Current Pharmaceutical Design, 2007, 13, 253-261 253
1381-6128/07 $50.00+.00 © 2007 Bentham Science Publishers Ltd.
Activity Based Probes for Proteases: Applications to Biomarker Discovery,
Molecular Imaging and Drug Screening
Marko Fonovi!1,2 and Matthew Bogyo1,*
1Department of Pathology, Stanford University School of Medicine, 300 Pasteur Drive, Stanford, California 94305, USA
and 2Department of Biochemistry and Molecular Biology, Jozef Stefan Institute, Jamova 39, SI-1000 Ljubljana, Slovenia
Abstract: Recent advances in global genomic and proteomic methods have lead to a greater understanding of how genes
and proteins function in complex networks within a cell. One of the major limitations in these methodologies is their in-
ability to provide information on the dynamic, post-translational regulation of enzymatic proteins. In particular proteases
are often synthesized as inactive zymogens that need to be activated in order to carry out specific biological processes.
Thus, methods that allow direct monitoring of protease activity in the context of a living cell or whole animal will be re-
quired to begin to understand the systems-wide functional roles of proteases. In this review, we discuss the development
and applications of activity based probes (ABPs) to study proteases and their role in pathological processes. Specifically
we focus on application of this technique for biomarker discovery, in vivo imaging and drug screening.
Key Words: Proteases, activity based probes, proteomics, biomarkers.
lished in the last five years and more than 1000 eukaryotic
and prokaryotic sequencing projects are currently underway
. While raw sequence data is highly informative it often
fails to provide insight into the functional roles of specific
gene products. Furthermore, differences in splicing and post-
translational modifications of a single gene can give rise to a
variety of products leading to a high degree of complexity.
For example it is thought that the 40,000 human genes can
generate greater than 1 million distinguishable functional
protein entities. To begin to address this complexity the field
of proteomics was established with the important goal of
analyzing the functional regulation of all proteins in a given
Almost 300 complete genome sequences have been pub-
many advances in technology, specifically this area of re-
search has flourished with the advent of user-friendly and
affordable mass spectrometers. Interestingly, one of the most
commonly used proteomics methods remains one that was
developed in the early days of the field, namely two dimen-
sional gel chromatography coupled to mass spectrometry
(2D/MS). Historically, 2D gel chromatography has been the
separation method of choice for resolving complex protein
mixtures as it can be performed easily without the need for
extensive expertise or special equipment. However, this ana-
lytical method has its limitations. It is not suitable for deter-
mination of extremely large or small proteins, very basic or
acidic proteins and membrane proteins. It also has limited
ability to detect low abundance proteins and has limited
quantification capabilities. To resolve some of these critical
Over the past 20 years the field of proteomics has seen
*Address correspondence to this author at the Department of Pathology,
Stanford University School of Medicine, 300 Pasteur Drive, Stanford, CA
94305, USA; Tel: (650)-725-4132; Fax: (650)-725-7424;
issues, additional approaches such as SELDI (solid phase
laser induced ionization)  and MudPIT (multidimensional
protein identification technique)  have been developed.
Both of these methods utilize protein separation with various
chromatographic resins prior to MS analysis. Separation can
be performed either on a chip (SELDI) or on a multi-
dimensional liquid chromatographic (LC) column (MudPIT).
Chromatographic separations significantly increase resolving
power and allow detailed analysis of sub-populations of pro-
teins within a complex proteome. In addition quantification
can be achieved by coupling isotope labeling of the pro-
teome to direct LC analysis (ICAT and SILAC) [4, 5].
global analysis of protein abundance they do not provide
information about the regulation of enzyme activity. Most
enzymes including proteases are often tightly regulated on a
post-translational level leading to a potentially significant
divergence of abundance and activity. To begin to address
some of these limitations chemical or activity based pro-
teomics has been established to characterize protein activity
and provide a means to directly monitor functional regula-
tion in complex proteomes (Fig. 1). This technique uses
small-molecule activity based probes (ABPs) that covalently
modify key active site residues through a highly specific
chemical reaction that depends on enzyme activity. These
probes, due to their high degree of selectivity, can be used in
complex samples such as cell lysates, intact cells and even
whole organisms. In this review we describe recent advances
and strategies for activity based proteomics of proteases and
their potential impact on target and drug discovery.
While the proteomic methods described above allow for
2. DESIGN AND SPECIFICITY OF ACTIVITY BASED
PROBES THAT TARGET PROTEASES
makes use of elements for targeting, subsequent chemical
modification and detection of labeled products. Perhaps the
Structurally, all ABPs have a shared basic design that
254 Current Pharmaceutical Design, 2007, Vol. 13, No. 3 Fonovi! and Bogyo
most crucial component of an ABP is the reactive functional
group, sometimes referred to as a “warhead”. This reactive
group covalently binds to amino acid residues of the target
enzyme (Fig. 2). The majority of activity based probes use
electrophilic groups that are able to form a covalent bond
with key catalytic nucleophiles located in the active site of
the enzyme. These warhead groups must be chemically reac-
tive towards an activated nucleophile of the target enzyme
but not reactive enough to modify other free neucleophiles
(i.e. cysteine or glutathione) in the proteome. Targeting of an
ABP to a specific subset of enzymes is often accomplished
by using a variable region that mimics a true substrate. In the
case of proteases, this region is almost always a peptide
which can be tailored to bind distinct sets of targets by vary-
ing the sequence of natural or non-natural amino acids. In
addition this recognition region is often used to separate the
reactive functional group from the tag to allow accessibility
and reduce steric hindrance from the often bulky labeling
group. A variety of tagging groups have been used in the
design of activity based probes. These reporters usually al-
low either direct detection or isolation of the labeled enzyme
or both. Biotin, I125 and various types of fluorophores are
most commonly used as tags. However, some types of tags
are not cell permeable and therefore cannot be used for ‘’in
vivo’’ labeling. Furthermore many of the cell permeable
fluorescent tags are bulky and may interfere with binding of
probes to target enzymes. To address these issues, an ap-
proach was recently developed where tags are chemically
linked to the probe after labeling and lysis of cells. This
strategy relies on the bio-orthogonal ‘’click’’ chemistry in
which copper catalyzes a cycloaddition reaction between an
alkyne group on the probe and an azide group on the tag .
