Design, Synthesis, and Biological Evaluation of Tetrahydroisoquinoline-Based Histone Deacetylase 8 Selective Inhibitors

Article abstract of DOI:10.1021/acsmedchemlett.7b00126

Histone deacetylase 8 (HDAC8) is a promising drug target for multiple therapeutic applications. Here, we describe the modeling, design, synthesis, and biological evaluation of a novel series of C1-substituted tetrahydroisoquinoline (TIQ)-based HDAC8 inhibitors. Minimization of entropic loss upon ligand binding and use of the unique HDAC8 "open" conformation of the binding site yielded a successful strategy for improvement of both HDAC8 potency and selectivity. The TIQ-based 3g and 3n exhibited the highest 82 and 55 nM HDAC8 potency and 330- and 135-fold selectivity over HDAC1, respectively. Selectivity over other class I isoforms was comparable or better, whereas inhibition of HDAC6, a class II HDAC isoform, was below 50% at 10 μM. The cytotoxicity of 3g and 3n was evaluated in neuroblastoma cell lines, and 3n displayed concentration-dependent cytotoxicity similar to or better than that of PCI-34051. The selectivity of 3g and 3n was confirmed in SH-SY5Y cells as both did not increase the acetylation of histone H3 and α-tubulin. Discovery of the novel TIQ chemotype paves the way for the development of HDAC8 selective inhibitors for therapeutic applications.

Full text of DOI:10.1021/acsmedchemlett.7b00126

Letter
pubs.acs.org/acsmedchemlett
Design, Synthesis, and Biological Evaluation of
Tetrahydroisoquinoline-Based Histone Deacetylase 8 Selective
Inhibitors
Taha Y. Taha,† Shaimaa M. Aboukhatwa,†,‡ Rachel C. Knopp,† Naohiko Ikegaki,§ Hazem Abdelkarim,†,⊥
Jayaprakash Neerasa,†,∥ Yunlong Lu,† Raghupathi Neelarapu,†,∇ Thomas W. Hanigan,†
Gregory R. J. Thatcher,† and Pavel A. Petukhov*,†
†Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, University of Illinois at Chicago, Chicago, Illinois
60612, United States
‡Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Tanta University, Tanta 31527, Egypt
§Department of Anatomy and Cell Biology, College of Medicine, University of Illinois at Chicago, Chicago, Illinois 60612, United
States
S
* Supporting Information
ABSTRACT: Histone deacetylase 8 (HDAC8) is a promising drug
target for multiple therapeutic applications. Here, we describe the
modeling, design, synthesis, and biological evaluation of a novel
series of C1-substituted tetrahydroisoquinoline (TIQ)-based
HDAC8 inhibitors. Minimization of entropic loss upon ligand
binding and use of the unique HDAC8 “open” conformation of the
binding site yielded a successful strategy for improvement of both
HDAC8 potency and selectivity. The TIQ-based 3g and 3n exhibited
the highest 82 and 55 nM HDAC8 potency and 330- and 135-fold selectivity over HDAC1, respectively. Selectivity over other
class I isoforms was comparable or better, whereas inhibition of HDAC6, a class II HDAC isoform, was below 50% at 10 μM.
The cytotoxicity of 3g and 3n was evaluated in neuroblastoma cell lines, and 3n displayed concentration-dependent cytotoxicity
similar to or better than that of PCI-34051. The selectivity of 3g and 3n was confirmed in SH-SY5Y cells as both did not increase
the acetylation of histone H3 and α-tubulin. Discovery of the novel TIQ chemotype paves the way for the development of
HDAC8 selective inhibitors for therapeutic applications.
KEYWORDS: Histone deacetylase, HDAC8, hydroxamate, inhibitor, tetrahydroisoquinoline
istone deacetylases (HDACs), also known as lysine
deacetylases, are enzymes that catalyze the removal of
approved pan-HDAC inhibitors.14 HDAC8, a member of the
class I HDAC family, has been investigated as a potential target in
a variety of disease states, including neuroblastoma,15,16 acute
myelogenous leukemia,17 colon cancer,18 schistosomiasis,19 and
others.20 Recent analysis of tissues from breast cancer patients
indicates that hypomethylation of HDAC8 promoter and
HDAC8 overexpression correlate with disease progression and
overall poor prognosis.21 To that end, HDAC8 represents a
promising therapeutic target with a somewhat limited set of
potent isoform-selective inhibitors (Figure 1).16,20,22 Only
aromatic hydroxamic acids are potent HDAC8 inhibitors,
whereas aliphatic ones are less potent and inhibit other
isoforms.23−25 In this Letter, we present computer-aided
molecular design, synthesis, and biological evaluation of novel
amine-based series 1 and 2 and tetrahydroisoquinoline (TIQ)-
based HDAC8 selective inhibitors 3.
