Selective activity-based probes for cysteine cathepsins

Article abstract of DOI:10.1002/anie.200702811

(Chemical Equation Presented) By "reverse design": The core structures of protease inhibitors, whose selectivity has been optimized by extensive medicinal chemistry, have been redesigned into selective protease substrates by replacing the reactive electro

Full text of DOI:10.1002/anie.200702811

Communications
DOI: 10.1002/anie.200702811
Fluorescence Probes
Selective Activity-Based Probes for Cysteine Cathepsins**
Anja Watzke, Gregor Kosec, Maik Kindermann, Volker Jeske, Hans-Peter Nestler, Vito Turk,
Boris Turk,*and K. Ulrich Wendt*
Proteases are critically involved in a multitude of vital
processes, essential for cellular signaling and tissue homeo-
stasis.[1–3] Since proteolytic activity has to remain under
stringent control, most of the proteases are synthesized as
inactive precursors (zymogens), which are activated by
limited proteolysis. Consequently, most of the classical
biochemical and proteomics methods fail to discriminate
between the expression level and proteolytic activity of a
given protease in vivo. Therefore, the development of
chemical tools for biomolecular imaging of protease activity
rather than protease expression has gained considerable
interest in recent years, and a series of internally quenched
activity-based probes (ABPs)for proteases has been devel-
oped.[4,5] These probes become fluorescent upon proteolytic
cleavage.
Analysis of the function of a distinct protease in the
complex milieu of the cell requires the development of
selective and cell-permeable ABPs, which has been challeng-
ing. To present, internally quenched imaging probes for
proteolytic enzymes have been derived from peptidic sub-
strates, coupled to appropriate reporter groups which are
spatially separated upon proteolytic cleavage of the cognate
sequence.[4] The preferred substrates are commonly identified
from positional scanning combinatorial libraries.[6] The syn-
thesis and scanning of such libraries can be time consuming,
and the peptidic nature of the resulting substrates may result
in unspecific cleavage by other proteases when the probes are
applied in cellular or whole animal imaging experiments. For
proteases from the cysteine cathepsin family, cell-permeable
activity-based probes have been designed as suicide sub-
strates which label the enzyme covalently and inactivate it
upon binding.[5,7] While such probes are powerful tools for the
localization of proteases, the sensitive real-time monitoring of
protease activity and inhibition in vivo would likely benefit
from probes that are substrates to the enzyme and amplify the
fluorescent signal upon proteolysis. Such probes are of high
interest for drug discovery and diagnostics since they could
enable direct observation of inhibitor efficacy by competition
studies in pharmacologically relevant models.[4d–f]
We report here a new concept for the design of specific,
stable, and cell-permeable ABPs. The probes that are
described herein are substrates to the target enzyme and
selectively differentiate between related cysteine cathepsins.
Cysteine cathepsins have been established as pharmaceutical
targets for the treatment of bone disorders and degenerative
joint diseases (cathepsin K)[8,9] as well as immune disorders
(cathepsin S).[10,11] Notably, the design of our probes is not
based on peptide substrate information gained through
positional scanning combinatorial libraries. We base the
design on the structures of highly potent and selective
cathepsin inhibitors which have undergone extensive medic-
inal chemistry optimization and thus display favorable
pharmacokinetic properties in addition to potency and
selectivity against related proteases.[1,11–14] These inhibitors
commonly carry a reactive group (often a nitrile group)as an
electrophilic mimic of the substrateꢀs scissile bond. Once
bound to a cathepsin, the electrophilic group is attacked by
the catalytic cysteine residue of the enzyme, resulting in a
covalent protein–inhibitor complex.[11–14] Since the catalytic
mechanism is conserved within the cysteine cathepsin family,
the selectivity of such inhibitors is most likely derived from
their chemical scaffold. Consequently, we hypothesized that
the replacement of the electrophilic group of such inhibitors
with a cleavable peptide bond should result in a nonsuicide
substrate with a selectivity pattern related to the original
inhibitor. Subsequent derivatization with appropriate
reporter groups would result in a selective activity-based
imaging probe (Scheme 1). Starting from optimized inhibitors
rather than peptide substrates, this concept allows one to
access a significant knowledge base of existing structure–
activity relationships and utilize it for the design of new
selective ABPs.