Using this method, smaller, cell permeable probes can be
used allowing proteins to be labeled in their native environ-
ment in intact cells .
activities using two different approaches. The directed ap-
proach involves the design and synthesis of probes that tar-
get a specific enzyme class based on knowledge of catalytic
mechanism. The second method is focused on the design of
generally reactive probes that can be used to search for target
classes that have interesting regulatory patterns (i.e. in dis-
ease progression). Both methods provide a means to simplify
the proteome to a set of relevant target enzymes that can then
be studied in more detail.
In general ABPs have been applied to profile enzyme
2.1. Directed Approach
target enzyme families with well-established catalytic
mechanisms, including proteases [8-11], kinases [12, 13],
and phosphatases . In most cases probe design has taken
advantage of mechanism based inhibitors that have been
developed by biochemist and medicinal chemists. In particu-
lar the abundance of structural and kinetic data on many
classes of proteases has enabled the design of small mole-
cules with specific irreversible inhibitory activities . By
using known specific synthetic inhibitors as a template, ac-
tivity based probes that target different types of protease
classes have been designed (Fig. 2).
Activity based probes have been specifically designed to
Fig. (1). Comparison of abundance based proteomics and activity based proteomics.
A. Standard proteomic analysis using 2D SDS-PAGE coupled to a mass spectrometry. In this method total proteins are separated by 2D
SDS-PAGE, proteins are visualized by silver or Commassie stain and individual protein spots are excised, digested with trypsin and identi-
fied by mass spectrometry. This methodology provide identification and quantification of overall protein abundance. B. The activity based
proteomics approach. A complex proteome is labeled with an active site probe allowing specific isolation of active targets by affinity chro-
matography. Unlabeled proteins are removed from the sample (dashed circles) and intensity of labeled protein bands correlates to their activ-
ity rather than abundance. Only active protein species are excised and analysed by mass spectrometry.
Activity Based Probes for Proteases Current Pharmaceutical Design, 2007, Vol. 13, No. 3 255
Probes for Cysteine Proteases
mechanistic classes of proteases. Their common feature is
their catalytic mechanism: they all utilize a Cys residue as
the primary nucleophile and a His residue as the general base
for proton transfer. They have been further divided into six
clans based on the structure of the active site. The papain
family of cysteine proteases is a member of clan CA and has
been extensively studied due to its involvement in various
important physiological and pathological processes [16-18].
E-64 is a natural product inhibitor containing a reactive ep-
Cysteine proteases represent one of the four primary
oxide warhead that is highly specific for papain family cys-
teine proteases . Since its discovery in 1976, it has been
widely used for studies of cysteine proteases and has been
the central structural element in several classes of activity
based probes [8, 20-22]. The E-64 derived probe, DCG-04
has been used to determine functional roles of papain-like
proteases in processes such as: tumor progression , cata-
ract formation , prohormone processing , parasitic
invasion  and cell cycle regulation . In addition to E-
64-derived epoxides, other reactive groups have also been
used for papain like protease profiling: diazomethyl ketones
, acyloxymethyl ketones [29-31] and vinyl sulfones .
Fig. (2). Structures of protease ABPs and their chemical reaction mechanisms leading to activity dependant target modification.
A number of diverse protease directed probes have been developed. A. Covalent labeling mechanism of ABPs degined to specifically target
cysteine proteases. The reactive functional groups shown display highly specific reactivity towards the cysteine nucleophile. B. Covalent
labeling of a serine protease by a peptide diphenyl phosphonate ester. The phosphonate reactive group is highly selective for the serine nu-
cleophile of a serine protease. C. Affinity labeling of metalloproteases using a tight binding peptide hydroxamate probe. Since metallo prote-
ases do not use a covalent attack mechanism by an active site amino acid, probes must carry an alternatively reactive functional group that
can target residues within the active site. In this example a photocrosslinking group is used to covalently link the probe to the target enzyme.
S Cys Enzyme
A. CYSTEINE PROTEASES
B. SERINE PROTEASES
256 Current Pharmaceutical Design, 2007, Vol. 13, No. 3 Fonovi! and Bogyo
proteases. Its members include caspases, gingipains, legum-
ains, clostripains and separases. Among this family, caspases
are of particular interest due to their crucial role in apoptosis.
Not surprisingly the first generation of ABPs that target CD
clan proteases were directed towards caspases. In fact an
activity based probe was used to identify the first caspase
(caspase 1, !-interluken converting enzyme -ICE) . Ini-
tial caspase probes used acyloxymethyl ketone (AOMK) and
aldehyde reactive groups coupled to a specific peptide se-
quence and a biotin tag. Acyloxymethyl and chloromethly
ketone based probes have also been used to study separase
and its role in cell division [33, 34]. Recently, AOMK based
probes that target diverse CD clan of cysteine proteases have
been reported .
Clan CD is the second most abundant clan of cysteine
Probes for Serine Proteases
of human genome and includes numerous proteases, lipases,
esterases, amidases and transacylases. Like cysteine prote-
ases, serine proteases participate in many important cellular
processes, such as blood coagulation , apoptosis ,
neurotransmitter catabolism [37, 38], cancer  and protein
maturation . The majority of serine hydrolases are irre-
versibly inhibited by fluorophosphonates (FP) thus prompt-
ing its use as a general activity based probe for this family. A
3H-labeled version of diisopropyl fluorophosphonate (DFP)
is commercially available however, this probe lacks sensitiv-
ity and also cannot be used to isolate and identify target en-
zymes. To circumvent these problems, a series of analogs of
DFP containing aliphatic or PEG based linkers and a variety
of tags have been developed [9, 10, 41]. These FP probes
have been used in a range of profiling experiments, most
notably in studies of serine hydrolase biomarkers for cancer
[see below 10, 42, 43]. Additionally, peptides modified with
a c-terminal phosphonate have been developed as selective
inhibitors of serine proteases  and probes containing di
phenyl phosphonate esters were shown to function as effec-
tive probes of typsin family proteases  as well as for a
viral serine protease . These probes contain a peptide
scaffold and therefore react only with serine proteases and
not other members of the serine hydrolyase family.
