Development of a dual fluorogenic and chromogenic dipeptidyl peptidase IV substrate

Article abstract of DOI:10.1016/j.bmcl.2006.02.045

A new far-red dual fluorogenic and chromogenic substrate, 5-glycylprolylglycylprolyl-9-di-3-sulfonyl-propylaminobenza[a]phenoxazonium perchlorate (GPGP-2SBPO), was developed for dipeptidyl peptidase IV (DPP-IV) sensing. The glycylprolylglycylprolyl tetrapeptide was chosen as the recognition sequence due to its stability under physiological conditions. In contrast, the truncated substrate, GP-2SBPO, containing only a glycylprolyl peptide, is unstable. Proteolysis of GPGP-2SBPO was assayed by monitoring the absorbance and fluorescence signals from the released fluorochrome, 2SBPO, at 625 and 670 nm, respectively.

Full text of DOI:10.1016/j.bmcl.2006.02.045

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2599–2602
Development of a dual fluorogenic and chromogenic
dipeptidyl peptidase IV substrate
Nan-Hui Ho, Ralph Weissleder and Ching-Hsuan Tung*
Center for Molecular Imaging Research, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA 02129, USA
Received 26 January 2006; revised 13 February 2006; accepted 15 February 2006
Available online 6 March 2006
Abstract—A new far-red dual fluorogenic and chromogenic substrate, 5-glycylprolylglycylprolyl-9-di-3-sulfonyl-propylaminoben-
za[a]phenoxazonium perchlorate (GPGP-2SBPO), was developed for dipeptidyl peptidase IV (DPP-IV) sensing. The glyc-
ylprolylglycylprolyl tetrapeptide was chosen as the recognition sequence due to its stability under physiological conditions. In
contrast, the truncated substrate, GP-2SBPO, containing only a glycylprolyl peptide, is unstable. Proteolysis of GPGP-2SBPO
was assayed by monitoring the absorbance and fluorescence signals from the released fluorochrome, 2SBPO, at 625 and 670 nm,
respectively.
ꢀ
2006 Elsevier Ltd. All rights reserved.
Dipeptidyl peptidase IV (DPP-IV) is an enzyme of con-
siderable biomedical interest because it is upregulated in
for improved tissue penetration and have low back-
ground in biological samples. For in vivo use, the ideal
fluorochromes would have emission maxima between
1
–3
certain diseases.
DPP-IV is a serine protease, which
9
cleaves Xxx-Pro dipeptides from the N-termini of poly-
peptides and proteins with a penultimate proline resi-
due. In vitro, the activity of DPP-IV is often assayed
using chromogenic substrates, such as glycylproline
650 and 900 nm.
Recently, we reported a new type of water-soluble far-
red fluorogenic molecule containing a disulfonated
4
5
10
b-naphthylamide and glycylproline p-nitroanilide, or
benzo[a]phenoxazine (2SBPO) scaffold. Peptide conju-
fluorogenic probes, such as (Ala-Pro) -cresyl violet and
6
gates of this fluorochrome are weakly fluorescent but
display strong fluorescence emission at 670 nm follow-
ing proteolytic release of the fluorochrome. In addition,
dramatic changes in absorbance at 625 nm are observed.
In the present report, the synthesis and preliminary bio-
chemical studies of a DPP-IV sensitive probe based on
the above scaffold are detailed.
