Novel HDAC6 isoform selective chiral small molecule histone deacetylase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2008.12.045

In an effort to identify HDAC isoform selective inhibitors, we designed and synthesized novel, chiral 3,4-dihydroquinoxalin-2(1H)-one and piperazine-2,5-dione aryl hydroxamates showing selectivity (up to 40-fold) for human HDAC6 over other class I/IIa HDACs. The observed selectivity and potency (IC₅₀ values 10-200 nM against HDAC6) is markedly dependent on the absolute configuration of the chiral moiety, and suggests new possibilities for use of chiral compounds in selective HDAC isoform inhibition.

Full text of DOI:10.1016/j.bmcl.2008.12.045

Bioorganic & Medicinal Chemistry Letters 19 (2009) 688–692
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel HDAC6 isoform selective chiral small molecule histone
deacetylase inhibitors
a
a
a
a
b
David V. Smil a, , Sukhdev Manku , Yves A. Chantigny , Silvana Leit , Amal Wahhab , Theresa P. Yan ,
Marielle Fournel b, Christiane Maroun b, Zuomei Li b, Anne-Marie Lemieux c, Alina Nicolescu c, Jubrail Rahil c,
Sylvain Lefebvre d, Anthony Panetta d, Jeffrey M. Besterman c, Robert Déziel a
*
a MethylGene Inc, Department of Medicinal Chemistry, 7220 Frederick-Banting, Montréal, Que., Canada H4S 2A1
b Department of Molecular Biology, 7220 Frederick-Banting, Montréal, Que., Canada H4S 2A1
c Department of Lead Discovery, 7220 Frederick-Banting, Montréal, Que., Canada H4S 2A1
d Department of Protein Biochemistry, 7220 Frederick-Banting, Montréal, Que., Canada H4S 2A1
a r t i c l e i n f o
a b s t r a c t
Article history:
In an effort to identify HDAC isoform selective inhibitors, we designed and synthesized novel, chiral 3,4-
dihydroquinoxalin-2(1H)-one and piperazine-2,5-dione aryl hydroxamates showing selectivity (up to 40-
fold) for human HDAC6 over other class I/IIa HDACs. The observed selectivity and potency (IC50 values
10–200 nM against HDAC6) is markedly dependent on the absolute configuration of the chiral moiety,
and suggests new possibilities for use of chiral compounds in selective HDAC isoform inhibition.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 5 November 2008
Revised 5 December 2008
Accepted 9 December 2008
Available online 14 December 2008
Keywords:
HDAC6
HDAC inhibitors
Aryl hydroxamic acids
Histone acetylation
Tubulin acetylation
3,4-Dihydroquinoxalin-2(1H)-ones
Piperazine-2,5-diones
The reversible hyperacetylation of lysine residues on the N-ter-
minal tails of core histone proteins H2A/B, H3, and H4 (correlated
with gene activation), and histone hypoacetylation (leading to
transcriptional repression), are catalyzed by histone acetyltransfer-
ases (HATs) and histone deacetylases (HDACs), respectively.1 The
18 known human HDACs, 11 of which are functionally related
Zn+2-dependent amidohydrolases, are grouped into three classes
(I, II, and IV). Class I/II HDACs show some structural homology to
each other within their catalytic domains, but are structurally dis-
tinct from the NAD+-dependent class III HDACs (sirtuins, SIRT1-7).2
HDAC subtypes 1, 2, 3 and 8 collectively make up class I (HDAC11
is assigned as class IV), and are ubiquitously expressed, predomi-
nantly nuclear enzymes. Class II is comprised of six HDAC isoforms,
and is itself divided into two subclasses, IIa (HDACs 4, 5, 7 and 9)
and IIb (HDAC6 and HDAC10). The IIb subclass enzymes uniquely
feature two deacetylase domains, and are primarily cytoplasmic,
while class IIa members shuttle between the nucleus and cyto-
