A time-resolved fluorescence probe for dipeptidyl peptidase 4 and its application in inhibitor screening

Article abstract of DOI:10.1002/chem.201001077

The prevalence of type 2 diabetes is increasing dramatically throughout the world. Recently, dipeptidyl peptidase 4 (DPP4) was identified as a potential antidiabetes target. Many DPP4 inhibitors, such as sitagliptin and vildagliptin, have been developed a

Full text of DOI:10.1002/chem.201001077

FULL PAPER
DOI: 10.1002/chem.201001077
A Time-Resolved Fluorescence Probe for Dipeptidyl Peptidase 4 and Its
Application in Inhibitor Screening
Mitsuyasu Kawaguchi,[a, b] Takayoshi Okabe,[c] Takuya Terai,[a, b] Kenjiro Hanaoka,[a, b]
Hirotatsu Kojima,[b, c] Izumi Minegishi,[c] and Tetsuo Nagano*[a, b, c]
Abstract: The prevalence of type 2 dia-
betes is increasing dramatically
throughout the world. Recently, dipep-
tidyl peptidase 4 (DPP4) was identified
cence, thus giving rise to many false-
positive and false-negative results.
Therefore, we have designed and syn-
thesised a novel DPP4 probe (Gly-Pro-
BCD-Tb; Gly=glycine, Pro=proline,
andBCD defines the backbone of the
probe comprising an aniline derivative
as on/off switch, a 7-amino-4-methyl-
2(1H)-quinolinone (cs-124) as antenna
with Gly-Pro-BCD-Tb showed high
sensitivity and reliability in the inhibi-
tory assay relative to Gly-Pro-MCA
(MCA=4-methylcoumarin-7-amide), a
conventional fluorescence probe for
DPP4. Further, we employed our
probe for high-throughput DPP4 inhib-
itor screening with 3841 randomly se-
lected compounds and found that epi-
bestatin, an epimer of bestatin (a well-
known anticancer drug and general
aminopeptidase inhibitor), showed
dose-dependent DPP4 inhibitory activi-
ty. Interestingly, bestatin did not exhib-
it DPP4 inhibitory activity. We believe
that this screening system will be useful
for the discovery of DPP4 inhibitors
with novel structural scaffolds.
as
a
potential antidiabetes target.
Many DPP4 inhibitors, such as sitaglip-
tin and vildagliptin, have been devel-
oped and marketed, but superior thera-
peutic agents are still required. There-
fore, we have developed new method-
ology for screening of DPP4 inhibitors.
Absorption-based measurements with
para-nitroaniline or fluorescence-based
measurements with the coumarin de-
rivative 7-amino-4-methylcoumarin are
often used for the screening of pro-
tease inhibitors, including DPP4 inhibi-
tors, but these strategies are not suffi-
ciently sensitive because of interfering
background absorption and fluores-
moiety, and
a
diethylenetriamine-
N,N,Nꢀ,Nꢀꢀ,Nꢀꢀ-pentaacetic acid (DTPA)
as chelator moiety, Tb=terbium) for
time-resolved
fluorescence
(TRF)
measurements. TRF measurements
Keywords: diabetes · fluorescence
spectroscopy
·
high-throughput
screening · lanthanides · lumines-
cence · proteases
Introduction
An estimated 246 million people worldwide have type 2 dia-
betes, and this number is predicted to grow to 366 million
by 2030.[1] Treatment of type 2 diabetes generally begins
with oral antidiabetic agents, such as biguanides and sulfo-
nylureas,[2] but these agents have serious adverse effects,
such as lactic acidosis and hypoglycemia. So, a novel thera-
peutic strategy for type 2 diabetes is desired. Incretin hor-
mones, such as glucagon-like peptide 1 (GLP-1) and glu-
cose-dependent insulinotropic polypeptide, are candidates
for the treatment of type 2 diabetes.[3] These hormones are
released from the gut in response to food uptake and play
an important role in glucose homeostasis. For example,
GLP-1 stimulates insulin biosynthesis and secretion in pan-
creatic b-cells and also inhibits glucagon release from pan-
creatic a-cells. Moreover, GLP-1 regulates insulin in a strict-
ly glucose-dependent manner, that is, it is ineffective during
[a] M. Kawaguchi, T. Terai, Dr. K. Hanaoka, Prof. Dr. T. Nagano
Graduate School of Pharmaceutical Sciences
The University of Tokyo, 7-3-1, Hongo
Bunkyo-ku, Tokyo 113-0033 (Japan)
Fax : (+81)3-5841-4855
E-mail: tlong@mol.f.u-tokyo.ac.jp
[b] M. Kawaguchi, T. Terai, Dr. K. Hanaoka, Dr. H. Kojima,
Prof. Dr. T. Nagano
CREST (Japan) Science and Technology Agency
4-1-8 Honcho Kawaguchi
Saitama 332-0012 (Japan)
[c] Dr. T. Okabe, Dr. H. Kojima, I. Minegishi, Prof. Dr. T. Nagano
Chemical Biology Research Initiative
The University of Tokyo, 7-3-1, Hongo
Bunkyo-ku, Tokyo (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001077.
Chem. Eur. J. 2010, 16, 13479 – 13486
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
13479

hypoglycemia. Therefore, GLP-1 therapy may be attractive
for treating diabetes without risk of hypoglycemia. Howev-
er, active GLP-1ACHTUNGTRENNUNG(7–36) amide is promptly hydrolysed in the
blood through the action of dipeptidyl peptidase 4 (DPP4),
a serine protease. DPP4 cleaves N-terminal Xxx-Pro dipep-
tides from polypeptides with a proline (Pro) residue at the
latter, the excitation wavelength is too short for screening
because of the use of the aniline moiety as an antenna.
Thus, these methods are not suitable for simple and sensitive
screening. Therefore, we focused on a simple quenching
principle of photo-induced electron transfer (PeT)[11] previ-
ously reported by our group. In that study, we reported the
measurement of leucine aminopeptidase (LAP) activity with
a TRF probe. However, high-throughput screening (HTS)
with a TRF probe based on the PeT mechanism has not
been reported yet. We expected that HTS with a TRF probe
would offer high sensitivity and good reliability. Herein, we
report the development of a DPP4 probe based on the PeT
mechanism and its application to HTS for DPP4 inhibitors.