[*] Dr. G. Kosec,[+] Prof. V. Turk, Prof. B. Turk
ˇ
Jozef Stefan Institute
Department of Biochemistry and Molecular Biology
Jamova 39, 1000 Ljubljana (Slovenia)
Fax: (+386)1477-3984
E-mail: Boris.Turk@ijs.si
Dr. A. Watzke,[+] Dr. M. Kindermann, V. Jeske, Dr. H.-P. Nestler,
Dr. K. U. Wendt
Sanofi-Aventis Deutschland GmbH
SMA Chemical and Analytical Sciences
IndustrieparkParkHöchst, 65926 Frankfurt (Germany)
Fax: (+49)69-305-16189
Several selective inhibitors for cathepsin K have entered
clinical trials.[13] To demonstrate our concept we chose nitrile
inhibitor 1 to provide us with the basic scaffold for an ABP for
cathepsin K (Scheme 1a). Compound 1 has a high affinity to
cathepsin K (IC50 = 0.5 nm)while exhibiting good selectivity
against the related cathepsins B (IC50 = 6200 nm)and S
(IC50 = 2200 nm).[13] Inhibitor 3 has been reported as specific
for cathepsin S and was therefore selected as a promising
E-mail: ulrich.wendt@sanofi-aventis.com
[+] These authors contributed equally.
[**] This workwas supported by the European Commission Framework
VI Program (CAMP project, LSHG-2006-018830). We thankProf.
Markus Grütter and Dr. Marc Nazare for critical comments on this
work.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
406
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 406 –409

Angewandte
Chemie
While cathepsin K can be secreted to the extracellular
medium,[10] cathepsin S has been found to be predominantly
localized in lysosomes where it is critical for antigen process-
ing and presentation.[10,11] In vivo imaging of cathepsin S
activity therefore requires cell-permeable probes with spec-
troscopic properties compatible with cellular imaging. To
match these requirements, the fluorophore/quencher pair of 4
was exchanged to Bodipy TMR-X and QSY-7, resulting in an
internally quenched red-shifted fluorescent probe 5. Sub-
strate 5 was synthesized by using a combination of solid- and
solution-phase chemistry.[16]
Enzymatic turnover and selectivity of ABPs 2 and 4 were
first tested in vitro with the related cysteine cathepsins S, K, B
and L.[17] Probe 2 was selectively cleaved by cathepsin K as
shown in Figure 1a. Within the detection limit of the assay,
absolute specificity against cathepsin S was observed. Like-
wise, ABP 4 was cleaved by cathepsins S and L, whereas no
increase of fluorescence was obtained with the related
Scheme 1. Converting of cathepsin inhibitors into activity-based probes
(ABPs). a) Selective cathepsin K inhibitor 1[13] and related ABP 2. In 2
the electrophilic nitrile group is converted into a peptide bond.
Fluorophore 5-((2-aminoethyl)amino)naphthalene-1-sulfonic acid and
quencher 4-{[4-(dimethylamino)-phenyl] azo}benzoic acid are attached
in the P1 and P1’ position, respectively. b) Selective cathepsin S
inhibitor 3[11] and corresponding ABP 4. ABP 5 is a corresponding red
fluorescent probe for in vivo studies containing Bodipy TMR-X and
QSY-7 as fluorophore/quencher pair.
starting point to build an ABP directed towards this enzyme
(Scheme 1b).[11]
For the design of the desired ABPs for cathepsins K and S,
we retained the core structures of inhibitors 1 and 3,
respectively. The electrophilic groups were converted into
cleavable peptide bonds (Scheme 1). Cocrystal structures of
cathepsins K and S inhibitors indicate that the P1 and P1’
moieties are openly accessible from the surrounding sol-
vent.[15] We therefore decided to attach the spatially demand-
ing fluorophore and quencher groups in these positions. The
cleavable peptide bond between the P1’ lysine residue
functionalized with the corresponding quencher (dabcyl)
and the glutamic acid carrying the fluorophore (Edans)in
P1 should be precisely oriented in the active site to be cleaved
by cathepsin K (probe 2)or cathepsin S (probe 4), respec-
tively. Substrates 2 and 4 were synthesized by standard solid-
phase peptide synthesis.[16]
Figure 1. Enyzmatic turnover of probes 2, 4, and 5 with the cysteine
[17]
~
&
^
*
^
cathepsins S ( ), K ( ), B( ), L ( ), and papain ( ); c=10 nm.
a) Progress curve for probe 2 (excitation wavelength lex =336 nm,
emission wavelength lem =468 nm); b) Progress curve for probe 4
(lex =336 nm, lem =468 nm); c) Progress curve for probe 5
(lex =544 nm, lem =573 nm).
Angew. Chem. Int. Ed. 2008, 47, 406 –409
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
407

Communications
cathepsins K and B under the conditions used (Figure 1b),
In summary, compounds 2, 4, and 5 demonstrate that
desirable properties of mechanism-based inhibitors for cys-
teine cathepsins can be transferred to cell-permeable and
selective ABPs when the electrophilic group is exchanged for
a cleavable peptide bond. The substrate nature and cell
permeability of the newly designed probes allows one to
monitor the activity of selected proteases in living cells. By
transforming an optimized inhibitor to a substrate ABP, we
have taken the reverse pathway compared to that which
protease drug discovery commonly
thus reflecting the selectivity of the underlying inhibitors, on
which these ABPs are based structurally.[11] Both probes are
stable in human plasma, and the turnover numbers are
comparable with commercially available substrates when
incubated with their target enzyme (Table 1). Corresponding
to the affinities of the underlying inhibitors 1 and 3, ABPs 2
and 4 exhibit low KM values of 35 mm and 1.3 mm for their
targets cathepsin K and S, respectively.
Table 1: Kinetic parameters of probes 2, 4, and 5 with different cathepsins.[17][a]
applies when transforming prefer-
red substrates into inhibitors, which
Cath S
Cath K
Cath B
Cath L
Papain
are subsequently optimized by the
means of medicinal chemistry. This
concept of “reverse design” retains
optimized properties of the inhib-
itor in the ABP and may provide a
general route towards selective
probes for cysteine cathepsins and
other cysteine proteases. “Reverse
design” makes a significant reser-
voir of existing structure–activity
relationships accessible for the
design of new ABPs. Notably, with
this concept not only the selectivity
but also the KM value of the ABP
should be tunable on the basis of
the properties of the parent inhib-
K
M [m]
2
2
2
n.d.
n.d.
n.d.
3.5”10À5
7.3
5420
8.8”10À4
2.6
283
7.1”10À3
9.7
424
6.3”10À6
0.04
296
vmax [mmolsÀ1
]
]
k
cat/KM [mÀ1 sÀ1
]
]
]
K
M [m]
4
4
4
1.3”10À6
0.5
15100
8.9”10À5
0.6
187
3.2”10À3
2.7
332
8.4”10À4
19.2
7070
1.6”10À5
0.1
581
vmax [mmolsÀ1
kcat/KM [mÀ1 sÀ1
K
M [m]
vmax [mmolsÀ1
cat/KM [mÀ1 sÀ1
5
5
5
8.8”10À6
0.35
3930
5.6”10À4
3.3
569
n.d.
n.d.
n.d.
3.5”10À6
0.3
7930
1.4”10À4
0.9
680
]
k
[a] Kinetic efficiencies of probes 2, 4, and 5. Probes 2 and 4 exhibit good selectivity amongst the related
cysteine cathepsins S, K, B, L, and papain (n.d.=no turnover detectable). Probe 5 is selective for
cathepsins S and L. All errors are within 10%.
ABP 5 contains a fluorophore/quencher pair with a
spectral range compatible with cellular imaging. HaCaT
cells (spontaneously immortalized human keratinocytes)
were incubated with a 8 mm solution of 5. Consistent with
the reported lysosomal expression of cathepsins S and L in
this cell line,[18] we observed enrichment of fluorescence in a
lysosome-compatible staining pattern (Figure 2a). However,
when HaCaT cells were preincubated with either 15 mm E-64d
(a broad-spectrum cysteine cathepsin inhibitor)or 15 mm of
an inhibitor selective for cathepsin S and L,[19] the fluores-
cence was almost completely supressed (Figure 2b,c), sug-
gesting that the fluorescence signal results from the cathe-
psin S and/or cathepsin L. Similar results were obtained in the
U937 human monocyte cell line (data not shown). These
results demonstrate that ABP 5, which was obtained by
redesigning a selective cathepsin inhibitor into a cleavable
substrate, allows one to selectively monitor the activity of two
cysteine cathepsins in living cells.
itor. The low KM values of probes 2, 4, and 5 render them
attractive for in vivo studies.
Received: June 25, 2007
Revised: September 19, 2007
Published online: November 19, 2007
Keywords: enzymes · fluorescent probes · inhibitors ·
.
medicinal chemistry
[1] B. Turk, Nat. Rev. Drug Discovery 2006, 5, 785 – 799.
[2] X. S. Puente, L. M Sanchez, C. M. Overall, C. Lopez-Otin, Nat.
Rev. Genet. 2003, 4, 544 – 558.
[3] O. Vasiljeva, T. Reinheckel, C. Peters, D. Turk, V. Turk, B. Turk,
Curr. Pharm. Des. 2007, 13, 385 – 401.
[4] a)M. Fonovic, M. Bogyo, Curr. Pharm. Des. 2007, 13, 253 – 261;
b)J. Rao, A. Dragulescu-Andrasi, H. Yao, Curr. Opin. Biotech-
nol. 2007, 18, 17 – 25; c)L. Shi, V. De Paoli, N. Rosenzweig, Z.
Rosenzweig, J. Am. Chem. Soc. 2006, 128, 10378 – 10379; d)R.
Weissleder, C. H. Tung, U. Mahmood, A. Bogdanov, Nat.
Biotechnol. 1999, 17, 375 – 378; e)B. Law, A. Curino, T. H.
Bugge, R. Weissleder, C. H. Tung, Chem. Biol. 2004, 11, 99 – 106;
f)M. Rudin, R. Weissleder, Nat. Rev. Drug Discovery 2003, 2,
123 – 131.
[5] G. Blum, S. R. Mullins, K. Keren, M. Fonovic, C. Jedeszko, M. J.
Rice, B. F. Sloane, M. Bogyo, Nat. Chem. Biol. 2005, 1, 203 – 209.
[6] a)M. Meldal, QSAR Comb. Sci. 2005, 24, 1141 – 1148; b)B. J.
Backes, J. L. Harris, F. Leonetti, C. S. Craik, J. A. Ellman, Nat.
Biotechnol. 2000, 18, 187 – 193; Y. Choe, F. Leonetti, D. C.
Greenbaum, F. Lecaille, M. Bogyo, D. Brömme, J. A. Ellman,
C. S. Craik, J. Biol. Chem. 2006, 281, 12824 – 12832.
Figure 2. Monitoring of cathepsin activity in HaCaT cell line. a) After
3 h incubation with 8 mm of 5, human keratinocytes exhibit a lyso-
some-compatible fluorescence pattern. b) When preincubated with
15 mm E64d, no fluorescence is observed. c) When preincubated with
15 mm selective cathepsin S/L inhibitor,[19] no fluorescence is observed.
[7] M. J. Evans, B. F. Cravatt, Chem. Rev. 2006, 106, 3279 – 3301.
408
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 406 –409

Angewandte
Chemie
[8] P. Saftig, E. Hunziker, O. Wehmeyer, S. Jones, A. Boyde, W.
Rommerkirch, J. D. Moritz, P. Schu, K. von Figura, Proc. Natl.
Acad. Sci. USA 1998, 95, 13453 – 13458.
[9] V. Turk, B. Turk, D. Turk, EMBO J. 2001, 20, 4629 – 4633.
[10] R. J. Riese, P. R. Wolf, D. Bromme, L. R. Natkin, J. A. Villan-
dangos, H. L. Ploegh, H. A. Chapman, Immunity 1996, 4, 357 –
366.
Catalano, D. N. Deaton, A. M. Hassell, C. H. Jurgensen, S. T.
Long, R. B. McFadyen, A. B. Miller, L. R. Miller, J. A. Payne,
J. A. Ray, V. Samano, L. M. Shewchuk, F. X. Tavares, K. J. Wells-
Knecht, D. H. Willard, L. L. Wright, H. Q. Zhou, Bioorg. Med.
Chem. Lett. 2005, 15, 3540 – 3546.
[16] ABPs 2 and 4 were synthesized by standard solid-phase peptide
synthesis using the 2-chlorotrityl chloride resin. ABP 5 was
generated using a combination of solid- and solution-phase
chemistry. For details, see the Supporting Information.
[11] J. Link, S. Zipfel, Curr. Opin. Drug Discov. Devel. 2006, 9, 471 –
482.
[12] J. C. Powers, J. L. Asgian, O. D. Ekici, K. E. James, Chem. Rev.
2002, 102, 4639 – 4750.
[13] U. B. Grabbowska, T. J. Chambers, M. Shiroo, Curr. Opin. Drug
Discov. Devel. 2005, 8, 619 – 630.
[17] For the in vitro enzyme activity assay, the active cysteine
proteases were dissolved in AHNP buffer (150 mm acetate/2-[4-
(2-hydroxyethyl)-1-piperazinyl]ethanesulfonic acid (HEPES)
pH 6.5, 300 mm NaCl; 0.001% pluronic, 5–100 mm cysteine
depending on the enzyme)at a final concentration of 10 n m. The
substrates 2, 4, and 5, dissolved in DMSO, were added at
concentrations of 98 mm (2)and 24 mm (4 and 5), and fluores-
cence was measured with a Tecan SAFIRE II spectrometer. The
final DMSO concentration in the assay did not exceed 1% (v/v).
Steady-state kinetics were fitted by nonlinear least-squares
regression using v = [A]V/(KM+(1+([A]/Ksi))[A]), where v is
the initial velocity, V is the maximal rate, KM is the Michaelis–
Menten constant, and Ksi is the constant for substrate inhibition.
[18] G. Schwarz, W. H. Boehncke, M. Braun, C. J. Schröter, T.
Burster, T. Flad, D. Dressel, E. Weber, H. Schmid, H. Kalbacher,
J. Invest. Dermatol. 2002, 119, 44 – 49.
[14] a)A. W. Patterson, W. J. L. Wood, M. Hornsby, S. Lesley, G.
Spraggon, J. A. Ellman, J. Med. Chem. 2006, 49, 6298 – 6307;
b)Y. D. Ward, D. S. Thomson, L. L. Frye, C. L. Cywin, T.
Morwick, M. J. Emmanuel, R. Zidell, D. McNeil, Y. Bekkali,
M. Giradot, M. Hrapchak, M. DeTuri, K. Crane, D. White, S.
Pav, Y. Wang, M. H. Hao, C. A. Grygon, M. E. Labadia, D. M.
Freeman, W. Davidson, J. L. Hopkins, M. L. Brown, D. M.
Spero, J. Med. Chem. 2002, 45, 5471 – 5482; c)T. A. Pauly, T.
Sulea, M. Ammirati, J. Sivaraman, D. E. Danley, M. C. Griffor,
A. V. Kamath, I. K. Wang, E. R. Laird, A. P. Seddon, R. Menard,
M. Cygler, V. L. Rath, Biochemistry 2003, 42, 3203 – 3213.
[15] a)J. G. Catalano, D. N. Deaton, E. S. Furfine, A. M. Hassell,
R. B. McFadyen, A. B. Miller, L. R. Miller, L. M. Shewchuk,
D. H. Willard, L. L. Wright, Bioorg. Med. Chem. Lett. 2004, 14,
275 – 278; b)D. G. Barrett, D. N. Deaton, A. M. Hassell, R. B.
McFadyen, A. B. Miller, L. R. Miller, J. A. Payne, L. M.
Shewchuk, D. H. Willard, L. L. Wright, Bioorg. Med. Chem.
Lett. 2005, 15, 3039 – 3043; c)D. G. Barrett, V. M. Boncek, J. G.
[19] Inhibition data for the used cathepsin S/L inhibitor: cath S:
IC50 = 5.3 nm; cath L: IC50 = 55 nm; cath K: IC50 = 1030 nm;
cath B: IC50 = 19000 nm (J. P. Falgueyret, S. Desmarais, R.
Oballa, C. Black, W. Cromlish, K. Khougaz, S. Lamontagne, F.
Masse, D. Riendeau, S. Toulmond, M. D. Percival, J. Med. Chem.
2005, 48, 7535 – 7543).
Angew. Chem. Int. Ed. 2008, 47, 406 –409
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
409