Total synthesis and full histone deacetylase inhibitory profiling of azumamides A-E as Well as β2- epi -Azumamide e and β3- epi -Azumamide e

Article abstract of DOI:10.1021/jm4008449

Cyclic tetrapeptide and depsipeptide natural products have proven useful as biological probes and drug candidates due to their potent activities as histone deacetylase (HDAC) inhibitors. Here, we present the syntheses of a class of cyclic tetrapeptide HDAC inhibitors, the azumamides, by a concise route in which the key step in preparation of the noncanonical disubstituted β-amino acid building block was an Ellman-type Mannich reaction. By tweaking the reaction conditions during this transformation, we gained access to the natural products as well as two epimeric homologues. Thus, the first total syntheses of azumamides B-D corroborated the originally assigned structures, and the synthetic efforts enabled the first full profiling of HDAC inhibitory properties of the entire selection of azumamides A-E. This revealed unexpected differences in the relative potencies within the class and showed that azumamides C and E are both potent inhibitors of HDAC10 and HDAC11.

Full text of DOI:10.1021/jm4008449

Subscriber access provided by FORDHAM UNIVERSITY
Article
Total Synthesis and Full Histone Deacetylase Inhibitory Profiling of
Azumamide A–E as well as #2- Epi-Azumamide E and #3-Epi-Azumamide E
Jesper S. Villadsen, Helle Marie Stephansen, Alex R. Maolanon, Pernille Harris, and Christian Adam Olsen
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm4008449 • Publication Date (Web): 18 Jul 2013
Downloaded from http://pubs.acs.org on July 22, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Total Synthesis and Full Histone Deacetylase Inhibitory
Profiling of Azumamide A–E as well as β2ꢀ Epiꢀ
Azumamide E and β3ꢀEpiꢀAzumamide E
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Jesper S. Villadsen, Helle M. Stephansen, Alex R. Maolanon, Pernille Harris, and Christian A. Olsen*
Department of Chemistry, Technical University of Denmark, Kemitorvet 207, Kongens Lyngby, DKꢀ
2800, Denmark.
*cao@kemi.dtu.dk
RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required
according to the journal that you are submitting your paper to)
TITLE RUNNING HEAD: Full HDAC Inhibitory Profiling of Azumamides.
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 33
1
2
3
4
5
6
7
8
9
ABSTRACT: Cyclic tetrapeptide and depsipeptide natural products have proven useful as biological
probes and drug candidates due to their potent activities as histone deacetylase (HDAC) inhibitors.
Here, we present the syntheses of a class of cyclic tetrapeptide HDAC inhibitors, the azumamides, by a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
concise route in which the key step in the preparation of the nonꢀcanonical disubstituted βꢀamino acid
building block was an Ellmanꢀtype Mannich reaction. By tweaking the reaction conditions during this
transformation, we gained access to the natural products as well as two epimeric homologs. Thus, the
first total syntheses of azumamides B–D corroborated the originally assigned structures, and the
synthetic efforts enabled the first full profiling of the HDAC inhibitory properties of the entire selection
of azumamides A–E. This revealed unexpected differences in the relative potencies within the class and
showed that azumamides C and E are both potent inhibitors of HDAC10 and HDAC11.
KEYWORDS: Nonꢀribosomal peptides, natural product synthesis, Mannich reaction, enzyme inhibitors,
histone deacetylases, epigenetics.
ACS Paragon Plus Environment
2

Page 3 of 33
Journal of Medicinal Chemistry
INTRODUCTION
Macrocyclic peptides have played important roles in the field of epigenetics due their potent activities
1
2
3
4
5
6
7
8
9
as inhibitors of HDAC enzymes. One of the two HDAC targeting drugs (11 and 3) that are approved by
the FDA for clinical treatment of cutaneous Tꢀcell lymphoma is the macrocyclic natural product
romidepsin (3).2 Furthermore, a cyclic tetrapeptide, trapoxin,3 played an instrumental role in the first
isolation of a mammalian HDAC enzyme.4, 5 Thus, this class of inhibitors holds promise as tool
compounds as well as potential drug candidates targeting HDACs.6ꢀ9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Though clearly bearing an overall resemblence to the classical cyclic tetrapeptide HDAC inhibitors
[including for example apicidin (4)],10 the azumamides (5–9) are structurally unique in that their
extended Zn2+ꢀcoordinating amino acid (shown in yellow in Figure 1) is a disubstituted ꢀamino acid.11
β
Furthermore, we found the azumamides interesting due to the relatively strong potencies reported for
azumamide E against classꢀI HDACs12 in spite of its weak Zn2+ꢀcoordinating carboxylic acid
functionality.13 Previously, azumamide A14, 15 and azumamide E12, 14ꢀ16 have been prepared by multistep
chemical syntheses, but only azumamide E was tested against recombinant HDAC isoforms 1–9.12
Furthermore, in vitro profiling using recombinant HDACs has witnessed important new developments
since the publication of those results.17, 18 We therefore found it relevant to explore the properties of
these macrocycles in more detail by preparing the complete selection of natural products (5–9), and
profiling their activities against the full panel of recombinant human Zn2+ꢀdependent HDAC enzymes,
HDAC1–11. As total syntheses of azumamides B–D had not been reported previously, this work would
also allow unequivocal validation of the proposed structures.11
For the syntheses of the azumamides, we envisaged two significant challenges. First, the efficient
stereoselective synthesis of the disubstituted
βꢀamino acid, and secondly, the macrocyclization step,
which is known to be difficult for small cyclic peptides in general,19 and furthermore proved challenging
in previously reported syntheses of azumamide analogs.12
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 33
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Structures of archetypical HDAC inhibitors (1–4) and target azumamides 5–9.
RESULTS AND DISCUSSION
Building Block Synthesis. For our synthesis of the
βꢀamino acid building block, we chose a
diastereoselective Ellmanꢀtype Mannich reaction to set the stereochemistry, as also previously reported
by Ganesan and coworkers.15 However, to avoid having this important transformation at a late stage in
our synthetic route, we decided to optimize this reaction between a propionate ester and a simple imine
as shown in eq. 1.
This should give an intermediate with the correct stereochemistry (2S,3R), which could be readily
elaborated to give the desired βꢀamino acid by robust organic synthetic transformations (vide infra).
Mannich reactions between esterꢀenolates and chiral sulfinylimines have been studied extensively,20, 21
and using previously reported conditions as our starting point we conducted an optimization study as
outlined in Table 1. The tertꢀbutylester showed superior selectivity (entry 5) compared to the less bulky
methyl, ethyl, allyl, and PMB esters (entries 1–4), and furthermore, the methylester did not proceed to
completion in our hands. Somewhat surprisingly, however, the major diastereoisomer in entry 5 proved
to have (2S,3S) configuration as determined by Xꢀray crystallography upon desilylation (Figure 2).
ACS Paragon Plus Environment
4