The serine hydrolase family represents approximately 1%
ases do not use an amino acid nucleophile for direct covalent
attack on a substrate but rather use a zinc activated water
molecule for peptide bond hydrolysis . Thus, metallo
proteases do not form acyl-enzyme intermediates and cannot
be targeted using simple electrophiles that mimic a protein
substrate. Although some examples of electrophile-based
covalent inhibitors for metalloproteases were reported, they
were not suitable for probe design because of their weak
potency and lack of selectivity [47-49]. Alternatively, activ-
ity based probes that are capable of labeling metalloproteases
were designed using a tight binding peptide hydroxamate
scaffold carrying a label and photocrosslinking group .
These probes selectively label metalloproteases after irradia-
tion with UV light. The primary disadvantage of this type of
ABPs is that they are not suitable for use in living cells and
In contrast to cysteine and serine proteases, metalloprote-
2.2. Non-Directed Approach
groups have not been developed thus limiting the ability to
use a directed approach. In order to expand the range of en-
zyme families that can be targeted using ABPs a non-
directed or combinatorial approach was introduced [50, 51].
In this approach, a library of candidate probes was synthe-
sized using a simple alkyl scaffold carrying a series of gen-
erally reactive electrophiles. Probes were composed of a
variable alkyl/aryl binding group, a sulfonate ester reactive
group, an aliphatic linker and a rhodamine or biotine tag. A
carbon electrophile (sulfonate ester) was selected as a reac-
tive group because many carbon electrophile bearing natural
products are known to covalently bind to multiple enzyme
classes [52-54]. Libraries of probes were screened in direct
labeling assays using a diverse set of tissue and cell proteo-
mes. Probe labeled protein bands that were sensitive to heat
denaturation prior to labeling were scored as positive hits.
Members of at least nine distinct mechanistic enzyme classes
were detected by this approach. Among them were dehydro-
genases, kinases, hydratases and transferases. Interestingly,
none of the enzymes targeted by the sulfonate library was
targeted by previously described classes of specific activity
probes. In addition, sulfonate probes labeled various active
site residues (aspartate, cysteine, glutamate, tyrosine) and in
some enzymes, reacted with non active site amino acids .
For many enzyme classes, specific reactive functional
CA) reactive group coupled to a variable dipeptide binding
group was also developed for non-directed activity based
profiling . The "-CA electrophile was selected for two
reasons: first, it is relatively small resulting in minimal influ-
ence on non-covalent probe-protein interactions and second,
like other carbon electrophiles (sulfonate esters, epoxides)
the "-CA can label a variety of active site residues. Complex
proteomes from tissue samples and cell lines were labeled
with the "-CA probe library resulting in the identification of
more than 10 different enzyme classes as targets. Interest-
ingly, the overlap of targets with the sulfonate ester library
was low, suggesting that both libraries target unique subsets
of the proteome.
A library of ABPs containing the "-chloroacetamide ("-
3. APPLICATION OF ACTIVITY BASED PROFILING
3.1. Identification of Biomarkers and Target Discovery
intensified the search for biomarkers of human disease. Ge-
nomic technologies such as DNA microarrays and single
nucleotide polymorphism (SNP) analysis have been leading
the way in the discovery of gene sequences linked to human
diseases. DNA microarrays are an efficient method that pro-
vides comparative analysis of the entire complement of
mRNA populations within a biological sample in a single
experiment. Unfortunately, many pathological conditions
result from interactions and processes that take place on the
protein level. In contrast to mRNA, proteins cannot be am-
plified and they exhibit a high level of heterogeneity due to
various post-translational modifications. Activity based pro-
filing coupled with mass spectrometry has proved to be an
excellent tool for identification of protein-based disease
markers in complex proteomes (Fig. 3).
Completion of the human genome sequencing project has
Activity Based Probes for Proteases Current Pharmaceutical Design, 2007, Vol. 13, No. 3 257
covery have made use of a fluorescent serine hydrolase reac-
tive probe FP-Rhodamine to profile activities in a series of
cancer cell lines . Probe labeled samples were separated
by SDS-PAGE and fluorescently labeled proteins visualized
by laser scanning of the resulting gel. In a parallel experi-
ment, proteomes were labeled by the biotinylated version of
the probe and target proteins isolated by affinity chromatog-
raphy and identified by mass spectrometry. Using the result-
ing activity profiles of a series of serine hydrolase targets,
cancer cell lines were classified into functional subtypes
based on tissue of origin and state of invasiveness. Different
populations of serine hydrolases were found to be expressed
in cell lines with different cellular invasiveness properties. A
similar approach was also used to profile hydrolases found in
different stages of breast cancer . Serine hydrolase spe-
cific probes (fluorophosphonate reactive group) and non-
directed probes (sulfonate ester reactive group) were used to
study the differences in activity profiles of MDA-MB-231
human breast cancer cells, when grown in cell culture and
after tumor formation in mouse mammary fat pads. These
studies showed that cancer cells exhibited different activity
profiles when grown in vitro (in culture) and in vivo
(xenograft tumors). More than seven types of enzyme activi-
ties with distinct expression patterns were identified. These
findings suggest that studies performed with human cancer
cell lines in culture, may not be predictive of the behaviour
of these lines in vivo. It was also noted that many dramatic
alterations in enzyme activities occurred as a result of post-
transcriptional events, again confirming the value of activity
based profiling methods.
