Selective imaging of cathepsin L in breast cancer by fluorescent activity-based probes

Article abstract of DOI:10.1039/c7sc04303a

Cysteine cathepsins normally function in the lysosomal degradation system where they are critical for the maintenance of cellular homeostasis and the MHC II immune response, and have been found to have major roles in several diseases and in tumor progression. Selective visualization of individual protease activity within a complex proteome is of major importance to establish their roles in both normal and tumor cells, thereby facilitating our understanding of the regulation of proteolytic networks. A generally accepted means to monitor protease activity is the use of small molecule substrates and activity-based probes. However, there are eleven human cysteine cathepsins, with a few of them displaying overlapping substrate specificity, making the development of small molecules that selectively target a single cathepsin very challenging. Here, we utilized HyCoSuL, a positional scanning substrate approach, to develop a highly-selective fluorogenic substrate and activity-based probe for monitoring cathepsin L activity in the breast cancer cell line MDA-MB-231. Use of this probe enabled us to distinguish the activity of cathepsin L from that of other cathepsins, particularly cathepsin B, which is abundant and ubiquitously expressed in normal and transformed cell types. We found that cathepsin L localization in MDA-MB-231 cells greatly overlaps with that of cathepsin B, however, several cathepsin L-rich lysosomes lacked cathepsin B activity. Overall, these studies demonstrate that HyCoSuL-derived small molecule probes are valuable tools to image cathepsin L activity in living cells. This approach thus enables evaluation of cathepsin L function in tumorigenesis and is applicable to other cysteine cathepsins.

Full text of DOI:10.1039/c7sc04303a

View Article Online
View Journal
Chemical
Science
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Poreba, W. Rut,
M. Vizovisek, K. Groborz, P. Kasperkiewicz, D. Finlay, K. Vuori, D. Turk, B. Turk, G. Salvesen and M. Drag,
Chem. Sci., 2018, DOI: 10.1039/C7SC04303A.
This is an Accepted Manuscript, which has been through the
Volume
7 Number 1 January 2016 Pages 1–812
Royal Society of Chemistry peer review process and has been
accepted for publication.
Chemical
Science
Accepted Manuscripts are published online shortly after
www.rsc.org/chemicalscience
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 2041-6539
EDGE ARTICLE
Francesco Ricci et al.
Electronic control of DNA-based nanoswitches and nanodevices
rsc.li/chemical-science

Page 1 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
Selective imaging of cathepsin L in breast cancer by fluorescent activity-based probes
Marcin Poreba1,2, Wioletta Rut1, Matej Vizovisek3, Katarzyna Groborz1, Paulina Kasperkiewicz1,2,
Darren Finlay2, Kristiina Vuori2, Dusan Turk3, Boris Turk3, Guy S. Salvesen2, Marcin Drag1
1Department of Bioorganic Chemistry, Faculty of Chemistry, Wroclaw University of Technology,
Wyb. Wyspianskiego 27, 50ꢀ370 Wroclaw, Poland
2Sanford Burnham Prebys Medical Discovery Institute, 10901 North Torrey Pines Road, La Jolla, CA
92037, USA
3Department of Biochemistry and Molecular and Structural Biology, Jožef Stefan Institute, SIꢀ1000
Ljubljana, Slovenia
Corresponding authors: marcin.poreba@pwr.edu.pl, marcin.drag@wpr.edu.pl

Chemical Science
Page 2 of 32
View Article Online
DOI: 10.1039/C7SC04303A
ABSTRACT
Cysteine cathepsins normally function in the lysosomal degradation system where they are
critical for maintenance of cellular homeostasis and MHC II immune response, and have been found to
have major roles in several diseases including tumor progression. Selective visualization of individual
protease activity within a complex proteome is of major importance to establish their roles in both
normal and tumor cells, thereby facilitating our understanding of regulation of proteolytic networks. A
generally accepted means to monitor protease activity are small molecule substrates and activityꢀbased
probes. However, there are eleven human cysteine cathepsins, with a few of them displaying
overlapping substrate specificity, making the development of small molecules selectively targeting a
single cathepsin very challenging. Here, we utilized HyCoSuL, a positional scanning substrate
approach, to develop a highlyꢀselective fluorogenic substrate and activityꢀbased probe for monitoring
cathepsin L activity in the breast cancer cell line MDAꢀMBꢀ231. Use of this probe enabled us to
distinguish activity of cathepsin L from other cathepsins, particularly cathepsin B, which is abundant
and ubiquitously expressed in normal and transformed cell types. We found that cathepsin L
localization in MDAꢀMBꢀ231 cells greatly overlaps with that of cathepsin B, however, several
cathepsin Lꢀrich lysosomes lacked cathepsin B activity. Overall, these studies demonstrate that
HyCoSuLꢀderived small molecule probes are valuable tools to image cathepsin L activity in living
cells. This approach thus enables evaluation of cathepsin L function in tumorigenesis and is applicable
to other cysteine cathepsins.

Page 3 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
1. INTRODUCTION
Cysteine cathepsins are a group of eleven structurally related proteases in humans, normally
found in lysosomes that are responsible for intracellular protein degradation, autophagy and immune
response (1). Cysteine cathepsins also have major roles in a number of diseases, including cancer,
arthritis and osteoporosis, and were therefore considered of major importance in drug development (2ꢀ
4). Cathepsin L is ubiquitously expressed and is one of the most abundant cathepsins (3). It is
synthesized as a proenzyme, which is activated in the acidic milieu of endolysosomal system
autocatalytically (5) or by other proteases (3). It is an endopeptidase with the highest catalytic activity
at pH 5.0ꢀ 6.0 (6, 7) and displays broad substrate specificity (3, 8). In addition to its role in protein
degradation, cathepsin L was found to be critical for the MHC IIꢀmediated antigen processing and
presentation in thymus (9). Furthermore, cathepsin L was found to have a major role in the nucleus by
processing the H3 histone (10) as well as the Cuxꢀ1 transcription factor (11), and has an important role
in keratinocyte differentiation (12). Similarly to several other cathepsins, cathepsin L was found to
have an important role in the progression of cancer. However, in contrast to cathepsins B and S that
are present in both stromal cells, such as macrophages, and tumor cells, cathepsin L is primarily
secreted by tumor cells (13ꢀ15).
Cathepsin L was suggested to participate in tumor cell mechanisms underlying chemoꢀ
resistance (16). Active forms of cathepsin L found in the nucleus reportedly serve as either tumor
promoters (17, 18) or suppressors (19), depending on the microenvironment. Finally, secreted
cathepsin L participates in degradation of extracellular proteins like the insulin receptor, matrix
fibronectin, elastin, and collagen (13, 20), as well as several membrane anchored cell adhesion
molecules (CAMs) and receptors (21), thereby additionally contributing to tumorigenesis.
While in mouse a single cathepsin L gene exists, in humans there are two closely related
cathepsin L orthologues, cathepsin L and cathepsin V. Both cathepsins L and V have maximum
activity in the same pH range and share 79.5% of amino acid identity (22, 23). Moreover, several
functions originally attributed to cathepsin L based on mouse studies, were found to be taken over by
cathepsin V in humans, including MHC II – mediated immune response in thymus (24). This is likely
true also for the nuclear functions of cathepsin L, as human cathepsin V, but not cathepsin L, was
found to bind DNA (25), however, a thorough validation is still lacking. Moreover, the two cathepsins
were found to have very similar substrate specificities, as revealed by the classical Positional Scanning
Synthetic Combinatorial Library (PSꢀSCL) (8). In addition, not only cathepsin V, but also other
cathepsins have similar and largely overlapping substrate specificities to cathepsin L (8), also
confirmed by several proteomic studies (26, 27). Hence, a cathepsin L selective peptide motif is still to
be discovered.

Chemical Science
Page 4 of 32
View Article Online
DOI: 10.1039/C7SC04303A
In order to be able to discriminate between cathepsin L and related cathepsins and to correctly
address their physiological functions, one has to understand their specificity (28). This is also of
crucial importance for development of selective and potent inhibitors, as well as appropriate tools for
monitoring their activity in situ. Among the latter are also selective substrates and activityꢀbased
probes (ABPs) (29ꢀ31). All early substrates aimed to target cathepsin L were unspecific and were
efficiently hydrolyzed by cathepsin B and other proteases, and were largely useful for detection of
cathepsin activity in biological samples rather than probes for cathepsin L (32, 33). Similarly, the
cresyl violet peptidic substrates, which allow monitoring of cathepsin activities in cells, are based on
nonselective peptidic motifs (34). A major step forward was application of the reverseꢀdesign
principle, which is based on the conversion of a selective medꢀchem optimized inhibitor into a
cleavable substrate by replacement of the warhead with a quencherꢀfluorophore pair. This lead to the
development of selective cathepsin S substrates (35, 36), whereas selective substrates for other
proteases are still lacking.
ABPs are mostly short peptides coupled to a reactive group (warhead) that binds to catalytic
machinery, and a detectable tag (biotin, radioisotope, or fluorophore) (37, 38). Use of ABPs in
protease and related enzyme studies was first proposed by Powers (39) and later exploited by others
(40ꢀ43). Application of ABPs to analysis of cysteine cathepsins was pioneered by the Bogyo group,
who initially designed probes based on simple broadꢀspectrum radiolabeled inhibitors (44), although
fluorophoreꢀlabeled probes (45) and quenched activityꢀbased probes have also been applied to living
cells and whole organisms (46, 47). Some successes have been achieved in the design of selective
cathepsin S probes (48, 49), but selective tools for other cathepsins remain to be developed.
Currently, the most successful approach to rapidly identify sequences with high activity and
selectivity for individual proteases is HyCoSuL, which utilizes a combination of natural and unnatural
amino acids and positional scanning library technology to explore extended chemical space around
protease active sites (40, 50, 51). We used two library scanning procedures to obtain highly active
substrates for cathepsin L. In the first, we employed a fixed P1 Arg HyCoSuL to dissect cathepsin L
specificity in the P4ꢀP2 positions. In the second, we also utilized a P1 individual substrate library to
explore cathepsin L preferences in the P1 position. With these two types of libraries we were able to
develop potent and highly selective fluorogenic substrates and ABPs for cathepsin L. Finally, we
confirmed the utility of identified probes by defining cathepsin L activity and localization in the breast
cancer line MDAꢀMBꢀ231.