Page 5 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
Table 1. Optimization of the Stereochemical Outcome of the Mannich Reaction Shown in
Equation 1
6
7
entry auxiliary*
R1
R2
Additive
enolatea
d.r.b
major isomer
NDc
8
9
1
R
R
R
R
R
R
R
R
R
S
Me
Et
OSi(iPr)3
OSi(iPr)3
OSi(iPr)3
OSi(iPr)3
OSi(iPr)3
OSi(iPr)3
OBn
TiCl(OiPr)3
TiCl(OiPr)3
TiCl(OiPr)3
TiCl(OiPr)3
TiCl(OiPr)3
HMPA
E
E
E
E
E
E
E
E
E
E
Z
47:39:10:4
49:29:11:11
46:34:10:10
46:33:11:10
60:26:8:6
71:15:14:0
70:18:12:0
77:13:10:0
75:21:4:0
77:18:5:0
64:25:8:2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ND
3
Allyl
PMB
tBu
tBu
tBu
tBu
tBu
tBu
tBu
ND
4
ND
5
(2S,3S)d
(2S,3S)
ND
6
7
TiCl(OiPr)3
TiCl(OiPr)3
TiCl(OiPr)3
HMPA
8
OPMB
(2S,3S)
(2S,3S)d
(2R,3R)e
9
OSi(Et)3
OSi(iPr)3
OSi(iPr)3
10
11
S
HMPAf
(2S,3R)g
t
aMajor configuration of the enolate as determined by NMR and by trapping with BuMe2SiCl.
bDiastereomeric ratio determined by H NMR. ND = not determined. Determined by Xꢀray
1
c
d
e
crystallography on its desilylated homologue. Determined spectroscopically by comparison with its
f
enantiomer from entries 5 and 9. HMPA (5.4 equiv) was added prior to the substrate to obtain the (Z)ꢀ
enolate (>80%). gDetermined by comparison of spectroscopic data of the fully elaborated Boc protected
β
ꢀamino acid with previously reported data.12
Figure 2. Xꢀray crystal structure of the (2S,3S) precursor obtained by desilylation of the major product
in entry 5.
This indicates that the pathway leading to our major isomer did not proceed through the sixꢀ
membered Zimmerman–Traxlerꢀtype transition state,22 which has been proposed to be responsible for
the diastereoselectivity with similar substrates.20, 23 By using HMPA as an additive instead of a Lewis
acid, this reaction has previously been shown to proceed through a different transition state,20 and
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 33
indeed we saw the same product distribution when using HMPA and TiCl(OiPr)3 as additives with our
substrates (entries 5 and 6). This indicates that the sixꢀmembered transition state, where coordination of
titanium is crucial, is highly unlikely to play a significant role in the formation of our major isomer. This
is not in agreement with the diastereoselectivities observed with the substrates reported by Ganesan and
coworkers.15 Thus, to address whether the steric bulk of the triisopropylsilylether was responsible for
interrupting the sixꢀmembered transition state, we performed the reaction with different means of
protecting the alcohol (entries 7–9). No significant effect was observed, however, indicating instead that
the steric bulk of the tertꢀbutylester caused the predominance of a different transition state when using
our substrates. This is also in agreement with the original study by Ellman and Tang where the level of
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
selectivity decreased for 2,3ꢀdisubstituted βꢀamino acids when increasing the bulk of the ester from
methyl to tertꢀbutyl.20
Because we were interested in taking advantage of solidꢀphase synthesis methods to prepare the liniar
tetrapeptide azumamide precursors with a minimum of chromatographic purification steps, we were
keen on keeping the acid labile tertꢀbutylester protecting group, which would allow easy protecting
group manipulation to give an Fmoc protected
βꢀamino acid building block. Hence, instead of
substituting this protecting group, we decided to optimize the Mannich reaction conditions to deliver the
desired stereochemistry. First, we changed the stereochemistry of the sulfinylimine to the Rꢀenantiomer,
which expectedly furnished the enantiomer of entries 5–9 (2R,3R) as the major isomer (entry 10). We
then hypothesized that the configuration of the 2ꢀposition would be sensitive to the E/Zꢀconfiguration of
the enolate. Using Ireland’s conditions forming the enolate in the presence of HMPA,24, 25 we achieved
>80% of Zꢀisomer, which gratifyingly afforded the (2S,3R) product as major isomer (entry 11). Using
the developed conditions, we prepared compound 12, which was further elaborated to give Fmoc
protected
βꢀamino acid 16 in 14% overall yield with just four column chromatographic purification
steps from compound 10 (Scheme 1).
ACS Paragon Plus Environment
6

Page 7 of 33
Journal of Medicinal Chemistry
Scheme 1. Synthesis of β
ꢀamino acid building block 16.a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aReagents and Conditions: (a) HMPA (6.4 equiv), LDA (2.6 equiv), 11 (2.5 equiv), THF, –78 ºC, 30
min; then 10, –78 ºC, 30 min. (b) AcOH (1.0 equiv), Bu4NF (2.0 equiv), THF, 0 ºC → rt, 1.5 h. (c)
NaHCO3 (1.5 equiv), DessꢀMartin periodinane (1.4 equiv), dry CH2Cl2, 0 ºC → rt, 1.5 h. (d) KHMDS
(1.9 equiv), Ph3PBr(CH2)3COOEt, (2.0 equiv), THF, –78 ºC → rt, 18 h. (e) TFA–CH2Cl2 (1:1, 10 mL,
80 equiv), 0 ºC → rt, 3 h. (f) HCl (4.0 M in dioxane, 3.0 equiv), dioxane, 3 h. (g) Na2CO3 (4.0 equiv),
FmocꢀOSuc (1.2 equiv), dioxane–H2O, 0 ºC → rt, 2 h.
The Boc protected homolog of 16 was also prepared to confirm the (2S,3R) stereochemistry by
comparison of spectroscopic data (optical rotation and NMR) with those previously reported
[Supplementary Figure S1].12 Furthermore, the β2ꢀ and β3ꢀepimeric building blocks were prepared by
elaboration of the major isomers from entries 10 and 5, respectively (see the Supporting Information for
details). Although the achieved diastereomeric ratios were not particularly impressive, this strategy very
nicely provided the correct stereochemistry along with two novel βꢀamino acids enabling investigation
of the biochemical effect of the stereochemical configuration at these two chiral centers.
Cyclic Peptide Synthesis. Because three different points of cyclization had been reported for
azumamide E and since these were all performed with different coupling reagents,12, 14, 15 we performed
cyclization experiments using a simplified model peptide to address the issue. Not too surprisingly, this
showed that macrolactamizations with the most sterically hindered amino acids at the Cꢀterminal were
particularly poor, resulting in significant amounts of Nꢀterminal guanidinylation, incomplete
cyclization, epimerization, and/or dimerization (Supplementary Table S1). Thus, we prepared the linear
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 33
tetrapeptides 17, 19 and 21 on solid support by standard Fmoc solidꢀphase synthesis using βꢀamino acid
1
2
3
16 and commercially available FmocꢀDꢀamino acids.
4
5
6
7
8
Scheme 2. Synthesis of azumamides A–E by solidꢀphase synthesis followed by headꢀtoꢀtail
macrolactamization in solution.a
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aReagents and Conditions: (a) TFA–CH2Cl2 (1:1), 2 × 30 min; (b) HATU, iPr2NEt (8.0 equiv), DMF
(0.4–0.5 mM peptide concentration), 17–21 h, then HATU (0.5 equiv), 1–3 h [A = 11% of 18 after
preparative HPLC, B = 25% of 20 after column chromatography, and C = 19% of 22 after column
i
chromatography]; (c) LiOH, THF–H2O (1:1); (d) DIC (11 equiv), HOBt (3.0 equiv), Pr2NEt (4.0
i
equiv), NH3–dioxane (25 equiv), DMF–CH2Cl2 (2:1), 5 d, 67%; (e) HATU (2.0 equiv), Pr2NEt (5.5
equiv), NH3–dioxane (25 equiv), DMF, 5.5 h, 40% (for steps c and e); (f) DIC (6.0 equiv), HOBt (3.0
i
equiv), Pr2NEt (4.0 equiv), NH3–dioxane and NH3–MeOH (30 equiv), CH2Cl2–DMF (2:1), 13 d, 11%
(for steps c and f).
In panel A the cyclization was then performed at the βꢀamino acid position and in panel B at the
alanine residue, whereas the preparation of azumamide D (8) was achieved by cyclization between the
two least sterically challenging alanine residues (Scheme 2, panel C). After cleavage from the 2ꢀ
chlorotrityl polystyreneꢀresin with dilute TFA, the linear tetramers were ringꢀclosed using HATU under
ACS Paragon Plus Environment
8