H
acetyl moieties from lysine ε-amino group residues in histones as
well as nonhistone substrates. Deacetylation of histone substrates
results in an overall change in the post-translational state of
histones, known as the “histone code.”1,2 The list of cellular
events controlled by HDACs has grown beyond DNA
replication, DNA repair, chromatin remodeling, and gene
transcription to the world of noncoding microRNA, and it
continues to expand.3,4 Normal regulation of these processes is
compromised in a variety of disease states, and altered HDAC
expression/function has been shown to be a hallmark of many
cancers and neurodegenerative and inflammatory diseases.4−9
Because of the roles HDACs play in these diseases, HDAC
inhibitors have emerged as therapeutics, with four FDA-
approved HDAC inhibitors available to treat peripheral or
cutaneous T-cell lymphoma and multiple myeloma.
Increasing evidence shows that HDAC isoforms carry out
unique biological functions.10−12 Therefore, it has been
proposed that desired therapeutic effects can be achieved by
inhibiting a single or a subset of HDAC isoforms.13 This would
also aid in minimizing clinical toxicities exhibited by the currently
Received: March 22, 2017
Accepted: July 26, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acsmedchemlett.7b00126
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
that shortening the linker in secondary amines 1a−c causes the
benzyl moiety to bind deeper in the binding site. Overall, the
secondary amines appear to be too small to cause pronounced
conformational changes in the binding site, such as those found
in the “open” conformation of HDAC8. Limited flexibility of the
linker associated with a fewer number of methylene groups also
limits the ability of the benzyl group to form more favorable
hydrophobic interactions with aromatic residues lining the gorge
region of the binding site. Altogether these observations may
explain general loss of activity of secondary amines 1a−c against
class I HDAC isoforms. In the “closed” conformation of HDAC8,
the two benzyl groups of tertiary amines 2a−c, and especially 2c,
share the rather narrow gorge of the binding site. The distance
between the hydroxamate group in these ligands and the zinc ion
in the binding site is longer, 2.6−2.9 Å for Zn2+···O bonds,
compared with <2 Å for the ligands crystallized with HDAC8,
and the overall geometry of the zinc complex is distorted.28 Such
pronounced deviations from the parameters typically observed in
HDAC X-ray structures may result in substantially weaker
binding.29 Since none of the crystal structures of HDAC1−3
have an “open” conformation, docking to these isoforms gave
results very similar to those for the “closed” conformation of
HDAC8, which is the likely reason for superior selectivity of 2c
for HDAC8. In the “open” conformation, however, 2c forms a
more extended contact with the gorge region residues by
occupying the additional lipophilic pocket (Figure 2A) between
F152 and Y306 not available in the “closed” conformation, and
the distance between the hydroxamate group and the zinc ion is
similar to that observed in the X-ray structures. We hypothesized
that an additional linker bridging the two benzyl groups in 2c as
shown in Figure 2A may further improve potency of the resulting
compounds by minimizing the entropic penalty paid by the
benzyl groups upon binding.
To test this hypothesis, we designed and synthesized a series of
TIQ-based compounds 3 and tested their activity against class I
HDAC isoforms. To facilitate the synthesis, the length of the
linker connecting the tertiary amine moiety and one of the
phenyl rings (Figure 2A) was increased from one to two
methylene groups. The general synthetic approach for the TIQ-
based compounds is shown in Scheme 2. For unsubstituted
compounds, 1,2,3,4-tetrahydroisoquinoline was alkylated with
the appropriate commercially available ω-bromo esters with six
or two methylene linkers. The resulting esters 6a−b were then
converted to the corresponding hydroxamic acids 3a−b. For C1-
substituted compounds, the starting commercially available acids
7a−j were first converted to acyl chlorides and then coupled with
phenethylamine resulting in amides 8a−j. Next, amides 8a−j
were subjected to Bischler−Napieralski cyclization and hydro-
genation to give tetrahydroisoquinolines 9a−j. Intermediates
9a−j were then alkylated with appropriate commercially
available ω-bromo esters with six, four, or two methylene linkers
Figure 1. Structures of HDAC8 selective inhibitors PCI-34051 and
Cpd-2 and general structures for secondary (1) and tertiary (2) amine-
based and TIQ-based (3) HDAC inhibitors.
