An unexpected example of protein-templated click chemistry

Article abstract of DOI:10.1002/anie.201002205

(Figure Presented) It all happened with a click: In a search for histone deacetylase (HDAC) inhibitors using in situ click chemistry, the first example of protein-Cu acceleration of the azide-alkyne cycloaddition reaction was uncovered. The copper center

Full text of DOI:10.1002/anie.201002205

Angewandte
Chemie
DOI: 10.1002/anie.201002205
Click Chemistry
An Unexpected Example of Protein-Templated Click Chemistry**
Takayoshi Suzuki,* Yosuke Ota, Yuki Kasuya, Motoh Mutsuga, Yoko Kawamura,
Hiroki Tsumoto, Hidehiko Nakagawa, M. G. Finn,* and Naoki Miyata*
Click chemistry is a popular approach to the synthesis of
functionalized molecules, and emphasizes the use of practical
and reliable reactions.[1] Copper(I)-catalyzed azide–alkyne
cycloaddition[2] (CuAAC), which selectively produces anti-
(1,4)-triazoles in preference to the syn isomer (1,5-triazole), is
regarded as a superlative example of click chemistry. The
CuAAC reaction can be accelerated by CuI-stabilizing
ligands, such as tris[(1-substituted-1H-1,2,3-triazol-4-yl)me-
thyl]amines[3] and tris(2-benzimidazolylmethyl)amines.[4] The
catalytic system has received a great deal of use in various
fields such as chemical biology and materials science.[1,5] The
1,3-dipolar cycloaddition of azides with unactivated alkynes
occurs much more slowly but is highly chemoselective. This
property stimulated the development of “in situ click
chemistry” for the field of drug discovery, in which target
enzymes are allowed to assemble new inhibitors by linking
azides and alkynes that bind to adjacent sites on the protein
surface.[6] The linkage reaction does not employ Cu catalysis,
but instead relies on acceleration of the otherwise sluggish
[3+2] cycloaddition reaction when the reaction partners are
held in proximity to each other, often in or near the enzyme
active site. In the course of an in situ click chemistry study on
histone deacetylase (HDAC), we unexpectedly observed
acceleration of the AAC reaction by trace copper associated
with the protein in a structurally sensitive manner. Herein we
report these findings, which constitute the first example of a
Cu-protein complex catalyzing the AAC reaction.
HDAC inhibitors are attractive drug candidates for
cancer, inflammation, and neurodegenerative disorders.[7]
As shown in Figure 1, most HDAC inhibitors consist of a
Figure 1. Structural characteristics of HDAC inhibitors.
Zn-binding group (ZBG) that coordinates with the Zn ion in
the active site, a capping region that interacts with residues on
the rim of the active site, and a linker that connects the cap
and ZBG at an appropriate distance. For example, the
clinically used HDAC inhibitor vorinostat (Figure 1) consists
of a hydroxamic acid (ZBG), an anilide (cap), and an alkyl
chain (linker). This general linked motif resembles that of
acetylcholinesterase, the first target of in situ click chemis-
try.[6a] Accordingly, we prepared two alkynes with hydroxamic
acid (1a and 1b) and 15 alkyl azides (2a–o) as building blocks
for in situ assembly screening (Scheme 1).
[*] Dr. T. Suzuki, Y. Ota, Y. Kasuya, Dr. H. Tsumoto, Dr. H. Nakagawa,
Prof. N. Miyata
Graduate School of Pharmaceutical Sciences
Nagoya City University
3-1 Tanabe-dori, Mizuho-ku, Nagoya, Aichi 467-8603 (Japan)
Fax: (+81)52-836-3407
E-mail: suzuki@phar.nagoya-cu.ac.jp
miyata-n@phar.nagoya-cu.ac.jp
Dr. T. Suzuki
PRESTO (Japan) Science and Technology Agency (JST)
4-1-8 Honcho Kawaguchi, Saitama 332-0012 (Japan)
Prof. N. Miyata
Division of Organic Chemistry
National Institute of Health Sciences
1-18-1 Kamiyoga, Setagaya-ku, Tokyo 158-8501 (Japan)
In conventional in situ click chemistry, a mixture of an
alkyne and an azide is incubated in the presence of the target
Dr. M. Mutsuga, Dr. Y. Kawamura
Division of Food Additives
National Institute of Health Sciences
1-18-1 Kamiyoga, Setagaya-ku, Tokyo 158-8501 (Japan)
Prof. M. G. Finn
Department of Chemistry
The Scripps Research Institute
10550 North Torrey Pines Road, La Jolla, CA 92037 (USA)
Fax: (+1)858-784-8850
E-mail: mgfinn@scripps.edu
[**] This work was supported in part by the JST PRESTO program (T.S.)
