase β
(PI3Kβ), phosphoinositide 3ꢀkinase δ (PI3Kδ), and phosphoinositide 3ꢀkinase δ
50
24
5
1
(PI3Kγ). Several PI3Ks inhibitors are currently in clinical trials for the treatment of
2
2
25
52
cancer such as apitolisib (GDCꢀ0980) and pictilisib (GDCꢀ0941) . However, the
efficacy of PI3Ks inhibitors is limited by concurrent activation of other survivalꢀ and
growthꢀrelated pathways. A potential strategy to overcome these limitations is to
5
3
54
ACS Paragon Plus Environment

Page 5 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
5
5
6
combine PI3Ks with other target inhibitors such as MEK inhibitors or HDAC
26
inhibitors to achieve synergistic antitumor activities.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
7
58
Figure 1. Clinically tested HDAC inhibitors.
5
9
In previous reports, Changgeng Qian group successfully achieved CUDCꢀ907 by
60
integrating HDAC inhibitory functionality (hydroxamic acid) into a core structure
27ꢀ29
6
1
scaffold (morpholinopyrimidine) shared by several PI3Ks inhibitors.
CUDCꢀ907
62
is both a potent panꢀinhibitor of HDACs and a PI3Ks inhibitor with IC50 values at low
nanomolar level. At present, CUDCꢀ907 has been in phase II clinical trials at Curis
and was assigned as orphan drug designation in the U.S.A. for the treatment of
relapsed or refractory diffuse large Bꢀcell lymphoma (DLBCL) in 2015. CUDCꢀ907 is
also in phase I clinical trials for the treatment of breast cancer and multiple
6
3
64
6
5
66
30
6
7
myeloma.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
6
6
7
7
7
7
7
7
7
7
8
9
0
1
2
3
4
5
6
7
Purine as the core scaffold structure has been applied in many drugs of PI3Ks
3
1
32
33
inhibitors, including VSꢀ5584 , ICꢀ87114 , and 3ꢀMA (Figure S1). Importantly,
34
the morpholine ring is also an important functional group in many PI3Ks inhibitors.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Encouraged by the success of Qian group, we thought of designing hybrid
pharmacophores to append HDAC recognition cap and hydroxamic acid as a
zincꢀbinding motif (Figure 2), in which morpholinopyrimidine was substituted with
morpholinopurine scaffold. We designed and synthesized
a
series of
morpholinopurineꢀbased hydroxamic acid derivatives and evaluated their inhibitory
activities on the PI3Ks and HDACs. The most potent derivatives were then subjected
to in vitro and in vivo antitumor activities. As expected, 10s and 10t with pyrimidine
at Cꢀ2 position of morpholinopurine displayed dualꢀacting PI3Kα and HDAC1
inhibitory activities. 10s showed the low IC50 values for HDAC1 and PI3Kα with an
IC50 of 1.04 nM and 1.33 nM in comparison to 1.7 nM and 19 nM of CUDCꢀ907,
78
7
9
80
30
8
1
respectively. Interestingly, when phenyl substituted groups were introduced at Cꢀ2
82
position of morpholinopurine, all these compounds still displayed excellent inhibitory
activities for HDAC1 with IC50 values ranging from 0.45 nM to 30.10 nM but lost the
activities of PI3Ks (IC50 > 1000 nM) as singleꢀtarget inhibitors of HDACs. Among
them, 10o showed better inhibitory activities for class I and class IIb isoforms of
HDACs than CUDCꢀ907. Further studies revealed that 10o showed greater in vitro
and in vivo tumor growth inhibition both in hematologic and solid malignancies than
LBHꢀ589 and SAHA. Structureꢀactivity relationship (SAR) and computer modeling
of these novel compounds were also investigated.
8
3
84
8
5
86
8
7
88
8
9
ACS Paragon Plus Environment

Page 7 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
9
0
1
Figure 2. Design of dualꢀacting HDACs and PI3Ks inhibitors.
CHEMISTRY
9
9
9
9
9
2
3
4
5
6
The general procedure to synthesize the target molecules 10a−y has been
outlined in Scheme 1. Commercially available 2,6ꢀdichloropurine (1) as the starting
material reacted with morpholine to give 2, which was treated with CH I to obtain 3.
3
Then the compound 3 was introduced an aldehyde group by reacting with DMF and
nꢀBuLi, which got the compound 4. Then the compound was treated with
97
9
8
methylamine methanol solution and NaBH
4
in sequence to produce the key
CN afforded 7. The
99
00
intermediate 5. Treatment of 5 with 6 at room temperature in CH
3
1
corresponding compounds 9a−y were obtained by 7 reacting with boronic acid
derivatives 8. Those compounds were directly converted into compounds 10a−y by
101
02
103
1
2
NH OH.
a
Scheme 1. Synthesis of the compounds 10a−y.
