New cathepsin inhibitors to explore the fluorophilic properties of the S2 pocket of cathepsin B: Design, synthesis, and biological evaluation

Article abstract of DOI:10.1002/chem.201100113

Fluor-in or out? Based on β,β-difluorinated cycloaliphatic amino acids, a library of new dipeptide nitriles was evaluated as human cathepsin inhibitors. The orientation of the fluorinated face relative to the protein structure of cathepsin B was elucidated by molecular modeling and NMR studies (see figure). For (R)-configured eutomers, the fluorine atoms are directed to the S² pocket, whereas in (S)-configured distomers, the fluorinated face is solvent-exposed.

Full text of DOI:10.1002/chem.201100113

DOI: 10.1002/chem.201100113
New Cathepsin Inhibitors to Explore the Fluorophilic Properties of the S2
Pocket of Cathepsin B: Design, Synthesis, and Biological Evaluation
Santos Fustero,*[a, b] Vanessa Rodrigo,[b] Marꢀa Sꢁnchez-Rosellꢂ,[b] Carlos del Pozo,[a]
Joaquꢀn Timoneda,[c] Maxim Frizler,[d, e] Mihiret T. Sisay,[d, f] Jꢃrgen Bajorath,[f]
Luis P. Calle,[g] F. Javier CaÇada,[g] Jesffls JimØnez-Barbero,*[g] and
Michael Gꢃtschow*[d, e]
In memory of Rafael Suau.
Cysteine cathepsins (Cat) represent a group of human ly-
sosomal cysteine proteases that share a common papain-like
structural fold. They are involved in several pathophysiolog-
ical conditions and represent new promising therapeutic tar-
gets.[1] For example, Cat B is implicated in tumor invasion
and metastasis,[2] apoptotic pathway crosstalk,[3] and process-
ing of the amyloid precursor protein.[4] Cat K, the predomi-
nant osteoclastic protease, is widely accepted as a target for
the treatment of diseases involving excessive bone resorp-
tion, such as osteoporosis.[5] Typically, cathepsin inhibitors
contain entities prone to covalently interact with the cys-
teine–thiol moiety. Among the synthetic cathepsin inhibi-
tors, peptide-based nitriles have received much attention in
recent years.[6,7] These nitriles inhibit cysteine proteases by
the reversible formation of a thioimidate adduct resulting
from the nucleophilic attack of the thiol to the carbon–nitro-
gen triple bond. The selectivity of the inhibitors towards
cathepsins is mainly determined by the interaction of the
peptidic residues of the inhibitor (P1, P2, P3) with the non-
primed subsites (S1, S2, S3) of each different enzyme.[6] Non-
chiral aminoacetonitriles such as 1 with benzamide moieties
at P3 and homocycloleucine (1-aminocyclohexane-1-carbox-
ylic acid) at P2 displayed subnanomolar inhibitory activity
against Cat K.[8] The fluorine-containing Cat K inhibitor
odanacatib is currently being evaluated in clinical trials for
the treatment of osteoporosis.[5,6,9]
[a] Prof. Dr. S. Fustero, Dr. C. del Pozo
Departamento de Quꢀmica Orgꢁnica, Universidad de Valencia
46100 Burjassot (Spain)
Fax : (+34)963-544-939
E-mail: santos.fustero@uv.es
[b] Prof. Dr. S. Fustero, Dr. V. Rodrigo, Dr. M. Sꢁnchez-Rosellꢂ
Laboratorio de Molꢃculas Orgꢁnicas
Centro de Investigaciꢂn Prꢀncipe Felipe
46012 Valencia (Spain)
[c] Prof. Dr. J. Timoneda
Departamento de Bioquꢀmica y Biologꢀa Molecular
Universidad de Valencia, 46100 Burjassot (Spain)
One of the important strategies in drug design is the iden-
tification of fluorophilic/fluorophobic environments within
the target protein for favorable/unfavorable interactions
with covalently bound fluorine atoms in the ligand. Such an
approach was explored successfully to develop inhibitors of
serine and cysteine proteases.[10] Based on the recently de-
scribed efficient strategy for the asymmetric synthesis of
b,b-difluorinated 1-aminocyclohexane- and 1-aminocyclo-
pentane-1-carboxylic acid derivatives of type 2 with a qua-
ternary stereocenter[11] (for structures, see Table 1), we envi-
sioned the possibility of using these substructures as building
blocks to generate new cathepsin inhibitors. Herein, we
report on the synthesis and biological evaluation of a small
library of compounds of type 5 (Table 1), as well as on mo-
lecular modeling studies and NMR spectroscopic measure-
ments to explore the interaction of these inhibitors with
cathepsin B.
