A turn-On Fluorescent Sensor for Methylglyoxal

Article abstract of DOI:10.1021/ja406077j

Methylglyoxal (MGO), a dicarbonyl metabolite produced by all living cells, has been associated with a number of human diseases. However, studies of the role(s) MGO plays biologically have been handicapped by a lack of direct methods for its monitoring and detection. To address this limitation, we have developed a fluorescent sensor (methyl diaminobenzene-BODIPY, or MBo ) that can detect MGO under physiological conditions. We show that MBo is selective for MGO over other biologically relevant dicarbonyls and is suitable for detecting MGO in complex environments, including that of living cells. In addition, we demonstrate MBo's utility in estimating plasma concentrations of MGO. The results reported herein have the potential to advance both clinical and basic science research and practice.

Full text of DOI:10.1021/ja406077j

Article
pubs.acs.org/JACS
A “Turn-On” Fluorescent Sensor for Methylglyoxal
Tina Wang,† Eugene F. Douglass, Jr.,† Kelly J. Fitzgerald,‡ and David A. Spiegel*,†,‡
†Department of Chemistry, Yale University, 225 Prospect Street, New Haven, Connecticut 06511, United States
‡Department of Pharmacology, Yale University School of Medicine, 333 Cedar Street, New Haven, Connecticut 06510, United States
S
* Supporting Information
ABSTRACT: Methylglyoxal (MGO), a dicarbonyl metabolite
produced by all living cells, has been associated with a number
of human diseases. However, studies of the role(s) MGO plays
biologically have been handicapped by a lack of direct methods
for its monitoring and detection. To address this limitation, we
have developed a fluorescent sensor (methyl diaminobenzene-
BODIPY, or “MBo”) that can detect MGO under physio-
logical conditions. We show that MBo is selective for MGO
over other biologically relevant dicarbonyls and is suitable for
detecting MGO in complex environments, including that of living cells. In addition, we demonstrate MBo’s utility in estimating
plasma concentrations of MGO. The results reported herein have the potential to advance both clinical and basic science
research and practice.
methods require cell lysis followed by extensive processing and
deproteinization under harsh conditions. As a result, estimates
of cellular MGO levels using this method have been extremely
variable, spanning 3 orders of magnitude.26,28
INTRODUCTION
■
Methylglyoxal (MGO) is a reactive dicarbonyl produced by all
living cells during glucose, fatty acid, and amino acid
metabolism.1,2 It is a potent glycating agent, capable of
nonenzymatically modifying both proteins and DNA.2
Reactions of biological macromolecules with MGO can also
lead to the formation of advanced glycation end-products
(AGEs), which have been shown to cause protein dysfunction,3
activate membrane receptors and trigger pro-inflamatory
signaling, and have been linked to aging disorders, diabetic
complications, and chronic inflammation.4,5 Interestingly,
elevated levels of MGO have also been associated with
pathologies such as diabetes,6,7 cardiovascular disease,8 hyper-
algesia,9,10 and kidney disease.11,12 MGO’s connection with
diabetes is particularly notable; levels of MGO have
consistently been reported to increase in the plasma of
diabetics.7 Furthermore, MGO by itself has been shown to
induce a number of deleterious effects in cellular systems,
including oxidative stress,13,14 inflammation,13,15,16 defects in
cell adhesion,17 and endothelial cell dysfunction.18−20 Thus,
MGO has been hypothesized to contribute directly to the
pathophysiology of diabetic complications.
