Identification of a novel aminotetralin class of HDAC6 and HDAC8 selective inhibitors

Article abstract of DOI:10.1021/jm5008962

Herein we report the identification of a novel class of HDAC6 and HDAC8 selective inhibitors through a unique chemistry and phenotypic screening strategy. Tetrahydroisoquinoline 12 was identified as a potent HDAC6 and HDAC8 dual inhibitor from a focused l

Full text of DOI:10.1021/jm5008962

Article
pubs.acs.org/jmc
Identification of a Novel Aminotetralin Class of HDAC6 and HDAC8
Selective Inhibitors
Guozhi Tang,* Jason C. Wong,† Weixing Zhang, Zhanguo Wang, Nan Zhang, Zhenghong Peng,
Zhenshan Zhang, Yiping Rong, Shijie Li, Meifang Zhang, Lingjie Yu, Teng Feng, Xiongwen Zhang,
Xihan Wu, Jim Z. Wu, and Li Chen
Roche Pharmaceutical Research and Early Development, Roche Innovation Center Shanghai, 720 Cailun Road, Shanghai, 201203
China
S
* Supporting Information
ABSTRACT: Herein we report the identification of a novel
class of HDAC6 and HDAC8 selective inhibitors through a
unique chemistry and phenotypic screening strategy. Tetrahy-
droisoquinoline 12 was identified as a potent HDAC6 and
HDAC8 dual inhibitor from a focused library through cellular
tubulin acetylation and p21 induction screening assays.
Scaffold hopping from 12 led to the discovery of an
aminotetralin class of HDAC inhibitors. In particular, the 3-R stereoisomer 32 showed highly potent inhibition against
HDAC6 and HDAC8 with IC50 values of 50 and 80 nM, respectively. Treatment of neuroblastoma BE(2)C cells with 32 resulted
in elevated levels of acetylated tubulin, TrkA, and neurite outgrowth with only marginal effects on p21 induction and histone H3
acetylation. Consistent with its weak enzymatic inhibition of HDAC1, 32 showed significantly less cytotoxicity than SAHA and
moderately inhibited the growth of myeloma NCI-H929 and OPM-2 cells.
hydrogen bonding interactions with the surrounding active site
residues; (2) a linker that passes through the narrow
hydrophobic channel of HDACs; and (3) a partially exposed
capping group that interacts with the rim region where HDAC
isoforms have significant structural variation and conforma-
tional plasticity. Practically, the structural difference of HDACs
has been exploited successfully and isoform selectivity can be
generated from modulation of ZBG and appropriate conforma-
tional constraints in the linker or capping groups. Tubacin (1),3
ACY-1215 (2),4 tubastatin (3),5 and HPOB (4)6 are reported
as HDAC6 selective inhibitors using bulky or conformationally
constrained capping groups as selectivity vector (Figure 1).
Conformational constraint at the linker part is also widely
applied to generate HDAC isoform selectivity, as exemplified
by HDAC6 inhibitor BRD9757 (5)7 and HDAC8 inhibitors
PCI-34051 (6)8 and NCC149 (7).9 In particular, 5 is capping-
free and it uses a cyclopentene moiety for HDAC6 selectivity.
For the ZBG part of HDAC inhibitors, it is generally limited to
hydroxamic acid, thiol, carboxylic acid, and ketone groups. But
there are a few successful cases where atypical ZBGs were
devised for generation of HDAC isoform selectivity. Different
from other HDAC isoforms, HDAC1−3 are reported to
possess a larger exit channel at the catalytic site that tolerates
bulky ZBGs like o-aminoanilide. Both SNDX-275 and
chidamide belong to this class of HDAC inhibitors that target
HDAC1−3 only.10,11
INTRODUCTION
■
Histone deacetylases (HDACs) are one of the major classes of
post-translational regulators that are responsible for deacetyla-
tion of lysine residues in histone and non-histone substrates.1
There are 18 HDAC isoforms that are divided into four classes
based on their sequence similarity, cellular localization, tissue
expression patterns, and enzymatic mechanisms. Class I
(HDACs 1, 2, 3, and 8), class II (HDACs 4, 5, 6, 7, 9, and
10), and class IV (HDAC11) HDACs are all zinc-dependent
deacetylases that are mechanistically distinct from NAD+-
dependent class III HDACs. Inhibition of HDACs is a proven
epigenetic strategy for the treatment of cancers, evidenced by
the fact that both vorinostat (SAHA) and romidepsin (FK-228)
have gained FDA approval for the treatment of cutaneous T-
cell lymphoma (CTCL). In addition, there are additional
HDAC inhibitors at various stages of clinical development for
the treatment of hematological malignancies, solid tumors, and
immunological or neurological disorders. Most of them are
class I selective, such as romidepsin, chidamide (CS-055),
entinostat (SNDX-275), or pan-HDAC inhibitors such as
SAHA and LBH-589. Selective HDAC inhibitors that affect
only a limited number of isoforms are valuable, as they can be
used as tools to study the roles of individual HDAC isoforms in
biological processes and as potential therapeutic agents to treat
certain diseases with potentially fewer side effects than pan-
HDAC inhibitors.2
The HDAC inhibitors typically share a well-recognized
pharmacophore that consists of three parts: (1) a zinc-binding
group (ZBG) that is buried deep in the catalytic pocket of the
HDAC enzyme to chelate with the zinc cation and form key
Received: June 12, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. SAHA and known HDAC6/HDAC8-selective inhibitors.
Scheme 1. Solid Phase Synthesis of Focused Libraries and Identification of 12 as HDAC6/8 Inhibitor
Previously, we disclosed 3R,4S-pyrrolidines as HDAC1−3
selective inhibitors that demonstrated dose dependent in vivo
antitumor efficacy in xenograft models with concurrent
induction of p21 and acetylated histone H3 in tumor tissues.12
Herein we report our efforts to discover and characterize novel
HDAC inhibitors with specific HDAC isoform selectivity,
especially toward HDAC6 and/or HDAC8. Selective inhibition
of HDAC6 has recently emerged as an attractive approach for
the treatment of cancers, Alzheimer’s disease, and rheumatoid
arthritis.13 For example, 2 currently is in phase I/II clinical trials
for treatment of multiple myeloma.4 While originally disclosed
as a neuroprotective agent, 3 and its β- and γ-carboline
analogues were found to enhance the immunosuppressive
effects of Foxp3+ regulatory T cells and have anti-inflammatory
effects.5 Notably, 4 did not induce cell death by itself but
enhanced the effectiveness of DNA-damaging anticancer drugs
in transformed cells but not normal cells.6 On the other hand,
HDAC8 has been implicated as a drug target for the treatment
of minimal residual disease in neuroblastoma and malignancies
such as T-cell lymphoma and acute myeloid leukemia.14
HDAC8 inhibitors such as 6 and 7 were reported to induce
apoptosis and exert growth-inhibitory effects on T-cell
lymphoma cells.8,9 Recently, E. Holson et al. reported the
identification of a meta-substituted phenylhydroxamic acid 8 as
an HDAC6/8 dual inhibitor based on enzymatic inhibition
profile against selected HDAC isoforms.15 It has been
suggested that simultaneous inhibition of HDAC6 and
HDAC8 has many potential therapeutic applications, providing
a larger therapeutic window than inhibition of HDAC1−3.13−15
RESULTS AND DISCUSSION
■
We considered conformational constriction of the linker as the
most effective approach to devise HDAC isoform-selectivity
with high ligand efficiency. Since hydroxamic acids are usually
complicated by hydrolysis issues and suboptimal solubility, a
focused library was designed and prepared by facile solid phase
synthesis using reductive amination to connect fused bicyclic
linkers and various capping groups (Scheme 1). In library
synthesis, “Wang” resin was loaded with N-hydroxyphthalimide
by Mitsunobu reaction and then treated with hydrazine to
generate resin 9. A coupling reaction of 9 and benzoic acids
that have six- or seven-member fused rings gave 10 (see details
in Supporting Information). After removal of the N-Fmoc
protective groups by piperidine, the secondary amine was
coupled with different aromatic aldehydes and the resulting
Schiff base was reduced with NaBH3CN to afford 11.