Page 9 of 33
Journal of Medicinal Chemistry
dilute conditions (0.4–0.5 mM),26ꢀ29 and furthermore slow addition of the linear peptide by syringe
pump to a solution of Hünig’s base and HATU, as described by Ganesan and coworkers, was tested.15
Judging from LCꢀMS analyses of the reaction mixtures, we could not observe any significant
differences between the cyclization yields obtained with the different methods. Although all the
couplings proceeded satisfactorily, with full conversion of linear peptides and minor amounts of the
corresponding dimers as the only observed byproducts, the resulting overall isolated yields were
relatively low (∼10%). We ascribe this to difficulties during the purification of the macrocyclic products
by preparative reversedꢀphase HPLC caused by poor water solubility, as we were able to recover more
material by purifying the macrocycles by column chromatography. Unfortunately, however, this did not
provide the final compounds in satisfyingly high purity for the bioassays, and thus the final compounds
were all subjected to preparative reversedꢀphase HPLC purification although this resulted in a loss of
material. Carbodiimideꢀmediated amidation of the side chain was attempted for conversion of 7 to 6 and
23 to 8, but the reaction was slow and gave varying yields (6 vs. 8, Scheme 2). Instead, HATUꢀmediated
coupling was attempted for conversion of 9 to 5, and this proved was faster and gave an acceptable yield
(5). The spectral data of all the natural products 5–9 were in excellent agreement with those originally
reported for the azumamides isolated from natural sources,5 thus corroborating the original structural
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
assignment (Figures S2–S6). Finally, the two epimeric βꢀamino acid building blocks were applied in
analogous syntheses of β3ꢀepiꢀazumamide E (26) and β2ꢀepiꢀazumamide E (29) as shown in Scheme 3.
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 33
Scheme 3. Synthesis of the two epimers 26 and 29 of azumamide E by solidꢀphase synthesis followed
by headꢀtoꢀtail macrolactamization in solution.a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aReagents and Conditions: (a) TFA–CH2Cl2 (1:1), 2 × 30 min; (b) HATU, iPr2NEt (8.0 equiv), DMF
(0.3–0.7 mM peptide concentration), 17–21 h; (c) LiOH, THF–H2O (1:1).
HDAC Screening. As an initial test of the HDAC inhibitory potency of all seven compounds, we first
screened against the full panel of recombinant human HDACs using two compound concentrations (50
ꢀ
M and 5 ꢀ
M). Protocols for HDAC1–9 were adapted from Bradner et al.,18 using the fluorogenic Acꢀ
LeuGlyLys(Ac)ꢀAMC substrate for HDAC1–3 and 6, while using the AcꢀLeuGlyLys(tfa)ꢀAMC
substrate for HDAC4, 5, 7, 8 and 9. For HDAC10 we used the tetrapeptide AcꢀArgThrLys(Ac)Lys(Ac)ꢀ
AMC,30 which was recently reported to perform well with this enzyme.31 Finally, for HDAC11, we used
also used AcꢀLeuGlyLys(Ac)ꢀAMC substrate.32
The siteꢀspecifically epimerized compounds exhibited no activity as previously reported for an analog
having both stereocenters inverted.7 It was not surprising that 26 was inactive, but it is noteworthy the
subtle change of inverting the stereochemistry of a single methyl group in compound 29 had such a
detrimental effect across the entire selection of enzymes (Figure 3). Furthermore, none of the
compounds 5–9 were able to inhibit the classꢀIIa HDAC activity against a trifluoroacetylated substrate
(Figure 3).
ACS Paragon Plus Environment
10

Page 11 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Figure 3. Single dose HDAC inhibitory screening. The assays were performed at 50
ꢀM (shown) and 5
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢀ
M (not shown) peptide concentrations. We chose <50% inhibition at 50 M as our cutꢀoff to sort away
ꢀ
inactive compounds before performing full dose–response experiments. All compound–enzyme
combinations that were discarded at this stage were tested in at least two individual assays performed in
duplicate. The error bars represent the standard deviation. *Fusion protein of GSTꢀtagged HDAC3 with
the deacetylase activation domain (DAD) of nuclear receptor coꢀrepressor (NCoR1).
Figure 4. MichaelisꢀMenten plot for HDAC10.
Inhibitor Ki Values. Next, we performed dose–response experiments for all compound–HDAC
combinations that gave above 50% inhibition in the initial assay (Supplementary Figure S7 and Table
S2). The obtained IC50 values were converted to Ki values using the ChengꢀPrusoff equation [Ki =
IC50/(1+[S]/Km)] by asuming a standard fastꢀon–fastꢀoff mechanism of inhibition. Reported Km values
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 33
were applied for the calculations except HDAC10, where we determined the Km for the used substrate to
be 1.5±0.2 M (Figure 4).
1
2
3
4
ꢀ
5
6
7
8
Low potencies were recorded against HDACs 6 and 8, which is in accordance with previous data for
azumamide E (Table 2),6 however, compounds 7 and 9 were both potent inhibitors of HDACs 10 and
11. Although they are classified together (classꢀIIb), HDACs 6 and 10 clearly interact very differently
with these inhibitors.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Generally, we found the compounds with a carboxylic acid Zn2+ꢀbinding group (7 and 9) to be more
potent than the carboxamides (5, 6, and 8), which is in contrast to the originally reported HDAC
inhibition data obtained for the natural products against an HDAC containing cell extract.5 However, the
data presented herein agrees with subsequent work from Ganesan and coworkers on azumamide A (5)
and azumamide E (9).9 We thus show that this applies to all the azumamides, which also confirms that a
carboxylate Zn2+ꢀbinding group renders HDAC inhibitors significantly more potent than a
corresponding carboxamide as would be expected from literature precendents.19, 26, 27 Furthermore,
compound 7 was more potent than 9 against HDACs 1–3, 6, 10, and 11, which is also in contrast to the
original evaluation that found azuE (9) more potent than azuC (7) against crude enzymes from K562
cell extract.5 The tyrosine containing compound (7) exhibited ∼2ꢀfold higher potency against HDACs 1,
3, 6, 10, and 11, whereas the phenylalanine containing azumamide E (9) was only more potent against
HDAC8, albeit at micromolar Ki values.
Finally, the inhibition of HDAC11 by azumamides C (7) and E (9) is, to the best of our knowledge,
the first demonstration of potent cyclic peptide inhibitors of this isozyme.20 Notably, these binding
affinities were achieved without the presence of a strong Zn2+ chelator, such as the hydroxamic acid.
ACS Paragon Plus Environment
12