We found an interesting DPP4 inhibitor among 3841 sam-
ples screened with this HTS.
penultimate position and generates an inactive GLP-1ACTHUNTGRNEUNG(9–36)
amide. Therefore, DPP4 inhibitors are expected to increase
the half-life of active GLP-1 in the blood and to prolong the
beneficial effects of incretin hormones. So far, several DPP4
inhibitors have been reported,[5–8] and some have already
been marketed as therapeutic agents for type 2 diabetes. For
example, sitagliptin (Januvia, Merck),[5] vildagliptin (Galvus,
Novartis),[6] and saxagliptin (Onglyza, Bristol–Myers
Squibb/AstraZeneca),[7] which lack the a-amino acid moiety
of early DPP4 inhibitors, are available.
Methods to screen DPP4 inhibitors include absorption
measurements based on the liberation of para-nitroani-
line[6,7,9] and fluorescence measurements with coumarin de-
rivatives.[5] HPLC analysis of inactive GLP-1 generated by
DPP4 is also performed as a gold-standard method because
it employs the native substrate and reaction. However, a
sensitive high-throughput method is still needed to find
potent and specific inhibitors by screening large numbers of
compounds with low levels of false-positive and false-nega-
tive results. Generally, background fluorescence derived
from the microplates and scattered excitation light limit the
sensitivity in inhibitor screening with 96- or 384-well micro-
plates. These phenomena are especially problematic when
fluorophores such as coumarin derivatives are used because
coumarin derivatives are excited by short-wavelength light.
However, coumarin derivatives are often used as fluorescent
substrates for proteases. Luminescent lanthanide complexes
offer an attractive solution to these problems. If an appro-
priate chelator is used, a lanthanide (especially Tb3+ or
Eu3+ ions) complex can show strong luminescence with an
extraordinarily long luminescence lifetime on the order of
milliseconds in contrast to typical organic fluorophores,
which have a short fluorescence lifetime in the nanosecond
region. By taking advantage of this feature, the influence of
short-lived background fluorescence and scattered light can
be reduced to a negligible level by means of time-resolved
fluorescence (TRF) measurements (see Figure S1 in the
Supporting Information). In TRF measurements, the fluo-
rescence signal is collected for a certain gate-time after an
appropriate delay following a pulsed excitation. By employ-
ing this method, it is easy to distinguish long-lived lantha-
nide luminescence from other signals. Furthermore, lantha-
nide complexes have a large Stokes shift (>200 nm), which
also helps to reduce background signals.[10,11]
Results and Discussion
Probe design and strategy: We developed a lanthanide-
based luminescence probe for DPP4 based on the PeT
mechanism (Scheme 1). The probe consists of three moiet-
Scheme 1. Schematic representation of the DPP4 probe. Off/on switch
moiety: aniline derivative; antenna moiety: 7-amino-4-methyl-2ACHTUNGTRENNUNG(1H)-qui-
nolinone (cs124); chelator moiety: diethylenetriamine-N,N,N’,N’’,N’’-pen-
taacetic acid (DTPA). Gly=glycine, Pro=proline.
ies, which are an antenna, an off/on switch, and a chelator.
The substrate peptide sequence of DPP4 glycylproline was
attached to the amino group of the luminescence off/on
switch moiety. Lanthanide metal ions Tb3+ and Eu3+ were
selected as they show long-lived luminescence, that is, after
an enzymatic reaction with DPP4 glycylproline amide is
converted into the aniline derivative NH2-BCD-Tb, which
has a luminescence intensity that is greatly decreased
through the a-PeT (acceptor-excited PeT) mechanism.[10]
The probes Gly-Pro-BCD-Tb and Gly-Pro-BCD-Eu (where
BCD defines the backbone of the probe comprising an ani-
line derivative as on/off switch, a 7-amino-4-methyl-2(1H)-
quinolinone (cs-124) as antenna moiety, and a diethylenetri-
amine-N,N,Nꢀ,Nꢀꢀ,Nꢀꢀ-pentaacetic acid (DTPA) as chelator
moiety) were synthesised in six steps (Scheme 2). A stan-
dard substrate of DPP4 is Gly-Pro-MCA, which is a conven-
Various protease activity assays based on TRF measure-
ments have been reported, such as time-resolved Fçrster
resonance energy transfer (TR-FRET) by Karvinen et al.[12]
and a luminescence off/on strategy that employed a simple
aniline structure by Mizukami et al.[13] However, there are
problems with these strategies. In the case of the former, the
synthesis of the probe is very laborious; in the case of the
13480
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13479 – 13486

Development of a Lanthanide-Based Luminescence Probe
FULL PAPER
the enzymatic reaction (FI30 min
/
FI0 min =12.5:1, data not shown).
The kinetic parameters of
Gly-Pro-BCD-Tb and Gly-Pro-
MCA were determined from
Michaelis–Menten
plots
(Table 1). The Km and kcat
values of Gly-Pro-BCD-Tb
were 33.6 mm and 28.3 sꢀ1, re-
spectively. The kcat/Km value
was 0.85 mmꢀ1 sꢀ1. The Km, kcat,
and kcat/Km values for Gly-Pro-
BCD-Tb were identical with
those values of Gly-Pro-MCA
(34.0 mm,
25.9 sꢀ1,
respectively).
and
0.76 mmꢀ1 sꢀ1,
These results indicate that the
rates of the enzymatic reaction
of Gly-Pro-BCD-Tb and Gly-
Pro-MCA are almost the same
when the substrate concentra-
tion is sufficiently low. Thus,
Gly-Pro-BCD-Tb is a good sub-
strate for DPP4 from the view-
point of kinetic parameters.
The conversion of Gly-Pro-
BCD-Tb into NH2-BCD-Tb
was confirmed by HPLC analy-
sis (Figure 1D). This outcome
indicates that the enzymatic re-
action proceeded as expected
(Figure 1D).
Scheme 2. Synthesis of Gly-Pro-BCD-Tb and Gly-Pro-BCD-Eu. Reagents and conditions: a) Boc2O, NEt3,
EtOH, 608C!RT (quant. yield); b) 2, HOBT, HBTU, DIEA, DMF, RT (43%); c) piperidine, DMF, RT
(93%); d) DTPA bisanhydride, DIEA, DMF, RT (50%); e) CH2Cl2/TFA, 08C (83%); f) LnCl3·6H2O, MeOH,
RT (quant. yield). Boc=tert-butoxycarbonyl, DIEA=N,N-diisopropylethylamine, Fmoc=9-fluorenylmethoxy-
carbonyl, HBTU=O-benzotriazole-N,N,N’,N’-tetramethyluronium hexafluorophosphate , HOBT=1-hydroxy-
benzotriazole, TFA=trifluoroacetic acid.
tional coumarin-based substrate, and was also synthesised
(Scheme 3).