and grants from Taisho Pharmaceutical Co., Ltd. (T.S.), the Uehara
Memorial Foundation (T.S.), and the Skaggs Institute for Chemical
Biology (M.G.F.).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002205.
Scheme 1. Structures of alkynes 1a and 1b and azides 2a–o.
Angew. Chem. Int. Ed. 2010, 49, 6817 –6820
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6817

Communications
protein and the formation of the product is monitored by LC/
absence (Figure 3d). The triazole was not formed when
bovine serum albumin was substituted for HDAC8 (Fig-
ure 3e), nor when the HDAC8 reaction was performed in the
presence of an excess of vorinostat, an active site-binding
HDAC inhibitor (Figure 3 f). These results show that the
triazole formation takes place in the active site. Compound
anti-3 was synthesized independently and found to be a
competent HDAC inhibitor (Table S1 in the Supporting
Information).
MS analysis. The alkyne fragment is not activated by electron-
withdrawing substituents, so as to make the uncatalyzed
cycloaddition with azide very slow. At concentrations typi-
cally used for in situ inhibitor discovery, the background
reaction has been estimated to have a half life of many
years,[6a] making false positives very rare. It also makes even
the target-templated reaction slow in absolute terms, such
that only a percent or two of inhibitor is formed and detected
by mass spectrometry. Independent synthesis is then neces-
sary to confirm that the assembled triazole is indeed a good
inhibitor.
Since we had a convenient assay of HDAC activity
available,[8] we chose to see if enough of an inhibitor could be
generated by in situ assembly to measurably affect enzyme
function. We therefore incubated a mixture of each known
alkyne ligand (at a concentration approximately equivalent to
its IC50) with each candidate azide (in large excess) in the
presence of HDAC and subsequently carried out a fluoro-
metric assay for HDAC activity directly on the reaction
mixture (Scheme 2). These experiments were conducted in
parallel in 96-well microtiter plates using human recombinant
HDAC8, which had been shown separately to be stable under
the incubation and assay conditions. As shown in Figure 2, a
significant decrease in fluorescence was observed only for the
combination of alkyne 1b and azide 2o relative to the effect
of 1b and 2o alone. Other alkyne/azide combinations resulted
in no statistically significant change in fluorescence signal in
the presence of both the azide and alkyne, relative to alkyne
alone (data not shown).
Figure 3. Identification of in situ click chemistry product generated
from 1b (1 mm) and 2o (1 mm) by HPLC analysis: a) authentic
sample anti-3 (13.1 min); b) a mixture of syn-3 (8.37 min) and anti-3
(12.9 min); c) reaction in the presence of HDAC8 (0.6 mm), in situ
product (12.7 min); d) reaction without enzyme; e) reaction in the
presence of BSA (0.6 mm); f) reaction in the presence of HDAC8
(0.6 mm) and vorinostat (10 mm). mAU=milli absorbance units.
The in situ and regiospecific formation of a triazole from
1b and 2o was also observed by HPLC (Figure 3). Incubation
of 1b and 2o gave rise to a significant peak corresponding to
anti-3 in the presence of HDAC8 (Figure 3c), but not in its
Several aspects of these apparently successful in situ click
chemistry results were unusual. First, while ordinary in situ
reactions produce only very small amounts of triazole per
enzyme, the yield of the triazole in this study was as much as
50%. Second, no syn-3 was found in the in situ process, even
though the syn-triazole isomer proved to be a better HDAC8
inhibitor (IC50 = 0.51 mm) than anti-3 (IC50 = 4.0 mm). These
factors suggested that the rapid and regioselective formation
of anti-3 in the presence of HDAC8 is consistent with
acceleration of the reaction by a small amount of enzyme
containing CuI instead of ZnII.
Scheme 2. In situ click chemistry screening using a fluorometric assay
for HDAC activity.
