Non-peptide macrocyclic histone deacetylase inhibitors

Article abstract of DOI:10.1021/jm801128g

Inhibition of histone deacetylase inhibitors (HDACi) hold great promise in cancer therapy because of their demonstrated ability to arrest proliferation of nearly all transformed cell types. Of the several structurally distinct small molecule HDACi reporte

Full text of DOI:10.1021/jm801128g

456
J. Med. Chem. 2009, 52, 456–468
Non-Peptide Macrocyclic Histone Deacetylase Inhibitors
Adegboyega K. Oyelere,*,¶,§ Po C. Chen,¶ William Guerrant,¶ Sandra C. Mwakwari,¶ Rebecca Hood,#,¶ Yunzhe Zhang,† and
Yuhong Fan†,§
School of Chemistry and Biochemistry, School of Biology, and Parker H. Petit Institute for Bioengineering and Bioscience, Georgia Institute of
Technology, Atlanta, Georgia 30332-0400
ReceiVed September 9, 2008
Inhibition of histone deacetylase inhibitors (HDACi) hold great promise in cancer therapy because of their
demonstrated ability to arrest proliferation of nearly all transformed cell types. Of the several structurally
distinct small molecule HDACi reported, macrocyclic depsipeptides have the most complex recognition
cap-group moieties and present an excellent opportunity for the modulation of the biological activities of
HDACi. Unfortunately, the structure-activity relationship (SAR) studies for this class of compounds have
been impaired largely because most macrocyclic HDACi known to date comprise complex peptide
macrocycles. In addition to retaining the pharmacologically disadvantaged peptidyl backbone, they offer
only limited opportunity for side chain modifications. Here, we report the discovery of a new class of
macrocyclic HDACi based on the macrolide antibiotics skeletons. SAR studies revealed that these compounds
displayed both linker-length and macrolide-type dependent HDAC inhibition activities with IC50 in the low
nanomolar range. In addition, these non-peptide macrocyclic HDACi are more selective against HDACs 1
and 2 relative to HDAC 8, another class I HDAC isoform, and hence have subclass HDAC isoform selectivity.
Introduction
backbone, they offer only limited opportunity for side chain
modifications.10 Identification of non-peptide macrocyclic HDA-
Ci will offer a new class of macrocyclic HDACi with potentially
more favorable druglike properties. Furthermore, this will aid
comprehensive SAR studies and further enhance our under-
standing of the roles of specific interactions between the enzyme
outer rim and inhibitor cap groups in HDACi activity and
selectivity. Herein, we report the discovery of a new class of
potent, non-peptide macrocyclic HDACi derived from the
macrolide macrocyclic ring structures.
Inhibition of histone deacetylases (HDACsa) has recently been
clinically validated as a novel therapeutic strategy for cancer
treatment.1 Because of their demonstrated ability to arrest
proliferation of nearly all transformed cell types,2 HDAC
inhibitors (HDACi) hold great promise as agents of choice,
either as stand-alone therapeutics or in combination with others,
in the fight against the cancer scourge. To date, several
structurally distinct small molecule HDACi have been reported
including aryl hydroxamates, benzamides, short-chain fatty
acids, electrophilic ketones, and macrocyclic peptides (Scheme
1).3-6 All HDACi so far reported fit a three-motif pharmacoph-
oric model, namely, a zinc-binding group (ZBG), a hydrophobic
linker, and a recognition cap group.3 The X-ray crystal structures
of a bacterial HDAC homologue, histone deacetylase-like
protein (HDLP) bound to suberoylanilide hydroxamic acid
(SAHA) and trichostatin A (TSA), and recently human HDAC8
and HDAC7 have validated this model.7,8 Of these HDACi,
macrocyclic peptides have the most complex recognition cap-
group moieties and present an excellent opportunity for the
modulation of the biological activities of HDACi. Although
cyclic-peptide HDACi possess potent HDAC inhibition activity
(nanomolar range), their broad application in cancer therapy
currently remains largely unproven.3 One promising exception,
FK-228 (Scheme 1), is currently in phase II study for the
treatment of cutaneous T-cell lymphoma.9
Results and Discussion
Macrolides are glycosylated polyketide antibiotics that have
been in use for over 50 years for the treatment of respiratory
tract infections. Additionally, macrolides have elicited other non-
antibiotic effects, including anti-inflammatory and immuno-
modulatory effects that make them promising candidates for
the management of diseases of chronic airway inflammation.11,12
More recently, macrolides derived from the 6-O-methyleryth-
romycin A ring have been reported to serve as nonpeptidic
surrogates for the peptide backbone of macrocyclic peptide
luteinizing hormone-releasing hormone (LHRH) receptor an-
tagonists.13 Drawing inspiration from the peptidomimetic prop-
erty of macrolides, we hypothesize that an appropriate substi-
tution of the cyclic peptide moiety of a prototypical cyclic-
peptide HDACi with macrolide skeletons will generate a new
class of potent non-peptide macrocyclic HDACi.
To test our hypothesis, we first sought a SAHA-macrolide
conjugate that incorporated the 15-membered azalide ring of
azithromycin as the macrolide template (Scheme 1). The
azithromycin skeleton is an attractive choice because of its
excellent pharmacokinetic (PK) profile and ease of chemical
transformation of key moieties on the skeleton.14 Our design
approach is to attach the HDAC inhibiting group to a macrolide
moiety that is remote from the macrocyclic ring. We anticipate
that this will minimize the potential steric clash at the HDAC
active site that might result from the introduction of the
azithromycin skeleton, a macrocyclic ring not optimized to bind
The dearth of clinically effective cyclic-peptide HDACi may
be in part due to development problems characteristic of large
peptides, most especially poor oral bioavailability. In addition
to retaining the pharmacologically disadvantaged peptidyl
* To whom the correspondence should be addressed. Phone: 404-894-
4047. Fax: 404-894-2291. E-mail: aoyelere@gatech.edu.
¶
School of Chemistry and Biochemistry.
Parker H. Petit Institute for Bioengineering and Bioscience.
Beckman Undergraduate Research Fellow.
School of Biology.
§
#
†
a Abbreviations: HDAC, histone deacetylase; HDACi, histone deacetylase
inhibitors; HDLP, histone deacetylase-like protein; SAHA, suberoylanilide
hydroxamic acid; TSA, trichostatin A; ZBG, zinc-binding group.
10.1021/jm801128g CCC: $40.75
2009 American Chemical Society
Published on Web 12/18/2008

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 457
Scheme 1. (a) Selected Examples of Acyclic HDAC Inhibitors, (b) Representative Examples of Cyclic-Peptide HDAC Inhibitors, and
(c) Representative Examples of Macrolide Antibiotics
the HDAC enzyme. The NMR structure15 and 3D model
revealed that the 3′-tertiary amine of the desosamine sugar and
the 4′′-OH of the cladinose sugar met our design requirement.
We opted for the former group because of the well established
facile transformations of the tertiary amine moiety.13,14 Almost
all modifications of this moiety have resulted in the attenuation
of the antibacterial activity of this class of compound.14b
Furthermore, coupling of such desosamine modifications with
cladinose sugar removal will result in compounds devoid of
antibacterial activity,16 a property of the parent macrolides
undesirable to our goals. Toward this end, we synthesized
compound 8, which incorporates a SAHA-like moiety (Figure
1) into the 3′-tertiary amine of azithromycin, as a prototype
molecule. Compound 8 was synthesized from azithromycin 5
and methyl 8-chloro-8-oxooctanoate 1 through a five-step
synthetic route as shown in Figure 1. To gain some preliminary
insights into the roles of key macrolide moieties in anti-HDAC
activity of 8, we synthesized compound 10, an analogue of 8
lacking the cladinose sugar. Acid promoted removal of the
cladinose sugar according to a published protocol13 quantita-
tively yielded compound 9, which was converted to the
hydroxamate 10 using the same protocol for the synthesis of 8
from 7 (Figure 1). We then tested for the HDAC inhibition
activity of 8 and 10 using a cell free kit assay17 and found that
both compounds caused a concentration dependent inhibition
of HDACs 1 and 2 from HeLa cell nuclear extract (Table 1).
In fact, both compounds have identical anti-HDAC activity, with
IC50 values in the low nanomolar range. Conversely, compound
7, the methyl ester precursor of 8, is completely inactive in this
assay. This result suggests that the 15-membered azalide ring
of the azithromycin macrolide is a suitable non-peptidyl
surrogate for the macrocyclic peptide moiety of a typical cyclic-
peptide HDACi. Additionally, the binding orientations of these
compounds at the HDAC active site may be such that the
hydroxamate group is efficiently presented to chelate the active
site Zn2+ ion while the cladinose moiety is oriented away from
the enzyme’s outer rim.
We then initiated structure-activity relationship studies on
8 and 10 to optimize the HDAC binding affinity of these
compounds. Preliminarily, we focused on the effects of the
modification of the linker-cap group connection moiety, mac-
rolide skeleton, and linker length on anti-HDAC activity. We
used the AutoDock program18 and the crystal structure of a
histone deacetylase-like protein (HDLP)7 to guide our structural
optimization. In an earlier study, we demonstrated the suitability
of a 1,2,3-triazole ring as an alternative linker-cap group
connection moiety in SAHA-like HDAC inhibitors.19 In addition
to serving as an alternative connection moiety with a potentially
favorable pharmacokinetic properties, the triazole ring is a more
synthetically tractable group20 and has been adopted in the
construction of active-site directed chemical probes for profiling
HDAC activities in proteomes and live cells.21 Its incorporation

458 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Oyelere et al.
Figure 1. Synthesis of prototypical non-peptide macrocyclic HDACi 8 and 10.
Table 1. In Vitro HDAC Inhibition (IC50) and Isoform Selectivity of
compound 6 led to the desired desosamine alkylation product
12 in good yields. Copper(I) catalyzed cycloaddition reaction
of alkyne 12 with O-silyl azidohydroxamate 1419,20 gave the
desired cycloadduct 15 in good yields. Similarly, the reaction
of alkynes 12 and 13 with azido ester 18 gave the desired
cycloadducts 19a and 19b, respectively. The deprotection of
the silyl group of compound 15 is accomplished with TBAF
treatment, leading to desired hydroxamate 16a in good yields.
However, difficulties were encountered in separating 16a from
the TBAF reagent. In much the same manner as in the synthesis
of 8 or 10 (Figure 1), the conversion of methyl ester 19a and
19b to the corresponding hydroxamate is dogged by low yields.
Subsequently, we found that a direct Cu(I) catalyzed cycload-
dition between unprotected azidohydroxamate 17 and alkyne
12 or 13 in anhydrous, degassed THF under exclusion of oxygen
uneventfully gave the desired hydroxamates (Figure 2a).
Non-Peptide Macrocyclic HDACia
compd
HDAC 1/2 (nM)
HDAC 8 (nM)
isoform selectivityb
7
8
10
ND
NT
c
62
21
51
42
72
97
121
94
38
107.1
109.8
91.6
88.8
13.9
10.6
58.9
72.4
145.5
226.7
37.0
44.3
4.1
6680
2320
4730
3740
994
1020
7130
6780
11050
ND
3990
4750
1890
1390
5880
4420
10550
ND
16a
16b
16c
16d
16e
16f
16g
16h
24a
24b
24c
24d
24e
24f
24g
24h
SAHA
c
108
107
462
743
106
36
56
c
29
1.9
55.6
123.0
169.8
223.4
65.0
HDAC inhibition studies revealed that the triazolyl com-
pounds 16a and 16b have virtually identical anti-HDAC activity
as the amide compounds 8 and 10 (Table 1). This result is in
contrast with our observation on simple aliphatic hydroxamates
where the introduction of the triazole ring led to an enhancement
of anti-HDAC activity.19 To gain insights on the molecular
interactions between these non-peptide macrocyclic HDACi and
HDAC active site, we performed molecular docking analysis
of 16b on HDLP, using the AutoDock program, as previously
described.18,19 We chose to use the HDLP structure because it
shared conserved active site residues with class I HDACs.
Additionally, direct docking experiments using this structure
or HDAC1 homology model built from the same HDLP
structure have given docking results that are essentially of the
same quality and also agreed with experimentally obtained data.
Either of these approaches has been extensively used in the
literature to interrogate the binding interactions of HDAC
inhibitors at the protein active site.18b,c The structures we
1860
a IC50 values were determined using a cell free kit assay.17 Each value
is obtained from three independent experiments. ND: not determinable. NT:
not tested. b Calculated by dividing the IC50 of HDAC 8 by the IC50 of
HDAC 1/2. c Undeterminable.
into our design could facilitate a facile, high yielding synthesis
of the triazole-based SAHA-macrolide conjugates and thus
provide solution to the problems of low yields that currently
plagued the synthesis of the amide-based compounds 8 and 10
(Figure 1).
Accordingly, we proceeded to test the compatibility of the
triazole ring with the anti-HDAC activity of 8 and 10. We
prepared compounds 16a and 16b, analogues of 8 and 10,
respectively, having the amide moiety connecting the cap group
and the linker region substituted with a 1, 2, 3-triazole ring
(Figure 2a). The reaction of 4-ethynylbenzyl mesylate 11 with

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 459
Figure 2. SAR studies on non-peptide macrocyclic HDACi: (a) synthesis of triazole-linked non-peptide macrocyclic HDACi based on azithromycin
15-membered ring; (b) synthesis of triazole-linked nonpeptide macrocyclic HDACi based on clarithromycin 14-membered ring.
obtained from our docking experiments indicated interesting
molecular surface complementarities between the macrolide
skeleton of 16b and the HDAC outer rim. Previous investiga-
tions have shown that there are four possible binding pockets
on the HDLP surface whose interactions with the HDACi cap
groups could enhance the inhibitor binding ability.18b Compound
16b adopts a docked structure that placed the macrolide
macrocyclic ring in binding pockets 1 and 3 (Figure 3a). In
addition, the hydrophobic components of the macrolide ring
optimally interact with the hydrophobic residues in pockets 1
and 3 while the hydrophilic hydroxyl groups are oriented away
from the pocket’s hydrophobic residues. Compared to the
structure of SAHA, the hydroxamate moiety of 16b is farther
oriented from the active site Zn2+ ion (Figure 3b). On the basis
of the preceding observation, we inferred that optimization of
the linker region could result in compounds with enhanced
HDAC affinity due to a better presentation of the hydroxamate
moiety to the catalytic Zn2+ ion. Subsequent docking analyses