Ultimately, hydroxamates were released from resin 11 upon
treatment with TFA and purified by preparative HPLC. The
synthesized compounds were then evaluated in tubulin
acetylation assay (Tub-Ac)16 as a surrogate for cellular
HDAC6 inhibition. As a surrogate assay for HDAC1−3
inhibition, p21 reporter gene assay17 was applied as a
counterscreen using SNDX-275 as the standard reference for
selectivity assessment.
B
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Among those moieties used as linkers, the 2,7-disubstituted
tetrahydroisoquinoline class was found to possess the desired
selectivity profile. For example, analogue 12 showed strong
Tub-Ac activities and low p21 induction in cellular assays. In
the cellular HDAC6 inhibition assay, 12 showed about 70% of
Tub-Ac when compared to SAHA at 10 and 30 μM,
respectively, with EC50 at 3.45 μM. In the counterscreen
against HDAC1−3 isoforms, 12 did not induce p21 when
compared to SNDX-275 at 3, 10, and 30 μM in HeLa cells
(Table 1). 12 was then biochemically profiled against a panel of
hydrolysis of the hydroxamic acid group and N−C cleavage on
the tetrahydroisoquinoline moiety based on microsome
metabolite identification studies. Compound 12 also had
poor solubility at less than 10 μg/mL in the lyophilized
solubility assay (LYSA). To address these metabolic stability
and solubility issues, an aminotetralin class of analogues 17−25
were designed in close mimicry to the linker of 12 and
synthesized according to the procedures in Scheme 2. Ketone
derivative 13 was converted to N-Boc aminotetralin 14 by an
amination reaction and then treatment with (Boc)2O. After the
Pd-catalyzed carboxylation and removal of Boc protective
group, 15 was obtained. It was treated with different aryl halides
under copper-catalyzed coupling or nucleophilic aromatic
substitution conditions to incorporate the capping groups.
The resultant ester 16 was treated with hydroxylamine under
basic conditions to give hydroxamic acids 17−25. In addition, a
few N-methylated analogues such as 26 and sulfonamides 27
and 28 were made to explore the structure−activity relationship
(SAR) and HDAC isoform selectivity.
Table 1. Tubulin Acetylation and p21 Reporter Gene
a
Induction by 17−28
These aminotetralin-based hydroxamates were evaluated for
their biological effects on Tub-Ac and p21 induction in cellular
assays. We anticipated that various capping groups would form
important hydrophobic interactions with the water-exposed
residues at the rim region and that heteroaromatic groups and
polar functional groups would be tolerated or even beneficial at
this position. Accordingly, the 2-aminopyrimidine analogue 20
showed significantly enhanced cellular HDAC6 inhibition over
aniline analogue 17 and the 2-aminopyridine analogue 19. It
remains unclear if this was due to extra hydrogen bonding or
increased π−π interactions between the electron-deficient
pyrimidine group and certain HDAC6 residues. It was found
that a phenyl group was well tolerated at the meta-position of
the aminopyrimidine ring but any further substitution on the
phenyl group did not necessarily contribute to potency or
selectivity (see examples 21−24). Finally, we found that
analogue 25 with a pyridine substituent at the meta-position
demonstrated potent Tub-Ac induction with an EC50 of 1.18
μM and had good solubility with LYSA of 66 μg/mL. N-
methylation typically led to decreased Tub-Ac induction as
indicated by analogue 26. While sulfonamide analogues had
good potency in Tub-Ac, it came with a loss of selectivity
against HDAC1−3. For example, sulfonamides 27 and 28 had
consistent p21 induction at 3 and 10 μM, respectively. We
speculate that the sulfonamide capping groups have a different
orientation relative to the linker and are thus better tolerated by
HDAC1−3 isoforms than non-sulfonamides 17−26.
We decided to further study racemate 25 to clarify the
impact of its stereochemistry on HDAC6/8 inhibition.
Optically pure tert-butyl N-(7-bromotetralin-2-yl)carbamates
29 and 30 were obtained by supercritical fluid chromatography
(SFC) chiral separation of 14, and the stereochemistry of 30
(2R) was confirmed by X-ray structure (Supporting Informa-
tion). By following the same procedure used to prepare 25,
chiral hydroxamates 31 (S) and 32 (R) were synthesized from
29 and 30 independently (Scheme 3).
In agreement with the results of 25, neither 31 nor 32
induced p21 in HeLa cells at 3 and 10 μM relative to SNDX-
275 (Figure 2A). In contrast, SAHA, a nonselective HDAC
inhibitor, demonstrated 8−40 times more p21 induction than
SNDX-275 under the same concentrations. Meanwhile, both
31 and 32 potently induced tubulin acetylation with EC50
values of 0.69 and 1.64 μM, respectively, which are similar to
the activity of SAHA (Figure 2B). The cellular assays indicated
a
EC50 values of tubulin acetylation are based on ELISA experiments
b
run in duplicate in A549 cells, SD 15%. p21-RP3 represents the
relative luciferase signal in the p21 reporter gene assay17 by individual
example compared to SNDX-27510 at 3 μM in triplicate and
normalized by internal GFP transfection control; p21-RP10 represents
the relative level of p21 induced by individual example compared to
SNDX-275 at 10 μM in triplicate. In all cases data varied by 10% or
c
less. The lyophilized solubility assay (LYSA) was run in duplicate.
The mean values are reported in μg/mL, SD 10%.
all 11 zinc-dependent HDACs. It inhibited enzymatic activities
of HDAC6 and HDAC8 with IC50 values of 0.05 and 0.03 μM,
respectively, but was barely active against other HDAC
isoforms with an IC50 of 6.6 μM against HDAC3 and above
20 μM against other HDACs (Table 2). These results indicated
that 12 is a selective HDAC6/8 inhibitor and a good starting
point for further chemistry optimization. We then evaluated the
in vitro metabolic stability and physiochemical properties of
compound 12. It had high intrinsic clearance largely due to
C
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 2. Biological Activities of 31 and 32
ac
,
bc
,
IC50 against HDAC isoforms (μM)
GI50 (μM)
compd HDAC1 HDAC2 HDAC3 HDAC4 HDAC5 HDAC6 HDAC7 HDAC8 HDAC9 HDAC10 HDAC11 NCI-H929 BE(2)-C
SAHA
12
0.31
21.3
11.2
6.31
0.24
109.0
−
0.13
6.59
−
76.0
67.8
−
27.2
36.0
−
0.20
0.05
0.04
0.05
105
60.7
−
0.31
0.03
0.39
0.08
141
65.8
−
0.43
71.1
−
0.20
75.3
−
0.8
−
−
1.1
5.1
31
16.1
5.4
32
>100
>100
>100
>100
30.8
35.0
>100
>100
7.7
a
IC50 values of enzymatic inhibition to HDAC enzymes run in duplicate, SD 15%. IC50 values of SAHA are cited from www.reactionbiology.com.