plasm. Furthermore, unlike class I HDACs, class IIa/b enzymes are
expressed in a more limited number of cell types. Significantly,
class IIb enzyme HDAC6 operates on a variety of substrates other
than histone proteins, and is involved in processing Lys40 of the
mitotic spindle protein a
-tubulin.3
It is known that HDAC inhibitors regulate differentiation, prolif-
eration, cell cycle, protein turnover, and apoptosis,4 and can func-
tion as targeted chemotherapeutic agents.5 Despite the broad
structural and chemical spectrum of HDAC inhibitors under clinical
development, few show notable or significant specificity for class I
versus class II enzymes, or any subtype selectivity within a given
class. Perhaps the most selective HDAC inhibitors are benzamides,
exemplified by MS275 and its aryl substituted analog (Fig. 1),
showing high potency and exclusive class I activity confined pri-
marily to the HDAC 1, 2, 3 subtypes.6 The prodrug FK228 (depsi-
peptide/RomidepsinÒ) also shows this class I selectivity, with no
additional subtype selectivity.6a
In the realm of hydroxamate-based inhibitors, such levels of
selectivity are rarely observed, with most inhibitors tending to
be pan class
I and II, such as SAHA, LBH589, PXD101,
JNJ26481585 and ITF2357.6a While limited reports have been
published outlining class IIb7 and HDAC88 selectivity, most pro-
gress has been made in the area of HDAC6 selectivity. The dis-
closure of tubacin (Fig. 2), a high molecular weight (>700 D)
* Corresponding author. Tel.: +1 514 337 3333x395; fax: +1 514 337 0550.
E-mail address: smild@methylgene.com (D.V. Smil).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.12.045

D. V. Smil et al. / Bioorg. Med. Chem. Lett. 19 (2009) 688–692
689
O
O
O
N
H
NH2
O
N
H
NH2
H
N
H
N
N
N
O
O
S
MethylGene 2007
MS275
HDAC1,2,3
HDAC1,2
O
O
N
H
O
HN
HN
O
S
S
O
H
N
NH
N
H
O
H
HN
N
OMe
NH
O
H
O
O
O
H
O
FK228
HDAC1,2,3,8
Apicidin
HDAC2,3
Figure 1. Known potent and selective class I HDAC inhibitors.
O
NHOH
HN
O
H
O
N
NHOH
O
N
N
N
O
O
S
O
HO
N
F
Kozikowski, 2007
Schreiber, 2003
O
O
H
N
O
O
MeO
NHOH
BocHN
N
H
NHOH
O
N
O
Br
Kozikowski, 2008
Jung, 2007
Figure 2. Reported HDAC6 selective inhibitors.
1,3-dioxane containing hydroxamate, by Schreiber and co-work-
ers was the first example of a compound capable of selectively
moiety to potentially confer this HDAC class/isoform selectivity
and potency to an aryl hydroxamate was considered an attractive
option given the literature data reported for apicidin (Fig. 1), a
macrocyclic tetrapeptide electrophilic ketone HDAC inhibitor
showing some HDAC2 and HDAC3 selectivity over HDAC1, HDAC8,
and all class II HDACs, with potencies in the nM range.14 Work by
Yoshida and co-workers involving the hybridization of hydroxa-
mate-based inhibitors with these cyclic tetrapeptide class struc-
tures, termed CHAPs, also served to demonstrate that the surface
recognition domain of the compound affects not only potency,
but selectivity as well. Several of the CHAPs synthesized showed
low nM potencies, and exhibited 10- to 90-fold selectivity for
HDAC1 over HDAC6.15 The solvent-exposed periphery of the en-
zyme’s catalytic site, a region showing significant sequence diver-
sity between class I and II HDACs,9a can provide a structural
rationale for the observed selectivity of apicidin and the CHAPs,
and suggests that selectivity for other HDAC isoforms may be
achievable by use of a similar strategy involving a chiral, enzyme
rim binding pharmacophore.