In our preliminary studies, we investigated the effect of linker
length on potency of amine-based HDAC inhibitors. A library of
novel secondary and tertiary amine-based compounds 1a−c and
2a−c, respectively, was synthesized via alkylation of the
corresponding primary amines and conversion of the methyl
ester moiety to hydroxamic acid (Scheme 1). Screening of these
compounds against class I HDAC isoforms (Table 1) showed
that potency of secondary amines 1a−c decreased for all class I
isoforms as the length of the linker was shortened from six to two
methylene groups. Potency of tertiary amines 2a−c exhibited a
similar trend only for HDAC1, 2, and 3, where it decreased from
1.1, 2.7, and 1.1 μM for 2a (n = 6) to 33, >100, and 52 μM for 2c
(n = 2), respectively. However, shortening the linker of the
tertiary amines led to a rather unexpected 2.9-fold improvement
in potency for HDAC8 from 4.0 μM for 2a to 1.4 μM for 2c.
Changes in HDAC8 selectivity corresponding to shortening the
linker were not appreciable for the secondary amines. The
tertiary amines, however, experienced increased selectivity for
HDAC8 when the linker was shortened with 2c being 24-, >71-,
and 37-fold selective for HDAC8 over HDAC1, 2, and 3,
respectively, compared to 2a. Such substantial improvement in
selectivity rendered it worth investigating further.
To guide further improvements in HDAC8 activity and
selectivity, amines 1a−c and 2a−c were docked to the binding
site of HDAC8 in the “closed” and “open” conformations26 and
to the other class I HDAC isoforms. It was reported previously
that binding of the “linkerless” ligands generally occurs in an
open conformation of loop B (residues S30−K36), whereas
ligands less bulky in the linker region bind to the conformation of
HDAC8 with loop B closed.22,26,27 Analysis of the docking poses
of the ligands in the “closed” conformation of HDAC8 has shown
a
Scheme 1. Synthesis of Amine-Based HDAC Inhibitors
a
Reagents and conditions: (a) Et3N (1.5 equiv), CH3CN, rt, 12 h; (b) NH2OH·HCl (20 equiv), KOH (25 equiv), MeOH, 0 °C−rt, 5 h.
B
DOI: 10.1021/acsmedchemlett.7b00126
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Table 1. IC50 Values of Amine-Based and TIQ-Based HDAC Inhibitors against Class I HDACs
a
IC50 (μM)
b
compd
1a
n
R
HDAC1
HDAC2
HDAC3
HDAC8
HDAC6 (% inhibition)
6
4
2
6
4
2
6
2
4
2
1
2
2
2
2
2
2
2
2
2
H
0.34 0.0035
14 3.1
3.1 0.79
58 0.96
>100
0.43 0.0038
22 0.15
72 4.2
1.1 0.012
1.4 0.30
52 3.0
0.097 0.0027
98 8.9
0.81 0.083
10 0.55
>100
1.8 0.13
ND
1b
H
10 3.2
ND
1c
H
50 3.7
>100
ND
2a
phenyl
phenyl
phenyl
H
1.1 0.059
1.8 0.13
33 1.1
2.7 0.67
5.4 0.23
>100
4.0 0.37
ND
2b
1.9 0.20
ND
2c
1.4 0.41
ND
3a
0.098 0.0038
57 3.2
0.33 0.012
>100
1.9 0.17
97 0.050
31 1.0
93 0.056
35 8.8
17 5.1
27 9.4
21 0.040
5.7 4.5
17 6.1
26 1.2
32 0.080
30 2.0
74 0.25
39 1.2
37 4.1
3b
H
19 2.4
3c
phenyl
phenyl
phenyl
0.77 0.069
4.4 0.74
>100
2.7 0.12
40 2.0
>100
0.044 0.026
0.95 0.012
19 4.9
3d
3e
3f
3-methoxyphenyl
4-methoxyphenyl
4-ethoxyphenyl
13 1.4
69 11
>100
34 4.3
>100
0.11 0.026
0.082 0.019
0.21 0.020
0.34 0.039
0.19 0.033
6.7 1.1
3g
27 3.7
3h
24 0.69
37 3.1
99 4.9
>100
>100
3i
3-(trifluoromethyl)phenyl
4-(trifluoromethyl)phenyl
cyclohexyl
>100
3j
18 3.6
>100
27 3.0
75 4.9
29 13
3k
>100
>100
3l
4-(tert-butyl)phenyl
naphthalene-2-yl
18 2.7
29 8.1
13 1.1
47 17
31 2.0
0.48 0.077
0.39 0.027
0.055 0.014
0.0066 0.0030
3m
2.4 0.23
7.3 0.48
3.8 2.7
3.4 0.21
38 2.2
7.1 0.38
3n
1,1′-biphenyl-4-yl
PCI-34051
a
b
IC50 values and percent inhibition of HDAC6 activity at 10 μM concentration are expressed as mean
independent experiments and rounded to two significant figures. ND: not determined.
standard deviation of at least two
methylene groups in 3b resulted in loss of activity but somewhat
improved selectivity for HDAC8. Modeling of the binding pose
of the TIQ scaffold showed that the C1-substituent points into
the HDAC8 extended pocket (Figure 2B), and varying this
substituent would enable probing of the unique features of this
pocket.