2
(
Ala-Pro) -Rhod110. Although these probes are com-
2
monly available, they have poor aqueous solubility
and frequently require organic solvents such as DMSO
for solubilization. The use of organic co-solvents in
enzymatic assays for DPP-IV may lead to unreliable
results. For example, DPP-IV activity has been reported
to be modulated by DMSO, with 1% DMSO reducing
7
activity by 50%. Moreover, both (Ala-Pro) -cresyl
2
All amino acids were purchased from Novabiochem
(San Diego, CA). The parent dye 9-di-3-sulfonyl-propyl-
aminobenzo[a]phenoxazonium perchlorate (2SBPO),
violet and (Ala-Pro) -Rhod110 are unstable in aqueous
2
solution and undergo hydrolysis in the absence of
6
10
DPP-IV. Therefore, fluorogenic (Ala-Pro) -cresyl violet
2
synthesized as previously described, was adjusted to
and (Ala-Pro) -Rhod110 are of limited use for sensing
2
pH 8.5 and lyophilized prior to use. Dipeptidyl amino-
peptidase (DPP-IV, EC 3.4.14.5, from porcine kidney)
and other common reagents were obtained from Sigma
Aldrich (St. Louis, MO) and used without further puri-
fication. Absorbance spectra were measured on a Varian
Cary 50-Bio UV/Visible spectrophotometer (Palo Alto,
CA). Fluorescence spectra were carried out on a
F-4500 spectrofluorometer (Hitachi, Danbury, CT).
Fluorescence measurements for stability studies and en-
zyme kinetics were determined on a monochromator-
based microplate detection system (Safire2, TECAN,
8
,6
DPP-IV activity in live cells or in vivo. There is thus
an urgent need for the development of improved probes
which are stable in aqueous solution, are water-soluble,
have far-red or near-IR excitation and emission maxima
Keywords: Dipeptidyl peptidase; DPP-IV; Fluorogenic; Chromogenic;
Fluorescent assay.
*
Corresponding author. Tel.: +1 617 726 5779; fax: +1 617 726
708; e-mail: tung@helix.mgh.harvard.edu
5
0
960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.02.045

2600
N.-H. Ho et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2599–2602
San Jose, CA) with excitation/emission band width of
0/20 nm. MALDI-TOF mass spectra were measured
on a Voyager linear mass spectrometer (PE Biosystems;
Framingham, MA). The kinetic parameters (Km and
Vmax) were determined using the direct linear plot of
Eisenthal and Cornish-Bowden.
GP-2SBPO was initially synthesized as a potential fluor-
ogenic substrate for DPP-IV (Scheme 1). Boc-protected
amino acids were activated in situ by HBTU and cou-
pled sequentially to 2SBPO using diisopropylethylamine
as base. The coupling reactions were typically complete
in 1 h at room temperature with near quantitative con-
version as determined by TLC. The final products were
purified by RP C18 chromatography.
2
Synthesis of Gly-Pro-2SBPO (GP-2SBPO). To a stirred
solution of N-a-t-Boc-L-proline (21 mg, 0.1 mmol) in
DMF (0.8 mL) were added 2SBPO (20 mg,
As expected, the synthesized dipeptide substrate,
GP-2SBPO, was only weakly fluorescent. Initial studies
with DPP-IV under the physiological buffer conditions
showed high fluorescent signal after incubation. Howev-
er, when the substrate was incubated under identical
conditions in the absence of DPP-IV, a similar increase
in fluorescence was also detected. Therefore, the stability
of GP-2SBPO in aqueous solution was systematically
examined over the pH range from 4 to 8 in 0.1 M Tris
buffers. After 3 h of incubation at 27 ꢁC, GP-2SBPO
was found only to be stable below pH 5; when the pH
was above 6, probe hydrolysis was significant (Fig. 1).
0
0
.033 mmol), N-hydroxybenzotriazole (HOBT) (37 mg,
.1 mmol), 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyl-
uronium hexafluorophosphate (HBTU) (14 mg,
.1 mmol), and N,N-diisopropylethylamine (DIPEA)
26 mg, 0.2 mmol). The solution was stirred at room
0
(
temperature for 1 h and monitored by silica gel TLC.
When the reaction was complete, the crude product
was precipitated with diethyl ether and the precipitate
was treated with TFA (1 mL) to remove the Boc-pro-
tecting group. After 15 min, diethyl ether was added
and the precipitate was collected. The crude product
was then coupled with Boc-glycine using the previous
coupling conditions. The product was precipitated with
diethyl ether and purified on a 10 g RP C18 cartridge
N
(Waters) eluting with 20% acetonitrile in water to give
GP-2SBPO as a blue solid (22 mg, 91%). MS (MAL-
+
H N
O
N
i), (ii)
N
2
+
DI-TOF), calcd for C H N O S (M+H) : 661.28.
Found: 661.75.
SO H
3
2
9
35
5
9 2
2SBPO
SO H
3
(
Synthesis of Gly-Pro-Gly-Pro-2SBPO (GPGP-2SBPO).
GPGP-2SBPO was synthesized and purified using the
same procedure as described for GP-2SBPO. Overall
yield was 90%. MS (MALDI-TOF), calcd for
+
H N-Pro-HN
2
O
N
+
SO H
C H N O S (M+H) : 815.49. Found: 815.59. The
7
3
3
6
45
11 2
molar extinction coefficient of GPGP-2SBPO at
6
SO H
3
(iii), (ii)
À1
30 nm is 7000 (M cm) in 0.1 M PBS buffer.
Aqueous stability of GPGP-2SBPO and GP-2SBPO.
Stock solutions of compounds GPGP-2SBPO and GP-
N
O
2
1
0
SBPO were prepared in water at a concentration of
· 10 M. The stock solutions were then added to
.1 M, pH 4.0–8, Tris buffers. The final concentration
+
N
À4
H N-Gly-Pro-HN
2
SO H
3
À6
GP-2SBPO
SO3H
of the solutions was 1 · 10 M. The fluorescence emis-
sion at 670 nm was measured with excitation at 630 nm
using a plate reader for 3 h.
Scheme 1. Synthetic scheme of GP-2SBPO. Reagents and conditions:
(
(
i) 3 equiv Boc-Pro-OH, 3 equiv HOBT/HBTU, 6 equiv DIPEA/DMF;
ii) TFA; (iii) 3 equiv Boc-Gly-OH, 3 equiv HOBT/HBTU, 6 equiv
Proteolysis of GPGP-2SBPO by DPP-IV. A solution of
DPP-IV (10 lL, 18 mU) was added to GPGP-2SBPO
DIPEA/DMF.
(
200 lL, 25 lM) in 0.1 M HEPES buffer (pH 7.4) con-
taining 140 mM NaCl, 10 mM KCl, and 0.1% bovine
serum albumin at 37 ꢁC. The increase in fluorescence
emission at 670 nm was measured using a fluorescence
plate reader with excitation at 630 nm. The absorption
increases at 625 nm were measured using a UV–vis spec-
trophotometer from t = 0 to 100 min. The same condi-
tions were applied to the control sample solution
without adding DPP-IV.
3000
GP-SBPO
2500
2000
1500
1000
GPGP-SBPO
500
0
Water-soluble 2SBPO was chosen for development of
the dual fluorogenic and chromogenic substrates be-
cause of its superior optical properties in aqueous buffer
as previously reported. Since glycylprolyl containing
peptide substrates for DPP-IV are in common use,
4
5
6
7
8
pH
1
0
Figure 1. Effect of pH on the stability of GP-2SBPO and GPGP-
2SBPO in 0.1 M Tris buffer.

N.-H. Ho et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2599–2602
2601
One possibility that may account for the observed pH-
dependent, non-enzymatic hydrolysis of GP-2SBPO is
via intramolecular cyclization to give cyclo[Gly-Pro]
N
O
+
Gly-Pro-Gly-Pro-HN
N
(
Scheme 2). Cyclization of dipeptides containing a pro-
SO H
3
GPGP-2SBPO
lineamide moiety to anhydrides (piperazine-2,5-diones)
has been shown previously to occur in aqueous solu-
SO H
3
11
tion. For GP-2SBPO, cyclization may be facilitated
by the rigid pyrrolidine ring which helps to position
the glycylamine for nucleophilic attack on the adjacent
DPP-IV
Gly-Pro
1
2
aromatic amide bond (Scheme 2). To corroborate this
hypothesis, mass analysis was performed on the GP-
N
O
+
N
2
0
SBPO samples after incubation for 3 h at 37 ꢁC in
Gly-Pro-HN
SO H
.1 M Tris buffer (pH 7.4) with and without DPP-IV.
GP-2SBPO
3
For the GP-2SBPO sample with DPP-IV, mass peaks
corresponding to linear glycylproline and cyclo[Gly-
Pro] were observed indicating both enzyme catalyzed
hydrolysis and spontaneous cyclization of GP-2SBPO.
On the other hand, only a cyclo[Gly-Pro] mass peak
was found for the GP-2SBPO sample without DPP-IV
indicating spontaneous cyclization of the dipeptide.
SO3H
DPP-IV
+
Gly-Pro
+
Cyclization
cyclo[Gly-Pro]
N
O
+
N
To overcome the problem of this unexpected non-
enzymatic hydrolysis, addition of a second glycylprolyl
sequence to the N-terminus of GP-2SBPO could be an
effective solution. Ideally, the extended peptide chain
would disfavor intramolecular cyclization, while the
initial proteolytic removal of the N-terminal Gly-Pro
by DPP-IV would lead to GP-2SBPO, and subsequent
liberation of free 2SBPO could be triggered by DPP-
IV catalyzed hydrolysis and spontaneous non-enzymatic
cyclization (Scheme 3).
H N
2
SO H
3
2SBPO
SO3H
Scheme 3. Proposed schematic representation for the two-step cleav-
age of GPGP-2SBPO.
2
1
0000
5000
As expected, the GPGP-2SBPO was found to be pH sta-
ble. Unlike GP-2SBPO, no hydrolysis was observed for
GPGP-2SBPO in the absence of DPP-IV between pH 4
and 8 (Fig. 1). The proteolytic hydrolysis of GPGP-
10000
5000
0
2
SBPO by DPP-IV was analyzed by fluorescence and
absorbance measurements. Under physiological condi-
tions (pH 7.4), no fluorescence signal changes were ob-
served for GPGP-2SBPO alone (Fig. 2). When the
probe was incubated with DPP-IV, a steady increase
in fluorescence emission was observed giving sixfold in-
0
20
40
60
80
100
Reaction time (min)
Figure 2. Fluorescence assay for the hydrolysis of GPGP-2SBPO with
DPP-IV (18 mU) in 0.1 M HEPES buffer (pH 7.4).
crease after 60 min. The fluorescence signal generated at
670 nm indicated catalytic cleavage of the amide bond
N
and release of the fluorescent 2SBPO. Similar to the
fluorescence measurements, a sixfold increase in absor-
bance at 625 nm is observed 60 min after treatment with
DPP-IV (Fig. 3). As with GP-2SBPO, the modified
GPGP-2SBPO probe is completely water-soluble and re-
quires no additional organic co-solvent for the fluoro-
genic and chromogenic assays.
O
N
+
N
H
N
NH
O
SO H
3
2
SO H
3
GP-2SBPO
O
Kinetic parameters for the hydrolysis of GPGP-2SBPO
with DPP-IV were further determined under physiolog-
ical conditions (Table 1). The corresponding parameters
O
NH
N
O
+
N
for commercially available Gly-Pro-AMC, (Ala-Pro) -
2
+
N
H N
2
7
O
Rhod110, and Ala-Pro-Rhod110-NH2 are included in
SO H
3
Cyclo[Gly-Pro]
2SBPO
the table for comparison. Under the conditions used,
the dissociation rate constants (Km values) indicate that
GPGP-2SBPO binds more effectively to DPP-IV than
the coumarin-based substrate (Gly-Pro-AMC), but less
SO H
3
Scheme 2. Proposed mechanism for hydrolysis via intramolecular
cyclization to release free 2SBPO.

2602
N.-H. Ho et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2599–2602
1
.8
.6
.4
.2
1
wavelength emission and absorption maxima. A six-fold
increase in fluorescence and absorbance was obtained.
