вход по аккаунту


Highly Sensitive Protease Assay Using Fluorescence Quenching of Peptide Probes Based on Photoinduced Electron Transfer.

код для вставкиСкачать
Highly Sensitive Protease Assay Using Fluorescence Quenching of Peptide Probes Based on
Photoinduced Electron Transfer**
Nicole Marm, Jens-Peter Knemeyer, Jrgen Wolfrum,
and Markus Sauer*
The interest in fast and sensitive assays for proteases, that is,
enzymes that specifically cleave peptide bonds, has increased
considerably in the last few years. Two medically important
facts in particular have accelerated the development of
proteolytic assays. One is that proteases are implicated in
more and more diseases. Because of their involvement in
tumor progression and metastasis, for example, matrix metalloproteinases, urokinase plasminogen activator (uPA), and
cathepsins such as cathepsin B and cathepsin D, proteases
play a central role in cancer diagnosis and follow-up of
malignant diseases.[1–6] In addition, viral infections such as
HIV could be detected directly by detection and monitoring
of their own specific proteases. This underscores the need for
new highly sensitive and fast assays for the specific detection
of proteolytic enzymes.[7–11]
To date, several different fluorescence-based assays have
been developed to prove the presence of a specific protease in
a sample using labeled enzyme substrates. Typically, enzyme
substrates, for example, a specific peptide sequence, are
doubly labeled with a donor and an acceptor fluorophore in a
way that ensures efficient fluorescence resonance energy
transfer (FRET).[12, 13] Upon cleavage of the peptide substrate
by a protease the interaction between the donor and acceptor
fluorophore is lost. This enables the direct monitoring of
protease activity by measuring the increase in donor fluorescence intensity. Alternatively, a peptide substrate might be
doubly labeled with the same fluorophore. In aqueous
solution hydrophobic interactions of the fluorophores might
force the peptide into a conformation in which the fluorophores can form non- or only weakly fluorescent dimers.[14–17]
Nevertheless, chemical modifications of peptide substrates
reduce the affinity of the resulting probe to the target
molecule and hence the detection sensitivity of the assay.
Thus, a minimum of chemical modification is favorable for the
design of high-affinity molecular probes for proteases. In
[*] Prof. Dr. M. Sauer
Fakult+t f,r Physik
Angewandte Laserphysik und Laserspektroskopie
Universit+t Bielefeld
Universit+tsstrasse 25, 33615 Bielefeld (Germany)
Fax: (+ 49) 521-106-2958
Dr. N. Marm>, Dr. J.-P. Knemeyer, Prof. Dr. J. Wolfrum
Physikalisch-Chemisches Institut
Universit+t Heidelberg
Im Neuenheimer Feld 253, 69120 Heidelberg (Germany)
[**] The authors thank K. H. Drexhage for the oxazine derivative MR121.
Financial support by the Bundesministerium f,r Bildung, Wissenschaft, Forschung und Technologie is gratefully acknowledged.
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
addition, doubly labeled peptide probes require site-specific
labeling with two extrinsic fluorophores, and synthesis of such
compounds is still relatively complicated and expensive.
Furthermore, incomplete labeling of peptide probes with
only one extrinsic fluorophore, which is not quenched,
complicates highly sensitive assays because of the stronger
background signal. A method that takes advantage of
properties of naturally occurring amino acids instead of a
second extrinsic fluorophore would be more useful.
Here we describe a novel fluorescence assay for the
ultrasensitive and specific detection of proteolytic enzymes in
homogeneous solution using singly labeled peptide probes,
the fluorescence of which is quenched by the substrate itself.
The basic idea of the experiment is that the fluorescence of
suitable fluorophores is efficiently quenched by the amino
acid tryptophan by means of photoinduced electron transfer
(Figure 1).[18–23] All other amino acids are much less efficient
at fluorescence quenching.[18–21]
Figure 1. General principle behind using selective fluorescence quenching of fluorophores by tryptophan residues to monitor the presence of
specific proteases. a) The fluorophore (gray/red) is coupled to one end
of the peptide in close proximity to a tryptophan residue (blue)
located, for example, at the other end of the peptide. The flexibility of
the peptide backbone will lead to conformations in which the fluorophore and the tryptophan residue come into contact. This contact formation results in efficient fluorescence quenching by photoinduced
electron transfer. b) In the presence of a protease that specifically recognizes the peptide sequence located between the fluorophore and
tryptophan residue, contact formation and subsequent fluorescence
quenching is prevented due to specific cleavage. Thus, the fluorescence intensity of the fluorophore increases.
In the first step a suitable fluorescent dye and a
tryptophan residue are incorporated into a peptide. Equilibrium fluctuations of the backbone lead to conformations in
which the fluorophore and the tryptophan residue come into
contact. After excitation of the fluorophore, it either emits a
fluorescence photon (dependent on the fluorescence quantum yield) or the fluorescence is quenched due to an electron
transfer reaction with the tryptophan residue. Therefore, the
fluorescence quantum yield of the peptide conjugate is
reduced depending on the conformational flexibility and the
DOI: 10.1002/anie.200453835
Angew. Chem. Int. Ed. 2004, 43, 3798 –3801
number of amino acids separating the fluorophore and the
tryptophan residue.
Recently,[20] selective fluorescence quenching of a redabsorbing oxazine derivative by tryptophan residues was used
to measure association (contact formation) and dissociation
rates in the range of a few hundred nanoseconds in flexible
peptides at the single-molecule level. The rates determined
are consistent with previously published data on the dynamics
of unstructured peptides obtained with a different ensemble
methodology.[24] The advantage of this fluorescence-quenching-based technique over the previously discussed comprises
the ease of experimental monitoring in water under air with
single-molecule sensitivity. If, however, contact formation
between the fluorophore and tryptophan residue is prevented
because of specific binding of the peptide to a target protein,
for example, binding to an antibody in the case of a peptide
epitope, or cleavage of the peptide by a protease enzyme
(Figure 1), the fluorescence intensity of the fluorophore
increases. This can be used advantageously, for example, to
monitor the presence of p53 autoantibodies in sera of cancer
patients. Two flexible fluorescently labeled peptide epitopes
of the amino terminal transactivation domain of human p53
both carrying a single tryptophan residue can be used, and the
presence of p53 autoantibodies can be detected by an increase
in fluorescence intensity.[19]
To demonstrate the potential of the method for the
development of a fast and highly sensitive protease assay, we
monitored the increase in fluorescence intensity of a simple
dipeptide, lysine–tryptophan, with different fluorescence
labels after addition of carboxypeptidase A (CPA)
(Figure 2).[25, 26] All peptide conjugates investigated so far
Figure 2. Relative fluorescence intensities (Irel) of solutions of differently labeled dipeptides Lys-Trp (10 7 m) versus time (t) in phosphatebuffered saline (PBS, pH 7.4, 25 8C) after addition of 10 9 m carboxypeptidase A (CPA). a) MR121-Lys-Trp (almost identical results were
obtained with ATTO655-Lys-Trp), b) Bodipy-FL-Lys-Trp, and c) Cy5-LysTrp. The inset shows an expanded view of the curve recorded for Cy5Lys-Trp. Excitation was performed at the absorption maximum of each
fluorophore (lex = 667, 650, and 490 nm, respectively). Fluorescence
intensities were recorded at the emission maximum of each fluorophore (lem = 690, 670, and 515 nm, respectively). Control experiments
without CPA show a stable fluorescence signal for several hours.
Angew. Chem. Int. Ed. 2004, 43, 3798 –3801
were substrates for CPA. In agreement with the constants for
intermolecular quenching,[21] the oxazine-labeled conjugates
MR121-Lys-Trp and ATTO655-Lys-Trp show the most pronounced increase in fluorescence intensity (Figure 2 a). The
fluorescence intensity of a 10 7 m peptide solution increases
roughly tenfold upon addition of 10 9 m CPA within four
minutes. The Bodipy-FL conjugate shows a sixfold increase
(Figure 2 b), whereas the fluorescence intensity of the Cy5
conjugate increases only slightly. This is consistent with
previous studies which demonstrated that the fluorescence of
Cy5 is only slightly quenched by tryptophan residues.[21] In
addition, the digestion rate differs significantly and is highest
for the two conjugates MR121-Lys-Trp and ATTO655-LysTrp (Figure 2 a).
The influence of the fluorophore structure on enzyme
activity is unknown and difficult to estimate, but a direct
interaction between the fluorophore and enzyme could
indeed strongly influence enzyme activity. Our results prove
that the contact-induced interactions between MR121 or
ATTO655 and tryptophan enable the fast and reliable
detection of nanomolar concentrations of CPA in aqueous
solutions. Therefore, we chose MR121 as the reporter
fluorophore in the following experiments.
When MR121 is coupled directly to tryptophan, the
relative fluorescence quantum yield is 0.02.[21] For the
dipeptide MR121-Lys-Trp, the fluorescence quantum yield
increases to 0.10. Therefore, it is expected that the quenching
efficiency will decrease with increasing distance between the
fluorophore and the tryptophan residue in longer peptides
due to less frequent contact formation. However, when nine
amino acids separate MR121 and a tryptophan residue,
relative quantum yields of 0.30 have been measured.[19, 20]
At the same time, the peptide length and structure can
influence enzyme activity. Figure 3 shows the time-dependent
fluorescence intensity of three MR121 peptide conjugates
with an increasing number of glycine residues separating the
fluorophore and tryptophan residue after addition of 10 7 m
CPA. Although the relative fluorescence quantum yields of
Figure 3. Relative fluorescence intensities (Irel) versus time (t)
(lex = 640 nm, lem = 690 nm) of different MR121-labeled peptides
(10 7 m in PBS, pH 7.4 at 25 8C) after addition of 10 7 m CPA.
a) MR121-Lys-Gly-Trp, b) MR121-Lys-Gly2-Trp, c) MR121-Lys-Gly3-Trp.
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
the three labeled peptides are similar due to the high
flexibility of the glycine residues, the enzyme activity
(digestion rate) differs significantly. With an increasing
number of glycine residues the digestion rate of CPA
decreases (Figure 3).
To evaluate the sensitivity of this assay, we measured the
fluorescence intensity of an aqueous solution of MR121-LysTrp in the presence of different concentrations of CPA
(Figure 4). The measurements were performed in pure water
Figure 4. Relative fluorescence intensities (Irel) measured for the peptide MR121-Lys-Trp (10 7 m) versus time (t) after addition of various
amounts of CPA: a) 10 6 m, b) 10 9 m, c) 10 11 m, d) 10 13 m, and
e) 10 15 m. Measurements were performed in pure water at 25 8C
(lex = 640 nm, lem = 690 nm). Data points were measured every 5–20 s
depending on the sample.
Figure 5. Relative fluorescence intensities (Irel) of the peptide MR121Lys-Arg3-Trp (10 7 m) versus time (t) after addition of various amounts
of trypsin: a) 10 9 m, b) 10 10 m, c) 10 11 m, and d) 10 12 m. e) Fluorescence increase upon addition of 10 12 m trypsin monitored on a longer
time scale. Measurements were performed in 50 mm Tris-HCl buffer
(pH 7.0) with 1 mm calcium chloride, at 25 8C (lex = 640 nm,
lem = 690 nm).
to reduce the salt-induced nonspecific adsorption of CPA
molecules on the glass walls of the cuvette. In pure water, a
maximum increase in fluorescence intensity of about 6.5-fold
was achieved after addition of an excess of CPA. At a
concentration of 10 6 m CPA (tenfold excess) the maximum
increase is reached within seconds (Figure 4 a). Even a CPA
concentration of 10 13 m, which correlates to one enzyme
molecule per 106 substrate molecules, effects a 2.5-fold
increase in fluorescence intensity in only ten minutes
(Figure 4 d). For a further increase of detection sensitivity
longer measurement times have presently to be accepted.
However, our data demonstrate that even a CPA concentration of 10 15 m can be detected by a 1.5-fold increase in
fluorescence intensity in approximately 25 minutes (Figure 4 e).
To demonstrate the general applicability of the new assay
we designed a peptide substrate for an endopeptidase. Trypsin
was chosen as the model protease because it is readily
available and often used as a standard in other protease
assays. Trypsin cleaves peptide bonds on the carboxyl side of
lysine and arginine residues.[27] We used MR121-Lys-Arg3-Trp
as the peptide substrate, which has a relative fluorescence
quantum yield of 0.20 in 50 mm Tris-HCl buffer (pH 7.0)
containing 1 mm calcium chloride. As Figure 5 demonstrates,
trypsin concentrations down to 10 11m can be detected easily
within minutes owing to the increase in fluorescence intensity.
For lower concentrations, the fluorescence signal has to be
monitored for several hours (Figure 5 e).
This approach for assaying proteolytic enzymes using
fluorescent-labeled peptide probes, in which fluorescence is
quenched by photoinduced electron transfer between fluorophores and tryptophan residues upon contact formation
offers several advantages. First, the synthesis of large
quantities of singly labelled peptide probes is uncomplicated
and inexpensive compared to the use of doubly labeled
peptide probes. Second, modification of the peptide substrate
is reduced to the minimum, that is, a single reporter
fluorophore, which potentially enables more accurate monitoring of enzyme activity. By monitoring the time-dependent
fluorescence intensity, the presented technique permits realtime analysis of proteolysis. Third, our new protease assay
enables fast, specific, and highly sensitive detection of
proteolytic enzymes with a broad dynamic range (more than
six orders of magnitude) and detection limits below the
picomolar range. Our results raise our hopes that singly
labeled peptide probes can be used advantageously for the
development of new fast and highly sensitive fluorescence
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2004, 43, 3798 –3801
assays for cancer diagnosis and follow-up of malignant
diseases through the detection of specific proteases.
Experimental Section
N-terminal labeling of the peptides with the fluorescent dye was
performed by classical N-hydroxysuccinimidyl ester (NHS ester)
chemistry using standard solvents purchased from Merck (Darmstadt,
Germany). The oxazine derivative MR121 was kindly provided by the
group of Prof. K. H. Drexhage (UniversitBt-Gesamthochschule,
Siegen). ATTO655 (ATTO-TEC, Siegen, Germany), Bodipy-FL
(Molecular Probes, GCttingen, Germany) and Cy5 (Amersham
Pharmacia Biotec, Freiburg, Germany) were purchased as NHS
esters. Fluorophores (1 mmol mL 1) and the synthetic peptides
(1 mmol mL 1) were dissolved in dimethylformamide (DMF). A 10mL (10 nmol) portion of the dye solution was added to 100 mL
(100 nmol, tenfold excess) of peptide and 2 mL diisopropylethylamine
(DIPEA). The solution was incubated for 3 h at room temperature in
the dark. The N-terminal conjugate was purified by reversed-phase
(Hypersil-ODS column) HPLC (Agilent Technologies, Waldbronn,
Germany) using a linear gradient of 0–75 % acetonitrile in 0.1m
aqueous triethylammonium acetate. Molecular weights were determined by mass spectrometry. The labeled peptides are stable for
several months when stored at 20 8C.
Absorption measurements were recorded with a Cary 500 UV/
Vis-NIR spectrometer (Varian, Darmstadt, Germany) at room
temperature. Steady-state emission spectra were measured with a
Cary Eclipse fluorescence spectrometer. All measurements were
conducted in PEG-coated standard quartz cuvettes. In all measurements the concentration was kept strictly below 1 mm to avoid
reabsorption and reemission effects. Relative fluorescence quantum
yields were measured with respect to the fluorescence intensity of the
free dye.
Carboxypeptidase A (CPA) is a zinc-containing exopeptidase
secreted by the pancreas. It acts preferentially, but not exclusively, on
peptide bonds adjacent to the C-terminus of amino acid residues with
aromatic or branched side chains at pH 7–8. Peptide bonds involving
glycine, aspartic acid, and glutamic acid are hydrolyzed slowly, while
those involving arginine, proline, and hydroxyproline are not hydrolyzed.[25, 26]
Trypsin is a digestive enzyme, which has a serine residue in the
active center. This endopeptidase hydrolyzes peptide bonds in which
the carbonyl group is connected to the basic amino acids lysine or
arginine. The optimum pH is in the range of 7–9.[27] As storage buffer
0.046 m Tris-HCl buffer, pH 8.0 with 0.0115 m calcium chloride was
used. For digestion experiments, 1 mL of the respective enzyme was
mixed with 600 mL of a 10 7 m solution of the peptide substrates in
PBS buffer (phosphate-buffered saline, pH 7.4).
Received: January 23, 2004 [Z53835]
Keywords: analytical methods · electron transfer · enzymes ·
fluorescence spectroscopy · proteases
Angew. Chem. Int. Ed. 2004, 43, 3798 –3801
[1] O. Simonetti, G. Lucarini, D. Brancorsini, P. Nita, M. L.
Bernardini, G. Biagini, A. Offidani, Cancer 2002, 95, 1963 – 1970.
[2] A. Franchi, M. Santucci, E. Masini, I. Sardi, M. Paglierani, O.
Gallo, Cancer 2002, 95, 1902 – 1910.
[3] A. Kugler, Anticancer Res. 1999, 19, 1589 – 1592.
[4] Y. Mochizuki, S. Tsuda, H. Kanetake, S. Kanda, Oncogene 2002,
21, 7027 – 7033.
[5] N. Harbeck, R. E. Kates, M. P. Look, M. E. Meijer-Van Gelder,
J. G. Klijn, A. Kruger, M. Kiechle, F. Janicke, M. Schmitt, J. A.
Foekens, Cancer 2002, 15, 4617 – 4622.
[6] K. Bajou, J. M. Lewalle, C. R. Martinez, C. Soria, H. Lu, A.
Noel, J. M. Foidart, Int. J. Cancer 2002, 100, 501 – 506.
[7] C. Jumper, E. Cobos, C. Lox, Anticancer Res. 2002, 22, 2073 –
[8] O. D. Liang, T. Chavakis, S. M. Kanse, K. T. Preissner, J. Biol.
Chem. 2001, 276, 28 946 – 28 953.
[9] M. R. Gyetko, S. Sud, G. H. Chen, J. A. Fuller, S. W. Chensue,
G. B. Toews, J. Immunol. 2002, 168, 801 – 809.
[10] J. E. Koblinski, M. Ahram, B. F. Sloane, Clin. Chim. Acta 2000,
291, 113 – 135.
[11] G. Berchem, M. Glondu, M. Gleizes, J. P. Brouillet, F. Vignon,
M. Garcia, E. Liaudet-Coopman, Oncogene 2002, 21, 5951 –
[12] E. D. Matayoshi, G. T. Wang, G. A. Krafft, J. Erickson, Science
1990, 247, 954 – 958.
[13] M. Cavrois, C. de Noronah, W. C. Greene, Nat. Biotechnol. 2002,
20, 1151 – 1154.
[14] M. J. Blackman, J. E. Corrie, J. C. Croney, G. Kelly, J. F.
Eccleston, D. M. Jameson, Biochemistry 2002, 41, 12 244 – 12 252.
[15] B. Z. Packard, A. Komoriya, V. Nanda, L. Brand, J. Phys. Chem.
1998, 102, 1820 – 1827.
[16] B. Z. Packard, D. D. Toptygin, A. Komoriya, L. Brand, Proc.
Natl. Acad. Sci. USA 1996, 93, 11 640 – 11 645.
[17] B. Z. Packard, D. D. Toptygin, A. Komoriya, L. Brand, Methods
Enzymol. 1997, 278, 15 – 23.
[18] R. M. Watt, E. W. Voss, Immunochemistry 1977, 14, 533 – 541.
[19] H. Neuweiler, A. Schulz, A. C. Vaiana, J. C. Smith, S. Kaul, J.
Wolfrum, M. Sauer, Angew. Chem. 2002, 114, 4964 – 4068;
Angew. Chem. Int. Ed. 2002, 41, 4769 – 4773.
[20] H. Neuweiler, A. Schulz, M. BChmer, J. Enderlein, M. Sauer, J.
Am. Chem. Soc. 2003, 125, 5324 – 5330.
[21] N. MarmK, J. P. Knemeyer, M. Sauer, J. Wolfrum, Bioconjugate
Chem. 2003, 14, 1133 – 1139.
[22] M. Sauer, Angew. Chem. 2003, 115, 1834 – 1837; Angew. Chem.
Int. Ed. 2003, 42, 1790 – 1793.
[23] A. Vaiana, H. Neuweiler, A. Schulz, J. Wolfrum, M. Sauer, J. C.
Smith, J. Am. Chem. Soc. 2003, 125, 14 564 – 14 572.
[24] L. J. Lapidus, W. A. Eaton, J. Hofrichter, Proc. Natl. Acad. Sci.
USA 2000, 97, 7220 – 7225.
[25] W. Lipscomb, Proc. Natl. Acad. Sci. USA 1973, 70, 3797 – 3801.
[26] R. H. Bradshaw, L. H. Ericsson, K. A. Walsh, H. Neurath, Proc.
Natl. Acad. Sci. USA 1969, 63, 1389 – 1394.
[27] W. Brown, F. Wold, Biochemistry 1973, 12, 835 – 840.
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Без категории
Размер файла
130 Кб
using, quenching, electro, assays, base, fluorescence, photoinduced, transfer, probes, sensitive, protease, highly, peptide
Пожаловаться на содержимое документа