Despite its links to disease, MGO’s roles in cellular processes
and pathogenesis remain poorly understood.21 This limitation
is partly due to a lack of straightforward methods to monitor
this small molecule in complex systems, and no method exists
for visualizing it in live biological samples. Available techniques
for measuring MGO levels include electrochemical and
absorbance-based approaches, but these are either not directly
applicable to living systems (the former),22 or too insensitive to
be of use (the latter).23 Due to the reactivity of MGO, most
methods for monitoring its levels involve derivatization with o-
phenylenediamine (OPD) or similar molecules to form stable
adducts, followed by HPLC or LC−MS analysis.24−,27 These
Here we present the first fluorescent “turn-on” sensor for
MGO detection, called methyl diaminobenzene-BODIPY or
MBo, that can be used in living systems. We provide data
showing that MBo is selective for MGO over other biological
dicarbonyls and also can be used to detect changes in MGO
levels in live cells by fluorescence microscopy. In addition, MBo
can be used to directly estimate MGO concentrations in plasma
or serum. The facility with which MBo can be used to both
measure MGO levels and image its presence in live cells has the
potential to represent an enabling advance both in clinical
diagnosis and basic science research.
RESULTS AND DISCUSSION
■
Existing methods for detection of MGO and other 1,2-
dicarbonyls involve the use of o-phenylenediamine (OPD) (1).
OPD reacts quickly with these metabolites to form the stable,
strongly UV-absorbing 2-methylquinoxaline (2), which is
detectable using HPLC or LC−MS (Figure 1a). We sought a
way to take advantage of this reactivity to develop “turn-on”
sensors for MGO. Along these lines, the Nagano group has
previously reported that DAF-2 (3), an agent based on an OPD
scaffold, could be used to detect nitric oxide (NO) (Scheme S2
in Supporting Information [SI]).29,30 In this sensor, fluo-
rescence is quenched via an intramolecular photoinduced
electron transfer from the OPD functionality to the acceptor
Received: June 19, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja406077j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. (a) o-Phenylenediamine (OPD, 1) traps MGO to form
stable 2-methylquinoxaline adduct 2. (b) OPD, an electron-rich arene,
quenches fluorescence in 3 by a-PeT. (c) Conversion of the pendant
OPD in 3 to a quinoxaline (4) lowers its EHOMO, preventing quenching
by a-PeT.
Figure 2. (a) Fluorescence of 4 is quenched by donor-excited
photoinduced electron transfer (d-PeT) due to the lower ELUMO of the
electron-poor quinoxaline. (b) Addition of electron-donating groups
raises the quinoxaline ELUMO in 6, preventing quenching and allowing
fluorescence to occur.
fluorophore, a process known as acceptor-excited photoinduced
electron transfer (a-PeT, Figure 1b).31,32
Reaction of the pendant OPD in 3 with NO affords a triazole
(S5, Scheme S2, SI). Because of its relatively low HOMO
energy, the triazole in S5 does not serve as an a-PeT donor,
rendering S5 highly fluorescent.31 By analogy, we hypothesized
that reaction with MGO would relieve acceptor-excited
photoinduced electron transfer, and “turn on” fluorescence in
compound 4. As a first step toward testing this hypothesis, we
compared the HOMO energy values between the 2-
methylquinoxaline and the benzotriazole moieties found in 4
and S5, respectively (Table S1, SI). The similarity between
these energy values suggested that reaction of 3 with a
dicarbonyl should relieve autoinhibitory effects on fluorescence
emission. We therefore decided to evaluate 3 as an MGO
sensor.
Treatment of 3 with either MGO or S-nitroso-N-
acetylpenicillamine (SNAP) (an NO donor) led to mixed
results. While 3 responded vigorously to the presence of NO, as
reported previously, the compound exhibited virtually no
response to MGO (Figure S1a, SI). LC−MS analysis confirmed
formation of quinoxaline product (Figure S1b, SI), which led us
to conclude that the quinoxaline adduct of 3 must lack
fluorescence. We speculated that perhaps the LUMO of
quinoxaline was sufficiently low in energy to accept an electron
donated from the excited fluorophore, thus quenching
fluorescence emission through donor-excited photoinduced
electron transfer (d-PeT) (Figure 2a).32 Indeed, calculations
indicated that quinoxaline S6 has a relatively low-lying LUMO
compared to that of triazole S11 (Table S1, SI).