Some of the first efforts to use ABPs for biomarker dis-
for diagnostic markers hydroxamate based probes for metal-
loproteases were used for profiling of an invasive melanoma
cell line . Neprilysin, a membrane associated metallopro-
tease was found to be highly upregulated in melanoma cell
lines even though it is known to degrade several mitogenic
peptides and is considered to be a negative regulator of tu-
In another example of an application of ABPs to search
morigenesis . This finding suggested that in some tumor
types, neprilysin may also contribute to the progression of
cancer. Neprylisin was also shown to be a good target for the
matrix metalloprotease inhibitor GM6001 (ilomostat) which
is currently in clinical trials for cancer . Many MMP
inhibitors failed in clinical trials due to toxicity that may be
related to their broad reactivity towards different classes of
metalloproteases whose functions are poorly understood.
These studies highlight the application of ABPs for identifi-
cation of drug ‘off targets’ in complex cellular and animal
obesity, have also been studied using ABPs. The generally
reactive !-chloroacetamide based probe was used to study
differentially expressed proteins in lean and obese mice .
Several distinct enzyme activities were identified, among
them hydroxypyruvate reductase implicated in the biosyn-
thesis of glucose from serine was found to be six fold
upregulated in obese mice. This suggest, that nonclassical
pathways of glucose biosynthesis may play a part in obesity
related syndromes such as type II diabetes [59, 60].
Biomarkers for additional pathological states, such as
proteases involved in the life cycle of malaria parasite Plas-
modium falciparum. The papain family-specific probe DCG-
04 was used to identify falcipain 1 as a protease that is
upregulated in the invasive merozoite stage of P. falciparum
growth . This finding suggests that falcipain 1 may play
an important role in host cell invasion, making it an interest-
ing target for antimalarial drugs.
Activity based profiling has also been used to identify
only a limited group of enzymes in a specific enzyme class
can be targeted at once. This is problematic when attempting
to monitor large families of enzymes such as the metallopro-
teases. To overcome this problem, Cravatt and co-workers
have a developed a strategy in which small libraries of struc-
turally diverse photoreactive hydroxamate probes with com-
plementary metalloprotease selectivity are used to profile
For most of the activity based probes developed so far
Fig. (3). Application of activity based profiling to biomarker discovery.
Activity based probes can be used to profile complex proteome to identify enzymatic targets that are regulated in disease progression. In this
example normal and pathological tissue proteomes are labeled by a fluorescently tagged ABP followed by analysis by SDS-PAGE. Probe
labeled proteins are visualized by laser scanning of the resultant gel and proteins whose activities are differentially regulated in pathological
and normal proteome can be identified. These targets can then be identified by mass spectrometry and may serve as potential biomarkers for
258 Current Pharmaceutical Design, 2007, Vol. 13, No. 3 Fonovi! and Bogyo
tissue proteomes. Following SDS-PAGE analysis, a set of
optimal probes can be selected that provides the greatest
coverage of the family. This optimal set of probes can then
be used for more extensive LC/MS based analysis in diverse
biological samples. In the initial application of this approach
more than 20 metalloproteases were identified, including
members from nearly all the main branches of this enzyme
strongly support the utility of activity based profiling in the
process of drug discovery by providing a means to identify
and validate targets in the initial stages of the process. ABPs
allow direct labelling of target enzymes in crude proteomes
thereby highlighting potential targets for selection based on
the correlation of activities within a given disease pathology
(Fig. 3). ABPs can then be used for detailed biochemical and
biological assays to validate their roles in disease progres-
Taken together these examples of applications of ABPs
3.2. Inhibitor Discovery
the pharmaceutical industry for discovery of novel therapeu-
tic agents. Generally, screens are substrate-based assays
which rely on either purified native enzymes or enzymes
from a recombinant source. They are therefore dependant on
the ability to express, isolate and purify an enzyme target in
a form that is capable of processing the reporter substrate.
However, such ‘in vitro’ assays provide only limited infor-
mation about in vivo potency and selectivity of a compound
for a series of related enzymes. Furthermore, enzymes that
High throughput inhibitor screens are important tools of
cannot be expressed in recombinant form or isolated from
tissues in sufficient quantities and enzymes with unknown
substrates and reaction conditions, can not be screened for
drug leads in such assays.
tor screens that solve many of the shortcomings of standard
in vitro kinetic assays. Since ABPs bind directly to a cata-
lytic active site residue of a target enzyme, they can be used
to measure small molecule inhibitor binding using a compe-
tition assay. Furthermore, the screen can be performed in
complex cellular mixtures that contain multiple related target
enzymes. Both potency and selectivity of inhibitor binding
can be quantified by monitoring loss of probe labeling as a
result of titration of a small molecule inhibitor into the sam-
ple (Fig. 4). Using this method enzymes can be screened in
their native environment, thereby eliminating the need for
recombinant expression, purification and development of
specific substrate assays. Furthermore, activity based profil-
ing allows inhibitor screening of more than one target within
a specific proteome in a single experiment.
ABPs have been used to develop small molecule inhibi-
can also be obtained by screening of positional scanning
libraries (PSLs) of inhibitors. This approach has been used to
profile specificities of multiple papain family proteases as
well as the multiple active sites of the proteasome [62, 63].
The library screening approach allows the contribution of
each amino acid residue in the inhibitor to be determined
individually. By screening PSLs against multiple related
targets it is possible to generate an ‘affinity fingerprint’ for
each enzyme. This information can then be used to create a
Important information about target enzyme specificity
Fig. (4). Application of activity based probes in drug discovery.
ABPs can be used to profile protease activities during disease progression to identify enzymes that are likely to contribute to disease pheno-
type. Initially staged sample proteomes are directly labeled using an ABPs. Labeled target proteins whose activities are correlated with dis-
ease progression can then be identified (left panel dashed circles). Subsequently, libraries of lead compounds (often based on the probe struc-
ture) can be screened to identify compounds that target specific enzyme activities linked to disease progression. Compounds that bind to the
active site of a target prevent labeling by the ABP and are seen as a loss of labeling (inhibitors 4 and 6 right panel dashed circles). Lead com-
pounds can then be used to validate the target enzyme and can serve as lead compounds for drug development.
Activity Based Probes for Proteases Current Pharmaceutical Design, 2007, Vol. 13, No. 3 259
database of reference affinity patterns for known enzymes.