[d] M. Frizler, Dr. M. T. Sisay, Prof. Dr. M. Gꢄtschow
Pharmaceutical Institute, Pharmaceutical Chemistry I
University of Bonn, 53121 Bonn (Germany)
Fax : (+49)228-732567
E-mail: guetschow@uni-bonn.de
[e] M. Frizler, Prof. Dr. M. Gꢄtschow
NRW International Graduate Research School Biotech-Pharma
53105 Bonn (Germany)
[f] Dr. M. T. Sisay, Prof. Dr. J. Bajorath
Department of Life Science Informatics, B-IT
University of Bonn, 53113 Bonn (Germany)
[g] Dr. L. P. Calle, Dr. F. J. CaÇada, Prof. Dr. J. Jimꢃnez-Barbero
Chemical and Physical Biology, Centro de Investigaciones Biolꢂgicas
CSIC, 28040 Madrid (Spain)
Fax : (+34)915-360-432
E-mail: jjbarbero@cib.csic.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100113.
5256
ꢅ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5256 – 5260

COMMUNICATION
Table 1. Synthesis and yields of fluorinated compounds 5[a]
.
give both chiral and racemic amides 4 (Table 1). Compounds
4
were reacted with n-tetrabutyl ammonium fluoride
(TBAF)[12] and the resulting amino acids were coupled, with-
out purification, with aminoacetonitrile hydrochloride,
under standard conditions, to obtain the desired fluorinated
dipeptidomimetics 5 (Table 1).[13] The non-fluorinated refer-
ence compounds 6a–6d (for structures, see Table 2) were
synthesized (see the Supporting Information, Figure S1) fol-
lowing the sequence described for the homocycloleucine de-
rivatives[8,13]
.
Table 2. Cathepsin inhibition by dipeptide nitriles 5a–5d and 6a–6d.
Compound
Ki [mm][a]
Cat B
Cat K
Cat L
Cat S
(S)-5a
(R)-5a
(S)-5b
(R)-5b
(S)-5c
(R)-5c
(S)-5d
(R)-5d
(Æ)-5a
(Æ)-5b
(Æ)-5c
(Æ)-5d
6a
7.2Æ0.2
0.25Æ0.01
3.8Æ0.2
0.29Æ0.02
28Æ1
0.36Æ0.01
0.036Æ0.001
0.10Æ0.01
0.034Æ0.001
3.5Æ0.2
>28
16Æ1
>28
>28
>28
>28
>28
>28
23Æ2
>28
>28
>28
21Æ4
>28
>28
>28
>85
>85
>85
>85
>85
>85
>85
>85
>85
>85
>85
>85
4.6Æ0.1
7.0Æ0.3
>85
Entry
n
R2
4 (Yield [%])
5 (Yield [%])
1
2
3
4
5
6
7
8
1
1
1
1
0
0
0
0
1
1
0
0
Br
Br
Ph
Ph
Br
Br
Ph
Ph
Br
Ph
Br
Ph
(S)-4a (72)
(R)-4a (51)
(S)-4b (68)
(R)-4b (48)
(S)-4c (73)
(R)-4c (78)
(S)-4d (40)
(R)-4d (66)
(Æ)-4a (85)
(Æ)-4b (75)
(Æ)-4c (74)
(Æ)-4d (88)
(S)-5a (57)
(R)-5a (40)
(S)-5b (58)
(R)-5b (78)
(S)-5c (77)
(R)-5c (30)
(S)-5d (72)
(R)-5d (46)
(Æ)-5a (66)
(Æ)-5b (60)
(Æ)-5c (65)
(Æ)-5d (37)
1.9Æ0.1
15Æ1
2.9Æ0.3
15Æ2
2.4Æ0.1
0.46Æ0.02
0.48Æ0.08
4.3Æ0.1
4.0Æ0.4
1.3Æ0.1
0.93Æ0.02
4.1Æ0.1
1.5Æ0.1
2.7Æ0.1
0.055Æ0.001
0.047Æ0.004
3.1Æ0.2
2.7Æ0.1
9
0.010Æ0.001
0.0037Æ0.0004
0.65Æ0.03
0.016Æ0.001
10
11
12
6b
6c
6d
>85
[a] PMP=p-methoxyphenyl;
CAN=cerium
ammonium
nitrate;
[a] All limits relate to IC50 values>200 mm. Data with standard errors
were calculated from duplicate experiments by using five different inhibi-
tor concentrations. The Ki values were calculated from the corresponding
IC50 and Km values and the substrate concentration, [S], by using the
equation Ki =IC50/(1+[S]/Km).