These data suggested that careful tuning of the energetic
matching between quencher and fluorophore would be
necessary for sensor development. Thus, we reasoned that
the o-arene-diamine functionality in an MGO sensor must give
rise to a quinoxaline with a sufficiently high LUMO energy to
avoid a d-PeT process; yet simultaneously, we must also avoid
raising the quinoxaline’s HOMO energy so much that it can
quench fluorescence by an a-PeT mechanism. We further
reasoned that lowering the HOMO/LUMO energies of the
fluorophore component would widen the energy gap between
the fluorophore’s excited state and the LUMO energy of the
quinoxaline acceptor, thus decreasing d-PeT quenching. With
these considerations in mind, we pursued a sensor containing
both an electron-deficient fluorophore and an electron-rich
OPD derivative.
To this end, we chose to adopt a BODIPY scaffold, as it is
easily derivatizable at both the aryl moiety and the
fluorophore.33 Gaussian calculations performed on various 2-
methylquinoxalines led us to select a 6-methyl substituent as
ideal for raising the LUMO energy without forcing
simultaneous increases in EHOMO (S9, Table S1, SI). Thus,
we synthesized methyl diaminobenzene-BODIPY derivative (5,
“MBo”) in four steps and in 13.5% overall yield starting from
N-(4-formyl-5-methyl-2-nitrophenyl)acetamide,34 as detailed in
the SI (Scheme S1).
Although a modified version of this compound (diamino-
benzene-BODIPY ethyl ester, DAMBO-CO2Et, 7), has
previously been employed as a sensor for NO,30 we reasoned
that 5 would not respond to NO under the conditions of our
experiments; as the pKa of DAMBO-triazole (S15) is
approximately 7.7, its deprotonation at physiological pH results
in the formation of an electron-rich triazolate. This electron-
rich system induces fluorescence quenching through an a-PeT
mechanism. (S16, Figure S2, SI).30 Due to the lack of acidic
protons in quinoxaline S9, on the other hand, we reasoned that
the MBo−MGO adduct should not experience similar
quenching, and therefore, MBo should exhibit a significantly
stronger fluorescence response to MGO over NO under
physiological conditions (pH 7.4). For purposes of comparison,
we also synthesized diaminobenzene-BODIPY ethyl ester
(DAMBO-CO2Et, 7) and diaminobenzene-BODIPY
(DAMBO, 8).30 Electronic calculations led us to predict that
5 should have the greatest fluorescence increase upon reaction
with MGO, followed by 7, while 8 should exhibit the smallest
change in signal.
The ability of dyes 5, 7, and 8 to detect MGO in an in vitro
assay was then tested. As anticipated, compound 5 displayed
over a 10-fold increase in fluorescence upon reaction with
MGO, making it the most sensitive MGO detector among the
compounds synthesized. The quantum yield of the MBo−
MGO adduct 6 was found to be 0.326, compared to 0.003 for
MBo (Table S2, SI). Compounds 3 and 8 exhibited no
response to MGO, while 7 showed a modest increase in
fluorescence (Figure 3).
As a next step, we went on to characterize MBo’s dicarbonyl
selectivity and reactivity in more detail. We therefore compared
MBo-induced fluorescence upon reaction with a variety of α-
dicarbonyl species (Figure 4), including MGO, glyoxal (GO),
B
dx.doi.org/10.1021/ja406077j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
plasma MGO concentrations have also been estimated to be as
high as single-digit micromolar.7,38 In addition to its low steady-
state concentration, NO is also likely to have a much shorter
half-life than MGO due its reactivity with cellular antioxidants
such as glutathione.39 Taken together, these observations
suggested that NO would not interfere with MGO detection in
cellular settings.
We also evaluated the sensitivity and kinetic properties for
the MBo−MGO reaction. The detection limit of MBo for
MGO by plate reader after 1 h of incubation in PBS at 37 °C is
50−100 nM (Figure S4a, SI); furthermore, the sensitivity
improves with longer incubation times (Figure S4b, SI). This
implies that reaction under these concentrations is not
complete at 1 h; indeed, the pseudo-first-order rate constant
was estimated to be 4.56 × 10−3 s−1 (Figure S5, SI), giving a
half-life of ∼2.5 h for a concentration of 100 nM MGO and 10
μM MBo. Assuming bimolecular kinetics, this rate is
comparable to what has been reported for other useful
fluorescent probes.40 Taken together, these results suggested
that MBo would be suitable for detecting MGO in biological
milieu.