Substituents in the phenyl ring B at C5−C8, however, were
expected to either point toward the solvent or to clash with the
binding site (Figure 2B) and were deprioritized for this SAR
study. To investigate the effect of C1 substitution in the TIQ
scaffold, a series of compounds was synthesized, and the results
are shown in Table 1. The TIQ scaffold was first substituted with
a phenyl group, which resulted in 3c−e closely resembling
tertiary amines 2a−c, respectively. Compound 3c with a four-
methylene linker had similar potency and selectivity for HDAC1,
2, and 3, but had a 43-fold increase in potency and 18-, 61-, and
18-fold selectivity for HDAC8 over HDAC1, 2, and 3,
respectively, compared with the corresponding tertiary amine
2b. Shortening of the linker to two methylene groups resulted in
compound 3d that exhibited 7.5-, 2.5-, 5.2-, and 1.5-fold
improvement for HDAC1, 2, 3, and 8 potency, respectively. Its
HDAC8 selectivity over HDAC1, 2, and 3 compared with the
corresponding tertiary amine 2c decreased by 5.1-, 1.7-, and 3.5-
fold, respectively. The observed slight improvement in potency
of 3d for all class I isoforms might be due to HDAC1, 2, and 3
undergoing small conformational changes, similar to those in the
“open” conformation of HDAC8. Further shortening of the
linker to one methylene group in 3e resulted in a complete loss of
Figure 2. (A) Binding site in the “open” conformation of HDAC8 (PDB
ID: 1VKG) and the docking pose of 2c. Possible rigidification is shown
with the orange dotted line. (B) Same view of the binding site showing
the docking pose of 3f.
to give esters 10a−l, which were then converted to the
corresponding racemic hydroxamic acids 3c−n. All the final
compounds were tested for inhibition of class I HDAC isoforms.
The IC50 inhibitory data for the known “linkerless” HDAC8
selective inhibitor PCI-34051 were generated in the same
conditions in-house for comparison.
Unsubstituted TIQ-based compound 3a with a six-methylene
linker had similar selectivity and up to 10-fold higher potency for
class I HDAC isoforms compared with the corresponding
secondary amine 1a (Table 1). Shortening of the linker to two
C
DOI: 10.1021/acsmedchemlett.7b00126
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
a
Scheme 2. Synthesis of TIQ-Based HDAC Inhibitors
a
Reagents and conditions: (a) methyl 3-bromopropanoate or methyl 7-bromoheptanoate (1.2 equiv), K2CO3 (2.0 equiv), CH3CN, rt, 12 h; (b)
NH2OH·HCl (20 equiv), KOH (25 equiv), MeOH, 0 °C−rt, 5 h; (c) (i) (COCl)2 (1.5 equiv), cat. DMF, CH2Cl2, 0 °C, 3 h; (ii) phenethylamine
(1.0 equiv), Et3N (1.5 equiv), CH2Cl2, 0 °C, 4 h; (d) (i) P2O5 (2.5 equiv), POCl3 (3.0 equiv), xylene, reflux, 160 °C, 6 h; (ii) H2/PtO2, ethanol, rt,
12 h; (e) methyl 3-bromopropanoate, methyl 5-bromopentanoate, or methyl 7-bromoheptanoate (1.2 equiv), K2CO3 (2.0 equiv), CH3CN, rt, 12 h.
activity for HDAC1, 2, and 3 and showed only 19 μM potency for
HDAC8. Based on these findings, we decided to expand on the
SAR studies of the C1-substituted TIQ-based compounds with
two methylene groups between the hydroxamic acid zinc binding
group and the nitrogen atom of the TIQ moiety.
pronounced loss of activity for HDAC1, 2, and 3 and only a 6.7
μM potency for HDAC8 compared with the phenyl-substituted
3d (Table 1). Interestingly, bulkier 4-(tert-butyl)phenyl C1-
substituent in 3l resulted in 4.1-fold lower potency for HDAC1,
similar potency for HDAC2, 2.9-fold decrease in potency for
HDAC3, and 2-fold improvement in potency for HDAC8
compared with 3d. In contrast, the naphthyl substituted 3m was
2- to 3-fold more potent for all the class I isoforms and
maintained similar selectivity when compared to 3d. The drastic
loss of HDAC8 activity seen in 3k but not in 3l and 3m may,
therefore, be explained by the general preference of the binding
site for linear and flat aromatic substituents and inability to
accommodate large nonflat groups, such as cyclohexyl, likely due
to poor fit and loss of pi−pi interactions with the aromatic
residues lining the gorge region of the binding site. To follow-up
on these observations, we synthesized C1-biphenyl substituted
compound 3n and found it to exhibit similar potency for
HDAC2, 1.6- and 3.8-fold lower potency for HDAC1 and 3,
respectively, and a 17-fold higher potency for HDAC8 when
compared with 3d. The improvement in potency for HDAC8 but
not for HDAC1, 2, or 3 of 3n compared to that of 3k, 3l, and 3m
is consistent with the conformational changes observed only in
the “open” conformation of HDAC8. These results suggest that
not only aromatic “linkerless” but also aliphatic hydroxamic acid-
based ligands are capable of causing conformational changes
similar to those in the “open” conformation of HDAC8.
The TIQ-based inhibitors were docked to HDAC1−3 and 8.
Inhibitors demonstrating dual HDAC6 and 8 selectivity have
been reported and include “linkerless”23,24 and substituted
aliphatic hydroxamic acids,25 and hence, HDAC6 was also
included in the docking studies and biological evaluation of the
TIQ-based inhibitors. In all the isoforms, except the “open”
conformation of HDAC8, the binding poses of 3d−n were found
to be similar to that of 2c, i.e., suboptimal in geometry of binding
to the zinc ion. Since some of the TIQ-based inhibitors show a
low micromolar activity against HDAC1, 2, and 3, it is possible
that these isoforms can also undergo small conformational
changes, albeit probably not as extensive as those in HDAC8.
HDAC6 activity assay showed that all the C1-substituted two-
methylene linker TIQ-based compounds, except for 3m,
To investigate the possible impact of electronic and steric
effects on potency and selectivity of the compounds, a series of
C1-substituted TIQ-based compounds were synthesized bearing
electronically diverse substituents in the phenyl moiety. The
meta- and para-methoxy substituted 3f and 3g, respectively, had
lower potency for HDAC1, 2, and 3 and 8.6- and 12-fold
improved potency and selectivity for HDAC8 compared with the
unsubstituted 3d, respectively (Table 1). The slight 1.3-fold
increase in potency and selectivity for HDAC8 of the para-
substituted 3g compared with the meta-substituted 3f suggests
that the HDAC8 extended binding pocket is too narrow and
prefers to accommodate linear C1 substituents (Figure 2B). The
slight increase in size of the para-substituent in 4-ethoxyphenyl-
TIQ 3h resulted in nearly identical potency and selectivity
compared to those of 3g. Replacement of the electron donating
methoxy substituent with an electron withdrawing trifluoro-
methyl moiety in 3i and 3j resulted in similar activity profiles to
3f and 3g, respectively. The difference in potency for HDAC8
between meta and para substitution observed for 3f and 3g,
respectively, was also present in the trifluoromethyl substituted
compounds, where 3j was 1.8-fold more potent than 3i.
Compounds 3i and 3j exhibited a 3.1- and 2.3-fold decrease in
potency for HDAC8 when compared with their meta- and para-
substituted methoxy analogues 3f and 3g, respectively. The
observed decrease in potency cannot be explained by electronic
differences alone because the trifluoromethyl moiety is expected
to increase potency via reducing the electron density on the face
of the ring thereby decreasing repulsive component of the pi−pi
interaction30,31 between the phenyl moiety in the ligands and
side chains of F152 and Y306 (Figure 2B). Overall, these results
indicate that steric bulk of C1-substituent in the TIQ-based series
plays a more important role in binding of the ligands to HDAC8.
To assess the possible effect of bulkier C1-substitunets on
activity, compounds 3k−n were synthesized and tested.
Compound 3k containing a cyclohexyl substituent exhibited a
D
DOI: 10.1021/acsmedchemlett.7b00126
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
exhibited less than 50% inhibition at 10 μM concentration and
were comparable to PCI-34051.
In summary, the modeling, design, synthesis, SAR studies, and
biological evaluation of the amine-based and C1-substituted
TIQ-based compounds led to the discovery of the novel series of
highly potent and selective HDAC8 inhibitors. Rigidification of
the substituents connected to the tertiary amino group to
minimize entropic loss upon binding presents an example of
successful strategy for improvement of both HDAC8 potency
and selectivity. The difference in propensity of class I HDAC
isoforms to undergo conformational changes such as those in the
“open” conformation of HDAC8 appears to account for the
unique potency and selectivity of the C1-substituted TIQ-based
compounds. The SAR studies suggest that binding of linear and
aromatic substituents in the additional lipophilic pocket in the
“open” conformation of HDAC8 is preferred. We envision that
such rational approach to inducing the conformational changes
in HDAC8 may be a valuable addition to the drug design toolbox
for HDACs and other targets. Future studies will focus on
improvement of HDAC8 potency and selectivity by varying the
substitution at the C1 position as well as exploring the effects of
substituting the TIQ scaffold at other positions.