1
04
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
o
1
1
1
1
1
1
1
1
1
1
05
06
07
08
09
10
11
12
13
14
Reagents and conditions: (a) morpholine, CH
3
OH, 0 C to rt, 2 h; (b) CH
3
I, K
2
CO
3
,
o
o
o
CH
3
CN, rt, 3 h; (c) nꢀBuLi, TEMED, DMF, THF, ꢀ78 C to ꢀ40 C to ꢀ78 C, 6 h; (d)
o
Methylamine, rt, 2 h; NaBH
4
, 0 C, 1 h; (e) 2ꢀchloroꢀpyrimidineꢀ5ꢀcarboxylate,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DIPEA, CH CN, rt, 5 h; (f) boronic acid derivatives, PdCl (dppf) , NaHCO ,
3
2
2
3
o
Toluene/EtOH/water (v/v/v, 7/3/2), 80 C, 8 h; (g) NH
2
3
OH, DCM/CH OH (v/v, 1/1),
rt, 2 h.
In order to study the structure and relationship analysis (SAR) of cap, we also
synthesized a series of compounds with NHꢀsubstitued analogues as the cap (Scheme
2). Alternatively, compound 2 was treated with 3,4−dihydropyran to obtain 11 and
compounds 15a−l were prepared according to the same synthetic procedures reported
above. The protecting group THP was removed in HCl (con.) methanol solution to
give the important intermediates 16a−l. A series of substituted groups were introduced
to N−9 position of morpholinopurine to afford 17a−l. Those compounds were directly
115
16
117
18
119
1
1
converted to hydroxamic acid compounds 18a−l by NH OH.
2
a
Scheme 2. Synthesis of the compounds 18a−l.
ACS Paragon Plus Environment

Page 9 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1
20
21
a
Reagents and conditions: (a) DHP, pꢀTSA, DCM, rt, 2 h; (b) nꢀBuLi, TEMED, DMF,
o
o
o
o
1
22
4
THF, ꢀ78 C to ꢀ40 C to ꢀ78 C, 6 h; (c) Methylamine, rt, 2 h; NaBH , 0 C, 1 h; (d)
123
24
125
26
127
28
129
30
131
32
133
34
2ꢀchloroꢀpyrimidineꢀ5ꢀcarboxylate, DIPEA, CH
3
CN, rt, 5 h; (e) boronic acid
o
1
derivatives, PdCl
2
2
(dppf) , NaHCO
3
, Toluene/EtOH/ater (v/v/v, 7/3/2), 80 C, 8 h; (f)
o
o
HCl, CH
3
OH, 50 C, 2 h; (g) R
1
X, Cs
2
CO
3
, DMF, 80 C, 4 h; (h) NH
2
OH,
1
DCM/CH OH (v/v, 1/1), rt, 2 h.
3
BIOLOGICAL RESULTS AND DISCUSSION
1
In vitro inhibitory activities of the phenyl substituted analogues. We
primarily began our research by screening the inhibitory activities of these compounds
on the PI3Kα and HDAC1 isoform. As presented in Table 1, phenyl substituted
analogues all showed potent HDAC1 inhibition activities but poor PI3Kα inhibitory
activities (IC50 > 1000 nM). When hydroxyl as an electronꢀdonating group was
introduced to metaꢀposition of benzene, 10b reached an IC50 as low as 0.55 nM
against HDAC1. Similarly, 10c showed equivalent potency with 10b with an IC50 of
1
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
1
1
1
1
1
1
1
1
35
36
37
38
39
40
41
42
43
44
0.45 nM when methoxyl group was introduced to paraꢀposition of benzene. In
contrast, 10f decreased HDAC1 inhibition activity with an IC50 of 5.10 nM when
trifluoromethyl as an electronꢀwithdrawing group was introduced to paraꢀposition of
phenyl. These results suggest that electronꢀdonating groups improve HDAC1
inhibition activity and electronꢀwithdrawing groups decrease HDAC1 inhibition
activity. As expected, 10e, 10g, 10h and 10j all displayed decreased inhibition
activities toward HDAC1 according to the similar rule. When NꢀH of the aniline was
substituted by methyl, ethyl, and isopentene group separately, 10p, 10q and 10r
showed significant decrease in inhibitory activities with the extension of chain length
of substituent groups.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
145
46
147
48
149
50
151
52
153
To investigate the effects of chain length on HDAC1 inhibitory activity, 10k−n
were synthesized. Dramatically, the activity gradually decreased with the increase of
chain length from methyl to tꢀbutyl and IC50 ranging from 1.00 nM to 10 nM.
Interestingly, most compounds such as 10a−e, 10h−k and 10o−q all showed better
inhibitory activity of HDAC1 isoform than CUDCꢀ907, suggesting that
morpholinopurine as a substituted core structure scaffold showed better HDAC1
activity than morpholinopyrimidine.
1
1
1
1
Table 1. IC50 values for enzymatic inhibition of HDAC1 and PI3Kα.
1
54
ACS Paragon Plus Environment

Page 11 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
55
a
a
Compd Aryl
IC50 (nM)
Compd Aryl
IC50 (nM)
HDAC1 PI3Kα
HDAC1 PI3Kα
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
0a
Cl
1.10
0.55
>10000 10k
1.00
1.98
>10000
>10000
10b
>10000 10l
10c
0.45
1.02
>10000 10m
>10000 10n
6.15
>10000
>10000
10d
10.01
10e
1.59
5.10
30.10
1.01
0.97
0.85
>10000 10o
>10000 10p
>10000 10q
>10000 10r
0.65
0.81
1.10
10.00
11
>1000
>10000
>10000
>10000
>10000
19
10f
1
0g
0h
1
10i
>1000
>1000
SAHA
10j
CUDCꢀ907
1.7
a
1
157
58
159
60
56
IC50 values for enzymatic inhibition of HDAC1 and PI3Kα. The IC50 values are the
means of at least two experiments, with intraꢀ and interꢀassay variations of < 10%.