Figure 3. Response of fluorophores to MGO. Fluorophore (10 μM)
was incubated with 50 μM MGO for 1 h at 37 °C in PBS (pH 7.4)
before fluorescence was read by plate reader. Fold fluorescence
represents fluorescence in the presence of MGO divided by the
fluorescence of the blank control (fluorophore only). Error bars
represent standard deviation (n = 3).
After confirming that MBo could detect MGO under
physiological conditions, we examined its capacity for MGO
imaging in live cells. HeLa cells were loaded for 1 h with 10 μM
MBo before being treated with MGO at three separate
concentrations (0, 5, or 10 μM) for an additional hour.
These concentrations are close to the physiologically relevant
range, which has been estimated to be 1−5 μM.2 Cells loaded
with MBo were also tested for cross-reactivity with nitric oxide
by treatment with NOC-7 for 1 h. Application of 5 or 10 μM
MGO was found to cause a significant fluorescence increase in
MBo-treated cells (Figure 5, Figure S7, SI). This fluorescent
Figure 4. Fluorescence response of MBo (5) upon reaction with
various dicarbonyls and reactive oxygen species. MBo (10 μM) was
incubated with 50 μM substrate for the indicated times at 37 °C in
PBS (pH 7.4) before fluorescence was read by plate reader. MGO:
methylglyoxal; GO: glyoxal; GLU: glucosone; PYR: pyruvate; PPD: 1-
phenyl-1,2-propanedione; PGO: phenylglyoxal; NOC-7: nitric oxide
donor, t1/2 ≈ 1.7 min;35 SNAP: nitric oxide donor, t1/2 ≈ 4.6 h;36
H2O2: hydrogen peroxide; blank: PBS only. Error bars represent
standard deviation (n = 3).
glucosone (GLU), pyruvate (PYR), 1-phenyl-1,2-propanedione
(PPD), and phenylglyoxal (PGO). Following incubations of
various dicarbonyls with MBo for 20−60 min, only MGO led
to significant fluorescence increases versus baseline (Figure 4).
Selectivity for MGO over GO is particularly noteworthy, given
their structural similarity, and we hypothesize that our
observations result from the lower LUMO energy of the GO-
derived quinoxaline (S13, Table S1, SI), which can quench
fluorescence by d-PeT. Other dicarbonyls are less reactive
toward MBo than MGO likely because they are less
electrophilic and/or more sterically encumbered (PPD), or
present mostly as the hemiacetal tautomer (GLU).
Since MBo is based on a scaffold initially designed to detect
NO, we wished to determine our fluorophore’s cross-reactivity
with this species. At 50 μM, MBo shows some fluorescence
upon addition of NOC-7, an NO donor with a half-life of 1.7
min (Figure 4).35 However, the estimated physiological
concentration of NO has been estimated at low to mid-
nanomolar range.37 By comparison, values of intracellular
MGO are at least 2 orders of magnitude larger,2 and human
Figure 5. Fluorescence microscopy detection of MGO in live HeLa
cells. Cells were loaded with 10 μM MBo in HEPES/Krebs/Ringer’s
(HKR) buffer for 1 h, then washed once and treated with HKR buffer
containing (a) 0 μM, (b) 5 μM, or (c) 10 μM MGO for an additional
1 h. (d−f) Corresponding phase contrast images. Scale bars represent
40 μm.
signal appears to be cytosolic, with no distinct compartmen-
talization.41 This is consistent with the observation that MGO
is membrane permeable.42 In contrast, cells pretreated with the
nitric oxide donor NOC-7 exhibited only a minimal increase in
fluorescence in the presence of MBo (Figure S8, SI). Overall,
these experiments indicate that MBo is both sufficiently
selective over NO and sensitive enough to image MGO in
live cells.