Such a database facilitates classification of unknown prote-
ases from complex proteomes by comparison of their affinity
fingerprints to the reference database. In addition, combining
specificity information for each residue enables the design of
optimized selective inhibitors. An example of this approach
is the recent development of selective caspase inhibitors
which were used for the study of caspase activation kinetics
during apoptosis . A significant technical advance in the
high-throughput screening of chemical libraries “in vitro” was
also made with the development of an enzyme microarray
technology. This method allows rapid analysis of inhibition
kinetics of enzyme targets arrayed on a chip and uses activity
based probes for the final readout of activity . Using this
technique it is possible to measure inhibition kinetics of mul-
tiple combinations of targets and inhibitors using extremely
small quantities of each.
for identification of irreversible enzyme inhibitors. Although
irreversible inhibitors are useful experimental tools, they are
less desirable as compounds for drug design because of their
inherent reactivity. While ABPs are covalent inhibitors and
therefore best suited for analysis of irreversible inhibitors, it
is possible to optimize competition assays to allow assess-
ment of reversible inhibitor binding. Fluorophosphonate-
based probes that target serine hydrolases have been used for
screening of mouse proteomes with a reversible inhibitor
library , demonstrating the utility of this approach.
So far, activity based profiling has been applied mostly
of drug development, they can also be utilized in the costly
late stages of drug development. This is the stage when drug
candidates are selected for initial animal studies for determi-
nation of general toxicity and pharmacodynamic properties.
Since ABPs can be used for in vivo competition and imaging
studies, they can provide important information regarding
specificity and potency of the drug candidate in a whole
animal. Candidates with unfavorable side effects can than be
screened for cross-reactivity with related enzyme targets.
Such approaches enable quick identification and elimination
of drug candidates likely to fail later in development thereby
potentially limiting losses associated with compound attri-
While activity based probes have value in the early stages
3.3. In Vivo Imaging of Enzyme Activities
levels by factors such as spatial and temporal expression,
small molecule or cofactor binding and posttranslational
modification. In vitro approaches are usually not capable of
reproducing complex intracelullar conditions. Therefore,
methods that allow in vivo visualization of enzyme activity
in intact cells or whole organisms would provide a much
better understanding of biochemical and physiological proc-
esses. Two main approaches have been taken toward imag-
ing of proteolytic activity in vivo. The first makes use of re-
porter substrates that produce a fluorescent signal when
processed by a protease while the second utilizes fluores-
cently labeled ABPs that directly label active proteases .
Protein activity inside the living cell is regulated on many
are composed of synthetic peptides attached to a fluorogenic
or colorimetric molecule that is released only after prote-
In the simplest of methods, protease imaging substrates
olytic processing by a protease. Such substrates were used
for detection of caspase activity in apoptotic cells . Ac-
tivity was visualized by detecting the fluorochrome or dye
molecules that were released after substrate cleavage. The
main limitation of this approach is high background signal
which results from illumination of cells with UV light. An
improved method makes use of quenched fluorogenic sub-
strates, utilizing fluorescence resonance energy transfer
(FRET). FRET-based probes contain two fluorochromes
situated less than 100 Å apart. The emission wavelength of
the donor fluorochrome overlaps with the exitation wave-
length of acceptor, allowing transfer of the donor emission to
the acceptor without radiation. Proteolytic cleavage of the
linker between two fluorochromes changes the fluorescence
intensity, which can be monitored and used as an indirect
readout of protease activity. FRET-based reporters eliminate
many drawbacks of the direct fluorogenic substrates by
eliminating artifacts associates with variations in probe con-
centrations and cell thickness. The main disadvantage of all
substrate based imaging methods is lack of specificity to-
ward enzyme targets. Peptide scaffolds of the probes can
usually serve as a substrate for more than one protease. Fur-
thermore, a lack of cell permeability and rapid diffusion of
the reporter limits their use in high resolution localization
ABP based approaches have the potential to provide bet-
ter selectivity since the reactive group tends to be highly
specific for specific protease classes and can be varied easily
to target distinct subsets of proteases. Although cell perme-
ability can often be an issue, many ABPs can freely pene-
trate cell membranes due to their hydrophobic nature. Fur-
thermore, since ABPs directly modify an enzyme target
through formation of covalent bond, any signal observed in a
whole cell or organism can be traced back to the enzyme by
direct biochemical analysis.
Multiple examples of applications of ABPs for visualiza-
tion of proteases in intact cells have been reported. Fluores-
cent ABPs have been used for monitoring of caspase activity
after induction of apoptosis in variety of cell lines [69, 70].
Fluorescent ABPs that target papain family cysteine protease
were also used for in vivo imaging studies in a mouse model
for pancreatic cancer . In this study a fluorescent ABP
was intravenously injected into a mouse. After several hours,
tissue samples were collected and protease activity visual-
ized by fluorescence microscopy. After imaging, SDS-PAGE
analysis of lysed tissue revealed the activity profiles and
identities of labeled proteases. In addition, treatment of ani-
mals with broad-spectrum inhibitors could be directly moni-
tored by visualization of loss of protease labeling in tissues.