DMAP=4-dimethylaminopyridine; TBAF=n-tetrabutyl ammonium
fluoride; DIC=N,N’-diisopropylcarbodiimide; HOBt=1-hydroxybenzo-
triazole.
The synthesis of the target dipeptide-based nitriles con-
taining the gem-difluorocycloalkyl moiety started from unsa-
turated cyclic a-amino ester derivatives 2 (Table 1). With
the aim of obtaining enantiomerically pure compounds, (R)
or (S)-phenylglycinol methyl ether was employed as a chiral
auxiliary, and p-methoxyphenyl (PMP) was chosen as pro-
tecting group for the racemic derivatives. Chiral compounds
2a–2d were deprotected with Pd(OH)2 in ethanol (for the
six-membered derivatives 2a and 2b, n=1) or ethyl acetate
(for the five-membered compounds 2c and 2d, n=0) under
a hydrogen atmosphere. The reaction proceeded smoothly,
with the simultaneous reduction of the double bond to give
the free amines 3a and 3b (Table 1). On the other hand, the
racemic substrates 2e and 2 f (n=1, 0) were subjected to
catalytic hydrogenation followed by treatment with cerium
ammonium nitrate (CAN) to remove the PMP group. The
amines 3 were coupled with two different acid chlorides to
Next, the inhibitory activity of the 16 dipeptide nitriles
was determined towards the following human enzymes,
Cat B, Cat K, Cat L, and Cat S, by using chromogenic or flu-
orogenic peptide substrates (Table 2).[13] All the dipeptide
nitriles were found to inhibit cathepsins B and K, however,
they were more or less inactive against cathepsins L and S.
Cathepsins L and S prefer large aromatic amino acids in the
P2 position; Cat S also accommodates an extended aliphatic
amino acid such as isobutylsulfonylcysteine.[6]
Thus, our data are in agreement with results of a previous
series of dipeptide nitriles with homocycloleucine in the P2
position.[8] These compounds showed a better selectivity for
Cat K over Cat S and Cat L, than over Cat B. Nearly all of
the compounds of type 5 and 6 were more active against
Cat K than Cat B. Significantly different inhibitory potency
was found for the corresponding homocycloleucine enantio-
mers, with the (R)-enantiomers being more active against
Chem. Eur. J. 2011, 17, 5256 – 5260
ꢅ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5257

M. Gꢄtschow, S. Fustero, J. Jimꢃnez-Barbero et al.
Cat B and K. This tendency was also found for the corre-
sponding cyclopentane enantiomers, but less pronounced.