MBo’s selectivity for MGO under physiological conditions
suggested that it might also be applicable to the estimation of
MGO levels in biological samples. Therefore, we first evaluated
the relationship between MBo fluorescence and MGO
C
dx.doi.org/10.1021/ja406077j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
concentration in both buffer and serum. The fluorescence
response of MBo to MGO at various concentrations was linear
in both PBS and fetal bovine serum (FBS), suggesting that
MBo is suitable for use in MGO quantification in both buffer
and serum (Figure S9, SI).
compound reported capable of visualizing MGO in living cells.
In addition, MBo can be applied to quantifying MGO levels in
serum or plasma samples as a faster, more time-efficient
alternative to existing HPLC-based methods. In vitro experi-
ments have shown that MBo is sensitive even to low
concentrations of MGO (50 nM) and is selective over other
dicarbonyls. We anticipate that this compound will be a useful
tool for clarifying MGO’s roles in various diseases and cellular
processes.
Next, we developed a method to measure MGO concen-
trations using a standard addition procedure (SI). The average
percent recovery of our method (defined as ([MGOestimated]/
[MGOactual]) × 100) was obtained by spiking known amounts
of MGO into PBS or FBS. These recoveries were 76% in PBS
and 64% in FBS (Figure S10, SI), values that are comparable to
the existing “gold standard” strategy for determining MGO
levels (the OPD derivatization method).24,27 Next, we
evaluated the MGO concentration of pooled mouse serum
using MBo and compared these values to the OPD
derivatization strategy. Analysis of pooled BALB/c mouse
serum using both protocols gave similar values of estimated
MGO concentrations (Figure S12, SI). The interday coefficient
of variance was 28% for the MBo method, and 47% for the
OPD dervatization method (Figure S12, SI).
ASSOCIATED CONTENT
■
S
* Supporting Information
Detailed experimental procedures and compound character-
izations. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
david.spiegel.@yale.edu
We then compared MBo- and OPD-based protocols to
measure MGO concentrations in the plasma of several B6 mice
(Figure S13, SI, and Table 1). The average MGO concentration
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Table 1. MGO Concentrations in Mouse Plasma Estimated
by MBo
We thank Dr. Takuji Shoda for invaluable discussions. We also
thank Dr. Thihan Padukkavidana for his assistance with animal
work. This work was supported by the SENS Foundation.
mouse
[MGO], μM
a
a
a
1 and 3
2
0.99
0.59
REFERENCES
(1) Kalapos, M. P. Toxicol. Lett. 1999, 110, 145.
(2) Thornalley, P. J. Drug Metab. Drug Interact. 2008, 23, 125.
(3) Matafome, P.; Sena, C.; Seica, R. Endocrine 2013, 43, 472.
(4) Singh, R.; Barden, A.; Mori, T.; Beilin, L. Diabetologia 2001, 44,
129.
■
4 and 5
6
0.54
0.51
7 and 8
average
0.68
0.66 0.20
a
Pooled to obtain sufficient material for analysis.
(5) Rabbani, N.; Thornalley, P. J. Amino Acids 2012, 42, 1133.
(6) Beisswenger, P. J.; Howell, S. K.; Touchette, A. D.; Lal, S.;
Szwergold, B. S. Diabetes 1999, 48, 198.
in these mice was found to be 0.66 0.20 μM using MBo and
0.63 0.07 μM using the OPD method (Figure S14a, SI).
(7) Kalapos, M. Diabetes Res. Clin. Pract. 2013, 99, 260.
(8) Kilhovd, B. K.; Juutilainen, A.; Lehto, S.; Ronnemaa, T.; Torjesen,
P. A.; Hanssen, K. F.; Laakso, M. Atherosclerosis 2009, 205, 590.
(9) Bierhaus, A.; Fleming, T.; Stoyanov, S.; Leffler, A.; Babes, A.;
Neacsu, C.; Sauer, S. K.; Eberhardt, M.; Schnolzer, M.; Lasitschka, F.;
Neuhuber, W. L.; Kichko, T. I.; Konrade, I.; Elvert, R.; Mier, W.;
Pirags, V.; Lukic, I. K.; Morcos, M.; Dehmer, T.; Rabbani, N.;
Thornalley, P. J.; Edelstein, D.; Nau, C.; Forbes, J.; Humpert, P. M.;
Schwaninger, M.; Ziegler, D.; Stern, D. M.; Cooper, M. E.; Haberkorn,
U.; Brownlee, M.; Reeh, P. W.; Nawroth, P. P. Nat. Med. 2012, 18,
926.
(10) Eberhardt, M. J.; Filipovic, M. R.; Leffler, A.; de la Roche, J.;
Kistner, K.; Fischer, M. J.; Fleming, T.; Zimmermann, K.; Ivanovic-
Burmazovic, I.; Nawroth, P. P.; Bierhaus, A.; Reeh, P. W.; Sauer, S. K.
J. Biol. Chem. 2012, 287, 28291.
(11) Odani, H.; Shinzato, T.; Matsumoto, Y.; Usami, J.; Maeda, K.
Biochem. Biophys. Res. Commun. 1999, 256, 89.
(12) Nakayama, K.; Nakayama, M.; Iwabuchi, M.; Terawaki, H.; Sato,
T.; Kohno, M.; Ito, S. Am. J. Nephrol. 2008, 28, 871.
(13) Wu, L.; Juurlink, B. H. Hypertension 2002, 39, 809.
(14) Kalapos, M. P. Chem.-Biol. Interact. 2008, 171, 251.
(15) Wang, H.; Meng, Q. H.; Gordon, J. R.; Khandwala, H.; Wu, L.
Clin. Biochem. 2007, 40, 1232.
(16) Welten, A. G.; Schalkwijk, C. G.; ter Wee, P. M.; Meijer, S.; van
den Born, J.; Beelen, R. J. Peritoneal Dial. Int. 2003, 23, 213.
(17) Dobler, D.; Ahmed, N.; Song, L.; Eboigbodin, K. E.; Thornalley,
P. J. Diabetes 2006, 55, 1961.
Taking into account interday variance, these values do not
differ significantly between the two methods (Figure S14b, SI)
and are comparable to MGO levels reported in the literature for
mice of the same genetic background.43 Taken together, the
results of these experiments indicate that MBo is effective for
estimating MGO concentrations in biological samples and
provides results that are comparable to those obtained by OPD
derivatization.
Notably, the MBo method holds some significant advantages
versus OPD derivatization. First, sample processing using MBo
is operationally more simple than using OPD because the
protein precipitation step can be skipped. Also, the MBo
method requires a shorter incubation time (3 h vs 24 h).
Finally, the use of a plate reader in lieu of analytical HPLC
allows for the simultaneous analysis of many samples, thus
increasing the efficiency of the MBo method.
CONCLUSIONS
■
In summary, by relating calculated HOMO and LUMO
energies to experimental observations, we have been able to
rationally design MBo, a BODIPY-based probe for MGO. MBo
takes advantage of the reactivity of OPD toward MGO, while
overcoming the electron-deficient nature of quinoxalines by
tuning the electronics of both the fluorophore and pendant aryl
moiety. To the best of our knowledge, MBo is the first
(18) Chan, W. H.; Wu, H. J. J. Cell. Biochem. 2008, 103, 1144.
D
dx.doi.org/10.1021/ja406077j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(19) Brouwers, O.; Niessen, P. M.; Haenen, G.; Miyata, T.;
Brownlee, M.; Stehouwer, C. D.; De Mey, J. G.; Schalkwijk, C. G.
Diabetologia 2010, 53, 989.
(20) Dhar, A.; Dhar, I.; Desai, K. M.; Wu, L. Br. J. Pharmacol. 2010,
161, 1843.
(21) Nemet, I.; Varga-Defterdarovic, L.; Turk, Z. Mol. Nutr. Food Res.
2006, 50, 1105.
(22) Chatterjee, S.; Chen, A. C. Anal. Chim. Acta 2012, 751, 66.
(23) Wild, R.; Ooi, L.; Srikanth, V.; Munch, G. Anal. Bioanal. Chem.
2012, 403, 2577.
(24) McLellan, A. C.; Phillips, S. A.; Thornalley, P. J. Anal. Biochem.
1992, 206, 17.
(25) Chaplen, F. W.; Fahl, W. E.; Cameron, D. C. Anal. Biochem.
1996, 238, 171.
(26) Dhar, A.; Desai, K.; Liu, J.; Wu, L. J. Chromatogr., B 2009, 877,
1093.
(27) Nemet, I.; Varga-Defterdarovic, L.; Turk, Z. Clin. Biochem. 2004,
37, 875.
(28) Chaplen, F. W.; Fahl, W. E.; Cameron, D. C. Proc. Natl. Acad.
Sci. U.S.A. 1998, 95, 5533.
(29) Kojima, H.; Nakatsubo, N.; Kikuchi, K.; Kawahara, S.; Kirino,
Y.; Nagoshi, H.; Hirata, Y.; Nagano, T. Anal. Chem. 1998, 70, 2446.
(30) Gabe, Y.; Urano, Y.; Kikuchi, K.; Kojima, H.; Nagano, T. J. Am.
Chem. Soc. 2004, 126, 3357.
(31) Tanaka, K.; Miura, T.; Umezawa, N.; Urano, Y.; Kikuchi, K.;
Higuchi, T.; Nagano, T. J. Am. Chem. Soc. 2001, 123, 2530.
(32) Ueno, T.; Urano, Y.; Setsukinai, K.; Takakusa, H.; Kojima, H.;
Kikuchi, K.; Ohkubo, K.; Fukuzumi, S.; Nagano, T. J. Am. Chem. Soc.
2004, 126, 14079.
(33) Boens, N.; Leen, V.; Dehaen, W. Chem. Soc. Rev. 2012, 41, 1130.
(34) Tanaka, A.; Ito, K.; Nishino, S.; Motoyama, Y.; Takasugi, H.
Chem. Pharm. Bull. 1994, 42, 560.
(35) Zhang, P.; Ohara, A.; Mashimo, T.; Sun, J.; Shibuta, S.; Takada,
K.; Kosaka, H.; Terada, M.; Yoshiya, I. Circulation 1996, 94, 2235.
(36) Ignarro, L. J.; Lippton, H.; Edwards, J. C.; Baricos, W. H.;
Hyman, A. L.; Kadowitz, P. J.; Gruetter, C. A. J. Pharmacol. Exp. Ther.
1981, 218, 739.
(37) Hall, C. N.; Garthwaite, J. Nitric Oxide 2009, 21, 92.
(38) We have observed that, at these relative concentrations, the
presence of NO does not impact MGO sensing by MBo (Figure S6).
(39) Sies, H. Eur. J. Biochem. 1993, 215, 213.
(40) Miller, E. W.; Albers, A. E.; Pralle, A.; Isacoff, E. Y.; Chang, C. J.
J. Am. Chem. Soc. 2005, 127, 16652.
(41) We cannot rule out that this staining is due solely to the
distribution of the MBo−MGO adduct. The slight punctate staining
observed may be due to aggregation of small amounts of insoluble
MBo or MBo−MGO adduct.
(42) Thornalley, P. J. Biochem. J. 1988, 254, 751.
(43) Watson, A. M.; Soro-Paavonen, A.; Sheehy, K.; Li, J.; Calkin, A.
C.; Koitka, A.; Rajan, S. N.; Brasacchio, D.; Allen, T. J.; Cooper, M. E.;
Thomas, M. C.; Jandeleit-Dahm, K. J. Diabetologia 2011, 54, 681.
E
dx.doi.org/10.1021/ja406077j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX