Design and synthesis of dual-action inhibitors targeting histone deacetylases and 3-hydroxy-3-methylglutaryl coenzyme a reductase for cancer treatment

Article abstract of DOI:10.1021/jm400179b

A series of dual-action compounds were designed to target histone deacetylase (HDAC) and 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGR) by having a hydroxamate group essential for chelation with the zinc ion in the active site of HDAC and the key structural elements of statin for binding with both proteins. In our study, the statin hydroxamic acids prepared by a fused strategy are most promising in cancer treatments. These compounds showed potent inhibitory activities against HDACs and HMGR with IC₅₀ values in the nanomolar range. These compounds also effectively reduced the HMGR activity as well as promoted the acetylations of histone and tubulin in cancer cells, but were not toxic to normal cells.

Full text of DOI:10.1021/jm400179b

Subscriber access provided by BALL STATE UNIV
Article
Design and Synthesis of Dual-action Inhibitors Targeting Histone
Deacetylases and HMG-CoA Reductase for Cancer Treatment
Jhih-Bin Chen, Ting-Rong Chern, Tzu-Tang Wei, Ching-Chow Chen, Jung-Hsin Lin, and Jim-Min Fang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm400179b • Publication Date (Web): 09 Apr 2013
Downloaded from http://pubs.acs.org on April 15, 2013
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 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design and Synthesis of Dual-action Inhibitors Targeting Histone Deacetylases and
HMG-CoA Reductase for Cancer Treatment
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
†,#
‡,#
¶,#
¶
Jhih-Bin Chen, Ting-Rong Chern, Tzu-Tang Wei, Ching-Chow Chen, Jung-Hsin
,
‡
,†
Lin,* and Jim-Min Fang*
†
Department of Chemistry, National Taiwan University, Taipei 106, Taiwan.
School of Pharmacy, National Taiwan University, Taipei 100, Taiwan.
‡
¶
Department of Pharmacology, College of Medicine, National Taiwan University, Taipei 100,
Taiwan.
#
These authors contributed equally to this work.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 50
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
ABSTRACT
A series of dual-action compounds were designed to target histone deacetylase (HDAC) and
-hydroxy-3-methylglutaryl coenzyme A reductase (HMGR) by having a hydroxamate group
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
essential for chelation with the zinc ion in the active site of HDAC and the key structural
elements of statin for binding with both proteins. In our study, the statin hydroxamic acids
prepared by fused strategy are most promising in cancer treatments. These compounds
showed potent inhibitory activities against HDACs and HMGR with IC values in nanomolar
50
range. These compounds also effectively reduced the HMGR activity as well as promoted the
acetylations of histone and tubulin in cancer cells, but were not toxic to normal cells.
2
ACS Paragon Plus Environment

Page 3 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Cancer cells may be developed from inherited defects or acquired damages of DNA. Cancer is
a highly complex multi-genetic disease involving multiple cross-talks between signaling
networks. Using drug-cocktails that combine multiple anti-cancer agents working in different
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
–4
mechanisms has been a standard treatment of cancers to avoid drug resistance.
Histone acetyltransferases (HATs) and histone deacetylases (HDACs) have recently
emerged as important targets for cancer therapy. Acetylation of the lysine residues on histone
H3 and H4 leads to a loose and active chromatin, which allows access of various transcription
factors to the promoters of target genes. In contrast, deacetylation of the lysine residues
5
results in a highly compact and transcriptionally inactive chromatin. The levels of histone
6
acetylation and deacetylation are regulated by HATs and HDACs, respectively. HDAC
overexpression has been found in a variety of human cancers, including myeloid neoplasia
7
and solid tumors. The association of HDACs with oncogenic DNA-binding fusion proteins
and other repressive transcription factors constitutively suppresses specific tumor suppressor
8
genes. Therefore, HDACs represent a rational target for cancer treatment. Several HDAC
inhibitors (HDACi’s) are currently under clinical trials on either monotherapy or combination
9
, 10
therapy for cancer treatment.
HDACi’s are categorized into four groups: short-chain fatty
8
acids, hydroxamic acids, cyclic tetrapeptides, and benzamides. Among them, the
hydroxamate-containing HDACi’s trichostatin A (TSA) and suberoylanilide hydroxamic acid
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 50
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
(
SAHA, Figure 1A) exhibit the most potent efficacy.
In another aspect, statins have recently been shown to be effective for cancer prevention
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
11
in observational, preclinical, and certain randomized controlled studies. Statins, such as
lovastatin (Figure 1B) and atorvastatin, are known to reduce serum cholesterol levels through
competitive inhibition at 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGR). The
HMGR inhibitors (HMGRi’s) are effectively used to decrease the incidence of cardiovascular
1
2
and cerebrovascular disorders, and to prevent cardiovascular disease (CVD). Statins possess
an established record of safety and efficacy in human CVD prevention.
It has been reported that the combination use of anticancer agents with statins may reduce
1
3
side effects to attain better treatment of cancers. Furthermore, the in vitro experiment using
combination of HDACi and HMGRi has shown the synergistic induction of apoptosis of
1
5
HeLa cells.
The underlying synergistic mechanism has been proposed that the
down-regulation of geranylgeranyltransferase (GGTase)-I β subunit, caused by HDACi (TSA
14
in that study), enhances the depletion of mevastatin-induced geranylgeranylated RhoA.
15
However, the direct HDAC inhibition by statins could have also made significant
contributions to this synergism.
Given the aforementioned evidence, we conceived that concurrent inhibition of HDAC
and HMGR would be a promising approach for cancer treatment. However, using
multi-component drug-cocktails for therapeutics has some drawbacks, such as complex
4
ACS Paragon Plus Environment

Page 5 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pharmacokinetics, unpredictable drug–drug interaction, and formulation problems due to
16, 17
different solubilities of individual drugs.
Alternatively, to design a single compound that
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
simultaneously modulates multiple targets, dubbed designed multiple ligand (DML), has
16, 18
become an emerging paradigm for drug discovery.
DMLs constructed by incorporation of
1
9, 20
HDACi’s into other active agents targeting inosine monophosphate dehydrogenase,
2
1
22, 23
24
nuclear vitamin D receptor, tyrosine kinase receptor
or topoisomerase II
have been
25
tested in cancer treatments.
In this study, we designed the dual-action anticancer agents 6a, 6b, 10, and 12–15 by the
knowledge- and structure-based approaches to target both HDAC and HMGR (Figure
1
7, 26
1C–E).
The rationale for the dual-inhibitor design was originated from our previous
1
6
modeling work, which showed that lovastatin could fit the adjacent transient pocket about 4
Å away from the active site of HDAC. We first designed the HMGR–HDAC dual inhibitors
27
6
a,b by direct connection of lovastatin with a triazole-linked SAHA (Figure 1C). According
28
to the structural information, the hydroxamate group plays an essential role in chelation of
zinc ion for HDAC inhibition, whereas the hydrophobic moiety of SAHA can be replaced by
lovastatin fragment to accommodate with the deep pocket of HDAC. In the meantime,
29
compounds 6a,b still retain the HMG-like structure to assure their potency to HMGR. These
conjugated molecules were predicted to hold the key interactions with the two enzymes to
exert the desired biological functions. To reduce the molecular weight of dual inhibitor, we
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 50
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
further prepared a merged molecule 10 (Figure 1D), in which the two essential
pharmacophoric elements, hydroxamic acid and hydrophobic moiety, are linked by a short
aliphatic chain. We finally realized that the aliphatic chains in SAHA and lovastatin could be
overlaid (Figure 1E), so that the fused molecules could act as HMGR–HDAC dual-targeting
inhibitors. We report herein the synthesis, molecular modeling and biological activities of
these conjugated, merged and hybrid HMGR–HDAC 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
(
A) HDAC inhibitors:
(B) HMGR inhibitor:

H
N
HO
O
HO
HO
OH
OH
N
O
O
H
O
O
O
H
Suberoylanilide hydroxamic acid (SAHA)
O
H
O
H

OH
N
N
H
N
N
SAHA analog
Lovastatin (1, lactone form)
Biologically active form
(
C) Designed conjugated molecules, e.g., 6a and 6b:
(D) Designed merged molecules, e.g. 10:
HO
O
H
HO
O
Spacer
Zn2+ binder
O
O
H
O
H
Zn2+ binder
O
Spacer
O
O
OH
O
8
N
O
N
N
H
N
H
H
H
N
N
5
HN OH
Hydrophobic region
Hydrophobic region
(
E) Designed hybrid HMGRHDAC inhibitors, e.g., 12:

H
N
OH
N
O
Zn2+ binder
H
Spacer
OH OH
O
SAHA
O
H
O
OH
N
O
H
O
H
OH OH
O
Hydrophobic region
Hybrid HMGRHDAC inhibitor
OH
Lovastatin (active form)
Figure 1. Designed dual inhibitors for HMG-CoA reductase (HMGR) and histone deacetylase
(HDAC). (A) Structure of a representative HMGR inhibitor. (B) Structures of representative
6
ACS Paragon Plus Environment

Page 7 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
HDAC inhibitors. (C) Design of lovastatin derivative bearing hydroxamic acid group by
conjugation strategy via click reaction. (D) Design of lovastatin derivative bearing
hydroxamic acid group by merged strategy using a short aliphatic spacer. (E) Design of
lovastatin hydroxamic acid by fused strategy to share a common 6-carbon unit from SAHA
and lovastatin. The hydroxamic acid (red) is a surrogate of the carboxylate group in lovastatin,
and acts as a zinc chelation group for HDAC inhibition. The HMG-like moiety (blue)
provides the appropriate interactions with HMGR and HDAC.
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
RESULTS AND DISCUSSION
Chemistry. To synthesize the conjugated compounds 6a,b (Scheme 1), lovastatin in the
lactone form was treated with KOH in H O/MeOH to open the lactone ring and to cleave the
2
ester bond. The lactone ring was regenerated by treatment with HCl (6 M) for 4.5 h at room
temperature without causing side reactions due to elimination of water molecules. A bulky
tetrabutyldimethylsilyl (TBS) group was selectively introduced to protect the less hindered
hydroxyl group, giving intermediate 2. The subsequent treatment of 2 with p-nitrophenyl
chloroformate in the presence of pyridine and 4-dimethylaminopyridine (DMAP) afforded the
carbonate 3. The substitution reactions of 3 with 4-ethynylaniline and 4-ethynylbenzylamine
were carried out to provide the carbamates 4a and 4b in 66% and 90% yields, respectively.
+
Compounds 4a and 4b were then subjected to the Cu -catalyzed 1,3-dipolar cycloadditions
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 50
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
(
click reactions) with 7-azidoheptanoic acid to give practically pure triazole products 5a and
5
b, which were isolated simply by extraction with EtOAc. The acids 5a and 5b were activated
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
by treatment with ethyl chloroformate to form mixed anhydrides, which were reacted in situ
with hydroxylamine to give hydroxamates 6a and 6b in modest yields.
Scheme 1. Synthesis of lovastatin derivatives 6a and 6b bearing hydroxamic acid group by
a
conjugation strategy.
HO
O
TBSO
O
TBSO
O
H
O
H
O
H
O
H
NO2
O
O
iiii
iv
v
O
OH
O
O
H
H
1
2
3
Lovastatin (lactone form)
TBSO
O
TBSO
O
O
H
O
H
O
vi
O
O
N
O
H
N
n
n
H
H
H
O
5
N
N
N
OH
4
a, n = 0
5a, n = 0
5b, n = 1
4b, n = 1
TBSO
O
O
H
O
vii
O
N
H
H
n
O
N
N
N
5
HN OH
6
6
a, n = 0
b, n = 1
a
Reagents and conditions: (i) KOH, H O/MeOH, reflux, 8 h; (ii) 6 M HCl, rt, 4.5 h; (iii)
2
TBSCl, imidazole, CH Cl , rt, 5.5 h; 58% for 3 steps; (iv) p-nitrophenyl chloroformate,
2
2
DMAP, pyridine, rt, 15 h; 81%; (v) 4-ethynylaniline or 4-ethynylbenzylamine, DMAP,
pyridine, rt, 3–21 h; 66% for 4a; 90% for 4b; (vi) 7-azidoheptanoic acid, CuSO ·(H O) ,
4
2
5
8
ACS Paragon Plus Environment

Page 9 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
o
o
sodium ascorbate, t-BuOH/H O, 60 C, 18 h; (vii) ClCO Et, Et N, THF, 0 C, 10 min; (viii)
2
2
3
o
NH OH·HCl, KOH, THF/MeOH, 0 C, 15 min; 37% for 6a from 4a; 53% for 6b from 4b.
2
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
In another approach (Scheme 2), carbonate 3 was reacted with the allyl ester of
7
-aminoheptanoic acid (7) to afford carbamate 8. The allyl group in 8 was removed by the
catalysis of palladium. The acid intermediate 9 was activated with ethyl chloroformate and
reacted with hydroxylamine to give hydroxamate 10.
Scheme 2. Synthesis of lovastatin derivative 10 bearing a hydroxamic acid group by merged
a
strategy.
TBSO
O
H
3
O
H
O
O
5
+
i
ii
O
N
H
O
O
NH2
O
5
7
8
TBSO
O
TBSO
O
O
H
O
O
O
O
O
iii, iv
H
OH
O
N
OH
5
O
N
N
H
H
H
H
5
H
9
10
a
Reagents and conditions: (i) DMAP, pyridine, rt, 2 h; 70%; (ii) Pd(PPh ) , PPh , Et N,
3
4
3
3
o
HCOOH, THF, rt, 3 h; (iii) ClCO Et, Et N, THF, 0 C, 10 min; (iv) NH OH·HCl, KOH,
2
3
2
o
THF/MeOH, 0 C, 15 min; 38% for 10 from 8.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 50
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
As the commercially available lovastatin is in the lactone form, we attempted to explore
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
its direct coupling reaction with hydroxylamine to afford lovastatin hydroxamic acid
30
(
lova-HA) without involvement of tedious protection–deprotection procedures. In our initial
attempt, hydroxylamine was freshly prepared by neutralization of the hydrochloric salt with a
base (KOH or NaOMe) in MeOH, and used as the nucleophile to react with lovastatin (1 in
the lactone form) in anhydrous THF/MeOH in the presence of Et N or DMAP at room
3
1
13
temperature. Though the desired product of lova-HA (12) was observed by MS, H and C
NMR spectral analyses, the reaction was complicated by recovery of lovastatin (in the acid
form) and formation of the methyl ester. To reduce the side reactions, Lewis acid was tested to
activate the lactone moiety of lovastatin while retain the nucleophilicity of hydroxylamine
(
method A in Scheme 3). Among the examined Lewis acids (LiCl, MgBr , ZnCl and CeCl ),
2 2 3
using MgBr (2 equiv) along with NaHCO (8 equiv) for in situ neutralization of
2
3
hydroxylamine hydrochloride (8.5 equiv) in THF/MeOH turned out to be a superior method
for conversion of lovastatin (in lactone form) to 12 (79% yield). By similar procedures
(
method A), the lactone forms of simvastatin, atorvastatin and rosuvastatin were also
successfully converted to their corresponding hydroxamic acids 13 (simva-HA), 14
atorva-HA) and 15 (rosuva-HA) in 48%, 56% and 51% yields, respectively. The ester
functionality and C=C double bonds existing in lovastatin, simvastatin and rosuvastatin were
(
1
0
ACS Paragon Plus Environment