In silico analysis of the physicochemical properties of the
amine-based and TIQ-based compounds demonstrated that
these compounds are druglike and are likely to have excellent
solubility and cell permeability (Supporting Information). The
computed properties of these compounds are also in line with
those active in the central nervous system (CNS).32 Considering
outstanding selectivity of our TIQ-based inhibitors even for very
homologous class I HDACs, there is a high probability that off-
target activity against other metalloproteases is minimal.
Somewhat similar TIQ-containing hydroxamic acids, although
without a basic nitrogen, showed only weak inhibition of
aminopeptidase N.33
Given their high potency and selectivity for HDAC8,
compounds 3g and 3n were evaluated in neuroblastoma cell
lines. Compound 3n exhibited significant concentration- and
time-dependent cytotoxicity similar to or better than PCI-34051
in all the cell lines (Figures 3A, Supporting Information).
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acsmedchem-
lett.7b00126.
■
S
Chemical, biological, and molecular modeling methods
and experimental data (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: pap4@uic.edu. Phone: 312-996-4174.
■
ORCID
Taha Y. Taha: 0000-0002-7344-7490
Thomas W. Hanigan: 0000-0003-0752-4707
Gregory R. J. Thatcher: 0000-0002-7757-1739
Pavel A. Petukhov: 0000-0003-1554-242X
Present Addresses
Figure 3. Cytotoxicity and HDAC8 selectivity of 3g and 3n in
neuroblastoma cells. (A) Cells were treated with 10 μM 3g or 3n for 48
h, and cell proliferation was assessed by MTS assay. (B) Cells were
treated with indicated concentrations for 24 h followed by Western blot
analysis of acetylation of histone H3 and α-tubulin. Cell viability values
⊥Department of Biochemistry and Molecular Genetics, College
of Medicine, University of Illinois at Chicago, Chicago, Illinois
60607, United States.
∥Department of Research & Development, DongBang Future
Tech & Life Co. Ltd., Seoul, South Korea.
represent mean
standard deviation of at least three independent
experiments. *Statistically significant difference compared with DMSO
control (Student’s t test, p < 0.001). Blot is representative of three
independent experiments.
∇Department of Chemistry and Biochemistry, University of
Delaware, Newark, Delaware 19716, United States.
Author Contributions
The manuscript was written through contributions of all authors.
All authors have approved the final version of the manuscript.
Notes
Compound 3n was most potent in SH-SY5Y (EC50 = 0.76 μM)
and IMR5 (EC50 = 5.8 μM) cell lines after 48 h treatment.
Compound 3g showed pronounced activity only at 10 μM, >72 h
treatment, and in SH-SY5Y cells (Supporting Information). The
differences in cytotoxicity of 3g and 3n (Figure 3A) despite the
similar HDAC inhibitory profile might be due to target
engagement differences in cells compared with recombinant
enzyme assays as well as off-target activity. The solubility or cell
permeability are unlikely to play a role based on their precited
properties. The selectivity of 3n and 3g for HDAC8 among other
class I and II HDACs was confirmed in SH-SY5Y cells. Both
compounds did not affect the acetylation of α-tubulin or histone
H3 similar to PCI-34051 compared with nonselective SAHA
(Figure 3B).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This study was funded by the National Cancer Institute/NIH
grants R21 CA183627, R01 CA131970, and R01 HL130760, the
Alzheimer’s Drug Discovery Foundation grant ADDF
#20101103, and UICentre for Drug Discovery.
ABBREVIATIONS
■
HDAC, histone deacetylase; HDACs, histone deacetylases; TIQ,
tetrahydroisoquinoline; SAR, structure−activity relationship;
CNS, central nervous system
E
DOI: 10.1021/acsmedchemlett.7b00126
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
(19) Heimburg, T.; Chakrabarti, A.; Lancelot, J.; Marek, M.; Melesina,
J.; Hauser, A. T.; Shaik, T. B.; Duclaud, S.; Robaa, D.; Erdmann, F.;
Schmidt, M.; Romier, C.; Pierce, R. J.; Jung, M.; Sippl, W. Structure-
Based Design and Synthesis of Novel Inhibitors Targeting HDAC8 from
Schistosoma mansoni for the Treatment of Schistosomiasis. J. Med.
Chem. 2016, 59, 2423−35.
(20) 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,
1026−34.
(21) Hsieh, C. L.; Ma, H. P.; Su, C. M.; Chang, Y. J.; Hung, W. Y.; Ho,
Y. S.; Huang, W. J.; Lin, R. K. Alterations in histone deacetylase 8 lead to
cell migration and poor prognosis in breast cancer. Life Sci. 2016, 151,
7−14.