The kinetic parameters indicate that the catalytic efficien-
cy of DPP-IV to the substrate is high. Potentially, the ap-
proach of repeating proteolytic sequence could be applied
to design other protease probes. In addition, since this im-
proved DPP-IV substrate emits at far-red region, it could
have applications for in vitro high-throughput screening
1
1
1
0
0
0
0
.8
.6
.4
.2
0
(HTS) and potentially in vivo imaging.
Acknowledgment
5
00
550
600
650
700
Wavelength (nm)
This research was supported in part by NIH P50-
CA86355, RO1 CA99385 and R21 CA114149.
Figure 3. Absorbance assay for the hydrolysis of GPGP-2SBPO with
DPP-IV (18 mU) in 0.1 M HEPES buffer (pH 7.4). The curves from
bottom to top are at 0, 10, 25, 35, 45, 55, 80, and 95 min, respectively.
References and notes
Table 1. Kinetic parameters of the fluorogenic substrates for DPP IV
1
. Holst, J. J.; Deacon, C. F. Diabetes 1998, 47, 1663.
Substrates
K
m
(lM)
V
max (nM/min)
2. Hartmann, B.; Thulesen, J.; Kissow, H.; Thulesen, S.;
Orskov, C.; Ropke, C.; Poulsen, S. S.; Holst, J. J.
Endocrinology 2000, 141, 4013.
3. Pospisilik, J. A.; Stafford, S. G.; Demuth, H. U.; Brown-
sey, R.; Parkhouse, W.; Finegood, D. T.; McIntosh, C. H.;
Pederson, R. A. Diabetes 2002, 51, 943.
a
Gly-Pro-AMC
(Ala-Pro)
Ala-Pro-Rhod110-NH
210.6 ± 3.6
15.8 ± 2.4
18.4 ± 1.6
56 ± 8
5.2 ± 0.2
0.44 ± 0.14
0.32 ± 0.13
630 ± 25
a
2
-Rhod110
a
2
Gly-Pro-Gly-Pro-2SBPO
Ref. 7.
a
4
. Hopsu-Havu, V. K.; Glenner, G. G. Histochemistry 1966,
, 197.
. Nagatsu, T.; Hino, M.; Fuyamada, H.; Hayakawa, T.;
7
5
effectively than the rhodamine-based substrates. A rath-
er large Vmax value (630 nM/min) for GPGP-2SBPO was
obtained. The selectivity constant kcat/Km value of
Sakakibara, S. Anal. Biochem. 1976, 74, 466.
6. Boonacker, E.; Elferink, S.; Bardai, A.; Fleischer, B.; Van
Noorden, C. J. J. Histochem. Cytochem. 2003, 51, 959.
7. Grant, S. K.; Sklar, J. G.; Cummings, R. T. J. Biomol.
Screen. 2002, 7, 531.
. Van Noorden, C. J.; Boonacker, E.; Bissell, E. R.; Meijer,
A. J.; van Marle, J.; Smith, R. E. Anal. Biochem. 1997,
3
À1 À1
2
.1 · 10 M
s calculated from the kinetic data indi-
cates that the catalytic efficiency of the 2SBPO-based
substrate is in the moderate range.
8
2
. Ntziachristos, V.; Ripoll, J.; Wang, L. V.; Weissleder, R.
52, 71.
In conclusion, we have synthesized a dual fluorogenic and
chromogenic substrate, GPGP-2SBPO, for DPP-IV sens-
ing. The substrate is soluble and stable in aqueous solu-
tion without the need for organic co-solvent. Release of
water-soluble 2SBPO by DPP-IV can be determined by
both fluorometric and colorimetric analysis with long
9
Nat. Biotechnol. 2005, 23, 313.
0. Ho, N. H.; Weissleder, R.; Tung, C. H. Tetrahedron 2006,
2, 578.
1
6
11. Moss, J.; Bundgaard, H. J. Pharm. Pharmacol. 1990, 42, 7.
12. Rydon, H. N.; Smith, P. W. G. J. Chem. Soc. 1956, 3642.