The cyclohexane (R)-enantiomers were two orders of mag-
nitude more active towards Cat K than the cyclopentane
(R)-enantiomers ((R)-5a vs. (R)-5c; (R)-5b vs. (R)-5d). Ten-
fold lower Ki values on Cat B were obtained for the same
eutomers. Also, the non-fluorinated, non-chiral dipeptide ni-
triles had a stronger affinity to Cat K when they contained a
six-membered ring (6a vs. 6c; 6b vs. 6d), as already report-
ed for (Z)-homocycloleucyl-glycine-nitrile.[8] The N-terminal
acyl group present in compounds 5 and 6 did not have a
clear impact on enzyme inhibition.
As already mentioned for Cat B and K inhibition, the
enantiomers showed a clearly different activity. In seven out
of eight cases, the racemate ((Æ)-5a–5d on Cat B; (Æ)-5a–
5c on Cat K) was somewhat less potent than the corre-
sponding eutomer, as to be predicted. The most potent
Cat B inhibitor identified in this study was the (R)-config-
ured homocycloleucyl derivative with a 4-bromobenzoyl
moiety ((R)-5a), which exhibited a Ki value of 250 nm. This
compound was selected for a detailed kinetic study to eluci-
date the type of inhibition. The analysis of the reactions per-
formed with different substrate concentrations indicated
competitive inhibition (see the Supporting Information, Fig-
ure S2). This behavior is in agreement with the active-site-
directed interaction of dipeptide nitriles with cysteine pro-
teases, leading to the reversible formation of thioimidate ad-
ducts.[6]
We next addressed the question whether a hydrogen–fluo-
rine exchange was advantageous for enzyme inhibition. In a
previous study on selective peptidomimetic inhibitors of
Cat K, fluorine was introduced into the P2 leucine moiety.
The resulting three inhibitors, including odanacatib, retained
the very strong activity of the parent leucine derivative
toward Cat K. Odanacatib, but not the two other fluorinated
analogues, also showed a somewhat improved affinity for
Cat B.[9] In our series of inhibitors, a hydrogen–fluorine ex-
change was not advantageous for Cat K inhibition. The non-
fluorinated, non-chiral derivatives were more potent than
the corresponding fluorinated eutomers, and the Ki values
increased between 3.6-fold to 170-fold. Thus, the b,b-di-
fluorination of cycloaliphatic amino acids in the P2 position
of dipeptide nitriles adversely affects Cat K inhibition. In
contrast, the three most potent fluorinated eutomers dis-
played lower Ki values towards Cat B when compared with
the corresponding non-fluorinated analogues ((R)-5a vs. 6a;
(R)-5b vs. 6b; (R)-5c vs. 6c). Moreover, the four distomers
were less active than the non-fluorinated compounds ((S)-
5a vs. 6a; (S)-5b vs. 6b; (S)-5c vs. 6c; (S)-5d vs. 6d). There-
fore, the introduction of fluorine affects the affinity of the
inhibitor for Cat B depending on the face of the aliphatic
ring where the hydrogen–fluorine exchange occurred.
were manually constructed and the resulting thioimidate
complexes were energy minimized.[13] Evaluation of the re-
sulting poses showed that, in all cases, the cycloaliphatic
group resides in the S2 pocket, which is mainly formed by
the residues Ala173, Ala200, Tyr75, and Pro76. As has been
observed for the related homocycloleucine derivative 1,[8]
the backbone of the six inhibitors forms a hydrogen-bonding
2
1
À
network with the backbone of the protein (NH of P
P
2
1
À
amide with CO of Gly198, CO of P P amide with NH of
3
2
À
Gly74 and NH of P P amide with CO of Gly74). In addi-
tion, the bromophenyl group occupies the S3 pocket forming
an edge-to-face interaction with Tyr75. The thioimidate NH
is located within a hydrogen-bonding distance (3.4 ꢆ) to the
side chain CO of Gln23. On the one hand, in the case of the
(R)-configured eutomers, (R)-5a and (R)-5c, the fluorinated
face is directed to the S2 pocket, whereas the non-fluorinat-
ed face is solvent exposed (Figure 1). On the other hand the
fluorine atoms of the (S)-configured distomers, (S)-5a and
(S)-5c, are directed to the solvent (Figure 1).