(22) Krennhrubec, K.; Marshall, B. L.; Hedglin, M.; Verdin, E.; Ulrich,
S. M. Design and evaluation of ’Linkerless’ hydroxamic acids as selective
HDAC8 inhibitors. Bioorg. Med. Chem. Lett. 2007, 17, 2874−8.
(23) Olson, D. E.; Wagner, F. F.; Kaya, T.; Gale, J. P.; Aidoud, N.;
Davoine, E. L.; Lazzaro, F.; Weiwer, M.; Zhang, Y. L.; Holson, E. B.
Discovery of the first histone deacetylase 6/8 dual inhibitors. J. Med.
Chem. 2013, 56, 4816−20.
REFERENCES
■
(1) Jenuwein, T.; Allis, C. D. Translating the histone code. Science
2001, 293, 1074−80.
(2) Gardner, K. E.; Allis, C. D.; Strahl, B. D. Operating on chromatin, a
colorful language where context matters. J. Mol. Biol. 2011, 409, 36−46.
(3) Delcuve, G. P.; Khan, D. H.; Davie, J. R. Roles of histone
deacetylases in epigenetic regulation: emerging paradigms from studies
with inhibitors. Clin. Epigenet. 2012, 4, 5.
(4) Xu, W. S.; Parmigiani, R. B.; Marks, P. A. Histone deacetylase
inhibitors: molecular mechanisms of action. Oncogene 2007, 26, 5541−
52.
(5) Falkenberg, K. J.; Johnstone, R. W. Histone deacetylases and their
inhibitors in cancer, neurological diseases and immune disorders. Nat.
Rev. Drug Discovery 2014, 13, 673−691.
(6) Johnstone, R. W. Histone-deacetylase inhibitors: novel drugs for
the treatment of cancer. Nat. Rev. Drug Discovery 2002, 1, 287−299.
(7) Marks, P.; Rifkind, R. A.; Richon, V. M.; Breslow, R.; Miller, T.;
Kelly, W. K. Histone deacetylases and cancer: causes and therapies. Nat.
Rev. Cancer 2001, 1, 194−202.
(8) 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.
(24) Rodrigues, D. A.; Ferreira-Silva, G. A.; Ferreira, A. C.; Fernandes,
R. A.; Kwee, J. K.; Sant’Anna, C. M.; Ionta, M.; Fraga, C. A. Design,
Synthesis, and Pharmacological Evaluation of Novel N-Acylhydrazone
Derivatives as Potent Histone Deacetylase 6/8 Dual Inhibitors. J. Med.
Chem. 2016, 59, 655−70.
(9) Li, Z.; Zhu, W. G. Targeting histone deacetylases for cancer
therapy: from molecular mechanisms to clinical implications. Int. J. Biol.
Sci. 2014, 10, 757−70.
(10) Dovey, O. M.; Foster, C. T.; Cowley, S. M. Histone deacetylase 1
(HDAC1), but not HDAC2, controls embryonic stem cell differ-
entiation. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 8242−8247.
(11) Deardorff, M. A.; Bando, M.; Nakato, R.; Watrin, E.; Itoh, T.;
Minamino, M.; Saitoh, K.; Komata, M.; Katou, Y.; Clark, D.; Cole, K. E.;
De Baere, E.; Decroos, C.; Di Donato, N.; Ernst, S.; Francey, L. J.;
Gyftodimou, Y.; Hirashima, K.; Hullings, M.; Ishikawa, Y.; Jaulin, C.;
Kaur, M.; Kiyono, T.; Lombardi, P. M.; Magnaghi-Jaulin, L.; Mortier, G.
R.; Nozaki, N.; Petersen, M. B.; Seimiya, H.; Siu, V. M.; Suzuki, Y.;
Takagaki, K.; Wilde, J. J.; Willems, P. J.; Prigent, C.; Gillessen-Kaesbach,
G.; Christianson, D. W.; Kaiser, F. J.; Jackson, L. G.; Hirota, T.; Krantz, I.
D.; Shirahige, K. HDAC8 mutations in Cornelia de Lange syndrome
affect the cohesin acetylation cycle. Nature 2012, 489, 313−7.
(12) Bhaskara, S.; Knutson, S. K.; Jiang, G.; Chandrasekharan, M. B.;
Wilson, A. J.; Zheng, S.; Yenamandra, A.; Locke, K.; Yuan, J.-l.; Bonine-
Summers, A. R.; Wells, C. E.; Kaiser, J. F.; Washington, M. K.; Zhao, Z.;
Wagner, F. F.; Sun, Z.-W.; Xia, F.; Holson, E. B.; Khabele, D.; Hiebert, S.