Page 13 of 33
Journal of Medicinal Chemistry
Table 2. Potencies of the Azumamides Against the Zn2+-Dependent Histone Deacetylases Given as
Ki Values (nM)a
1
2
3
4
5
6
7
8
9
classꢀI
classꢀIIa
classꢀIIb
classꢀIV
HDAC11
>5,000
>5,000
35±3
compound
5 (azuA)
6 (azuB)
7 (azuC)
8 (azuD)
9 (azuE)
HDAC1 HDAC2 HDAC3b
HDAC8 HDAC4 HDAC5 HDAC7 HDAC9 HDAC6 HDAC10
>5,000
5,000
32±1
>5,000
3,000
3,200
3,000
14±1
3,700
25±5
>5,000
IA
52%c
IA
IAd
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
40±20
>5,000
50±30
>5,000
IA
IA
2,000
IA
10±4
IA
>5,000
67±7
IA
>5,000
60±16
IA
4,400
IA
>5,000
20±12
26
azuE)
(
β
3ꢀepiꢀ
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
29
azuE)
(
β
2ꢀepiꢀ
IA
1 (SAHA)
8±1.5
0.002
7±1.5
0.038
12±4
0.15
700±20
0.15
IA
IA
IA
IA
22±9
10
NTe
NT
13±2
NT
3 (FKꢀ228)f
20.5
550
1250
1100
aIC50 values were determined from at least two individual dose–response experiments performed in
duplicate (Fig. S7), and the Ki values were calculated using the ChengꢀPrusoff equation. bFusion protein
c
of GSTꢀtagged HDAC3 with the deacetylase activation domain NCoR1. Percent inhibition at 50
ꢀM
d
e
inhibitor concentration. IA = inactive (<50% inhibition at 50
ꢀM [inhibitor], Fig. 3). NT = not tested.
fData from Bradner et al.18
CONCLUSIONS
In summary, we report total syntheses of all five azumamides, including for the first time azumamides
B–D, which corroborate the originally proposed structures. Our synthetic route furthermore enabled
preparation of siteꢀspecifically edited analogs for exploration of structureꢀactivity relationships
(SAR).33ꢀ35 The HDAC profiling results show that the
βꢀamino acid residue, present in all the
azumamides, is sensitive to even slight modifications. In addition, the original HDAC testing using cell
extract indicated that azumamide E was the most potent of the series, but the comprehensive profiling
presented herein, show that azumamide C is in fact ∼2ꢀfold more potent than azumamide E against the
majority of the isozymes.
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 33
By taking advantage of the modular methodologies described in this article, and building on the
gained SAR information, we are currently investigating collections of azumamide analogs in search of
more potent and selective ligands based on this promising scaffold.
1
2
3
4
5
6
7
8
9
EXPERIMENTAL SECTION
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General. All chemicals and solvents were analytical grade and used without further purification.
Vacuum liquid chromatography (VLC) was performed on silica gel 60 (particle size 0.015−0.040 ꢀm).
UPLC−MS analyses were performed on a Phenomenex Kinetex column (1.7 ꢁm, 50 × 2.10 mm) using a
Waters Acquity ultra highꢀperformance liquid chromatography system. A gradient with eluent I (0.1%
HCOOH in water) and eluent II (0.1% HCOOH in acetonitrile) rising linearly from 0% to 95% of II
during t = 0.00–2.50 min was applied at a flow rate of 1 mL/min (gradient A) or during t = 0.00–5.20
min (gradient B). Analytical HPLC was performed on a Phenomenex Luna column [150 mm × 4.6 mm,
C18 (3 ꢁm)] using an Agilent 1100 LC system equipped with a UV detector. A gradient, C, with eluent
III (95:5:0.1, water−MeCN−TFA) and eluent IV (0.1% TFA in acetonitrile) rising linearly from 0% to
95% of IV during t = 2−20 min was applied at a flow rate of 1 mL/min. Preparative reversedꢀphase
HPLC was performed on a Phenomenex Luna column [250 mm × 20 mm, C18 (5 ꢁm, 100 Å)] using an
Agilent 1260 LC system equipped with a diode array UV detector and an evaporative light scattering
detector (ELSD). A gradient C with eluent III and eluent IV) rising linearly from 0% to 95% of IV
during t = 5−45 min was applied at a flow rate of 20 mL/min. All tested compounds were purified to
homogeneity and shown by both analytical HPLC (gradient C) and LCꢀMS (gradient A) to be of more
than 95% purity. 1D and 2D NMR spectra were recorded on a Varian Mercury 300 instrument or a
Varian INOVA 500 MHz instrument. All spectra were recorded at 298 K. Correlation spectroscopy
(COSY) spectra were recorded with a relaxation delay of 1.5 sec before each scan, a spectral width of
6k × 6k, collecting 8 FIDs and 1k × 512 data points. Heteronuclear single quantum coherence (HSQC)
spectra were recorded with a relaxation delay of 1.5 sec before each scan, a spectral width of 6k × 25k,
ACS Paragon Plus Environment
14

Page 15 of 33
Journal of Medicinal Chemistry
collecting 16 FIDs and 1k × 128 datapoints. Heteronuclear 2ꢀbond correlation (H2BC) spectra were
recorded with a relaxation delay of 1.5 sec before each scan, a spectral width of 4k × 35k, collecting 16
FIDs at 295 K and 1k × 256 datapoints. Heteronuclear multipleꢀbond correlation (HMBC) spectra were
recorded with a relaxation delay of 1.5 sec before each scan, a spectral width of 6k × 35k, collecting 32
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
FIDs and 1k × 256 datapoints. Chemical shifts are reported in ppm relative to deuterated solvent peaks
as internal standards (δH, DMSOꢀd6 2.50 ppm; δC, DMSOꢀd6 39.52 ppm, δH, CD3OH 3.30 ppm, δH,
CDCl3 7.26 ppm; δC, CDCl3 77.16 ppm). Coupling constants (J) are given in hertz (Hz). Multiplicities
of 1H NMR signals are reported as follows: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet.
General procedure for the Mannich reactions. A solution of LDA (2.1 equiv) was added dropwise
to a solution of the ester (2.0 equiv) in dry THF at –78 °C. After stirring for 30 min Ti(OꢀiPr)3Cl (4.2
equiv) in dry THF was added dropwise. The orange solution was stirred for 30 min and the imine (1.0
equiv) in dry THF was added dropwise. The mixture was stirred for 3 h or until TLC showed full
conversion of the imine. The mixture was quenched with sat. aq. NH4Cl and allowed to reach room
temperature. Water was added and the mixture decanted into a separatory funnel. The remaining Ti
precipitate was added EtOAc–water (1:1) and stirred vigorously for 5 min before being added to the
separatory funnel. The aqueous phase was extracted with EtOAc and the combined organic phases were
washed again with water, dried (MgSO4), filtered, and concentrated in vacuo.
Azumamide
A
[(Z)-6-((2R,5R,8R,11R,12S)-8-Benzyl-2-isopropyl-5,12-dimethyl-3,6,9,13-
tetraoxo-1,4,7,10-tetraazacyclotridecan-11-yl)hex-4-enamide] (5). LiOH (89 mg, 3.72 mmol, 85
equiv) in water (4.0 mL) was added to a stirred solution of the impure cyclic peptide 20 (24.2 mg,
approx. 0.045 mmol) in THF (4 mL). After 2.5 hours of stirring the organic solvent was removed in
vacuo. The aqueous phase was acidified with 1 M HCl to pH 2 and extracted with EtOAc (4 × 30 mL)
and CH2Cl2 (40 mL). The organic phases were dried (Na2SO4), filtered, and concentrated to afford crude
azumamide E (102 mg), which was used without further purification. HPLC gradient A, tR = 1.47 min.
To a solution of the above crude azumamide E (102 mg, 0.045 mmol) in DMF (3.0 mL) was added
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 33
HATU (34 mg, 0.09 mmol, 2 equiv), iPr2NEt (43 ꢂL, 0.25 mmol, 5.5 equiv) and after 5 min. NH3–
dioxane (0.9 mL, 0.45 mmol, 10 equiv). After 1 h NH3–dioxane (0.45 mL, 0.23 mmol, 5 equiv) was
added. UPLCꢀMS analysis showed on 50% conversion after 3 hours and HATU (34 mg, 0.09 mmol, 2
equiv) and NH3–dioxane (0.45 mL, 0.23 mmol, 5 equiv) was added. After additionally 1 hour DMF (1.0
mL) followed by HATU (17 mg, 0.045 mmol, 1 equiv) and NH3–dioxane (0.45 mL, 0.23 mmol, 5
equiv) was added and stirring continued for 1 hour before concentration in vacuo. The residue was
dissolved in MeCN–H2O and purified by preperative HPLC to give azumamide A (5) (4.8 mg, 12%
overall). [α]D: +56° (c = 0.2, MeOH); previously reported11 [α]D: +33° (c = 0.1, MeOH). 1H NMR (500
MHz, CD3OH) δ 8.18 (d, J = 8.6 Hz, 1H), 8.16 (d, J = 9.0 Hz, 1H), 7.72 (d, J = 8.2 Hz, 1H), 7.59 (m,
2H), 7.20 (m, 5H), 6.74 (bs, 1H), 5.48 (m, 1H), 5.37 (m, 1H), 4.33 (dt, J = 9.0, 7.0 Hz, 1H), 4.24 (m,
2H), 3.81 (dd, J = 10.4Hz and 8.4 Hz, 1H), 3.10 (m, 2H), 2.72 (m, 1H), 2.57 (dt, J = 14.1 Hz and 6.9
Hz, 1H), 2.41 (m, 2H), 2.27 (m, 4H), 1.30 (d, J = 7.5 Hz, 3H), 1.23 (d, J = 7.3 Hz, 3H), 0.96 (d, J = 6.5
Hz, 3H), 0.94 (d, J = 6.7 Hz, 3H). HRMS (ESIꢀTOF): m/z calculated for C27H38N4O7H+: 514.3029;
found: 514.3032 [M+H]+. HPLC gradient C, tR = 11.62 min (>95%).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Azumamide
B
[(Z)-6-((2R,5R,8R,11R,12S)-8-(4-Hydroxybenzyl)-2-isopropyl-5,12-dimethyl-
3,6,9,13-tetraoxo-1,4,7,10-tetraazacyclotridecan-11-yl)hex-4-enamide] (6). An aqueous solution of
LiOH (0.5 M, 55 ꢂL, 2.0 mmol, 2.5 equiv) was added to the cyclic peptide 18 (6.1 mg) in THF–H2O
((1:1), 2 mL) at 0 oC. After 30 min the iceꢀbath was removed. Additional portions of LiOH solution (55
ꢂL, 2.0 mmol, 2.5 equiv) were added after 2, 4, and 6 hours, and stirring was continued for additional
19 hours to ensure full conversion. Then water (0.5 mL) was added and the organic solvent removed in
vacuo. The aqueous phase was acidified with 1 M HCl and extracted with EtOAc (5 × 20 mL). The
organic phase was dried (Na2SO4), filtered and concentrated in vacuo to afford the crude azumamide C
(7) (20 mg), which was used without further purification. To a solution of crude azumamide C (5.8 mg,
10.9 ꢁmol) in DMF (2 mL) was added HOBt (4.4 mg, 33 ꢂmol, 3 equiv), DIC (5.1 ꢂL, 34 ꢂmol, 3
equiv) and iPr2NEt (7.6 ꢂL, 44 ꢂmol, 4 equiv). After 10 minutes NH3–dioxane (0.5 M, 0.11 mL, 55
ꢂmol, 5 equiv) was added. After 1.5 hour DIC (5 ꢂL, 34 ꢂmol, 3 equiv) followed by NH3–dioxane (0.5
ACS Paragon Plus Environment
16