460 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Oyelere et al.
Figure 3. Docked structures of SAHA and 14- and 15-membered macrocyclic HDACi at the active site of HDLP. (a) Superposition of the low
energy conformation of 16b (blue) and SAHA (yellow) revealed the pocket binding preferences of inhibitors at the HDLP surface. Shown are ball
and stick models of the orientations of the hydroxamate groups of SAHA and 16b (b) and of 16b and 16d (orange) (c) with respect to the active
site Zn2+ ion (gray ball). (d) Relative orientation of the macrocyclic rings of 16d and 24d (yellow) with respect to Phe 338 at the HDLP surface.
with analogues of 16b having varied methylene-linker lengths
revealed that 16d, a C7-linker compound, optimally interacts
with the Zn2+ ion (Figure 3c). Interestingly, compound 24d, an
analogue of 16d in which the 15-membered azithromycin ring
has been substituted with the 14-membered clarithromycin ring,
has a slight preference for the enzyme. A closer analysis of the
docked structures of 16d and 24d revealed that the C12-C14
region of the larger 15-membered ring is about 0.5 Å closer to
the phenyl ring of Phe338 that defines one of the hydrophobic
pockets at the enzyme outer rim compared to that of the 14-
membered compound (Figure 3d). This might compromise the
binding affinity of the 15-membered compounds relative to the
14-membered analogues.
To experimentally test these in silico observations, we
synthesized compounds 16c-h and 24a-h, the 15- and 14-
membered non-peptide macrocyclic hydroxamates respectively
(parts a and b of Figure 2, respectively). Results from HDAC
inhibition assay on these compounds revealed HDAC inhibition
activities that essentially paralleled the in silico prediction (Table
1). The compounds displayed both linker-length and macrolide-
type dependent HDAC inhibition activities. For compounds
derived from the same macrolide ring, an increase in the linker
length from C6 to C7 conferred a better anti-HDAC activity.
Further linker length increase did not improve HDAC inhibition
activity; in fact such an increase is detrimental to function in
some cases. For compounds with C6 and C7 linkers, a head-to-
head comparison between 14- and 15-membered macrolides
revealed that the 14-membered compounds are about 2- to 5-fold
better HDACi than their 15-membered counterparts (Table 1;
see 16c and 24c, for example). However, this preference
dissipated with increase in linker length. This is presumably
due to a relief of steric clash between the macrocyclic ring and
the phenyl ring of Phe338 at the enzyme’s outer rim, conferred
by the longer linkers.
To obtain evidence for the HDAC isoform selectivity of the
macrocyclic HDACi described herein, we tested their HDAC8
inhibition activity. We chose HDAC8 because it is in the same
subclass as HDACs 1 and 2, the principal HDACs contained in
the HeLa cell nuclear extract used in the assay kit employed in
this study. There are very few examples of HDAC inhibitors
that are selective for HDAC isoforms within the same class;
hence, this choice should permit a quick, yet rigorous assessment
of HDAC isoform selectivity of our compounds. Compared to
SAHA, all non-peptide macrocyclic hydroxamates tested are
more selective for HDAC 1/2. In particular, C7-linked, 14-
membered compounds 24c and 24d are several-fold more
selective than their 15-membered counterparts (Table 1).
Although HDLP and HDAC8 shared similar amino acid
sequences and topology at the active site, observation from the
analysis of X-ray data, however, revealed significant inhibitor
specific changes in the enzyme active site topology of HDAC8.8b
Our docking analysis operates in the rigid receptor mode, and
it is incapable of capturing such crystallographically observed
ligand induced conformational changes. Nevertheless, we
performed molecular docking analysis of 16b on the HDAC8
structure reported by Somoza et al.8b In contrast to its docked
structure on HDLP, the orientation of 16b that has a chance of
making any interaction with the active site Zn2+ ion is that which
adopted a closed conformation nestled atop of the entrance to
the enzyme active site. In this conformation, the linker group
wrapped around the macrocyclic ring to orient the hydroxamate
moiety toward the Zn2+ ion, albeit much farther away to make
any stabilizing interaction (see Figure S1, Supporting Informa-
tion). An alternative lower energy confirmation of 16b oriented

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 461
Table 2. Cell Growth Inhibition Data for Non-Peptide Macrocyclic
HDACia
SKMES 1 NCI-H69 DU-145 lung fibroblast HMEC
compd
(µM)
(µM)
(µM)
(µM)
(µM)
8
10
NT
NT
NT
NT
>25
<25
1.45
1.24
1.88
1.97
5.89
5.68
>10
2.98
3.29
1.12
1.05
6.97
5.78
8.14
2.12
NT
NT
NT
NT
16a
16b
16c
16d
16e
16f
16g
24a
24b
24c
24d
24e
24f
24g
SAHA
1.79
1.68
2.33
2.56
4.89
4.67
7.54
2.15
1.95
1.33
1.28
4.89
4.45
7.12
2.42
1.92
1.77
3.45
3.01
4.56
3.99
8.45
2.67
1.92
1.45
1.49
5.67
5.09
7.29
2.06
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
Figure 4. Western blot detection of p21WAF1/CIP1 expression levels in
SK-MES-1 cells exposed to SAHA and 24d. Cells are exposed to test
agents for 8 h. Compound numbers and concentrations are displayed
in the top part. Immunoblotting with anti-actin antibody (in green) is
used as a control for equivalent protein loading.
hydroxamates tested shows any growth inhibitory effects on the
normal human primary lung fibroblast and mammary epithelial
cell lines at concentrations in excess of 10 µM. These data
showed that the macrocyclic compounds are selectively toxic
to the transformed cells, a trait that tracks with those of many
HDACi.
a EC50 values were determined from trypan blue exclusion data. Each
value is obtained from a duplicate of four simultaneous experiments. NT:
not tested.
the hydroxamate moiety away from the Zn2+ ion. It is therefore
possible that the observed isoform selectivity may be due to
the inability of the non-peptide macrocyclic hydroxamates herein
described to efficiently induce active site conformational changes
that facilitate HDAC8 specific inhibitor association with the
enzyme active site.22a Alternatively, HDAC8 activity has been
observed to depend on the sequence of its peptide substrate;22b
hence, it is also conceivable that the extent of enzyme inhibition,
and consequently isoform selectivity, may depend on the
substrate used in the inhibition study.
To further test for HDAC isoform selectivity, we investigated
the effect of selected macrocyclic HDACi on the deacetylase
activity of HDAC6, a representative member of class II HDAC.
For this preliminary study, we chose compounds 16c and 16d,
and 24c and 24d, the most potent representative 15- and 14-
membered macrocyclic HDACi, respectively (Table 1). Cell free
HDAC6 inhibition assay was performed as recommended by
the supplier.17 Unlike SAHA, which equally inhibited HDAC1/2
and HDAC6, we observed that the macrocyclic HDACi still
displayed significant preference for HDAC1/2 (see Table S1,
Supporting Information). This observation is in agreement with
the literature reports that suggest that complex headgroups tend
to promote isoform selectivity.3,22c
To screen for the whole cell activity of compounds described
in this study, we studied their effect on the viability of SK-
MES-1 (human NSCLC cell line), NCI-H69 (human SCLC cell
line), DU-145 (a human prostate cancer cell line), and non-
transformed human primary lung fibroblasts and mammary
epithelial cell lines. The nontransformed cell lines were
investigated to obtain evidence for compound selective toxicity.
Drug concentrations necessary for 50% inhibition of cell
viability (EC50) were quantitatively measured using trypan blue
exclusion,23 as previously described.19 Table 2 shows the EC50
values of each compound. The EC50 values obtained for SAHA
are in close agreement with the reported values under similar
experimental conditions.24 Macrocyclic methyl ester 7, the
precursor to compound 8 (Figure 1), has no effect on cell
viability (data not shown). This result may not be unexpected,
since compound 7 has no HDAC inhibition activity (Table 1).
However, all macrocyclic hydroxamates inhibit the proliferation
of all transformed cells studied. Most importantly, compounds
16b, 24c, and 24d are at least twice as potent as SAHA in DU-
145 cells (Table 2). Gratifyingly, none of the macrocyclic
An important biomarker that is primarily associated with
intracellular HDAC inhibition is the expression levels of the
p21WAF1/CIP1 gene. Up-regulation of the p21WAF1/CIP1 gene has
been generally observed with cellular HDAC inhibition.25,26 As
part of a study aimed at cellular target and mechanistic
validation, we investigated the effect of 24d, a representative
macrocyclic HDACi, on the intracellular status of p21WAF1/CIP1
protein in SK-MES-1 cells. We used SAHA as a positive control
for HDAC inhibition. Cultured cells were exposed to various
concentrations of our test agents for 8 h, and the cellular
p21WAF1/CIP1 expression levels were determined by Western
blotting according to literature protocol.27 We found that both
24d and SAHA induced a dose dependent increase in p21WAF1/
CIP1 expression (Figure 4). This result provided additional
evidence that the likely mechanism of the antiproliferative
activities of the non-peptide macrocyclic hydroxamates de-
scribed here is through intracellular HDACs inhibition.
Conclusion
We have identified a new class of non-peptide HDACi
derived from the macrocyclic skeletons of clinically useful
macrolides. These compounds will enable a molecular descrip-
tion of the interaction between the HDAC enzyme’s outer rim
and the inhibitors’ macrocyclic cap groups, thereby further
aiding our understanding of the roles of this interaction in
inhibitors’ binding affinity and possibly HDAC isoform selec-
tivity. In addition, because of the selective tissue distribution
that may be conferred by the appended macrolide moiety, some
of these HDACi are anticipated to have targeted anticancer
activity. Specifically, compounds incorporating azithromycin
skeleton could be selectively accumulated in the lungs,28 thereby
possessing lung-selective anticancer activity. The prospect of
tissue-specific HDACi delivery is a particularly enticing alterna-
tive to isoform selective HDACi and could lead to the
identification of new chemotherapeutic agents for use in targeted
cancer therapy applications. Efforts are underway in our
laboratory to profile the tissue distribution of the new class of
HDACi described here.
Experimental Section
Clarithromyin and azithromycin were purchased from Greenfield
Chemical and Pfizer, respectively. 4-Ethynylbenzyl alcohol, ethyl
6-bromohexanoate, ethyl 7-bromoheptanoate, 8-bromooctanoic acid,