Table 1. Kinetic parameters of Gly-Pro-BCD-Tb and Gly-Pro-MCA.[a]
Substrate
Gly-Pro-BCD-Tb
Gly-Pro-MCA
Km [mm]
33.6
5.9
28.3
0.85
34.0
5.4
25.9
0.76
Photochemical properties of the probes: The enzymatic re-
actions of Gly-Pro-BCD-Tb and Gly-Pro-BCD-Eu with
DPP4 were monitored. Both probes were good substrates
for DPP4 and large changes in the luminescence intensity
were observed after the enzymatic reaction (Figure 1A–C).
We adopted Gly-Pro-BCD-Tb as a DPP4 substrate in the
following experiments because it showed a greater change
in luminescence intensity (FI0 min/ FI30 min =55.3:1) than Gly-
Pro-BCD-Eu (FI0 min/FI30 min =19.5:1). We also confirmed
that Gly-Pro-MCA showed an increase in fluorescence after
Vmax [mmminꢀ1
]
kcat [sꢀ1
]
kcat/Km [mmꢀ1 sꢀ1
]
[a] All data were obtained at pH 8.0 (50 mm HEPES buffer).
Comparison of detection sensitivity: Next, we compared the
sensitivity of Gly-Pro-BCD-Tb and Gly-Pro-MCA (10 and
100 nm, respectively) by using
a
microplate reader
(Figure 2). We measured the inhibitory activity of diprotin
A,[14] a conventional DPP4 inhibitor, with the two above
Scheme 3. Synthesis of Gly-Pro-MCA. Reagents and conditions: a) 2, HATU, DIEA, DMF, RT (crude product); b) CH2Cl2/TFA, 08C (8% over 2 steps).
MCA=4-methylcoumarin-7-amide, HATU=O-(7-azabenzotriazol-1-yl)tetramethyluronium hexafluorophosphate.
Chem. Eur. J. 2010, 16, 13479 – 13486
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13481

T. Nagano et al.
Thus, we could detect even
weak inhibitors, which are often
seen in initial screening, with
Gly-Pro-BCD-Tb.
Further evaluation of Gly-Pro-
BCD-Tb: We next examined
the usefulness of Gly-Pro-BCD-
Tb in a HTS system relative to
Gly-Pro-MCA. First,
1
and
10 nm Gly-Pro-BCD-Tb were
sufficient for the TRF measure-
ments with a microplate reader,
whereas 100 nm Gly-Pro-MCA
was needed for general fluores-
cence
measurements.
This
result is due to the low back-
ground luminescence of TRF
measurements and relatively
high background fluorescence
of ordinary fluorescence mea-
surement, as already noted. The
difference in IC50 values of di-
Figure 1. Fluorescence spectra of 5 mm solution of Gly-Pro-BCD-Tb (A) and Gly-Pro-BCD-Eu (B) at 0, 2, 4, 6,
8, 10, 14, 20, and 30 min after the addition of 7 nm (final) of DPP4. The reactions were performed in 50 mm
HEPES buffer (pH 8.0) containing 1% dimethyl sulfoxide (DMSO) at room temperature. Excitation wave-
length was l=330 nm. C) Time course of the luminescence intensity of Gly-Pro-BCD-Tb with DPP4 (7 nm).
Excitation and emission wavelengths were l=330 and 546 nm, respectively. D) HPLC charts of the enzymatic
reaction of Gly-Pro-BCD-Tb to afford NH2-BCD-Tb: a) A solution of Gly-Pro-BCD-Tb, b) a solution of NH2-
BCD-Tb, c) reaction mixture after the enzymatic reaction, d) mixed solution of (b) and (c). The chromatogram
was monitored at l=330 nm for all the spectra. Stationary phase: reversed-phase column (Inertsil ODS-3 4.6ꢁ
250 mm; GL Sciences). HPLC analysis: eluent A (100 mm triethylammonium acetate (TEAA) buffer) and
eluent B (acetonitrile/H2O 80:20); A/B 90:10!70:30 (30 min).
protin
A and sitagliptin (a
novel DPP4 inhibitor)[5] are
shown in Figure 3 and Table 2,
as determined by using 1 and
10 nm Gly-Pro-BCD-Tb and
100 nm Gly-Pro-MCA. The
IC50 values depend on the affin-
ity and the concentration of
Table 2. IC50 values of sitagliptin and diprotin A with Gly-Pro-BCD-Tb
and Gly-Pro-MCA.[a]
Sitagliptin [nm]
Diprotin A [mm]
[b]
Gly-Pro-BCD-Tb (1 nm)
Gly-Pro-BCD-Tb (10 nm)
Gly-Pro-MCA (100 nm)
2.0
6.0
15.5
–
9.4
34.1
[a] All data were obtained at pH 8.0 (50 mm HEPES buffer). [b] Not de-
termined.
each probe (Figure 3D), and a lower concentration of probe
Figure 2. A comparison of sensitivity between Gly-Pro-BCD-Tb (10 nm)
and Gly-Pro-MCA (100 nm). The indicated concentration of diprotin A
was added to each sample. The concentration of DPP4 was 1 nm. TRF
measurements were performed with a delay time of 0.05 ms and a gate-
time of 0.95 ms. Excitation wavelengths were l=330 and 380 nm for Gly-
Pro-BCD-Tb and Gly-Pro-MCA, respectively, and the luminescence in-
tensity was measured at l=546 and 440 nm for Gly-Pro-BCD-Tb and
Gly-Pro-MCA, respectively. The measurements were performed with a
microplate reader. * indicates P <0.05, *** indicates P <0.001, NS=not
significant, N.I.=no inhibitor, N.E.=no enzyme. The data are the
meanꢁSD (n=6).
gives
a smaller IC50 value (Figure 3A,B). In inhibitor
screening, a one-point measurement (i.e., the concentration
of all the samples is fixed, e.g., at 10 mm) is often used, and
it is important that the IC50 value should be small. For exam-
ple, if the IC50 value is one order larger, it may not be possi-
ble to estimate the inhibitory activity of a sample accurately
(Figure 3C). For screening, it is important to identify sam-
ples even with weak inhibitory activity. Thus, Gly-Pro-BCD-
Tb is superior to Gly-Pro-MCA.
Next, the inhibitory activity of sitagliptin was measured
with Gly-Pro-BCD-Tb and Gly-Pro-MCA in the presence of
fluorescent contaminants (Figure 4). Fluorescent contami-
nants were selected from screening samples as inactive com-
pounds that showed strong absorption and fluorescence
sensor probes. TRF measurements with Gly-Pro-BCD-Tb
detected inhibition by 1 mm diprotin A, whereas measure-
ments with Gly-Pro-MCA required a concentration of 3 mm
diprotin A before inhibitory activity could be observed.
13482
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13479 – 13486

Development of a Lanthanide-Based Luminescence Probe
FULL PAPER
Interestingly, bestatin itself did
not show any inhibitory activity
(Figure 5C). To our knowledge,
there are no reported examples
of epibestatin that have shown
a
stronger inhibitory activity
than bestatin.[15,16] For example,
bestatin is comparatively
a
strong inhibitor of general ami-
nopeptidases, such as leucine
aminopeptidase (LAP; IC50 =
0.032 mm), whereas epibestatin
is
a weak inhibitor (IC50 =
24 mm).[16] However, DPP4 is a
serine aminopeptidase that rec-
ognises substrates that contain
a proline residue, and we specu-
late that the unique structural
characteristics of the proline
residue account for this appar-
ent anomaly. Among samples
with similar chemical structures
to epibestatin, there were none
that exhibited a higher activity
than epibestatin (data not
shown). Furthermore, when we
compared the assay results that
used Gly-Pro-BCD-Tb and
Figure 3. Inhibition curve of sitagliptin (A) and diprotin A (B). Diamond: 1 nm Gly-Pro-BCD-Tb, square:
10 nm Gly-Pro-BCD-Tb, circle: 100 nm Gly-Pro-MCA. Each data point represents the meanꢁSD (n=6).
C) Difference in inhibition (%) between Gly-Pro-BCD-Tb and Gly-Pro-MCA at a one-point measurement.
D) Equation for the IC50 value.
overlapping with those of Gly-Pro-BCD-Tb or Gly-Pro-
MCA. (The absorption and fluorescence spectra and the
structure of each sample are shown in Figure S3 in the Sup-
porting Information.) When the inhibitory activity of sita-
gliptin was measured in the presence of a contaminant, the
inhibitory activity was successfully detected with Gly-Pro-
BCD-Tb. In contrast, Gly-Pro-MCA was ineffective because
of the very strong background fluorescence. In the case of
Gly-Pro-BCD-Tb, the measured IC50 value was almost the
same in the presence of samples 1–3 as fluorescent contami-
nants, that is, the inhibitory activity could be measured accu-
rately with Gly-Pro-BCD-Tb, even in the presence of fluo-
rescent contaminants. This result indicates that we can pre-
cisely evaluate the inhibitory activities of screening samples
with Gly-Pro-BCD-Tb, even if the samples are strongly fluo-
rescent.
Gly-Pro-MCA, epibestatin was not detected with Gly-Pro-
MCA, whereas another sample (square point in Figure 5D)
was detected as a strong inhibitor. However, further exami-
nation showed that this strong inhibitor was a false-positive
result. (The structure and photochemical properties of this
inhibitor, a 4-amide-1,8-naphthalimide derivative, are shown
in Figure S4 in the Supporting Information.) Naphthalimides
are known to be strongly fluorescent,[17] and indeed the com-
pound showed excitation and emission absorptions at l=
350 and 485 nm, respectively, in 2-amino-2-(hydroxymethyl)-
1,3-propanediol hydrochloride (Tris–HCl) buffer (pH 8.0).
Therefore, it is considered that accurate measurements
could not be carried out with Gly-Pro-MCA (Figure 5D)
due to overlap of the excitation and emission wavelengths.
False-positive samples were seen in other sample plates
when Gly-Pro-MCA was used (not shown) but not in TRF
measurements with Gly-Pro-BCD-Tb. We also examined
the dose dependency of epibestatin with Gly-Pro-MCA (see
Figure S5 in the Supporting Information). Marginal inhibito-
ry activity of epibestatin (10 mm) towards DPP4 was detect-
ed. Only weak inhibitory activity was seen with Gly-Pro-
MCA relative to the activity with Gly-Pro-BCD-Tb (Fig-
ure 5C). It is considered that this outcome is due to the dif-
ference in sensitivity between Gly-Pro-MCA and Gly-Pro-
BCD-Tb (Figures 2 and 3). These results indicate that sensi-
tive and reliable screening can be carried out with Gly-Pro-
BCD-Tb.
DPP4 inhibitor screening: From above results, we consid-
ered that Gly-Pro-BCD-Tb is superior to Gly-Pro-MCA for
high-sensitivity screening of DPP4 inhibitors. We performed
screening of DPP4 inhibitors among 3841 randomly selected
samples (Figure 5A). Among the 3841 samples (each micro-
plate contains 80 samples), there were seven that showed
more than 50% inhibition at 10 mm and several samples also
exhibited about 40% inhibition at 10 mm. We obtained titra-
tion curves of the above seven compounds for the inhibition
of DPP4. Only one sample showed dose-dependent inhibito-
ry activity (Figure 5B). This sample was epibestatin (IC50 =
17 mm), which is an epimer of the anticancer drug bestatin.
Chem. Eur. J. 2010, 16, 13479 – 13486
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13483

T. Nagano et al.
DTPA and the Tb3+ ion is very strong (logKM =22.71 and
22.39 for Tb and Eu, respectively),[18] and the concentration
of Gly-Pro-BCD-Tb in this screening is sufficiently low.
Therefore, we think that this on/off type methodology with
TRF measurements is suitable for screening. The present
methodology is also expected to be applicable for screening
inhibitors of other proteases, if the substrate sequence of
Gly-Pro-BCD-Tb is appropriately modified.
Experimental Section
General information: HBTU, HOBT, HATU, and glycylproline were pur-
chased from Watanabe Chemical Industries, Ltd (Japan). DPP4 was pur-
chased from R&D Systems (USA). All the other reagents and solvents
were purchased from Tokyo Kasei Co., Ltd (Japan) or Wako Pure Chem-
ical Industries, Ltd (Japan) and were of the highest grade available and
used as received. The 1H NMR spectra were recorded on JEOL JNM-
LA300 and JEOL JNM-LA400 instruments; d values are given in ppm
from the solvent resonance as the internal standard. Data are reported as
follows: chemical shift, multiplicity (s=singlet, d=doublet, t=triplet,
br=broad, and m=multiplet), integration, J coupling, and assignment.
Mass spectra were recorded on a JEOL JMS-T100 LC mass spectrome-
ter. Column chromatography was performed on silica gel 60 (Kanto
Chemical Co.; Tokyo, Japan) and Chromatorex-NH silica gel (Fuji Silysia
Chemical; Kasugai, Japan). Purification by preparative HPLC was per-
formed on a reverse-phase ODS column (GL Sciences, Japan; Inertsil
ODS-3 10ꢁ250 mm) with 0.1% TFA in water as eluent A and 0.1%
TFA in acetonitrile/water (80:20) as eluent B or 100 mm TEAA in water
as eluent A and acetonitrile/water (80:20) as eluent B on a JASCO PU-
2080 plus system at a flow rate of 5.0 mLminꢀ1. HPLC analyses were per-
formed on a reversed-phase ODS column (GL Sciences; Inertsil ODS-3
4.6ꢁ250 mm) fitted onto a JASCO PU-980 plus system at a flow rate of
1.0 mLminꢀ1. UV/Vis spectra were obtained on a Shimazu UV-1600 spec-
trometer (Tokyo, Japan). Fluorescence spectra were obtained on a Hita-
chi F4500 (Tokyo, Japan). Time-resolved fluorescence assay on 96-well
plates were performed using Perkin–Elmer EnVision Multilabel Plate
Readers (Beaconsfield, Buckinghamshire, England).
Figure 4. Inhibition curves of sitagliptin with Gly-Pro-BCD-Tb and Gly-
Pro-MCA, containing 10 mm solution of samples 1 and 3 and 1 mm solu-
tion of sample 2 as contaminants (the fluorescence intensity of 10 mm
sample 2 was too strong to be measured with the microplate reader).
Each data point represents the meanꢁSD (n=6).
Determination of kinetic parameters: Various concentrations of probes
(Gly-Pro-BCD-Tb and Gly-Pro-MCA) were dissolved in 2-[4-(2-hydrox-
yethyl)piperazin-1-yl]ethanesulfonic acid (HEPES) buffer (50 mm,
pH 8.0). DPP4 was added to the solution, and the fluorescence intensity
was recorded continuously. The initial reaction velocity was calculated
and plotted against probe concentration. The points were fitted to a Mi-
chaelis–Menten curve, and the Km and kcat values were determined by the
least-squares method.
Conclusion
We have developed the novel DPP4 probe Gly-Pro-BCD-
Tb. This probe showed almost the same reactivity with
DPP4 as Gly-Pro-MCA, but is superior in two respects: 1) it
offers greater sensitivity, so that even weak inhibitors can be
detected and 2) it is suitable for TRF measurements, which
can be accurately carried out, even in the presence of strong
background fluorescence and scatted light. The false-posi-
tive rate with Gly-Pro-BCD-Tb was much smaller than with
Gly-Pro-MCA, as expected. We carried out high-throughput
TRF-based DPP4 inhibitor screening of about 4000 samples
with Gly-Pro-BCD-Tb, and epibestatin has been newly iden-
tified as a DPP4 inhibitor. Interestingly, bestatin itself was
not an inhibitor. This difference in inhibitory activity against
DPP4 between bestatin and epibestatin may lead to the dis-
covery of novel scaffolds for DPP4 inhibitors. The present
methodology is based on the fluorescence decrease that re-
sulted from DPP4 activity. It is necessary to consider the
possibility of false-negative results in the inhibitor screening
that result from competitive chelation of the metal ion or
collisional quenching. However, the chelation between
HPLC analysis: Reversed-phase HPLC analysis was performed using an
Inertsil ODS-3 column (4.6ꢁ250 mm; GL Science, Japan) fitted onto a
JASCO PU980 HPLC system to confirm that the enzymatic reaction had
taken place. The conditions and recorded absorption wavelength are
given in the figure legends.
DPP4 inhibition assay on 96-well plates: Experiments were performed
on Costar 96-well half-area black plates. Gly-Pro-BCD-Tb (40 nm, 10 mL)
was added to Tris–HCl buffer (50 mm, 10 mL; pH 8.0). To each well, the
inhibitor (40 mm, 10 mL; 2% DMSO) was added at the same time and
DPP4 (3.5 nm, 10 mL) was added. The plate was incubated at room tem-
perature for 90 min. The fluorescence intensity (lex =330 nm and lem
=
545 nm) was measured with a delay time of 0.05 ms and a gate-time of
0.95 ms. An assay with Gly-Pro-MCA was performed similarly, except for
the measured wavelength (lex =355 nm and lem =460 nm) and without a
delay time.
Screening of DPP4 inhibitors: High-throughput screening was performed
against a part of the chemical library (3841 compounds) of the Chemical
Biology Research Initiative (The University of Tokyo). All the samples
(buffer, Gly-Pro-BCD-Tb, DPP4, and candidate inhibitors (samples))
were dispensed through a dispenser. The fluorescence intensities at 0 and
13484
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13479 – 13486

Development of a Lanthanide-Based Luminescence Probe
FULL PAPER
The reaction mixture was stirred at
room temperature overnight under
argon and evaporated. The residue
was dissolved in AcOEt and the solu-
tion was washed with sodium phos-
phate buffer (pH 4.4), sodium phos-
phate buffer (pH 9.2), and brine; dried
over Na2SO4; and evaporated. The res-
idue was purified by column chroma-
tography on silica gel (eluent: CH2Cl2/
MeOH 19:1!9:1) to obtain 4 as a col-
ourless powder (51 mg, 0.068 mmol,
43%).