Indeed, ICP–MS analysis of a mixture of alkyne 1b
(1 mm), azide 2o (1 mm), and HDAC8 (0.6 mm) revealed the
presence of both 0.95 mm of Zn and 0.10 mm of Cu. ICP–MS
analysis of each component independently showed that
alkyne 1b contained 0.01 mol% Cu, which probably came
from the Sonogashira reaction used for its synthesis
(Scheme S2 in the Supporting Information).
Figure 2. In situ click chemistry screening using fluorometric assay for
HDAC activity. **P<0.01; ANOVA and Bonferroni-type multiple t test
results indicated differences between 1b or 2o alone and a combina-
tion of 1b with 2o.
To examine whether CuI is required for the cycloaddition,
we repeated the in situ experiment in the presence of various
concentrations of bathocuproine disulfonic acid (BCDSA),[9]
6818
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6817 –6820

Angewandte
Chemie
a CuI-specific chelator and inhibitor of the CuAAC reaction.
As shown in Figures 4 and 5, BCDSA inhibited the triazole
formation in dose-dependent fashion. In addition, Zn(OAc)2
was added to the mixture of alkyne 1b, azide 2o, and HDAC8
to displace Cu from the protein. This treatment completely
suppressed the formation of anti-3 (Figure S2 in the Support-
ing Information), whereas in solution ZnII has no effect on the
CuAAC reaction.[10] Conversely, kinetic measurements of
HDAC8 catalytic activity showed that added CuI increased
the Km value for substrate relative to the value observed in the
absence of added cuprous ion (Lineweaver–Burk analysis,
Figure S3 in the Supporting Information).
Cu–ligand catalysts of the CuAAC reaction have to be
used at 10 mm or higher concentrations.[3b,4] These results
suggest that protein–Cu complexes can be developed as
highly active CuAAC catalysts.
2) It shows that a single Cu center may be enough to catalyze
the reaction,[11] in contrast to most of our kinetics experi-
ments, which indicate that two Cu atoms are required.[12]
However, many possibilities exist for coordination of Cu
in the HDAC binding pocket. In addition to the two Asp
(178 and 267) and one His (180) residues that are shown to
bind Zn in the X-ray crystal structures of HDAC8, four
other potential metal-binding side chains are in the
immediate vicinity (His142, His143, Met274, and
Tyr306).[13] Thus, it is conceivable for two metal centers
(Zn/Cu or Cu/Cu) to occupy the active site together, or for
a single Cu center to do so in different ways. Further
investigation is needed to determine the CuAAC active
structure.
3) It shows that HDAC8 preserves Cu in the + 1 oxidation
state, even though the solution is not protected from air.
The histidine and methionine residues in the active site of
HDAC8 may contribute to the stabilization of CuI in
analogy to natural CuI environments in copper-containing
enzymes.[14]
4) It shows that the HDAC8–Cu complex, while being a fast
catalyst, is not a general one, since it provides triazole only
for the 1b + 2a combination among the possibilities
tested, and therefore that the reaction is guided by the
protein structure. The X-ray crystal structure of HDAC8
shows a large hydrophobic pocket next to the active site.[15]
The adamantyl group of 2o could be located in this pocket
in an orientation that allows reaction with a Cu–acetylide
generated from alkyne 1b and CuI in the active site.
Figure 4. HPLC analysis of triazole formation from 1b (1 mm) and 2o
(1 mm) in the presence of HDAC8 (0.6 mm): a) authentic sample of
anti-3; b) reaction in the absence of the CuI-specific chelator BCDSA;
c–f) reaction in the presence of the indicated concentration of BCDSA.
If the protein holds the reaction components in such a way
as to allow Cu in the Zn binding site to selectively assemble a
triazole in situ, the metal center would interact with the
alkyne portion of 1b rather than the hydroxamate, essentially
inverting the orientation shown schematically in Scheme 2.