462 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Oyelere et al.
and methyl 10-bromodecanoate were purchased from Sigma Ald-
rich. 9-Bromononanoic acid was purchased from Karl Industries
Inc. Common reaction solvents were either high performance liquid
chromatography (HPLC) grade or American Chemical Society
(ACS) grade and used without further purification. HDAC fluori-
metric assay kit and recombinant HDACs were procured from
BIOMOL International, PA. Analtech silica gel plates (60 F254) were
used for analytical TLC, and Analtech preparative TLC plates (UV
254, 2000 µm) were used for purification. UV light was used to
examine the spots. The 200-400 mesh silica gel was used in
column chromatography. Nuclear magnetic resonance (NMR)
spectra were recorded on a Varian-Gemini 400 magnetic resonance
brine (30 mL) and dried over Na2SO4. Solvent was evaporated off
to give compound 4 as a white solid.
A mixture of 4′-desmethylazithromycin 6 (0.315 g, 0.430 mmol)
and crude compound 4 in anhydrous DMSO (7 mL) and Hunig’s
base (0.7 mL) was stirred at 85 °C for 1.5 h. The mixture was
cooled and diluted with EtOAc (60 mL) and washed with saturated
NaHCO3 (40 mL) and saturated brine (40 mL). The organic layer
was dried over Na2SO4 and concentrated in vacuo and the crude
was purified by preparative TLC, eluting with EtOAc/hexanes/Et3N
3:2:0.1 to give 152 mg (35%) of compound 7 as a white solid. 1H
NMR (CDCl3, 400 MHz) δ 0.78-0.82 (7H, m), 0.94-1.03 (11H,
m), 1.09-1.32 (22H, m), 1.34-1.58 (5H, m), 1.59- 1.73 (4H, m),
1.77-2.01 (5H, m), 2.12 (3H, s), 2.20-2.28 (8H, m), 2.42-2.53
(2H, m), 2.54-2.71 (3H, m), 2.87-3.01 (2H, m), 3.11 (3H, s),
3.25-3.33 (3H, m), 3.42 (2H, m), 3.54-3.65 (6H, m), 3.97 (1H,
m), 4.18 (1H, m), 4.36 (1H, d, J ) 6.8 Hz), 4.61 (2H, m), 5.03
(1H, d, J ) 4.4 Hz), 7.14 (2H, d, J ) 8.4 Hz), 7.42 (2H, d, J ) 8.4
Hz), 7.73 (1H, s), 8.97 (1H, bs); 13C NMR (CDCl3, 100 MHz) δ
7.5, 9.1, 11.3, 14.8, 16.2, 16.9, 18.2, 20.5, 21.2, 21.4, 21.5, 21.9,
24.6, 25.3, 26.7, 27.5, 28.7, 29.6, 33.9, 34.7, 36.2, 36.7, 37.3, 39.1,
41.9, 42.2, 45.1, 48.5, 49.3, 51.4, 57.3, 62.2, 64.3, 65.4, 68.5, 69.9,
70.5, 72.7, 73.5, 73.8, 74.1, 77.7, 77.9, 83.4, 94.4, 102.6, 119.5,
129.0, 134.2, 137.0, 171.0, 173.8, 178.4 MS; HRMS (FAB, mnba)
calcd for [C53H91N3O15 + H]+ 1010.6529, found 1010.6450.
Azithromycinarylalkylhydroxamic Acid (8). To a solution of
compound 7 (0.050 g, 0.050 mmol) in 1:1 THF/MeOH (2 mL)
was added hydroxylamine (50% in H2O) (0.07 mL, 1.260 mmol)
and a catalytic amount of KCN. The mixture was stirred at room
temperature for 24 h. The mixture was partitioned between 5%
MeOH in CH2Cl2 (30 mL) and saturated sodium bicarbonate (25
mL). The two layers were separated, and the aqueous layer was
extracted with 5% MeOH in CH2Cl2 (2 × 20 mL). The combined
organic layer was washed with saturated brine (40 mL) and dried
over Na2SO4. Solvent was evaporated off and the crude was purified
by preparative TLC, eluting with CH2Cl2/MeOH/NH4OH 10:1:0.1
to give 26 mg (52%) compound 8 as brown-white solid. 1H NMR
(CD3OD, 400 MHz) δ 0.87-0.92 (6H, m), 1.03-1.12 (12H, m),
1.17-1.37 (m), 1.43-1.69 (m), 1.75-1.88 (6H, m), 1.99 (4H, m),
2.08 (3H, m), 2.13-2.19 (3H, m), 2.24-2.41 (13H, m), 2.54 (1H,
d, J ) 11.2 Hz), 2.75-2.80 (4H, m), 3.00 (1H, d, J ) 9.6 Hz),
3.19 (4H, m), 3.47-3.51 (1H, m), 3.60-3.78 (6H, m), 4.14-4.22
(3H, m), 4.50 (1H, d, J ) 7.2 Hz), 5.02 (1H, d, J ) 4.8 Hz), 7.29
(2H, d, J ) 8.0 Hz), 7.49 (2H, d, J ) 8.4 Hz); 13C NMR (acetone-
d6, 100 MHz) δ 7.6, 9.7, 11.6, 14.3, 15.3, 17.4, 18.9, 21.7, 21.9,
22.2, 22.5, 26.0, 26.1, 27.3, 27.9, 30.9, 32.6, 33.2, 35.5, 36.5, 36.8,
37.6, 43.0, 43.1, 46.1, 49.6, 58.6, 63.1, 64.1, 64.5, 66.2, 68.9, 70.0,
71.6, 73.3, 73.6, 74.2, 74.9, 75.3, 78.0, 78.9, 84.3, 95.6, 103.7,
119.8, 130.0, 135.0, 139.5, 171.9, 178.8; HRMS (EI) calcd for
[C52H90N4O15 + H]+ 1011.6481, found 1011.6580
1
spectrometer. H NMR spectra were recorded in parts per million
(ppm) relative to the peak of CDCl3 (7.24 ppm), CD3OD (3.31
ppm), DMSO-d6 (2.49 ppm), or acetone-d6 (2.04 ppm). 13C spectra
were recorded relative to the central peak of the CDCl3 triplet (77.0
ppm), CD3OD (49.0 ppm), DMSO-d6 septet (39.7 ppm), or acetone-
d6 (2.04 ppm) and were recorded with complete heterodecoupling.
High-resolution mass spectra were recorded at the Georgia Institute
of Technology mass spectrometry facility in Atlanta. 4-Ethynyl-
benzyl methylsulfonate 11 and azidoalkyl esters were synthesized
by adapting literature protocol.29-31 6-Azido-O-silyl hexahydrox-
amate 14 was prepared from the corresponding azidocarboxylic acid,
t-BuPh2SiCl and NaH, according to the procedure described by Muri
et al.32
Synthesis of Methyl 8-(4-(Hydroxymethyl)phenylamino)-8-
oxooctanoate (3). To a solution of (4-aminophenyl)methanol 2
(0.314 g, 2.500 mmol) in anhydrous pyridine (5 mL) was added
chlorotrimethylsilane (0.32 mL, 2.500 mmol) at room temperature,
and stirring continued for 2 h. The mixture was cooled in an ice
bath to 0 °C, and to the mixture was added methyl 8-chloro-8-
oxooctanoate 1 (0.32 mL, 2.200 mmol) and a catalytic amount of
DMAP. The mixture was allowed to warm to room temperature,
and stirring continued overnight. Water (5 mL) and 1 M TBAF in
tetrahydrofuran (THF) (0.25 mL, 0.250 mmol) were added, and
stirring continued for an additional 30 min. EtOAc (50 mL) and 1
N HCl (30 mL) were added, the two layers were separated, the
organic layer was washed with 1 N HCl (30 mL) and saturated
brine (30 mL) and dried over Na2SO4. Solvent was evaporated off
and the crude was purified by preparative TLC, eluting with EtOAc/
hexanes 2:1 to give 275 mg (43%) compound 3 as a yellow-white
solid. 1H NMR (DMSO-d6, 400 MHz) δ 1.26 (4H, m), 1.47-1.56
(4H, m), 2.26 (4H, m), 3.55 (3H, s), 4.40 (2H, d, J ) 5.6 Hz), 5.06
(1H, t, J ) 5.6 Hz), 7.19 (2H, d, J ) 8.8 Hz), 7.51 (2H, d, J ) 8.8
Hz), 9.79 (1H, s); 13C NMR (DMSO-d6, 100 MHz) δ 24.3, 24.9,
28.2, 28.3, 33.2, 36.3, 51.2, 62.6, 118.8, 126.9, 136.9, 138.0, 171.2,
173.4; HRMS (FAB, thioglycerol) calcd for [C16H23NO4 + H]+
294.1705, found 294.1652.
4′-Desmethylazithromycin (6). To a solution of azithromycin
(2.000 g, 2.548 mmol) and sodium acetate (1.78 g, 21.500 mmol)
in 80% aqueous methanol (30 mL) at 90 °C was added iodine (0.700
g, 2.756 mmol) in three batches within 5 min. The mixture was
maintained at pH 8-9 by addition of 1 M NaOH (2 mL, once at
10 min of reaction time), and stirring continued for 3 h. The mixture
was poured into cold water containing 5% sodium thiosulfate (80
mL) and extracted with CH2Cl2 (2 × 40 mL). The aqueous layer
was basified with concentrated NH4OH and extracted with 10%
MeOH in CH2Cl2 (3 × 40 mL), and the organic layer was dried
over Na2SO4. Solvent was evaporated off to give 1.57 g of
compound 6 as an off-white solid (>90% purity, TLC, CHC2Cl2/
MeOH/NH4OH 12:1:0.1). The crude 6 was used without further
purification.
Azithromycinarylalkyl Methyl Ester (7). To a solution of
compound 3 (0.175 g, 0.597 mmol) in CH2Cl2 (7 mL) and
triethylamine (Et3N) (0.24 mL, 1.800 mmol) was added mesyl
chloride (0.10 mL, 1.200 mmol) at 0 °C, and the mixture was
allowed to warm to room temperature. Stirring continued for 1 h,
during which TLC revealed a quantitative conversion into a higher
Rf product. CH2Cl2 (40 mL) and saturated sodium bicarbonate (30
mL) were added, and the two layers were separated. The organic
layer was washed with sodium bicarbonate (2 × 30 mL), saturated
Desclasinose Azithromycinarylalkyl Hydroxamate (10). A mix-
ture of compound 7 (0.050 g, 0.050 mmol) in 0.25 N HCl (15 mL)
was stirred at room temperature for 20 h and poured into EtOAc
(20 mL). The two layers were separated, and the aqueous layer
was washed with EtOAc (2 × 20 mL), basified with concentrated
NH4OH, and then extracted with 5% MeOH in CH2Cl2 (2 × 30
mL). The combined organic layer was washed with saturated brine
(30 mL) and dried over Na2SO4. Solvent was evaporated off to
give compound 9, which was used for the next reaction without
further purification.
To a solution of compound 9 in 1:1 THF/MeOH (2 mL) was
added hydroxylamine (50% in H2O) (0.05 mL, 0.790 mmol) and a
catalytic amount of KCN. The mixture was stirred at room
temperature for 24 h. The mixture was partitioned between 5%
MeOH in CH2Cl2 (30 mL) and saturated brine (20 mL). The two
layers were separated, and the organic layer was dried over Na2SO4.
Solvent was evaporated off and the crude was purified by
preparative TLC, eluting with CH2Cl2/MeOH/NH4OH 9:1:0.1 to
1
give 7 mg (16%) compound 10 as brown-white solid. H NMR
(CD3OD, 400 MHz) δ 0.78 (3H, m), 0.85 (3H, d, J ) 7.2 Hz),
0.91 (3H, d, J ) 8.0 Hz), 0.99 (3H, s), 1.08-1.14 (10H, m),
1.18-1.79 (20H, m), 1.87 (2H, m), 1.98 (2H, t, J ) 7.4 Hz),

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 463
2.05-2.13 (2H, m), 2.17 (3H, s), 2.25 (2H, t, J ) 7.4 Hz),
2.52-2.65 (4H, m), 2.95 (1H, bs), 3.24 (m), 3.37-3.56 (6H, m),
3.67 (1H, d, J ) 13.2 Hz), 4.53 (2H, d, J ) 7.6 Hz), 7.19 (2H, d,
J ) 8.4 Hz), 7.39 (2H, d, J ) 8.4 Hz); HRMS (EI) calcd for
[C44H76N4O12 + H]+ 853.5538, found 853.5488.
100 MHz) δ 7.4, 9.1, 11.3, 14.7, 16.3, 18.2, 21.3, 21.4, 21.5, 22.0,
26.7, 27.5, 29.6, 29.7, 29.9, 34.6, 36.2, 36.9, 42.2, 45.2, 49.3, 49.9,
57.6, 62.4, 64.1, 65.4, 68.5, 69.9, 70.5, 72.7, 73.5, 74.1, 77.7, 77.9,
83.4, 94.3, 102.7, 119.2, 125.5, 127.6, 129.0, 135.5, 147.1, 178.4;
HMRS (ESI) calcd for [C68H106N6O14Si + H]+1259.7609, found
1259.7570.
4′-Ethynylbenzylazithromycin (12). To a solution of 4′-desm-
ethylazithromycin 6 (0.940 g, 1.280 mmol) in anhydrous DMSO
(15 mL) was added Hunig’s base (1.5 mL) and 4-ethynylbenzyl
methanesulfonate 11 (0.380 g, 1.800 mmol). The reaction mixture
was heated with stirring under argon at 85 °C for 2 h. The mixture
was cooled and diluted with EtOAc (100 mL) and washed with
saturated NaHCO3 (3 × 60 mL) and saturated brine (60 mL). The
organic layer was dried over Na2SO4 and concentrated in vacuo.
The crude product was purified by flash chromatography (silica,
12:1:0.05 CH2Cl2/MeOH/concentrated NH4OH) to give 850 mg
Methyl Azithromycin-N-benzyltriazolylhexanoate (19a). 4′-Ethy-
nylbenzylazithromycin 12 (0.045 g, 0.053 mmol) and azido ester
18 (0.014 g, 0.080 mmol) were dissolved in anhydrous THF (5
mL) and stirred under argon at room temperature. Copper(I) iodide
(0.010 g, 0.053 mmol) and Hunig’s base (0.05 mL) were then added
to the reaction mixture, and stirring continued for 12 h. The reaction
mixture was diluted with CH2Cl2 (30 mL) and washed with 1:4
NH4OH/saturated NH4Cl (3 × 25 mL) and again with saturated
NH4Cl (25 mL). The organic layer was dried over Na2SO4 and
concentrated under vacuum. The crude product was purified by
preparative TLC, eluting with hexane/EtOAc/Et3N 3:2:0.1 to give
1
(78%) of 12 as a brown-white solid. H NMR (CDCl3, 400 MHz)
δ 0.86 (6H, m), 0.99 (3H, d, J ) 7.6 Hz), 1.05 (7H, m), 1.11-1.32
(19H, m), 1.36-1.55 (2H, m), 1.68-1.77 (2H, m), 1.81-2.07 (4H,
m), 2.22 (3H, s), 2.27-2.31 (4H, m), 2.51 (2H, m), 2.65-2.75
(2H, m), 2.87 (1H, bs), 2.98 (1H, t, J ) 9.8 Hz), 3.04 (1H, s), 3.10
(3H, s), 3.29-3.34 (2H, m), 3.40-3.47 (2H, m), 3.58 (1H, d, J )
6.8 Hz), 3.65 (1H, s), 3.74 (1H, d, J ) 13.2 Hz), 3.99 (1H, m),
4.21 (1H, dd, J ) 2 Hz, 4.4 Hz), 4.38 (1H, d, J ) 7.2 Hz), 4.65
(1H, dd, J ) 2.8 Hz, 10 Hz), 5.06 (1H, d, J ) 4.4 Hz), 7.23 (2H,
d, J ) 8 Hz), 7.41 (2H, d, J ) 8 Hz); 13C NMR (CDCl3, 100 MHz)
δ 7.5, 9.2, 11.3, 14.9, 16.3, 18.3, 21.3, 21.4, 21.6, 22.0, 26.8, 27.6,
29.6, 34.7, 36.3, 36.9, 42.0, 42.3, 45.2, 49.2, 57.7, 62.3, 63.7, 65.4,
68.5, 70.0, 70.6, 72.7, 73.5, 73.8, 74.2, 77.1, 77.9, 78.0, 83.4, 83.7,
94.5, 102.6, 120.8, 128.5, 132.0, 139.6, 178.3; HRMS (FAB, mnba)
calcd for [C46H76N2O12 + H]+ 849.5476, found 849.5411.
1
50 mg (92%) of 19a as a white-brown solid. H NMR (CDCl3,
400 MHz) δ 0.82-0.90 (3H, m), 0.98 (3H, d, J ) 7.6 Hz),
1.05-1.52 (24H, m), 1.60-1.74 (6H, m), 1.80-2.06 (9H, m),
2.22-2.37 (9H, m), 2.56 (3H, m), 2.67 (3H, m), 2.95 (2H, t, J )
9.8 Hz), 3.07 (3H, m), 3.29-3.34 (1H, m), 3.46 (3H, m), 3.54 (1H,
d, J ) 6.8 Hz), 3.61 (3H, s), 3.68 (1H, bs), 3.77 (1H, m), 3.97
(1H, m), 4.18 (1H, m), 4.34-4.38 (3H, m), 4.69 (1H, m), 5.06
(1H, d, J ) 4 Hz), 7.32 (2H, d, J ) 6.4 Hz), 7.73-7.75 (3H, m);
13C NMR (CDCl3, 100 MHz) δ 8.7, 9.2, 11.3, 14.2, 14.7, 16.5,
18.2, 21.4, 21.5, 22.2, 24.2, 25.9, 26.6, 27.3, 29.7, 30.0, 33.6, 34.6,
36.4, 36.9, 42.4, 45.3, 45.8, 49.3, 50.0, 51.5, 57.7, 63.9, 65.5, 68.6,
69.4, 70.5, 72.7, 73.8, 74.2, 77.2, 77.6, 78.0, 83.4, 94.4, 102.7,
119.3, 125.5, 129.1, 129.4, 147.2, 173.4, 178.1; HRMS (FAB,
mnba) calcd for [C53H89N5O14 + H]+ 1020.6484, found 1020.6430.
Methyl Descladinoseazithromycin-N-benzyltriazolylhexanoate
(19b). Compound 13 (0.080 g, 0.115 mmol) and azido ester 18
(0.030 g, 0.173 mmol) were dissolved in anhydrous THF (5 mL)
and stirred under argon at room temperature. Copper(I) iodide
(0.010 g, 0.053 mmol) and Hunig’s base (0.05 mL) were then added
to the reaction mixture, and stirring continued for 12 h. The reaction
mixture was diluted with CH2Cl2 (30 mL) and washed with 1:4
NH4OH/saturated NH4Cl (3 × 25 mL) and again with saturated
NH4Cl (25 mL). The organic layer was dried over Na2SO4 and
concentrated in vacuo. The crude product was purified by prepara-
tive TLC, eluting with hexane/EtOAc/Et3N 3:2:0.1 to give 65 mg
(65%) of 19b as a white-brown solid. 1H NMR (CDCl3, 400 MHz)
δ 0.79-0.86 (6H, m), 1.00-1.07 (6H, m), 1.17-1.26 (m),
1.42-1.51 (m), 1.55-1.72 (6H, m), 1.80-1.94 (6H, m), 2.00-2.05
(2H, m), 2.1 (3H, s), 2.23-2.27 (4H, m), 2.33 (3H, s), 2.47 (1H,
d, J ) 10.4 Hz), 2.58-2.72 (5H, m), 3.32-3.41 (3H, m), 3.52-3.73
(6H, m), 3.92-4.00 (2H, m), 4.34 (2H, t, J ) 7.0 Hz), 4.41 (1H,
d, J ) 7.6 Hz), 4.69 (1H, d, J ) 10.8 Hz), 7.24 (2H, d, J ) 8.4
Hz), 7.71 (2H, d, J ) 8 Hz), 7.73 (1H, s); 13C NMR (CDCl3, 100
MHz) δ 7.7, 7.9, 8.7, 10.9, 16.1, 16.1, 20.9, 21.2, 24.2, 25.8, 25.9,
26.6, 29.2, 29.6, 30.0, 33.6, 36.0, 36.3, 37.1, 42.0, 44.5, 45.8, 50.0,
51.5, 57.7, 62.6, 65.1, 69.9, 70.4, 73.1, 74.1, 75.3, 79.4, 94.8, 106.4,
119.3, 125.5, 128.9, 129.6, 138.2, 147.1, 173.4, 177.2. HRMS (EI)
calcd for [C45H75N5O11 + H]+ 862.5541, found 862.5566.
Descladinose-4′-ethynylbenzylazithromycin (13). A solution of
4′-ethynylbenzylazithromycin 12 (0.12 g, 0.14 mmol) in 0.25 N
HCl (15 mL) was stirred at room temperature for 20 h and poured
into EtOAc (20 mL). The two layers were separated, and the
aqueous layer was washed with EtOAc (2 × 20 mL), basified with
concentrated NH4OH, and then extracted with 5% MeOH in CH2Cl2
(2 × 30 mL). The combined organic layer was washed with
saturated brine (30 mL) and dried over Na2SO4. Solvent was
evaporated off to give 89 mg (91%) of descladinose compound 12
1
as a white solid. H NMR (CDCl3, 400 MHz) δ 0.80-1.52 (16H,
m), 1.65-1.90 (2H, m), 1.97 (3H, s), 2.20-2.30 (6H, m), 2.42-2.69
(3H, m), 2.77 (1H, s), 2.89 (1H, s), 3.02 (1H, s), 3.18 (1H, s),
3.30-3.37 (3H, m), 3.49-3.63 (6H, m), 3.72 (2H, d, J ) 10.6
Hz), 3.84 (1H, s), 3.89 (2H, s), 4.04 (3H, q, J ) 14.4, 7.2 Hz),
4.41 (1H, d, J ) 7.2 Hz), 4.64 (1H, d, J ) 10.8 Hz), 7.15 (2H, d,
J ) 8.0 Hz), 7.37 (2H, d, J ) 8.4 Hz); 13C NMR (CDCl3, 100
MHz) δ 7.7, 7.9, 10.9, 14.2, 16.1, 20.9, 21.0, 21.2, 25.8, 26.6, 29.2,
35.9, 36.4, 37.0, 42.0, 44.5, 57.6, 60.3, 62.4, 65.3, 69.8, 70.4, 70.9,
73.0, 74.1, 75.4, 79.4, 83.3, 94.9, 106.4, 120.8, 128.2, 131.9, 139.2,
171.0, 177.2; HRMS (FAB, thioglycerol) calcd for [C38H62N2O9
+ H]+ 691.4533, found 691.4513.
Azithromycin-N-benzyltriazolyl-O-silylhexahydroxamate (15).
4′-Ethynylbenzylazithromycin 12 (0.045 g, 0.050 mmol) and
6-azido-O-silyl hexahydroxamate 14 (0.060 g, 0.146 mmol) were
dissolved in anhydrous THF (5 mL) and stirred under argon at room
temperature. Copper(I) iodide (0.010 g, 0.050 mmol), Hunig’s base
(0.5 mL), and tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine
(TBTA) (0.016 g, 0.030 mmol) were then added to the reaction
mixture, and stirring continued for 2 h. The reaction mixture was
diluted with CH2Cl2 (40 mL) and washed with 1:4 NH4OH/saturated
NH4Cl (2 × 30 mL) and saturated NH4Cl (30 mL). The organic
layer was dried over Na2SO4 and concentrated in vacuo. The crude
product was purified by preparative TLC (silica, 12:1:0.1 CH2Cl2/
MeOH/concentrated NH4OH) to give 38 mg (60%) of silyl protected
compound 15 as a brown-white solid. 1H NMR (CDCl3, 400 MHz)
δ 0.82-1.52 (39H, m) 1.71-2.04 (13H, m), 2.21 (3H, s), 2.28 (6H,
s), 2.44-2.57 (2H, m), 2.65-2.70 (2H, m), 2.96 (1H, br s), 3.09
(2H, s), 3.31-3.35 (2H, m), 3.40-3.48 (6H, m), 3.56-3.60 (1H,
m), 3.76 (1H, d, J ) 13.2 Hz), 4.00 (1H, br s), 4.20 (3H, br s),
4.39 (1H, d, J ) 6.8 Hz), 4.64 (1H, d, J ) 9.2 Hz), 5.09 (1H, br
s) 7.29-7.41 (10H, m), 7.64-7.83 (5H, m); 13C NMR (CDCl3,
Representative Procedure for Conversion of Methyl or Ethyl
Ester to Hydroxamic Acid. 6-Azidohexahydroxamic Acid (17a).
To a solution of ethyl 6-azidohexanoate (1.00 g, 5.840 mmol) in
1:1 THF (5 mL) and anhydrous methanol (5 mL) were added
aqueous hydroxylamine (4 mL, 70.100 mmol) and KCN (0.070 g,
1.170 mmol), and the stirring continued for 24 h. The mixture was
diluted with EtOAc (30 mL) and washed with saturated NaHCO3
(2 × 30 mL) and saturated brine (30 mL). The organic layer was
dried over Na2SO4 and concentrated in vacuo to give 797 mg (80%)
of 17a as white solid. 1H NMR (DMSO-d6, 400 MHz) δ 1.24-1.30
(2H, m), 1.45-1.53 (4H, m), 1.93 (2H, t, J ) 7.2 Hz), 3.29 (2H,
t, J ) 6.8 Hz), 8.65 (1H, s), 10.3 (1H, s); 13C NMR (CDCl3, 100
MHz) δ 24.8, 26.1, 28.4, 32.6, 51.1, 171.4; HRMS (ESI) calcd for
[C6H12N4O2 + H]+ 173.0947, found 173.0983.
7-Azidoheptahydroxamic Acid (17b). Reaction of ethyl 7-azi-
doheptanoate (1.00 g, 5.400 mmol) and aqueous hydroxylamine