Page 17 of 33
Journal of Medicinal Chemistry
M, 0.11 mL, 55 ꢂmol, 5 equiv) were added. After stirring for 16 hours, additional DIC (2 equiv) and
NH3–dioxane (5 equiv) were added and this procedure was repeated once more after 18 hours. Finally,
CH2Cl2 (1 mL) followed by DIC (3 equiv) and NH3–dioxane (10 equiv) were added and after 2 days of
stirring at room temperature, the reaction mixture was concentrated, dissolved in MeCN–H2O (2:1), and
purified by preparative HPLC to give azumamide B (6) (3.6 mg, 62%, two steps) as a white solid. [α]D:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
+65° (c = 0.15, MeOH); previously reported11 [α]D: +45° (c = 0.1, MeOH). H NMR (500 MHz,
CD3OH) δ 8.13 (d, J = 7.9 Hz, 1H), 8.00 (d, J = 8.9 Hz, 1H), 7.85 (d, J = 7.2 Hz, 1H), 7.61 (d, J = 8.2
Hz, 1H), 7.01 (d, J = 8.4 Hz, 3H), 6.67 (d, J = 8.4 Hz, 3H), 5.49 (m, 1H), 5.37 (dd, J = 18.0 Hz and 7.3
Hz, 2H), 4.29 (pentet, J = 7.2 Hz, 1H), 4.15 (m, 1H), 4.05 (m, 1H), 3.60 (m, 1H), 3.13 (dd, J = 13.7,
10.1 Hz, 1H), 3.00 (dd, J = 13.8, 6.3 Hz, 1H), 2.70 (m, 2H), 2.36 (ddd, J = 22.3, 21.5, 7.1 Hz, 11H),
1.29 (d, J = 7.2 Hz, 3H), 1.27 (d, J = 7.4 Hz, 3H), 0.95 (d, J = 5.7 Hz, 3H), 0.93 (d, J = 6.0 Hz, 3H).
HRMS (ESIꢀTOF): m/z calculated for C27H39N5O6H+: 530.2978; found: 530.2973 [M+H]+. HPLC
gradient C, tR = 10.31 min (>95%).
Azumamide
C
[(Z)-6-((2R,5R,8R,11R,12S)-8-(4-Hydroxybenzyl)-2-isopropyl-5,12-dimethyl-
3,6,9,13-tetraoxo-1,4,7,10-tetraazacyclotridecan-11-yl)hex-4-enoic acid] (7). LiOH (49 mg, 2.0
mmol, 35 equiv) in water (5.0 mL) was added to the crude cyclic peptide 18 (61 mg) in THF (5.0 mL).
The solution was stirred for 16 hours and concentrated in vacuo. The resulting residue was dissolved in
THF–H2O (1:1), 10 mL) by adding a few drops of TFA and then purification by preparative HPLC
afforded azumamide C (7) (2.2 mg, 9% overall) as a white solid. [α]D: +49° (c = 0.14, MeOH);
previously reported11 [α]D: +21° (c = 0.1, MeOH). 1H NMR (500 MHz, CD3OH) δ 8.08 (d, J = 7.7 Hz,
1H), 7.99 (d, J = 8.4 Hz, 1H), 7.84 (s, 1H), 7.62 (d, J = 8.1 Hz, 1H), 7.01 (d, J = 8.4 Hz, 2H), 6.67 (d, J
= 8.4 Hz, 2H), 5.48 (t, J = 8.8 Hz, 1H), 5.38 (dt, J = 10.7, 7.0 Hz, 1H), 4.29 (pentet, J = 7.3 Hz 1H),
4.16 (m, 1H), 4.01 (m, 1H), 3.58 (m, 1H), 3.15 (dd, J= 13.7, 10.2 Hz 1H), 3.00 (dd, J = 13.7 Hz and 6.0
Hz, 1H), 2.72 (m, 1H), 2.67 (m, 1H) 2.39 (m, 5H), 1.29 (d, J = 7.2 Hz, 3H), 1.27 (d, J = 7.3 Hz, 3H),
0.95 (d, J = 6.0 Hz, 3H), 0.93 (d, J = 6.0 Hz, 3H). HRMS (ESIꢀTOF): m/z calculated for C27H38N4O7H+:
531.2819; found: 531.2815 [M+H]+. HPLC gradient C, tR = 11.04 min (>95%).
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 33
Azumamide D [(Z)-6-((2R,5R,8R,11R,12S)-8-Benzyl-2,5,12-trimethyl-3,6,9,13-tetraoxo-1,4,7,10-
tetraazacyclotridecan-11-yl)hex-4-enamide] (8). LiOH (53 mg, 2.21 mmol) in water (5.0 mL) was
added to a stirred solution of the impure cyclic peptide 22 (22.7 mg, approx. 0.044 mmol) in THF (3
mL). After 4 hours the organic solvent was removed in vacuo and the water (0.5 mL) was added to the
aqueous phase, which was then acidified with 1 M HCl (2 mL) and extracted with EtOAc (4 × 20 mL).
The organic phases were dried (MgSO4), filtered, and concentrated to give the crude acid 23 (53 mg),
which was used without further purification. To a solution of the crude acid 23 (11 mg, 23 ꢂmol) in
CH2Cl2–DMF (8:1, 2.3 mL) was added HOBt (10 mg, 66 ꢂmol, 3.0 equiv), DIC (10 ꢂL, 66 ꢂmol, 3
equiv) and iPr2NEt (15 ꢂL, 88 ꢂmol, 4 equiv). After 5 minutes NH3–dioxane (0.5 M, 0.22 mL, 110
ꢂmol, 5 equiv) was added. After 1 hour NH3–dioxane (0.5 M, 0.22 mL, 110 ꢂmol, 5 equiv) was added.
After stirring for 18 hours additional DMF (0.5 mL) followed by NH3–MeOH (2.0 M, 0.11 mL, 230
ꢂmol, 10 equiv) was added. After additionally 5 hours DIC (7 ꢂL, 46 ꢂmol, 2 equiv) was added. The
next day NH3–MeOH (2.0 M, 0.06 mL, 111 ꢂmol, 5 equiv) was added and the mixture was stirred for
10 days. Finally, DIC (3.4 ꢂL, 23 ꢂmol, 1 equiv) followed by NH3–MeOH (2.0 M, 0.055 mL, 210
ꢂmol, 5 equiv) was added and after 2 days the mixture was concentrated, dissolved in MeCN–H2O (2:1)
and purified by preparative HPLC to afford azumamide D (8) (1.2 mg, 4% overall) as a white solid.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
[α]D: +32° (c = 0.08, MeOH); previously reported11 [α]D: +25° (c = 0.1, MeOH). H NMR (500 MHz,
CD3OH) δ 8.03 (d, J = 7.8 Hz, 1H), 8.01 (d, J = 8.8 Hz, 1H) 7.94 (d, J = 6.8 Hz, 1H), 7.58 (s, 1H), 7.32
(d, J = 7.5 Hz, 1H), 7.25–7.14 (m, 5H), 6.75 (s, 1H), 5.47 (m, 1H), 5.39 (m, 1H), 4.35 (m, 1H), 4.19 (m,
1H) 4.17– 4.11 (m, 2H), 3.09 (m, 2H), 2.69 (m, 1H), 2.62 (m, 1H), 2.41 (m, 2H), 2.28 (m, 1H), 1.47 (d,
J = 7.4 Hz, 3H), 1.28 (d, J = 7.4 Hz, 3H), 1.22 (d, J = 7.3 Hz, 3H). HRMS (ESIꢀTOF): m/z calculated
for C25H35N5O5H+: 486.2716; found: 486.2710 [M+H]+. HPLC gradient C, tR = 10.55 min (>95%).
Azumamide
E
[(Z)-6-((2R,5R,8R,11R,12S)-8-Benzyl-2-isopropyl-5,12-dimethyl-3,6,9,13-
tetraoxo-1,4,7,10-tetraazacyclotridecan-11-yl)hex-4-enoic acid] (9). LiOH (18.5 mg, 0.77 mmol, 40
equiv) in water (4 mL) was added to a stirred solution of the impure cyclic peptide 20 (10.5 mg, approx.
0.02 mmol) in THF (4 mL). After 1 hour LiOH (10 mg, 0.42 mmol, 20 equiv) in water (1 mL) was
ACS Paragon Plus Environment
18