From the atom-based scoring, it was observed that the
overall predicted interaction energies with Cat B were
slightly lower for the (R)-stereoisomers than for the (S)-ste-
reoisomers and the non-fluorinated analogues. Direct corre-
lation of docking energy scores with the experimentally ob-
tained activity is not possible because the energy values are
not actual binding free-energies. However, in our case, the
compounds only differ by the presence of fluorine and its
position. Therefore, the energy score differences are as-
sumed to reflect the different contribution of the fluorine
atoms at the corresponding face to the protein–ligand inter-
action. In the case of the eutomers (R)-5a and (R)-5c, the
lower predicted energy scores can be rationalized by the
suggested binding mode, in which the fluorine atoms are
deeply buried in the hydrophobic S2 pocket, thus providing
additional sources for binding affinity.
Protein-bound ligand conformations can be determined
by detecting transferred NOEs (trNOEs) on the basis of the
shorter correlation time of small free ligands than that of re-
ceptor-bound ligands. Free ligands provide positive NOEs,
whereas bound ligands give negative cross-peaks. We per-
1
formed H NMR trNOE experiments for the complexes of
(S)-5a and (R)-5a with Cat B.[13] In the bound state, nega-
tive NOE cross-peaks were observed for both derivatives in
the presence of the enzyme at a 20:1 ligand/receptor molar
ratio. In contrast, in the free state, NOE cross-peaks were
exclusively positive, which indicates binding of the ligands
to the enzyme. The analysis of the cross-peaks revealed basi-
cally only the trivial cross-peaks, and no inter-residual cross-
peaks (see the Supporting Information, Figure S5). No
peaks were evidenced between protons belonging to the ar-
omatic and aliphatic moieties, or between any of these two
rings with the CH2CN protons. These results let us discard
the possibility that a folded conformer binds to the enzyme.
The bound conformer, in both cases, is expected to adopt an
extended conformation as evidenced by the docking results.
Key information on the binding event can also be de-
duced from a saturation transfer difference (STD) experi-
To elucidate the orientation of the fluorinated face rela-
tive to the protein structure, we performed molecular mod-
eling studies of bromophenyl representatives in the active
site of Cat B (see the Supporting Information). Using a co-
valent docking procedure, the corresponding thioimidates
5258
www.chemeurj.org
ꢅ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5256 – 5260

Design, Synthesis, and Biological Evaluation of Cathepsin Inhibitors
COMMUNICATION
(S)-5a and (R)-5a. Indeed, at
all saturation times, the transfer
of saturation from the protein
to the cyclohexyl ring was sig-
nificantly higher for (S)-5a
(9% in total, with a saturation
time of 2 s) than for (R)-5a
(6% in total, with the same sat-
uration time). For (S)-5c, the
pattern was fairly similar to
that of (S)-5a. Notably, accord-
ing to the best poses of the
docking analysis, for the (R)-
isomer (R)-5a the two fluorine
atoms are directly pointing to-
wards the S2 pocket of Cat B,
whereas for the (S)-isomer (S)-
5a, the corresponding positions
are taken by two hydrogen
atoms, which are now in more
intimate contact. Thus, the ex-
perimental observations of the
STD experiments, with some-
how larger STD intensities for
the aliphatic protons in (S)-5a
compared with (R)-5a, are in
agreement with the docking re-
sults.
In summary, we have devel-
oped a synthetic access to new
dipeptide nitriles with a b,b-di-
fluoro 1-amino-1-cyclopentane-
and cyclohexanecarboxylic acid
unit. The fluorinated enantio-
pure compounds 5a–5d are not
only potent dual inhibitors of
Cat B and K, but also attractive
Figure 1. Covalent docking of Cat B-inhibitor complexes: top; modeled complex of (R)-5a inside the active
site of Cat B; the fluorine atoms are within van der Waals interaction distances of His199 Ca, Ala200 Cb,
Pro76 Cg, Cd, and Trp30 Cb; bottom; modeled complex of (S)-5a inside the active site of Cat B.
ment.[13] As a consequence of the stimulated protein–ligand
intermolecular saturation-transfer process, the ligand satura-
tion is higher for those protons that are in closer contact
tools to investigate enzyme–ligand interactions. A covalent
docking protocol was applied to elucidate the binding mode
towards Cat B, which was supported by trNOE and STD
NMR spectroscopic experiments. The gem-difluoro moiety
of (R)-configured inhibitors is directed towards the S2
pocket of Cat B. Kinetic measurements revealed that the
(R)-configured inhibitors are significantly more potent than
the (S)-configured stereoisomers. Therefore, a fluorophilic
nature of the S2 pocket of Cat B could be deduced.