464 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Oyelere et al.
(4 mL, 70.100 mmol) within 24 h as described for the synthesis of
17a gave 820 mg (82%) of 17b as a white solid. 1H NMR (DMSO-
d6, 400 MHz) δ 1.23-1.32 (4H, m), 1.45-1.52 (4H, m), 1.92 (2H,
t, J ) 7.6 Hz), 3.29 (2H, t, J ) 7.2 Hz), 8.63 (1H, br s), 10.3 (1H,
br s); 13C NMR (CDCl3, 100 MHz) δ 25.1, 26.2, 28.5, 28.6, 32.7,
51.2, 171.6; HRMS (FAB, thioglycerol) calcd for [C7H14N4O2 +
H]+ 187.1195, found 187.1163.
8-Azidooctahydroxamic Acid (17c). Reaction of methyl 8-azi-
dooctanoate (0.580 g, 2.910 mmol) and aqueous hydroxylamine
(2.49 mL, 37.800 mmol) within 24 h as described for the synthesis
of 17a gave 432 mg (74%) of 17c as white solid. 1H NMR (CDCl3,
400 MHz) δ 1.33-1.38 (6H, m), 1.54-1.63 (4H, m), 2.13 (2H, t,
J ) 7.6 Hz), 3.24 (2H, t, J ) 6.8 Hz), 8.85 (1H, br); 13C NMR
(CDCl3, 100 MHz) δ 25.5, 26.7, 29.0, 29.1, 29.2, 33.1, 51.6, 172.0;
HRMS (FAB, thioglycerol) calcd for [C8H16N4O2 + H]+ 201.1351,
found 201.1352.
9-Azidononahydroxamic Acid (17d). Reaction of methyl 9-azi-
dononanoate (0.290 g, 1.370 mmol) and aqueous hydroxylamine
(1.18 mL, 17.900 mmol) within 24 h as described for the synthesis
of 17a gave 225 mg (77%) of 17d as white solid. 1H NMR (CDCl3,
400 MHz) δ 1.30-1.40 (8H, m), 1.56-1.61 (4H, m), 2.13 (2H, t,
J ) 7.2 Hz), 3.25 (2H, t, J ) 8); 13C NMR (CDCl3, 100 MHz) δ
25.6, 26.8, 29.0, 29.2, 29.3, 33.1, 51.6, 172.2; HRMS (FAB,
thioglycerol) calcd for [C9H18N4O2 + H]+ 215.1508, found
215.1529.
10-Azidodecahydroxamic Acid (17e). Reaction of methyl 10-
azidodecanoate (0.300 g, 1.310 mmol) and aqueous hydroxylamine
(1.13 mL, 17.100 mmol) within 24 h as described for the synthesis
of 17a gave 254 mg (84%) of 17e as white solid. 1H NMR (CDCl3,
400 MHz) δ 1.24-1.39 (8H, m), 1.54-1.65 (4H, m), 2.12 (2H, t,
J ) 7.6 Hz), 3.24 (2H, t, J ) 7.2 Hz), 9.0 (1H, br); 13C NMR
(CDCl3, 100 MHz) δ 25.6, 26.9, 29.0, 29.2, 29.3, 29.4, 29.5, 33.2,
51.7, 172.2; HRMS (FAB, thioglycerol) calcd for [C10H20N4O2 +
H]+ 229.1665, found 229.1666
Hz), 1.24-1.29 (15H, m), 1.33-1.47 (3H, m), 1.54 (1H, dd, J )
4.8 Hz, 15.2 Hz), 1.60-1.69 (5H, m), 1.80-2.01 (m), 2.06-2.12
(1H, m), 2.18-2.24 (1H, m), 2.26 (3H, s), 2.28-2.31 (1H, m),
2.35-2.41 (4H, m), 2.51 (1H, d, J ) 10 Hz), 2.65-2.96 (m), 3.12
(3H, s), 3.22-3.29 (1H, m), 3.47 (1H, m), 3.54-3.69 (6H, m),
3.81 (1H, d, J ) 13.2 Hz), 4.11 (1H, m), 4.24 (1H, m), 4.45 (3H,
t, J ) 7.0 Hz), 4.50 (1H, d, J ) 6.8 Hz), 4.75 (1H, d, J ) 7.2 Hz),
4.97 (1H, d, J ) 5.2 Hz), 7.42 (2H, d, J ) 8.0 Hz), 7.84 (2H, d,
J ) 8.0 Hz), 8.35 (1H, s); 13C NMR (CDCl3, 100 MHz) δ 6.5, 8.7,
11.5, 14.4, 16.7, 17.7, 21.3, 21.6, 21.8, 24.6, 25.7, 26.7, 27.0, 29.2,
29.6, 29.9, 33.0, 34.5, 35.6, 36.7, 41.8, 42.7, 45.3, 49.3, 50.0, 53.4,
57.8, 62.6, 63.4, 65.9, 68.6, 69.3, 70.4, 72.6, 73.3, 73.8, 77.8, 78.3,
78.4, 83.4, 94.5, 102.8, 119.6, 125.7, 129.3, 129.7, 138.6, 147.4,
171.3, 178.4; HMRS (ESI) calcd for [C52H88N6O14 + H]+ 1021.6437,
found 1021.6409.
Desclasinose Azithromycin-N-benzyltriazolylhexahydroxamic
Acid (16b). Method A. To a solution of compound 19b (0.040 g,
0.050 mmol) in 1:1 THF/MeOH (3 mL) was added hydroxylamine
(50% in H2O) (0.04 mL, 0.540 mmol) and a catalytic amount of
KCN. The mixture was stirred at room temperature for 24 h. The
mixture was partitioned between 5% MeOH in CH2Cl2 (30 mL)
and saturated sodium bicarbonate (25 mL). The two layers were
separated, and the aqueous layer was extracted with 5% MeOH in
CH2Cl2 (2 × 20 mL). The combined organic layer was washed
with saturated brine (40 mL) and dried over Na2SO4. Solvent was
evaporated off and the crude was purified by preparative TLC,
eluting with CH2Cl2/MeOH/NH4OH 10:1:0.1 to give 9.0 mg (23%)
of compound 16b as brown-white solid.
Method B. Reaction of descladinose compound 13 (0.134 g,
0.188 mmol) and 6-azidohexahydroxamic acid 17a (0.130 g, 0.755
mmol) within 8 h, according to the protocols of method B described
for the synthesis of compound 16a, followed by preparative TLC
(silica, 10:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 73 mg
1
(43%) of 16b as a brown-white solid. H NMR (acetone-d6, 400
MHz) δ 0.82-0.90 (9H, m), 1.02 (3H, d, J ) 7.2 Hz), 1.07 (3H,
s), 1.09 (3H, d, J ) 6.8 Hz), 1.18-1.23 (m), 1.28 (3H, s), 1.31-1.39
(m), 1.46-1.56 (4H, m), 1.61-1.65 (3H, m), 1.80-1.99 (7H, m),
2.05-2.11 (2H, m), 2.18-2.21 (1H, m), 2.24 (3H, s), 2.25-2.29
(1H, m), 2.35 (3H, s), 4.47 (1H, d, J ) 9.2 Hz), 2.61-2.67 (1H,
m), 2.70-2.77 (1H, m), 3.30-3.34 (1H, m), 3.41 (1H, m),
3.52-3.65 (5H, m), 3.81 (1H, d, J ) 13.2 Hz), 4.44 (2H, t, J )
7.0 Hz), 4.59 (1H, d, J ) 7.6 Hz), 4.87 (1H, dd, J ) 1.8 Hz, 11.0
Hz), 7.43 (2H, d, J ) 8.4 Hz), 7.83 (2H, d, J ) 8.4 Hz), 8.34 (1H,
s); 13C NMR (CDCl3, 100 MHz) δ 7.3. 7.9, 10.7, 16.0, 16.2, 20.9,
21.1, 24.4, 25.6, 25.7, 26.5, 29.0, 29.6, 29.7, 35.8, 36.3, 36.8, 42.1,
44.4, 50.0, 57.9, 62.7, 64.0, 69.8, 70.5, 70.8, 73.3, 74.1, 75.0, 79.5,
94.8, 106.5, 120.0, 125.8, 129.2, 129.5, 138.4, 147.3, 170.5, 177.5;
HMRS (ESI) calcd for [C44H74N6O11 + H]+ 863.5494, found
863.5544.
Azithromycin-N-benzyltriazolylhexahydroxamic Acid (16a).
Method A. To a solution of compound 19a (0.040 g, 0.040 mmol)
in 1:1 THF/MeOH (3 mL) was added hydroxylamine (50% in H2O)
(0.03 mL, 0.540 mmol) and a catalytic amount of KCN. The mixture
was stirred at room temperature for 24 h. The mixture was
partitioned between 5% MeOH in CH2Cl2 (30 mL) and saturated
sodium bicarbonate (25 mL). The two layers were separated, and
the aqueous layer was extracted with 5% MeOH in CH2Cl2 (2 ×
20 mL). The combined organic layer was washed with saturated
brine (40 mL) and dried over Na2SO4. Solvent was evaporated off
and the crude was purified by preparative TLC, eluting with CH2Cl2/
MeOH/NH4OH 10:1:0.1 to give 6.5 mg (16%) of compound 16a
as a brown-white solid.
Method B. 4′-Ethynylbenzylazithromycin 12 (0.100 g, 0.109
mmol) and 6-azidohexahydroxamic acid 17a (0.081 g, 0.117 mmol)
were dissolved in anhydrous THF (5 mL) and stirred under argon
at room temperature. Copper(I) iodide (0.011 g, 0.070 mmol) and
Hunig’s base (0.5 mL) were then added to the reaction mixture,
and stirring continued for 4 h. The reaction mixture was diluted
with CH2Cl2 (40 mL) and washed with 1:4 NH4OH/saturated NH4Cl
(3 × 30 mL) and saturated NH4Cl (30 mL). The organic layer was
dried over Na2SO4 and concentrated in vacuo. The crude product
was purified by preparative TLC (silica, 12:1:0.1 CH2Cl2/MeOH/
concentrated NH4OH) to give 71 mg (59%) of 16a as a brown-
white solid.
Azithromycin-N-benzyltriazolylheptahydroxamic Acid (16c).
Reaction of 4′-ethynylbenzylazithromycin 12 (0.134 g, 0.158 mmol)
and 7-azidoheptahydroxamic acid 17b (0.125 g, 0.672 mmol) within
4 h, according to the protocols of method B described for the
synthesis of compound 16a, followed by preparative TLC (silica,
12:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 93 mg (56%)
1
of 16c as a brown-white solid. H NMR (CDCl3, 400 MHz) δ
0.81-1.51 (30H, m), 1.54-1.65 (6H, m), 1.70-2.14 (9H, m),
2.20-2.38 (9H, m), 2.46-2.56 (2H, m), 2.60-2.70 (2H, m), 3.00
(3H, s), 3.31 (2H, t, J ) 8.8 Hz), 3.38-3.54 (6H, m), 3.60 (1H, s),
3.78 (1H, d, J ) 12.8 Hz), 3.98-4.20 (2H, m), 4.36 (3H, d, J )
7.2 Hz), 4.49 (1H, d, J ) 7.2 Hz), 5.11 (1H, d, J ) 4.0 Hz), 7.32
(2H, d, J ) 7.6 Hz), 7.73 (1H, s), 7.75 (2H, d, J ) 7.6 Hz); 13C
NMR (CDCl3, 100 MHz) δ 6.6, 8.8, 11.5, 14.4, 16.6, 17.7, 21.3,
21.6, 21.8, 25.1, 26.0, 26.7, 27.1, 28.2, 29.2, 29.6, 30.0, 33.1, 34.5,
35.7, 36.7, 41.8, 42.7, 45.3, 49.3, 50.3, 50.7, 57.9, 62.7, 63.0, 65.8,
68.6, 69.4, 70.4, 72.6, 73.2, 73.8, 77.8, 78.1, 78.2, 83.5, 94.4, 102.8,
119.3, 125.7, 129.4, 129.7, 138.4, 147.4, 171.3, 178.4; HMRS (ESI)
calcd for [C53H90N6O14 + H]+ 1035.6587, found 1035.6628.
Descladinose Azithromycin-N-benzyltriazolylheptahydroxamic
Acid (16d). Reaction of descladinose-4′-ethynylbenzylazithromycin
13 (0.130 g, 0.188 mmol) and 7-azidoheptahydroxamic acid 17b
Method C. To a solution of silyl protected compound 15 (0.025
g, 0.020 mmol) in THF (1 mL) was added 1 M TBAF in THF
(0.030 mL, 0.030 mmol), and the mixture was stirred at room
temperature for 2 h, during which TLC revealed a near-quantitative
conversion to a lower Rf product. The mixture was partitioned
between CH2Cl2 (30 mL) and saturated NH4Cl (25 mL). The two
layers were separated, and the organic layer was dried over Na2SO4
and concentrated in vacuo. The crude product was purified by
preparative TLC (silica, 12:1:0.1 CH2Cl2/MeOH/Et3N) to give 15
1
mg (73%) of 16a as a brown-white solid. H NMR (acetone-d6,
400 MHz) δ 0.83-0.92 (6H, m), 1.02 (3H, d, J ) 7.6 Hz),
1.08-1.11 (8H, m), 1.14 (3H, d, J ) 7.6 Hz), 1.18 (3H, d, J ) 6