Inhibition assays were performed by Reaction Biology Corporation according to their standard assay protocols; see Supporting Information.
b
c
Growth inhibition in cancer cell lines. Values are the mean of three determinations run in duplicate, SD 20%. “−” represents not determined.
a
Scheme 2. Synthesis of Aminotetralin Linked Analogues 17−28
a
Reagents and conditions: (a) NH4OAc, NaBH3CN, rt, overnight; (b) (Boc)2O, NEt3, rt, 2 h; (c) Pd(OAc)2, dppp, CO, 80 °C; (d) HCl/MeOH (1
M), rt, 2 h; (e) aryl iodide or bromide, CuI, L-proline, NEt3, DMF, 120 °C; or aryl chloride, DIPEA, DMF, 150 °C; (f) HONH2, NaOH, MeOH, rt;
(g) MeI, NaH, DMF, rt; (h) arylsulfonyl chloride, NEt3, DCM, rt.
a
Scheme 3. Synthesis of Enantiomers 31 and 32
a
Reagents and conditions: (c) Pd(OAc)2, dppp, CO, 80 °C; (d) HCl/MeOH (1 M), rt, 2 h; (e) aryl iodide or bromide, CuI, L-proline, NEt3, DMF,
120 °C; or aryl chloride, DIPEA, DMF, 150 °C; (f) HONH2, NaOH, MeOH, rt.
that both 31 and 32 have preferential inhibitory effects on
HDAC6 rather than HDAC1−3.
Compound 32 also induced significantly less cell death in those
cell lines than SAHA.
To generate hypotheses about the structural basis of HDAC
isoform selectivity, we performed computational docking of
enantiomers 31 and 32 into the substrate binding pockets of
HDAC1, HDAC6 and HDAC8 using MacroModel and Glide
To confirm their HDAC isoform selectivity profile, we tested
31 and 32 in enzymatic inhibition assays. Indeed, both
compounds were highly potent against HDAC6 with over
100-fold selectivity over HDAC1 (Table 2). The R-config-
uration may be more favored by HDAC8, as 32 was 4−5 times
more potent than 31 in the HDAC8 inhibition assay.
Moreover, 32 was only weakly active or inactive to other
HDAC isoforms with IC50 values above 30 μM. We then tested
antiproliferative effects of 32 in several multiple myeloma and
neuroblastoma cell lines. Compound 32 was several times
weaker than SAHA in cell growth inhibition assays, with GI50
values of 7.7 and 5.4 μM against NCI-H929 (multiple
myeloma) and BE(2)-C (neuroblastoma) cells, respectively.
of Schrodinger (Supporting Information). An HDAC6
̈
homology model was generated from structural alignment of
HDAC6b with HDAH (histone deacetylase-like amidohydro-
lase, 1ZZ1 in PDB)18 in Molecular Operating Environment
(MOE) modules. Among several crystal structures of HDAC8,
we used the X-ray structure of CRA-A in complex with HDAC8
(1VKG in PDB) for docking studies, considering that CRA-A
and its close analogues are selective toward HDAC8 over
HDAC1.19
D
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 2. Compounds 31 and 32 do not have impact on p21 induction but induce tubulin acetylation. (A, left) p21 report gene assay in HeLa cells.
p21-RP3 represents the relative luciferase signal compared to SNDX-275 at 3 μM. p21-RP10 represents the relative level compared to SNDX-275 at
10 μM. In all cases SD < 10%. (B, right) Tubulin acetylation assay in A549 cells, SD 15%.
On the basis of docking predictions, both 31 and 32 should
take optimal binding poses and superimpose well with SAHA
and 12 in the HDAC6 homology model (Figure 3A). The
region of HDAC6 and are partially water exposed. This may
explain the increased induction of tubulin acetylation by
pyrimidine-capped analogues relative to phenyl analogues (e.g.,
20 vs 17) and similar activity against HDAC6 for stereoisomers
31 and 32.
In contrast, it is predicted that 31 and 32 adopt distinct
orientations in the rim region of HDAC8 (Figure 3B). The 3-
pyridylpyrimidine capping group of 32 fits into a deep and
narrow cavity and forms hydrophobic interactions with Pro35,
Phe152, and Tyr306 of HDAC8, while the same group of 32 is
not in close contact with HDAC8. This is consistent with
stronger HDAC8 inhibition by 32 in enzymatic evaluations.
Overall, the 2-R configuration of the aminotetralin linker is
important to HDAC6 and HDAC8 dual inhibition as 32
showed IC50 values of 0.05 and 0.08 μM to them. Furthermore,
32 demonstrated moderate stability in human and mouse liver
microsome tests and moderate solubility at 59 μg/mL based on
LYSA assay. Accordingly, compound 32 was chosen for further
characterization in cellular assays. It has been reported that
inhibition of HDAC8 can induce cell differentiation as assessed
by morphology and biomarker changes.14,20 Thus, BE(2)-C
cells were treated with 32 at 2 μM for 3 and 9 days and
compared with BE(2)-C cells treated with 13-cis-retinoic acid
(RA) as positive control. As a potent HDAC8 inhibitor, 32
induced neuroblastoma cell differentiation as indicated by the
number of neurite outgrowths (Figure 4). Of note, the cell
differentiation effect of 32 in BE(2)-C cells appeared to be time
dependent. As a comparison, SAHA also induced cell
differentiation but was more cytotoxic than 32 under the
same conditions.
Figure 3. Predicted binding models of 31 and 32 in complex with
HDAC6 (A) and HDAC8 (B), respectively. (A) Docking models of
SAHA (magenta), 12 (bronze), 31 (blue), and 32 (green) in the
HDAC6 homology model to HDAH (1ZZ1 in PDB). (B) Docking
models of 31 (blue) and 32 (green) in the CRA-A binding pocket of
HDAC8 (1VKG in PDB). For clarity, SAHA and 12 are omitted from
(B). HDAC6 and HDAC8 are shown in mesh representation with
critical hydrophobic residues labeled in white and carbon atoms in
green and gray colors. Oxygen, nitrogen, and sulfur atoms are shown
in red, blue, and orange, respectively.
aminotetralin linker of 31 and 32 is sandwiched between
Phe620 and Phe680 in the hydrophobic channel and it forms
important hydrophobic interactions similar to the tetrahydro-
isoquinoline moiety of 12. These models suggest that the
HDAC6 hydrophobic channel could accommodate a bulkier
linker than HDAC1 because of the presence of loop 1 in
HDAC6 (Supporting Information). In addition, the capping
groups are positioned in a wide and shallow pocket in the rim
We then treated BE(2)-C cells with increasing concen-
trations of SAHA and 32 for 24 h and assessed the resulting
changes in several proteins of interest by immunoblot (Figure
5). Both SAHA and 32, at all tested concentrations, increased
levels of acetylated tubulin and TrkA (a cell differentiation
marker)20 compared to vehicle control. In a separate time
Figure 4. Aminotetralin 32 induces differentiation of neuroblastoma BE(2)-C cells time dependently. Green arrows indicate neurite outgrowth, and
red arrows indicate dead cells.
E
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
5 μm) column. For SFC chiral separation, intermediates were
separated by chiral column (Daicel chiralpak IC, 5 μm, 30 mm ×
250 mm) column using Mettler Teledo SFC-Multigram system:
solvent system of 95% CO2 and 5% IPA (0.5% TEA in IPA), back
pressure 100 bar, detection UV 254 nm. Optical rotation was
measured in a Rudolph Autopol V automatic polarimeter with
wavelength at 589 nm. LC/MS spectra were obtained using a
MicroMass Platform LC (Waters Alliance 2795-ZQ2000). NMR
spectra were obtained using a Bruker Avance 400 MHz NMR
spectrometer. All final compounds have purities greater than 95%
1
based upon LC/MS, and H NMR.