Page 19 of 33
Journal of Medicinal Chemistry
added and after 2 hours LiOH (5.0 mg, 0.21 mmol, 1 equiv) in water (0.5 mL) was added. The solution
was stirred for 16 h and another portion of LiOH (6.0 mg, 0.25 mmol, 1.3 equiv) in water (0.5 mL) was
added. After additionally 2.5 hours of stirring the organic solvent was removed in vacuo. The aqueous
phase was acidified with 1 M HCl and extracted with EtOAc (4 × 25 mL). The combined organic phases
were dried (Na2SO4), filtered, and concentrated. The resulting residue was dissolved in MeCN–water
[(3:2), 2.5 mL] and purified by preparative HPLC to afford azumamide E (9) (4.3 mg, 15% overall) as a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
white solid. [α]D: +66° (c = 0.2, MeOH); previously reported11 [α]D: +53° (c = 0.1, MeOH). H NMR
(500 MHz, CD3OH) δ 8.10 (d, J = 7.7 Hz, 1H), 7.95 (d, J = 8.8 Hz, 1H), 7.84 (bs, 1H), 7.63 (d, J = 8.4
Hz, 1H), 7.28–7.16 (m, 5H), 5.48 (m, 1H), 5.37 (m, 1H), 4.28 (pentet, J = 7.5 Hz, 1H), 4.16 (m, 1H),
4.08 (m, 1H), 3.59 (m, 1H), 3.25 (dd, J = 13.6 Hz and 10.4 Hz, 1H), 3.11 (dd, J = 13.6 Hz and 6.1 Hz,
1H), 2.72 (m, 1H), 2.68 (m, 1H), 2.39 (d, J = 1.7 Hz, 6H), 2.39 (m, 6H), 1.28 (d, J = 7.1 Hz, 3H), 1.27
(d, J = 7.4 Hz, 3H), 0.94 (m, 6H). HRMS (ESIꢀTOF): m/z calculated for C27H38N4O6H+: 515.2869;
found: 515.2869 [M+H]+. HPLC gradient C, tR = 12.53 min (>95%).
Assay Materials. HDAC1 (Purity >45% by SDSꢀPAGE according to the supplier), HDAC4 (Purity
>90% by SDSꢀPAGE according to the supplier), and HDAC 7 (Purity >90% by SDSꢀPAGE according
to the supplier) were purchased from Millipore (Temecula, CA 92590). HDAC2 used for dose–response
experiments (Full length, purity ≥94% by SDSꢀPAGE according to the supplier) and HDAC 5 (Full
length, purity ≥4% by SDSꢀPAGE according to the supplier) and HDAC8 used for doseꢀresponse
experiments (Purity ≥90% by SDSꢀPAGE according to the supplier) were purchased from BPS
Bioscience (San Diego, CA 92121). HDAC2 used for initial screening experiments (Full length, purity
50% by SDSꢀPAGE according to the supplier), HDAC3ꢀ“NCoR1” complex [(Purity 90% by SDSꢀ
PAGE according to supplier; fusion protein of GSTꢀtagged HDAC3 with the deacetylase activation
domain (DAD) of NCoR1 (nuclear receptor coꢀrepressor)], HDAC6 (Purity >90% by SDSꢀPAGE
according to the supplier), HDAC8 for initial screening experiments (Purity >50% by SDSꢀPAGE
according to the supplier), HDAC10 (Purity >50% by SDSꢀPAGE according to the supplier), and
HDAC11 (Purity >50% by SDSꢀPAGE according to the supplier) were purchased from Enzo Life
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 33
Sciences (Postfach, Switzerland). HDAC9 (Full length, purity 12% by SDSꢀPAGE according to the
supplier) was purchased from Abnova (Taipei, Taiwan). The HDAC assay buffer [50 mM tris/Cl, pH
8.0, 137 mM NaCl, 2.7 mM KCl, 1 mM MgCl2, and bovine serum albumin (0.5 mg/mL)]. Trypsin
(10,000 units/mg, TPCK treated from bovine pancreas) was from Sigma Aldrich (Steinheim, Germany).
1
2
3
4
5
6
7
8
9
All peptides were purified to homogeneity (>95% purity by HPLC230
using reversedꢀphase
nm
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
preparative HPLC), and the white fluffy materials obtained by lyophilization were kept at –20 °C. For
assaying, peptides were reconstituted in DMSO to give 5–10 mM stock solutions, the accurate
concentrations of which were determined by coꢀinjection on HPLC with a standard of known
concentration.
In Vitro Histone Deacetylase Inhibition Assays. For inhibition of recombinant human HDACs the
dose–response experiments with internal controls were performed in black low binding NUNC 96ꢀwell
microtiter plates. Dilution series (3ꢀfold dilution, 10 concentrations) were prepared in HDAC assay
buffer from 5–10 mM DMSO stock solutions. The appropriate dilution of inhibitor (10
ꢀ
L of 5 × the
L) containing
M) for HDAC6 and 11; Acꢀ
M) for HDAC7; (400 M)
M) for HDAC9; and AcꢀArgꢀHisꢀLys(Ac)ꢀLys(Ac)ꢀAMC (100 M) for
HDAC10]. Finally, a solution of the appropriate HDAC (15 L) was added and the plate was incubated
desired final concentration) was added to each well followed by HDAC assay buffer (25
ꢀ
substrate [AcꢀLeuꢀGlyꢀLys(Ac)ꢀAMC (40
LeuꢀGlyꢀLys(Tfa)ꢀAMC (40 M) for HDAC4; (240
for HDAC8; and (160
ꢀ
M) for HDAC1, 2, and 3; (80
ꢀ
ꢀ
ꢀM) for HDAC5; (80
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
at 37 °C for 30 min [HDAC1: 150 ng/well, HDAC2: 100 ng/well, HDAC3: 10 ng/well, HDAC4: 2
ng/well, HDAC5: 40 ng/well, HDAC6: 60 ng/well, HDAC7: 2 ng/well, HDAC8: 5 ng/well, HDAC9: 40
ng/well, HDAC10: 500 ng/well, HDAC11: 500 ng/well]. Then trypsin (50 ꢀL, 0.4 mg/mL) was added
and the assay development was allowed to proceed for 15–30 min at room temperature, before the plate
was read using a Perkin Elmer Enspire plate reader with excitation at 360 nm and detecting emission at
460 nm. Each assay was performed in duplicate. The data were analyzed by nonꢀlinear regression using
GraphPad Prism to afford IC50 values from the doseꢀresponse experiments, and Ki values were
ACS Paragon Plus Environment
20