We would then expect the molecule so formed to be less likely
than most hydroxamates to bind to the Zn atom, and
therefore to be a less potent inhibitor of the enzyme. In
addition to syn-3, preliminary experiments showed that other
triazoles available from the azide–alkyne library, such as the
1b/2 f combination, were also more potent inhibitors than
anti-3, but were not formed in the enzyme. These compounds
can presumably benefit from hydroxamate–Zn binding to a
greater degree than anti-3. While helpful and used by most
HDAC inhibitors, such binding is not a requirement for
HDAC8 inhibition (Figure S4 in the Supporting Informa-
tion).[16] The fact that anti-3 is only a moderate inhibitor is
probably responsible for the catalytic nature of its production
(more triazole is formed than there is Cu in the sample): with
a binding affinity of 4 mm, most of the protein is not bound by
anti-3 during the course of the reaction, and the off-rate is
likely to be fast, allowing for catalytic turnover.
Figure 5. Inhibition of triazole anti-3 formation by the CuI-specific
chelator BCDSA.
These data strongly suggest that CuI, acting in the active
site of HDAC8, accelerates the rate of cycloaddition between
alkyne 1b and azide 2o. Indeed, CuAAC reactions attempted
under analogous conditions but without HDAC8 (CuBr at 0.1
or 0.6 mm; CuSO4 at 0.1 or 0.6 mm and sodium ascorbate at 0.5
or 3.0 mm) resulted in no triazole formation (Figure S1 in the
Supporting Information). This study is important for four
reasons:
1) It shows for the first time that a protein–Cu complex can
be a far better catalyst than Cu alone. HDAC8–Cu
complex at 0.1 mm catalyzes triazole formation, whereas
CuI at submicromolar concentration provides very little
reaction (Figure S1 f).[3b,4] In order to be effective, most
In conclusion, we have established that a CuI complex of
HDAC8 accelerates a selective reaction between an azide and
an alkyne, thereby forming a compound with greater inhib-
Angew. Chem. Int. Ed. 2010, 49, 6817 –6820
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6819

Communications
2010, 39, 1252 – 1261; d) X. Hu, R. Manetsch, Chem. Soc. Rev.
2010, 39, 1316 – 1324.
[7] a) T. Suzuki, Chem. Pharm. Bull. 2009, 57, 897 – 906; b) M. Paris,
M. Porcelloni, M. Binaschi, D. Fattori, J. Med. Chem. 2008, 51,
1505 – 1529.
itory power than either of the pieces alone. In this case, the
unwavering regioselectivity of the CuAAC process overrides
the tendency of the Cu-free in situ click reaction to form the
more potent of the two triazole isomers. We regard this as an
important precedent in the search for highly active protein
catalysts of the CuAAC reaction, as well as an interesting and
potentially useful example of in situ inhibitor formation.
[8] a) T. Suzuki, Y. Nagano, A. Kouketsu, A. Matsuura, S. Mar-
uyama, M. Kurotaki, H. Nakagawa, N. Miyata, J. Med. Chem.
2005, 48, 1019 – 1032; b) T. Suzuki, A. Kouketsu, Y. Itoh, S.
Hisakawa, S. Maeda, M. Yoshida, H. Nakagawa, N. Miyata, J.
Med. Chem. 2006, 49, 4809 – 4812; c) Y. Itoh, T. Suzuki, A.
Kouketsu, N. Suzuki, S. Maeda, M. Yoshida, H. Nakagawa, N.
Miyata, J. Med. Chem. 2007, 50, 5425 – 5438; d) N. Suzuki, T.
Suzuki, Y. Ota, T. Nakano, M. Kurihara, H. Okuda, T. Yamori,
H. Tsumoto, H. Nakagawa, N. Miyata, J. Med. Chem. 2009, 52,
2902 – 2922; e) T. Asaba, T. Suzuki, R. Ueda, H. Tsumoto, H.
Nakagawa, N. Miyata, J. Am. Chem. Soc. 2009, 131, 6989 – 6996.
[9] A. T. Faizullah, A. Townshend, Anal. Chim. Acta 1985, 172, 291 –
296.
[10] See section G in the Supporting Information of reference [3b].
[11] C. Nolte, P. Mayer, B. F. Straub, Angew. Chem. 2007, 119, 2147 –
2149; Angew. Chem. Int. Ed. 2007, 46, 2101 – 2103.
[12] a) V. O. Rodionov, V. V. Fokin, M. G. Finn, Angew. Chem. 2005,
117, 2250 – 2255; Angew. Chem. Int. Ed. 2005, 44, 2210 – 2215;
b) V. O. Rodionov, S. I. Presolski, D. D. Dꢀaz, V. V. Fokin, M. G.