Page 5 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
2. EXPERIMENTAL
2.1. Reagents. All chemicals were purchased from commercial suppliers and used without
further purification. For ACCꢀlabeled individual substrate synthesis, we used Rink amide RA resin
(loading 0,48 mmol/g, Iris Biotech GmbH), Fmocꢀprotected amino acids (purity > 98%, Iris Biotech
GmbH, Bachem, Creosalus, P3 BioSystems, QM Bio), Nꢀhydroxybenzotriazole (HOBt monohydrate
purity > 98%, Creosalus), N,Nꢀdiisopropylethylamine (DIPEA, peptide grade, Iris Biotech GmbH),
diisopropylcarbodiimide (DICI, peptide grade, Iris Biotech GmbH), HATU and HBTU (peptide grade,
ChemPep Inc.), 2,4,6ꢀtrimethylpyridine (2,4,6ꢀcollidine, peptide grade, SigmaꢀAldrich Sp. z o.o.),
N,N`ꢀdimethylformamide (DMF, peptide grade, WITKO Sp. z o.o.), acetonitrile (ACN, HPLC pure,
WITKO Sp z o.o.), piperidine (PIP, peptide grade, Iris Biotech GmbH), trifluoroacetic acid (TFA,
purity 99%, Iris Biotech GmbH), triisopropylsilane (TIPS, purity 99%, SigmaꢀAldrich Sp. z o.o.),
methanol (MeOH, pure for analysis, POCh), dichloromethane (DCM, pure for analysis, POCh),
diethyl ether (Et2O, pure for analysis, POCh), acetic acid (AcOH, purity 98%, POCh), and phosphorus
pentoxide (P2O5, purity 98%, POCh). Fmocꢀ6ꢀahxꢀOH, biotin, and 2ꢀchlorotrityl chloride resin (100ꢀ
200 mesh, 1.59 mmol/g) were purchased from Iris Biotech GmbH. TFE (2,2,2ꢀtrifluoroethanol), HBr
(30% wt. in AcOH), anhydrous THF (tetrahydrofuran), NMM (4ꢀmethylmorpholine), IBCF
(isobutylchloroformate) and 2,6ꢀDMBA (2,6ꢀdimethylbenzoic acid) were from SigmaꢀAldrich.
Fluorescent tags (Cyanine5 NHS/Cy5ꢀNHS, Cyanine3 NHS/Cy3ꢀNHS, and BODIPYꢀFL
NHS/BDPFLꢀNHS) were purchased from Lumiprobe. Diazomethane was generated according to the
Aldrich Technical Bulletin (ALꢀ180) protocol.
2.2. Preparation of recombinant cathepsins B, L, K, V, and S. Detailed description of
expression and purification is found elsewhere (52, 53). The concentration of active cathepsins was
determined via active site titration with Eꢀ64 inhibitor (Peptide Institute).
2.3. P4-P2 Hybrid Combinatorial Substrate Library (HyCoSuL) synthesis. A detailed
protocol for synthesis of the P4ꢀP2 fluorogenic substrate combinatorial library is provided elsewhere
(54).
2.4. Enzymatic kinetic studies. All kinetic experiments were performed using fMax
(Molecular Devices) and CLARIOStar (BMG LABTECH) spectrofluorimeter operating in kinetic
mode using 96ꢀwell plates. ACCꢀlabeled substrates were screened using 355 nm and 460 nm
wavelengths (excitation and emission, respectively). Cathepsin assay buffer (for library, substrates and
kinetic inhibitors) contained 100 mM sodium acetate, 100 mM sodium chloride, 10 mM DTT, 1 mM
EDTA, pH 5.5. Buffer was prepared at room temperature and enzyme kinetic studies were performed
at 37 °C.

Chemical Science
Page 6 of 32
View Article Online
DOI: 10.1039/C7SC04303A
2.5. Characterization of cathepsin L P4-P2 substrate specificity using HyCoSuL. P4, P3,
and P2 subꢀlibraries were each screened at 100 µM concentration with cathepsin L in 100 µL final
volume. The active enzyme concentration was in the 0.5ꢀ2.0 nM range, depending on subꢀlibrary.
Assay time was 30 min, but only the linear portions of progression curves (5ꢀ15 min) were used for
velocity (RFU/s) calculations. Each subꢀlibrary screening was repeated at least 3 times, and the
average value was used to create the substrate specificity matrix. The best recognized amino acid at
each position was set to 100%, and other amino acids were normalized accordingly.
2.6. Characterization of cathepsin L P1 substrate specificity. To determine cathepsin L
preferences in the P1 position we used a fluorogenic substrate library of 139 components with the
general formula AcꢀAlaꢀArgꢀLeuꢀP1ꢀACC. The library contains 19 natural and 120 unnatural amino
acids. Synthesis and detailed structure of the library are found elsewhere (51). The library was
screened with cathepsin L at two concentrations: 4 µM (below KM values) and 200 µM (above KM
values). The substrate cleavage assay was carried out for ~30 min, but only linear portions of
hydrolysis curves (10ꢀ15 min) were analyzed. Both screenings were repeated at least three times, and
average values are presented. The AcꢀAlaꢀArgꢀLeuꢀArgꢀACC substrate served as a positive control,
and the rate of its hydrolysis was set as 100%. Values of other substrates were normalized to the P1
Arg substrate.
2.7. Synthesis of individual optimized substrates. Each ACCꢀlabeled cathepsin L substrate
analyzed was synthesized and purified as described elsewhere (55).
2.8. Screening of potential cathepsin L-selective substrates with other cathepsins. All
potentially selective cathepsin L substrates were tested for their selectivity against a panel of five
different cathepsins containing cathepsins L, V, S, K and V at concentrations of 2 µM and 100 µM .
To obtain reliable results, the concentration of each cathepsin was 5 nM. Substrate hydrolysis was
carried out for 10 min (cathepsin L) and up to 60 min (for other cathepsins). Hydrolysis rates (RFU/s)
are presented as average values from at least three screenings.
2.9. Determination of kinetic parameters (kcat, KM, kcat/KM) for individual substrates.
Classic substrate velocity plots were analyzed according to MichaelisꢀMenten kinetic and a detailed
protocol for determination of kinetic parameters for ACCꢀtagged substrates is found elsewhere (55).
Because the substrates demonstrated a high degree of selectivity for cathepsin L, we had to employ
higher concentrations of other cathepsins to observe substrate hydrolysis, thus the following cathepsin
concentrations were used: cathepsin L, 0.1ꢀ0.5 nM; cathepsin B, 4ꢀ15 nM; cathepsin K, 5ꢀ20 nM;
cathepsin V, 10ꢀ20 nM; and cathepsin S, 5ꢀ20 nM. Experiments were repeated at least three times, and
results are presented as an average. S.D. were below 15%.

Page 7 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
2.10. Synthesis of irreversible activity-based probes. Two biotinꢀlabeled probes (MP-cL1
and MP-cL2) were synthesized as described by Poreba et al. (30). In brief, a biotinꢀ6ꢀahxꢀHis(Trt)ꢀ
DThr(tBu)ꢀPhe(F5)ꢀOH peptide was synthesized on solid support using 2ꢀchlorotrityl chloride resin
and used for further synthesis without purification. In parallel, two amino acylꢀwarheads amino acids
conjugated to electrophilic binding groups. (BocꢀLys(Cbz)ꢀAOMK and BocꢀArg(Cbz)2ꢀAOMK) were
synthesized in solution followed by use of diazomethane in diethyl ether (Lys/ArgꢀCH2N2), HBr in
AcOH (Lys/ArgꢀCH2Br) and 2,6ꢀDMBA in DMF (Lys/ArgꢀAOMK). Next, BocꢀLys(Bzl)ꢀAOMK and
BocꢀArg(Cbz)2ꢀAOMK were Nꢀterminally deꢀprotected in TFA/DCM (1:1 v/v with 3% of TIPS, 30
min) and coupled with the above peptide to obtain (1) biotinꢀ6ꢀahxꢀHis(Trt)ꢀ
DThr(tBu)ꢀPhe(F5)ꢀ
Lys(Cbz)ꢀAOMK and (2) biotinꢀ6ꢀahxꢀHis(Trt)ꢀ Thr(tBu)ꢀPhe(F5)ꢀArg(Cbz)2ꢀAOMK, followed by
D
purification on HPLC. Next, Trt/tBu protecting groups were removed in TFA/DCM and Cbz groups
were removed via hydrogenolysis (H2, Pd/C, DMF). The final compounds were purified again on
HPLC. Fluorescence probes were similarly synthesized. Cy5-labeled MP-cL3 probe. Bocꢀ6ꢀahxꢀ
His(Trt)ꢀDThr(tBu)ꢀPhe(F5)ꢀOH peptide was synthesized on solid support and used without further
purification. BocꢀCys(Bzl)ꢀAOMK was synthesized and Nꢀterminally deꢀprotected as described
above. Next, the peptide and the H2NꢀCys(Bzl)ꢀAOMK were coupled in DMF to obtain Bocꢀ6ꢀahxꢀ
His(Trt)ꢀDThr(tBu)ꢀPhe(F5)ꢀCys(Bzl)ꢀAOMK. All protecting groups were then removed in TFA/DCM
(1:1 v/v with 3% of TIPS, 30 min), and excess TFA was removed in a rotary evaporator under reduced
pressure (the mixture was washed several times with DCM). The crude product (1eq) was dried over
P2O5 in a desiccator for several hours. In a separate flask, 1eq of Cy5ꢀNHS was dissolved in DMF,
followed by the addition of 5eq of DIPEA. The solution was immediately mixed with crude peptideꢀ
AOMK (1eq), and the reaction was carried out at room temperature for 30 min. Reaction progress was
monitored by HPLC (220 nm for peptideꢀAOMK or 646 nm for Cy5ꢀNHS). The final product was
purified on HPLC to obtain Cy5ꢀ6ꢀahxꢀHisꢀDThrꢀPhe(F5)ꢀCys(Bzl)ꢀAOMK. Cy5/BDPFL/Cy3-labeled
MP-pc1/2/3 pan-cathepsin specific probes. First, the Bocꢀ6ꢀahxꢀAlaꢀArg(Pbf)ꢀLeuꢀOH peptide was
synthesized on a solid support and used without further purification. In parallel, BocꢀArg(Boc)2ꢀ
AOMK was synthesized and Nꢀterminally deꢀprotected as described above. Next, the peptide and
warhead were coupled in DMF to obtain Bocꢀ6ꢀahxꢀAlaꢀArg(Pbf)ꢀLeuꢀArg(Boc)2ꢀAOMK. The crude
product was then purified on HPLC, deꢀprotected with TFA/DCM, and dried over P2O5 to obtain H2Nꢀ
6ꢀahxꢀAlaꢀArgꢀLeuꢀArgꢀAOMK. Finally, the fluorescent tag (BDPFL, Cy3) was attached to the Nꢀ
terminus and the product was purified on HPLC to obtain BDPFLꢀ6ꢀahxꢀAlaꢀArgꢀLeuꢀArgꢀAOMK
(MPꢀpc2) and Cy3ꢀ6ꢀahxꢀAlaꢀArgꢀLeuꢀArgꢀAOMK (MPꢀpc3). Cy5ꢀAlaꢀArgꢀLeuꢀArgꢀAOMK (MPꢀ
pc1) was synthesized without a linker; thus first the BocꢀAlaꢀArg(Pbf)ꢀLeuꢀOH peptide was
synthesized and then coupled to the H2NꢀArgꢀAOMK warhead. Next, the compound was deꢀprotected
and labeled with a Cy5 tag in a manner similar to BDPFL and Cy3 probes.

Chemical Science
Page 8 of 32
View Article Online
DOI: 10.1039/C7SC04303A
2.11. Determination of kinetic parameters (kobs/I) for activity-based probes. For kinetic
studies, five recombinant human cathepsins (B, L, V, K, and S) were expressed and purified and active
site titrated using E64 inhibitor. The kobs/I parameter was measured under pseudoꢀfirst order kinetic
conditions, as described (40). The cathepsin of interest was mixed with various concentrations of
probe (at least 5ꢀfold excess over cathepsin) and 100 µM ACCꢀfluorescence substrate. Experiments
were repeated at least three times, and results are presented as an average. S.D. were below 20%.
2.12. Detection of cathepsin L in HEK293T cells with biotinylated ABPs. To evaluate
applicability of bMPꢀcL1 and bMPꢀcL2 probes using affinityꢀbased enrichment assays, HEK293T
cells were seeded on two 10 cm plates and cultured 2 days. Medium was changed, and cells were then
transfected using the pcDNA3::ppCatL construct (prepared inꢀhouse for expression of human
preprocathepsin L in mammalian cells) and Lipofectamine 2000TM based on the manufacturer’s
guidelines (Thermo Scientific). Fortyꢀeight hours later, cells were harvested and whole cell lysates
were prepared in a lysis buffer maintaining cathepsin activity (50 mM sodium phosphate, 100 mM
NaCl, 1 mM EDTA, 1 mM DTT, 0.5% NPꢀ40 (nonyl phenoxypolyethoxylethanol), pH 6.0) as
described elsewhere (26). Nonꢀtransfected HEK293T cells served as controls. Transfection efficiency
was calculated by measuring cathepsin L activity increase in HEK293T cells after transfection. For
cathepsin L labeling, cell lysates were mixed with various (0 ꢀ 10 ꢁM) concentrations of MPꢀcL1 and
MPꢀcL2 probes and incubated 2 hours at 37 °C. To control for nonꢀspecific binding, 25 ꢁM Eꢀ64d, a
panꢀcathepsin cell permeable inhibitor was added to lysates 1 hour prior to probe addition. After
incubations, samples were mixed with 6x SDSꢀPAGE loading buffer and analyzed by immunoblotting.
Nitrocellulose membranes were blocked after transfer with 5 ml 3% BSA (bovine serum albumin) in
PBS, pH 7.4, for 1 hour before the streptavidinꢀHRP conjugate was added to membranes at a 1:2000
dilution, according to the manufacturer's guidelines (Alpha diagnostics). Membranes were incubated
45 minutes with gentle rocking and then washed 4 times for 15 min in 10 ml PBS, pH 7.4. Membranes
were then developed using standard ECL detection reagents (GE Healthcare), and signals were
detected on bioluminescence sensitive film (Kodak).
2.13. ABP uptake into cells. To evaluate probe uptake, 250,000 HEK293T cells transfected
with pcDNA3:ppCatL were seeded into 6ꢀwell plates and incubated with various concentrations of
bMPꢀcL1 and bMPꢀcL2 (0 ꢀ 10 ꢁM) for up to 24 hours. In control experiments cells were preincubated
with 25 ꢁM Eꢀ64d, a panꢀcathepsin inhibitor, for 2 hours prior to probe addition. After incubations,
samples were collected and immunoblotted. Biotinylated proteins were detected as described above.
2.14. Evaluation of probe selectivity by immunocapture. To evaluate MPꢀcL1 and MPꢀcL2
selectivity, 250,000 HEK293T cells overexpressing cathepsin L were plated and treated with probes
and inhibitors as described above. Streptavidinꢀbound magnetic beads were prepared in separate tubes
according to the manufacturer's guidelines (Pierce™ Streptavidin Magnetic Beads, Thermo Scientific).

Page 9 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
Briefly, 50 ꢁL beads were washed with sample lysis buffer and then lysates were added. Tubes were
incubated at room temperature on a rocking platform for 2 hours. Beads were then extensively washed
with 50 mM Trisꢀbuffered saline containing 0.1% Tweenꢀ20 and collected between each washing step.
Beads were then reꢀsuspended in 1xSDS loading buffer at a final volume of 100 ꢁL, and the samples
were placed at 95 °C and prepared for immunoblotting as described above to detect biotinylated
proteins in the lysates, as well as in the unbound and eluted fractions. To confirm cathepsin L/B
selectivity, we seeded HEK293T cells as described above. Probes were added at 5 ꢁM final
concentration and incubated 18 hours before cells were harvested and lysed to isolate biotinylated
proteins using streptavidin magnetic beads, as described above. Lysates, plus the unbound and eluting
fractions were then evaluated to detect cathepsin B (inꢀhouse mouse polyclonal Ab [51]; dilution
1:1000) and cathepsin L (Abcam ab62710).
2.15. Cathepsin reactivity with MP-cL3 and MP-pc1 ABPs based on SDS-PAGE analysis.
To determine Cy5ꢀprobes potency and selectivity, each active siteꢀtitrated cathepsin was preincubated
separately in the assay buffer for 15min at 37 °C at 10 nM (experiment A) or 100 nM (experiment B)
concentration, and then mixed with various concentrations of either MPꢀcL3 or MPꢀpc1 probe for 30
min. In the experiment A, probe concentrations were 2 nM, 10 nM, 50 nM, and 250 nM, while in the
experiment B probe concentrations were 20 nM, 100 nM, 500 nM, and 2.5 µM. Reaction volume for
all samples was 200 µL. After 30 min, 100 µL of 3xSDS/DTT was added to reaction mixtures,
samples were boiled for 5 min, and then 20 µL of sample was run on 4ꢀ12% BisꢀTris Plus 15ꢀwell gels
for 30 min at 200 V with 2 µL of PageRulerTM Prestained Protein Ladder. Gels were then directly
scanned at 700 nm (red channel for Cy5 detection; excitation 685 nm) using an Odyssey fluorescence
imaging system (LIꢀCOR) and images analyzed using Image Studio software.
2.16. Detection of cathepsin L in MDA-MB-231 cells by SDS-PAGE and Western
blotting. Low passage MDAꢀMBꢀ231 cells were seeded on 12ꢀwell plates (200,000 cells per well per
1 mL of medium) and cultured in Dulbecco`s Modification of Eagle`s Medium (DMEM)
supplemented with 10% of Fetal Bovine Serum, 2 mM of Lꢀglutamine, 100 units/mL penicillin and
100 g/mL streptomycin. The next day, MPꢀcL3 and MPꢀpc1 probes (1 µM) were added to the
medium, and cells were incubated for 0ꢀ18 hours. To control for nonꢀspecific binding, cells were
preincubated with E64d (25 µM) for 2 hours prior to probe addition. At various time points, the
medium was aspirated and cells were gently washed twice with 1 mL of Dulbecco`s Phosphateꢀ
Buffered Saline (DPBS) and then trypsinized, harvested, and centrifuged for 10 min (500×g).
Supernatants were removed by aspiration and pellets washed with 1 mL of DPBS and centrifuged (5
min, 500×g). Supernatants were again removed and pellets solubilized in 100 µL 1xSDS/DTT and
then boiled 5 min. Each sample was then cooled to room temperature and sonicated (Branson Sonifier
250, duty cycle: 50, output control: 5, cycles: 5) and then 30 µL was loaded on 4ꢀ12% BisꢀTris Plus

Chemical Science
Page 10 of 32
View Article Online
DOI: 10.1039/C7SC04303A
gels and run for 30 min at 200V with a SDSꢀPAGE Standards/Broad Range ladder. Proteins were
transferred to a 0.2 µm nitrocellulose membrane for 1 hour at 10 V followed by Ponceau staining to
verify equal loading. Membranes were blocked with 2.5% BSA in Trisꢀbuffered saline/0.1% v/v
Tween 20 (TBSꢀT) for 1 hour at room temperature. Cathepsins L and B were detected by primary goat
antiꢀhCathepsin L (R&D, AF952, 0.2mg/mL, 1:2,000) and goat antiꢀhCathepsin B (R&D, AF953,
0.2mg/mL, 1:1,000) antibodies, respectively, followed by incubation with secondary antibody
(IRDye® 800CW, LIꢀCOR, donkey antiꢀgoat, 1:10,000). Membranes were incubated with primary
antibodies for 2 hours, and with (fluorescent) secondary antibodies for 30 min (both in TBSꢀT, at room
temperature). Fluorescence was read at 700nm (Cyanine5; excitation 685 nm) and 800nm (secondary
antibody; excitation 785 nm) using an Odyssey fluorescence imaging system (LIꢀCOR). Images were
analyzed using Image Studio software.
2.17. Detection of cathepsin L in MDA-MB-231 cells by immunofluorescence. Glass
coverslips 15 mm in diameter were placed in 12ꢀwell plates and coated with polyꢀDꢀlysine (500 µL, 1
mg/mL in water, Specialty Media) for 30 minutes. PolyꢀDꢀlysine was aspirated and coverslips were
washed twice with sterile water and airꢀdried. Next, low passage MDAꢀMBꢀ231 cells in DMEM
(100,000 cells/mL) were added to wells (1mL per well) and allowed to attach for 2 hours. Cells were
then treated with Cy5ꢀlabeled probes (1 µM) for 8 or 24 hours. Control cells were preincubated with
E64d for 2 hours. After various incubation times, medium was removed and cells were gently washed
twice with DPBS (1mL each). Cells were then fixed with iceꢀcold methanol (500 µL per well) for 15
minutes at ꢀ20 °C. Methanol was gently removed and fixed cells were washed twice with DPBS,
blocked and permeabilized with 10% BSA in DPBS (with 0.5% TritonꢀX, v/v) for 1 hour at room
temperature, and washed twice with 3% BSA in DPBS (with 0.5% TritonꢀX, v/v). Cathepsins L and B
were detected by primary goat antiꢀhCathepsin L (R&D, AF952, 0.2mg/mL, 1:200) and goat antiꢀ
hCathepsin B (R&D, AF953, 0.2mg/mL, 1:80), respectively, in 3% BSA in DPBS (with 0.5% Tritonꢀ
X, v/v) for 2 hours at room temperature. Antibodies were removed and cells were washed twice with
3% BSA in DPBS (with 0.5% TritonꢀX, v/v) and labeled with secondary antibody (Alexa FluorTM 488
donkey antiꢀgoat IgG (H+L), A11055, Invitrogen, 1:500) in DPBS (with 0.5% TritonꢀX, v/v) for 1
hour at room temperature. Next, cells were washed twice with DPBS, and coverslips were mounted
using Vectashield mounting medium fluorescence containing DAPI (Vector Lab. Hꢀ1000). In brief, a
drop of mounting medium was placed on the Superfrost Plus microscope slide (Fisher Scientific, no.
12ꢀ550ꢀ15) and the coverslip was placed on the drop upsideꢀdown. Excess mounting medium was
removed with a kimwipe, and nail polish was used to seal coverslips before subjecting slides to
confocal microscope analysis (Axio Observer.A1 Inverted Microscope, Zeiss, with BD Carv II
confocal unit, objective 40x, and Zeiss LSM 710 NLO Confocal Microscope, objective
PlanApochromat 63x/1.4 DIC oil). DAPI was detected by the UV channel, the Cy5 probe was detected
with Cy5 filter (excitation 647 nm (Axio Observer) or 633 nm (Zeiss LSM), emission 667 nm; red

Page 11 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
channel) and cathepsins/secondary antibodies were read using FITC filters (excitation 495 nm (Axio
Observer) or 488 nm (Zeiss LSM), emission 517 nm; red channel; green channel). All images were
acquired in .tiff format in MetaMorph software either in .lsm format in ZEN software and analyzed
using ImageJ and ZEN software. Images are representative views of cells from two coverslips.
2.18. Weighted colocalization coefficients. To quantify ABP/cathepsin colocalization we
calculated weighted colocalization coefficient between MPꢀcL3 ABP (red) and cathepsin L (green)
and between MPꢀcL3 ABP (red) and cathepsin B (green) using ZEN 2011 software. These
calculations were performed by summing the pixels in the colocalized regions (red and green) and
dividing by the sum of pixels in red channel (ABP). The value of each pixel was equal to its intensity
value (from 0 to 1), thus a weighted colocalization coefficient is more accurate here than the standard
colocalization coefficient where all the pixels, regardless of their intensity, have a value of 1. To
eliminate the red and green staining background in these calculations, we set crosshairs according to
single label controls. A graphical example of the calculated weighted colocalization coefficient is
presented in Figure S12.
3. RESULTS & DISCUSSION
3.1.P4-P1 combined approach with unnatural amino acids.
To develop novel active and selective substrates and ABPs for cathepsin L, we undertook a
combined approach employing HyCoSuL and a P1ꢀindividual substrate library containing unnatural
amino acids. Figure 1 shows an outline of this method.

Chemical Science
Page 12 of 32
View Article Online
DOI: 10.1039/C7SC04303A
Figure 1. A general procedure to optimize cathepsin L fluorogenic substrates and ABP. P4ꢀP2 enzyme
preferences are determined using previously reported HyCoSuL technology, and the P1 position is screened
using a fixed P4ꢀP3ꢀP2 sequence, but varying P1 (AcꢀAlaꢀArgꢀLeuꢀP1ꢀACC) allowing determination of the full
substrate specificity profile (P4ꢀP1). Next, the most selective substrate is converted into fluorescentꢀlabeled ABP
which allows for the selective imaging of cathepsin L in cells.
3.2. P4-P2 Hybrid Combinatorial Substrate Library. Cathepsin L substrate specificity was
determined using a substrate combinatorial library containing Arg at P1 (54). Arg is reportedly
recognized by cysteine cathepsins in the S1 pocket (8). The overall library architecture was AcꢀP4ꢀP3ꢀ
P2ꢀArgꢀACC, and the library was composed of three subꢀlibraries: P4, AcꢀXꢀMixꢀMixꢀArgꢀACC; P3,
AcꢀMixꢀXꢀMixꢀArgꢀACC; and P2, AcꢀMixꢀMixꢀXꢀArgꢀACC (Mix: an equimolar mixture of natural
amino acids, omitting cysteine and substituting norleucine for methionine). In addition to natural
amino acids, we also used 101 unnatural derivatives that can explore extended chemical space around
the cathepsin L active site pockets. As a fluorescent tag in this library (and all individual substrates for
cathepsin L) we selected an ACC (7ꢀaminoꢀ4ꢀcarbamoylcoumarin) reporter. This group is widely used
for generation of fluorescent substrates for proteases due to its high quantum yield and great signalꢀtoꢀ
noise ratio. Moreover, ACC when quenched by a peptide bond displays almost nonꢀdetectable
fluorescence (56, 57). The library was screened with cathepsin L and the data were used to create a
broad substrate specificity matrix (Figure 2). Screening the enzyme with the natural amino acids
library confirmed the specificity profile obtained by Choe et al. (8); however, use of the HyCoSuL
approach allowed us to obtain a highly detailed picture of cathepsin L active site preferences. We
found that cathepsin L has relatively narrow specificity in the S2 pocket, in agreement with proteomic
studies (26, 27, 58). Moreover, unnatural Phe derivatives were wellꢀrecognized (the best being Phe(3ꢀ
Cl)). Other amino acids were only poorly recognized or ignored by cathepsin L. Based on the P2
specificity matrix, we conclude that the cathepsin L S2 pocket is hydrophobic but not large, as some

Page 13 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
Phe analogs (for example, hPhe or 1ꢀNal) were very weakly hydrolyzed. The P3 position appeared
more tolerant than P2. Cathepsin L mostly preferred basic (Arg, Lys, Orn, Dab) and some
hydrophobic (Phg, Nle(OBzl)) amino acids. Interestingly, some Dꢀamino acids were better recognized
by cathepsin L than their Lꢀenantiomers, indicating that the S3 pocket is not stereochemically specific.
Moreover, several amino acids (mostly Phe derivatives) were completely ignored by cathepsin L.
Similarly to P3, P4 also displayed broad specificity with preference for basic amino acids (Dab, Dap,
Orn, Agp) and ability to tolerate Dꢀamino acids. Among natural amino acids, only branched Leu, Ile
and Val and hydrophobic Phe and Tyr were very poorly recognized by cathepsin L. Among unnatural
amino acids, only Phe derivatives were not recognized.
Finally, HyCoSuL screening represents a valuable tool in design of potent or selective enzyme
substrates; however, we note that HyCoSuL (like PSꢀSCL (8, 59)) is a combinatorial technique, so
potential cooperativity of protease subsites must be considered. To rule out this possibility in terms of
cathepsin L, we synthesized three sets of substrates with the formulas AcꢀP4ꢀLysꢀPheꢀArgꢀACC, Acꢀ
HisꢀP3ꢀPheꢀArgꢀACC, and AcꢀHisꢀLysꢀP2ꢀArgꢀACC and performed kinetic studies. When we
compared their kcat/Km profiles to HyCoSuL screening data, we found that cathepsin L displays almost
no S4ꢀS2 subsite cooperativity (Figure S1).
3.3. P1 individual substrate library. Cathepsin L reportedly cleaves substrates at the
carboxylic site of Arg (P1 position); however, other amino acids are tolerated by this protease in the
S1 pocket. Choe et al. demonstrated that at P1 cathepsin L also recognizes Lys, Gln, Thr, Nle and Met
(8), in agreement with proteomic data (26, 58). To gain further insight into cathepsin L P1 preferences
we utilized a 139ꢀmembered individual substrate library containing over 100 unnatural amino acids.
As a P4ꢀP2 scaffold we selected an AlaꢀArgꢀLeu sequence, which is recognized by almost all cysteine
cathepsins (8). The AcꢀAlaꢀArgꢀLeuꢀP1ꢀACC library contains diverse chemical groups at P1: basic,
acidic, hydrophilic, hydrophobic, and linear, among others. The specificity matrix is presented in
Figure 3 (the reference Arg and the best amino acids are shown in red). Our analysis revealed that
some unnatural amino acids, namely, Cys(Bzl), Cys(MeBzl), Cys(MeOBzl) and Nle(OBzl), are
significantly better recognized at P1 than the natural residue Arg. This observation demonstrates that
the cathepsin L S1 pocket has dual properties: it is negatively charged as it accepts Arg and hArg, but
it can also accommodate large, hydrophobic amino acid side chains.