Preparation of Resin 9. An amount of 2 g of 4-(hydroxymethyl)-
phenoxymethyl-copoly(styrene−l% divinylbenzene) resin (“Wang”
resin, 100−200 mesh, 0.75 mmol/g loading, 1.5 mmol) was suspended
in dry THF (25 mL) and gently agitated for 30 min. N-
Hydroxyphthalimide (735 mg, 4.5 mmol) and triphenylphosphane
(1.18 g, 4.5 mmol) were added, and the mixture was agitated until
these reagents were dissolved. DEAD (783 mg, 4.5 mmol) was added
by cannula to give a bright red solution, and the mixture was shaken at
rt for 18 h. The resin was washed successively with THF, DMF, DCM,
methanol, and finally DCM before being dried in vacuo. The resin was
suspended in ethanol/THF (1:3, 20 mL) at 0 °C, and hydrazine
hydrate (3 mL) was added. The pale yellow solution was warmed to rt
and then gently agitated overnight. Resin 1 was washed with DMF (50
mL), THF (50 mL) under 60 °C and finally with DCM (3 × 20 mL)
before dried in vacuo.
General Procedure for the Preparation of Resin 10. Resin 9 (2 g,
0.75 mmol/g loading, 1.5 mmol) was suspended in DMF (40 mL) and
gently agitated for 30 min. To this slurry were added N-Fmoc
protected carboxylic acid (2.25 mmol), HATU (2.28 g, 6 mmol), and
DIPEA (774 mg, 6 mmol). The mixture was shaken for 24 h. The
resin was collected by filtration, washed successively with DMF (2 ×
100 mL), DCM (100 mL), methanol (2 × 100 mL), DCM (100 mL),
ether (100 mL), and dried.
General Procedure for the Preparation of Resin 11. Resin 10 (2 g,
0.75 mmol/g loading, 1.5 mmol) was suspended in 20% piperidine/
DMF (40 mL) and gently agitated for 30 min. The mixture was shaken
for 1 h. The resin was collected by filtration, washed successively with
DMF (2 × 100 mL), DCM (100 mL), methanol (2 × 100 mL), DCM
(100 mL), ether (100 mL), and dried. Then 200 mg of the obtained
resin (0.75 mmol/g loading, 0.15 mmol) was suspended in MeOH/
DMF (2:1, 2 mL) and gently agitated for 30 min. To this slurry was
added substituted aromatic aldehyde (0.3 mmol) and NaBH3CN (47
mg, 0.75 mmol). The mixture was shaken for 24 h. The resin was
collected by filtration, washed successively with DMF (2 × 100 mL),
DCM (100 mL), methanol (2 × 100 mL), DCM (100 mL), ether
(100 mL), and dried.
General Procedure for the Resin Cleavage and Preparation of
Focused Screening Library. Resin 11 (200 mg, 0.75 mmol/g loading,
0.15 mmol) was suspended in TFA/DCM (1:1, 2 mL) and gently
agitated for 30 min. After filtration, the solution was concentrated and
the crude product was purified by prep-HPLC.
2-[[5-(3-Nitrophenyl)-2-furyl]methyl]-3,4-dihydro-1H-isoquino-
line-7-carbohydroxamic Acid (12). This compound was prepared by
following the SPS procedures as above, with 2-(9H-fluoren-9-
ylmethoxycarbonyl)-3,4-dihydro-1H-isoquinoline-7-carboxylic acid
used in the coupling reaction to form resin 2. HRMS: calcd (MH+)
394.1397, exp (MH+) 394.1396. 1H NMR (MeOD, 400 MHz) δ 8.53
(s, 1H), 8.23−8.25 (d, 1H, J = 8.0 Hz), 8.11−8.13 (d, 1H, J = 8.0 Hz),
7.65−7.74 (m, 4H), 7.47−7.48 (d, 1H, J = 4.0 Hz), 7.40−7.42 (d, 1H,
J = 8.0 Hz), 4.83 (s, 2H), 4.57 (s, 2 H), 3.68−3.71 (m, 2H), 3.27−3.29
(m, 2H).
3-(4-Fluoroanilino)tetralin-6-carbohydroxamic Acid (17). A
mixture of HCl salt of amine 15 (180 mg, 0.88 mmol), 1-fluoro-4-
iodobenzene (124 mg, 0.58 mmol), TEA (0.16 mL, 1.2 mmol), CuI
(11 mg, 0.058 mmol), and L-proline (13 mg, 0.116 mmol) in 3 mL of
DMF was heated in a microwave reactor at 120 °C for 30 min. After
workup with EtOAc and water, the intermediate was mixed with 50%
aqueous NH2OH (1 mL) and KOH (50 mg) in 1 mL of MeOH. After
being stirred at rt for 1 h, the mixture was dissolved in EtOAc, washed
Figure 5. Compound 32 increased Ac-Tub and TrkA but not p21 and
Ac-H3 in BE(2)-C cells.
course study, the TrkA expression level correlated with
differentiation of BE2C cells (data not shown). Interestingly,
SAHA significantly induced HDAC6 but not HDAC8 in
BE(2)-C cells. In contrast to SAHA, 32 did not stimulate
HDAC6 expression and showed minimal effects on p21
induction and histone H3 acetylation even at 10 μM.
CONCLUSIONS
■
We have discovered a novel class of HDAC6/8 dual inhibitors
with 2-R-aminotetralin as a linker and confirmed their HDAC
isoform selectivity in biochemical as well as cellular assays. To
effectively explore the linker and capping groups, we designed
and synthesized a focused library of hydroxamic acids by solid
phase synthesis technology. The reductive amination and resin
carrier were well suited for the preparation of hydroxamic acids
as chemical probe to individual HDAC isoforms. There have
been certain debates on the application of enzymatic assays for
determination of HDAC selectivity, as the substrates used for
individual HDAC isoform profiling may or may not be the
relevant substrate for investigating a specific functional role of a
given HDAC enzyme that exists as multiprotein complex within
the cell.21 For example, artificial trifluoroacetylated substrates
are commonly used to evaluate compound inhibition against
purified HDAC4 and HDAC5 enzymes in vitro.22 Thus, instead
of conducting primary screens with standard enzymatic
inhibition assays, we used Tub-Ac and p21 induction assays
for the identification of putative HDAC6 selective inhibitors
and then subsequently confirmed their selectivity profile in
enzyme inhibition assays. Our efforts led to the discovery of
tetrahydroisoquinoline and aminotetralin moieties as selectivity
vectors and linkers for HDAC6/8 dual inhibitors. Treatment of
BE(2)-C cells with R-aminotetralin 32 resulted in elevated
levels of Tub-Ac and induction of TrkA but not p21.
Accordingly, compound 32 had weaker antiproliferative effects
on cancer cell lines compared with the prototypical pan-HDAC
inhibitor SAHA and thus induced differentiation of neuro-
blastoma cells without significant cytotoxicity. This class of
hydroxamates warrants further exploration as selective
HDAC6/8 inhibitors for the treatment of multiple myeloma
and neuroblastoma based upon the emerging biology of
HDAC6 and HDAC8 in these two diseases.
EXPERIMENTAL SECTION
■
Synthetic Chemistry General Comments. All starting materials
were obtained commercially and were used without further
purification. All reported yields are of isolated products and are not
optimized. Solid phase synthesis was carried out with a MiniBlock
from Mettler Toledo. All final compounds were purified by preparative
HPLC on reverse phase column using Waters XBridge OBD Phenyl
(30 mm × 100 mm, 5 μm) column or OBD RP18 (30 mm × 100 mm,
F
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
with water, and concentrated. The crude product was purified by
HPLC. MS: calcd (MH+) 301.1, exp (MH+) 301.1. HRMS: calcd
(MH+) 301.1347, exp (MH+) 301.1346. 1HNMR (CD3OD, 400
MHz) δ 7.52 (d, 2H, J = 8.0 Hz), 7.24 (m, 1H), 7.12 (m, 4H), 3.85
(m, 1H), 3.22 (m, 1H), 3.05−2.84 (m, 3H), 2.26 (m, 1H), 1.78 (m,
1H).