Page 21 of 33
Journal of Medicinal Chemistry
determined from the ChengꢀPrusoff equation [Ki = IC50/(1+[S]/Km)] assuming a standard fastꢀon–fastꢀ
1
2
3
off mechanism of inhibition.
4
5
6
7
8
ASSOCIATED CONTENT
9
Supporting Information Available: Supplementary Tables S1–S2, Figures S1–S7, and Schemes S1–
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
S2, full experimental details, compound characterization data, as well as copies of H and 13C NMR
spectra and CIF file (CCDC 933151) for the Xꢀray crystal structures are available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
Eꢀmail: cao@kemi.dtu.dk. Phone: (+45) 45252105.
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
This work was supported by the Lundbeck Foundation (Young Group Leader Fellowship, C.A.O.),
the Danish Independent Research Council–Natural Sciences (Steno Grant No. 10ꢀ080907, C.A.O.), and
the Carlsberg Foundation. Novo Nordisk A/S is thanked for a generous donation of peptide coupling
reagents used in this work. We thank Ms. Anne Hector and Dr. Charlotte H. Gotfredsen for assistance
with NMR spectroscopy and Ms. Tina Gustafsson for technical assistance with UPLCꢀMS and HRMS.
Dr. A. S. Madsen is gratefully acknowledged for assistance with the biochemical assays.
ABBREVIATIONS USED
AMC, 7ꢀaminoꢀ4ꢀmethylcoumarin; DAD, deacetylase activation domain; DIC, N,N’ꢀ
diisopropylcarbodiimide; H3, histone 3 protein; H4, histone 4 protein; HATU, Oꢀ(7ꢀazabenzotriazolꢀ1ꢀ
yl)ꢀN,N,N',N'ꢀtetramethyluronium hexafluorophosphate; HDAC, histone deacetylase; HMPA,
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 33
hexamethylphosphoramide; HOBt, hydroxybenzotriazole; HPLC, high performance liquid
chromatography; h, hour; KDAC, lysine deacylase; Kac, εꢀNꢀacetyllysine; Ktfa, εꢀNꢀ
1
2
3
4
5
6
7
8
trifluoroacetyllysine; KHMDS, potassium hexamethyldisilazide; MS, mass spectrometry; NcoR, nuclear
receptor coꢀrepressor; NMR, nuclear magnetic resonance; RFU, relative fluorescence units; rt, room
temperature; TFA, trifluoroacetic acid; tR, retention time; UPLC, ultraꢀhigh performance liquid
chromatography.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
REFERENCES
1.
deacetylase inhibitor as an anticancer drug. Nat. Biotechnol. 2007, 25, 84ꢀ90.
2. Furumai, R.; Matsuyama, A.; Kobashi, N.; Lee, K. H.; Nishiyama, M.; Nakajima, H.; Tanaka,
Marks, P. A.; Breslow, R. Dimethyl sulfoxide to vorinostat: development of this histone
A.; Komatsu, Y.; Nishino, N.; Yoshida, M.; Horinouchi, S. FK228 (depsipeptide) as a natural prodrug
that inhibits class I histone deacetylases. Cancer Res. 2002, 62, 4916ꢀ4921.
3.
Kijima, M.; Yoshida, M.; Sugita, K.; Horinouchi, S.; Beppu, T. Trapoxin, an antitumor cyclic
tetrapeptide, is an irreversible inhibitor of mammalian histone deacetylase. J. Biol. Chem. 1993, 268,
22429ꢀ22435.
4.
transcriptional regulator Rpd3p. Science 1996, 272, 408ꢀ411.
5. Taunton, J.; Collins, J. L.; Schreiber, S. L. Synthesis of natural and modified trapoxins, useful
reagents for exploring histone deacetylase function. J. Am. Chem. Soc. 1996, 118, 10412ꢀ10422.
6. Beumer, J. H.; Tawbi, H. Role of histone deacetylases and their inhibitors in cancer biology and
treatment. Curr. Clin. Pharmacol. 2010, 5, 196ꢀ208.
7. Haberland, M.; Montgomery, R. L.; Olson, E. N. The many roles of histone deacetylases in
development and physiology: implications for disease and therapy. Nat. Rev. Genet. 2009, 10, 32ꢀ42.
8. Kazantsev, A. G.; Thompson, L. M. Therapeutic application of histone deacetylase inhibitors for
Taunton, J.; Hassig, C. A.; Schreiber, S. L. A mammalian histone deacetylase related to the yeast
central nervous system disorders. Nat. Rev. Drug Discov. 2008, 7, 854ꢀ868.
ACS Paragon Plus Environment
22