Finn, J. Am. Chem. Soc. 2007, 129, 12705 – 12712.
[13] A. Vannini, C. Volpari, G. Filocamo, E. C. Casavola, M.
Brunetti, D. Renzoni, P. Chakravarty, C. Paolini, R. De Fran-
cesco, P. Gallinari, C. Steinkꢁhler, S. Di Marco, Proc. Natl. Acad.
Sci. USA 2004, 101, 15064 – 15069.
[14] a) Y. Rondelez, G. Bertho, O. Reinaud, Angew. Chem. 2002, 114,
1086 – 1088; Angew. Chem. Int. Ed. 2002, 41, 1044 – 1046; b) S. T.
Prigge, B. A. Eipper, R. E. Mains, L. M. Amzel, Science 2004,
304, 864 – 867.
[15] a) J. R. Somoza, R. J. Skene, B. A. Katz, C. Mol, J. D. Ho, A. J.
Jennings, C. Luong, A. Arvai, J. J. Buggy, E. Chi, J. Tang, B. C.
Sang, E. Verner, R. Wynands, E. M. Leahy, D. R. Dougan, G.
Snell, M. Navre, M. W. Knuth, R. V. Swanson, D. E. McRee,
L. W. Tari, Structure 2004, 12, 1325 – 1334; b) K. KrennHrubec,
B. L. Marshall, M. Hedglin, E. Verdin, S. M. Ulrich, Bioorg.
Med. Chem. Lett. 2007, 17, 2874 – 2878.
Received: April 14, 2010
Revised: June 13, 2010
Published online: August 16, 2010
Keywords: alkynes · azides · click chemistry · copper · proteins
.
[1] a) H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001,
113, 2056 – 2075; Angew. Chem. Int. Ed. 2001, 40, 2004 – 2021;
b) M. G. Finn, V. V. Fokin, Chem. Soc. Rev. 2010, 39, 1231 – 1232.
[2] a) C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002,
67, 3057 – 3062; b) V. V. Rostovtsev, L. G. Green, V. V. Fokin,
K. B. Sharpless, Angew. Chem. 2002, 114, 2708 – 2711; Angew.
Chem. Int. Ed. 2002, 41, 2596 – 2599.
[3] a) T. R. Chan, R. Hilgraf, K. B. Sharpless, V. V. Fokin, Org. Lett.
2004, 6, 2853 – 2855; b) V. Hong, S. I. Presolski, C. Ma, M. G.
Finn, Angew. Chem. 2009, 121, 10063 – 10067; Angew. Chem. Int.
Ed. 2009, 48, 9879 – 9883.
[4] V. O. Rodionov, S. I. Presolski, S. Gardinier, Y. H. Lim, M. G.
Finn, J. Am. Chem. Soc. 2007, 129, 12696 – 12704.
[5] For recent reviews and original papers, see a) J. F. Lutz, Angew.
Chem. 2007, 119, 1036 – 1043; Angew. Chem. Int. Ed. 2007, 46,
1018 – 1025; b) C. M. Salisbury, B. F. Cravatt, J. Am. Chem. Soc.
2008, 130, 2184 – 2194; c) C. Xu, E. Soragni, C. J. Chou, D.
Herman, H. L. Plasterer, J. R. Rusche, J. M. Gottesfeld, Chem.
Biol. 2009, 16, 980 – 989; d) A. Carlmark, C. Hawker, A. Hult, M.
Malkoch, Chem. Soc. Rev. 2009, 38, 352 – 362.
´
[6] a) W. G. Lewis, L. G. Green, F. Grynszpan, Z. Radic, P. R.
Carlier, P. Taylor, M. G. Finn, K. B. Sharpless, Angew. Chem.
2002, 114, 1095 – 1099; Angew. Chem. Int. Ed. 2002, 41, 1053 –
1057; b) H. C. Kolb, K. B. Sharpless, Drug Discovery Today 2003,
8, 1128 – 1137; c) S. K. Mamidyala, M. G. Finn, Chem. Soc. Rev.
[16] G. Estiu, N. West, R. Mazitschek, E. Greenberg, J. E. Bradner,
O. Wiest, Bioorg. Med. Chem. 2010, 18, 4103 – 4110.
6820 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6817 –6820