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 465
(0.130 g, 0.755 mmol) within 8 h, according to the protocols of
Method B described for the synthesis of compound 16a, followed
by preparative TLC (silica, 10:1:0.1 CH2Cl2/MeOH/concentrated
0.85-1.36 (30H, m), 1.41-2.24 (22H, m), 2.28, 2.36 (6H, s),
2.33-3.10 (8H, m), 3.05 (3H, s), 3.23-3.82 (9H, m), 4.06-4.10
(1H, m), 4.36-4.41 (3H, m), 4.49 (1H, d, J ) 8.0 Hz), 5.15 (1H,
d, J ) 4.0 Hz), 7.34 (2H, d, J ) 8 Hz), 7.75 (1H, s), 7.78 (2H, d,
J ) 8.0 Hz); 13C NMR (DMSO-d6, 100 MHz) δ 6.7, 9.0, 11.7,
14.6, 16.9, 17.9, 21.6, 21.9, 22.0, 25.6, 26.4, 26.8, 27.0, 27.3, 28.8,
29.0, 21.9, 29.3, 29.5, 29.9, 30.3, 33.6, 34.9, 35.8, 37.0, 42.1, 43.0,
45.6, 49.6, 50.6, 51.6, 58.0, 62.8, 63.9, 66.2, 68.9, 69.6, 70.7, 72.9,
73.5, 74.0, 74.1, 78.2, 78.6, 83.6, 94.6, 103.0, 119.6, 126.0, 129.6,
129.9, 138.9, 147.6, 178.6; HRMS (MALDI) calcd for [C56H96N6O14
+ H]+ 1077.7057, found 1077.6971.
4′-Desmethylclarithromycin (20). To a solution of clarithromycin
(3.320 g, 4.440 mmol) and sodium acetate (3.280 g, 39.900 mmol)
in 80% aqueous methanol (50 mL) at 85 °C was added iodine (1.240
g, 4.890 mmol) in three batches within 5 min. The mixture was
maintained at pH 8-9 by additions of 1 M NaOH (2 × 3 mL,
once at 10 and 45 min of reaction time). Stirring was continued at
85 °C for 3 h, and TLC analysis indicated about 90% consumption
of the starting material. A solution of 5% sodium thiosulfate (120
mL) and dichloromethane (80 mL) was added, and the two layers
were separated. The aqueous layer was extracted with CH2Cl2 (60
mL) and the combined organic layers were washed with saturated
brine, dried over Na2SO4, and concentrated in vacuo to give 2.4 g
of 20, which was used without further purification.
4′-Ethynylbenzylclarithromycin (21). To a solution of 4′-
desmethylclarithromycin 20 (2.400 g, 3.340 mmol) in anhydrous
DMSO (30 mL) was added Hunig’s base (3 mL) and 4-ethynyl-
benzyl methanesulfonate 11 (0.920 g, 4.340 mmol). The reaction
mixture was then heated with stirring under argon at 85 °C for
2.5 h. The mixture was cooled and diluted with EtOAc (100 mL)
and washed with saturated NaHCO3 (3 × 60 mL) and saturated
brine (60 mL). The organic layer was dried over Na2SO4 and
concentrated in vacuo. The crude product was purified by flash
chromatography (silica, gradient 12:1; 10:1; 8:1; CH2Cl2/acetone)
to give 1.8 g (63%) of 21 as a brown-white solid. 1H NMR (CDCl3,
400 MHz) δ 0.82 (3H, t, J ) 7.2 Hz), 1.03-1.28 (18H, m), 1.37
(3H, s), 1.40-1.55 (3H, m), 1.65-1.90 (6H, m), 2.03 (1H, d, J )
10.0 Hz), 2.22 (3H, s), 2.30 (1H, d, J ) 15.2 Hz), 2.40-2.60 (2H,
m), 2.80-2.90 (2H, m), 2.94-3.00 (6H, m), 3.04 (1H, s), 3.09
(3H, s), 3.16 (1H, s), 3.24-3.29 (1H, m), 3.38-3.46 (3H, m), 3.59
(1H, d, J ) 6.8 Hz), 3.70-3.75 (3H, m), 3.88-3.95 (1H, m), 4.37
(1H, d, J ) 7.2 Hz), 4.88 (1H, d, J ) 4.4 Hz), 5.02 (1H, dd, J )
11.6, 2.4 Hz), 7.23 (2H, d, J ) 12.0 Hz), 7.42 (2H, d, J ) 8.0 Hz);
13C NMR (CDCl3, 100 MHz) δ 9.2, 10.7, 12.4, 16.1, 18.1, 18.7,
19.9, 21.1, 21.5, 29.3, 32.4, 34.8, 36.9, 37.2, 39.2, 45.0, 45.2, 49.3,
50.6, 53.4, 57.6, 63.3, 65.6, 68.5, 69.0, 70.6, 72.5, 74.2,76.5, 77.8,
78.1, 78.2, 80.8, 95.8, 102.5, 120.9, 128.6, 132.0, 133.5, 139.4,
175.4; HRMS (ESI) calcd for [C46H73NO13 + H]+ 848.5155, found
848.5181.
1
NH4OH) gave 78 mg (47%) of 16d as a brown-white solid. H
NMR (CDCl3, 400 MHz) δ 0.66-2.32 (42H, m), 2.47 (2H, d, J )
10.8 Hz), 2.63-2.70 (4H, m), 3.34-3.51 (6H, m), 3.62-3.69 (5H,
m), 4.20-4.40 (5H, m), 4.74 (2H, br s), 7.26 (2H, br s), 7.73 (3H,
br s); 13C NMR (CDCl3, 100 MHz) δ 7.4, 7.9, 10.7, 16.0, 16.1,
20.9, 21.1, 25.0, 25.7, 26.5, 28.0, 28.9, 29.8, 35.8, 36.3, 36.9, 42.0,
44.4, 50.1, 57.9, 62.7, 63.9, 69.9, 70.4, 70.8, 73.3, 74.1, 75.2, 79.5,
94.9, 106.6, 119.7, 125.7, 129.2, 129.6, 138.4, 147.3, 177.5; HMRS
(ESI) calcd for [C45H76N6O11 + H]+ 877.5645, found 877.5665.
Azithromycin-N-benzyltriazolyloctahydroxamic Acid (16e). Re-
action of 4′-ethynylbenzylazithromycin 12 (0.100 g, 0.120 mmol)
and 8-azidooctahydroxamic acid 17c (0.047 g, 0.240 mmol) within
2.5 h, according to the protocols of method B described for the
synthesis of compound 16a, followed by preparative TLC (silica,
12:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 72 mg (58%)
1
of 16e as brown-white solid. H NMR (CDCl3, 400 MHz) δ 0.85
(3H, t, J ) 4.0 Hz), 0.87-1.22 (18H, m), 1.29 (9H, s), 1.30-2.28
(18H, m), 2.29 (6H, s), 2.30-3.00 (8H, m), 3.10 (3H, s), 3.20-3.79
(9H, m), 3.99-4.03 (1H, m), 4.35-4.40 (3H, m), 4.65 (1H, d, J )
8.0 Hz), 5.11 (1H, d, J ) 4.8 Hz), 7.34 (2H, d, J ) 8.0 Hz), 7.72
(1H, s), 7.77 (2H, d, J ) 8.0 Hz); 13C NMR (DMSO-d6, 100 MHz)
δ 7.5, 9.8, 11.6, 15.4, 18.2, 19.1, 21.5, 22.1, 22.7, 25.6, 26.3, 26.6,
28.7, 29.0, 29.6, 30.2, 32.0, 32.8, 35.2, 36.4, 37.2, 42.2, 45.3, 49.2,
50.1, 58.3, 63.2, 65.4, 67.7, 70.8, 73.3, 74.2, 77.0, 78.4, 83.4, 102.8,
121.6, 125.6, 129.7, 130.0, 135.0, 147.0, 177.8; HRMS (FAB,
thioglycerol) calcd for [C54H92N6O14 + H]+ 1049.6749, found
1049.6648.
Descladinose Azithromycin-N-benzyltriazolyloctahydroxamic
Acid (16f). Reaction of descladinose-4′-ethynylbenzylazithromycin
13 (0.100 g, 0.144 mmol) and 8-azidooctahydroxamic acid 17c
(0.049 g, 0.246 mmol) within 2.5 h, according to the protocols of
method B described for the synthesis of compound 16a, followed
by preparative TLC (silica, 10:1:0.1 CH2Cl2/MeOH/concentrated
NH4OH) gave 94 mg (73%) of 16f as a brown-white solid. 1H NMR
(CDCl3, 400 MHz) δ 0.84-0.91 (9H, m), 1.04-1.11 (9H, m),
1.23-1.30 (12H, m), 1.37-2.16(14H, m), 2.2 (3H, s), 2.35 (3H,
s), 2.49-2.74 (5H, m), 3.22-3.75 (6H, m), 4.10-4.12 (1H, m),
4.34-4.42 (3H, m), 4.77 (1H, d, J ) 12 Hz), 7.30 (2H, d, J ) 7.6
Hz), 7.79 (3H, m); 13C NMR (CDCl3, 100 MHz) δ 7.6, 8.2, 11.0,
14.4, 16.2, 16.4, 21.1, 21.4, 25.2, 25.9, 26.1, 26.7, 28.4, 28.8, 28.9,
29.1, 29.9, 30.2, 36.1, 36.6, 42.3, 44.6, 50.5, 51.6, 58.2, 60.6, 63.0,
64.0, 70.1, 70.7, 73.6, 74.4, 75.4, 79.7, 95.1, 106.8, 120.0, 126.0,
129.5, 129.9, 138.6, 147.6, 170.6, 177.8; HRMS (FAB, thioglycerol)
calcd for [C46H78N6O11 + H]+ 891.5806, found 891.5776.
Azithromycin-N-benzyltriazolylnonahydroxamic Acid (16g). Re-
action of 4′-ethynylbenzylazithromycin 12 (0.100 g, 0.120 mmol)
and 9-azidononahydroxamic acid 17d (0.043 g, 0.200 mmol) within
2.5 h, according to the protocols of method B described for the
synthesis of compound 16a, followed by preparative TLC (silica,
12:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 64 mg (51%)
of 16g as brown-white solid. 1H NMR (CDCl3, 400 MHz) δ
0.84-1.30 (30H, m), 1.33-2.26 (20H, m), 2.30 (6H, s), 2.38-2.68
(8H, m), 2.99 (3H, s), 3.32-3.84 (9H, m), 4.03-4.08 (1H, m),
4.35-4.41 (3H, m), 4.53 (1H, d, J ) 8.0 Hz), 5.13 (1H, d, J ) 4.0
Hz), 7.35 (2H, d, J ) 8.0 Hz), 7.75 (1H, s), 7.78 (2H, d, J ) 8.0
Hz); 13C NMR (DMSO-d6, 100 MHz) δ 6.9, 9.0, 11.6, 14.7, 16.9,
18.0, 21.6, 21.8, 22.1, 25.6, 26.4, 26.9, 27.3, 28.8, 29.0, 29.1, 29.5,
29.9, 30.4, 34.8, 36.0, 37.0, 42.1, 43.0, 45.6, 49.5, 50.5, 58.1, 63.5,
66.1, 68.8, 70.7, 72.9, 74.1, 78.1, 78.3, 78.5, 83.7, 94.4, 94.7, 103.1,
119.6, 126.0, 129.7, 130.0, 147.6, 178.7; HRMS (ESI) calcd for
[C55H94N6O14 + H]+ 1063.6900, found 1063.6861.
Descladinose-4′-ethynylbenzylclarithromycin (22). To a solution
of 4′-ethynylbenzylclarithromycin 21 (0.500 g, mmol) in ethanol
(20 mL) was added 1 N HCl (20 mL), and stirring continued for
22 h at room temperature. The reaction mixture was basified with
concentrated NH4OH to about pH 9. The reaction mixture was
diluted with distilled water (40 mL) and extracted with EtOAc (3
× 60 mL). The combined organic layers were washed with saturated
brine (40 mL), dried over Na2SO4, and concentrated in vacuo. The
crude product was purifed by flash chromatography (silica, 8:1
CH2Cl2/acetone) to give 320 mg (79%) of 22 as a brown-white
1
solid. H NMR (CDCl3, 400 MHz) δ 0.82 (3H, t, J ) 7.6 Hz),
1.09-1.28 (12H, m), 1.34 (3H, s), 1.40-1.55 (3H, m), 1.70-1.74
(2H, m),1.87-1.94 (3H, m), 2.08-2.15 (6H, m), 2.54-2.66 (2H,
m), 2.94-2.98 (3H, m), 3.05 (1H, s), 3.25 (1H, s), 3.31-3.42 (2H,
m), 3.48-3.56 (2H, m), 3.66 (2H, d, J ) 10.0 Hz), 3.82 (1H, s),
3.90 (1H, s), 4.35 (1H, d, J ) 7.6 Hz), 5.14 (1H, dd, J ) 10.8, 2.0
Hz), 7.18 (2H, d, J ) 8.0 Hz), 7.42 (2H, d, J ) 8.4 Hz); 13C NMR
(CDCl3, 100 MHz) δ 8.4, 10.5, 12.7, 15.3, 16.3, 17.8, 18.8, 21.4,
29.2, 35.9, 36.6, 37.5, 38.7, 44.5, 45.5, 49.6, 57.8, 65.0, 69.7, 70.1,
70.6, 74.1, 77.9, 78.9, 83.3, 88.5, 106.5, 121.0, 128.4, 132.1, 139.1,
174.7; HRMS (ESI) calcd for [C38H59NO10 + H]+ 690.4212, found
690.4259.
Azithromycin-N-benzyltriazolyldecahydroxamic Acid (16h). Re-
action of 4′-ethynylbenzylazithromycin 12 (0.100 g, 0.120 mmol)
and 10-azidodecahydroxamic acid 17e (0.045 g, 0.20 mmol) within
4.5 h, according to the protocols of method B described for the
synthesis of compound 16a, followed by preparative TLC (silica,
12:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 70 mg (56%)
1
of 16h as a brown-white solid. H NMR (CDCl3, 400 MHz) δ