1
with the receptor. The STD H NMR spectra[14] of (S)-5a,
(R)-5a, and (S)-5c in the presence of Cat B were recorded
(see the Supporting Information, Figure S6). Two different
molar ratios between ligand and receptor (10:1 and 20:1)
and four different saturation times, between 500 ms and 2 s,
were employed. In all cases, major saturation transfer to the
aromatic protons of the p-bromobenzoyl moiety were ob-
served, and additional transfer to the aliphatic protons of
the cyclohexane ((S)-5a, (R)-6a) or cyclopentane ((S)-5c)
rings, with basically no transfer to the CH2CN protons.
Thus, the recognition epitope mainly involves the aromatic
and aliphatic rings, with slightly different features. The STD
intensities at the aromatic protons, larger by approximately
20% at each meta- and ortho-positions, reflect the accom-
modation of the p-bromophenyl ring in the P3 pocket of
Cat B, as discussed above. Careful integration of the STD
signals revealed minor differences between the enantiomers
Acknowledgements
This work was supported by a grant from the Ministry of Science and In-
novation of Spain (CTQ2007-61462/BQU, CTQ2006-10874-C02-01/BQU,
CTQ2009-08536, CTQ2010-19774-C02, and GV/PROMETEO/2010/061).
V.R. expresses their thanks for a predoctoral fellowship. C.d.P. and
M.S.R thank the same institution for a Ramꢂn y Cajal and a Juan de la
Cierva contract, respectively. M.F. is supported by a fellowship from the
NRW Forschungsschule Biotech-Pharma, and M.T.S. by a fellowship
Chem. Eur. J. 2011, 17, 5256 – 5260
ꢅ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5259

M. Gꢄtschow, S. Fustero, J. Jimꢃnez-Barbero et al.
from the Graduiertenkolleg (GRK) 677 of the Deutsche Forschungsge-
meinschaft.
Venuti, G. Wesolowski, R. M. Young, R. Zambonia, W. C. Black,
Bioorg. Med. Chem. Lett. 2008, 18, 923–928; Riendeau, J. Robi-
chaud, G. A. Rodan, S. B. Rodan, C. Seto, M. Thꢃrien, V. L. Truong,
M. C. Venuti, G. Wesolowski, R. M. Young, R. Zambonia, W. C.
Black, Bioorg. Med. Chem. Lett. 2008, 18, 923–928.
Keywords: fluorinated ligands · hydrolases · inhibitors ·
peptides · stereoselectivity
[10] a) K. Mꢄller, C. Faeh, F. Diederich, Science 2007, 317, 1881–1886;
b) J. A. Olsen, D. W. Banner, P. Seiler, U. Obst Sander, A. D’Arcy,
M. Stihle, K. Mꢄller, F. Diederich, Angew. Chem. 2003, 115, 2611–
2615; Angew. Chem. Int. Ed. 2003, 42, 2507–2511; c) R. Lçser, K.
Schilling, E. Dimmig, M. Gꢄtschow, J. Med. Chem. 2005, 48, 7688–
7707; d) A. K. Podichetty, S. Wagner, S. Schrçer, A. Faust, M. Schꢇ-
fers, O. Schober, K. Kopka, G. Haufe, J. Med. Chem. 2009, 52,
3484–3495.
[11] a) S. Fustero, M. Sꢁnchez-Rosellꢂ, V. Rodrigo, J. F. Sanz-Cervera, J.