Page 11 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
unchanged under such mild reaction conditions of our method, whereas these functional
groups might not be retained on acid-catalyzed hydrolysis or hydrogenation involved in the
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
30
previously reported protection–deprotection procedures. In these cases, only modest yields
of the desired statin hydroxamic acids were obtained due to the competitive formation of the
related methyl esters in MeOH solution. To reduce the formation of statin methyl esters in
MeOH solution, we further investigated the substitution reactions of statin lactones with 50%
hydroxylamine aqueous solution (method B in Scheme 3). To our satisfaction, high yields
(88–95%) of statin hydroxamic acids 12–15 were obtained by treatment of the THF solutions
of statin lactones with 50% aqueous hydroxylamine (5 equiv) for a short reaction time (1 h) at
ambient temperature.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 50
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
Scheme 3. Synthesis of statin hydroxamic acids 12–15.
HO
O
O
O
H
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
O
H
R
1
, R = H
11, R = Me
Method A:
Method B:
NH OH-HCl, NaHCO , MgBr
NH OH, THF/H O, rt, 1 h.
2
3
2
2
2
THF/MeOH, rt, 22 h.
O
HO
HN OH
OH
O
O
H
R
1
1
2, R = H
3, R = Me
O
O
HO
HO
HN OH
OH
HN OH
OH
F
F
N
N
N
PhHN
Ph
N
Me
SO Me
2
O
1
4
15
Molecular Modeling. In this work, we modeled the structures of class I HDACs (in
particular HDACs 1 and 2) to focus on the design of HMGR–HDAC dual-targeting
compounds for applications in cancer treatment, though inhibition of HDAC6 is reported to
31
be a promising strategy for treatment of some neurodegenerative diseases. HDAC1 and
HDAC2 are homologs with high sequence identify and similarity, especially for the active site
residues (Figure s1 in Supporting Information (SI)). There were no HDAC1/2 crystal
structures available when we launched the design project. To simplify the design process, we
mainly used the homology modeled HDAC1 as our primary protein structure of design. The
1
2
ACS Paragon Plus Environment

Page 13 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RMSD of catalytic residues between our homology modeled HDAC1 and the HDAC2 crystal
32
structure (3MAX) published in 2010 was only 0.921Å , justifying our result of homology
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
modeling (Figure s2 in SI).
Figures 2 and 3 illustrate the predicted binding modes and the detailed protein–inhibitor
interactions of compound 12 with HDAC1 and HMGR, respectively. The molecular modeling
results for compounds 13–15 are collected in SI (Figure s3). These statin hydroxamic acids all
show the 3,5,N-trihydroxyheptamide moiety consistently inserts into the catalytic outer-tunnel
of HDAC. The hydroxamate group also chelates a zinc ion, which is important for the
catalytic process of HDAC. In addition, the 3,5-dihydroxyl groups also contribute to the
hydrogen-bonding interactions with the residues in the catalytic tunnel. In contrast to the
previously developed HDAC inhibitors that simply contain aliphatic chains to exert
hydrophobic interactions with the catalytic tunnel of HDAC, our results of molecular
modeling reveal that placing polar substituents at proper locations of ligand, e.g., the OH
groups at C-3 and C-5 positions in 12, can still maintain the binding affinity to HDAC
inhibitors. Thus, statin hydroxamates bearing 3,5-dihydroxyl substituents on the alkyl chain
may become better HDAC inhibitors.
In another aspect, hydroxamic acid is considered as a bioisostere of carboxylic acid.
2
9
Similar to the crystal structures of statin–HMGR complexes (Figure s4 in SI), our molecular
docking studies also indicate that the hydroxamate group in compounds 12–15 contributes to
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 50
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
substantial hydrogen bondings with at least two residues of Lys A735, Ser B684, and Lys
B692 in HMGR, wherein A and B represent A and B chains of the protein. Thus, compounds
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
2–15 are also predicted to exhibit reasonable affinity to HMGR. Our alignment analysis
using various binding poses of compound 12 further indicates that the binding pockets of
HMGR and HDAC1 are dissimilar while the highly flexible nature of compound 12 still fits
into each active site with different conformations (Figure s5 in SI).
Figure 2. Predicted binding mode and receptor–ligand interaction diagrams of compound 12
on homology modeled HDAC1. The structural template, HDLP structure, was obtained from
Protein Data Bank (PDB ID: 3CSR). Amino acid residues within 4.5 Å of ligand are
presented in the two-dimensional interaction diagram. The blue circle on the ligand represents
its exposure to the solvent. The larger circle indicates more exposure of the ligand to the
solvent. The green and magenta dashed lines represent hydrogen bonding and metal chelation,
respectively.
1
4
ACS Paragon Plus Environment

Page 15 of 50
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
Figure 3. Predicted binding mode and receptor–ligand interaction diagrams of compound 12
on HMGR. The illustration is similar to that described in Figure 2. The letters A and B before
the residue numbers represent the A and B chains, respectively, in the PDB entry of HMGR
(PDB ID: 1HW9).
HMGR and HDAC Inhibition. We first used in vitro enzymatic assays to examine the
inhibition of synthetic compounds on HMGR and HDAC activities. As shown in Table 1,
compounds 6a, 6b, 10, and 12–15 all inhibited HMGR activity with IC50 values similar to
statins in nanomolar range. However, SAHA had no inhibition on HMGR activity even at a
dose of 10 μM. Compounds 6a, 6b, 10, and 12–15 appreciably inhibited HDAC1 (class I),
HDAC2 (class I) and HDAC6 (class II) activities with IC50 values in nanomolar range.
However, lovastatin and atorvastatin inhibited HDACs at a concentration of more than 10 μM.
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 50
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
These experiments clearly indicated that compounds 6a, 6b, 10, and 12–15 acted as potent
dual functional inhibitors against HMGR 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 1. Inhibitory activities (IC ) against HMGR and HDACs.
5
0
a
IC (nM)
5
0
compound
HMGR
HDAC1
11911 ± 681
11619 ± 382
20.9 ± 7.1
HDAC2
HDAC6
16285 ± 1575
14466 ± 567
19.4 ± 6.0
lovastatin
atorvastatin
SAHA
29.5 ± 3.5
12.9 ± 1.3
>10,000
25933 ± 651
22547 ± 1618
100.9 ± 10.0
463.3 ± 28.0
881.8 ± 9.7
6
a
36.5 ± 5.3
53.8 ± 5.2
54.1 ± 2.1
16.8 ± 1.9
13.1 ± 1.5
12.3 ± 2.7
43.7 ± 1.6
159.0 ± 8.4
124.7 ± 6.3
122.0 ± 12.7
64.8 ± 5.4
127.4 ± 21.5
34.0 ± 4.3
6
b
1
1
1
1
1
0
2
3
4
5
657.7 ± 21.1
468.3 ± 27.2
414.2 ± 15.5
467.2 ± 19.0
600.1 ± 34.0
139.7 ± 5.7
51.0 ± 6.1
63.4 ± 4.5
69.5 ± 4.1
122.9 ± 9.5
125.2 ± 7.1
86.6 ± 6.0
133.3 ± 5.5
a
Data are shown as mean ± SD of three experiments.
To further evaluate their inhibition on HMGR in lung cancer cells, compounds 6a, 6b, 10,
1
6
ACS Paragon Plus Environment