inhibiting class IIb subtype HDAC6, and inducing
a-tubulin acet-
ylation.9 Several years later, Jung et al. described phenylalanine-
containing hydroxamates that exhibited activity against HDACs
(IC50 values typically >0.30 lM), and modest HDAC6 selectivity
(5- to 14-fold over HDAC1).10 More recently, Kozikowski et al.
have described potent and HDAC6 selective triazolylphenyl
capped hydroxamate inhibitors (HDAC6 IC50 = 1.9 nM and 51-
fold selectivity over HDAC1),11 and related phenylisoxazole
capped hydroxamates (HDAC6 IC50 = 0.002 nM and 210-fold
selectivity over HDAC3).12 In all instances, the inhibitors are sub-
eric acid or lysine-based, relying almost exclusively on this
structural feature in order to achieve the observed potencies
against the enzyme, and employing the capping moiety (enzyme
surface recognition domain) as a presumed selectivity element.
Guided by the documented role of HDAC6 regulation in neuro-
protection,13 we were also interested in the selective inhibition of
this HDAC isoform. The possibility of employing a chiral capping

690
D. V. Smil et al. / Bioorg. Med. Chem. Lett. 19 (2009) 688–692
Although the cyclic tetrapeptide capping region of apicidin and
Construction of the less rigid piperazine-2,5-dione caps found in
the CHAPs is structurally complex, it was anticipated that the sig-
nificant contribution it makes to rim binding could be replicated
effectively in a simplified system employing similar structural ele-
ments; principally, the H-bond donor/acceptors and chirality avail-
able in the cap’s constituent amino acid derived amides, and its’
‘space filling’ topology. With these elements in mind, it was envis-
aged that introduction of a chiral component into the capping por-
tion of the novel hydroxamate inhibitors would be most rapidly
and efficiently accomplished through the use of commercially
the second series of hydroxamates 12 was initiated with the ester-
ification of several (D)-amino acids, giving rise to building blocks 8
(Scheme 2). Unlike in the previous sequence, subsequent reductive
amination using methyl 4-formylbenzoate and sodium borohy-
dride as the reductant was employed for the earlier introduction
of the aryl linker moiety, providing the intermediates 9. Further
elaboration through amidation with N-Cbz-glycine under either
EDCI/HOBt or HATU/DIPEA conditions allowed for access to inter-
mediates 10. As in the previous synthetic route, it was expected
that reductive cleavage of the Cbz protecting group from the gly-
cine fragment using catalytic palladium and hydrogen atmosphere
conditions would result in concomitant cyclization to the penulti-
mate intermediates 11. This strategy did work for the cases where
R = phenyl and 3-indole, but failed for the conversion of the sub-
strate with R = 2-thiophene, resulting in partial reduction of the
thiophene moiety. In this later case, an alternative strategy relying
on HBr/AcOH mediated cleavage of the Cbz group and subsequent
high temperature cyclization was used to access that material.
Transformation of 11 to the final hydroxamic acids 12 was then ef-
fected directly from the methyl esters by means of a mild and gen-
eral protocol employing aqueous hydroxylamine under basic
conditions.
All of the hydroxamate inhibitors in the 7 and 12 scaffold series,
along with reference compound ( )-trichostatin (TSA), were ini-
tially analyzed for their potency and selectivity against a panel of
recombinant human HDACs. This panel was comprised of
HDAC1-8 isoforms that had been affinity purified together with
their cofactors. The data for HDAC6 (class IIb) and HDAC2 and 8
(chosen as representative class I enzyme subtypes) is collected in
Table 1. All other HDAC inhibition values are consistent with the
observed trends discussed below, and are thus omitted for pur-
poses of clarity.