This application has the potential to significantly improve
the process of drug testing in animal models by allowing
direct assessment of pharmacodynamic properties of lead
One of the major limitation to the application of fluores-
cent ABPs to cell-based imaging applications is that they
produce a fluorescencene signal both when bound to a target
enzyme and when they are free in solution. To address this
limitation and enable direct real time studies of protease ac-
tivity in live cells, quenched activity based probes (qABPs)
have been designed . In this approach, a quencher group
was placed in close proximity to the fluorophore, thereby
260 Current Pharmaceutical Design, 2007, Vol. 13, No. 3 Fonovi! and Bogyo
preventing fluorescence emission. When the qABP cova-
lently binds to the target enzyme, the quencher group is lost
as part of the leaving group resulting from the nucleophilic
substitution reaction with the active site residue. qABPs
therefore only emit fluorescence when they are bound to
their target enzyme. These probes have been successfully
applied for real time imaging of cysteine protease activity in
living cells and have great potential for whole-body imaging
4. CONCLUSIONS AND FUTURE DIRECTIONS
firmly established, future efforts in this field will shift to-
wards broadening the technology. The available spectrum of
activity based probes will be expanded to allow profiling of
additional enzyme classes as well as individual targets within
a specific class. In addition, development of new classes of
structurally diverse probe libraries will inevitably lead to
identification of new probe scaffolds. Applications of activ-
ity based profiling to direct mass spectrometry methods is
likely to lead to methods that will improve the limit of detec-
tion and omit the time consuming ‘in gel’ analytical methods
of current standard proteomic techniques. The first steps in
this direction have already been reported using biotinylated
serine hydrolase probes for enrichment of target proteins that
could be identified by MudPit LC-MS/MS . The advent
of click chemistry applications to activity based probes will
enable even more accurate target profiling and this approach
had already been used for development of small-molecule
cell based screens . Sensitive fluorescent reporter groups
make ABPs potentially useful tools for ‘in vivo’ analysis and
quantification of protease activity using highly sensitive
techniques such as CE-LIF (capillary electrophoresis with
laser induced fluorescence). This platform was recently used
for profiling of serine hydrolyases and cysteine proteases in
various murine proteomes . The high degree of repro-
ducibility and low detection limits, makes this approach an
important tool for pharmacokinetic and pharmacodynamic
studies of drug candidates.
With the techniques of activity based proteomics now
Lab for critical evaluation of the manuscript. This work was
supported by a NIH National Technology Centers for Net-
works and Pathways grant as part of the Center on Prote-
olytic Pathways grant #U54 RR020843.
The authors would like to thank members of the Bogyo
References 73-75 are related articles recently published in
Current Pharmaceutical Design.
 Medina M. Genomes, phylogeny, and evolutionary systems biol-
ogy. Proc Natl Acad Sci USA 2005; 102(Suppl 1): 6630-5.
Seibert V, Ebert MP, Buschmann T. Advances in clinical cancer
proteomics: SELDI-ToF-mass spectrometry and biomarker discov-
ery. Brief Funct Genomic Proteomic 2005; 4(1): 16-26.
Chen EI, Hewel J, Brunhilde Felding-Habermann B, Yates JR 3rd.
Large scale protein profiling by combination of protein fractiona-
tion and multidimensional protein identification technology (Mud-
PIT). Mol Cell Proteomics 2006; 5(1): 53-6.
Gygi SP, Rist B, Gerber SA, Turecek F, Gelb MH, Aebersold R.
Quantitative analysis of complex protein mixtures using isotope-
coded affinity tags. Nat Biotechnol 1999; 17(10): 994-9.
 Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H,
Pandey A, et al. Stable isotope labeling by amino acids in cell cul-
ture, SILAC, as a simple and accurate approach to expression pro-
teomics. Mol Cell Proteomics 2002; 1(5): 376-386.
Speers AE, Adam GC, Cravatt BF. Activity based protein profiling
in vivo using a copper(I)-catalyzed azide-alkyne [3+2] cycloaddi-
tion. J Am Chem Soc 2003; 125(16): 4686-7.
Speers AE, Cravatt BF. Profiling enzyme activities in vivo using
click chemistry methods. Chem Biol 2004; 11(4): 535-46.
Bogyo M, Verhelst S, Bellingard-Dubouchaud V, Toba S, Green-
baum D. Selective targeting of lysosomal cysteine proteases with
radiolabeled electrophilic substrate analogs. Chem Biol 2000; 7:
Liu Y, Patricelli MP, Cravatt BF. Activity-based protein profiling:
the serine hydrolases. Proc Natl Acad Sci USA 1999; 96(26):
Kidd D, Liu Y, Cravatt BF. Profiling serine hydrolase activities in
complex proteomes. Biochemistry 2001; 40(13): 4005-15.
Saghatelian A, Jessani N, Joseph A, Humphrey M, Cravatt BF.
Activity-based probes for the proteomic profiling of metalloprote-
ases. Proc Natl Acad Sci USA 2004; 101(27): 10000-5.
Bishop AC, Buzko O, Shokat KM. Magic bullets for protein
kinases. Trends Cell Biol 2001; 11(4): 167-72.
Cohen MS, Zhang C, Shokat KM, Taunton J. Structural Bio-
informatics-Based Design of Selective, Irreversible Kinase Inhibi-
tors. Science 2005; 308: 1318-1321.
Kumar S, Zhou B, Liang F, Wang WQ, Huang Z, Zhang ZY. Ac-
tivity-based probes for protein tyrosine phosphatases.Proc Natl
Acad Sci USA 2004; 101(21): 7943-8.
Powers JC, Asgian JL, Ekici OD, James KE. Irreversible inhibitors
of serine, cysteine and threonine proteases. Chem Rev 2002; 102:
Turk V, Turk B, Turk D. Lysosomal cysteine proteases: facts and
opportunities. EMBO J 2001; 20(17): 4629-33.
Yasuda Y, Kaleta J, Bromme D. The role of cathepsins in osteopo-
rosis and arthritis: rationale for the design of new therapeutics. Adv
Drug Deliv Rev 2005; 57(7): 973-93.
Fehrenbacher N, Jaattela M. Lysosomes as targets for cancer ther-
apy. Cancer Res 2005; 65(8): 2993-5.
Barrett AJ, Kembhavi AA, Brown MA, Kirschke H, Knight CG,
Tamai M, et al. L-trans-Epoxysuccinyl-leucylamido(4-guanidino)
butane (E-64) and its analogues as inhibitors of cysteine protein-
ases including cathepsins B, H and L. Biochem J 1982; 201(1):
Greenbaum D, Medzihradsky KF, Burlingame A, Bogyo M. Epox-
ide electrophiles as activity-dependant cysteine protease profiling
and discovery tools. Chem Biol 2000; 7(8): 569-81.
Greenbaum D, Baruch A, Hayrapetian L, Darula Z, Burlingame A,
Medzihradsky KF, Bogyo M. Chemical approaches for functionally
probing the proteome. Mol Cell Proteomics 2002; 1(1): 60-8.