466 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Oyelere et al.
Clarithromycin-N-benzyltriazolyl-O-silyl-hexahydroxamate (23).
4′-Ethynylbenzylclarithromycin 21 (0.100 g, 0.118 mmol) and
6-azido-O-silyl hexahydroxamate 14 (0.103 g, 0.251 mmol) were
dissolved in anhydrous THF (6 mL) and stirred under argon at room
temperature. Copper(I) iodide (0.011 g, 0.057 mmol), TBTA (0.060
g, 0.094 mmol), and Hunig’s base (0.1 mL) were then added to
the reaction mixture, and stirring continued for 24 h. The reaction
mixture was diluted with CH2Cl2 (30 mL) and washed with 1:4
NH4OH/saturated NH4Cl (3 × 30 mL) and saturated NH4Cl (30
mL). The organic layer was dried over Na2SO4 and concentrated
in vacuo. The crude product was purified by preparative TLC (silica,
20:1 CH2Cl2/MeOH) to give 147 mg (99%) of compound 23 as a
brown-white solid. 1H NMR (CDCl3, 400 MHz) δ 0.81 (3H, t, J )
7.2 Hz), 1.03-1.51 (30H, m), 1.67-1.88 (12H, m) 2.10-2.29 (8H,
m), 2.53-2.58 (3H, m), 2.82-2.90 (2H, m), 2.93-3.01 (6H, m),
3.10 (3H, s), 3.17 (2H, s), 3.28-3.32 (1H, m), 3.40-3.48 (3H,
m), 3.60-3.78 (4H, m), 3.91-3.97 (1H, m), 4.22 (1H, br s), 4.41
(2H, d, J ) 8.0 Hz), 4.87 (1H, d, J ) 4.8 Hz), 5.03 (1H, d, J )
10.8 Hz), 7.21-7.23 (2H, m), 7.30-7.42 (10H, m), 7.64-7.76 (3H,
m); 13C NMR (CDCl3, 100 MHz) δ 9.1, 10.6, 12.3, 16.0, 18.0,
18.6, 19.8, 21.0, 21.3, 21.5, 26.7, 29.3, 29.6, 29.9, 34.7, 36.8, 37.1,
39.0, 39.2, 45.0, 45.2, 47.0, 49.3, 49.9, 50.5, 53.4, 54.0, 65.5, 68.5,
68.9, 70.6, 72.3, 74.1, 76.4, 77.7, 78.1, 78.2, 80.7, 95.7, 102.5,
119.3, 123.5, 125.4, 127.7, 128.4, 128.8, 129.1, 133.5, 134.4, 135.5,
143.9, 147.0, 175.4; HMRS (ESI) calcd for [C68H103N5O15Si +
H]+1258.7292, found 1258.7313.
Clarithromycin-N-benzyltriazolylhexahydroxamic Acid (24a).
Method A. 4′-Ethynylbenzylclarithromycin 21 (0.100 g, 0.120
mmol) and 6-azidohexahydroxamic acid 17a (0.080 g, 0.470 mmol)
were dissolved in anhydrous THF (5 mL) and stirred under argon
at room temperature. Copper(I) iodide (0.011 g, 0.057 mmol) and
Hunig’s base (0.5 mL) were added to the reaction mixture, and
stirring continued for 2.5 h. The reaction mixture was diluted with
CH2Cl2 (50 mL) and washed with 1:4 NH4OH/saturated NH4Cl (3
× 30 mL) and saturated NH4Cl (30 mL). The organic layer was
dried over Na2SO4 and concentrated in vacuo. The crude product
was purified by preparative TLC (silica, 12:1:0.1 CH2Cl2/MeOH/
concentrated NH4OH) to give 70 mg (58%) of 24a as a brown-
white solid.
Method B. To a solution of compound 23 (0.085 g, 0.067 mmol)
in anhydrous THF (1.5 mL) was added 1 M TBAF in THF (0.1
mL, 0.100 mmol). The mixture was stirred under argon for 2 h.
The reaction mixture was diluted with 5% MeOH in dichlo-
romethane (10 mL) and washed with saturated NH4Cl (15 mL).
The organic layer was dried over Na2SO4 and concentrated in vacuo.
The crude product was purified on preparative TLC (silica, 12:1:
0.1 CH2Cl2/MeOH/Et3N). The purified product was dissolved in
5% MeOH in CH2Cl2 (10 mL) and washed with distilled water (10
mL) and saturated brine (10 mL) to remove the last trace of
associated TBAF. The organic layer was dried over Na2SO4 and
concentrated in vacuo to give 26 mg (38%) of 24a as a brown-
white product. 1H NMR (CDCl3, 400 MHz) δ 0.81 (3H, t, J ) 7.6
Hz), 1.03-1.52 (21H, m), 1.62-1.92 (12H, m) 2.04-2.29 (8H,
m), 2.48-2.60 (3H, m), 2.82-2.90 (2H, m), 2.93-2.99 (6H, m),
3.09 (3H, s), 3.19 (2H, s), 3.28-3.33 (1H, m), 3.42-3.46 (3H,
m), 3.60 (1H, d, J ) 7.6 Hz), 3.70-3.80 (3H, m), 3.90-3.98 (1H,
m), 4.37-4.40 (3H, m), 4.87 (1H, d, J ) 4.8 Hz), 5.03 (1H, dd, J
) 11.6, 2.4 Hz), 7.34 (2H, d, J ) 7.6 Hz), 7.77 (2H, d, J ) 7.6
Hz), 7.82 (1H, s); 13C NMR (CDCl3, 100 MHz) δ 9.1, 10.5, 12.2,
15.9, 17.9, 18.5, 19.7, 20.9, 21.2, 21.4, 24.3, 25.5, 29.4, 29.6, 34.7,
36.8, 37.1, 39.0, 39.1, 45.0, 45.1, 49.3, 49.9, 50.5, 53.3, 57.5, 63.6,
65.5, 68.5, 69.0, 70.7, 72.4, 74.2, 77.8, 78.2, 80.9, 95.9, 102.6,
119.8, 125.6, 129.4, 147.4, 175.8; HMRS (ESI) calcd for
[C52H85N5O15 + H]+ 1020.6114, found 1020.6121.
NMR (CDCl3, 400 MHz) δ 0.79 (3H, t, J ) 7.2 Hz), 1.08-1.32
(22H, m), 1.39-1.64 (3H, m), 1.71-1.81 (2H, m), 1.82-1.96 (3H,
m), 2.04-2.18 (6H, m), 2.51-2.70 (4H, m), 2.92-2.98 (3H, m),
3.18-3.38 (3H, m), 3.45-3.55 (2H, m), 3.60-3.74 (3H, m), 3.81
(3H, s), 3.90 (1H, s), 4.33 (3H, br s), 5.13 (1H, d, J ) 10.4 Hz),
7.29 (2H, br s), 7.74 (2H, br s); 13C NMR (CDCl3, 100 MHz) δ
8.6, 10.7, 12.9, 15.5, 16.4, 18.0, 19.0, 21.5, 21.6, 29.5, 29.9, 36.1,
36.7, 37.7, 39.0, 44.7, 45.7, 49.8, 50.3, 58.2, 64.6, 70.0, 70.3, 70.9,
74.4, 78.3, 79.1, 88.5, 106.7, 120.3, 126.1, 129.6, 129.9, 147.7,
175.4; HMRS (ESI) calcd for [C44H71N5O12 + H]+ 862.5172, found
862.5155.
Clarithromycin-N-benzyltriazolylheptahydroxamic Acid (24c).
Reaction of 4′-ethynylbenzylclarithromycin 21 (0.130 g, 0.153
mmol) and 7-azidoheptahydroxamic acid 17b (0.105 g, 0.565 mmol)
within 2.5 h, according to the protocols of method A described for
the synthesis of compound 24a, followed by preparative TLC (silica,
12:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 105 mg (67%)
of 24c as a brown-white solid. 1H NMR (CDCl3, 400 MHz) δ 0.82
(3H, t, J ) 8.0 Hz), 1.04-1.52 (21H, m), 1.67-1.92 (14H, m),
2.14-2.29 (8H, m), 2.52-2.60 (3H, m), 2.82-2.90 (2H, m),
2.95-3.00 (6H, m), 3.10 (3H, s), 3.16 (2H, s), 3.27-3.32 (1H,
m), 3.41-3.46 (3H, m), 3.59 (1H, d, J ) 6.8 Hz), 3.69-3.79 (3H,
m), 3.90-3.95 (1H, m), 4.34-4.39 (3H, m), 4.87 (1H, d, J ) 4.4
Hz), 5.02 (1H, d, J ) 9.2 Hz), 7.33 (2H, d, J ) 6.4 Hz), 7.77 (3H,
d, J ) 8.4 Hz); 13C NMR (CDCl3, 100 MHz) δ 9.1, 10.5, 12.2,
15.9, 17.9, 18.5, 19.8, 20.9, 21.2, 21.4, 24.7, 25.5, 27.7, 29.8, 34.7,
36.8, 37.1, 39.1, 45.0, 45.2, 49.3, 50.0, 50.5, 57.6, 63.6, 65.6, 68.6,
69.0, 70.7, 72.4, 74.2, 77.8, 78.3, 80.9, 95.9, 102.7, 119.5, 125.7,
129.4, 147.5, 175.8; HMRS (ESI) calcd for [C53H87N5O15 + H]+
1034.6271, found 1034.6246.
Descladinose Clarithromycin-N-benzyltriazolylheptahydroxamic
Acid (24d). Reaction of descladinose-4′-ethynylbenzylclarithromy-
cin 22 (0.075 g, 0.109 mmol) and 7-azidoheptahydroxamic acid
17b (0.040 g, 0.233 mmol) within 4 h, according to the protocols
of method A described for the synthesis of compound 24a, followed
by preparative TLC (silica, 10:1:0.1 CH2Cl2/MeOH/concentrated
1
NH4OH) gave 80 mg (84%) of 24d as a brown-white solid. H
NMR (CDCl3, 400 MHz) δ 0.78 (3H, t, J ) 7.2 Hz), 1.06-1.31
(22H, m), 1.40-1.53 (5H, m), 1.71 (2H, d, J ) 11.6 Hz), 1.80-1.91
(3H, m), 2.01-2.20 (6H, m), 2.50-2.65 (4H, m), 2.91-2.97 (3H,
m), 3.16 (1H, t, J ) 6.4 Hz), 3.26-3.35 (2H, m), 3.42-3.54 (2H,
m), 3.64-3.71 (3H, m), 3.80 (1H, br s), 3.90 (1H, br s), 4.30-4.34
(3H, m), 5.12 (1H, dd, J ) 11.6, 2.4 Hz), 7.27 (2H, d, J ) 8.0
Hz), 7.72 (2H, d, J ) 7.2 Hz), 7.80 (1H, s); 13C NMR (CDCl3, 100
MHz) δ 8.3, 10.3, 12.5, 15.2, 16.1, 17.6, 18.6, 21.1, 21.3, 24.8,
25.1, 25.5, 26.2, 27.8, 28.5, 29.1, 29.6, 29.7, 32.3, 35.8, 36.3, 37.4,
38.6, 44.3, 45.4, 49.5, 50.0, 51.2, 57.8, 64.2, 69.7, 69.9, 70.6, 74.1,
77.9, 78.7, 88.0, 106.3, 119.9, 125.7, 129.3, 129.5, 138.3, 147.3,
175.1; HMRS (ESI) calcd for [C45H73N5O12 + H]+ 876.5329, found
876.5301.
Clarithromycin-N-benzyltriazolyloctahydroxamic Acid (24e).
Reaction of 4′-ethynylbenzylclarithromycin 21 (0.101 g, 0.120
mmol) and 8-azidooctahydroxamic acid 17c (0.047 g, 0.240 mmol)
within 2.5 h, according to the protocols of method A described for
the synthesis of compound 24a, followed by preparative TLC (silica,
12:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 92 mg (74%)
of 24e as a brown-white solid. 1H NMR (CDCl3, 400 MHz) δ 0.81
(3H, t, J ) 7.2 Hz), 1.04-2.05 (45H, m), 2.22 (3H, s), 2.19-2.82
(m, 7H), 3.00 (3H, s), 3.08 (3H, s), 2.91-3.80 (6H, m), 3.95 (3H,
m), 4.38 (3H, m), 4.88 (1H, d, J ) 4.4 Hz), 5.04 (1H, dd, J )
10.8, 2.0 Hz), 7.33 (2H, d, J ) 7.6 Hz), 7.71 (1H, s), 7.77 (2H, d,
J ) 8.0 Hz); 13C NMR (CDCl3, 100 MHz) δ 8.8, 9.4, 10.8, 12.5,
16.2, 18.2, 18.8, 20.0, 21.2, 21.5, 21.6, 25.1, 26.0, 26.7, 28.2, 28.6,
28.9, 29.9, 30.1, 35.0, 36.9, 37.4, 39.2, 39.4, 45.2, 45.4, 46.1, 49.6,
50.4, 50.8, 51.6, 58.1, 63.6, 65.9, 68.4, 69.3, 70.9, 72.8, 74.4, 78.0,
78.5, 81.2, 96.1, 120.1, 102.6, 126.1, 130.2, 147.4, 176.0; HRMS
(ESI) calcd for [C54H90N5O15 + H]+ 1048.6486, found 1048.6427.
Descladinose Clarithromycin-N-benzyltriazolyloctahydroxamic
Acid (24f). Reaction of descladinose-4′-ethynylbenzylclarithromycin
22 (0.10 g, 0.144 mmol) and 8-azidooctahydroxamic acid 17c (0.049
g, 0.246 mmol) within 2.5 h, according to the protocols of method
Descladinose Clarithromycin-N-benzyltriazolylhexahydroxamic
Acid (24b). Reaction of descladinose-4′-ethynylbenzylclarithromy-
cin 22 (0.075 g, 0.109 mmol) and 6-azidohexahydroxamic acid 17a
(0.040 g, 0.233 mmol) within 4 h, according to the protocols of
method A described for the synthesis of compound 24a, followed
by preparative TLC (silica, 10:1:0.1 CH2Cl2/MeOH/concentrated
1
NH4OH) gave 47 mg (51%) of 24b as a brown-white solid. H

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 467
A described for the synthesis of compound 24a, followed by
0.1%. To control wells, only fresh medium was added. Viability
was assessed after 72 h by trypan blue staining.
preparative TLC (silica, 10:1:0.1 CH2Cl2/MeOH/concentrated
1
NH4OH) gave 117 mg (90%) of 24f as a brown-white solid. H
Western Blotting. SK-MES-1 cells were passaged 24 h prior to
dosing with test compounds. Compounds were diluted to appropriate
concentrations and were added to fresh medium so that the final
concentration of DMSO was 0.1%. Cells were dosed with com-
pounds for 8 h. Cells were washed with ice-cold PBS, recovered
by centrifugation, and then lysed on ice in RIPA buffer containing
protease inhibitors. Lysates were mixed repeatedly by pipetting and
centrifuged at 14 000 rpm at 4 °C. Protein concentration was
determined by Bradford assay (Bio-Rad). Protein denaturation and
electrophoresis were performed as described.27 Proteins were
transferred to a nitrocellulose membrane for 1 h at 4 °C. The
membrane was then blocked for 1 h in Odyssey blocking buffer
(LI-COR Biosciences). Mouse anti-p21waf1 (Zymed) and rabbit
antiactin (Sigma) primary antibodies and IR-labeled secondary
antibodies (LI-COR Biosciences) were added as recommended (LI-
COR Biosciences), and two-color detection was used to image actin
and p21waf1 by incubating with different host animal secondary
antibodies. Detection was performed on an Odyssey infrared
imaging system (LI-COR Biosciences). Blots were imaged simul-
taneously in both 700 and 800 nm channels at 169 µm resolution
and analyzed on the imager software.
NMR (CDCl3, 400 MHz) δ 0.81 (3H, t, J ) 7.2 Hz), 1.10-2.09
(37H, m), 2.18 (3H, s), 2.19-2.68 (7H, m), 2.98-3.73 (4H, m),
3.83 (2H, s), 3.93 (1H, m), 4.36 (3H, m), 5.16 (1H, d, J ) 8.0 Hz),
7.31 (2H, d, J ) 8.0 Hz), 7.77 (2H, d, J ) 8.0 Hz), 7.79 (1H, s);
13C NMR (CDCl3, 100 MHz) δ 8.5, 10.6, 12.8, 15.4, 16.4, 17.9,
18.9, 21.4, 21.6, 29.4, 36.1, 36.7, 37.7, 38.9, 44.7, 45.7, 49.8, 58.0,
65.2, 70.0, 70.4, 70.8, 74.4, 76.8, 78.2, 79.2, 83.6, 88.8, 106.9,
121.3, 128.7, 132.5, 139.6, 175.2; HRMS (FAB, thioglycerol) calcd
for [C46H76N5O12 + H]+ 890.5490, found 890.5562.
Clarithromycin-N-benzyltriazolylnonahydroxamic Acid (24g).
Reaction of 4′-ethynylbenzylclarithromycin 21 (0.100 g, 0.120
mmol) and 9-azidononahydroxamic acid 17d (0.043 g, 0.20 mmol)
within 2.5 h, according to the protocols of method A described for
the synthesis of compound 24a, followed by preparative TLC (silica,
12:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 54 mg (42%)
of 24g as a brown-white solid. 1H NMR (CDCl3, 400 MHz) δ 0.81
(3H, t, J ) 7.2 Hz), 1.04-2.02 (47H, m), 2.24 (3H, s), 2.10-2.97
(7H, m), 3.00, 3.09 (6H, s), 3.20-3.82 (6H, m), 3.88 (3H, m), 4.39
(3H, m), 4.88 (1H, d, J ) 4.0 Hz), 5.05 (1H, d, J ) 10.0 Hz), 7.35
(2H, d, J ) 8.0 Hz), 7.77 (3H, d, J ) 4.0 Hz); 13C NMR (CDCl3,
100 MHz) δ 9.0, 9.3, 10.8, 12.5, 16.1, 18.2, 18.8, 21.2, 21.5, 21.6,
21.7, 21.8, 26.3, 26.8, 28.7, 28.9, 29.1, 29.9, 30.3, 35.0, 37.0, 37.4,
39.3, 39.4, 45.2, 45.4, 49.6, 50.5, 50.8, 51.6, 57.8, 63.8, 65.8, 68.8,
69.2, 70.9, 72.7, 74.5, 76.8, 78.1, 78.4, 78.5, 81.1, 96.1, 102.9,
119.7, 125.9, 129.6, 129.9, 138.7, 147.6, 176.1; HRMS (ESI) calcd
for [C55H91N5O15 + H]+ 1062.6584, found 1062.6586.
Acknowledgment. We thank Professor Jim Powers for a
helpful discussion on the manuscript. We are grateful to
Professors Al Merrill and Donald Doyle for the use of their
cell culture facilities. We thank Professor Thomas Barker for a
generous gift of primary lung fibroblasts. We are indebted to
Chiaolong Hsiao for his assistance on molecular docking. This
work was financially supported by Georgia Institute of Technol-
ogy (A.K.O. and Y.F.), by the Blanchard fellowship to A.K.O.,
by Georgia Cancer Coalition Distinguished Cancer Clinicians
and Scientists Program, and by NIH Grant GM085261 (Y.F.).
P.C.C. and W.G. are recipients of the GAANN predoctoral
fellowship from the Georgia Tech Center for Drug Design,
Development and Delivery. R.H. is a Beckman Undergraduate
Research Fellow.
Clarithromycin-N-benzyltriazolyldecahydroxamic Acid (24h).
Reaction of 4′-ethynylbenzylclarithromycin 21 (0.10 g, 0.120 mmol)
and 10-azidodecahydroxamic acid 17e (0.045 g, 0.197 mmol) within
2.5 h, according to the protocols of Method A described for the
synthesis of compound 24a, followed by preparative TLC (silica,
12:1:0.1 CH2Cl2/MeOH/concentrated NH4OH) gave 68 mg (53%)
1
of 24h as a brown-white solid. H NMR (CDCl3, 400 MHz) 0.82
(3H, t, J ) 7.2 Hz), 1.05-2.12 (49H, m), 2.24 (3H, s), 2.26-2.97
(7H, m), 3.01, 3.10 (6H, s), 3.19-3.80 (6H, m), 3.95 (3H, m), 4.39
(3H, m), 4.89 (1H, d, J ) 4.0 Hz), 5.04 (1H, d, J ) 8.0 Hz), 7.35
(2H, d, J ) 8.0 Hz), 7.76 (1H, s), 7.79 (2H, d, J ) 8.0 Hz); 13C
NMR (CDCl3, 100 MHz) δ 9.3, 10.8, 12.5, 16.1, 18.2, 18.8, 20.0,
21.2, 21.5, 21.7, 25.4, 26.2, 28.9, 29.0, 29.3, 29.6, 29.9, 30.2, 35.0,
37.0, 37.4, 39.3, 39.4, 45.2, 45.4, 49.6, 50.5, 50.8, 51.6, 57.8, 63.8,
65.8, 68.8, 69.2, 70.9, 72.7, 74.5, 76.8, 78.1, 78.5, 81.1, 96.1, 102.9,
119.7, 125.9, 129.6, 129.8, 138.9, 147.6, 176.1; HRMS (ESI) calcd
for [C56H93N5O15 + H]+ 1076.6740, found 1076.6667.
1
Supporting Information Available: H NMR and 13C NMR
spectral information, docked structure of 16d, and isoform selectiv-
ity results. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
HDAC Activity Assay. In vitro HDAC inhibition was assayed
using the HDAC Fluorimetric Assay/Drug Discovery Kit as
previously described.17,19 Briefly, 15 µL of HeLa nuclear extract
was mixed with 5 µL of 10× compound and 5 µL of assay buffer.
Fluorogenic substrate (25 µL) was added, and reaction was allowed
to proceed for 15 min at room temperature and then stopped by
addition of a developer containing TSA. Fluorescence was moni-
tored after 15 min at excitation and emission wavelengths of 360
and 460 nm, respectively. IC50 values were determined using logit
plots.
Cell Culture and Viability. SK-MES-1 and NCI-H69 lung cancer
cell lines and DU-145 prostate cancer cell line were obtained from
ATCC (Manassas, VA) and were maintained in the recommended
complete growth mediums. MCF-7 cell line (a generous gift from
Dr. Al Merill) was maintained in EMEM containing 10% fetal
bovine serum. Human mammary epithelial cells (HMEC) (Lonza
Biosciences) were maintained in complete MEGM per product
instructions. Human normal lung fibroblasts (a generous gift from
Dr. Barker) were maintained in EMEM containing 10% FBS. All
cell lines were maintained in a 37 °C environment containing 5%
CO2. All compounds to be tested were dissolved to a concentration
of 10 mM in DMSO and stored at -80 °C. For cell viability
experiments, cells were passaged 24 h prior to dosing. All
compounds were diluted to appropriate concentrations in DMSO
and fresh medium such that the final concentration of DMSO was
(1) Marks, P. A. Discovery and development of SAHA as an anticancer
agent. Oncogene 2007, 26, 1351–1356.
(2) Kelly, W. K.; O’Connor, O. A.; Marks, P. A. Histone deacetylase
inhibitors: from target to clinical trials. Expert. Opin. InVest. Drugs
2002, 11, 1695–1713.
(3) Miller, T. A.; Witter, D. J.; Belvedere, S. Histone deacetylase inhibitors.
J. Med. Chem. 2003, 46, 5097–5116.
(4) Rosato, R. R.; Grant, S. Histone deacetylase inhibitors in clinical
development. Expert Opin. InVest. Drugs 2004, 13, 21–38.
(5) Monneret, C. Histone deacetylase inhibitors. Eur. J. Med. Chem. 2005,
40, 1–13.
(6) Yoo, C. B.; Jones, P. A. Epigenetic therapy of cancer: past, present
and future. Nat. ReV. Drug DiscoVery 2006, 5, 37–50.
(7) Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.; Rifkind,
R. A.; Marks, P. A.; Breslow, R.; Pavletich, N. P. Structures of a
histone deacetylase homologue bound to the TSA and SAHA
inhibitors. Nature 1999, 401, 188–193.
(8) (a) 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, com-
plexed with a hydroxamic acid inhibitor. Proc. Natl. Acad. Sci. U.S.A.
2004, 101, 15064–15069. (b) Somoza, J. R.; Skene, R. J.; Katz, B. A.;
Mol, C.; Ho, J. D.; Jennings, A. J.; Luong, C.; Arvai, A.; Buggy,
J. J.; Chi, E.; Tang, J.; Sang, B.-C.; Verner, E.; Wynands, R.; Leahy,
E. M.; Dougan, D. R.; Snell, G.; Navre, M.; Knuth, M. W.; Swanson,
R. V.; McRee, D. E.; Tari, L. W. Structural snapshots of human
HDAC8 provide insights into the class I histone deacetylases. Structure
2004, 12, 1325–1334. (c) Schuetz, A.; Min, J.; Allali-Hassani, A.;