On examination, we found that the parent 3,4-dihydroquinoxa-
lin-2(1H)-one capped hydroxamate 7a (which lacks a chiral moi-
ety) shows no appreciable selectivity for any of the class I or II
HDACs. Remarkably, however, it’s evident that not only is HDAC6
selectivity achievable using some of the chiral variants in the 7
scaffold series, but that this selectivity is dependent on the abso-
lute configuration of the chiral centre. For example, while the (S)
configured hydroxamate 7c exhibits essentially no selectivity
across the entire HDAC panel, the isomeric (R) compound 7b shows
a 26-fold selectivity for HDAC6 over HDAC2, and a 53-fold selectiv-
ity for HDAC6 over HDAC8. The selectivity of 7b for HDAC6 with
respect to all other members of the HDAC panel (not shown)
tended to be comparable, or greater. The isomeric pair 7d/e also
available chiral starting materials such as (D)- and (L)-amino acids.
To this end, the chosen amino acids could either be appended to a
rigidifying aryl moiety to form 3,4-dihydroquinoxalin-2(1H)-one
caps (Scheme 1), or be incorporated into less rigid piperazine-
2,5-dione caps (Scheme 2); a strategy complementary to the use
of 1,4-benzodiazepine-2,5-diones and 1,4-benzodiazepines as con-
strained peptidomimetics for the capping of lysine-based HDAC
inhibitors reported by Tranoy-Opalinski et al .16 and Romanelli
et al.17, respectively. Both types of caps are inherently appealing
in terms of their ease of synthesis, and in the different topology
and steric bulk accessible through the judicious selection of start-
ing materials. In order to reduce the total number of rotatable
bonds in the final inhibitor, it was also established at the outset
to employ an aryl linker in the construction of the overall molecu-
lar scaffold. Based on extensive observations made in our laborato-
ries, use of an aryl linker was viewed as a viable option, and was
not expected to impair the potency of any hydroxamates prepared
for this study.
Synthesis of the rigid 3,4-dihydroquinoxalin-2(1H)-ones 5 began
with the facile SNAr coupling of glycine or various (D)- and (L)-amino
acids to 1-fluoro-2-nitrobenzene 1, giving rise to the 2-nitrophenyl-
amino acid derivatives 3 (Scheme 1). Subsequent esterification with
iodomethane providing intermediates 4 proceeded in essentially
quantitative yield, and supplied material requiring no additional
purification prior to the next step. Reduction of the nitrophenyl moi-
ety that followed was accompanied by in situ cyclization to form the
target 3,4-dihydroquinoxalin-2(1H)-ones 5 in high yield. Intermedi-
ates 6 were then obtained via reductive amination employing 4-
formylbenzoic acid, catalytic dibutyltin dichloride, and phenyl si-
lane as the reductant.18 Final transformation to the desired hydroxa-
mic acids 7 was performed using the BOP mediated coupling of
hydroxylamine hydrochloride. The conditions employed in this step,
like in all prior steps, were selected to avoid racemization of the sol-
itary chiral center within the structure. Overall, this five step se-
quence proved to be efficient, high yielding, and general, allowing
access to a diverse set of hydroxamates 7.
O
R
O
R
O
NO2
NO2
NO2
H
N
H
N
a
b
NH2
*
*
+
*
OH
O
M
e
HO
F
Y
Y
Y
R
2
1
3
4
c
O
O
O
R
Y
O
O
R
Y
R
Y
OH
HN
N
HN
HN
OH
d
e
*
*
*
H
N
NH
N
6
7a-j
5
Scheme 1. Reagents and conditions: (a) K2CO3, EtOH/H2O (5:1), 100 °C, sealed tube, 16 h (>95%); (b) K2CO3, MeI, DMF, rt, 16 h (>95%); (c) H2, 10% Pd/C, MeOH/EtOAc (2:1), rt,
16 h (>80%); (d) 4-formylbenzoic acid, PhSiH3, Bu2SnCl2, THF, rt, 16 h (>80%); (e) NH2OHÁHCl, BOP, TEA, Pyr., rt, 8 h (>50%).