Chemical Science
Page 14 of 32
View Article Online
DOI: 10.1039/C7SC04303A
Figure 2. Human cathepsin L preferences at the P4, P3, and P2 positions. Human cathepsin L preferences
were determined using three HyCoSuL subꢀLibraries, AcꢀP4ꢀMixꢀMixꢀArgꢀACC, AcꢀMixꢀP3ꢀMixꢀArgꢀACC,
AcꢀMixꢀMixꢀP2ꢀArgꢀACC, all containing natural and unnatural amino acids. The x axis shows abbreviated
names of amino acids and the y axis displays relative activity presented as a percentage of the bestꢀrecognized
amino acid. Standard deviations calculated from three screening were 15% of values shown in the figure.

Page 15 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
Figure 3. Human cathepsin L preferences at the P1 position. P1 human cathepsin L preferences were
determined using an individual fluorogenic substrate library with the general formula AcꢀAlaꢀArgꢀLeuꢀP1ꢀACC.
In this library, the P1 position is occupied by either natural or a diverse set of unnatural amino acids. The x axis
shows abbreviated names of amino acids and the y axis displays relative activity presented as a percentage of the
bestꢀrecognized amino acid. Standard deviations calculated from three screening were 15% of values shown in
the figure. Red bars indicate the best natural (Arg) or unnatural amino acids.
3.4. Development of novel, potent cathepsin L substrates. To develop active cathepsin L
substrates we combined the HyCoSuL approach with results from P1 library screening (Figure 4).
Initially, we synthesized four potentially best substrates with various P4ꢀP2 regions (Table S1) and
compared the hydrolysis efficiency with the reference AcꢀHisꢀArgꢀPheꢀArgꢀACC substrate (Figure 4).
Kinetic analysis revealed the best cathepsin L substrate (with fixed Arg in P1) to be AcꢀDapꢀOrnꢀ
Phe(3ꢀCl)ꢀArgꢀACC, which was recognized almost three times better (kcat/KM) than the reference
substrate. Next, we synthesized a 10ꢀmembered library of substrates in which the P4ꢀP2 region was
fixed with the best recognition sequence, while varying the P1 position (Figure 4, Tables S1,S2).
kcat/KM analysis allowed us to synthesize two highly active substrates, AcꢀDapꢀOrnꢀPhe(3ꢀCl)ꢀ
Cys(MeOBzl)ꢀACC and AcꢀDapꢀOrnꢀPhe(3ꢀCl)ꢀNle(OBzl)ꢀACC, which are the most potent cathepsin
L tetrapeptide substrates reported to date: both are hydrolyzed almost 4ꢀtimes faster by cathepsin L
than is the AcꢀHisꢀArgꢀPheꢀArgꢀACC substrate.

Chemical Science
Page 16 of 32
View Article Online
DOI: 10.1039/C7SC04303A
Figure 4. New, potent cathepsin L substrates. Panel A Schematic representation of the twoꢀstep procedure
used to develop novel, active cathepsin L substrates. The protocol combines the HyCoSuL approach with use of
the P1 individual substrate library. Using the former, we extracted the DapꢀOrnꢀPhe(3ꢀCl) motif as the optimal in
P4ꢀP2 position. Then we added the best amino acids from the P1 library screening to obtain the AcꢀDapꢀOrnꢀ
Phe(3ꢀCl)ꢀCys(OMeBzl)ꢀACC substrate. Panel B The kinetic parameters (kcat/KM) for the most active cathepsin
L substrates developed through combining HyCoSuL profiling with P1ꢀoptimization; aa – amino acids.
3.5. Analysis of cathepsin L-selective substrates. Although the family of AcꢀDapꢀOrnꢀ
Phe(3ꢀCl)ꢀP1ꢀACC substrates are very potent cathepsin L substrates, they display only moderate or
even poor selectivity towards other cysteine cathepsins (data not shown). Next, to develop a highly
selective cathepsin L substrate, we synthesized eleven tetrapeptide substrates with various P4ꢀP1
sequences (Figure 5, Table S3). For each position we selected the most characteristic amino acids
from the cathepsin L P4ꢀP2 profile (Figure 2). For the P4 position we selected a variety of amino acid
side chains: Tic, His, Ala, and Met(O). In P3 we chose the best amino acids (Orn, Dab) and several Dꢀ
amino acids, which are also well tolerated. P2 was occupied by Phe derivatives, the group preferred by
the enzyme. For P1, we selected mostly basic amino acids (Arg, Orn, and Lys). We then hydrolyzed
these firstꢀgeneration substrates (10 µM each) with 5nM of cysteine cathepsins B, K, L, S, and V
(Table S4) and utilized the most selective substrate (AcꢀHisꢀDValꢀPhe(F5)ꢀArgꢀACC) as a candidate
for further structure optimization. We next synthesized a secondꢀgeneration library of substrates,
replacing one position at a time (from P4 to P1) (Figure 5, Table S5 and S6). For P4 we selected His,

Page 17 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
Arg and hArg; for P3, several Dꢀamino acids; for P2, Phe analogues; and for P1, 10 amino acids best
recognized by cathepsin L (AcꢀAlaꢀArgꢀLeuꢀP1ꢀACC library profiling, Figure 3). kcat/KM profiling of
these substrates with L, V, B, K, and S cathepsins revealed several “cathepsin L selectivity
requirements” (Figure 5). In P4, the most selective amino acid appeared to be His, as this substrate
displayed a higher kcat/KM value towards cathepsin L than the P4 Arg derivatives, while maintaining
similar catalytic parameters for the other four cathepsins. P3 was occupied by four different Dꢀamino
acids, among which the DꢀThr displayed the highest affinity for cathepsin L and only residual activity
in the presence of other enzymes (kcat/KM of 286 Mꢀ1sꢀ1 for cathepsin V and 94 Mꢀ1sꢀ1 for cathepsin B).
Moreover, P2 was critical in terms of discriminating cathepsin L from other cysteine cathepsins.
Monoꢀ and diꢀsubstituted Phe are more active than 2,3,4,5,6ꢀpentafluorophenylalanine (Phe(F5)) in P2
on cathepsin L, however Phe(F5) is poorly recognized by other cathepsins, rendering Phe(F5) in the P2
position the most selective amino acid for cathepsin L.
Analysis of P1 specificity showed that various amino acids can be used to construct cathepsin
Lꢀselective substrates. Overall, the kinetic analysis performed on the secondꢀgeneration cathepsin L
substrates revealed that the most selective substrate from this library was AcꢀHisꢀDThrꢀPhe(F5)ꢀArgꢀ
ACC, which displayed almost nonꢀdetectable hydrolysis when tested with other cysteine cathepsins
(Figure 5). Moreover, a detailed kinetic analysis of two cathepsin L selective substrates revealed that
discrimination between cathepsin L and other cathepsins is mainly driven by kcat, whereas the KM
parameter has almost no impact on selectivity (Table S7).

Chemical Science
Page 18 of 32
View Article Online
DOI: 10.1039/C7SC04303A
Figure 5 Analysis of cathepsin L-selective substrates. Panel A Outline of the twoꢀstep procedure used to
develop cathepsin Lꢀselective substrates. Initially, HyCoSuL and P1ꢀlibrary screening data is used to select P4ꢀ
P1 cathepsin Lꢀcharacteristic amino acids, which are then selected to synthesize the first generation set of
substrates. After kinetic analysis of five cathepsins, the lead substrate is selected and used for second generation
substrate synthesis. The second round of kinetic analysis reveals highly specific cathepsin L substrates. Panel B
Kinetic analysis (kcat/KM) of defined secondꢀgeneration cathepsin Lꢀ substrates toward five cysteine cathepsins.
Experiments were triplicated and S.D. were below 15%.
3.6. Design and kinetic analysis of cathepsin ABPs. We next designed probes to track
cathepsin L activity in cells by synthesizing two biotinꢀlabeled ABPs containing the most selective
cathepsin L sequence (MPꢀcL1 and MPꢀcL2, Table S8). Both were equipped with an electrophilic
AOMK warhead. Kinetic analysis revealed that although probes are selective for cathepsin L over V,
S, and K, they slightly crossꢀreact with cathepsin B (Figure 6). Changing the biotin tag to cyanineꢀ5
and the P1 Arg to Cys(Bzl) significantly improved probe activity and selectivity (MPꢀcL3, Figure 6,
Table S8). We then compared compound specificity with generic (panꢀspecific) cathepsin probes

Page 19 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
containing the AlaꢀArgꢀLeuꢀArg sequence by synthesizing three fluorescentlyꢀlabeled probes and
measuring their kobs/I values in the presence of recombinant cathepsins (MPꢀpc1, MPꢀpc2 and MPꢀpc3,
Table S8). All were broad spectrum probes (Figure 6) however, the MPꢀpc1 probe displayed the
highest activity for all five enzymes and was thus further analyzed as a representative panꢀspecific
cathepsin probe. Interestingly, we observed a significant difference in potency between Cy3ꢀ and Cy5ꢀ
labeled probes (Figure 6). Since these tags have almost identical structures, we conclude that the
differences are caused by the 6ꢀahx spacer, which may be reducing the probe’s potency.
3.7. Detection of cathepsin L in cells using biotinylated probes. Biotinꢀtagged ABPs
currently represent the gold standard in terms of biological application and have been successfully
used to isolate labelled proteins by affinity enrichment/purification on appropriate beads (29, 60, 61).
Previous in vitro kinetic assays with recombinant cathepsins show that MPꢀcL1 and MPꢀcL2 probes
are potent in detecting cathepsin L and display only minor offꢀtarget activity against other cathepsins
(mainly cathepsin B) (Figure 6). Thus we wanted to validate probe utility in living cells. Given that
the biotin tag can significantly reduce permeability of a ABPs, we first investigated probes efficiency
in human embryonic kidney (HEK293T) cathepsin Lꢀtransfected cells. After incubation of these
probes in the 0 ꢀ 10 ꢁM range with cells, we observed a moderately strong signal in the cathepsins
MW range at probe concentrations as low as 100 nM (Figure S2, Panel A). The signal was increased
when the probes were used at 1µM and 10µM, respectively. Moreover, there was no cathepsin
labelling, when cells were preincubated with 25µM of Eꢀ64d, a cellꢀpermeable, broad spectrum
cathepsins inhibitor. Next, to assess applications of these two probes in biological studies, we tested
their ability to interact with cathepsin L as a potential means to identify active proteases. To do so, we
performed a pulldown assay using streptavidinꢀcoated magnetic beads and observed strong signals
corresponding to cathepsin L in the eluting fractions, signals that were completely abolished following
use of the Eꢀ64d inhibitor (Figure S2B). Furthermore, to estimate cathepsin L/B crossꢀreactivity in
HEK293T cells overexpressing cathepsin L, we performed a pulldown assay on streptavidin magnetic
beads then performed immunoblotting of cathepsin L and B in eluting fractions. To do so, we
incubated HEK293T cells overexpressing cathepsin L with both probes, prepared cell lysates, and then
evaluated the enrichment of biotinylated proteins using streptavidin magnetic beads. We found that
both MPꢀcL1 and MPꢀcL2 probes could be used to enrich cathepsin L, based on appearance of distinct
bands in the cathepsin MW range (Figure S2C). Strong bands corresponding to cathepsin L were seen
in eluting fractions (Figure S2C), but only a weak band corresponding to cathepsin B was detected in
those fractions (Figure S2D). Based on the signal intensity, we observed only minor crossꢀreactivity
of these probes with cathepsin B even after prolonged incubation, in agreement with the kinetic data
(Table 3).
3.8. Cathepsin labeling by two Cy5 ABPs. Since biotinꢀlabeled probes are not suitable for
direct visualization of protease activity, we utilized the cathepsin L selective sequence (HisꢀDThrꢀ

Chemical Science
Page 20 of 32
View Article Online
DOI: 10.1039/C7SC04303A
Phe(F5)) to construct a fluorescentlyꢀtagged probe (MPꢀcL3). To confirm probe specificity (Figure 6)
and assess activity of the MPꢀpc1 probe, we performed in vitro labeling using five human cathepsins
(L, V, B, S, and K). Both probes were incubated separately at various concentrations with active siteꢀ
titrated cathepsins (10 nM or 100 nM) for 30 min, followed by SDSꢀPAGE analysis (Figure 6). In
both experiments, regardless of cathepsin concentration, the probe/enzyme ratio was kept constant at
0.2, 1, 5, and 25. This analysis revealed that the broad spectrum MPꢀpc1 probe can label cathepsins L,
V, B, and S (10 nM) even at a 1:1 ratio. When cathepsins were used at the higher concentration of 100
nM this probe also labels cathepsin K. Importantly, the MPꢀcL3 probe displayed a very high degree of
selectivity, regardless of cathepsin concentration used (Figure 6). Cathepsin V, the closest cathepsin L
homolog, and cathepsin B are slightly labeled and only at high probe/enzyme concentrations,
demonstrating that HyCoSuL screening data can be directly applied to ABP design.
Figure 6 Broad spectrum and cathepsin L selective ABPs. Panel A Kinetic parameters (kobs/I, Mꢀ1sꢀ1) of six
ABPs assessed in the presence of five recombinant human cathepsins. Experiments were triplicated and S.D.
were below 20%. Panel B,C Labeling of recombinant cathepsins using two Cy5ꢀlabeled ABPs: a panꢀcathepsin
specific MPꢀpc1 probe (panel B) and the cathepsin Lꢀspecific MPꢀcL3 probe (panel C). Enzymes (10 nM or 100
nM) were incubated separately with various probe concentrations for 30 min and then subjected to SDSꢀPAGE
analysis. Fluorescence was scanned using the 700 nm channel (LIꢀCOR). In both experiments (enzyme at 10 nM
or 100 nM) the probe/enzyme ratio was kept constant at 0.2, 1, 5, and 10.

Page 21 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
3.9. Detection and selective labeling of cathepsin L in cells. Our ultimate goal is to develop
a chemical tool capable of selective labeling active cathepsin L in cells. This task is challenging, since
cathepsin B is highly expressed in numerous cells (including cancer cells) and cathepsins B and L
share some substrate preferences. Our initial analyses indicate that cathepsin L can be distinguished
from cathepsin B (Figure 6). To validate these results in a more biological setting we tested Cy5ꢀ
labeled MPꢀpc1 and MPꢀcL3 using HEK293T cells, which reportedly express higher levels of
cathepsin L than B [55]. To do so we incubated HEK293T cells with each probe for 0ꢀ18 hours, and in
parallel we performed immunoblotting analysis. Both probes could label active cathepsins in living
cells, as they were taken up within an hour of incubation into cellular lysosome/endosome
compartments (Figure S3). Moreover, the panꢀspecific MPꢀpc1 probe labeled both cathepsins L
(25kDa and 30kDa) and B (30kDa), while the HyCoSuLꢀderived selective MPꢀcL3 probe bound only
cathepsin L, even after prolonged (18 hour) incubation. Interestingly, analysis using a cathepsin L
antibody shows that the major form of cathepsin L in HEK293T cells is the heavy chain (25 kDa) form
this was the primary form labeled by the MPꢀcL3 probe (Figure S3). The single chain (30 kDa) form,
which also displays catalytic activity, was labeled by the probe (Figure S3).
We next challenged the selectivity of the MPꢀcL3 probe in MDAꢀMBꢀ231 cells, similarly to
HEK293T cells, since these cells contain more cathepsin B. We found that MPꢀcL3 is rapidly taken up
by cells and detectable after 1 hour and demonstrated no labeling of cathepsin B up to 8 hours (Figure
7, Figure S4). Thereafter, the fluorescence signal appeared saturated, suggesting that 8 hours is
sufficient to inhibit (and detect) cathepsin L activity. Prolonged (18 hour) incubation of the probe with
cells resulted in nonꢀspecific but faint labeling of cathepsin B (Figure 7A). As all three bands were
assigned to cathepsin L, or cathepsin B, this analysis also demonstrated that MPꢀcL3 probe does not
label cathepsin V (L2), a close cathepsin L homolog. Interestingly, MDAꢀMBꢀ231 cells may process
cathepsin L differently than HEK293T cells, as the former do not display clear differences in levels of
cathepsin L single and heavy chains. Moreover, as expected, the broad spectrum MPꢀpc1 probe bound
mostly to cathepsin B, since cathepsin L labeling was almost negligible (Figure 7B).

Chemical Science
Page 22 of 32
View Article Online
DOI: 10.1039/C7SC04303A
Figure 7 Cathepsin labeling in MDAꢀMBꢀ231 cells using Cy5ꢀlabeled ABPs. Panel A. MPꢀcL3 begins to label
active cathepsin L in living cells after 1 hour and remains selective for cathepsin L up to 8 hours (when the
fluorescence signal becomes almost saturated). After 8 hours the MPꢀcL3 probe labels also cathepsin B;
however, even after 18 hours this labeling is poor. As indicated by antibody analysis, the MPꢀcL3 probe labels
both single chain (30kDa) and heavy chain (25kDa) forms of cathepsin L. Preꢀincubation of cells with 25 µM of
E64d blocked all cysteine cathepsin activity (panel A, last lane). Panel B. The cathepsins generic probe, MPꢀc1,
labels mostly cathepsin B, as indicated by antibody analysis.
3.10. Detection of active cathepsin L by fluorescence microscopy. Our data indicate that we
could achieve high selectivity for cathepsin L by incubating MPꢀcL3 in cells for up to 8 hours (Figure
7A). We utilized this information to visualize cathepsin L activity in MDAꢀMBꢀ231 cells using
confocal fluorescence microscopy. We incubated 1µM of MPꢀcL3 probe with MDAꢀMBꢀ231 cells for
8 hours, and then methanolꢀfixed cells were labeled with either cathepsin L or cathepsin B antibodies
for imaging (Figure 8, Figures S5-S8). Since the MPꢀcL3 is not a quenched probe (qABP), it has to
be stressed here, that all nonꢀbound probe must be washed from fixed cells prior to imaging in order to
avoid falseꢀpositive signals. We found that probe signals (red) greatly overlapped with those of
cathepsin L antibodies (green), and importantly, the MPꢀcL3 probe labeled only active cathepsin L

Page 23 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
(Figure 8A, Figures S5, S6). Our Western blot analysis demonstrates that there is also a portion of
inactive (pro)cathepsin L, however, we could not detect this on fluorescence microscopy as it seems
that active and inactive cathepsin L share the same localization (Figure S10). More crucially, we
detected very few red spots alone in that overlay image, strongly suggesting that there is negligible
offꢀtarget binding by the MPꢀcL3 probe. In parallel experiments, we stained cells with cathepsin B
antibody and merged images with MPꢀcL3 probe localization (Figure 8B, Figures S7, S8). The
presence of overlapping signals strongly suggests that cathepsins L and B reside in the same
lysosomes/endosomes; however, we observed some organelles in which only cathepsin B, but not
cathepsin L, is active (Figure 8B, Figures S7, S8). Additional analysis confirmed that prolonged
incubation of the MPꢀcL3 probe with MDAꢀMBꢀ231 cells (24 hours) results in additional labeling
likely due to other cathepsins, confirming the MPꢀcL3 probe becomes nonꢀselective after 8 hours
(Figures S9, S10). Thus, the probe concentration and time of incubation with cells are both critical to
achieve selectivity in cells. Finally, we also incubated MDAꢀMBꢀ231 cells with the broad spectrum
MPꢀpc1 probe (1µM, 8 hours) and found that it labeled primarily cathepsin B, since the degree of
probe/enzyme overlap was higher than for cathepsin L enzyme (Figure S11).

Chemical Science
Page 24 of 32
View Article Online
DOI: 10.1039/C7SC04303A
Figure 8 Cathepsin L labeling of MDAꢀMBꢀ231 cells. Panel A. The cathepsin L selective probe MPꢀcL3 (1
µM) was incubated with MDAꢀMBꢀ231 cells for 8 hours and cells were then subjected for confocal fluorescence
microscopy. Cathepsin L antibody detects both the proenzyme and active enzyme; however, Cy5 detects only
active cathepsin L. In the overlay, green spots indicate inactive enzyme, as yellow show only active cathepsin L.
The lack of red spots in the overlay image strongly indicates high selectivity of the MPꢀcL3 probe. Panel B. The
same experiment as in Panel A, but here localization of the MPꢀcL3 probe was paired with staining with a
cathepsin B antibody. The results demonstrate that cathepsin B is not distributed equally with cathepsin L, as
there are lysosomes/endosomes where both enzymes are active, and others in which only active cathepsin B is
present. Both experiments (Panel A and B) were performed three times and representative pictures are presented.
Scale bar 20 µm.

Page 25 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
3.11. Colocalization between cathepsin L, cathepsin B and MP-cL3 ABP. To investigate
the colocalization of MPꢀcL3 with cathepsin L protein in MDAꢀMBꢀ231 cells we performed a
quantitative pixelꢀbyꢀpixel analysis of multiple fluorescence microscopy images. A reliable parameter
for such analysis is a weighted colocalization coefficient (wcc), which provides information not only
about single pixels overlapping, but also correlates their intensities, making the analysis more accurate
(62, 63). For this analysis we used: (1) MPꢀcL3 probe (selective labeling – 8 hours) + antiꢀcathepsin L
(8 images, 98 single cells analyzed), (2) MPꢀcL3 probe (nonꢀselective labeling – 24 hours) + antiꢀ
cathepsin L (8 images, 59 single cells analyzed) and (3) MPꢀcL3 probe (selective labeling – 8 hours) +
antiꢀcathepsin B (7 images, 68 single cells analyzed) (Figure 9, Figure S12). We demonstrated that
the highest weighted colocalization coefficient was obtained when MPꢀcL3 probe was incubated with
MDAꢀMBꢀ231 cells for 8 hours (wwc = 0.937 and 0.928) (Figure S5). Prolonged incubation of this
probe with cells (24 hours) resulted in a decline of this parameter (wwc = 0.725 and 0.695) (Figure
S9). Colocalization of MPꢀcL3 probe with cathepsin B antibody resulted in the lowest wwc value
(0.646 and 0.635) (Figure S7). To further demonstrate the usefulness of the wwc parameter we
constructed histograms from randomly selected lines from the microscope images, and overlaid probe
signal (red) with cathepsin L either cathepsin B antibody signal (green) (Figures S6 and S8).
Histograms for cathepsin L antibody (green) and MPꢀcL3 probe (red) overlapped very well (Figure
S6B), whereas histograms for cathepsin B antibody and MPꢀcL3 matched poorly (Figure S8B).
Figure 9 Colocalization of MPꢀcL3 probe with catepsin L and cathepsin B in MDAꢀMBꢀ231 cancer cells. Panel
A Weighted colocalization coefficients calculated for ABP/catꢀL and ABP/catꢀB based on (1) whole picture
analysis, or (2) single cell analysis. Points on the graph indicate single pictures (left) or single cells (right); with
“n” defining the number of pictures/single cells taken for analysis. Numbers in brackets on the xꢀaxis indicate
the incubation time of MPꢀcL3 with MDAꢀMBꢀ231 cells. Panel B Representative images used for the

Chemical Science
Page 26 of 32
View Article Online
DOI: 10.1039/C7SC04303A
calculation of weighted colocalization coefficients between MPꢀcL3 probe and cathepsin L (or B) antibody. Red
is ABP, green is cathepsin antibody, and blue is DAPI. Scale bar is 20 µm.
4. CONCLUSIONS
Development of proteaseꢀspecific substrates offering direct readout (fluorogenic or
chromogenic) is of a great significance, since such tools could be extremely useful in biochemical and
biological studies of proteases. Starting from broad spectrum chromogenic substrates developed in the
early 1940`s and 1950`s, the field has progressed through individual internally quenched fluorescence
substrate libraries, and combinatorial mixture libraries with natural amino acids (8, 64ꢀ67). However,
most of these tools are insufficient to develop selective substrates to distinguish closelyꢀrelated
proteases that display overlapping substrate specificities (among them, caspases, matrix
metalloproteases, or cysteine cathepsins). To address this challenge, we applied HyCoSuL technology
(40, 50, 54), because the use of such diverse chemical structures in a library allowed discovery of
highly selective cathepsin L substrates and probes. Combined analysis of enzyme preferences in the
P4ꢀP1 region allowed us to design an extremely selective tetrapeptide substrate (AcꢀHisꢀDThrꢀ
Phe(F5)ꢀArgꢀACC). Notably, given that we found that selectivity towards cathepsin L lies in the P4ꢀP2
sequence, the Arg at position P1 could be exchanged for nonꢀcharged amino acids to (1) make the
substrate and inhibitor more active in the presence of cathepsin L and (2) to avoid potential crossꢀ
reactivity with the numerous proteases that prefer Arg at P1. We then synthesized three small
molecule probes, with MPꢀcL3 (Cy5ꢀ6ꢀahxꢀHisꢀDThrꢀPhe(F5)ꢀCys(Bzl)ꢀAOMK) being the most
potent and selective toward cathepsin L. This probe bound exclusively to cathepsin L in an in vitro
experiment testing five recombinant cysteine cathepsins. To demonstrate that the enzymes were active,
we also used a broad spectrum cathepsin probe, MPꢀpc1 (Cy5ꢀAlaꢀArgꢀLeuꢀArgꢀAOMK). More
importantly, we showed that the MPꢀcL3 probe selectively labels cathepsin L in HEK293T and MDAꢀ
MBꢀ231 cell lines, both of which express cathepsins L and B at different levels. MPꢀcL3 probe
selectivity over cathepsin B is of special importance, as cathepsin B is overexpressed in many types of
cancers and masks cathepsin L proteolytic activity, making its analysis more complex. Finally, using
the MPꢀcL3 probe, we not only showed specific labeling of cathepsin L in cells but performed
quantitative pixelꢀbyꢀpixel localization studies using fluorescence microscopy. That analysis indicated
that cathepsin L coꢀlocalizes with cathepsin B in MDAꢀMBꢀ231 cancer cells and that some cathepsin
Bꢀrich lysosomes lack cathepsin L. We also prepared two biotinylated cathepsin L probes that have
only minor crossꢀreactivity with cathepsin B and could serve as tools in a diverse biological
experiments. In summary, we have developed the first selective cathepsin L ABP, one that allows
accurate detection of cathepsin L activity in cancer cells. Given that this probe does not react with

Page 27 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
other cathepsins, it has potential to facilitate understanding of cathepsin L function and provide new
information about the biological framework of cathepsin Lꢀdependent or ꢀrelated cancer progression.
CONFLICT OF INTEREST
There are no conflicts to declare
ACKNOWLEDGMENTS
This work was supported by the National Science Center in Poland (grant Preludium, UMOꢀ
2013/09/N/ST5/02448 to MP), National Institutes of Health in the USA (grant GM99040 to
GSS). This project has received founding from the European Union`s Horizon 2020 research
and innovation program under the Marie SkłodowskaꢀCurie grant agreement No. 661187.
B.T. and D.T were supported by the grants P1ꢀ0140 and P1ꢀ0048 by the Slovene Research
Agency. The Drag laboratory is supported by the Foundation for Polish Science.
AUTHOR CONTRIBUTIONS
M.P., G.S.S., and M.D. designed research; M.P., W.R., M.V., K.G., and P.K. performed
research and collected data; D.F., K.V., D.T., and B.T contributed new reagents and enzymes;
M.P., G.S.S., and M.D. analysed and interpreted the data, and wrote the manuscript; W.R.,
M.V., K.G., P.K., D.F., K.V., D.T, and B.T. critically revised the manuscript.
REFERENCES
1.
enzymes, their substrates and inhibitors. Nucleic Acids Res. 2016;44(D1):D343-50.
2. Kramer L, Turk D, Turk B. The Future of Cysteine Cathepsins in Disease Management. Trends
Pharmacol Sci. 2017.
Rawlings ND, Barrett AJ, Finn R. Twenty years of the MEROPS database of proteolytic
3.
Turk V, Stoka V, Vasiljeva O, Renko M, Sun T, Turk B, et al. Cysteine cathepsins: from
structure, function and regulation to new frontiers. Biochim Biophys Acta. 2012;1824(1):68-88.
4.
cathepsins in disease and their potential as drug targets. Curr Pharm Des. 2007;13(4):387-403.
5. Menard R, Carmona E, Takebe S, Dufour E, Plouffe C, Mason P, et al. Autocatalytic processing
Vasiljeva O, Reinheckel T, Peters C, Turk D, Turk V, Turk B. Emerging roles of cysteine
of recombinant human procathepsin L. Contribution of both intermolecular and unimolecular events
in the processing of procathepsin L in vitro. J Biol Chem. 1998;273(8):4478-84.
6.
regulator of human cathepsin L activity. Eur J Biochem. 1999;259(3):926-32.
7. Turk B, Dolenc I, Turk V, Bieth JG. Kinetics of the pH-induced inactivation of human cathepsin
L. Biochemistry. 1993;32(1):375-80.
Turk B, Dolenc I, Lenarcic B, Krizaj I, Turk V, Bieth JG, et al. Acidic pH as a physiological

Chemical Science
Page 28 of 32
View Article Online
DOI: 10.1039/C7SC04303A
8.
Choe Y, Leonetti F, Greenbaum DC, Lecaille F, Bogyo M, Bromme D, et al. Substrate profiling
of cysteine proteases using a combinatorial peptide library identifies functionally unique specificities.
J Biol Chem. 2006;281(18):12824-32.
9.
degradation and CD4 T cell selection in the thymus. Science. 1998;280(5362):450-3.
10. Duncan EM, Muratore-Schroeder TL, Cook RG, Garcia BA, Shabanowitz J, Hunt DF, et al.
Nakagawa T, Roth W, Wong P, Nelson A, Farr A, Deussing J, et al. Cathepsin L: critical role in Ii
Cathepsin L proteolytically processes histone H3 during mouse embryonic stem cell differentiation.
Cell. 2008;135(2):284-94.
11.
Goulet B, Baruch A, Moon NS, Poirier M, Sansregret LL, Erickson A, et al. A cathepsin L
isoform that is devoid of a signal peptide localizes to the nucleus in S phase and processes the
CDP/Cux transcription factor. Mol Cell. 2004;14(2):207-19.
12.
Reinheckel T, Hagemann S, Dollwet-Mack S, Martinez E, Lohmuller T, Zlatkovic G, et al. The
lysosomal cysteine protease cathepsin L regulates keratinocyte proliferation by control of growth
factor recycling. J Cell Sci. 2005;118(Pt 15):3387-95.
13.
therapeutic response. Nat Rev Cancer. 2015;15(12):712-29.
14. Reiser J, Adair B, Reinheckel T. Specialized roles for cysteine cathepsins in health and disease.
J Clin Invest. 2010;120(10):3421-31.
15. Mohamed MM, Sloane BF. Cysteine cathepsins: multifunctional enzymes in cancer. Nat Rev
Cancer. 2006;6(10):764-75.
Olson OC, Joyce JA. Cysteine cathepsin proteases: regulators of cancer progression and
16.
Zheng X, Chu F, Chou PM, Gallati C, Dier U, Mirkin BL, et al. Cathepsin L inhibition suppresses
drug resistance in vitro and in vivo: a putative mechanism. Am J Physiol Cell Physiol.
2009;296(1):C65-74.
17.
Frade R, Rousselet N, Jean D. Intratumoral gene delivery of anti-cathepsin L single-chain
variable fragment by lentiviral vector inhibits tumor progression induced by human melanoma cells.
Cancer Gene Ther. 2008;15(9):591-604.
18.
Yagel S, Warner AH, Nellans HN, Lala PK, Waghorne C, Denhardt DT. Suppression by
cathepsin L inhibitors of the invasion of amnion membranes by murine cancer cells. Cancer Res.
1989;49(13):3553-7.
19.
Dennemarker J, Lohmuller T, Mayerle J, Tacke M, Lerch MM, Coussens LM, et al. Deficiency
for the cysteine protease cathepsin L promotes tumor progression in mouse epidermis. Oncogene.
2010;29(11):1611-21.
20.
Fonovic M, Turk B. Cysteine cathepsins and extracellular matrix degradation. Biochim Biophys
Acta. 2014;1840(8):2560-70.
21.
Sobotic B, Vizovisek M, Vidmar R, Van Damme P, Gocheva V, Joyce JA, et al. Proteomic
Identification of Cysteine Cathepsin Substrates Shed from the Surface of Cancer Cells. Mol Cell
Proteomics. 2015;14(8):2213-28.
22.
Adachi W, Kawamoto S, Ohno I, Nishida K, Kinoshita S, Matsubara K, et al. Isolation and
characterization of human cathepsin V: a major proteinase in corneal epithelium. Invest Ophthalmol
Vis Sci. 1998;39(10):1789-96.
23.
Santamaria I, Velasco G, Cazorla M, Fueyo A, Campo E, Lopez-Otin C. Cathepsin L2, a novel
human cysteine proteinase produced by breast and colorectal carcinomas. Cancer Res.
1998;58(8):1624-30.
24.
Tolosa E, Li W, Yasuda Y, Wienhold W, Denzin LK, Lautwein A, et al. Cathepsin V is involved in
the degradation of invariant chain in human thymus and is overexpressed in myasthenia gravis. J Clin
Invest. 2003;112(4):517-26.
25.
inhibition of human cathepsin V by serpins. J Biol Chem. 2007;282(51):36980-6.
26. Vizovisek M, Vidmar R, Van Quickelberghe E, Impens F, Andjelkovic U, Sobotic B, et al. Fast
Ong PC, McGowan S, Pearce MC, Irving JA, Kan WT, Grigoryev SA, et al. DNA accelerates the
profiling of protease specificity reveals similar substrate specificities for cathepsins K, L and S.
Proteomics. 2015;15(14):2479-90.

Page 29 of 32
Chemical Science
View Article Online
DOI: 10.1039/C7SC04303A
27.
Vidmar R, Vizovisek M, Turk D, Turk B, Fonovic M. Protease cleavage site fingerprinting by
label-free in-gel degradomics reveals pH-dependent specificity switch of legumain. EMBO J.
2017;36(16):2455-65.
28.
29.
73.
30.
Turk B, Turk D, Turk V. Protease signalling: the cutting edge. EMBO J. 2012;31(7):1630-43.
Sanman LE, Bogyo M. Activity-based profiling of proteases. Annu Rev Biochem. 2014;83:249-
Poreba M, Solberg R, Rut W, Lunde NN, Kasperkiewicz P, Snipas SJ, et al. Counter Selection
Substrate Library Strategy for Developing Specific Protease Substrates and Probes. Cell Chem Biol.
2016;23(8):1023-35.
31.
Zhou S, Lin J, Du W, Zhang Z, Luo Q, Liu BF, et al. Characterization of proteinase activation
dynamics by capillary electrophoresis conjugating with fluorescent protein-based probe. J
Chromatogr B Analyt Technol Biomed Life Sci. 2006;844(1):158-62.
32.
Tchoupe JR, Moreau T, Gauthier F, Bieth JG. Photometric or fluorometric assay of cathepsin
B, L and H and papain using substrates with an aminotrifluoromethylcoumarin leaving group. Biochim
Biophys Acta. 1991;1076(1):149-51.
33.
Higashi T, Hashimoto M, Watanabe M, Yamauchi Y, Fujiwara M, Nakatsukasa H, et al. Assay
procedures for cathepsin B, H and L activities in rat tissue homogenates. Acta Med Okayama.
1986;40(1):27-32.
34.
Creasy BM, Hartmann CB, White FK, McCoy KL. New assay using fluorogenic substrates and
immunofluorescence staining to measure cysteine cathepsin activity in live cell subpopulations.
Cytometry A. 2007;71(2):114-23.
35.
probes for cysteine cathepsins. Angew Chem Int Ed Engl. 2008;47(2):406-9.
36. Hu HY, Vats D, Vizovisek M, Kramer L, Germanier C, Wendt KU, et al. In vivo imaging of
mouse tumors by a lipidated cathepsin S substrate. Angew Chem Int Ed Engl. 2014;53(29):7669-73.
37. Fonovic M, Bogyo M. Activity based probes for proteases: applications to biomarker
discovery, molecular imaging and drug screening. Curr Pharm Des. 2007;13(3):253-61.
38. Sadaghiani AM, Verhelst SH, Bogyo M. Tagging and detection strategies for activity-based
proteomics. Curr Opin Chem Biol. 2007;11(1):20-8.
39. Kam CM, Abuelyaman AS, Li Z, Hudig D, Powers JC. Biotinylated isocoumarins, new inhibitors
Watzke A, Kosec G, Kindermann M, Jeske V, Nestler HP, Turk V, et al. Selective activity-based
and reagents for detection, localization, and isolation of serine proteases. Bioconjug Chem.
1993;4(6):560-7.
40.
Kasperkiewicz P, Poreba M, Snipas SJ, Parker H, Winterbourn CC, Salvesen GS, et al. Design of
ultrasensitive probes for human neutrophil elastase through hybrid combinatorial substrate library
profiling. Proc Natl Acad Sci U S A. 2014;111(7):2518-23.
41.
Natl Acad Sci U S A. 1999;96(26):14694-9.
42. Blum G, von Degenfeld G, Merchant MJ, Blau HM, Bogyo M. Noninvasive optical imaging of
Liu Y, Patricelli MP, Cravatt BF. Activity-based protein profiling: the serine hydrolases. Proc
cysteine protease activity using fluorescently quenched activity-based probes. Nat Chem Biol.
2007;3(10):668-77.
43.
in proteomes. J Am Chem Soc. 2010;132(10):3264-5.
44. Bogyo M, Verhelst S, Bellingard-Dubouchaud V, Toba S, Greenbaum D. Selective targeting of
Tully SE, Cravatt BF. Activity-based probes that target functional subclasses of phospholipases
lysosomal cysteine proteases with radiolabeled electrophilic substrate analogs. Chem Biol.
2000;7(1):27-38.
45.
Paulick MG, Bogyo M. Development of activity-based probes for cathepsin X. ACS Chem Biol.
2011;6(6):563-72.
46.
Edgington-Mitchell LE, Bogyo M, Verdoes M. Live Cell Imaging and Profiling of Cysteine
Cathepsin Activity Using a Quenched Activity-Based Probe. Methods Mol Biol. 2017;1491:145-59.
47.
Ofori LO, Withana NP, Prestwood TR, Verdoes M, Brady JJ, Winslow MM, et al. Design of
Protease Activated Optical Contrast Agents That Exploit a Latent Lysosomotropic Effect for Use in
Fluorescence-Guided Surgery. ACS Chem Biol. 2015;10(9):1977-88.