Piera, A. Simꢂn-Fuentes, C. del Pozo, Chem. Eur. J. 2008, 14, 7019–
7029; b) S. Fustero, V. Rodrigo, M. Sꢁnchez-Rosellꢂ, F. Mojarrad,
A. Ricedo, T. Moscardꢂ, C. del Pozo, J. Fluorine Chem. 2008, 129,
943–950; c) S. Fustero, M. Sꢁnchez-Rosellꢂ, V. Rodrigo, C. del Pozo,
J. F. Sanz-Cervera, A. Simꢂn, Org. Lett. 2006, 8, 4129–4132.
[12] S. Fustero, M. Sꢁnchez-Rosellꢂ, V. Rodrigo, A. Garcꢀa, S. Catalꢁn,
C. del Pozo, J. Org. Chem. 2008, 73, 5617–5620.
[13] The synthesis and analytical characterization of the compounds, ca-
thepsin inhibition assays, details of molecular modeling, and NMR
spectroscopy experiments are given in the Supporting Information.
[14] a) J. Jimꢃnez-Barbero, E. Dragoni, C. Venturi, F. Nannucci, A.
Ardꢁ, M. Fontanella, S. Andrꢃ, F. J. CaÇada, H. J. Gabius, C. Nativi,
Chem. Eur. J. 2009, 15, 10423–10431; b) M. Kolympadi, M. Fonta-
nella, C. Venturi, S. Andrꢃ, H. J. Gabius, J. Jimꢃnez-Barbero, P.
Vogel, Chem. Eur. J. 2009, 15, 2861–2873.
[1] O. Vasiljeva, T. Reinheckel, C. Peters, D. Turk, V. Turk, B. Turk,
Curr. Pharm. Des. 2007, 13, 387–403.
[2] M. M. Mohamed, B. F. Sloane, Nat. Rev. Cancer 2006, 6, 764–775.
[3] M. R. Pratt, M. D. Sekedat, K. P. Chiang, T. W. Muir, Chem. Biol.
2009, 16, 1001–1012.
[4] V. Y. Hook, M. Kindy, G. Hook, J. Biol. Chem. 2008, 283, 7745–
7753.
[5] W. C. Black, Curr. Top. Med. Chem. 2010, 10, 745–751.
[6] M. Frizler, M. Stirnberg, M. T. Sisay, M. Gꢄtschow, Curr. Top. Med.
Chem. 2010, 10, 294–322.
[7] R. Lçser, M. Frizler, K. Schilling, M. Gꢄtschow, Angew. Chem. 2008,
120, 4403–4406; Angew. Chem. Int. Ed. 2008, 47, 4331–4334.
[8] J. T. Palmer, C. Bryant, D. X. Wang, D. E. Davis, E. L. Setti, R. M.
Rydzewski, S. Venkatraman, Z. Q. Tian, L. C. Burrill, R. V. Men-
donca, E. Springman, J. McCarter, T. Chung, H. Cheung, J. W. Janc,
M. McGrath, J. R. Somoza, P. Enriquez, Z. W. Yu, R. M. Strickley,
L. Liu, M. C. Venuti, M. D. Percival, J. P. Falgueyret, P. Prasit, R.
Oballa, D. Riendeau, R. N. Young, G. Wesolowski, S. B. Rodan, C.
Johnson, D. B. Kimmel, G. Rodan, J. Med. Chem. 2005, 48, 7520–
7534.
[9] J. G. Gauthier, N. Chauret, W. Cromlish, S. Desmarais, L. T. Duong,
J. P. Falgueyret, D. B. Kimmel, S. Lamontagne, S. Lꢃger, T. LeRiche,
C. S. Li, F. Massꢃ, D. J. McKay, D. A. Nicoll-Griffith, R. M. Oballa,
J. T. Palmer, M. D. Percival, D. Riendeau, J. Robichaud, G. A.
Rodan, S. B. Rodan, C. Seto, M. Thꢃrien, V. L. Truong, M. C.
Received: January 12, 2011
Published online: March 30, 2011
5260
www.chemeurj.org
ꢅ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5256 – 5260