D. V. Smil et al. / Bioorg. Med. Chem. Lett. 19 (2009) 688–692
691
O
O
O
O
OMe
a
b
NH2
R
NH2
R
H
N
MeO
HO
MeO
R
2
8
9
c
O
NHCbz
O
O
O
R
O
O
O
O
R
OH
O
OMe
HN
OMe
e
d
HN
N
H
N
R
N
N
MeO
12a-c
11
10
Scheme 2. Reagents and conditions: (a) HCl/MeOH, rt, 16 h (>70%); (b) methyl 4-formylbenzoate, NaBH4, MeOH, rt, 16 h (>40%); (c) N-Cbz-glycine with EDCI, HOBt, DMF, rt,
16 h (>80%) for 12a–b or HATU, DIPEA, DMF, rt, 16 h (54%) for 12c; (d) 10% Pd/C, MeOH, rt, 16 h (>50%) for 12a–b or 33% HBr/AcOH, rt, 1 h, followed by 10% NH4OH/DCM wash
and diphenyl ether, 170 °C, sealed tube, 1 h; (95%) for 12c; (e) NH2OH, NaOH, THF/MeOH/H2O, rt, 16 h (>60%).
Table 1
In vitro determination of IC50
(l
M) values for the inhibition of HDACs 2, 6, and 8 by compounds 7a–j, 12a–c, and TSA
Selectivity indexb
HDAC6 versus HDAC2 HDAC6 versus HDAC8
a
Compound
Isomer
Substituent
‘R’
IC50
(
l
M)
‘Y’
HDAC2
HDAC6
HDAC8
TSA
7a
7b
7c
7d
7e
7f
7g
7h
7i
—
—
R
S
R
S
R
R
R
R
R
R
R
R
—
H
—
—
Ph
Ph
0.01
0.56
0.26
0.46
0.87
1.48
0.46
0.33
0.52
1.48
1.44
5.54
1.81
3.47
0.02
0.15
0.01
0.22
0.04
0.31
0.04
0.01
0.01
0.04
0.05
0.17
0.11
0.16
0.77
0.52
0.53
0.21
0.69
1.19
0.42
0.73
0.64
1.15
0.25
2.34
1.21
0.44
1
4
26
2
21
5
11
41
40
34
28
33
17
22
39
3
53
1
17
4
10
91
49
27
5
14
11
3
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
—
2-Thiophene
2-Thiophene
3,4-diF-Ph
4-OH-Ph
–CH2Ph
–(CH2)3NHBoc
4-tBuO-Ph
Ph
3-Indole
2-Thiophene
7j
12a
12b
12c
—
—
a
Results shown are mean values calculated from experiments conducted in triplicate.
b
The selectivity index is equivalent to HDAC1 or 8 IC50/HDAC6 IC50
.
demonstrates this chirality-induced HDAC6 selectivity, and (R) iso-
meric compounds 7f–j further maintain the pattern.
tial induction of a-tubulin acetylation (compared to histone H3
acetylation) should be observed.
This selectivity pattern is also evident with the (R) isomers in
the piperazine-2,5-dione capped hydroxamate 12 scaffold series.
Indeed, when this trend was initially observed after the first sev-
eral inhibitors in the 7 scaffold series were prepared, only the (R)
isomers in the 12 scaffold series were synthesized, and produced
the expected result. All of the (R) isomers in both the 7 and 12 scaf-
fold series show superior HDAC6 over HDAC2 selectivity with re-
spect to TSA. Generally speaking, the nature of the pendent cap
substituent ‘R’ appears to play less of a role in the observed selec-
tivity than the absolute configuration at the chiral centre. Further-
more, the substituent also has little effect on the observed
inhibitory potency of the compounds against HDAC6; all of the
(R) isomers in the 7 scaffold series have activity in the 10–50 nM
range (equipotent to TSA), and are more potent as a whole than
compounds in the 12 scaffold series, whose values range between
100 and 200 nM.