W. Hdac3 is essential for the maintenance of chromatin structure and
genome stability. Cancer Cell 2010, 18, 436−447.
(13) Balasubramanian, S.; Verner, E.; Buggy, J. J. Isoform-specific
histone deacetylase inhibitors: the next step? Cancer Lett. 2009, 280,
211−21.
(14) Lane, A. A.; Chabner, B. A. Histone deacetylase inhibitors in
cancer therapy. J. Clin. Oncol. 2009, 27, 5459−68.
(15) Oehme, I.; Deubzer, H. E.; Wegener, D.; Pickert, D.; Linke, J. P.;
Hero, B.; Kopp-Schneider, A.; Westermann, F.; Ulrich, S. M.; von
Deimling, A.; Fischer, M.; Witt, O. Histone deacetylase 8 in
neuroblastoma tumorigenesis. Clin. Cancer Res. 2009, 15, 91−9.
(16) Rettig, I.; Koeneke, E.; Trippel, F.; Mueller, W. C.; Burhenne, J.;
Kopp-Schneider, A.; Fabian, J.; Schober, A.; Fernekorn, U.; von
Deimling, A.; Deubzer, H. E.; Milde, T.; Witt, O.; Oehme, I. Selective
inhibition of HDAC8 decreases neuroblastoma growth in vitro and in
vivo and enhances retinoic acid-mediated differentiation. Cell Death Dis.
2015, 6, e1657.
(25) Negmeldin, A. T.; Padige, G.; Bieliauskas, A. V.; Pflum, M. K.
Structural Requirements of HDAC Inhibitors: SAHA Analogues
Modified at the C2 Position Display HDAC6/8 Selectivity. ACS Med.
Chem. Lett. 2017, 8, 281−286.
(26) Brunsteiner, M.; Petukhov, P. A. Insights from comprehensive
multiple receptor docking to HDAC8. J. Mol. Model. 2012, 18, 3927−
3939.
(27) Tabackman, A. A.; Frankson, R.; Marsan, E. S.; Perry, K.; Cole, K.
E. Structure of ’linkerless’ hydroxamic acid inhibitor-HDAC8 complex
confirms the formation of an isoform-specific subpocket. J. Struct. Biol.
2016, 195, 373−8.
(28) Dowling, D. P.; Gantt, S. L.; Gattis, S. G.; Fierke, C. A.;
Christianson, D. W. Structural studies of human histone deacetylase 8
and its site-specific variants complexed with substrate and inhibitors.
Biochemistry 2008, 47, 13554−63.
(29) Harding, M. M. The geometry of metal-ligand interactions
relevant to proteins. II. Angles at the metal atom, additional weak metal-
donor interactions. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2000, 56,
857−67.
(30) Rashkin, M. J.; Waters, M. L. Unexpected substituent effects in
offset pi-pi stacked interactions in water. J. Am. Chem. Soc. 2002, 124,
1860−1.
(31) Sinnokrot, M. O.; Sherrill, C. D. Substituent effects in pi-pi
interactions: sandwich and T-shaped configurations. J. Am. Chem. Soc.
2004, 126, 7690−7.
(32) Ghose, A. K.; Herbertz, T.; Hudkins, R. L.; Dorsey, B. D.;
Mallamo, J. P. Knowledge-Based, Central Nervous System (CNS) Lead
Selection and Lead Optimization for CNS Drug Discovery. ACS Chem.
Neurosci. 2012, 3, 50−68.
(33) Flipo, M.; Beghyn, T.; Charton, J.; Leroux, V. A.; Deprez, B. P.;
Deprez-Poulain, R. F. A library of novel hydroxamic acids targeting the
metallo-protease family: design, parallel synthesis and screening. Bioorg.
Med. Chem. 2007, 15, 63−76.
(17) Durst, K. L.; Lutterbach, B.; Kummalue, T.; Friedman, A. D.;
Hiebert, S. W. The inv(16) fusion protein associates with corepressors
via a smooth muscle myosin heavy-chain domain. Mol. Cell. Biol. 2003,
23, 607−19.
(18) Kang, Y.; Nian, H.; Rajendran, P.; Kim, E.; Dashwood, W. M.;
Pinto, J. T.; Boardman, L. A.; Thibodeau, S. N.; Limburg, P. J.; Lohr, C.
V.; Bisson, W. H.; Williams, D. E.; Ho, E.; Dashwood, R. H. HDAC8 and
STAT3 repress BMF gene activity in colon cancer cells. Cell Death Dis.
2014, 5, e1476.
F
DOI: 10.1021/acsmedchemlett.7b00126
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX