Peptide-Based Fluorescence Resonance Energy Transfer Protease Substrates for the Detection and Diagnosis of Bacillus Species
We describe the development of a highly specific enzyme-based fluorescence resonance energy transfer (FRET) assay for easy and rapid detection both in vitro and in vivo of Bacillus spp., among which are the members of the B. cereus group. Synthetic substrates for B. anthracis proteases were designed and exposed to secreted enzymes of a broad spectrum of bacterial species. The rational design of the substrates was based on the fact that the presence of D-amino acids in the target is highly specific for bacterial proteases. The designed D-amino acids containing substrates appeared to be specific for B. anthracis but also for several other Bacillus spp. and for both vegetative cells and spores. With the use of mass spectrometry (MS), cleavage products of the substrates could be detected in sera of B. anthracis infected mice but not in healthy mice. Due to the presence of mirrored amino acids present in the substrate, the substrates showed high species specificity, and enzyme isolation and purification was redundant. The substrate wherein the D-amino acid was replaced by its L-isomer showed a loss of specificity. In conclusion, with the use of these substrates a rapid tool for detection of B. anthracis spores and diagnosis of anthrax infection is at hand. We are the first who present fluorogenic substrates for detection of bacterial proteolytic enzymes that can be directly applied in situ by the use of D-oriented amino acids.
XXXX American Chemical Society
Anal. Chem. XXXX, XXX, 000–000
Peptide-Based Fluorescence Resonance Energy Transfer Protease
Substrates for the Detection and Diagnosis of Bacillus Species
Wendy E. Kaman,*
Albert G. Hulst,
Pleunie T. W. van Alphen,
Marcel J. van der Schans,
Alex van Belkum,
and Floris J. Bikker
Department of Medical Microbiology and Infectious Diseases, Erasmus MC,
s-Gravendijkwal 230, 3015 CE Rotterdam,
TNO Defence, Security and Safety, Lange Kleiweg 137, 2288 GJ Rijswijk, The Netherlands
Department of Oral Biochemistry, Academic Centre for Dentistry Amsterdam , University of Amsterdam and VU University
Amsterdam, Gustav Mahlerlaan 3004, 1081 LA Amsterdam, The Netherlands
CBER Laboratory, U.S. Food and Drug Admini stration, 29 Lincoln Drive, Bethesda, Maryland 20892, United States
BioMerieux, 3 Route de Port Michaud, 38390, La Balme les Grottes, France
he Gram-positive Bacillus anthracis bacterium is the causa-
tive agent of anthrax. The stability, ease of production, and
infectious capacity of the spores confer upon B. anthracis a high
potential as a biological weapon.
During pulmonary anthrax,
spores germinate in the lungs ultimately followed by the emer-
gence of vegetative anthrax in the circulation.
infection frequently results in secondary shock and multiple
organ failure, which, if untreated, results in death.
point-of-care diagnosis is critical for eﬀective treatment of
B. anthracis possesses two major virulence components: the
pXO1 and pXO2 plasmids.
These plasmids encode the bio-
synthesis pathway for
D-glutamic acid and the anthrax toxins
lethal factor (LF), edema factor (EF), and protective antigen
Because of their speciﬁcity to B. anthracis
and their absence in the closely related species B. cereus and B.
thuringiensis, to date the pXO1 and pXO2 plasmids are the main
targets used in the detection of B. anthracis and the diagnosis of
anthrax. For instance, numerous antibody-based detection meth-
ods for B. anthracis target the pXO1 encoded anthrax toxins.
Other DNA-based techniques used in the detection of
B. anthracis are (real-time and multiplex) polymerase chain
reaction (PCR) targeting pXO1 and pXO2.
their high speciﬁcity, these tests do not provide evidence on
pathogen viability and disease progression.
Bacterial enzymes, such as proteases, are in theory ideally
suited as biomarkers for quick and sensitive identiﬁcation of
microorganisms in clinical samples.
Many of these enzymes are
released into the surrounding microenvironment and are acces-
sible for detection based on sensitive ﬂuorogenic and/or lumi-
nogenic substrates. However, in practice lack of speciﬁcity has
Received: October 20, 2010
Accepted: February 12, 2011
ABSTRACT: We describe the development of a highly speciﬁc
enzyme-based ﬂuorescence resonance energy transfer (FRET)
assay for easy and rapid detection both in vitro and in vivo of
Bacillus spp., among which are the members of the B. cereus group.
Synthetic substrates for B. anthracis proteases were designed and
exposed to secreted enzymes of a broad spectrum of bacterial
species. The rational design of the substrates was based on the fact
that the presence of
D-amino acids in the target is highly speciﬁc
for bacterial proteases. The designed
D-amino acids containing
substrates appeared to be speciﬁc for B. anthracis but also for
several other Bacillus spp. and for both vegetative cells and spores.
With the use of mass spectrometry (MS), cleavage products of the
substrates could be detected in sera of B. anthracis infected mice
but not in healthy mice. Due to the presence of mirrored amino
acids present in the substrate, the substrates showed high species
speciﬁcity, and enzyme isolation and puriﬁcation was re dundant.
The substrate wherein the
D-amino acid was replaced by its
L-isomer showed a loss of speciﬁcity. In conclusion, with the use of these substrates a rapid tool for detection of B. anthracis
spores and diagnosis of anthrax infection is at hand. We are the ﬁrst who present ﬂuorogenic substrates for detection of bacterial
proteolytic enzymes that can be directly applied in situ by the use of
D-oriented amino acids.
B dx.doi.org/10.1021/ac102764v |Anal. Chem. XXXX, XXX, 000–000
proven to be a large hurdle, which has seriously hampered
practical application for diagno sis.
Proteases occur abundantly in all organisms, from viruses to
men. They are involved in a myriad of processes and functions,
from simple digestion of food proteins to highly regulated
cascades such as the blood-clotting and the complement
It is therefore obvious that proteolytic activity by
itself is not a useful indicator for the presence of bacteria, let alone
for a speciﬁc pathogen. To overcome this problem, studies have
been undertaken to develop substrates with exquisite speciﬁcity
for protease(s) from a speciﬁc pathogen.
substrates developed thus far still suﬀer from a lack of speciﬁcity,
as they can be hydrolyzed by a variety of bacterial and human
For their use as diagnostic tools, therefore, processi ng
of the sample to isolate the target enzyme under investigation is
still needed. This is laborious, time-consuming, and costly and,
moreover, prone to yielding erratic results. Multiple enzymatic
assays for the detection of B. anthr acis using the ﬂuor escence
resonance energy transfer (FRET) technology have been de-
scribed by others.
Most of these assays are based on the
detection of the anthrax toxins LF and EF.
Boyer and co-
workers studied the LF activity on a 45 amino acid artiﬁcial
substrate using matrix-assisted laser desorption ionization time-
of-ﬂight mass spectrometry (MALDI-TOF MS).
applied to sera of infected monkeys, the assay detected the toxin
in the femtomolar range. However, the activity of unrelated
proteolytic enzymes will lead to false-po sitive results, and hence,
the detection of LF had to be facilitated by a time-consuming
In the present study a diﬀerent approach was developed to
enhance the speciﬁcity of protease substrates. This approach
exploited an intriguing di ﬀ erence between eukaryotic and pro-
karyotic cells with regard to the use of
L-amino acids are primarily incorporated in natural
proteins, the presence of
D-amino acids is highly speciﬁc for
bacterial proteases, in which they are abundantly present as a
component of the bacterial cell walls.
This led us to hypothesize
that bacteria may express proteases which possess a unique
property for recognizing and hydrolyzing
D-amino acid contain-
ing sub strates. Such activities can be translated into diagnostic
The aim of the present study was to develop FRET-based
substrates for the speci ﬁc detection of proteolytic enzymes of the
B. cereus group which includes B. anthracis .
’ MATERIALS AND METHODS
Bacteria. The bacterial isolates used in this study are
B. anthracis Vollum (ATCC 14578), B. cereus, B. thuringiensis
var. kurstaki aizawai, B. globigii, B. mycoides (ATCC 14579), B.
subtilis Ma rburg (ATCC 6051), B. megaterium, (ATCC 15374),
Yersinia pseudotuberculosis (ATCC 29833), Y. pestis HmsF-
(ATCC 19428), Brucella suis Thomsen, Br. melitensis 16M,
Escherichia coli (ATCC 11775), Micrococcus luteus, Erwinia
herbicola (ATCC 33243), Listeria monocytogenes EGDe (ATCC
BAA-679), Salmonella typhimurium, S. Montevideo, Pseudomonas
aeruginosa, Staphylococcus aureus, Staph. aureus MRSA (ATCC
43300), Staph. aureus MSSA (ATCC 25923), Acetinobacter
lwoffii calcoaceticus Ruh88, Vibrio cholerae, and Clostridium botu-
linum A (NCTC 2916). Bacteria were grown overnight in 5 mL
of brain heart infusion (BHI) medium (BioTrading, Mijdrecht,
The Netherlands) at 35 °C, and at 26 °C for the Yersiniae. The
next day, the bacteria were pelleted by centrifugation for 10 min
at 10 000 rpm. Supernatant, containing secreted enzymes, was
sterilized by filtration through a 0.22 μm filter (Millipore,
Amsterdam, The Netherlands). The crude samples were used
directly or stored at -20 °C for later use.
FRET Assay. FRET substrates were designed by using the
and provided by PepScan Presto B.V.
(Lelystad, The Netherlands) with a purity of approximately 90%.
The identity of the substrates was confirmed by mass spectrometry.
Thesubstratesweredenotedas“BikKams” (Table 1). Assays were
performed in Blackwell, clear bottom 96-well plates (Corning,
Lowell, U.S.A.). Enzyme activity in bacterial supernatants was
determined by incubating 16 μM substrate with 100 μLoffiltered
culture supernatant at 37 °C. Filtered BHI medium was used as a
negative control. Plates were read with 10 min intervals on a
CytoFluor 4000 (Applied Biosystems, Foster City, U.S.A.) with
excitation at 485 nm and emission at 530 nm. Relative fluorescence
(RF) is the value obtained after correction with the negative
control, BHI medium. The measured enzyme activity is defined
in RF per minute (RF/min).
Preparation of Spores. B. anthracis strain Vollum (ATCC
14578) and B. subtilis strain Marburg were cultured by shaking at
35 °C in 250 mL of sporulation broth (SB). At a sporulation
efficiency of 99% the suspensions were centrifuged at 4000g for
40 min. The pellets were washed with distilled water and
resuspended in 1 mL of water. The number of viable spores in
the suspensions was determined by plating 10-fold serial
dilutions on trypticase soy agar (TSA) plates (BioTrading,
Mijdrecht, The Nethe rlands). Plates were incubated at 35 °C,
and spores were enumerated after 1 day of incubation.
Preculture Detection of B. anthracis Spores. B. anthr acis
and B. subtilis spores were precultured in 1 mL of BHI medium at
35 °C. Samples were taken 0, 1, 2, 3, and 4 h after incubation.
After centrifugation the samples were incubated using 16 μMof
the BikKam1 substrate. The increase in fluorescence was mea-
sured for 1 h with 10 min intervals on a CytoFluor 4000 (Applied
Biosystems, Foster City, U.S.A.) with excitation using a 485 nm
filter and emission using a 530 nm filter. Relative fluorescence is
the value obtained after correction with the negative control, BHI
medium. The measured enzyme activity is defined in RF per
In Vivo Infection Model. Anesthetized male Balb/C mice
(Harlan, Horst, The Netherlands) were intranasally inoculated
with 1.25 10
B. anthracis spores in a 50 μL volume. For each
time point 10 mice were used. At 0, 12, and 48 h postinfection,
serum was isolated and pooled per time point. Because of the
small amount of sera, samples were analyzed using liquid
chromatography-electrospray tandem mass spectrometry
(LC-ES MS/MS) as described below. During the experiment
Table 1. FRET Substrates Designed in This Study
C dx.doi.org/10.1021/ac102764v |Anal. Chem. XXXX, XXX, 000–000
the animals were kept in sterile isolators (UNO, Zevenaar, The
Netherlands) in a biohazard animal unit. They were fed irra-
diated food (Harlan, Horst, The Netherlands) and acidified
water ad libitum. The mice were monitored regularly for clinical
status and weighed daily. All exper imental procedures performed
on the animals were approved by the The Ethical Committee on
Animal Experimentation of TNO (DEC 2727).
LC-ES MS/MS. For the LC-ES MS/MS analysis of the
infected mouse serum, 1:10 in 0.06 M EDTA diluted serum was
incubatedwith16μM FRET substrate for 3 h at 37 °C. Assay
mixtures were loaded onto an Amicon 10 000 MWCO filter
(Millipore, Amsterdam, The Netherlands) that had been precondi-
tioned with 300 μL of 50% acetonitrile and 300 μLof0.2%(v/v)
formic acid in water, respectively. After the sample loading, the filter
was washed with 100 μL of 0.2% (v/v) formic acid in water. The
collected samples were analyzed using LC-ES MS/MS.
LC-ES MS/MS experiments were conducted on a Q-TOF
hybrid instrument (Micromass, Altri ncham, U.K.) equipped with
a standard Z-spra y ES interface (Micromass) and an Alliance
type 2690 liquid chromatograph (Waters, Milford, MA, U.S.A.).
The chromatographic hardware consisted of a precolumn splitter
(type Accurate; LC Packings, Amsterdam, The Netherlands), a
six-port valve (Valco , Schenkon, Switzerland) with a 10 or 50 μL
injection loop mounted, and a PepMap C
column (15 cm
1 mm i.d., 3 μm particles; LC Packings, Amsterdam, The
A gradient of eluents A (H
O with 0.2 v/v % formic acid) and
B (acetonitrile with 0.2 vol % formic acid) was used to achieve
separation, as follows: 100% A (at time 0 min, 0.6 mL/min ﬂow)
to 10% A and 90% B (at 45 min, 0.6 mL/min ﬂow). The ﬂow
delivered by the LC equipment was split precolumn to allow a
ﬂow of approximately 40 μL/min through the column and into
the ES MS interface.
The Q-TOF was operated at a cone voltage of 20-25 V,
employing nitrogen as the nebulizer and desolvation gas (at a
ﬂow of 20 and 400 L/h, respectively). MS/MS product ion
spectra were recorded using a collision energ y of 10- 11 eV, with
argon as the collision gas (10
Concept and Peptide Design for a FRET-Based Anthrax
To design substrates specific for B. anthracis
we used the MEROPS database to search for known peptidases
produced by B. anthracis. Peptidases suitable for our purpose
should be able to cleave substrates wherein
acids are present to protect the substrate from cleavage by other
enzymes. One of the selected enzymes was dipeptidase AC which
recognizes substrates with the sequence Leu-
D-Leu. Besides in
B. anthracis this peptidase is also produced by several other
bacterial species such as A. calcoaceticus and Brucella spp.
check the specificity of the Leu-
D-Leu substrate (BikKam1) the
substrate was incubated with culture supernatant of numerous
bacterial species. It appeared that BikKam1 was only cleaved by
culture supernatants of Bacillus spp., whereas no cleavage of
BikKam1 by culture supernatants of A. calcoaceticus or Brucella
spp. was observed (Table 2). From these results we hypothesized
that probably the cleavage of this substrate is not related to
dipeptidase AC. To obtain more insight in the mechanism of
action variants on the BikKam1 substrate were designed (Table 1).
All substrates designed contained fluorescein isothiocyanate
(FITC) as probe and Dabcyl (Dbc) as its quencher. The
importance of the placement of
D-Leu was investigated by switch-
ing the position of Leu and
D-Leu in thesequence (BikKam2). The
hypothesis of the specificity of the use of
D-amino acids was
checked by a substrate in which no
D-amino acids are present
(BikKam3). BikKam4 was designed to see if the specificity could
be increased by the replacement of Leu by another
D-Leu and in
addition to try to increase the cleavage activity another Leu was
added to the Leu-
D-Leu sequence (BikKam5). The importance of
the presence of Leu in the sequence was investigated by the use of
substrates in which one of the
D-Leu was replaced by another
closely related amino acid
D-Val (BikKam6). The substrates Gly-
D-Leu (BikKam7) and Gly-D-Ala (BikKam8) were designed by
Adachi and Tsujimoto
and are both cleaved by dipeptidase AC
with a higher efficiency than the Leu-
In Vitro Evaluation of BikKam Substrates. To further
explore the specificity of all the BikKams (Table 1), the sub-
strates were incubated with culture supernatants, potentially
containing secreted pro teases deriving from a broad spectrum
of bacterial species. In all substrates which contain
cleavage of the substrate by bacilli of the B. cereus group as well
as B. megaterium and B. licheniformis was observed (Table 2). The
only exception was the substrate which consists of two
amino acids (BikKam4). No cleavage of this substrate by
B. anthracis or B. mycoides was observed. All other bacteria tested
did not show activity with any of the
D-Leu containing substrates
(Table 2) including two bacilli which belong to another part of
the Bacillus tree: the B. subtilis group (Table 2). In case
replaced by its
L-isomer (BikKam3) a loss in specificity was
observed; besides B. subtilis and B. globigii the substrate was
cleaved by culture supernatants of a number of other bacterial
species including P. aeruginosa and V. cholerae (Table 2).
To further substantiate the role of the
D-Leu in the BikKam1
D-Leu was replaced by either D-Val (BikKam 6), which
structure is closely related to
D-Leu, or D-Ala (BikKam8), where
the R-carbon atom is covalently bound to a methyl group.
BikKam6 was cleaved by all Bacillus spp. though with a lower
eﬃciency. No cleavage of the
D-Ala substituted substrate by
Bacillus spp. was observed (Table 2). Replacement of Leu by Gly
had no eﬀect on the cleavage pattern or cleavage eﬃciency. No
cleavage activity by A. calcoaceticus or Brucella spp. was observed
on the Gly-
D-Leu and Gly-D-Ala substrates (Table 2).
Identification of the B. anthracis Cleavage Sites by LC-ES
To identify the cleavage sites, BikKam substrates were
incubated with B. cereus culture supernatant and analyzed using
LC-ES MS/MS. All analyzed substrates appeared to be cleaved
directly after the
D-amino acid (Tab le 3). To verify these cleavage
sites for B. anthracis the experiment was repeated for two
substrates using B. anthracis culture supernatant. As expected
B. anthracis cleaved the two substrates at the same position as
B. cereus did.
Detection of Precultured B. anthracis Spores Using Bik-
To verify the applicability of the BikKam1 substrate to
detect anthrax spores, different amounts of B. anthracis spores
were triggered into vegetative state by precultivating and incu-
bated with the substrate. As a negative control spores of B. subtilis
were used. After 3 h of preculturing 10
B. anthracis spores, a
significant increase of fluorescence was observed (Figure 1A). In
case the spores were incubated for 4 h 10
B. anthracis spores
could be detected. No cleavage was observed when BikKam1 was
incubated with precultured B. subtilis spores (Figure 1B).
In Vivo Diagnosis of Inhalational Anthrax by BikKam1. To
further explore the possibilities for the BikKam1 substrate to
D dx.doi.org/10.1021/ac102764v |Anal. Chem. XXXX, XXX, 000–000
detect B. anthracis, mice were intranasally infected with 1.25
spores/mouse. Serum was isolated during the infection at 0 ,
12, and 48 h postinfection (p.i.). BikKam1 cleavage products
could be detected in the pooled sera of B. anthracis infected mice
at 48 h p.i. (Figure 2C). At this time point the mice were very ill
and showed onset of severe clinical signs. No cleavage products
were found in the sera at 0 (Figure 2A) and 12 h p.i. (Figure 2B).
In search for a rapid and simple tool for the detection of
bacterial protease acti vity in situ, short and speciﬁc substrates
D-amino acid were designed. Although L-amino
acids represent the vast majority of amino acids found in natural
proteins, the presence of
D-amino acids is highly speciﬁc for
D-amino acids are abundantly present as a
Table 3. Determination of the B. cereus and B. anthracis
BikKam Cleavage Sites Using MS Analysis
B. cereus B. anthracis
BikKam1 FITC-Leu-D-Leu 1 LysDbc FITC-Leu-D-Leu 1 LysDbc
D-Leu 1 Leu-LysDbc n.d.
D-Leu 1 Leu-LysDbc n.d.
D-Val 1 LysDbc FITC-Leu-D-Val 1 LysDbc
The cleavage sites of the substrates are denoted with (1).
Table 2. Proteolytic Activity of Bacterial Culture Supernatants against the Designed FRET Substrates
Bacteria were grown at appropriate temperature for 16 h in BHI. The measured enzyme activity is deﬁned in RF/min. RF is the value obtained after
correction with the negative control, BHI medium. RF/min: <5 (-), 5-24 (þ) less active, 25-124 ( þþ) middle activity, >125 (þþþ) very active.
Phylogenetic tree is adjusted from Kolsto et al. (ref 26).
E dx.doi.org/10.1021/ac102764v |Anal. Chem. XXXX, XXX, 000–000
component of the bacterial cell walls.
The fact that bacteria are
able to process
D-amino acids led us to hypothesize that bacteria
may express proteases which possess a unique property for
recognizing and hydrolyzing
D-amino acid containing substrates
that can be used for detection and diagnostic purposes.
The BikKam1 substrate appeared to be highly speciﬁc for the
detection of B. anthr acis and its close relatives (Table 2). No
cleavage could be detected in case the substrate was incubated
with culture supernatants of B. subtilis and B. globigii (also known
as B. subtilis var. niger). These two bacterial species are present in
the same branch of the phyloge netic tree of Bacillus.
To obtain more insight in the mechanism of action, several
BikKam1 analogues were designed. The presence of one or more
D-oriented amino acids appeared to be important to maintain the
speciﬁcity of the substrate. Replacement of the
D-Leu by its L-
isomer (BikKam3) led to a signiﬁcant decrease in speciﬁcity. The
BikKam3 substrate was recognized adequately by B. subtilis and
B. globigii proteases, but it was also cleaved by proteases of a
number of other pathogens, including P. aeruginosa (data not
shown). The position of the
D-isomer in the substrate seemed to
be of limited importance; BikKam1 and BikKam2 were cleaved
with similar eﬃciency. MS analysis of the cleaved substrates
revealed that all analyzed BikKams were cleaved directly after the
D-oriented amino acid. Both B. anthracis as well as B. cereus
cleaved the substrates at the same position. Changing the
position of the
D-isomer (BikKam2) or addition of an extra
Leu (BikKam5) had no eﬀec t on the cleavage pattern. Also in
D-Leu was substituted by D-Val (BikKam6) the substrate
was still cleaved directly after the
D-isomer, though it was with
lower eﬃciency. However, sub stitution of
D-Leu by D-Ala
(BikKam8) led to a total loss of cleavage activity. This is probably
due to the fact that Ala and Leu diﬀer more in structu re than Val
and Leu do. Both Leu and Val have two methyl groups in their
structure, whereas only one methyl group is present in the
structure of Ala. Thus, substitution of Leu by Ala has large eﬀects
on the steric design of the substrate.
On the basis of our observations we are tempted to suggest
that the BikKam1 substrate might be useful for the detection of
B. anthracis spores in the so-called “anthrax letters”. For this
purpose the spores had to be triggered into a vegetative state by
spores) or 4 h (10
spores) of preculturing to observe
a signiﬁcant increase in ﬂuorescence in time. No increase in
Figure 1. Detection of anthrax spores by BikKam1. Precultivated B. anthracis (A) and B. subtilis (B) spores were incubated with BikKam1 at 37 °C. After
3 h of preculturing 10
B. anthracis spores an activity of 11 RF/min, with an increase to 87 RF/min after 4 h, was measured (9). For the detection of 10
B. anthracis spores, the spores had to be precultured for 4 h ( 2). No increase in ﬂuorescence was observed in case B. subtilis spores were used (B). Results
are expressed as mean ( SEM (n = 3).
Figure 2. LC-ES MS/MS detection of BikKam1 speciﬁc activity in sera of infected mice. Sera taken from B. anthracis infected mice were incubated with
16 μM BikKam1 substrate for a 2 h cleavage reaction at 37 °C. At 0 (A) and 12 h (B) postinfection (p.i.) no BikKam1 fragments could be detected.
However, at 48 h p.i. a clear peak, identiﬁed by MS/MS as the BikKam1 fragment Lys-Dbc-NH
(MW 396.2), was observed (C).
F dx.doi.org/10.1021/ac102764v |Anal. Chem. XXXX, XXX, 000–000
ﬂuorescence was observed in case B. subtili s, ﬂour, washing
powder, or talc were used. Currently, culture still is the most
common technique to be used in the detection of anthrax.
The relative ease of this method facilitates a higher throughput
compared to immunochemical or PCR methods, which involve
multiple and complex sample prepara tion and assay steps.
One of the methods currently used in the U.S. Postal Oﬃces is
the PCR-based technique of Cepheid, with which it is possible to
detect 30 anthrax spores in 45 min (http://www.cepheid.com/
tests-and-reagents/anthrax/). In contrast, the approach pre-
sented in this study is easy to perform, requiring a minimum of
experimental steps. Compared to other enzyme-based detection
techniques it is fast; yet in 4 h 10
B. anthracis spores can be
detected. Due to the speciﬁc character of the BikKam1 substrate
there is no need for time-consuming enzyme pre-enrichment or
puriﬁcation. We envisage that by using our FRET assay as rapid
prescreening “anthrax letters” can be reliably analyzed within 1
day for the presence of Bacillus spp. in the ﬁeld, without the need
of highly trained personnel. To conﬁrm the outcome of FRET-
mediated testing eventually a B. anthracis speciﬁc PCR or
culturing can be executed in addition.
To explore the opportunities for the assa y to diagnose an
anthrax infection, sera of B. anthracis infected mice were analyzed
using the BikKam1 substrate. Before infection and 12 h p.i. no
BikKam1 degradation fragments could be detected. However, at
48 h after infection BikKam1 fragments could be detected by MS
analysis. At this time point B. anthracis was systemic, and the mice
were clearly aﬀected by the disease. To detect inhalational
anthrax before the onset of severe clinical signs, the infection
needs to be diagnosed before the bacterium is present in the
vascular system. However, in order to detect B. anthracis secreted
enzymes in blood the infection probab ly has to be systemic. In
future experiments bronchoalveolar lavage (BAL) ﬂuid will be
used to detect anthrax protease activity at an earlier stage.
Further characterization and identiﬁcation of the enzyme(s)
involved will enhance the possibilities to improve the current
methodology. In case the responsible enzyme is known, speciﬁc
stimulators of the assay can be added to the FRET assay to
increase the limit of detection. Moreover, the substrate can be
optimized by replacement or addition of amino acids or by the
usage of other conjugates.
More likely dipeptidase AC, the enzyme on which the original
BikKam1 substrate was based, is not involved in cleavage of the
substrates used in this study. Instead of dipeptidase AC we now
hypothesize that the cleavage of the substrate is due to a
peptidase which plays a role in cell wall metabolism of Bacillus
D-amino acids are incorporated in the peptidoglycan
Bacteria release these D-amino acids during their sta-
tionary growt h phase, probably to sync hronize growth inhibition
and PG synthesis.
Alternatively, we cannot exclude that either
FITC or Dabcyl plays a role in enzyme recognition and thereby
sterically hinders dipeptidase AC from A. calcoaceticus and
Brucella spp. from cleaving the BikKam substrates.
A candidate enzyme was recently described by Sela-Abramo-
vich et al. who discovered the cysteine peptidase NlpC/p60
This peptidase is present in sera of anthrax-
Members of the NlpC/p60 family from the genus of Bacillus
have shown to be
D,L-endopeptidases that hydr olyze the D-γ-
glutamyl-meso-diaminopimelate linkage in the cell wall
Moreover, BA1952 orthologs are present in the
secretomes of Bacillus spp. of the cereus group.
BikKam1 substrate was also cleaved by the more remotely related
bacilli such as B. mycoides, B. licheniformis , and B. megaterium.
Probably, these bacteria produce other enzymes that possess the
same functional capacities as the BA1952 peptidase found in B.
In conclusion, we re port a novel enzyme-based approach for
the detection of B. anthracis. In this study it is shown that the new
test can be applied for the detection of B. anthracis spores in
anthrax letters. Our in vivo study in mice might form the basis for
in vivo diagnosis of B. anthracis infection in humans.
We are the ﬁrst to use
D-amino acid containing substrates that
can be used for enzyme-based diagnostic purposes. We feel it
tempting to suggest that, besides B. anthracis, the described
method can be applied in the detection and diagnosis of other
D-amino acid containing FRET substrates
can be designed for the detection and/or identiﬁcation of other
’ AUTHOR INFORMATION
*Phone: (þ31) 10 703 2177. Fax: (þ31) 10 703 3875. E-mail:
This work was ﬁnancially supported by the V502 program of
the Dutch Ministry of Defence. We thank Ingrid Visser-Voskamp
for her help on the preparation of the B. anthracis and B. subtilis
(1) Inglesby, T. V.; Henderson, D. A.; Bartlett, J. G.; Ascher, M. S.;
Eitzen, E.; Friedlander, A. M.; Hauer, J.; McDade, J.; Osterholm, M. T.;
O’Toole, T.; Parker, G.; Perl, T. M.; Russell, P. K.; Tonat, K. JAMA, J.
Am. Med. Assoc. 1999, 281, 1735–1745.
(2) Mock, M.; Fouet, A. Annu. Rev. Microbiol. 2001 , 55, 647–671.
(3) Mignot, T.; Mock, M.; Fouet, A. Mol. Microbiol. 2003,
(4) Green, B. D.; Battisti, L.; Koehler, T. M.; Thorne, C. B.; Ivins,
B. E. Infect. Immun. 1985, 49, 291–297.
(5) Mikesell, P.; Ivins, B. E.; Ristroph, J. D.; Dreier, T. M. Infect.
Immun. 1983, 39, 371–376.
(6) Uchida, I.; Sekizaki, T.; Hashimoto, K.; Terakado, N. J. Gen.
Microbiol. 1985, 131, 363–367.
(7) Kobiler, D.; Weiss, S.; Levy, H.; Fisher, M.; Mechaly, A.; Pass, A.;
Altboum, Z. Infect. Immun. 2006, 74, 5871–5876.
(8) Mabry, R.; Brasky, K.; Geiger, R.; Carrion, R., Jr.; Hubbard, G. B.;
Leppla, S.; Patterson, J. L.; Georgiou, G.; Iverson, B. L. Clin. Vaccine
Immunol. 2006, 13, 671–677.
(9) Selinsky, C. L.; Whitlow, V. D.; Smith, L. R.; Kaslow, D. C.;
Horton, H. M. Biologicals 2007, 35, 123–129.
(10) Edwards, K. A.; Clancy, H. A.; Baeumner, A. J. Anal. Bioanal.
Chem. 2006, 384,73–84.
(11) Manaﬁ, M.; Kneifel, W.; Bascomb, S. Microbiol. Rev. 1991,
(12) Perpetuo, E. A.; Juliano, L.; Juliano, M. A.; Fratelli, F.; Prado,
S. M.; Pimenta, D. C.; Lebrun, I. Protein Pept. Lett. 2008, 15, 1100–1106.
(13) Yolken, R. H. Clin. Chem. 1981, 27, 1490–1498.
(14) Rawlings, N. D.; Barrett, A. J.; Bateman, A. Nucleic Acids Res.
2010, 38, D227–D233.
(15) Rasooly, R.; Stanker, L. H.; Carter, J. M.; Do, P. M.; Cheng,
L. W.; He, X.; Brandon, D. L. Int. J. Food Microbiol. 2008, 126, 135–139.
(16) Wikstrom, M.; Potempa, J.; Polanowski, A.; Travis, J.; Renvert,
S. J. Periodontol. 1994, 65,47–55.
G dx.doi.org/10.1021/ac102764v |Anal. Chem. XXXX, XXX, 000–000
(17) Boyer, A. E.; Quinn, C. P.; Woolﬁtt, A. R.; Pirkle, J. L.;
McWilliams, L. G.; Stamey, K. L.; Bagarozzi, D. A.; Hart, J. C., Jr.; Barr,
J. R. Anal. Chem. 2007, 79, 8463–8470.
(18) Aravamudhan, S.; Joseph, P. J.; Kuklenyik, Z.; Boyer, A. E.; Barr,
J. R. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2009, 1071–1074.
(19) Cummings, R. T.; Salowe, S. P.; Cunningham, B. R.; Wiltsie, J.;
Park, Y. W.; Sonatore, L. M.; Wisniewski, D.; Douglas, C. M.; Hermes,
J. D.; Scolnick, E. M. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 6603–6606.
(20) Duriez, E.; Goossens, P. L.; Becher, F.; Ezan, E. Anal. Chem.
2009, 81, 5935–5941.
(21) Zakharova, M. Y.; Kuznetsov, N. A.; Dubiley, S. A.; Kozyr, A. V.;
Fedorova, O. S.; Chudakov, D. M.; Knorre, D. G.; Shemyakin, I. G.;
Gabibov, A. G.; Kolesnikov, A. V. J. Biol. Chem. 2009, 284, 17902–17913.
(22) Kimura, R. H.; Steenblock, E. R.; Camarero, J. A. Anal. Biochem.
(23) Holtje, J. V. Microbiol. Mol. Biol. Rev. 1998 , 62, 181–203.
(24) Adachi, H.; Tsujimoto, M. J. Biochem. 1995, 118, 555–561.
(25) Shockman, G. D.; Daneo-Moore, L.; Kariyama, R.; Massidda,
O. Microb. Drug Resist. 1996, 2,95–98.
(26) Kolsto, A. B.; Tourasse, N. J.; Okstad, O. A. Annu. Rev.
Microbiol. 2009, 63, 451–476.
(27) Hodges, L. R.; Rose, L. J.; O’Connell, H.; Arduino, M. J.
J. Microbiol. Methods 2010, 81, 141–146.
(28) Beecher, D. J. Appl. Environ. Microbiol. 2006,
(29) Colston, B. W., Jr. Discovery Med. 2003, 3,35–37.
(30) Lam, H.; Oh, D. C.; Cava, F.; Takacs, C. N.; Clardy, J.; de
Pedro, M. A.; Waldor, M. K. Science 2009, 325, 1552–1555.
(31) Sela-Abramovich, S.; Chitlaru, T.; Gat, O.; Grosfeld, H.; Cohen,
O.; Shaﬀerman, A. Appl. Environ. Microbiol. 2009, 75, 6157–6167.
(32) Anantharaman, V.; Aravind, L. Genome Biol. 2003, 4, R11.