In vitro inhibition activities of sixꢀmembered Nꢀcontaining heterocyclic
analogues. When sixꢀmembered Nꢀcontaining heterocyclic was introduced to instead
of phenyl at Cꢀ2 position of morpholinopurine, we synthesized compounds 10s−y.
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
1
1
1
1
1
1
1
1
61
62
63
64
65
66
67
68
69
70
The enzyme inhibitory activities on HDAC1 and PI3Kα were evaluated (Table 2).
10s−y all showed excellent inhibitory activity toward HDAC1 with nanomolar IC50
values, but inhibitory activity for PI3Kα differed obviously. 10s and 10t exhibited
potent inhibitory activity both on HDAC1 and PI3Kα when pyrimidine was
introduced to the Cꢀ2 position of morpholinopurine. 10s with amino group at
paraꢀposition of the pyrimidine displayed stronger inhibitory activities on HDAC1
and PI3Kα with respective IC50 values of 1.04 nM and 1.33 nM in comparison to 1.7
nM and 19 nM of CUDCꢀ907. 10t without amino group displayed slightly lower
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
activity on PI3Kα with an IC of 28.06 nM. In contrast, pyridineꢀbased analogues
50
10u−y dramatically decreased PI3Kα inhibition activity above micromolar level.
These results suggested that aminopyrimidine plays an important role to maintain the
dualꢀtarget inhibitory activities on HDAC1 and PI3Kα.
171
72
173
1
Table 2. IC50 values for enzymatic inhibition of HDAC1 and PI3Kα.
1
74
a
Compd
Aryl
IC50 (nM)
HDAC1
PI3Kα
1
1
1
0s
1.04
1.33
0t
1.14
1.11
28.06
>1000
0u
ACS Paragon Plus Environment

Page 13 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
0v
1.60
>1000
1
1
1
0w
0x
0y
N
1.00
0.49
0.55
11
>1000
>1000
>1000
>10000
19
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
N
N
SAHA
CUDCꢀ907
1.7
a
1
1
75
76
IC50 values for enzymatic inhibition of HDAC1 and PI3Kα. The IC50 values are the
means of at least two experiments, with intraꢀ and interꢀassay variations of < 10%.
In vitro inhibition activities of NꢀH substituted analogues. To understand the
structureꢀactivity relationship of NꢀH substituted analogues, we synthesized 18a−l to
probe the effects of different substitutions on inhibitory potency of HDAC1 and
177
78
179
80
181
82
183
84
185
86
187
1
1
PI3Kα. As shown in Table 3, when R
1
changed from methyl to ethyl, the potency on
changed from ethyl to isopropyl and
HDAC1 was retained. However, when R
1
1
cyclopentyl, the potency on HDAC1 decreased significantly, the inhibitory activity of
18a on HDAC1 was 100ꢀfold over 18b and 18c. The similar results were also
observed in 18d−f, 18g−i, and 18j−l. These results indicated that methyl was crucial
to contribute to improving the inhibition activities on HDAC1 and PI3Kα. However,
all these compounds did not improve the potency on PI3Kα in comparison with 10s.
Table 3. In vitro HDAC1 and PI3Kα inhibitory activities.
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
88
a
a
Compd Aryl
R
1
IC50 (nM)
Compd Aryl
R
1
IC50 (nM)
HDAC1 PI3Kα
HDAC1 PI3Kα
1
1
1
1
0c
8a
8b
8c
Me
0.45
1.00
110
125
0.65
1.37
165
190
11
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
10s
18g
18h
18i
Me
1.04
2.62
240
300
1.00
1.60
62
1.33
28.42
>1000
>1000
>1000
>1000
>1000
>1000
19
10o
Me
10w
18j
18k
18l
Me
1
1
1
8d
8e
8f
67
SAHA
>10000 CUDCꢀ907
1.7
a
1
1
1
1
1
1
89
90
91
92
93
94
IC50 values for enzymatic inhibition of HDAC1 and PI3Kα. The IC50 values are the
means of at least two experiments, with intraꢀ and interꢀassay variations of < 10%.
In vitro cell growth inhibitory effects in MV4ꢀ11, A2780s, and HCT116 cells.
After the synthesis and evaluation for enzymatic activity assays of these compounds,
antiꢀproliferative effects of these compounds against MV4ꢀ11 cells, A2780s cells, and
HCT116 cells were tested using MTT assay. SAHA, LBHꢀ589, and CUDCꢀ907 were
used as references (Table 4). A total of 29 compounds showed strong
195
ACS Paragon Plus Environment

Page 15 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
2
2
2
2
2
2
96
97
98
99
00
01
02
03
04
05
antiꢀproliferative activity with most IC50 values less than 100 nM. Among them, 10o,
10p, 10q, 10w and 10y showed comparable inhibitory activities to CUDCꢀ907 and
LBHꢀ589, but much stronger than SAHA against all three types of cancer cells.