1H NMR
(300 MHz,
[D6]DMSO): d=1.41 (s, 9H, CH3),
1.74–2.15 (m, 4H, CH2-CH2), 3.40–
3.60 (m, 2H, CH2-CH2), 3.63–3.87 (m,
2H, CH2), 4.05 (s, 2H, CH2), 4.31 (t,
ꢀ
1H, J=6.6 Hz, Ar CH), 4.37–4.43 (m,
1H, CH), 4.48 (d, 2H, J=6.6 Hz,
CH2), 6.10 (s, 1H, ArH), 6.74–6.85 (m,
1H, CONH), 7.11 (d, 1H, J=8.1 Hz,
ArH), 7.20 (d, 2H, J=8.4 Hz, ArH),
7.30–7.46 (m, 4H, ArH), 7.51 (d, 2H,
J=8.4 Hz, ArH), 7.58–7.67 (m, 2H,
ArH), 7.75 (d, 2H, J=7.3 Hz, ArH),
7.90 (d, 2H, J=8.1 Hz, ArH), 9.92 (s,
1H, CONH), 10.02 (s, 1H, CONH),
11.60 ppm (s, 1H, CONH); 13C NMR
(100 MHz, [D6]DMSO): d=24.4, 28.2,
29.2, 36.7, 36.8, 42.4, 46.5, 60.2, 65.8,
77.9, 103.5, 112.8, 114.1, 119.1, 119.3,
120.2, 125.1, 125.5, 127.1, 127.7, 129.0,
129.1, 133.0, 137.4, 140.0, 140.8, 140.8,
143.7, 150.3, 153.2, 155.8, 162.0,
167.5 ppm; HRMS (ESI+): m/z: calcd
for C43H43N5NaO7: 764.3079 [M+Na+
]; found: 764.3060.
Figure 5. A) Results of the screening assay for DPP4 inhibitors. A total of 3841 samples were assayed. The
large triangle and arrow indicate epibestatin and the dashed box indicates a microplate containing epibestatin.
Each microplate contained 80 compounds. B) Inhibition curve of epibestatin. The calculated IC50 value was
17 mm. C) A comparison of DPP4 inhibitory activity of epibestatin and bestatin. The chemical structures are
also shown. The data are meansꢁSD (n=6). D) Assay result of a microplate (80 samples, dashed box shown
in (A)); left: Gly-Pro-BCD-Tb, right: Gly-Pro-MCA. Triangular and square points indicate epibestatin and a
false-positive compound, respectively. SD=12.9 for left, SD=21.4 for right.
Synthesis of 5: Piperidine (200 mL)
was added to a solution of 4 (23 mg,
0.031 mmol) in dehydrated DMF
(4 mL). The reaction mixture was
90 min were measured with a microplate reader after the addition of the
enzyme. The microplates were incubated at room temperature during the
measurement. The conditions were the same as described for the DPP4
inhibition assay.
stirred at room temperature for 1 h and evaporated. The residue was pu-
rified by column chromatography on NH silica gel (eluent: CH2Cl2/
MeOH 19:1!9:1) to obtain
5 as a colourless powder (15 mg,
0.029 mmol, 93%). 1H NMR (300 MHz, [D6]DMSO): d=1.43 (s, 9H,
ꢀ
ꢀ
CH3), 1.80–2.20 (m, 4H, CH2 CH2), 3.46–3.63 (m, 2H, CH2 CH2), 3.70–
3.96 (m, 2H, CH2), 4.03 (s, 2H, CH2), 4.44–4.52 (m, 1H, CH), 5.82 (s,
2H, NH2), 5.90 (s, 1H, ArH), 6.40–6.48 (m, 2H, ArH), 6.82–6.93 (m, 1H,
CONH), 7.24 (d, 2H, J=8.8 Hz, ArH), 7.44 (d, 1H, J=8.8 Hz, ArH),
7.57 (d, 2H, J=8.8 Hz, ArH), 9.99 (s, 1H, CONH), 11.27 ppm (s, 1H,
CONH); 13C NMR (100 MHz, [D6]DMSO): d=24.4, 28.2, 29.3, 36.9, 42.4,
45.8, 60.2, 77.9, 96.8, 109.4, 110.4, 115.0, 119.3, 125.7, 129.0, 129.1, 133.5,
137.3, 141.2, 150.6, 150.9, 155.8, 162.4, 167.5 ppm; HRMS (ESI+): m/z:
calcd for C28H34N5O5: 520.2560 [M+H+]; found: 520.2551.
Estimation of inhibition (%): All the values of inhibition were calculated
in accordance with Equation (1), in which L0 min =initial luminescence in-
tensity, L90 min =luminescence intensity at 90 min.
Inhibition ð%Þ ¼ ½ðL90 min=L0 minÞsampleꢀðL90 min=L0 minÞno inhibitorꢂ=
½ðL90 min=L0 minÞno enzymeꢀðL90 min=L0 minÞno inhibitorꢂ ꢃ 100
ð1Þ
Synthesis of 2: Boc2O (253 mg, 1.16 mmol) was added to a solution of
glycylproline (1; 100 mg, 0.58 mmol) and NEt3 (800 mL, 5.8 mmol) in dis-
tilled EtOH (10 mL). The reaction mixture was stirred at 60 8C for 1 h
and at room temperature for 3 h. The reaction mixture was evaporated,
and the residue was dissolved in CH2Cl2, washed with 0.1m HCl and
brine, dried over Na2SO4, and evaporated. The residue was purified by
column chromatography on silica gel to obtain 2 as a colourless solid
(157 mg, 0.58 mmol, quant. yield). 1H NMR (300 MHz, CDCl3): d=1.44
Synthesis of 6: DTPA (20 mg, 0.056 mmol) was added to a solution of 5
(15 mg, 0.029 mmol) and DIEA (15 mL, 0.084 mmol) in dehydrated DMF
(5 mL). The reaction mixture was stirred at 0 8C under argon and
warmed to room temperature overnight. H2O (2 mL) was added to
quench the reaction, and the reaction mixture was evaporated. The resi-
due was purified by reverse-phase HPLC to afford 6 as a colourless solid
(13 mg, 0.015 mmol, 50%). 1H NMR (300 MHz, [D6]DMSO): d=1.39 (s,
9H, CH3), 1.80–2.18 (m, 4H, CH2-CH2), 2.82–3.10 (m, 8H, CH2), 3.31–
3.63 (m, 12H, CH2), 3.66–3.92 (m, 2H, CH2), 4.10 (s, 2H, CH2), 4.40–4.48
(m, 1H, CH), 6.13 (s, 1H, ArH), 6.79–6.90 (m, 1H, CONH), 7.24 (d, 2H,
J=8.4 Hz, ArH), 7.39 (d, 1H, J=8.8 Hz, ArH), 7.54 (d, 2H, J=8.4 Hz,
ArH), 7.67 (d, 1H, J=8.8 Hz, ArH), 8.02 (s, 1H, ArH), 9.98 (s, 1H,
CONH), 10.60 (br, 1H, CONH), 11.71 ppm (s, 1H, CONH); 13C NMR
(100 MHz, [D6]DMSO): d=24.4, 28.2, 29.3, 36.7, 42.4, 45.8, 50.2, 50.9,
50.9, 51.9, 55.0, 58.2, 59.3, 60.3, 77.9, 78.0, 104.8, 113.9, 114.7, 119.4, 119.7,
ꢀ
ꢀ
(s, 9H, CH3), 1.87–2.35 (m, 4H, CH2 CH2), 3.40–3.70 (m, 2H, CH2
CH2), 3.81–4.11 (m, 2H, CH2), 4.54–4.62 (m, 1H, CH), 9.48 ppm (brs,
1H, CONH); 13C NMR (75 MHz, CDCl3): d=24.5, 28.3, 42.9, 46.2, 59.2,
79.9, 80.5, 156.0, 168.7, 174.0 ppm; HRMS (ESIꢀ): m/z: calcd for
C12H19N2O5: 271.1294 [MꢀHꢀ]; found: 271.1298.
Synthesis of 4: Compound 2 (87 mg, 0.32 mmol) was added to a solution
of 3 (78 mg, 0.16 mmol), HBTU (150 mg, 0.4 mmol), HOBT (61 mg,
0.4 mmol), and DIEA (120 mL, 0.64 mmol) in dehydrated DMF (5 mL).
Chem. Eur. J. 2010, 16, 13479 – 13486
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13485

T. Nagano et al.
125.3, 129.1, 133.0, 137.4, 139.8, 140.4, 150.4, 155.8, 162.0, 167.6, 169.6,
169.9, 170.4, 172.6, 172.9 ppm; HRMS (ESIꢀ): m/z: calcd for
C42H53N8O14: 893.3681 [MꢀHꢀ]; found: 893.3642.
Synthesis of 7: TFA (2 mL) was added dropwise at 08C to a solution of 6
(13 mg, 0.015 mmol) in CH2Cl2 (3 mL). The reaction mixture was stirred
at 0 8C for 2 h and evaporated. The residue was purified by reverse-
[2] R. Perfetti, P. S. Barnett, R. Mathur, J. M. Egan, Diabetes/Metab.
Rev. 1998, 14, 207–225.
[3] D. J. Drucker, Diabetes Care 2007, 30, 1335–1343.
[4] M. Ashiya, R. E. T. Smith, Nat. Rev. Drug Discovery 2007, 6, 777–
778.
[5] D. Kim, L. Wang, M. Beconi, G. J. Eiermann, M. H. Fisher, H. He,
G. J. Hickey, J. E. Kowalchick, B. Leiting, K. Lyons, F. Marsilio,
M. E. McCann, R. A. Patel, A. Petrov, G. Scapin, S. B. Patel, R. S.
Roy, J. K. Wu, M. J. Wyvratt, B. B. Zhang, L. Zhu, N. A. Thornberry,
A. E. Weber, J. Med. Chem. 2005, 48, 141–151.
[6] E. B. Villhauer, J. A. Brinkman, G. B. Naderi, B. F. Burkey, B. E.
Dunning, K. Prasad, B. L. Mangold, M. E. Russell, T. E. Hughes, J.
Med. Chem. 2003, 46, 2774–2789.
[7] D. J. Augeri, J. A. Robl, D. A. Betebenner, D. R. Magnin, A.
Khanna, J. G. Robertson, A. Wang, L. M. Simpkins, P. Taunk, Q.
Haung, S.-P. Han, B. Abboa-Offei, M. Cap, L. Xin, L. Tao, E. Tozzo,
G. E. Welzel, D. M. Egan, J. Marcinkeviciene, S. Y. Chang, S. A.
Biller, M. S. Kirby, R. A. Parker, L. G. Hamann, J. Med. Chem.
2005, 48, 5025–5037.
[8] a) R. A. Pederson, H. A. White, D. Schlenzig, R. P. Pauly, C. H. Mc-
Intosh, H. U. Demuth, Diabetes 1998, 47, 1253; b) E. B. Villhauer,
J. A. Brinkman, G. B. Naderi, B. E. Dunning, B. L. Mangold, M. D.
Mone, M. E. Russell, S. C. Weldon, T. E. Hughes, J. Med. Chem.
2002, 45, 2362–2365; c) H. Sakashita, F. Akahoshi, H. Kitajima, R.
Tsutsumiuchi, Y. Hayashi, Bioorg. Med. Chem. 2006, 14, 3662–3671;
d) M. Eckhardt, E. Langkopf, M. Mark, M. Tadayyon, L. Thomas,
H. Nar, W. Pfrengle, B. Guth, R. Lotz, P. Sieger, H. Fuchs, F. Him-
melsbach, J. Med. Chem. 2007, 50, 6450–6453; e) K. Tanaka-Amino,
K. Matsumoto, Y. Hatakeyama, I. Shima, S. Takakura, S. Muto, Eur.
J. Pharmacol. 2008, 590, 444–449; f) Z. Pei, X. Li, K. Longenecker,
T. W. von Geldern, P. E. Wiedeman, T. H. Lubben, B. A. Zinker, K.
Stewart, S. J. Ballaron, M. A. Stashko, A. K. Mika, D. W. A. Beno,
M. Long, H. Wells, A. J. Kempf-Grote, D. J. Madar, T. S. McDer-
mott, L. Bhagavatula, M. G. Fickes, D. Pireh, L. R. Solomon, M. R.
Lake, R. Edalji, E. H. Fry, H. L. Sham, J. M. Trevillyan, J. Med.
Chem. 2006, 49, 3520–3535.
[9] Y. B. Kim, L. M. Kopcho, M. S. Kirby, L. G. Hamann, C. A. Weigelt,
W. J. Metzler, J. Marcinkeviciene, Arch. Biochem. Biophys. 2006,
445, 9–18.
[10] a) A. K. Saha, K. Kross, E. D. Kloszewski, D. A. Upson, J. L. Toner,
R. A. Snow, C. D. V. Black, V. C. Desai, J. Am. Chem. Soc. 1993,
115, 11032–11033; b) W. Lei, A. Drꢂkop, Z. Lin, M. Wu, O. S. Wolf-
beis, Microchim. Acta 2003, 143, 269.
phase HPLC to obtain 7 as a colourless solid (10 mg, 0.012 mmol, 83%).
1
ꢀ
H NMR (300 MHz, [D6]DMSO): d=1.75–2.16 (m, 4H, CH2 CH2), 2.74–
3.60 (m, 20H, CH2), 3.74 (s, 2H, CH2), 4.05 (s, 2H, CH2), 4.39–4.45 (m,
1H, CH), 6.19 (s, 1H, ArH), 7.19 (d, 2H, J=8.0 Hz, ArH), 7.33 (d, 1H,
J=8.8 Hz, ArH), 7.44–7.61 (m, 3H, ArH), 8.00 (s, 1H, ArH), 10.09 (s,
1H, CONH), 10.80 (br, 1H, CONH), 11.67 ppm (br, 1H, CONH);
13C NMR (100 MHz, [D6]DMSO): d=22.1, 24.2, 29.4, 37.0, 45.8, 49.8,
51.3, 51.5, 51.9, 54.7, 57.6, 58.4, 59.9, 60.3, 104.8, 114.1, 114.4, 119.4, 119.7,
125.3, 128.9, 133.4, 133.8, 137.2, 139.9, 140.5, 150.0, 162.0, 169.4, 170.0,
170.1, 173.2, 174.1 ppm; HRMS (ESIꢀ): m/z: calcd for C37H45N8O12
793.3157 [M-Hꢀ]; found: 793.3123.
:
Synthesis of Gly-Pro-BCD-Ln complexes 8 and 9: TbCl3·6H2O (2.4 mg,
0.007 mmol) or EuCl3·6H2O (2.4 mg, 0.007 mmol) were added to a solu-
tion of 7 (5 mg, 0.006 mmol) in MeOH (3 mL). The reaction mixture was
stirred at room temperature for 1 h and evaporated. The residue was dis-
solved in H2O and the solution was loaded on a C18 Sep-Pak cartridge
(eluent: H2O!MeOH) to remove excess metal. The eluent was collected
and evaporated to afford 8 (5.7 mg, 0.006 mmol, quant. yield) and 9
(5.7 mg, 0.006 mmol, quant. yield) as colourless solids. 8: HRMS (ESI+):
m/z: calcd for C37H43N8NaO12Tb: 973.2152 [M+Na+]; found: 973.2157. 9:
HRMS (ESI+): m/z: calcd for C37H43N8NaO12Eu: 945.2291 [M+Na+];
found: 945.2243.
Synthesis of Gly-Pro-MCA 12: Compound 2 (300 mg, 1.1 mmol) was
added to
a solution of 10 (100 mg, 0.57 mmol), HATU (324 mg,
1.14 mmol), and DIEA (300 mL, 1.7 mmol) in dehydrated DMF (10 mL).
The reaction mixture was stirred at room temperature over night under
argon and evaporated. The residue was dissolved in CH2Cl2, washed with
0.1m HCl, dried over Na2SO4, and evaporated. The residue was partly
purified by column chromatography on silica gel (eluent: CH2Cl2/MeOH
19:1!9:1) to afford 11 as a crude product (35 mg). TFA (5 mL) was
added at 08C to a solution of 11 (35 mg) in CH2Cl2 (5 mL). The reaction
mixture was stirred at 08C for 2 h and evaporated. The residue was puri-
fied by column chromatography on NH silica gel (eluent: CH2Cl2/MeOH
19:1!93:7) to obtain 12 as a colourless solid (5 mg, 0.046 mmol, 8%).
1
ꢀ
H NMR (300 MHz, [D6]DMSO): d=1.75–2.21 (m, 4H, CH2 CH2), 2.39
ꢀ
(s, 3H, Ar CH3), 3.15–3.72 (m, 6H, NH2, CH2), 4.38–4.54 (m, 1H, CH),
6.26 (s, 1H, ArH), 7.47 (dd, 1H, J=1.5, 8.8 Hz, ArH), 7.71 (d, 1H, J=
8.8 Hz, ArH), 7.75 (d, 1H, J=1.5 Hz, ArH), 10.47 ppm (s, 1H, CONH);
13C NMR (100 MHz, [D6]DMSO): d=18.0, 24.4, 29.3, 43.5, 45.6, 60.3,
105.6, 112.3, 115.0, 115.2, 125.9, 142.4, 153.2, 153.7, 160.1, 171.4,
171.5 ppm; HRMS (ESI+): m/z: calcd for C17H19N3NaO4: 352.1273
[M+Na+]; found: 352.1237.
[11] T. Terai, K. Kikuchi, S. Iwasawa, T. Kawabe, Y. Hirata, Y. Urano, T.
Nagano, J. Am. Chem. Soc. 2006, 128, 6938–6946.
[12] J. Karvinen, V. Laitala, M. L. Mꢃkinen, O. Mulari, J. Tamminen, J.
Hermonen, P. Hurskainen, I. Hemmilꢃ, Anal. Chem. 2004, 76, 1429–
1436.
[13] S. Mizukami, K. Tonai, M. Kaneko, K. Kikuchi, J. Am. Chem. Soc.
2008, 130, 14376–14377.
[14] J. Rahfeld, M. Schierborn, B. Hartrodt, K. Neubert, J. Heins, Bio-
chim. Biophys. Acta 1991, 1076, 314–316.
[15] H. Suda, T. Aoyagi, T. Takeuchi, H. Umezawa, Arch. Biochem. Bio-
phys. 1976, 177, 196–200.
Acknowledgements
This study was supported in part by the Ministry of Education, Culture,
Sports, Science and Technology of Japan (Grant Nos. 20689001 and
21659024 to K.H. and 21750135 to T.T.) and by a grant from the Industri-
al Technology Development Organization (NEDO) of Japan (to T.T.).
K.H. was also supported by Sankyo Foundation of Life Science and M.K.
was supported by a Grant-in-Aid for JSPS Fellows. This study was sup-
ported in part by funds from the Target Protein Research Programs from
the Ministry of Education, Culture, Sports, Science and Technology of
Japan.
[16] R. Nishizawa, T. Saino, T. Takita, H. Suda, T. Aoyagi, H. Umezawa,
J. Med. Chem. 1977, 20, 510–515.
[17] D. Srikun, E. W. Miller, D. W. Domaille, C. J. Chang, J. Am. Chem.
Soc. 2008, 130, 4596–4597.
[18] T. Moeller, L. C. Thompson, J. Inorg. Nucl. Chem. 1962, 24, 499–
510.
Received: April 23, 2010
Revised: July 28, 2010
Published online: October 11, 2010
[1] a) S. Wild, G. Roglic, A. Green, R. Sicree, H. King, Diabetes Care
2004, 27, 1047–1053; b) American Diabetes Association (ADA),
National Diabetes Fact Sheets, 2005.
13486
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13479 – 13486