Page 17 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and 12–15 at 1–50 μM were applied to cells for 24 h, and the HMGR activity in whole cell
lysates was measured. Compounds 6a, 6b, 10, and 12–15 effectively reduced HMGR activity
in a dose-dependent manner with IC values similar to lovastatin in nanomolar range (Table
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
0
2
). However, SAHA did not affect HMGR at any dose.
Table 2. Inhibition on the HMGR activity in A549 lung cancer cells.
a
compound IC (μM)
5
0
lovastatin
SAHA
19.8 ± 2.2
b
ND
6
a
22.3 ± 5.9
16.1 ± 4.8
13.2 ± 5.1
15.9 ± 0.8
7.9 ± 1.4
6
b
1
1
1
1
1
0
2
3
4
5
a
5.7 ± 1.2
14.9 ± 1.7
a
b
Data are shown as mean ± SD of three experiments. ND: not determined (>10 μM) due to
high toxicity of SAHA to cells.
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 50
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
Intracellular histone acetylation status is a direct marker of class I HDAC inhibition,
whereas α-tubulin is a substrate of HDAC6 (class II). As shown in Figure 4, compounds 6a,
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
6
b, and 12–15 promoted histone and tubulin acetylations in a dose-dependent manner. SAHA
and lovastatin were included for comparison.
(A)
(B)
Figure 4. Effect of hydroxamate compounds on acetylation of histone H3 and tubulin in A549
lung cancer cell lines. A549 cells were treated with indicated doses of drugs for 24 h: (A)
conjugated compounds 6a and 6b; (B) hybrid compounds 12–15. SAHA (5 μM) and
lovastatin (30 or 50 μM) were used for comparison. Total protein lysates were subjected to
1
8
ACS Paragon Plus Environment

Page 19 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Western blotting using antibodies specific for acetyl-histone H3, acetyl-tubulin or actin.
Equivalent protein loading with similar level of β-actin was applied.
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
Cell Growth Inhibition. Cell viability assays were performed to determine the
cytotoxicity and specificity of compounds 12–14 (Table 3). A549 human lung cancer cells,
MEF normal mouse fibroblast cells, and HS68 normal human fibroblast cells were treated
with different doses of drug for 72 h, and the IC values were evaluated. SAHA was toxic to
50
both cancer and normal cells without specificity. The selectivity index of statins for cancer
and normal cells was low. Compounds 12–14 significantly induced cytotoxicity in the cancer
cells (IC < 20 μM), but were not toxic to the normal cells at 100 μM. Our results indicate
5
0
that compounds 12–14 have a potential to be developed as safer drugs than statins and SAHA
in cancer treatment and other therapeutic uses.
Table 3. Inhibition on the growth of cancer and normal cells.
a
b
IC (μM)
S.I.
5
0
compound
a
a
a
A549
MEF
HS68
A549/MEF
A549/HS68
lovastatin
simvastatin
atorvastatin
11.4 ± 6.3
16.3 ± 0.1
8.7 ± 1.1
35.0 ± 5.9
36.7 ± 4.4
30.7 ± 3.2
23.2 ± 3.5
26.4 ± 2.1
22.7 ± 2.0
3.1
2.3
3.5
2.0
1.6
2.6
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 50
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
SAHA
4.5 ± 0.8
18.2 ± 3.4
20.0 ± 3.1
17.5 ± 4.7
4.4 ± 1.4
> 100
4.6 ± 0.7
> 100
1.0
1.0
1
1
1
2
3
4
> 5.6
> 5.0
> 5.7
> 5.6
> 5.0
> 5.7
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
> 100
> 100
> 100
> 100
a
Cells were treated with indicated doses of test compounds for 72 h, and the cell viability was
measured by MTT assay. Data are shown as mean ± SD of three experiments. A549: human
lung cancer cells; MEF: normal mouse fibroblast cells; HS68: normal human fibroblast cells.
b
Selectivity index: the ratio of IC on cancer cell to IC on normal cell.
5
0
50
CONCLUSION
HDAC has been a validated target for cancer therapy. Though a potent HDAC inhibitor
SAHA has been approved for treatment of cutaneous T cell lymphoma, the high toxicity of
SAHA is a serious concern. In contrast, statins have shown safety and efficacy in preventing
human cardiovascular diseases. The recent studies also indicate that statins may have
beneficial effects in prevention and treatment of cancers. We thus conceive a new therapeutic
approach for cancer treatment by concurrent inhibition of HDAC and HMGR. We have
successfully synthesized a series of dual-action compounds to target HDAC and HMGR by
having a hydroxamate group essential for chelation with the zinc ion in the active site of
HDAC and the key structural elements of statin for binding with both proteins. In addition to
2
0
ACS Paragon Plus Environment