Although 10s was the same as CUDCꢀ907 to be a dualꢀtarget inhibitor, its inhibitory
activity on tumor cells was inferior to CUDCꢀ907.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For the series of compounds 10a−y, compounds with aromatic rings substituted
at the Cꢀ2 position of the purine ring (10b−y) were generally more potent inhibitors
than the chlorine substituted (10a). Proper substitution of Cꢀ4 position amino of 10o,
such as double methyl (10p) and single ethyl (10q), caused a little decrease of the
antiꢀproliferative activities than the nonꢀsubstituted (10o); while double dimethylallyl
substituted analogue of amino (10r) resulted in a sharp decrease in potency over 10o.
As the substitution group length at Cꢀ4 position of benzene increased, from methyl to
tꢀbutyl (10k−n), the compounds suffered a sharp decrease in antiꢀproliferative activity.
In corporation of trifluoromethyl group of phenyl ring (10f−h), the antiꢀproliferative
activities decreased.
206
07
208
09
210
11
212
2
2
2
Table 4. Cell growth inhibitory effects for all compounds, SAHA, LBHꢀ589 and
CUDCꢀ907.
a
Compd
IC50 ± SEM (nM)
MV4ꢀ11
A2780s
HCT116
10a
37.39±6.32
7.65±1.08
4.80±1.43
85.73±12.16
13.65±0.49
66.55±2.86
13.02±1.32
14.84±5.47
1
1
0b
0c
31.81±3.08
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
0d
0.95±0.07
2.69±0.62
3.67±0.32
39.31±3.56
224±12
15.38±1.58
54.35±5.18
10.63±1.12
89.32±5.16
381±25
10e
1
1
0f
45.78±1.58
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0g
65.04±4.45
148±12
560±16
10h
300±6
1
1
1
0i
4.09±0.71
1.04±0.09
3.32±2.94
30.80±3.20
35.30±3.20
160±12
18.48±0.10
12.53±1.58
11.30±1.29
3.74±0.59
52.06±8.37
286±25
9.25±1.08
7.85±2.10
5.77±1.20
7.05±0.32
57.63±6.52
160±16
0j
0k
10l
1
1
0m
0n
10o
0.15±0.02
0.89±0.10
2.51±0.50
44.77±13.27
7.65±0.79
59.8±3.20
4.00±0.41
7.46±0.72
0.69±0.05
1.15±0.28
0.19±0.05
78.82±12.16
4.85±0.87
6.98±1.82
2.36±0.89
9.17±3.56
127.10±12.40
117.60±3.42
75.08±4.93
181±22
0.74±0.10
4.73±1.07
10.26±1.27
76.46±9.83
214.20±16.35
46.66±3.20
40.24±7.04
9.22±0.34
16.13±2.30
13.99±2.35
1.99±0.30
241.10±25.72
27.65±2.39
1
1
1
0p
0q
0r
10s
1
1
0t
0u
10v
129.50±8.15
12.94±1.34
51.69±0.51
90.63±1.53
107.60±24.73
30.27±0.18
1
1
1
0w
0x
0y
18a
8d
1
ACS Paragon Plus Environment

Page 17 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
8g
14.63±1.42
240±15
317±20
18j
8.85±0.87
230±24.2
0.43±0.09
2.97±0.11
37.28±0.18
870±68.4
6.15±1.02
8.32±2.33
17.95±2.09
1450±259.7
7.34±1.98
3.36±0.36
SAHA
CUDCꢀ907
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
LBHꢀ589
a
2
2
2
2
2
2
2
2
2
2
2
2
2
13
14
15
16
17
18
19
20
21
22
23
24
25
IC50 = compound concentration required to inhibit tumor cell proliferation by 50%;
data are expressed as the mean ± SEM from the doseꢀresponse curves of at least three
independent experiments.
In vitro inhibition of the enzymatic activities of HDACs and PI3Ks. In our
progress of research, we found that although 10b, 10c, 10d, 10i−k, 10o−q, 10v, and
10w−y all showed excellent HDAC inhibitory effects and antiꢀproliferative activities
on tumor cells, most of these compounds had poor solubility in several solvents even
dimethyl sulfoxide. Their disadvantages limited further dose administration in vivo
evaluation. Considering the solubility and suitable administration as properties
required for therapeutic application, we chose 10c, 10o, 10w, 10s, and 10y to
ascertain the potency of selected compounds across the broader family of HDACs and
PI3Ks isoforms, CUDCꢀ907 and LBHꢀ589 were also chosen as positive controls. The
data were shown in Tables 5 and 6. As described previously, 10s displayed the most
potent inhibition activities toward PI3Ks, as well as high potency toward HDACs as a
dualꢀacting inhibitor. 10c, 10o, 10w and 10y all displayed remarkable inhibitory
activities toward class I (HDACs 1ꢀ3, 8) and class IIb (HDACs 6, 10) isoforms of
HDACs. In comparison to LBHꢀ589 and CUDCꢀ907, most of these compounds
showed better isoforms selectivity for class I and class IIb isoforms over class IIa
226
27
228
29
230
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
2
2
31
32
33
34
(HDACs 4, 5, 7, 9), suggesting that 10c, 10o, 10w and 10y are all class I and class IIb
isoform inhibitors. Among them, only 10s was identified as a dualꢀacting inhibitor
toward PI3Ks and HDACs.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 5. The enzymatic activities for HDACs of the selected compounds.