Verhelst SH, Bogyo M. Solid-phase synthesis of double-headed
epoxysuccinyl activity-based probes for selective targeting of pa-
pain family cysteine proteases. Chembiochem 2005; 6(5): 824-7.
Joyce JA, Baruch A, Chehade K, Meyer-Morse N, Giraudo E, Tsai
FY, et al. Cathepsin cysteine proteases are effectors of invasive
growth and angiogenesis during multistage tumorigenesis. Cancer
Cell 2004; 5(5): 443-53.
Baruch A, Greenbaum D, Levy ET, Nielsen PA, Gilula NB, Kumar
NM, et al. Defining a link between gap junction communication,
proteolysis, and cataract formation. J Biol Chem 2001; 276(31):
Yasothornsrikul S, Greenbaum D, Medzihradszky KF, Toneff T,
Bundey R, Miller R, et al. Cathepsin L in secretory vesicles func-
tions as a prohormone-processing enzyme for production of the
enkephalin peptide neurotransmitter. Proc Natl Acad Sci USA
2003; 100(16): 9590-5.
Greenbaum DC, Baruch A, Grainger M, Bozdech Z, Medzihrad-
szky KF, Engel J, et al. A role for the protease falcipain 1 in host
cell invasion by the human malaria parasite. Science 2002;
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 lo-
calizes to the nucleus in S phase and processes the CDP/Cux tran-
scription factor. Mol Cell 2004; 14(2): 207-19.
Mason RW, Wilcox D, Wikstrom P, Shaw EN. The identification
of active forms of cysteine proteinases in Kirsten-virus-transformed
Activity Based Probes for Proteases Current Pharmaceutical Design, 2007, Vol. 13, No. 3 261 Download full-text
mouse fibroblasts by use of a specific radiolabelled inhibitor. Bio-
chem J 1989; 257(1): 125-9.
Bromme D, Smith RA, Coles PJ, Kirschke H, Storer AC, Krantz A.
Potent inactivation of cathepsins S and L by peptidyl (acy-
loxy)methyl ketones. Biol Chem Hoppe Seyler 1994; 375(5): 343-
Kato D, Boatright KM, Berger AB, Nazif T, Blum G, Ryan C, et
al. Activity-based probes that target diverse cysteine protease fami-
lies. Nat Chem Biol 2005; 1(1): 33-8.
Blum G, Mullins SR, Keren K, Fonovic ˇ M, Jedeszko C, Rice MJ,
et al. Dynamic imaging of protease activity with fluorescently
quenched activity-based probes. Nat Chem Biol 2005; 1(4): 203-9.
Thornberry NA, Bull HG, Calaycay JR, Chapman KT, Howard
AD, Kostura MJ, et al. A novel heterodimeric cysteine protease is
required for interleukin-1 beta processing in monocytes. Nature
1992; 356(6372): 768-74.
Thornberry NA, Peterson EP, Zhao JJ, Howard AD, Griffin PR,
Chapman KT. Inactivation of interleukin-1 beta converting enzyme
by peptide (acyloxy)methyl ketones. Biochemistry 1994; 33(13):
Uhlmann F, Wernic D, Poupart MA, Koonin EV, Nasmyth K.
Cleavage of cohesin by the CD clan protease separin triggers ana-
phase in yeast. Cell 2000; 103(3): 375-86.
Kalafatis M, Egan JO, van 't Veer C, Cawthern KM, Mann KG.
The regulation of clotting factors. Crit Rev Eukaryot Gene Expr
1997; 7(3): 241-80.
Vandennabeele P, Orrenious S, Zhivotovsky P. Serine proteases
and calpains fulfill important supporting roles in the apoptotic trag-
edy of cellular opera. Cell Death Differ 2005; 12(9): 1219-24.
Cravatt BF, Giang DK, Mayfield SP, Boger DL, Lerner RA, Gilula
NB. Molecular characterization of an enzyme that degrades neuro-
modulatory fatty-acid amides. Nature 1996; 384(6604): 83-7.
McKinney MK, Cravatt BF. Structure and function of fatty acid
amide hydrolase. Annu Rev Biochem 2005; 74: 411-32.
Jessani N, Humphrey M, McDonald WH, Niessen S, Masuda K,
Gangadharan B, et al. Carcinoma and stromal enzyme activity pro-
files associated with breast tumor growth in vivo. Proc Natl Acad
Sci USA 2004; 101(38): 13756-61.
Steiner DF. The proprotein convertases. Curr Opin Chem Biol
1998; 2(1): 31-9.
Patricelli MP, Giang DK, Stamp LM, Burbaum JJ. Direct visualiza-
tion of serine hydrolase activities in complex proteomes using fluo-
rescent active site-directed probes. Proteomics 2001; 1(9): 1067-71.
Jessani N, Liu Y, Humphrey M, Cravatt BF. Enzyme activity pro-
files of the secreted and membrane proteome that depict cancer cell
invasiveness. Proc Natl Acad Sci USA 2002; 99(16): 10335-40.
Jessani N, Niessen S, Wei BQ, Nicolau M, Humphrey M, Ji Y, et
al. A streamlined platform for high-content functional proteomics
of primary human specimens. Nat Methods 2005; 2(9): 691-697.
Hawthorne S, Hamilton R, Walker BJ, Walker B. Utilization of
biotinylated diphenyl phosphonates for disclosure of serine prote-
ases. Anal Biochem 2004; 326(2): 273-5.
Marnett AB, Nomura AM, Shimba N, Ortiz de Montellano PR,
Craik CS. Communication between the active sites and dimer inter-
face of a herpesvirus protease revealed by a transition-state inhibi-
tor. Proc Natl Acad Sci USA 2004; 101(18): 6870-5.
Coleman JE. Zinc enzymes. Curr Opin Chem Biol 1998; 2(2): 222-
Rasnick D, Powers JC. Active site directed irreversible inhibition
of thermolysin. Biochemistry 1978; 17(21): 4363-9.