468 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Oyelere et al.
Schapira, M.; Shuen, M.; Loppnau, P.; Mazitschek, R.; Kwiatkowski,
N. P.; Lewis, T. A.; Maglathin, R. L.; McLean, T. H.; Bochkarev, A.;
Plotnikov, A. N.; Vedadi, M.; Arrowsmith, C. H. Human HDAC7
harbors a class IIa histone deacetylase-specific zinc binding motif and
cryptic deacetylase activity. J. Biol. Chem. 2008, 283, 11355–11363.
(9) http://www.clinicaltrials.gov/ct/show/NCT00106431.
(10) (a) Furumai, R.; Komatsu, Y.; Nishino, N.; Khochbin, S.; Yoshida,
M.; Horinouchi, S. Potent histone deacetylase inhibitors built from
trichostatin A and cyclic tetrapeptide antibiotics including trapoxin.
Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 87–92. (b) Yurek-George, A.;
Cecil, A. R. L.; Mo, A. H. K.; Wen, S.; Rogers, H.; Habens, F.; Maeda,
S.; Yoshida, M.; Packham, G.; Ganesan, A. The first biologically active
synthetic analogues of FK228, the depsipeptide histone deacetylase
inhibitor. J. Med. Chem. 2007, 50, 5720–5726. (c) Gomez-Paloma,
L.; Bruno, I.; Cini, E.; Khochbin, S.; Rodriquez, M.; Taddei, M.;
Terracciano, S.; Sadoul, K. Design and synthesis of cyclopeptide
analogues of the potent histone deacetylase inhibitor FR235222.
ChemMedChem 2007, 2, 1511–1519.
(11) Tsai, W. C.; Standiford, T. J. Immunomodulatory effects of macrolides
in the lung: lessons from in-vitro and in-vivo models. Curr. Pharm.
Des. 2004, 10, 3081–3093.
(12) Shinkai, M.; Park, C. S.; Rubin, B. K. Immunomodulatory effects of
macrolide antibiotics. Interstitial, inflammatory, and occupational lung
disease. Clin. Pulm. Med. 2005, 12, 341–348.
(13) Randolph, J. T.; Waid, P.; Nichols, C.; Sauer, D.; Haviv, F.; Diaz,
G.; Bammert, G.; Besecke, L. M.; Segreti, J. A.; Mohning, K. M.;
Bush, E. N.; Wegner, C. D.; Greer, J. Nonpeptide luteinizing hormone-
releasing hormone antagonists derived from erythromycin A: design,
synthesis, and biological activity of cladinose replacement analogues.
J. Med. Chem. 2004, 47, 1085–1097.
(14) (a) Freiberg, L. A. U.S. Patent 3725385, 1973. (b) Lartey, P. A.;
Nellans, H. N.; Faghih, R.; Petersen, A.; Edwards, C. M.; Freiberg,
L.; Quigley, S.; Marsh, K.; Klein, L. L.; Plattner, J. J. Synthesis of
4′′-deoxy motilides: identification of a potent and orally active
prokinetic drug candidate. J. Med. Chem. 1995, 38, 1793–1798. (c)
Stenmark, H. G.; Brazzale, A.; Ma, Z. Biomimetic synthesis of
macrolide/ketolide metabolites through a selective N-demethylation
reaction. J. Org. Chem. 2000, 65, 3875–3876.
(20) Numerous examples of Cu(I) catalyzed Huigsen cycloaddition reaction
have appeared in the literature (a comprehensive list is available at
www.scripps.edu/chem/sharpless/click.html). Cited here are two pio-
neering examples: (a) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.;
Sharpless, K. B. A stepwise huigsen cycloaddition process: copper
(I)-catalyzed regioselective “ligation” of azides and terminal alkynes.
Angew. Chem., Int. Ed. 2002, 41, 2596–2599. (b) Tornoe, C. W.;
Christensen, C.; Meldal, M. Peptidotriazoles on solid phase: [1,2,3]-
triazoles by regiospecific copper(I)-catalyzed 1,3-dipolar cycloadditions
of terminal alkynes to azides. J. Org. Chem. 2002, 67, 3057–3064.
(21) (a) Salisbury, C. M.; Cravatt, B. F. Activity-based probes for proteomic
profiling of histone deacetylase complexes. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 1171–1176. (b) Salisbury, C. M.; Cravatt, B. F.
Optimization of activity-based probes for proteomic profiling of histone
deacetylase complexes. J. Am. Chem. Soc. 2008, 130, 2184–2194.
(22) (a) 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–
2878. (b) Gurard-Levin, Z. A.; Mrksich, M. The activity of HDAC8
depends on local and distal sequences of its peptide substrates.
Biochemistry 2008, 47, 6242–6250. (c) Haggarty, S. J.; Koeller, K. M.;
Wong, J. C.; Grozinger, C. M.; Schreiber, S. L. Domain-selective
small-molecule inhibitor of histone deacetylase 6 (HDAC6)-mediated
tubulin deacetylation. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 4389–
4394.
(23) Freshney, R. Culture of Animal Cells: A Manual of Basic Technique;
Alan R. Liss, Inc.: New York, 1987; p 117.
(24) Suzuki, T.; Nagano, Y.; Kouketsu, A.; Matsuura, A.; Maruyama, S.;
Kurotaki, M.; Nakagawa, H.; Miyata, N. Novel inhibitors of human
histone deacetylases: design, synthesis, enzyme inhibition, and cancer
cell growth inhibition of SAHA-based non-hydroxamates. J. Med.
Chem. 2005, 48, 1019–1032.
(25) Rundlett, S. E.; Carmen, A. A.; Kobayashi, R.; Bavykin, S.; Turner,
B. M.; Grunstein, M. HDA1 and RPD3 are members of functionally
distinct yeast histone deacetylase complexes. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 14503–14508.
(26) 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.
(15) Awan, A.; Brennan, R. J.; Regan, A. C.; Barber, J. The conformations
of the macrolide antibiotics erythromycin A, azithromycin and
clarithromycin in aqueous solution: a 1H NMR study. J. Chem. Soc.,
Perkin Trans. 2000, 2, 1645–1652.
(16) Elliott, R. L.; Pireh, D.; Nilius, A. M.; Johnson, P. M.; Flamm, R. K.;
Chu, D. T. W.; Plattner, J. J.; Or, Y. S. Novel 3-deoxy-3-descladinosyl-
6-O-methyl erythromycin A analogues. Synthesis and in vitro activity.
Bioorg. Med. Chem. Lett. 1997, 7, 641–646.
(27) Richon, V. M.; Sandhoff, T. W.; Rifkind, R. A.; Marks, P. A. Histone
deacetylase inhibitor selectively induces p21WAF1 expression and gene-
associated histone acetylation. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,
10014–10019.
(28) Danesi, R.; Lupetti, A.; Barbara, C.; Ghelardi, E.; Chella, A.; Malizia,
T.; Senesi, S.; Angeletti, C. A.; Del Tacca, M.; Campa, M. Compara-
tive distribution of azithromycin in lung tissue of patients given oral
daily doses of 500 and 1000 mg. J. Antimicrob. Chemother. 2003,
51, 939–945.
(17) HDAC Fluorimetric Assay/Drug DiscoVery Kit. AK-500 Manual.
Fluorescent Assay System; BIOMOL International, L.P.: Plymouth
Meeting, PA, 2005.
(29) Savariar, E. N.; Aathimanikandan, S. V.; Thayumanavan, S. Supramo-
lecular assemblies from amphiphilic homopolymers: testing the scope.
J. Am. Chem. Soc. 2006, 128, 16224–16230.
(18) (a) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,
W. E.; Belew, R. K.; Olson, A. J. Automated docking using a
Lamarckian genetic algorithm and empirical binding free energy
function. J. Comput. Chem. 1998, 19, 1639-1662; http://www.
scripps.edu/pub/olson-web/doc/autodock/. (b) Wang, D.-F.; Wiest, O.;
Helquist, P.; Lan-Hargest, H.-Y.; Wiech, N. L. On the function of the
14 Å long internal cavity of histone deacetylase-like protein: implica-
tions for the design of histone deacetylase inhibitors. J. Med. Chem.
2004, 47, 3409–3417. (c) Lu, Q.; Wang, D.-S.; Chen, C.-S.; Hu, Y.-
D.; Chen, C.-S. Structure-based optimization of phenylbutyrate-derived
histone deacetylase inhibitors. J. Med. Chem. 2005, 48, 5530–5535.
(19) Chen, P. C.; Patil, V.; Guerrant, W.; Green, P.; Oyelere, A. K.
Synthesis and structure-activity relationship of histone deacetylase
(HDAC) inhibitors with triazole-linked cap group. Bioorg. Med. Chem.
2008, 16, 4839–4853.
(30) Heine, H. W.; Becker, E. B.; Lane, J. F. On cyclic intermediates in
substitution reactions. III. The alkaline hydrolysis of ε-bromocaproic
and ꢀ-bromoenanthic acids. J. Am. Chem. Soc. 1953, 75, 4514–4515.
(31) (a) Collman, J. P.; Devaraj, N. K.; Chidsey, C. E. D. “Clicking”
functionality onto electrode surfaces. Langmuir 2004, 20, 1051–1053.
(b) Shon, Y.-S.; Kelly, K. F.; Halas, N. J.; Lee, T. R. Fullerene-
terminated alkanethiolate SAMs on gold generated from unsymmetrical
disulfides. Langmuir 1999, 15, 5329–5332.
(32) Muri, D.; Bode, J. W.; Carreira, E. M. A novel, general method for
the synthesis of nitrile oxides: dehydration of O-silylated hydroxamic
acids. Org. Lett. 2000, 2, 539–541.
JM801128G