3-(3-Chloro-4-cyanoanilino)tetralin-6-carbohydroxamic Acid
(18). MS: calcd (MH+) 342.1, exp (MH+) 342.1. HRMS: calcd
(MH+) 342.1004, exp (MH+) 342.1004. 1H NMR (CD3OD, 400
MHz) δ 7.64 (m, 2H), 7.51 (m, 2H), 7.24 (d, 1H, J = 8.8 Hz), 6.95 (d,
1H, J = 8.8 Hz), 3.98 (m, 1H), 3.27 (m, 1H), 3.04 (m, 2H), 2.97 (m,
1H), 2.24 (m, 1H), 1.87 (m, 1H).
The ester intermediate was treated with aqueous NH2OH (1 mL) and
KOH (50 mg), and the crude product was purified by HPLC. MS:
calcd (MH+) 365.1, exp (MH+) 365.4. HRMS: calcd (MH+) 365.0966,
exp (MH+) 365.0965. 1H NMR (CD3OD, 400 MHz), 7.75 (d, 1H, J =
8.0 Hz) δ 7.60−7.66 (m, 2H), 7.46 (d, 1H, J = 8.0 Hz), 7.41 (m, 1H),
7.36 (s, 1H), 7.14 (d, 1H, J = 8.0 Hz), 3.55 (m, 1H), 2.95 (m, 2H),
2.82 (m, 1H), 2.69 (m, 1H), 1.95 (m, 1H) 1.73 (m, 1H).
3-[(3-Chlorophenyl)sulfonylamino]tetralin-6-carbohydroxamic
Acid (28). MS: calcd (MH+) 381.1, exp (MH+) 381.2. HRMS: calcd
(MH+) 381.0670, exp (MH+) 381.0668. 1H NMR (CD3OD, 400
MHz), 7.90 (m, 1H), 7.84 (d, 1H, J = 8.0 Hz) δ 7.66 (d, 1H, J = 8.4
Hz), 7.59 (t, 1H, J = 8.0 Hz), 7.47 (d, 1H, J = 8.0 Hz), 7.36 (s, 1H),
7.15 (d, 1H, J = 8.0 Hz), 3.55 (m, 1H), 2.95 (m, 2H), 2.82 (m, 1H),
2.69 (m, 1H), 1.95 (m, 1H) 1.73 (m, 1H).
3-(2-Pyridylamino)tetralin-6-carbohydroxamic Acid (19). MS:
calcd (MH+) 284.1, exp (MH+) 284.3. HRMS: calcd (MH+)
1
284.1394, exp (MH+) 284.1395. H NMR (CD3OD, 400 MHz) δ
(3S)-3-[[4-(3-Pyridyl)pyrimidin-2-yl]amino]tetralin-6-carbohy-
droxamic Acid (31). Compound 31 was prepared from chiral
intermediate 29 after SFC separation of 14. MS: calcd (MH+)
7.85−7.93 (m,2H), 7.55 (m, 2H), 7.25 (d, 1H, J = 8 Hz), 7.12 (d, 1H,
J = 8.8 Hz), 6.92 (t, 1H, J = 6.4 Hz), 4.79 (m, 1H), 2.8−3.2 (m, 4H),
2.2 (m, 1H), 1.9 (m, 1H).
1
362.1, exp (MH+) 362.1. H NMR (CD3OD, 400 MHz) δ 9.55 (s,
3-(Pyrimidin-2-ylamino)tetralin-6-carbohydroxamic Acid (20). A
mixture of 15 (60 mg), 2-chloropyrimidine (50 mg), and TEA (0.2
mL) in 2 mL of DMF was heated at 150 °C for 1 h in a microwave
reactor. After concentration, the crude product was dissolved in 1 mL
of MeOH and mixed with a solution of 50% aqueous NH2OH (1 mL)
and KOH (50 mg). After being stirred at rt for 1 h, the mixture was
concentrated and purified by HPLC. MS: calcd (MH+) 285.1, exp
(MH+) 285.3. HRMS: calcd (MH+) 285.1346, exp (MH+) 285.1346.
1H NMR (CD3OD, 400 MHz) δ 8.42 (d, 2H, J = 4.8 Hz), 7.22−7.52
1H), 9.13 (d, 1H, J = 6.8 Hz), 8.95 (s, 1H), 8.52 (d, 1H, J = 4.8 Hz),
8.07 (m, 1H), 7.53 (m, 3H), 7.24(m, 1H), 3.34(m,1H), 2.9−3.07 (m,
4H), 2.29 (m, 1H), 1.98 (m, 1H).
(3R)-3-[[4-(3-Pyridyl)pyrimidin-2-yl]amino]tetralin-6-carbohy-
droxamic Acid (32). Compound 32 was prepared from chiral
20
intermediate 29 after SFC separation of 14. Optical rotation, [α]D
+45.2° (0.115% g/100 mL, MeOH). MS: calcd (MH+) 362.1, exp
(MH+) 362.2. HRMS: calcd (MH+) 362.1612, exp (MH+) 362.1610.
1H NMR (CD3OD, 400 MHz) δ 9.51 (s, 1H), 9.08 (d, 1H, J = 7.6
Hz), 8.92 (m, 1H), 8.51 (d, 1H, J = 4.8 Hz), 8.00 (m, 1H), 7.52 (m,
3H), 7.25 (d, 1H, 7.6 Hz), 3.34 (m,1H), 2.9−3.07 (m, 4H), 2.29 (m,
1H), 1.98 (m, 1H).
(m, 3H), 6.8 (d, 1H, J = 4.8 Hz), 4.30 (m, 1H), 2.8−3.2 (m, 4H), 2.25
(m, 1H), 1.9 (m, 1H).
3-[(4-Phenylpyrimidin-2-yl)amino]tetralin-6-carbohydroxamic
Acid (21). MS: calcd (MH+) 361.1, exp (MH+) 361.4. HRMS: calcd
(MH+) 361.1659, exp (MH+) 361.1658. 1H NMR (CD3OD, 400
MHz) δ 8.34 (d, 1H, J = 6 Hz), 8.22 (d, 2H, J = 7.6 Hz), 7.5−7.7 (m,
5H), 7.42 (d, 1H, J = 5.6 Hz), 7.25 (d, 1H, J=8 Hz), 4.41 (s, 1H), 3.30
(m, 1H), 2.9−3.07 (m, 3H), 2.29 (m, 1H), 1.98 (m, 1H).
3-[[4-(4-Chlorophenyl)pyrimidin-2-yl]amino]tetralin-6-carbohy-
droxamic Acid (22). MS: calcd (MH+) 395.1, exp (MH+) 395.2.
HRMS: calcd (MH+) 395.1269, exp (MH+) 395.1268. 1H NMR
(CD3OD, 400 MHz) δ 8.35 (d, 1H, J = 6.0 Hz), 8.20 (d, 2H, J = 8.4
Hz), 7.55 (m, 4H), 7.35 (d, 1H, J = 5.6 Hz), 7.25 (d, 1H, J = 8.0 Hz),
4.41 (s, 1H), 3.30 (m, 1H), 3.05 (t, 2H, J = 5.6 Hz), 2.90 (m, 1H),
2.28 (m, 1H), 1.94 (m, 1H).
3-[[4-(3-Chlorophenyl)pyrimidin-2-yl]amino]tetralin-6-carbohy-
droxamic Acid (23). MS: calcd (MH+) 395.1, exp (MH+) 395.2.