Substantiation of this data and the associated trend was sought
from the measurement of histone H3 acetylation (H3Ac) versus a-
tubulin acetylation (TubAc) in cultured human bladder carcinoma
T24 cells (Table 2). The T24 cells were treated with escalating
doses of inhibitors in the 7 and 12 scaffold series, along with
TSA. It was expected that if (R) isomers in the chiral 7 and 12 scaf-
fold series were genuinely HDAC6 selective inhibitors, a preferen-
All of the compounds tested did induce both core histone H3
and
a-tubulin acetylation in a dose-dependent manner, with the
calculated EC50 values suggesting that the latter acetylation is fa-
vored (Table 2). The ability of these compounds to preferentially
Table 2
In vitro determination of EC50
(l
M) values for histone H3 and
a
-tubulin acetylation
by compounds 7a–e,i,j, 12a–c, and TSA
a
Compound
Isomer
Substituent
‘R’
EC50
H3Ac
(
l
M)
‘Y’
TubAc
TSA
7a
7b
7c
7d
7e
7i
—
—
R
S
R
S
R
R
R
R
R
—
H
—
—
Ph
Ph
0.01
3.58
0.52
1.90
1.22
2.30
6.75
2.75
0.01
1.01
0.04
0.32
0.07
0.23
0.09
0.08
1.49
2.87
2.00
CH2
CH2
CH2
CH2
CH2
CH2
—
2-Thiophene
2-Thiophene
–(CH2)3 NHBoc
4-tBuO-Ph
Ph
3-Indole
2-Thiophene
7j
12a
12b
12c
23.0
12.1
9.00
—
—
a
Results shown are mean values calculated from experiments conducted in
triplicate (in T24 cells).

692
D. V. Smil et al. / Bioorg. Med. Chem. Lett. 19 (2009) 688–692
Overall, compounds 7b,d,i,j and 12a can serve as important tools
for further investigations into the selectivity and utility of this class
of inhibitors, and to further probe HDAC isoform-specific biological
functions or yield insights into HDAC isoform structures.
Supplementary data
Supplementary data associated with this article can be found in
U.S. Patent Application 2007/0155730 A1. The enzymatic assays
followed the fluorescent signal obtained from the HDAC catalyzed
deacetylation of coumarin-labeled lysine. The substrate used for
Figure 3. Dose-dependent induction of
acetylation in vitro in T24 human bladder cancer cells determined for compounds
7d, 7e and TSA. Cells were incubated with the HDAC inhibitors at 0.1, 1.0, and
a-tubulin acetylation versus histone H3
HDAC1, 2, 3, 6, and 8 was Boc-Lys(e-acetyl)-AMC (Bachem Biosci-
10
lM for 16 h before whole-cell lysates were analyzed for a-tubulin and histone
ences Inc.) and Boc-Lys-( -trifluormethylacetyl)-AMC (synthesized
e
H3 acetylation by SDS–PAGE and Western immunoblot with antibodies specific for
in-house) for HDAC4, 5, and 7. Recombinant enzymes expressed in
baculovirus were used. HDAC1, 2, and 3 were C-terminal FLAG-
tagged and HDAC4 (612-1034), HDAC5 (620–1122), HDAC6,
HDAC7 (438–915), and HDAC8 are N-terminal His-tagged. The en-
zymes were incubated with the compounds in assay buffer (25 mM
Hepes, pH 8.0, 137 mM NaCl, 1 mM MgCl2 and 2.7 mM KCl) for
10 min at ambient temperature in black 96-well plates. The sub-
strate was added into enzyme-compound mixture and incubated
at 37 °C. Reaction was quenched by adding trypsin and TSA to a fi-
nal concentration of 1 mg/mL and 1 uM, respectively. Fluorescence
either acetylated histone H3 or
loading levels.