Page 23 of 33
Journal of Medicinal Chemistry
9.
Minucci, S.; Pelicci, P. G. Histone deacetylase inhibitors and the promise of epigenetic (and
1
2
3
4
5
more) treatments for cancer. Nat. Rev. Cancer 2006, 6, 38ꢀ51.
10. DarkinꢀRattray, S. J.; Gurnett, A. M.; Myers, R. W.; Dulski, P. M.; Crumley, T. M.; Allocco, J.
6
7
8
9
J.; Cannova, C.; Meinke, P. T.; Colletti, S. L.; Bednarek, M. A.; Singh, S. B.; Goetz, M. A.;
Dombrowski, A. W.; Polishook, J. D.; Schmatz, D. M. Apicidin: a novel antiprotozoal agent that
inhibits parasite histone deacetylase. Proc. Natl. Acad. Sci. U. S. A. 1996, 93, 13143ꢀ13147.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11.
Nakao, Y.; Yoshida, S.; Matsunaga, S.; Shindoh, N.; Terada, Y.; Nagai, K.; Yamashita, J. K.;
Ganesan, A.; van Soest, R. W.; Fusetani, N. Azumamides AꢀE: histone deacetylase inhibitory cyclic
tetrapeptides from the marine sponge Mycale izuensis. Angew. Chem. Int. Ed. 2006, 45, 7553ꢀ7557.
12.
Maulucci, N.; Chini, M. G.; Micco, S. D.; Izzo, I.; Cafaro, E.; Russo, A.; Gallinari, P.; Paolini,
C.; Nardi, M. C.; Casapullo, A.; Riccio, R.; Bifulco, G.; Riccardis, F. D. Molecular insights into
azumamide e histone deacetylases inhibitory activity. J. Am. Chem. Soc. 2007, 129, 3007ꢀ3012.
13.
Wang, D.; Helquist, P.; Wiest, O. Zinc binding in HDAC inhibitors: a DFT study. J. Org. Chem.
2007, 72, 5446ꢀ5449.
14.
Izzo, I.; Maulucci, N.; Bifulco, G.; De Riccardis, F. Total synthesis of azumamides A and E.
Angew. Chem. Int. Ed. 2006, 45, 7557ꢀ7560.
15.
Wen, S.; Carey, K. L.; Nakao, Y.; Fusetani, N.; Packham, G.; Ganesan, A. Total synthesis of
azumamide A and azumamide E, evaluation as histone deacetylase inhibitors, and design of a more
potent analogue. Org. Lett. 2007, 9, 1105ꢀ1108.
16.
Chandrasekhar, S.; Rao, C. L.; Seenaiah, M.; Naresh, P.; Jagadeesh, B.; Manjeera, D.; Sarkar,
A.; Bhadra, M. P. Total synthesis of azumamide E and sugar amino acidꢀcontaining analogue. J. Org.
Chem. 2009, 74, 401ꢀ404.
17.
Lahm, A.; Paolini, C.; Pallaoro, M.; Nardi, M. C.; Jones, P.; Neddermann, P.; Sambucini, S.;
Bottomley, M. J.; Lo Surdo, P.; Carfi, A.; Koch, U.; De Francesco, R.; Steinkuhler, C.; Gallinari, P.
Unraveling the hidden catalytic activity of vertebrate class IIa histone deacetylases. Proc. Natl. Acad.
Sci. U. S. A. 2007, 104, 17335ꢀ17340.
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 33
18.
Mazitschek, R. Chemical phylogenetics of histone deacetylases. Nat. Chem. Biol. 2010, 6, 238ꢀ243.
19. White, C. J.; Yudin, A. K. Contemporary strategies for peptide macrocyclization. Nat. Chem.
2011, 3, 509ꢀ524.
Bradner, J. E.; West, N.; Grachan, M. L.; Greenberg, E. F.; Haggarty, S. J.; Warnow, T.;
1
2
3
4
5
6
7
8
9
20.
Tang, T. P.; Ellman, J. A. Asymmetric synthesis of betaꢀamino acid derivatives incorporating a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
broad range of substitution patterns by enolate additions to tertꢀbutanesulfinyl imines. J. Org. Chem.
2002, 67, 7819ꢀ7832.
21.
Butanesulfinamide. Chem. Rev. 2010, 110, 3600ꢀ3740.
22. Zimmerman, H. E.; Traxler, M. D. The Stereochemistry of the Ivanov and Reformatsky
Reactions .1. J. Am. Chem. Soc. 1957, 79, 1920ꢀ1923.
23. Cutter, A. C.; Miller, I. R.; Keily, J. F.; Bellingham, R. K.; Light, M. E.; Brown, R. C. D. Total
Robak, M. T.; Herbage, M. A.; Ellman, J. A. Synthesis and Applications of tertꢀ
Syntheses of (–)ꢀEpilupinine and (–)ꢀTashiromine Using IminoꢀAldol Reactions. Org. Lett. 2011, 13,
3988ꢀ3991.
24.
Ireland, R. E.; Mueller, R. H.; Willard, A. K. Ester Enolate Claisen Rearrangement ꢀ
Stereochemical Control through Stereoselective Enolate Formation. J. Am. Chem. Soc. 1976, 98, 2868ꢀ
2877.
25.
Oare, D. A.; Heathcock, C. H. Acyclic Stereoselection. 47. Stereochemistry of the Michael
Addition of Ester and Ketone Enolates to Alpha,BetaꢀUnsaturated Ketones. J. Org. Chem. 1990, 55,
157ꢀ172.
26.
Montero, A.; Beierle, J. M.; Olsen, C. A.; Ghadiri, M. R. Design, Synthesis, Biological
Evaluation, and Structural Characterization of Potent Histone Deacetylase Inhibitors Based on Cyclic
alpha/betaꢀTetrapeptide Architectures. J. Am. Chem. Soc. 2009, 131, 3033ꢀ3041.
27.
Olsen, C. A.; Ghadiri, M. R. Discovery of potent and selective histone deacetylase inhibitors via
focused combinatorial libraries of cyclic alpha3betaꢀtetrapeptides. J. Med. Chem. 2009, 52, 7836ꢀ7846.
ACS Paragon Plus Environment
24

Page 25 of 33
Journal of Medicinal Chemistry
28.
Vickers, C. J.; Olsen, C. A.; Leman, L. J.; Ghadiri, M. R. Discovery of HDAC Inhibitors That
1
2
3
4
5
6
7
8
9
Lack an Active Site Zn2+ꢀBinding Functional Group. ACS Med. Chem. Lett. 2012, 3, 505ꢀ508.
29. Olsen, C. A.; Montero, A.; Leman, L. J.; Ghadiri, M. R. Macrocyclic PeptoidꢀPeptide Hybrids as
Inhibitors of Class I Histone Deacetylases. ACS Med. Chem. Lett. 2012, 3, 749ꢀ753.
30. Madsen, A. S.; Olsen, C. A. Substrates for Efficient Fluorometric Screening Employing the
NADꢀDependent Sirtuin 5 Lysine Deacylase (KDAC) Enzyme. J. Med. Chem. 2012, 55, 5582ꢀ5590.
31. Madsen, A. S.; Olsen, C. A. Profiling of Substrates for Zincꢀdependent Lysine Deacylase
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Enzymes: HDAC3 Exhibits Decrotonylase Activity In Vitro. Angew. Chem. Int. Ed. 2012, 51, 9083ꢀ
9087.
32.
of HDAC6. ACS Chem. Biol. 2012, 7, 331ꢀ337.
33. Szpilman, A. M.; Carreira, E. M. Probing the biology of natural products: molecular editing by
diverted total synthesis. Angew. Chem. Int. Ed. 2010, 49, 9592ꢀ9628.
34. Wender, P. A.; Buschmann, N.; Cardin, N. B.; Jones, L. R.; Kan, C.; Kee, J. M.; Kowalski, J. A.;
Inks, E. S.; Josey, B. J.; Jesinkey, S. R.; Chou, C. J. A Novel Class of Small Molecule Inhibitors
Longcore, K. E. Gateway synthesis of daphnane congeners and their protein kinase C affinities and cellꢀ
growth activities. Nat. Chem. 2011, 3, 615ꢀ619.
35.
Wilson, R. M.; Danishefsky, S. J. On the reach of chemical synthesis: creation of a miniꢀpipeline
from an academic laboratory. Angew. Chem. Int. Ed. 2010, 49, 6032ꢀ6056.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 33
Table of Contents graphic.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
26

Page 27 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1
82x70mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 33
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
80x23mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 29 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
90x62mm (300 x 300 DPI)
ACS Paragon Plus Environment