HRMS: calcd (MH+) 395.1269, exp (MH+) 395.1267. 1H NMR
(CD3OD, 400 MHz) δ 8.35 (d, 1H, J = 5.2 Hz), 8.14 (m, 1H), 8.02
(m, 1H), 7.46−7.54 (m, 4H), 7.25 (d, 1H, J = 8.4 Hz), 7.15 (d, 1H, J =
5.2 Hz), 4.41 (s, 1H), 3.30 (m, 1H), 3.05 (t, 2H, J = 5.6 Hz), 2.90 (m,
1H), 2.28 (m, 1H), 1.94 (m, 1H).
Molecular Modeling. Homology modeling and molecular
dynamics simulations were conducted in the molecular operating
environment (MOE), and geometric optimizations of compounds and
molecular docking were finished by MacroModel and Glide of
Schrodinger, respectively. The sequence corresponding to HDAC6b
̈
was aligned with that of HDAH in the MOE-Align module. The
HDAC6 homology model was constructed by using 1ZZ1 (PDB) as
template, and the final model was taken as the Cartesian average of all
the intermediate models and was further refined by molecular dynamic
simulations and energy minimizations, during which only the side
chains of all residues were allowed to move. The molecular dynamics
simulations were set to 5 ps of heating to 300 K, 10 ps of equilibrium
at 300 K, and 5 ps of cooling to 0 K. For docking studies in HDAC8,
the cocrystal structure of CRA-A in complex with HDAC8 (1VKG in
PDB) was used. OPLS_2005 force field of MacroModel was used to
optimize the initial geometries of the compounds with a final root-
mean-square gradient of 0.01 kcal mol−1 Å−1. The default values of the
optimization parameters and thresholds were kept. All torsion angles
of compound were released to freely rotate, and 10 000 docking runs
were adopted during the XP docking with NE2 atom of His residues
and zinc ion as docking constraints. Docking poses were clustered
based on heavy atom rmsd values (0.5 Å), and top-ranked poses of
each compound were selected and viewed graphically within Maestro
3-[[4-(2-Chlorophenyl)pyrimidin-2-yl]amino]tetralin-6-carbohy-
1
droxamic Acid (24). MS: calcd (MH+) 395.1, exp (MH+) 395.2. H
NMR (CD3OD, 400 MHz) δ 8.35 (d, 1H, J = 4.8 Hz), 7.60 (m, 1H),
7.54 (m, 3H), 7.43 (m, 2H), 7.22 (d, 1H, J = 8.4 Hz), 6.90 (d, 1H, J =
5.2 Hz), 4.41 (s, 1H), 3.30 (m, 1H), 3.05 (t, 2H, J = 5.6 Hz), 2.90 (m,
1H), 2.28 (m, 1H), 1.94 (m, 1H).
of the Schrodinger program and finally displayed in Pymol.
̈
Microsomal Stability Assay. Each incubated mixture contained
0.5 mg/mL liver microsome (human or mouse), 100 mM potassium
phosphate buffer (pH 7.4), 10 mM NADPH, and 1 μM test
compound in a total volume of 400 μL. After prewarming at 37 °C for
10 min, the NADPH was added to initiate the reaction. Reaction was
terminated after 0, 3, 6, 9, 15, and 30 min by adding 150 μL of 100 ng/
mL of tolbutamide (internal standard) in ice cold methanol into 300
μL of incubation mixtures. Samples were then centrifuged at 4000 rpm
for 10 min at 4 °C. The supernatant was then analyzed by LC−MS/
MS.
LYSA Solubility Assay. An amount of 150 μL of 10 mM DMSO
stock solution of compound was prepared and divided into two
portions. For one portion the DMSO solution was evaporated to
dryness at 35 °C in a centrifugal vacuum evaporator from Genevac
Technologies, and the remaining was redissolved in 50 mM phosphate
buffer (pH 6.5). The mixture was stirred and shaken for 1−2 h. After
3-[[4-(3-Pyridyl)pyrimidin-2-yl]amino]tetralin-6-carbohydroxa-
mic Acid (25). MS: calcd (MH+) 362.1, exp (MH+) 362.4. HRMS:
calcd (MH+) 362.1612, exp (MH+) 362.1610. 1H NMR (CD3OD, 400
MHz) δ 9.48 (s, 1H), 8.99 (d, 1H, J = 8 Hz), 8.88 (s, 1H), 8.48 (d,
1H, J = 4.8 Hz), 7.99 (m, 1H), 7.52 (m, 2H), 7.43 (d, 1H, J = 4.4 Hz),
7.24 (d, 1H, J = 8.4 Hz), 2.8−3.2 (m, 5H), 2.29 (m, 1H), 2.00 (m,
1H).
3-[Methyl-[4-(3-pyridyl)pyrimidin-2-yl]amino]tetralin-6-carbohy-
droxamic Acid (26). MS: calcd (MH+) 376.1, exp (MH+) 376.4.
HRMS: calcd (MH+) 376.1768, exp (MH+) 376.1766. 1H NMR
(CD3OD, 400 MHz) δ 9.34 (s, 1H), 8.72 (m, 2H), 8.50 (d, 1H, J = 4.8
Hz), 7.5 (m, 1H), 7.4−7.5 (m, 4 Hz), 5.20 (s, 2H), 3.2 (s, 3H), 2.9−
3.07 (m, 4H), 2.12 (m, 2H).
3-[(3-Fluorophenyl)sulfonylamino]tetralin-6-carbohydroxamic
Acid (27). A mixture of HCl salt of amine 11, 3-fluorobenzenesulfonyl
chloride, and TEA in DCM was stirred for 30 min and concentrated.
G
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
an overnight stand, the solution was filtered before HPLC analysis.
The other portion was used for the calibration curve preparation by
diluting the DMSO stock solution using the same PBS buffer
mentioned above to a series of solutions with concentration range of
50−500 μM.
Cell Proliferation (Water-Soluble Tetrazolium Dye) Assay.
Cells were seeded in 96-well culture plates (200 μL/well at different
seeding concentrations depending on cell type) and incubated
overnight at 37 °C in 5% CO2. After addition of compound (typically
eight-point half-log serial dilutions in triplicate) to the cells (DMSO
concentration kept below 0.5%), the cell culture was incubated at 37
°C in 5% CO2 for 72 h. The effects on proliferation were determined
by addition of CCK-8 reagent (Dojindo) according to the
manufacturer’s instruction, followed by a 2 h incubation at 37 °C in
5% CO2 and detection of the absorbance at 450 nm using an ELISA
plate reader.
Biology General Comments. Brief descriptions of the biological
protocols used to generate data in this manuscript can be found below.
All growth media included 100 units/mL penicillin and 100 μg/mL
streptomycin, and all cell lines were grown in 5% CO2 at 37 °C. NCI-
H929, A549, HeLa, OPM-2, and BE(2)-C cells were obtained from
ATCC. Mammalian protein extraction reagent and tissue protein
extraction reagent were obtained from Thermo Scientific. EDTA-free
protease inhibitors were obtained from Roche Diagnostics. Mono-
clonal anti-p21 antibody and antiacetylated histone H3 (Ac 1−20)
antibodies were purchased from Calbiochem. Anti-TrkA and anti-
HDAC6 were from Cell Signaling Technology. Anti-HDAC8 was from
Millipore. HRP-conjugated monoclonal anti-GAPDH, peroxidase-
conjugated goat anti-rabbit IgG and anti-mouse IgG were purchased
from KangChen. Enhanced chemiluminescence (ECL) detection kit
was obtained from Thermo Scientific.
Tubulin Acetylation (Tub-Ac) Cytoblot Assay. Tubulin
acetylation was detected by the antiacetylated tubulin antibody
(Sigma) and horseradish peroxidase (HRP) conjugated secondary
antibody (KangChen Bio-Tech). A549 cells were seeded into assay
plates (Corning 3912) at a concentration of 1 × 105 cells/mL and
incubated for 16−18 h at 37 °C in the presence of 5% CO2. Then 20
μL of diluted compound solution was transferred to the cell culture
plate and incubated for 17−18 h. After medium removal and fixation
by formaldehyde (3.7% paraformaldehyde in TBS), the cells in the
plates were treated with 180 μL of −20 °C MeOH and incubated for 5
min at rt. The cell lysis was incubated with 75 μL of primary
antiacetylated tubulin antibody and secondary HRP conjugated
antibody solution (1:750 antiacetylated tubulin, 1:750 HRP
conjugated anti-mouse IgG in antibody dilution buffer not containing
sodium azide) for 4 h at 4 °C. Then 75 μL of enhanced
chemiluminescence (ECL) reagent was added into the wells, and
the luminescence from each well was immediately quantified by the
plate reader. On the basis of the luminescence reading, the EC50 values
against Tub-Ac were calculated by plotting the curve with XLfit 4.0
software.
Western Blotting. 1 ×106 BE(2)-C cells were seeded overnight
and then incubated with indicated concentrations of SAHA and 32 for
24 h. Cell extract was prepared by lysing cultured cells with a
mammalian protein extraction reagent supplemented with EDTA-free
protease inhibitor for 15 min. Supernatants were collected following
centrifugation of lysed cells (15 000 g) for 10 min at 4 °C. To analyze
the cell lysate, 30 μg of total protein per sample was resolved by SDS−
PAGE and transferred onto nitrocellulose membrane. Membranes
with immobilized proteins were probed with antibodies for HDAC6,
HDAC8, TrkA, Tub-Ac, and p21. The reactive protein bands were
visualized using an enhanced chemiluminescence (ECL) detection
system.
ASSOCIATED CONTENT
* Supporting Information
■
S
Detailed experimental procedures for the synthesis of solid
phase library, and additional biological profiling of 12, 31, and
32. This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Phone: +86 21 38954910, extension 3350. Fax: +86 21
50790291. E-mail: gordon.tang@roche.com.
■
Present Address
†J.C.W.: 14 Moonbeam Drive, Mountain View, CA 94043, U.S.
Notes
p21 Reporter Gene Assay. Compounds were tested for their
ability to induce p21 gene expression using a reporter gene assay
involving HeLa cells transfected with a p21 promoter−luciferase
construct. The p21 promoter has the Sp1/Sp3 binding site for HDAC
but not the upstream p53 binding site. Briefly, the day before
transfection, HeLa cells were seeded at 11 000 cells/well in a 96-well
culture plate and incubated at 37 °C in 5% CO2 overnight. For
transfection, the medium was removed and replaced with 100 μL/well
transfection media prepared by following procedure: 5 μL of serum-
free DMEM, 0.15 μL of Fugene 6 reagent, 40 ng of p21-luc, and 10 ng
of GFP were mixed and incubated at room temperature for 30 min.
Then 98 μL of DMEM (with 10% FBS, 1% penicillin and
streptomycin) was added to the DNA/Fugene 6 reagent complex
and mixed gently. After incubation of the cells for 24 h at 37 °C under
5% CO2, fresh medium and test compounds were added to the wells
and incubated for 15 h. Cells were lysed by adding 80 μL/well of
culture lysis reagent (Promega). An amount of 50 μL of each lysate
was taken for GFP detection using an excitation wavelength of 486 nm
and detection at 527 nm. Then 100 μL of luciferase assay reagent
(Promega) was added to 20 μL of cell lysate for luminometer reading.
The relative gene level of p21 induced by individual compound was
compared to SNDX-275 at 1, 3, and 10 μM concentrations.
HDAC Enzyme Inhibition Assay. HDAC enzyme inhibition
assays were conducted by Reaction Biology Corporation using a 10-
point dose−response curve with half-log serial dilutions, fluorogenic
peptides at 50 μM as enzymatic substrate, and trichostatin A as a
positive control unless otherwise indicated. TMP-269 was used as
positive control for the testing of 31 and 32 against class IIa HDACs
(HDAC4, HDAC5, HDAC7, and HDAC9). See details at www.
reactionbiology.com.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to Sheng Zhong, Wenzhi Cheng, Qinshan Gao,
and Peilan Ding for their analytical assistance and purification
of final compounds. We also thank Yuxia Zhang, Yi Zhang,
Hongxia Qiu, and Xiaoyue Wu for carrying out the
physicochemical and microsome stability studies.
ABBREVIATIONS USED
■
HDAC, histone deacetylase; RA, 13-cis-retinoic acid; SAHA,
suberoylanilide hydroxamic acid; SAR, structure−activity
relationship; ZBG, zinc-binding group
REFERENCES
■
(1) Minucci, S.; Pelicci, P. G. Histone deacetylase inhibitors and the
promise of epigenetic (and more) treatments for cancer. Nat. Rev.
Cancer 2006, 6, 38−51.
(2) (a) Weiwer, M.; Lewis, M. C.; Wagner, F. F.; Holson, E. B.
Therapeutic potential of isoform selective HDAC inhibitors for the
treatment of schizophrenia. Future Med. Chem. 2013, 5, 1491−1508.
(b) McKinsey, T. A. Isoform-selective HDAC inhibitors: closing in on
translational medicine for the heart. J. Mol. Cell. Cardiol. 2011, 51,
491−496.
(3) Wong, J. C.; Hong, R.; Schreiber, S. L. Structural biasing
elements for in-cell histone deacetylase paralog selectivity. J. Am.
Chem. Soc. 2003, 125, 5586−5587.
H
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(4) Santo, L.; Hideshima, T.; Kung, A. L.; Tseng, J. C.; Tamang, D.;
Yang, M.; Jarpe, M.; van Duzer, J. H.; Mazitschek, R.; Ogier, W. C.;
Cirstea, D.; Rodig, S.; Eda, H.; Scullen, T.; Canavese, M.; Bradner, J.;
Anderson, K. C.; Jones, S. S.; Raje, N. Preclinical activity,
pharmacodynamic, and pharmacokinetic properties of a selective
HDAC6 inhibitor, ACY-1215, in combination with bortezomib in
multiple myeloma. Blood 2012, 119, 2579−2589.
(5) (a) Butler, K. V.; Kalin, J.; Brochier, C.; Vistoli, G.; Langley, B.;
Kozikowski, A. P. Rational design and simple chemistry yield a
superior, neuroprotective HDAC6 inhibitor, tubastatin A. J. Am. Chem.
Soc. 2010, 132, 10842−10846. (b) Kalin, J. H.; Butler, K. V.; Akimova,
T.; Hancock, W. W.; Kozikowski, A. P. Second-generation histone
deacetylase 6 inhibitors enhance the immunosuppressive effects of
Foxp3+ T-regulatory cells. J. Med. Chem. 2012, 55, 639−651.
(6) Lee, J. H.; Mahendran, A.; Yao, Y. S.; Ngo, L.; Venta-Perez, G.;
Choy, M. L.; Kim, N.; Ham, W. S.; Breslow, R.; Marks, P. A.
Development of a histone deacetylase 6 inhibitor and its biological
effects. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 15704−15709.
(7) Wagner, F. F.; Olson, D. E.; Gale, J. P.; Kaya, T.; Weiwer, M.;
Aidoud, N.; Thomas, M.; Davoine, E. L.; Lemercier, B. C.; Zhang, Y.
L.; Holson, E. B. Potent and selective inhibition of histone deacetylase
6 (HDAC6) does not require a surface-binding motif. J. Med. Chem.
2013, 56, 1772−1776.