Page 21 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
using hydroxamic acid as a surrogate of carboxylic acid, we also found that the aliphatic
chains in SAHA and statin could be overlaid. Thus, the hybrid molecules 12–15 exhibited
high inhibitory activities against both HDAC and HMGR. Our cell-based assays also showed
that these statin hydroxamic acids effectively reduced the HMGR activity and promoted the
acetylations of histone and tubulin in cancer cells (IC50 < 20 μM), but were not toxic to
normal cells at the concentration as high as 100 μM.
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
The in vivo experiments, pharmacokinetics and metabolic studies of the HDAC–HMGR
dual-action inhibitors are currently under investigation. Our results indicated that oral
administration of compound 12 could prevent and treat azoxymethane (AOM)/dextran
sulphate sodium (DSS)-induced colitis-associated colorectal cancer in mice. Treatment with
compound 12 did not show any toxicity by biochemical examinations and hematoxylin and
eosin (H&E) staining of various organs (manuscript in preparation). The pharmacokinetics of
compound 12 has been evaluated in healthy rats through intravenous or oral administration.
We need to stress that it is still possible to find more efficient dual-action compounds by
taking other possible combinations of different HMGR and HDAC inhibitors. For example,
using different zinc binding groups may provide a higher affinity toward the target proteins
and thus render a better therapeutic efficiency. On the other hand, changing the cap region in
the dual-action compounds may achieve better selectivity to differentiate HDAC subtypes. In
summary, we have demonstrated the first example of dual-action inhibitors targeting HDAC
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 50
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
and HMGR as a promising approach to cancer therapy.
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
EXPERIMENTAL SECTION
Materials and Methods. All the reagents were commercially available and used without
further purification unless indicated otherwise. All solvents were anhydrous grade unless
indicated otherwise. Dichloromethane was distilled from CaH ; triethylamine was distilled
2
from MgSO . Lovastatin and simvastatin were obtained from Lotus Pharmaceutical Company
4
(Nantou, Taiwan). Atorvastatin was obtained from Synpac Kingdom Pharmaceutical
Company (Taipei, Taiwan). SAHA was purchased from Merck (Frankfurter Sparkasse,
Germany). Anti-acetylhistone H3 antibody was purchased from Millipore (Bedford, MA,
USA). Anti-β-actin antibody was purchased from Gene Tex (Irvine, CA, USA). HDAC
Fluorimetric Assay/Drug Discovery Kit (AK-500) was purchased from Biomol (Plymouth
Meeting, PA, USA). HMG-CoA reductase activity kit (CS-1090), and anti-acetyltubulin
antibody were purchased from Sigma (St. Louis, MO, USA).
All non-aqueous reactions were carried out in oven-dried glassware under a slight
positive pressure of argon unless otherwise noted. Reactions were magnetically stirred and
monitored by thin-layer chromatography on silica gel. Analytical thin layer chromatography
(TLC) was performed on 0.25 mm silica gel 60 F254 plates. Compounds were visualized by
UV, or using p-anisaldehyde, ninhydrin, phosphomolybdic acid, KMnO or I as visualizing
4
2
2
2
ACS Paragon Plus Environment

Page 23 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
agent. Flash chromatography was carried out on columns packed with silica gel 60
(
0.040–0.063 mm particle sizes). High-performance liquid chromatography (HPLC) was
performed on Agilent 1100 Series instrument equipped with a degasser, Quat pump, and UV
detector.
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
Melting points were recorded on an Electrothermal MEL-TEMP 1101D melting point
apparatus and were not corrected. Optical rotations were measured on digital polarimeter of
–1
2 –1
Japan JASCO Co. DIP-1000; [α] values are given in units of 10 deg cm g . Infrared (IR)
D
spectra were recorded on a Nicolet Magna 550-II FT-IR spectrometer. Nuclear magnetic
resonance (NMR) spectra were obtained on Varian Unity Plus-400 (400 MHz) or Bruker
Avance-III 400 MHz NMR spectrometers. Chemical shifts (δ) were recorded in parts per
million (ppm) relative to δ 7.24/ δ 77.0 (central line of t) for CHCl /CDCl or δ 0.00/ δ
C
H
C
3
3
H
7
7.0 (central line of t) for TMS/CDCl , δ 3.31/ δ 49.0 for CH OH/CD OD, and δ 2.50
3 H C 3 3 H
3
9.5 (m) for (CH ) SO/(CD ) SO. The splitting patterns are reported as s (singlet), d
3 2 3 2
C
(doublet), t (triplet), q (quartet), m (multiplet) and br (broad). Coupling constants (J) are given
in hertz. The ESI–MS experiments were conducted on a Bruker Daltonics BioTOF III
high-resolution mass spectrometer.
General Compound Characterization. New compounds were characterized by their
1
13
physical and spectroscopic properties (mp, TLC, [α], IR, ESI–MS, H and C NMR). Purity
of synthetic compounds 6a, 6b, 10, and 12–15 was assessed to be  95% by HPLC analysis
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 50
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
(
Agilent HP-1100) on an HC-C18 column (250 mm × 4.6 mm i.d., 5 μm particle size) using
gradient elution of aqueous CH CN for 20–30 min at a flow rate of 1 mL/min with detection
3
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
at 254 nm wavelength.
Homology Modeling. The sequence alignment between human HDAC1 and histone
deacetylase-like protein (HDLP) was determined by the multiple sequence alignment between
human class I HDACs (HDAC1, HDAC2, HDAC3, and HDAC8) and HDLP using the
ClastalW module with BLOSUM scoring matrix in Discovery Studio 2.55 (Accelrys, Inc. San
Diego, CA, USA). The homology model structure was generated and optimized by
MODELLER in Discovery Studio 2.55 (Accelrys, Inc. San Diego, CA, USA). The structural
template of the homology modeling of HDAC1 and HDAC2, the HDLP structure was
obtained from Protein Data Bank (PDB ID: 3CSR). The one with the lowest energy score was
selected as the final model.
Molecular Docking. The crystal structure of HMG-CoA reductase was taken from
Protein Databank (PDB 1HW9). The structure of HDAC1 was prepared from the homology
modeling described above. All non-protein molecules were deleted except the zinc ion of the
3
3
HDAC complex. The receptor is prepared by AutoDock 4 suite. The grid box size was
adjusted to cover the original ligand and its surrounding binding pocket residues with 48 × 40
×
40 in HDAC1 and 40 × 40 × 52 in HMG-CoA reductase with grid spacing of 0.375Å .
The 2D ligand structures were prepared using ChemDraw Ultra 12 and the 3D structures were
2
4
ACS Paragon Plus Environment

Page 25 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
4
generated by OpenBabel 2.3.1.
The molecular docking simulations were performed by
RAP 35
AutoDock4, with the recent reparameterized scoring function (AutoDock4 ). It has been
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
shown that the native ligand of HDAC can be reproduced with the new scoring function and a
36
new divide-and-conquer docking approach. The binding pose with lowest score in each case
is selected to represent the predicted binding mode. The 3D and 2D protein–ligand interaction
plots were presented using ICM Browser version 3.7-2d (Molsoft L.L.C., San Diego, USA)
and Molecular Operating Environment version 2009.10 (Chemical Computing Group,
Montreal, Canada), respectively.
Cell Lines and Culture Conditions. A549 human lung carcinoma cells from American
Type Culture Collection (ATCC, Manassas, VA) were cultured in Dulbecco’s Modified
Eagle’s Medium (DMEM). MEF normal mouse fibroblast cells and HS68 normal human
fibroblast cells were cultured in RPMI 1640. All media were supplemented with 10% FBS,
1
00 U/mL penicillin G and 100 mg/mL streptomycin sulfate. Cells were maintained in a
humidified incubator containing 5% CO in air. Cells were subcultured by trypsinization in
2
laminar flow when grew about 80% in the culture dishes.
HDAC Activity Assay. The HDAC activity was performed using the HDAC fluorimetric
activity assay kit (BIOMOL, Plymouth Meeting, PA, USA) according to the manufacturer’s
instructions. Briefly, recombinant proteins of HDAC1, HDAC2 or HDAC6 were incubated
with test compounds, and HDAC reaction was initiated by addition of Fluor-de-Lys substrate.
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 50
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
Samples were incubated for 10 min at room temperature, followed by adding developer to
stop the reaction. Fluorescence was measured by fluorimetric reader with excitation at 360 nm
and emission at 460 nm. The HDAC activity was expressed as arbitrary fluorescence units
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
(AFU). The HDAC activity was calculated as a percentage of activity compared with the
control group. The 50% of inhibition concentration (IC ) values for the test compounds were
5
0
calculated using SigmaPlot software.
HMG-CoA Reductase Activity Assay. The HMGR activity was performed using the
HMGR assay kit according to the manufacturer’s instructions (Sigma–Aldrich, St. Louis, MO,
USA). Briefly, recombinant HMGR protein incubated with lovastatin or test compounds for
10 min at room temperature, or total cell lysate from A549 treated with lovastatin or test
o
compounds was mixed with HMG-CoA and NADPH, and incubated for 5 min at 37 C. The
absorbance at 340 nm was measured. The 50% of inhibition concentration (IC ) values for
5
0
the test compounds were calculated using SigmaPlot software.
Western Blot Analysis. Following treatment with test compounds, cells were lysed on ice.
Total cell lysates were prepared and subjected to SDS–PAGE using adequate percentage
polyacrylamide gels. Immunoblotting was performed using specific antibodies to evaluate the
expression of different proteins.
Cell Proliferation Assay. Cells were seeded at 3000 cells/well in 96-well plates and
maintained for 14–16 h. Cells were treated with DMSO or various concentrations of test
2
6
ACS Paragon Plus Environment