a
Classes of
IC50 (nM)
isoforms
HDACs
class I
10c
0.45
3.90
2.55
92.92
10o
10w
10s
10y CUDCꢀ907 LBHꢀ589
HDAC1
HDAC2
HDAC3
HDAC8
0.65
1.00
1.04
0.55
2.80
2.70
11
1.7
5.0
1.8
191
409
674
426
554
27
1.26
3.28
2.90 11.46 2.62
3.40
24
4.65
217
3.29
22
2.27
4.86
HDAC4 >1000 >1000 >1000 653 >1000
HDAC5 >1000 750 >1000 484 >1000
HDAC7 >1000 >1000 >1000 650 895
HDAC9 >1000 >1000 >1000 >1000 >1000
337.8
190.3
4354
887.8
4.16
class IIa
class IIb
HDAC6
7.78
6.85
6.40 18.22
22
2.70
0.86
HDAC10
2.80 17.55 2.83
2.8
5.4
4.45
class IV HDAC11 >10000 >1000 >1000 >1000 835
4112
a
2
2
2
35
36
37
IC50 values for enzymatic inhibition of HDACs. The IC50 values are the means of at
least two experiments, with intraꢀ and inter assay variations of < 10%.
Table 6. The enzymatic activities for class I PI3Ks of the selected compounds.
a
IC50 (nM)
Class I
10c
10o
10w
10s
10y
CUDCꢀ907 LBHꢀ589
ACS Paragon Plus Environment

Page 19 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
PI3Kα
PI3Kβ
PI3Kγ
PI3Kδ
>10000
>1000
>1000 1.33 >1000
19
54
>10000
>10000
>10000
>10000
>10000 >10000 >1000 34 >10000
>10000 >10000
>10000 >10000
255
493
8.10 >10000
15 >10000
311
39
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
238
39
240
41
242
43
244
45
246
47
248
49
250
51
252
53
254
55
IC values for enzymatic inhibition of PI3Ks. The IC values are the means of at
50 50
2
least two experiments, with intraꢀ and inter assay variations of <10%.
In order to investigate whether the HDAC1 IC50 value of selected compounds is
likely close to its actual IC50, we performed a HDAC enzyme activity assay utilizing
nuclear extracts from the cell line of HeLa. As shown in Table 7, 10c, 10o, 10s, 10w,
and 10y showed the similar inhibition to HDACs to CUDCꢀ907 but superior to SAHA,
the IC50 values of 10c, 10o, 10s, 10w, and 10y on HDAC were 3.4, 3.2, 3.3, 6.1, 6.3
nM, respectively. LBHꢀ589 exhibited the most inhibition on HDACs among these
compounds.
2
2
2
2
Tubulin acetylation and histone H3 acetylation induction of selected
compounds. Cytoblot experiments were performed to further evaluate the histone H3
acetylation (AcꢀH3) and tubulin acetylation (AcꢀTub) of selected compounds. As
exhibited in Table 7, 10s showed comparable AcꢀH3 and AcꢀTub induction in cellular
assays to CUDCꢀ907, which were in line with the in vitro enzymatic data. 10c, 10o,
10w and 10y all showed the similar activities to LBHꢀ589 but better activities than
SAHA in tubulin acetylation and histone H3 acetylation induction.
2
2
2
Table 7. Tubulin acetylation and histone H3 acetylation induction and against HeLa
cells of selected compounds, LBHꢀ589, CUDCꢀ907, and SAHA.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
b
Compd
IC50 (nM)
EC50 (nM)
HeLa
AcꢀTub
AcꢀH3
1
0c
0o
3.40±0.1
3.20±0.1
3.30±0.8
6.10±1.0
6.30±2.3
0.83±0.1
6.80±0.4
11.5±0.7
248.35±152.52
211.05±37.69
143.7±46.53
198.6±17.11
187.9±10.89
150.71±111.01
221.75±52.40
399.25±34.29
98.58±13.60
112.96±46.58
123.2±22.77
117.65±5.44
142.01±105.92
169.5±4.67
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
10s
0w
1
10y
LBHꢀ589
CUDCꢀ907
SAHA
126.25±4.03
377.48±23.16
a
b
2
2
2
2
2
2
2
2
2
2
2
2
2
56
57
58
59
60
61
62
63
64
65
66
67
68
IC50 values of HeLa Nuclear Extract HDAC inhibition. EC50 values of tubulin
acetylation and AcꢀH3 are based on cytoblot experiments run in duplicate in A2780s
cells. Data are expressed as the mean ± SEM from the dose−response curves of at
least three independent experiments.
In order to confirm whether compound 10s is a dualꢀacting selective inhibitor
toward PI3Ks and HDACs and 10c, 10o, 10w and 10y are all class I and class IIb
isoform inhibitors, we selected two compounds 10s and 10o to examine the protein
kinases inhibitory profile against 100 selected recombinant human protein kinases.
As shown in Tables 6 and 7, compound 10s exhibited excellent inhibitory activities
for HDACs and PI3Ks, the IC50 values of 10s on HDAC1, PI3Kα, PI3Kβ, PI3Kγ and
PI3Kδ were 1.04, 1.33, 34, 8.10 and 15 nM, respectively. In contrast, the IC50 values
of CUDCꢀ907 on HDAC1, PI3Kα, PI3Kβ, PI3Kγ and PI3Kδ were 1.7, 19, 54, 311
and 39 nM, respectively. 10s showed weak inhibitory activity against cKit, cSRC (h)
ACS Paragon Plus Environment

Page 21 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
2
2
2
2
2
2
2
2
2
69
70
71
72
73
74
75
76
77
78
and Src (T341M) (the binding affinities were 38, 56 and 39 at 1 ꢁM). However,
compound 10s showed almost no inhibitory activity against the other selected protein
kinases (Table S1). These data demonstrates that 10s is a dualꢀacting selective
inhibitor toward PI3Ks and HDACs. Compound 10o only showed strong inhibitory
activity against HDACs but had no inhibitory activity against the 100 selected
recombinant human protein kinases, demonstrating that compound 10o is a
singleꢀtarget class I and class IIb HDAC inhibitor.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To further confirm the selectivity of 10s and 10o on PI3Ks and HDACs, The
effects of 10s and 10o on the activities of mTOR, PI3K and Akt in HCT116 cells
treated with indicated concentrations for 6 h were investigated by western blot
analysis (Figure S2). 10s reduced the levels of mTOR, PI3K and Akt
phosphorylations in a concentration dependent manner. However, the levels of
phosphorylation of mTOR, PI3K and Akt were not obviously altered by 10o. The data
further demonstrates that 10s is a dualꢀacting selectivity inhibitor toward PI3Ks and
HDACs and 10o is a singleꢀtarget class I and class IIb HDAC inhibitor.
279
80
281
82
283
84
285
86
287
88
289
90
2
2
2
Molecular Docking Study. To better understand the different activities of 10s
and 10o on PI3Kα, the molecular docking was performed (Figure 3A and Figure S3).
From the docking results, we found that the oxygen of the morpholine group of 10s
forms a hydrogen bond interaction with the backbone NH group of Valꢀ851 in the
hinge region, the morpholine ring has the hydrophobic contacts with the residues of
Valꢀ851, Metꢀ922, Ileꢀ932, and the purine core forms hydrophobic interactions with
the side chains of Metꢀ772 and Ileꢀ932. These interactions are considered as critical
2
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
2
2
2
2
2
2
2
3
91
92
93
94
95
96
97
98
99
00
forces which determine the binding capacity of PI3Kα inhibitors. And the tail of
hydroxamic acid could interact with Serꢀ854 on the entrance of the ATP pocket. The
pyrimidine ring linked with hydroxamic acid could interact with the indole ring of
Trpꢀ780 by means of πꢀπ stacking. More importantly, the amino group on the pyridine
ring of 10s forms two hydrogen bonds to the residues of Aspꢀ810 and Aspꢀ933, while
the pyrimidine nitrogen atoms form additional hydrogen interactions with the
phenolic hydroxyl of Tyrꢀ836 and the catalytic lysine Lysꢀ802. In contrast, due to the
replacement of pyrimidine ring, the aniline of 10o could not form additional hydrogen
interactions with the residues of Tyrꢀ836 and Lysꢀ802. And, with the stronger
hydrogen bonding capabilities, compound 10s may have higher inhibitory activity
against PI3Kα than compound 10o. In according with the enzymatic inhibition level
described previously, 10s could inhibit PI3Kα efficiently with IC50 value of 1.33 nM,
but 10o showed weak inhibitory activity on PI3Kα (IC50 > 1000 nM).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
301
02
303
04
305
06
307
08
309
10
3
3
The molecular docking with compound 10s and 10o in HDAC1 was also
2
+
performed (Figure 3B and Figure S4). The catalytic Zn is coordinated by the
hydroxamic acid, Aspꢀ264, Aspꢀ176 and Hisꢀ178. Also, the hydroxamic acid tail
forms hydrogen interaction with the residues of Hisꢀ140 and Tyrꢀ303. In the cap
region, the purine ring forms a πꢀπ stacking interaction with residues of Pheꢀ150 and
Pheꢀ205. And the nitrogen atoms, linking the pyrimidine ring in 10s and benzene ring
in 10o, form hydrogen interaction with Glnꢀ26, respectively.
3
3
3
ACS Paragon Plus Environment

Page 23 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
3
11
12
Figure 3. (A) Proposed binding mode of compound 10s (green stick) and 10o (yellow
stick) within the ATP pocket of PI3Kα (PDB code 4L23). And the dash lines represent
the Hꢀbond interaction between compound 10s and PI3Kα. (B) Representation of the
predicted binding modes of 10s (yellow stick) and 10o (green stick) in the active site
of HDAC1 (PDB code 4BKX), yellow dash lines indicate the Hꢀbond interaction
between 10o and HDAC1. For clarity, oxygen and nitrogen atoms are shown in red
and blue respectively.
3
3
13
14
315
16
317
18
319
20
321
22
323
24
3
3
Antiꢀtumor activity in vivo. Since 10s, 10c, 10o, 10w and 10y all exhibited
good in vitro antiꢀproliferative activities, we primarily established a MV4ꢀ11
xenograft NOD/SCID mouse model to investigate the effects of 10c, 10s, 10w, 10y,
and 10o on tumor growth, with SAHA administrated for comparison. The
administration, dosing schedules, and results are presented in Table 8.
Table 8. Summary of tumor growth inhibition of selected compounds.
administration
3
3
3
max body
weight
tumor
statistical
survivors
Tumor
compd
mass
b
dose
significance
(day)
a
schedule
route
loss(%)(day)
change(%)
(
mg/kg)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
0c
10s
0w
Q2D×11
Q2D×11
Q2D×11
Q2D×11
Q2D×11
Q2D×11
Q2D×8
Q2D×8
Q2D×8
Q2D×8
Q2D×6
Q2D×6
Q2D×6
Q2D×6
Q2D×6
10
10
10
10
10
50
2.5
5
iv
iv
iv
iv
iv
ip
iv
iv
iv
ip
iv
ip
iv
iv
iv
ip
***
**
8.1(7)
6.4(13)
14.9(11)
2.1(19)
3.6(19)
3.0(17)
3.9(4)
4/6
4/6
5/6
5/6
6/6
6/6
6/6
6/6
6/6
6/6
8/8
5/8
7/7
7/7
7/7
7/7
65.3
45.1
63.4
64.3
65.6
0
1
***
***
***
ns
MV4ꢀ11
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0y
0o
1
SAHA
1
1
0o
0o
***
***
***
**
49.4
53.3
68.2
32.4
83.9
16.6
45.1
69.2
75.1
49.2
2.1(12)
1.4(10)
5.2(8)
HCT116
Ramos
10o
SAHA
10o
10
50
10
10
2.5
5
***
***
***
***
***
***
0.8(12)
1.9(7)
LBHꢀ589
1
1
1
0o
0o
0o
2.4(12)
3.1(9)
MM1S
10
10
2.1(12)
5.1(12)
LBHꢀ589
Q2D×6
a
b
3
3
3
25
26
27
Q2D, every 2 days. Oneꢀway ANOVA followed by Dunnett's multiple comparison
test was used to determine the statistical significance of tumor volume between a
treatment group and the control (vehicle) group: ns=not significant, (**) p<0.01, and
5
3
3
28
29
(***) p<0.001.
As displayed in Table 8, 10c, 10w, 10y, and 10o all exhibited antitumor
activities. The tumor growth on MV4ꢀ11 was significantly suppressed by 65.3, 63.4,
64.3 and 65.6% (tumor growth inhibitory rate [TGI], as indicated the percentage of
tumor mass change values) after intravenous administration of 10c, 10w, 10y, and 10o
330
331
3
32
ACS Paragon Plus Environment

Page 25 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
333
at 10 mg/kg, respectively. In contrast, SAHA had no inhibitory activity at 50 mg/kg ip
administration on the MV4ꢀ11 xenograft model. 10s showed the worst inhibitory
effect with the TGI of 45.1% which is in consistent with in vitro potency. In addition,
10c and 10s exhibited some toxicity and two of mice died during the treatment period.
At the same dose, 10w and 10y both retained potent inhibition on MV4ꢀ11 but both
still caused one mouse death. 10o could inhibit tumor growth with TGI of 65.6%.
Among the five compounds, 10o exhibited the most potent antitumor activity without
significant toxicity: slight body weight change and no death of mice were observed
during the treatment period. Furthermore, 10o showed the best waterꢀsolubility among
the five compounds. Hence, 10o was selected for further in vitro and in vivo
antitumor evaluation in solid tumor and hematological models.
3
3
3
34
35
36
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
337
3
3
3
3
38
39
40
41
342
3
3
3
43
44
45
We then established HCT116 xenograft model since this model had been
35
36
successfully used for HDAC inhibitors evaluation such as dacinostat and belinostat .
As displayed in Table 8 and Figure 4A, iv administration of 10o doseꢀdependently
inhibited tumor growth, the significant antitumor activity was observed at dose as low
as 2.5 mg/kg (TGI = 49.4%, p<0.001), 5 and 10 mg/kg of 10o caused respective TGI
of 53.3% and 68.2%. In contrast, SAHA only showed 32.4% of TGI at 50 mg/kg ip
administration. In addition, 10o did not cause significant weight loss in HCT116
xenograft model during the treatment period.
346
3
3
3
3
47
48
49
50
351
3
3
3
52
53
54
In vitro and in vivo antiꢀproliferative activities of 10o in hematological tumor
cells.
As most HDAC inhibitors are used for hematological cancer treatment, 10o was
further investigated the antiꢀproliferative activities against 8 kinds of hematological
tumors including multiple myeloma (U266, RPMIꢀ8226 and MM1S cells) and B cell
lymphoma (OCIꢀLY1, HBLꢀ1, Ramos, Raji and SUDHL4 cells) by MTT, and
355
3
3
56
57
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
358
LBHꢀ589 was chosen as a positive control (Table S2). 10o showed better or
comparable antiꢀproliferative potential to LBHꢀ589, and the IC50 values were ranging
from 0.78 to 12.85 nM in these tumor cell lines.
3
3
59
60
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
61
It was recently announced that LBHꢀ589 had been approved by FDA for the
treatment of multiple myeloma as a new HDAC panꢀinhibitor. LBHꢀ589 has also been
licensed or at various stages of clinical evaluation for the treatment of B cell
lymphoma. We further established a multiple myeloma MM1S and a B cell lymphoma
Ramos NOD/SCID xenograft models to validate the antitumor effect of 10o (Table 8，
Figure 4B and 4C). In MM1S tumor model, 10o doseꢀdependently inhibited tumor
growth, lowꢀdose administration of 10o (2.5 mg/kg iv) could effectively inhibit tumor
growth and the TGI was 45.1%, which was comparable to LBHꢀ589 at 10 mg/kg ip
treatment (TGI, 49.2%). However, 10o at 5 mg/kg iv treatment showed much more
efficacious in inhibiting tumor growth than LBHꢀ589 and the TGI was up to 69.2%.
Further increased dose of 10 mg/kg of 10o caused the TGI to 75.1%. In addition, there
is no significant weight loss of NOD/SCID mice during the treatment period.
Surprisingly, in B cell lymphoma Ramos NOD/SCID xenograft model, 10o at 10
mg/kg iv treatment caused significantly tumor growth inhibition, the TGI was 83.89%.
In contrast, LBHꢀ589 was not sensitive to this model and had no activity (TGI,
16.6%), and three of eight mice died during experimental period. These results
confirmed that 10o showed superior antitumor to LBHꢀ589 and could be used as a
novel HDAC inhibitor to be further researched on the therapy of both hematological
tumors and solid tumors.
362
63
364
65
366
67
368
69
370
71
372
73
374
75
376
3
3
3
3
3
3
3
3
3
3
77
78
79
ACS Paragon Plus Environment

Page 27 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
3
80
81
Figure 4. Antitumor activities of 10o, SAHA and LBHꢀ589 on the xenograft models.
(A) Antitumor growth of 10o on the HCT116 xenograft model. (B) Antitumor
growth of 10o on Ramos xenograft model. (C) Antitumor growth of 10o on MM1S
xenograft model. 10o was dissolved in physiological saline containing 20% captisol
with the pH adjusted to 6). SAHA (50 mg/kg) was dissolved in physiological saline
containing 5% DMSO and 30% PEG400. LBHꢀ589 was dissolved in physiological
saline containing 2% DMSO, 2% Tween 80 and 15% PEG400.
3
3
3
3
3
3
3
3
3
3
82
83
84
85
86
87
88
89
90
91
Formation of salts is a wellꢀknown technique to modify and optimize the
solubility to improve the pharmacokinetics and increase the bioavailability. As
depicted in Table S3, the oral bioavailability of 10o was 4.83%, which is comparable
37
to LBHꢀ589 (4.62%, Female Balb/c nude mice). However, the oral bioavailability of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
3
3
3
3
3
3
3
4
4
92
93
94
95
96
97
98
99
00
01
10o·CH
3
SO
3
H in SD rats was increased to 14.21%, much higher than LBHꢀ589.
3 3
Encouragingly, the oral bioavailability of 10o·CH SO H in beagle dogs was up to
41.82%. The low clearance, rapid elimination halfꢀlife and good bioavailability
indicated that 10o may be suitable both for oral and intravenous dosing for further
development as an anticancer agent.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Upregulation effects of histone H3 and αꢀtubulin. We attempted to validate the
observed biochemical potencies of the 10o in a cellular context by examining its
HDACs activity in comparison with SAHA and LBHꢀ589. The effects of 10o on the
acetylation level of histone H3 (a known substrate for HDACs 1ꢀ3) and αꢀtubulin (a
38
known substrate for HDAC6), the biomarkers of HDAC inhibition , in MV4ꢀ11 and
402
03
404
05
406
07
408
09
410
HCT116 cells treated with different concentrations are shown in Figure 5. As
expected, in agreement with the relative potency in class I (HDAC 1ꢀ3, 8) inhibition,
10o induced a concentrationꢀdependent increase of the acetylated histone H3 from
0.02 ꢁM to 2 ꢁM, which is comparable to LBHꢀ589, however, SAHA only slightly
upregulated the acetylated histone H3 at concentration of 2 ꢁM. Furthermore, both
10o and LBHꢀ589 upregulated acetylated αꢀtubulin in a concentrationꢀdependent
manner, but SAHA induced acetylated αꢀtubulin increase at slightly higher
concentration. These results showed that 10o was a class I and class IIb isforms
HDAC inhibitor.
4
4
4
4
ACS Paragon Plus Environment

Page 29 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
411
4
4
4
4
4
4
4
4
4
4
12
13
14
15
16
17
18
19
20
21
Figure 5. Western blot analysis of acetylated αꢀtubulin, acetylated histone H3 in
MV4ꢀ11 and HCT116 cell lines after 6 h treatment with compounds 10o, LBHꢀ589 at
0.02, 0.2, 2 ꢁM and SAHA at 0.2, 2 ꢁM. GAPDH was used as a loading control.
CONCLUSIONS
We chose morpholinopurine as a core structure scaffold of cap group, and a
series of hydroxamic acid analogues were prepared and evaluated for bioactivity in
vitro and in vivo. The inhibitory activities on the HDAC1 and PI3Kα isoforms of the
synthesized compounds and a SAR analysis revealed that the strategy successfully
achieved the target to find novel and highly efficient small molecule inhibitors of
HDACs. The most potent dualꢀtarget inhibitor 10s displayed efficient activities for
ACS Paragon Plus Environment