(8) Balasubramanian, S.; Ramos, J.; Luo, W.; Sirisawad, M.; Verner,
E.; Buggy, J. J. A novel histone deacetylase 8 (HDAC8)-specific
inhibitor PCI-34051 induces apoptosis in T-cell lymphomas. Leukemia
2008, 22, 1026−1034.
(9) (a) Suzuki, T.; Ota, Y.; Ri, M.; Bando, M.; Gotoh, A.; Itoh, Y.;
Tsumoto, H.; Tatum, P. R.; Mizukami, T.; Nakagawa, H.; Iida, S.;
Ueda, R.; Shirahige, K.; Miyata, N. Rapid discovery of highly potent
and selective inhibitors of histone deacetylase 8 using click chemistry
to generate candidate libraries. J. Med. Chem. 2012, 55, 9562−9575.
(b) Suzuki, T.; Muto, N.; Bando, M.; Itoh, Y.; Masaki, A.; Ri, M.; Ota,
Y.; Nakagawa, H.; Iida, S.; Shirahige, K.; Miyata, N. Design, synthesis,
and biological activity of NCC149 derivatives as histone deacetylase 8-
selective inhibitors. ChemMedChem 2014, 9, 657−664.
(15) Olson, D. E.; Wagner, F. F.; Kaya, T.; Gale, J. P.; Aidoud, N.;
Davoine, E. L.; Lazzaro, F.; Weïwer, M.; Zhang, Y.-L.; Holson, E. B.
Discovery of the first histone deacetylase 6/8 dual inhibitors. J. Med.
Chem. 2013, 56, 4816−4820.
(16) Hubbert, C.; Guardiola, A.; Shao, R.; Kawaguchi, Y.; Ito, A.;
Nixon, A.; Yoshida, M.; Wang, X. F.; Yao, T. P. HDAC6 is a
microtubule-associated deacetylase. Nature 2002, 417, 455−458.
(17) Wong, J. C.; Guo, L.; Peng, Z.; Zhang, W.; Zhang, N.; Lai, W.;
Zhang, Z.; Zhang, C.; Zhang, X.; Song, S.; Pan, D.; Xie, C.; Li, J.; Ning,
Z.; Lu, X.; He, Y.; Chen, L. Application of p21 and klf2 reporter gene
assays to identify selective histone deacetylase inhibitors for cancer
therapy. Bioorg. Med. Chem. Lett. 2011, 21, 110−116.
(18) Nielsen, T. K.; Hildmann, C.; Dickmanns, A.; Schwienhorst, A.;
Ficner, R. Crystal structure of a bacterial class 2 histone deacetylase
homologue. J. Mol. Biol. 2005, 354, 107−120.
(19) Vannini, A.; Volpari, C.; Filocamo, G.; Casavola, E. C.; Brunetti,
M.; Renzoni, D.; Chakravarty, P.; Paolini, C.; De Francesco, R.;
Gallinari, P.; Steinkuhler, C.; Di Marco, S. Crystal structure of a
eukaryotic zinc-dependent histone deacetylase, human HDAC8,
complexed with a hydroxamic acid inhibitor. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 15064−15069.
(20) Shikata, A.; Shikata, T.; Sotozono, Y.; Hosoi, H.; Matsumura, T.;
Sugimoto, T.; Sawada, T. Neuronal differentiation in human
neuroblastoma cells by nerve growth factor following TrkA up-
regulation by interferon-gamma. Med. Pediatr. Oncol. 2000, 34, 394−
401.
(21) Bantscheff, M.; Hopf, C.; Savitski, M. M.; Dittmann, A.; Grandi,
P.; Michon, A. M.; Schlegl, J.; Abraham, Y.; Becher, I.; Bergamini, G.;
Boesche, M.; Delling, M.; Dumpelfeld, B.; Eberhard, D.; Huthmacher,
C.; Mathieson, T.; Poeckel, D.; Reader, V.; Strunk, K.; Sweetman, G.;
Kruse, U.; Neubauer, G.; Ramsden, N. G.; Drewes, G. Chemo-
proteomics profiling of HDAC inhibitors reveals selective targeting of
HDAC complexes. Nat. Biotechnol. 2011, 29, 255−265.
(22) Lahm, A.; Paolini, C.; Pallaoro, M.; Nardi, M. C.; Jones, P.;
Neddermann, P.; Sambucini, S.; Bottomley, M. J.; Lo Surdo, P.; Carfi,
A.; Koch, U.; De Francesco, R.; Steinkuhler, C.; Gallinari, P.
Unraveling the hidden catalytic activity of vertebrate class Ila histone
deacetylases. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 17335−17340.
(10) Khan, N.; Jeffers, M.; Kumar, S.; Hackett, C.; Boldog, F.;
Khramtsov, N.; Qian, X. Z.; Mills, E.; Berghs, S. C.; Carey, N.; Finn, P.
W.; Collins, L. S.; Tumber, A.; Ritchie, J. W.; Jensen, P. B.;
Lichenstein, H. S.; Sehested, M. Determination of the class and
isoform selectivity of small-molecule histone deacetylase inhibitors.
Biochem. J. 2008, 409, 581−589.
(11) Liu, L.; Chen, B.; Qin, S.; Li, S.; He, X.; Qiu, S.; Zhao, W.; Zhao,
H. A novel histone deacetylase inhibitor chidamide induces apoptosis
of human colon cancer cells. Biochem. Biophys. Res. Commun. 2010,
392, 190−195.
(12) Wong, J. C.; Tang, G.; Wu, X.; Liang, C.; Zhang, Z.; Guo, L.;
Peng, Z.; Zhang, W.; Lin, X.; Wang, Z.; Mei, J.; Chen, J.; Pan, S.;
Zhang, N.; Liu, Y.; Zhou, M.; Feng, L.; Zhao, W.; Li, S.; Zhang, C.;
Zhang, M.; Rong, Y.; Jin, T. G.; Zhang, X.; Ren, S.; Ji, Y.; Zhao, R.;
She, J.; Ren, Y.; Xu, C.; Chen, D.; Cai, J.; Shan, S.; Pan, D.; Ning, Z.;
Lu, X.; Chen, T.; He, Y.; Chen, L. Pharmacokinetic optimization of
class-selective histone deacetylase inhibitors and identification of
associated candidate predictive biomarkers of hepatocellular carcinoma
tumor response. J. Med. Chem. 2012, 55, 8903−8925.
(13) (a) Yang, P. H.; Zhang, L.; Zhang, Y. J.; Zhang, J.; Xu, W. F.
HDAC6: physiological function and its selective inhibitors for cancer
treatment. Drug Discoveries Ther. 2013, 7, 233−242. (b) Zhang, L.;
Sheng, S.; Qin, C. The role of HDAC6 in Alzheimer’s disease. J.
Alzheimer’s Dis. 2013, 33, 283−295. (c) Valenzuela-Fernandez, A.;
Cabrero, J. R.; Serrador, J. M.; Sanchez-Madrid, F. HDAC6: a key
regulator of cytoskeleton, cell migration and cell-cell interactions.
Trends Cell Biol. 2008, 18, 291−297. (d) Dallavalle, S.; Pisano, C.;
Zunino, F. Development and therapeutic impact of HDAC6-selective
inhibitors. Biochem. Pharmacol. 2012, 84, 756−765.
(14) Oehme, I.; Deubzer, H. E.; Lodrini, M.; Milde, T.; Witt, O.
Targeting of HDAC8 and investigational inhibitors in neuroblastoma.
Expert Opin. Invest. Drugs 2009, 18, 1605−1617.
I
dx.doi.org/10.1021/jm5008962 | J. Med. Chem. XXXX, XXX, XXX−XXX