Brown S, Bernardo M, Li ZH, Kotra LP, Tanaka Y, Fridman R, et
al. Potent and Selective Mechanism-Based Inhibition of Ge-
latinases. J Am Chem Soc 2000; 122(28): 6799-800.
Ikejiri M, Bernardo MM, Meroueh SO, Brown S, Chang M, Frid-
man R, et al. Design, synthesis, and evaluation of a mechanism-
based inhibitor for gelatinase A. J Org Chem 2005; 70(14): 5709-
Adam GC, Cravatt BF, Sorensen EJ. Profiling the specific reactiv-
ity of the proteome with non-directed activity-based probes. Chem
Biol 2001; 8(1): 81-95.
Adam GC, Sorensen EJ, Cravatt BF. Proteomic profiling of
mechanistically distinct enzyme classes using a common chemo-
type. Nat Biotechnol 2002; (8): 805-9.
 Wymann MP, Bulgarelli-Leva G, Zvelebil MJ, Pirola L, Vanhaese-
broeck B, Waterfield MD, et al. Wortmannin inactivates phosphoi-
nositide 3-kinase by covalent modification of Lys-802, a residue
involved in the phosphate transfer reaction. Mol Cell Biol 1996;
Runnegar M, Berndt N, Kong SM, Lee EY, Zhang L. In vivo and in
vitro binding of microcystin to protein phosphatases 1 and 2A.
Biochem Biophys Res Commun 1995; 216(1): 162-9.
Liu S, Widom J, Kemp CW, Crews CM, Clardy J. Structure of
human methionine aminopeptidase-2 complexed with fumagillin.
Science 1998; 282(5392): 1324-7.
Adam GC, Burbaum J, Kozarich JW, Patricelli MP, Cravatt BF.
Mapping enzyme active sites in complex proteomes. J Am Chem
Soc 2004; 126(5): 1363-8.
Barglow KT, Cravatt BF. Discovering disease-associated enzymes
by proteome reactivity profiling. Chem Biol 2004; 11(11): 1523-
Turner AJ, Isaac RE, Coates D. The neprilysin (NEP) family of
zinc metalloendopeptidases: genomics and function. Bioessays
2001; 23(3): 261-9.
Overall CM, Lopez-Otin C. Strategies for MMP inhibition in can-
cer: innovations for the post-trial era. Nat Rev Cancer 2002; 2(9):
Kurukulasuriya R, Link JT, Madar DJ, Pei Z, Richards SJ, Rohde
JJ, et al. Potential drug targets and progress towards pharmacologic
inhibition of hepatic glucose production. Curr Med Chem 2003;
Ross SA, Gulve EA, Wang M. Chemistry and biochemistry of type
2 diabetes. Chem Rev 2004; 104(3): 1255-82.
Sieber SA, Niessen S, Hoover HS, Cravatt BF. Proteomic profiling
of metalloprotease activities with cocktails of active-site probes.
Nat Chem Biol 2006; 2(5): 274-81.
Nazif T, Bogyo M. Global analysis of proteasomal substrate speci-
ficity using positional-scanning libraries of covalent inhibitors.
Proc Natl Acad Sci USA 2001; 98(6): 2967-72.
Greenbaum DC, Arnold WD, Lu F, Hayrapetian L, Baruch A,
Krumrine J, et al. Small molecule affinity fingerprinting. A tool for
enzyme family subclassification, target identification, and inhibitor
design. Chem Biol 2002; 9(10): 1085-94.
Berger AB, Witte MD, Denault JB, Sadaghiani AM, Sexton KM,
Salvesen GS, et al. Identification of early intermediates of caspase
activation using selective inhibitors and activity based probes. Mol
Cell 2006; 23(4): 509-21.
Funeriu DP, Eppinger J, Denizot L, Miyake M, Miyake J. Enzyme
family-specific and activity-based screening of chemical libraries
using enzyme microarrays. Nat Biotechnol 2005; 23(5): 622-7.
Leung D, Hardouin C, Boger DL, Cravatt BF. Discovering potent
and selective reversible inhibitors of enzymes in complex proteo-
mes. Nat Biotechnol 2003; 21(6): 687-91.
Baruch A, Jeffery DA, Bogyo M. Enzyme activity--it's all about
image. Trends Cell Biol 2004; 14(1): 29-35.
Gurtu V, Kain SR, Zhang G. Fluorometric and colorimetric detec-
tion of caspase activity associated with apoptosis. Anal Biochem
1997; 251(1): 98-102.
Bedner E, Smolewski P, Amstad P, Darzynkiewicz Z. Activation of
caspases measured in situ by binding of fluorochrome-labeled in-
hibitors of caspases (FLICA): correlation with DNA fragmentation.
Exp Cell Res 2000; 259(1): 308-13.
Amstad PA, Yu G, Johnson GL, Lee BW, Dhawan S, Phelps DJ.
Detection of caspase activation in situ by fluorochrome-labeled
caspase inhibitors. Biotechniques 2001; 31(3): 608-10, 612, 614.
Evans MJ, Saghatelian A, Sorensen EJ, Cravatt BF. Target discov-
ery in small-molecule cell-based screens by in situ proteome reac-
tivity profiling. Nat Biotechnol 2005; 23(10): 1303-7.
Okerberg ES, Wu J, Zhang B, Samii B, Blackford K, Winn DT, et
al. High-resolution functional proteomics by active-site peptide
profiling. Proc Natl Acad Sci USA 2005; 102(14): 4996-5001.
Pusch W, Kostrzewa M. Application of MALDI-TOF mass spec-
trometry in screening and diagnostic research. Curr Pharm Des
2005; 11(20): 2577-91.
Riedel W. J, Mehta M. A, Unema P. J. Human cognition assess-
ment in drug research. Curr Pharm Des 2006; 12(20): 2525-39.
Takahashi H, Sano H, Chiba H, Kuroki Y. Pulmonary surfactant
proteins A and D: innate immune functions and biomarkers for
lung diseases. Curr Pharm Des 2006; 12(5): 589-598.