Page 27 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
compounds for 72 h, and then washed with PBS twice. A medium containing
3
-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT, 0.5 mg/mL) was added,
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
o
and the cells were incubated for 4 h at 37 C under 5% CO . During this period, cells having
2
functional succinate dehydrogenase in mitochondria would convert MTT to formazan. The
medium was replaced with 200 μL of DMSO for 30 min at room temperature, and the 96-well
plate was read by an ELISA reader at 570 nm to get the values of absorbance density. The
IC values were calculated using SigmaPlot software.
50
General Methods for Preparation of Statin Hydroxamates. Method A. To a solution of
lovastatin (1, in lactone form, 0.74 mmol) and MgBr (1.48 mmol) in anhydrous THF/MeOH
2
(7:3, 3 mL) was added hydroxylamine hydrochloride (6.3 mmol) and sodium bicarbonate (5.9
mmol). The mixture was stirred at ambient temperature for 22 h, and concentrated under
reduced pressure. The residue was extracted with EtOAc and brine. The organic phase was
dried over MgSO , filtered, concentrated, and purified by flash chromatography (silica gel,
4
CH Cl /MeOH (15:1)) to give the desired statin hydroxamic acid.
2
2
Method B. To a solution of statin (in lactone form, 0.25 mmol) in THF (469 μL) was
added 50% aqueous hydroxylamine (1.24 mmol, 76 μL). The mixture was stirred at ambient
temperature for 1 h, and then concentrated under reduced pressure. The residue was purified
by flash chromatography (silica gel, CH Cl /MeOH (15:1)) to give the desired statin
2
2
hydroxamic acid.
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 50
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
(
4R,6R)-6-[2-((1S,2S,6R,8S,8aR)-8-Hydroxy-2,6-dimethyl-1,2,6,7,8,8a-hexahydronap
hthyl)ethyl]-4-tert-butyldimethylsilyloxy-3,4,5,6-tetrahydro-2H-pyran-2-one
(2).
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
Lovastatin (1, 9.0 g, 22.3 mmol) was heated with potassium hydroxide (12.6 g, 224.5 mmol)
in H O/MeOH (1:6, 63 mL) at refluxing for 8 h. After adding H O (49.5 mL) to the mixture,
2
2
MeOH was removed under reduced pressure. To the residue were added H O (180 mL),
2
CH Cl (45 mL), and 6 M HCl aqueous solution until pH = 2. The mixture was stirred at room
2
2
temperature for 4.5 h, was and then neutralized with saturated NaHCO aqueous solution. The
3
mixture was extracted with CH Cl . The combined organic phase was dried over anhydrous
2
2
MgSO , filtered and concentrated to give a deacylation product as orange oil. TLC (EtOAc) Rf
4
= 0.33.
The crude product was treated with tert-butyldimethylsilyl chloride (8.7 g, 57.9 mmol)
and imidazole (8.8 g, 129.1 mmol) in CH Cl (82 mL) at room temperature for 5.5 h. The
2
2
mixture was concentrated under reduced pressure, and the residue was extracted with CH Cl
2
2
and H O. The organic phase was dried over MgSO , filtered, concentrated, and purified by
2
4
flash chromatography (silica gel, EtOAc/hexane (2:8) to EtOAc) to give the compound 2 (5.6
o
g, 58% overall yield from lovastatin). C H O Si; white solid, mp 141.1–142.3 C; TLC
25
42
4
1
(
EtOAc/hexane (6:4)) R = 0.61; H NMR (CDCl , 400 MHz) δ 5.98 (1 H, d, J = 9.6 Hz),
f
3
5
.78–5.82 (1 H, m), 5.55 (1 H, br s), 4.66–4.70 (1 H, m), 4.29–4.30 (1 H, m), 4.23–4.25 (1 H,
m), 2.54–2.65 (2 H, m), 2.35–2.45 (2 H, m), 2.16–2.18 (1 H, m), 1.70–1.93 (7 H, m),
2
8
ACS Paragon Plus Environment

Page 29 of 50
Journal of Medicinal Chemistry
1
2
3
4
1
.44–1.55 (2 H, m), 1.15 (3 H, d, J = 7.6 Hz), 0.88–0.91 (12 H, m), 0.08 (6 H, d, J = 1.6 Hz);
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
13
C NMR (CDCl , 100 MHz) 170.4, 133.6, 131.3, 129.9, 128.4, 76.3, 65.1, 63.5, 39.2, 38.7,
3
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
6.8, 36.3, 35.7, 32.9, 30.7, 27.3, 25.6 (3 ×), 24.2, 23.7, 17.9, 13.9, –5.0 (2 ×).
(4R,6R)-6-(2-{(1S,2S,6R,8S,8aR)-8-[(p-Nitrophenoxy)carbonyloxy]-2,6-dimethyl-1,2,
6
,7,8,8a-hexahydronaphthyl}ethyl)-4-tert-butyldimethylsilyloxy-3,4,5,6-tetrahydro-2H-p
yran-2-one (3). A mixture of alcohol 2 (5.0 g, 11.5 mmol), p-nitrophenyl chloroformate (16.2
g, 80.2 mmol) and DMAP (98.0 g, 80.2 mmol) was stirred in anhydrous pyridine (80.2 mL) at
room temperature for 15 h. Pyridine was removed under reduced pressure, and the residue
was extracted with CH Cl and 1 M HCl aqueous solution. The combined organic phase was
2
2
washed with saturated NaHCO aqueous solution and brine, dried over MgSO , filtered and
3
4
concentrated under reduced pressure. The residue was purified by flash chromatography
silica gel, CH Cl /hexane (6:4 to 8:2)) to give the carbonate compound 3 (5.6 g, 81%).
(
2
2
o
25
C H NO Si; white powder, mp 146.5–147.3 C; [α] = +233.9 (EtOAc, c = 1.0); TLC
3 D
2
45
8
(EtOAc/hexane (2:8)) R = 0.24; IR ν
(neat) 2955, 2930, 2857, 2360, 1760, 1594, 1525,
f
max
–1
1
1
347, 1258, 1216, 1082 cm ; H NMR (CDCl , 400 MHz) δ 8.27 (2 H, d, J = 9.2 Hz), 7.40 (2
3
H, d, J = 8.8 Hz), 6.01 (1 H, d, J = 9.6 Hz), 5.79–5.83 (1 H, m), 5.57 (1 H, br s), 5.34 (1 H, br
s), 4.68–4.70 (1 H, m), 4.27–4.29 (1 H, m), 2.52–2.59 (3 H, m), 2.34–2.41 (2 H, m),
2
.17–2.23 (1 H, m), 1.66–2.00 (6 H, m), 1.45–1.53 (2 H, m), 1.17 (3 H, d, J = 7.6 Hz), 0.93 (3
13
H, d, J = 6.8 Hz), 0.86 (9 H, s), 0.06 (6 H, d, J = 5.6 Hz); C NMR (CDCl , 100 MHz) 170.2,
3
2
9
ACS Paragon Plus Environment