a-tubulin. Total histone H3 was used to reveal blot
induce
a-tubulin acetylation in cancer cells is largely correlated
with their ability to inhibit HDAC6 selectively in vitro. While TSA
and achiral compound 7a, both non-HDAC6 selective inhibitors,
showed no propensity to induce
tially over histone H3 acetylation, compounds 7b,c,d,e,i,j and
12a,b,c did exhibit this tendency. Once again, the nature of substi-
tuent ‘R’ appears to have little effect on the observed selectivity of
the inhibitors, but the chirality of the capping moiety does appear
to play a significant role. The less potent and HDAC6 selective (S)
a-tubulin acetylation preferen-
was measured using
a fluorimeter (SPECTRAMAX GeminiXS,
Molecular Devices). The 50% inhibitory concentrations (IC50) for
inhibitors were determined by analyzing dose-response inhibition
curves with GraFit.
isomers 7c and 7e were, as expected, less efficient a-tubulin acet-
ylation inducers (by about 2-fold versus histone H3 acetylation)
than their (R) isomer counterparts, 7b and 7d, respectively. Fur-
thermore, the overall lower potency of the 12 scaffold series com-
References and notes
pounds against HDAC6 is reflected in their higher EC50 values for a-
tubulin acetylation, although their level of selectivity remains
significant.
1. Strahl, B. D.; Allis, C. D. Nature 2000, 403, 41.
2. Grozinger, C. M.; Schreiber, S. L. Chem. Biol. 2002, 9, 3.
3. Johnstone, R. W.; Licht, J. D. Cancer Cell 2003, 4, 13.
4. Sasakawa, Y.; Naoe, Y.; Sogo, N.; Inoue, T.; Sasakawa, T.; Matsuo, M.; Manda, T.;
Mutoh, S. Biochem. Pharmacol. 2005, 69, 603.
5. Dokmanovic, M.; Marks, P. A. J. Cell. Biochem. 2005, 96, 293.
6. a Mahboobi, S.; Sellmer, A.; Höcher, H.; Garhammer, C.; Pongratz, H.; Maier, T.;
Ciossek, T.; Beckers, T. J. Med. Chem. 2007, 50, 4405; b Fournel, M.; Bonfils, C.;
Hou, Y.; Yan, P. T.; Trachy-Bourget, M.-C.; Kalita, A.; Liu, J.; Lu, A.-H.; Zhou, N. Z.;
Robert, M.-F.; Gillespie, J.; Wang, J. J.; SteCroix, H.; Rahil, J.; Lefebvre, S.;
Moradei, O.; Delorme, D.; MacLeod, A. R.; Besterman, J. M.; Li, Z. Mol. Cancer
Ther. 2008, 7, 759; c Moradei, O. M.; Mallais, T. C.; Frechette, S.; Paquin, I.;
Tessier, P. E.; Leit, S. M.; Fournel, M.; Bonfils, C.; Trachy-Bourget, M.-C.; Liu, J.;
Yan, T. P.; Lu, A.-H.; Rahil, J.; Wang, J.; Lefebvre, S.; Li, Z.; Vaisburg, A. F.;
Besterman, J. M. J. Med. Chem. 2007, 50, 5543.
Compound 7i, a highly selective HDAC6 inhibitor in the in vitro
enzyme assay, is a particularly selective promoter of
a-tubulin
acetylation (75-fold versus histone H3 acetylation), while 7b, 7d,
7j, and 12a also show relatively high selectivity indices (>13-fold).
These results are comparable to the 75-fold selectivity reported for
tubacin, but the potencies of the most active
inducers 7b,d,i,j (with EC50 values <100 nM) are far superior to that
of tubacin, with an EC50 = 2.9
M.9c The data obtained from the
a-tubulin acetylation
l
cell-based assays are consistent with the results obtained in the
7. Mai, A.; Massa, S.; Pezzi, R.; Simeoni, S.; Rotili, D.; Nebbioso, A.; Scognamiglio,
A.; Altucci, L.; Loidl, P.; Brosch, G. J. Med. Chem. 2005, 48, 3344.
8. KrennHrubec, K.; Marshall, B. L.; Hedglin, M.; Verdin, E.; Ulrich, S. M. Bioorg.
Med. Chem. Lett. 2007, 17, 2874.
9. a Sternson, S. M.; Wong, J. C.; Grozinger, C. M.; Schreiber, S. L. Org. Lett. 2001, 3,
4239; b Haggarty, S. J.; Koeller, K. M.; Wong, J. C.; Grozinger, C. M.; Screiber, S. L.
Proc. Natl. Acad. Sci. 2003, 100, 4389; c Wong, J. C.; Hong, R.; Schreiber, S. L. J.
Am. Chem. Soc. 2003, 125, 5586.
10. Schäfer, S.; Saunders, L.; Eliseeva, E.; Velena, A.; Jung, M.; Schwienhorst, A.;
Strasser, A.; Dickmanns, A.; Ficner, R.; Schlimme, S.; Sippl, W.; Verdin, E.; Jung,
M. Bioorg. Med. Chem. 2008, 16, 2011.
11. Chen, Y.; Lopez-Sanchez, M.; Savoy, D. N.; Billadeau, D. D.; Dow, G. S.;
Kozikowski, A. P. J. Med. Chem. 2008, 51, 3437.
in vitro enzyme inhibition assays.
Western blot analysis for isomeric compound pair 7d/e is illus-
trated in Figure 3. Clear induction of
a
-tubulin acetylation is evi-
dent with highly selective and potent (R) inhibitor 7d at 1.0
lM,
while less selective and potent (S) inhibitor 7e shows a less signif-
icant induction effect (comparable to the non-selective control
inhibitor TSA). At the compound concentrations investigated in
the assay (0.1–10 lM), the Histone H3 acetylation level is seen to
vary little with respect to the DMSO control for inhibitors 7d and
7e.
12. Kozikowski, A. P.; Tapadar, S.; Luchini, D. N.; Kim, K. H.; Billadeau, D. N. J. Med.
Chem. 2008, 51, 4370.
13. Dompierre, J. P.; Godin, J. D.; Charrin, B. C.; Cordelières, F. P.; King, S. J.;
Humbert, S.; Saudou, F. J. Neurosci. 2007, 27, 3571.
14. Khan, N.; Jeffers, M.; Kumar, S.; Hackett, C.; Boldog, F.; Khramtsov, N.; Qian, X.;
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. Biochem. J. 2008, 409, 581.
15. Furumai, R.; Komatsu, Y.; Nishino, N.; Khochbin, S.; Yoshida, M.; Horinouchi, S.
Proc. Natl. Acad. Sci. 2001, 98, 87.
16. Loudni, L.; Roche, J.; Potiron, V.; Clarhaut, J.; Bachmann, C.; Gesson, J.-P.;
Tranoy-Opalinski, I. Bioorg. Med. Chem. Lett. 2007, 17, 4819.
17. Guandalini, L.; Cellai, C.; Laurenzana, A.; Scapecchi, S.; Paoletti, F.; Romanelli,
M. N. Bioorg. Med. Chem. Lett. 2008, 18, 5071.
In summary, by developing two series of novel aryl hydroxa-
mate HDAC inhibitors incorporating distinct chiral capping motifs,
it was determined that in vitro selectivity for HDAC 6 (class IIb)
over all other class I and IIa isoforms could be achieved. Moreover,
it was established that this selectivity was dependent not only on
the presence of the chiral centre, but also on its’ absolute configu-
ration. It was also possible to observe a parallel, chiral structure
dependent selectivity for
acetylation using in vitro cell-based assays. The majority of com-
pounds effectively induced -tubulin acetylation at a level compa-
rable to that of tubacin, but were found to be far more potent.
a-tubulin acetylation over histone H3
a
18. Apodaca, R.; Xiao, W. Org. Lett. 2001, 3, 1745.