A unique approach to development of near-infrared fluorescent sensors for in vivo imaging

Article abstract of DOI:10.1021/ja305802v

Near-infrared (NIR) fluorescent sensors have emerged as promising molecular tools for imaging biomolecules in living systems. However, NIR fluorescent sensors are very challenging to be developed. Herein, we describe the discovery of a new class of NIR fluorescent dyes represented by 1a/1c/1e, which are superior to the traditional 7-hydroxycoumarin and fluorescein with both absorption and emission in the NIR region while retaining an optically tunable hydroxyl group. Quantum chemical calculations with the B3LYP exchange functional employing 6-31G(d) basis sets provide insights into the optical property distinctions between 1a/1c/1e and their alkoxy derivatives. The unique optical properties of the new type of fluorescent dyes can be exploited as a useful strategy for development of NIR fluorescent sensors. Employing this strategy, two different types of NIR fluorescent sensors, NIR-H₂O₂ and NIR-thiol, for H₂O₂ and thiols, respectively, were constructed. These novel sensors respond to H₂O₂ or thiols with a large turn-on NIR fluorescence signal upon excitation in the NIR region. Furthermore, NIR-H₂O₂ and NIR-thiol are capable of imaging endogenously produced H₂O₂ and thiols, respectively, not only in living cells but also in living mice, demonstrating the value of the new NIR fluorescent sensor design strategy. The new type of NIR dyes presented herein may open up new opportunities for the development of NIR fluorescent sensors based on the hydroxyl functionalized reactive sites for biological imaging applications in living animals.

Full text of DOI:10.1021/ja305802v

Article
pubs.acs.org/JACS
A Unique Approach to Development of Near-Infrared Fluorescent
Sensors for in Vivo Imaging
Lin Yuan, Weiying Lin,* Sheng Zhao, Wensha Gao, Bin Chen, Longwei He, and Sasa Zhu
State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University,
Changsha 410082, P. R. China
S
* Supporting Information
ABSTRACT: Near-infrared (NIR) fluorescent sensors have
emerged as promising molecular tools for imaging biomole-
cules in living systems. However, NIR fluorescent sensors are
very challenging to be developed. Herein, we describe the
discovery of a new class of NIR fluorescent dyes represented
by 1a/1c/1e, which are superior to the traditional 7-
hydroxycoumarin and fluorescein with both absorption and
emission in the NIR region while retaining an optically tunable
hydroxyl group. Quantum chemical calculations with the B3LYP exchange functional employing 6-31G(d) basis sets provide
insights into the optical property distinctions between 1a/1c/1e and their alkoxy derivatives. The unique optical properties of the
new type of fluorescent dyes can be exploited as a useful strategy for development of NIR fluorescent sensors. Employing this
strategy, two different types of NIR fluorescent sensors, NIR-H2O2 and NIR-thiol, for H2O2 and thiols, respectively, were
constructed. These novel sensors respond to H2O2 or thiols with a large turn-on NIR fluorescence signal upon excitation in the
NIR region. Furthermore, NIR-H2O2 and NIR-thiol are capable of imaging endogenously produced H2O2 and thiols,
respectively, not only in living cells but also in living mice, demonstrating the value of the new NIR fluorescent sensor design
strategy. The new type of NIR dyes presented herein may open up new opportunities for the development of NIR fluorescent
sensors based on the hydroxyl functionalized reactive sites for biological imaging applications in living animals.
Significantly, the new class of NIR fluorescent dyes is superior
to the traditional 7-hydroxycoumarin and fluorescein with
absorption and emission in the NIR region while retaining an
optically tunable hydroxyl group. In other words, the new type
of fluorescent dyes enjoys the advantages of integrating tunable
optical properties with NIR absorption and emission. This
unique feature can be exploited for development of NIR
fluorescent sensors for living animal imaging applications.
Based on the design strategy, two innovative NIR fluorescent
sensors for H2O2 and thiols were constructed. We demonstrate
the utility of the novel NIR sensors in living animal imaging,
highlighting the value of the new NIR fluorescent sensor design
strategy.
INTRODUCTION
■
Fluorescence imaging is a very powerful technique for
localization and dynamic monitoring of biomolecules in living
systems.1−3 The development of new fluorescent imaging
sensors has facilitated the recent significant advances in cell
biology and medical diagnostic imaging.4−6 Near-infrared
(NIR) light (650−900 nm) is advantageous to be employed
in biological imaging due to minimum photodamage to
biological samples, deep tissue penetration, and minimum
interference from background autofluorescence by biomole-
cules in the living systems.7−9 However, small-molecule NIR
fluorescent sensors are very challenging to be constructed.
Thus, innovative strategies for development of NIR fluorescent
sensors are actively sought after.
The traditional 7-hydroxycoumarins and fluorescein dyes
contain a characteristic hydroxyl group (Figure 1), which can
be modified to regulate their optical properties. This feature of
the optically tunable hydroxyl group renders 7-hydroxycoumar-
ins and fluorescein extensively employed in biochemical assays,
optical sensing, and molecular optical imaging.10 However, the
absorption and emission wavelengths of the classic 7-
hydroxycoumarin and fluorescein dyes only locate in the UV
to visible region. Thus, they are not amenable for living animal
imaging, in which the absorption and emission in the NIR
region are highly desirable.
RESULTS AND DISCUSSION
■
Synthesis of Fluorescent Dyes 1a−g. Unexpectedly,
treatment of chloro-substituted cyanine 2a or 2b with resorcin
3a in the presence of a base, NaH or triethylamine, at 50 °C for
4 h in DMF did not furnish the straightforward nucleophilic
substitution product, but instead 2,3-dihydro-1H-xanthene-6-ol
1a or 1c (Scheme 1), respectively, was obtained. Similarly,
compound 1e, the analogue of compound 1a with a chlorine
atom nearby the phenolic alcohol, was also synthesized by the
same route. The structures of compounds 1a, 1c, and 1e were
In this study, we described the discovery of a new class of
NIR fluorescent dyes represented by 1a, 1c, and 1e (Figure 1).
Received: June 26, 2012
Published: July 20, 2012
© 2012 American Chemical Society
13510
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
Figure 1. The chemical structures of the classic and new dyes bear an optically tunable hydroxyl group. The chemical structures and absorption/
emission wavelengths of the classic dyes, such as 7-hydroxycoumarin and fluorescein, and the new NIR dyes reported herein (represented by 1a, 1c,
and 1e) with an optically tunable hydroxyl group.
a
Scheme 1. Structures and Synthesis of Compounds 1a−g
a
Conditions: (a) alkali, H2O, and/or heating; (b) CH3CH2I, Cs2CO3; and (c) CH3I, Cs2CO3.
Scheme 2. A Proposed Reaction Mechanism for the Unique Formation of 2,3-Dihydro-1H-xanthene-6-ol Dyes Represented by
1a
characterized by 1H and 13C NMR and HRMS (ESI).
Representative 2D COSY NMR spectra of compound 1c
further confirms the structure (Figure S1 and its description).
To get insight into the likely mechanism of the formation of
the new compounds exemplified by 1a, a number of control
experiments were performed: 1) Compound 2a was reacted
with resorcin 3a under the basic conditions at a lower
temperature (25 °C) for 2 h to afford compound 4 (Schemes
2 and S1), the product arising from the straightforward
nucleophilic substitution of chlorine atom; 2) The product 4
was isolated in pure form and further heated at 50 °C for 2 h in
the presence of a base, resulting in formation of product 1a
(Scheme S1). Thus, the results of these control experiments
13511
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
a
Scheme 3. Synthesis of Compounds 1c and 1d
a
Using a rational synthetic route based on the retrosynthesis analysis (Scheme S5) to corroborate the structures of this new type of compounds.
Conditions: (a) EtOH, AcOH, piperidine, reflux; (b) Ac2O, 50 °C; (c) BBr3, CH2Cl2, 0 °C.
■
◆
●
▲
Figure 2. Absorption (A) and fluorescence emission spectra (B) of compounds 1a ( , blue), 1b ( , green), 1c ( , purple), 1d (★, black), 1e ( ,
□
△
red), 1f ( , orange, open), and 1g ( , green, open) (10 μM) in pH 7.4 PBS/MeOH (1:1). (C) Fluorescence emission spectra of compounds 1a
■
◆
▲
(
●
, blue), 1b ( , green), 1c ( , purple), 1d (★, black), and 1e ( , red) (10 μM) in pH 7.4 PBS/MeOH (1:1). The emission spectra were
obtained by excitation at the maximum absorption wavelength (B) or 690 nm (C), respectively. (D) pH-dependence of the absorption spectra of
compound 1a (10 μM) with the arrows indicating the change of the absorption intensities with pH enhancement from 2.3 to 7.4.
suggest that compound 4 is the possible intermediate in the
mechanistic pathway.
a retro-Knoevenagel reaction by heating cyanine 2a at 50 °C
under the basic conditions (Schemes S2 and S3). This further
supports the proposed mechanism.
Based on the above key findings, a plausible mechanistic
pathway for the formation of 2,3-dihydro-1H-xanthene-6-ol
dyes represented by 1a is proposed in Scheme 2. We
hypothesized that under stringent conditions, cyanine 4 could
lose its Fischer’s base by a retro-Knoevenagel reaction to give a
putative half cyanine enol intermediate L. This hypothesis is
based on the chemistry of cyanine dye formation that a half
cyanine enol intermediate could condensate with Fischer’s base
to yield a cyanine dye,11 and the Knoevenagel-type
condensation reaction is reversible. The enol intermediate L
could then undergo cyclization and dehydration reactions to
furnish a much more stable product 1a. Although the enol
intermediate L was not isolated likely due to the facile
subsequent steps of cyclization and dehydration, the similar
chloro-substituted half cyanine enol could be furnished through
Additionally, the chloro-substituted half cyanine enol
generated by heating cyanine 2a at 50 °C under the basic
conditions was isolated (Scheme S2). Then, the chloro-
substituted half cyanine enol in pure form was further treated
with resorcin 3a in the presence of a base at 50 °C, which did
not afford product 1a (Scheme S4). Thus, this observation
excludes the possibility that chloro-substituted half cyanine enol
is the intermediate in the formation of product 1a, in good
agreement with the proposed mechanism as shown in Scheme
2. The conversion of compound 4 to the product 1a upon
further heating at a higher temperature (Scheme S1) is
interesting, indicating the delicate nature of cyanine chemistry
that the reaction conditions can dramatically affect the
structures of final products. Importantly, the formation of 2,3-
13512
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
dihydro-1H-xanthene-6-ols 1a, 1c, and 1e from cyanines 2a and
2b provides a unique pathway for the preparation of a wide
variety of 2,3-dihydro-1H-xanthene-6-ol derivatives from the
classic cyanines. It is worthy to note that this synthetic pathway,
to our best knowledge, is unprecedented.
Compounds 1b and 1d (Scheme 1), the ethyloxy and
methyloxy derivatives of compounds 1a and 1c, respectively,
were synthesized as controls to investigate the role of the
hydroxyl group on the modulation of absorption/emission
profiles. Reaction of compound 1a or 1c with iodoethane or
iodomethane, readily afforded compound 1b or 1d, respec-
tively. In addition, we also prepared compounds 1f and 1g
(Scheme 1), in which the hydroxyl/methoxy group is in the
tetrahydroacridine ring. Compounds 1f and 1g could be used as
references for compounds 1a−e to examine the potential effect
of different heteroatoms at the X-position on the optical
properties.
Table 1. Photophysical Data of the Dyes 1a−g in pH 7.4
PBS/MeOH (1:1)
a
b
c
d
λabs/nm
εmax (104 M−1 cm−1
)
λem/nm
Φf
pKa
1a
1b
1c
1d
1e
1f
690
7.9
5.4/4.9
8.7
716
677
716
677
718
565
565
0.36
0.007
0.37
5.6
−
5.6
656/608
690
654/608
698
4.9/5.3
7.9
0.006
0.30
−
4.5
454
1.5
0.009
0.006
5.1; 7.4
5.1
1g
454
2.0
a
b
The maximal absorption of the dye. The maximal emission of the
c
dyes. Φf is the relative fluorescence quantum yield estimated by using
indocyanine green (ICG, Φf = 0.13 in DMSO) for 1a−e or rhodamine
6G (Φf = 0.95 in water) for 1f−g as a fluorescence standard.14 The
d
pKa was calculated according to the Henderson−Hasselbach-type mass
action equation.12
To further corroborate the structures of this new type of
compounds, we wondered whether they could also be
independently prepared by a “rational” synthetic route based
on a retrosynthesis analysis (Scheme S5). We envisioned that
the intermediate oxonium 8 is unstable, thus these new dyes
may be synthesized in one pot without isolation of the
intermediate 8. For a representative case, as shown in Scheme
3, reaction of commercially available hydroxylbenzaldehyde 6
and ketone 7 in EtOH with a few drops of AcOH and
piperidine as the catalysts afforded a yellow residue, which was
further treated with the commercially available Fisher aldehyde
5 in Ac2O to give compound 1d. Removal of the methyl group
of 1d by BBr3 readily afforded compound 1c. To verify that the
compound 1c synthesized from the different routes (unique
route in Scheme 1 and rational route in Scheme 3) is the same
compound, an equivalent mixture of them was taken via 1H and
13C NMR spectra. As shown in Figures S2 and 3, the mixture
shows a single NMR spectrum. The same phenomenon is
observed for compound 1d (Figures S4 and 5). Furthermore,
the mass, absorption, and emission spectra of compounds 1c
and 1d synthesized from the different routes are essentially
identical (Figures S6 and 7), further supporting that this class
of new compounds can be obtained by different routes.
Significantly, the observation that compounds 1c and 1d could
be independently prepared by distinct synthetic routes (unique
route in Scheme 1 and rational route in Scheme 3), further
substantiating the structures of the new class of dyes.
Photophysical Properties of Fluorescent Dyes 1a−g.
The absorption and emission profiles of compounds 1a−g are
shown in Figure 2, and the photophysical data are compiled in
Table 1. Gratefully, compound 1a displays both absorption and
emission peaks in the NIR region with maximum at 690 and
716 nm, respectively, in pH 7.4 PBS/MeOH (1:1).
Importantly, compound 1a has a fluorescence quantum yield
of 0.36, which is relatively large for a NIR dye. Furthermore, the
Stokes shift of compound 1a is 26 nm, comparable to typical
cyanines. Thus, compound 1a appears to be a promising NIR
dye. On the other hand, the maximal absorption and emission
peaks of the reference 1b are at 656/606 and 677 nm,
respectively, which are significantly blue-shifted when com-
pared to those of compound 1a. Furthermore, the fluorescence
quantum yield of compound 1b is only 0.007, much less than
that of compound 1a. This is attributed to the alkyloxy
substituent of 1b which reduces the electron-donating ability of
the oxygen atom1c and thus forbids the formation of the
zwitterionic resonance form. Thus, these data suggest that the
optical properties of new NIR dye 1a can be regulated by
alkylation on the hydroxyl group. Like compounds 1a and 1b,
the absorption and emission spectra of dyes 1c−d exhibit the
similar changes. Interestingly, compounds 1b/1d exhibit a
shoulder peak in the emission profiles, which may be attributed
to the existence of distinct species in the ground state due to
charge localization.8l,m Notably, compounds 1a, 1c, and 1e
display strong fluorescence when excited at 690 nm, while
compounds 1b and 1d show almost no fluorescence when
excited at 690 nm (Figure 2C). The large fluorescence
difference (>200-fold) between these fluorophores (1a/1e vs
1b or 1c vs 1d) implies a method to develop a new NIR
fluorescence turn-on switch by breaking the C−O bond
through a chemical reaction.
To further understand the role of the hydroxyl group playing
on the photophysical properties, we investigated the pH effect
on the absorption/emission profiles of compound 1a. As shown
in Figure 2D, with the enhancement of pH from 2.3 to 7.4, the
absorption bands at around 656/608 nm, ascribed to the
phenolic form of 1a, undergo a red-shift to a peak at around
690 nm, attributed to the phenolate form of 1a (Scheme S6).
There is a well-defined isosbestic point at 662 nm in the
absorption spectra, indicating the equilibrium of the phenolic
and phenolate forms of 1a. The pH-dependence of the
emission profiles of compound 1a is consistent with that of
the absorption profiles. When excited at 690 or 645 nm,
enhancement of pH from 2.3 to 7.4 leads to a fluorescence
turn-on (Figure S8) or ratiometric (with a red-shift) response
(Figure S9), respectively. The pKa of compound 1a was
calculated to be 5.6 based on the Henderson−Hasselbach-type
mass action equation.12 Like compound 1a, compounds 1c and
1e also exhibit the absorption and emission peaks in the NIR
region (Figure 2 and Table 1) and show the similar pH-
dependent absorption (Figures S10 and S12) and fluorescence
(Figures S11 and S13) changes. However, the pKa of
compound 1e was calculated to be 4.5, which is significantly
lower when compared to that of compound 1a. This indicates
that the pKa of 2,3-dihydro-1H-xanthene-6-ol-based NIR
merocyanine dyes can be readily tailored by introducing
electron-withdrawing groups nearby the phenolic alcohol.
We also examined the absorption/emission properties of
tetrahydroacridines 1f and 1g (Figure 2 and Table 1). However,
compounds 1f and 1g show absorption and emission only in
the visible region in pH 7.4 PBS/MeOH (1:1). This
phenomenon may be ascribed to the pH-dependent equili-
13513
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
brium between the neutral and the ionic forms of push−pull
chromophores 1f and 1g (Scheme S7). In pH 7.4 PBS/MeOH
(1:1), the neutral form is dominant. The neutral form may have
relatively short absorption/emission wavelengths due to the
weak push−pull system.13 However, with the decrease of pH
values, the ionic forms which have the strong push−pull system
may become dominant, and a red-shift in absorption/emission
is thus observed (Figures S14−15). Since tetrahydroacridines
1f and 1g only exhibit emission in the visible region at pH 7.4
and we were interested in the development of NIR dyes with
potential for in vivo imaging applications, we decided to focus
on 2,3-dihydro-1H-xanthene-6-ols 1a, 1c, and 1e for further
studies, as they have emission in the NIR region at pH 7.4.
Significantly, the above findings that there is a striking
distinction between the optical properties of compounds 1a/1e
vs 1b or 1c vs 1d indicate that the new class of NIR fluorescent
dyes represented by 1a, 1c, and 1e may be exploited as a novel
strategy to design NIR fluorescent sensors by easy
modifications on the hydroxyl group. By sharp contrast,
regulation of the emission properties of the classic cyanines
by photoinduced electron transfer (PET) is very challenging
based on the Rehm−Weller equation.9
We further examined the photophysical properties of
compounds 1a, 1c, and 1e in biological media. Although
compounds 1a, 1c, and 1e appear to have limited solubility in
water, they exhibit micromolar solubility in biological media.
For example, compounds 1a, 1c, and 1e have a solubility of
approximately 6, 10, or 6 μM, respectively, in pure newborn calf
serum (without addition of any co-organic solvent) estimated
from the corresponding absorption spectra (Figure S16). In
addition, these NIR dyes show strong fluorescence in pure
newborn calf serum solution (Figure S17) with good
fluorescence quantum yields over 0.2 (Table S1). The living
cell fluorescence images and in situ (in cellulo) emission spectra
of compounds 1a, 1c, and 1e are shown in Figure S18. These
results indicate that the new NIR dyes are cell permeable and
have strong NIR fluorescence in living cells. Notably, the in situ
(in cellulo) excitation and emission spectra of compounds 1a,
1c, and 1e in HeLa cells resemble those of compounds 1a, 1c,
and 1e in pH 7.4 PBS/THF (1:1) (Figures S19−21),
suggesting that the NIR emission of compounds 1a, 1c, and
1e in living cells can be attributed to the single molecule of
dyes. To demonstrate the feasibility of developing water-soluble
NIR dyes, we have further synthesized compound 1h, which
bears several water-soluble groups (Schemes 4 and S8). As
Theoretical Calculations. To get insight into the optical
properties of the new functional dyes 1a−g, density functional
theory (DFT) calculations with the B3LYP exchange functional
employing 6-31G(d) basis sets using a suite of Gaussian 09
programs were performed. Figures 3 and S24−28 show the
representative optimized structures and molecular orbital plots
(LUMO and HOMO) of the dyes 1a−g. The C−C (N or O)
bond lengths (in pm) of 1a−g determined by DFT calculations
are depicted in Tables S2−11. In the case of 1a (phenolate
form), the indolium moiety is essentially coplanar and
conjugated with the 2,3-dihydro-1H-xanthene core (Figure
3). It is worthy to note that all the C−C bond lengths of the
conjugated 2,3-dihydro-1H-xanthene-indolium backbone are
very similar, at around 140 pm (Table S2). This value is the
intermediate between the typical carbon−carbon single (154
pm) and the double (134 pm) bonds, attributed to the strong
electronic delocalization and the partial decrease in the
carbon−carbon bond length alternation (BLA, difference
between consecutive single and double bonds) along the π-
conjugated system.13 This is further corroborated by the studies
of the terminal C−N and C−O bond lengths in the conjugated
system. The bond length of C14−N15 is 138 pm, which is the
intermediate between the lengths of C−N bond (single bond,
around 147 pm) and the iminium CN (double bond, around
128 pm). The bond length of C1−O17 is 127 pm, which is
relatively closer to CO (double bond, around 121 pm) than
C−O (single bond, around 144 pm), indicating that the O17
atom contributes pronouncedly to the π-conjugated system of
the dye 1a (phenolate form). The π electrons on the HOMO of
1a (phenolate form) are mainly located on the whole π-
conjugated 2,3-dihydro-1H-xanthene-indolium framework (in-
cluding the O17 atom), but the LUMO is mostly positioned at
the center of the conjugated 2,3-dihydro-1H-xanthene-vinyl
bridge including the N15 atom (Figure 3), where the BLA is
minimal. This distribution pattern of the π electrons on the
HOMO/LUMO is a typical feature of cyanine dyes. Cyanines
in the “cyanine limit” state feature a relatively sharp and intense
absorption band in the NIR spectral range in light of the
reduced vibronic contribution in the nonalternating (minimal
BLA) structure.13 Thus, the DFT calculation results are
consistent with the absorption spectrum of 1a (Figure 2A).
The C1−O17 bond length of the compound 1b is 135.3 pm
(Table S3), which is longer than that (126.8 pm) of C1−O17
of the compound 1a (phenolate form). This weakens the p−π
conjugation between the O17 atom and the π-conjugated
xanthene−indolium framework, and the energy levels of both
HOMO and LUMO of the compound 1b decrease relative to
those of the compound 1a (Figure 3). Upon close examination
reveals that the π electrons on the HOMO and LUMO of 1b
are primarily located on the π-conjugated xanthene−indolium
framework and the conjugated 2,3-dihydro-1H-xanthene-vinyl
bridge, respectively. When compared to compound 1a, the O17
atom of compound 1b contributes less to the HOMO/LUMO.
This suggests that the alkylation on the O17 atom indeed
decreases the electron-donating ability of the oxygen atom, and
compound 1b loses its “cyanine limit” structure to some extent.
In other words, the alkylation on O17 atom hinders the
resonance between the O17 atom and the xanthene−indolium
core. In addition, time-dependent DFT (TDDFT) calculations
indicate that the excitation HOMO−LUMO energy gap
between the S0 state and the S1 excited state of 1b (2.27 or
2.09 eV based on the optimized S0 or S1 state geometry) is
markedly higher than that of 1a (phenolate form) (2.13 or 2.02
Scheme 4. Structure of the Water-Soluble NIR Dye 1h
expected, dye 1h is soluble in pure PBS and newborn calf
serum (without addition of any co-organic solvent) with a
solubility of at least 2 mM in PBS and 3 mM in newborn calf
serum. The spectral properties of 1h in pure PBS and newborn
calf serum (Figures S22−23 and Table S1) resemble those of
1a.
13514
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
Figure 3. DFT optimized structures and molecular orbital plots (LUMO and HOMO) of dyes 1a (phenolate form), 1b, and 1e (phenolate form) in
MeOH based on the optimized ground-state geometry (S0). In the ball-and-stick representation, carbon, nitrogen, and oxygen atoms are colored in
gray, blue, and red, respectively.
eV based on the optimized S0 or S1 state geometry) (Figure
S24), which is in good agreement with the observation that the
measured absorption and emission wavelength of 1b is shorter
than that of 1a. The pH-dependence of the absorption spectra
of compound 1a (Figure 2D) can be explained in a similar
manner (Figure S25 and Tables S4). In addition, the drastic
distinction between the optical properties of the dyes 1c and 1d
also can be rationalized analogously (Figure S26 and Tables S5
and S6).
The dye 1e is the analogue of 1a with an electron-
withdrawing chlorine (Cl18) nearby the O17 atom. However,
the energy levels of HOMO/LUMO of 1e (phenolate form)
are relatively lower than those of 1a (phenolate form) due to
the electron-withdrawing effect of chlorine, which stabilizes the
aromatic system by electron delocalization (Figure 3). This may
explain the above observation that the dye 1e has a lower pKa
value than dye 1a. In addition, TDDFT calculations indicate
that the presence of the electron-withdrawing chlorine on the
aromatic system also decreases the excited-state energy (about
0.03 eV), which corresponds to a bathochromic shift of 7 nm in
the absorption maximum of 1e with respect to 1a (Table 2).
Interestingly, this red-shift between compounds 1e and 1a
obtained based on the TDDFT calculations is in good
agreement with the value (8 nm) acquired experimentally
(Table 2).
Dyes 1f and 1g, the tetrahydroacridine derivatives, show the
pH-dependent equilibrium between the neutral and the ionic
forms of push−pull chromophores (Scheme S7). DFT
calculation indicates that the C9−C11 bond (148 pm) in the
neutral form of the dye 1f is relatively closer to the carbon−
carbon single bond (154 pm) than the carbon−carbon double
bond (134 pm) (Figure S27, Table S8) and that the π electrons
on the LUMO of 1f (neutral form) are principally located on
the whole π-conjugated tetrahydroacridine moieties, but the
HOMO is mostly positioned at the indolium−vinyl bridge
moieties. Thus, at pH 7.4, when compared to 1a (phenolate
form), 1f (neutral form) loses the cyanine limit to some extent.
By contrast, in the ionic form of the dye 1f, all the C−C bond
lengths of the conjugated tetrahydroacridine−indolium back-
Table 2. TDDFT Excitation Energies (and Corresponding
Excitation/Absorption Wavelengths) and Experimental
Absorption Wavelengths of the Dyes 1a−g in MeOH
a
theory
experiment
b
form
ΔE, eV
λabs, nm
λabs, nm
1a
phenolate
phenolic
2.13
2.29
2.27
2.13
2.29
2.27
2.10
2.27
2.59
2.33
2.58
2.32
583
542
547
583
543
546
590
545
479
533
480
535
690
608
606
690
608
606
698
610
454
578
454
582
1b
1c
phenolate
phenolic
1d
1e
phenolate
phenolic
neutral
1f
ionic form
neutral
1g
ionic
a
The calculations are determined using B3LYP/6-31G(d) exchange−
b
correlation functionals and basis sets. The excitation HOMO−
LUMO energy gap.
bone are very similar, at around 140 pm (Figure S27 and Table
S9), ascribed to the strong electronic delocalization and the
partial decrease in the carbon−carbon BLA along the π-
conjugated system.13 Furthermore, the π electrons on the
HOMO of 1f (the ionic form) are mainly located on the whole
π-conjugated tetrahydroacridine−indolium framework, but the
LUMO is mostly positioned at the center of the conjugated
tetrahydroacridine−vinyl bridge (Figure S27), where the BLA is
minimal. In addition, TDDFT calculations indicate that the
excitation HOMO−LUMO energy gap of 1f (neutral form,
2.59 eV) is significantly higher than that of 1f (ionic form, 2.33
ev) (Table 2), which is in good agreement with the observation
that 1f exhibits a red-shift at lower pH values (Figure S14). The
pH-dependent absorption spectra of 1g (Figure S15a) also can
be explained in a similar manner (Figure S28 and Tables S10−
11).
13515
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
Scheme 5. Design and Synthesis of NIR Fluorescent Turn-on H2O2 Sensor NIR-H2O2
Figure 4. (a) Fluorescence spectra of NIR-H2O2 (1 μM) in the presence of various concentrations of H2O2 (0−250 μM) in phosphate buffer (pH
7.4, 1% DMSO) with excitation at 690 nm. Inset: Fluorescence intensity of NIR-H2O2 (1 μM) at 708 nm vs H2O2 concentration (0−250 μM). (b)
Fluorescence intensity (at 708 nm) of NIR-H2O2 (1 μM) with excitation at 690 nm in the presence of various species (250 μM): 1, blank; 2, H2O2;
−
−
−
3, CH3COOOH; 4, HOCl; 5, O2 ·; 6, NO; 7, ·OH; 8, Fe3+; 9, GSH; 10, NO2 ; 11, NO3 ; 12, TBHP; 13, vitamin C; 14, ·OtBu.
TDDFT calculations suggest that the maximal absorption
bands of 1a−g are attributed to the S1←S0 transition and that
the maximal emission bands of 1a−b are attributed to the S0←
S1 transition. The calculated maximum of absorption or
emission is shorter than that obtained experimentally (Table
2 and Figure S24), which is consistent with the previous report
that TDDFT calculations may underestimate the excitation or
emission wavelength due to the limitation of the exchange−
correlation functional.15 However, the deviations between the
experimental values and TDDFT data are often systematic and
can be corrected using appropriate linear scaling approaches.
Indeed, for 1a−g, as illustrated in Figure S29, the under-
estimated absorption maximum is rather systematic, and the
linear regression is characterized by a correlation coefficient
close to one. The experimental values and TDDFT data can be
well correlated by the following expression:
generation and dynamic fluctuation of H2O2 in living systems.
The construction of fluorescent sensors for H2O2 has attracted
great attention.20,21 However, the development of fluorescent
H2O2 sensors with both maximal absorption and emission
wavelengths in the NIR region (650−950 nm) is very
challenging, although it is highly desirable for biological
imaging of H2O2 in living animals.
Herein, employing the approach established above based on
the optical property studies of the new type of NIR fluorescent
dyes, we present NIR-H2O2 (Scheme 5) as an innovative NIR
fluorescent sensor for H2O2. The key features of NIR-H2O2
include absorption and emission in the NIR region, a very large
fluorescence turn-on response, high selectivity for H2O2, and
suitability for imaging H2O2 in both living cells and living mice.
Figure 4a shows the changes in the fluorescence emission
spectra when H2O2 was added in phosphate buffer containing
the sensor NIR-H2O2 (1 μM). The free sensor is almost
nonfluorescent in the absence of H2O2 when excited at around
690 nm. However, upon addition of increasing concentrations
of H2O2 (0−250 μM), a large fluorescence turn-on response
(up to 180-fold enhancement at 708 nm) was observed (Figure
4a). Notably, this is a great enhancement factor for an NIR
fluorescent sensor. The changes in the absorption spectra
(Figure S30) are consistent with the variations in the emission
profiles. Notably, the NIR sensor NIR-H2O2 can respond to
H2O2 in pure newborn calf serum (without addition of any co-
organic solvent) (Figure S31).
λexperiment = −618.532 + 2.245 × λtheory
(1)
Development of a New NIR Fluorescent Turn-on
Sensor for H2O2 and the Applications for Biological
Imaging in Living Mice. Reactive oxygen species (ROS) play
important roles in many physiological and pathological
processes.16 Hydrogen peroxide (H2O2), one of the key ROS,
is produced by activation of NADPH oxidase complexes during
cellular stimulation with cytokines, neurotransmitters, and
peptide growth.16e−h H2O2 serves as a signaling molecule in a
wide variety of signaling transduction processes, an oxidative
stress marker in aging and disease, and a defense agent in
response to pathogen invasion.16e−h,17 However, aberrant
production of H2O2 is associated with various diseases
including cancer, diabetes, neurodegenerative disorders, and
cardiovascular.18,19 Thus, it is of importance to monitor
Kinetics measurements of reaction of NIR-H2O2 (1 μM)
with H2O2 (100 μM) under the pseudo-first-order conditions
give an observed rate constant of kobs = 2.95 × 10−3 s−1. As
exhibited in Figure 4b, NIR-H2O2 is highly selective to H2O2
over other typical ROS and biorelevant species, such as
13516
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
Figure 5. Images of HeLa cells treated with the sensor NIR-H2O2. (a) Brightfield image of HeLa cells incubated with only NIR-H2O2 (1 μM) for 30
min; (b) fluorescence image of (a); (c) brightfield image of HeLa cells incubated with NIR-H2O2 (1 μM) for 30 min and then further treated with
H2O2 (10 μM); and (d) Fluorescence image of (c).
Figure 6. Imaging of H2O2 in living cells. (a) Fluorescence image of RAW 264.7 macrophages cells costained with NIR-H2O2 for 60 min from red
channel. (b−h) Fluorescence images of RAW 264.7 macrophages cells stimulated with PMA (3.0 μg/mL) and costained with NIR-H2O2, Hoechst
33258 (4.5 μM), and MitoTracker green for 60 min: (b) from blue channel (nuclear staining); (c) from green channel (mitochondria staining); (d)
from red channel; (e) overlay of the green and red channels; (f) overlay of the blue and green channels; (g) overlay of the blue and red channels; and
(h) overlay of the blue, green, and red channels.
Figure 7. Representative fluorescent images (pseudo-color) of in vivo H2O2 production from the peritoneal cavity of the mice with NIR-H2O2 during
an LPS-mediated inflammatory response: (a) Neither LPS nor NIR-H2O2 was injected for the negative control; (b) saline was injected in the ip
cavity of mice, followed by ip injection of NIR-H2O2 (160 nanomoles); and (c) LPS was injected into the peritoneal cavity of the mice, followed by
ip injection of NIR-H2O2 (160 nanomoles).The mice were imaged using a FMT 2500 LX quantitative tomography in vivo imaging system with an
excitation filter of 670 nm and an emission range of 690−740 nm 30 min after the NIR-H2O2 injection. (d) Quantification of fluorescence emission
intensity from the abdominal area of the mice of groups a−c.
CH3COOOH, HClO/−OCl, O2 ·, NO, ·OH, Fe3+, GSH,
emission spectrum of the sensor NIR-H2O2 in the presence of
H2O2 in HeLa cells resembles to that of compound 1e in pH
7.4 PBS/THF (1:1) (Figure S32), suggesting that the NIR
emission of the sensor NIR-H2O2 in living cells can be
attributed to the single molecule state.
Encouraged by the above promising results, we decided to
further examine the feasibility of the sensor NIR-H2O2 to
detect endogenously produced H2O2 in living macrophage
cells. When stimulated by phorbol myristate acetate (PMA),
macrophage cells may produce endogenous H2O2.22 The living
RAW264.7 macrophage cells loaded with only the NIR sensor
NIR-H2O2 (1 μM) display almost no fluorescence (Figure 6a).
However, the macrophage cells coincubated with PMA (3.0
μg/mL) and the sensor NIR-H2O2 (1 μM) exhibit a dramatic
−
−
−
NO2 , NO3 , tert-butyl hedroperoxide (TBHP), vitamin C, and
·OtBu. The high selectivity and the sensing mechanism of the
sensor NIR-H2O2 to H2O2 may be attributed to the unique
chemical properties of H2O2, as described previously by
Chang’s group.20d
We proceeded to evaluate the ability of the sensor NIR-
H2O2 to operate in live cells. HeLa cells incubated with NIR-
H2O2 (5 μM) for 30 min at 37 °C provide almost no
fluorescence (Figure 5b). However, when the living HeLa cells
loaded with NIR-H2O2 were further treated with H2O2, they
gave strong fluorescence (Figure 5d). These results imply that
NIR-H2O2 is cell membrane permeable and responsive to
H2O2 in the living cells. In addition, the in situ (in cellulo)
13517
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
Scheme 6. Design and Synthesis of NIR Fluorescent Turn-on Thiol Sensor NIR-Thiol
enhancement in the red emission (Figure 6d). These data
indicate that the new NIR sensor NIR-H2O2 is capable of
fluorescent imaging of endogenously produced H2O2 in the
living RAW264.7 macrophage cells. Notably, the MTT assays
(Figure S33)28 indicate that NIR-H2O2 of concentrations
below 60 μM have no marked cytotoxicity. In addition, the
mitochondria staining experiments (Figure 6c,e,h, and Figure
S34) suggest that the sensor mainly associates with the
mitochondria of RAW264.7 macrophage cells.
sensors have been developed.26 However, the applications of
them have been largely limited to cell culture, thin specimens,
and blood specimens, owing to the need for ultraviolet or
visible light excitation. Thus, the development of NIR
fluorescent thiol sensors with both absorption and emission
in the NIR for imaging of thiols in living mice remains a major
challenge in the field of optical imaging. Toward this end, we
prepared compound NIR-thiol as a novel candidate of NIR
fluorescent thiol sensor by modifications on the hydroxyl group
of NIR dye 1a with 2,4-dinitrobenzenesulfonate moiety27
(Scheme 6) for optical imaging of endogenous thiols in living
mice for the first time.
The major advantage of the new NIR-H2O2 over the sensors
based on traditional 7-hydroxycoumarin and fluorescein is that
our NIR-H2O2 has both absorption and emission in the NIR
region. Thus, the unique NIR properties of NIR-H2O2 render
NIR-H2O2 highly favorable for fluorsecence imaging of
endogenously produced H2O2 in living animals. To examine
this possibility, the sensor was treated with living mice. The
H2O2 production in vivo was generated by activated macro-
phages and neutrophils in a lipopolysaccharide (LPS) model of
acute inflammation.23 The ICR mice were divided into three
groups: Group one was untreated as the control group; the
second group was given saline (400 μL) in the peritoneal
cavity, followed by intraperitoneal (ip) injection with the sensor
NIR-H2O2 (160 nanomoles); and the third group was given an
ip injection of LPS (1 mg in 400 μL saline) and followed by ip
injection with the sensor NIR-H2O2 (160 nanomoles) after 4 h.
The mice were anaesthetized, and the abdominal fur was
removed. The mice were imaged using a FMT 2500 LX
quantitative tomography in vivo imaging system 30 min after
the sensor injection. The mice treated with both LPS and the
sensor exhibit a significantly higher fluorescence readout
(pseudo-color) (Figure 7c) than the mice untreated (Figure
7a) or treated with only the sensor (Figure 7b). The fluorescent
intensity from the abdominal area of the mice was quantified,
and the data indicate that the mice loaded with LPS and the
sensor have approximately 10- and 20-fold higher fluorescence
intensity than the mice loaded with saline and the sensor and
the mice loaded with saline, respectively (Figure 7d). Thereby,
these results demonstrate that the new NIR sensor NIR-H2O2
is suitable for imaging endogenously produced H2O2 in the
living animals.
As designed, the free sensor NIR-thiol is essentially
nonfluorescent in PBS buffer (Figure 8). However, the addition
Figure 8. Fluorescence spectra of NIR-thiol (10 μM) in the presence
of various concentrations of Cys (0−70 μM) in pH 7.4 PBS/CH3CN
(7:3) with excitation at 690 nm. Inset: Fluorescence intensity of NIR-
thiol (10 μM) at 716 nm vs Cys concentration (0−70 μM).
of representative thiol, cysteine (Cys), elicited a dramatic
change in the fluorescence spectra. A strong new emission peak
at 716 nm appeared with a 50-fold enhancement. As displayed
in the inset of Figure 8, low-micromolar thiols can be readily
detected. Thus, the sensor should be sensitive enough to
monitor intracellular thiols, which typically are in the millimolar
level. The changes in the absorption spectra (Figure S35) are
consistent with the variations in the emission profiles. The time
course of the fluorescence intensity at 716 nm in the absence or
presence of Cys was exhibited in Figure S36. Upon
introduction of Cys, a significant enhancement in the emission
intensity was noted, and the intensity essentially reached
maximum within 60 s. This indicates that NIR-thiol responds
very rapidly to thiols, and the sensor may have potential for
monitoring of thiols in real time. Notably, the free sensor NIR-
thiol is stable in the assay conditions over 24 h (Figure S36B),
and the free sensor in the solid form can be stored for more
than one year. As shown in Figure S37, NIR-thiol is highly
selective to typical small molecular weight biological thiols
(Cys, Hcy, and GSH) over other biorelevant species, such as
Phe, Gly, Arg, Lys, Tyr, Leu, glucose, Ser, and Val, suggesting
Development of a Novel NIR Fluorescent Turn-on
Sensor for Thiols and the Applications for Biological
Imaging in Living Mice. To further test the robust nature of
the approach for NIR sensor development, we constructed
NIR-thiol (Scheme 6) as a novel NIR fluorescent thiol sensor.
Small molecular weight biological thiols play a critical role in
many biological processes. However, abnormal levels of thiols
are implicated in a variety of diseases including liver damage,
skin lesions, and slowed growth.24,25 Thus, it is of high interest
to monitor biological thiols. A number of fluorescent thiol
13518
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
Figure 9. (a−c) Live Bel 7702 cells co-incubated with NIR-thiol (5 μM) and Hoechst 33258 (4.5 μM) for 30 min. (a) Emission from the blue
channel (nuclear staining); (b) emission from the red channel; and (c) overlay of the blue and red channels. (d) Fluorescence image (red channel)
of Bel 7702 cells pretreated with N-ethylmaleimide (1 mM) for 30 min and then incubated with NIR-thiol (5 μM) for 30 min.
Figure 10. Representative fluorescence images (pseudo-color) for the mice ip injected with a different amount [(a) 0, (b) 20, (c) 40, or (d) 160
nanomoles] of NIR-thiol. (e) Quantification of the fluorescence intensities from the mice (ip) injected with a different amount (0, 20, 40, or 160
nanomoles) of NIR-thiol. The images were obtained 10 min after the injection of NIR-thiol into the ip cavity of the mice. The mice were imaged
using a FMT 2500 LX quantitative tomography in vivo imaging system, with an excitation filter of 670 nm and an emission filter of 690−740 nm.
that the sensor is promising for applications in living systems.
Furthermore, NIR-thiol can respond to thiols in pure newborn
calf serum solution (without addition of any co-organic
solvent) (Figure S38). The sensing mechanism of the sensor
NIR-thiol can be attributed to the cleavage of the dinitrophenyl
moiety by thiols.27
fluorescence (pseudo-color) was noted for the mice injected
with 20 nanomoles of NIR-thiol, whereas almost no
fluorescence was observed in the control mice without being
injected with the sensor (Figure 10a). In addition, the mice
injected with a higher amount (40 or 160 nanomoles) of NIR-
thiol provided brighter fluorescence (pseudo-color) (Figure
10c,d). Quantification of the fluorescence intensities from the
mice (ip) injected with a different amount (0, 20, 40, or 160
nanomoles) of NIR-thiol reveals a dose-dependent increase in
signal as a function of the sensor amount (Figure 10e). A very
large signal-to-noise contrast ratio (ca. 700-fold enhancement)
was observed for the mice injected with 20 nanomoles of NIR-
thiol. In addition, the signal-to-noise contrast ratio jumped to
ca. 2900- or 7400-fold enhancement when the mice were
injected with 40 or 160 nanomoles of NIR-thiol, respectively.
Taken together, these data indicate that NIR-thiol is a robust
imaging agent for visualizing endogenous thiols in the context
of living mice for the first time.
Previously, we have developed a class of NIR dyes, Changsha
(CS) NIR dyes, which are superior to the traditional rhodamine
dyes with both absorption and emission in the NIR region
while retaining the rhodamine-like fluorescence on−off switch-
ing mechanism.8n Notably, there are drastic distinctions
between the CS NIR dyes and the novel type of NIR dyes
reported herein. First, the design strategies for these two types
of dyes are largely different. The CS NIR dyes were rationally
designed based on the “integration” of a benzoic acid moiety
with a merocyanine unit inspired by the fact that the electronic
pulling pathway for the CS NIR functional dye spirocyclization
highly resembles that of the traditional rhodamine dye
spirocyclization. By sharp contrast, the new type of NIR dyes
described herein was constructed based on the new reaction
between chloro-substituted cyanine with resorcin under basic
conditions. In addition, the mechanistic studies reveal that the
new reaction likely involves a cascade sequence of putative
To assess the ability of the sensor NIR-thiol for NIR imaging
of thiols in living cells, the sensor was incubated with living
HeLa or Bel 7702 cells, and bright fluorescence was observed
(Figures 9b, S39B). By contrast, in a control experiment, the
HeLa or Bel 7702 cells were pretreated with N-ethylmaleimide
(a thiol-reactive agent) and further incubated with NIR-thiol,
where almost no fluorescence was noted (Figures 9d, S39D),
suggesting the selective reaction of the sensor with thiols. Thus,
the sensor appears to be cell membrane permeable and capable
of NIR imaging of thiols in living cell. The NIR bioimaging is
advantageous over visible bioimaging in terms of negligible
background fluorescence of native cellular species. The in situ
(in cellulo) emission spectrum of the sensor NIR-thiol in HeLa
cells is very similar to that of compound 1a in pH 7.4 PBS/
THF (1:1) (Figure S40), suggesting that the NIR emission of
NIR-thiol in living cells can be attributed to the single molecule
state. In addition, the MTT assays (Figure S41)28 indicate that
NIR-thiol of concentrations below 50 μM does not show
noticeable cytotoxicity.
Thus, the prominent features of NIR-thiol include NIR
absorption and emission, rapid response, high sensitivity,
excellent selectivity, good cell membrane permeability, and
low cytotoxicity. These desirable attributes prompted us to
further examine the suitability of the sensor for visualizing
endogenous thiols in living animals. A different amount (0, 20,
40, or 160 nanomoles) of NIR-thiol was injected into the ip
cavity of mice. The mice were imaged using a FMT 2500 LX
quantitative tomography in vivo imaging system 10 min after
the NIR-thiol injection. As shown in Figure 10b, strong
13519
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
retro-Knoevenagel, cyclization, and dehydration. Second, the
CS NIR dyes may be “considered” as the NIR surrogates of the
classic rhodamine dyes. By contrast, the new type of NIR dyes
described herein appear to be the NIR surrogates of the
traditional fluorescein/7-hydroxycoumarin dyes. Thus, more
importantly, the distinction between the CS NIR and the new
NIR dyes is in their use as platforms for fluorescent sensor
design. Just like the rhodamine platform,5j−l the CS NIR dyes
are mainly applicable for the design of fluorescent sensors based
on the carboxylic acid functionalized reactive sites (Scheme 7).
performed quantum chemical calculations with the B3LYP
exchange functional employing 6-31G(d) basis sets to shed
light on the structure-optical properties of the new class of NIR
dyes. Moreover, we have demonstrated that the approach can
be easily employed to develop different types of NIR
fluorescent sensors for biological imaging applications in the
living animals: NIR-H2O2 for endogenously produced H2O2
and NIR-thiol for endogenous thiols, highlighting the value of
the new NIR fluorescent sensor design approach. We expect
that the new NIR dyes described herein will serve as useful
fluorescent platforms for the development of fluorescent
sensors based on the hydroxyl functionalized reactive sites, a
useful complement to our previously reported CS NIR dyes,
which are mainly applicable for the design of fluorescent
sensors based on the carboxylic acid functionalized reactive
sites.
Scheme 7. Schematic Illustration of the New NIR and CS
NIR dyes, Fluorescein/7-Hydroxycoumarin, And
a
Rhodamine As the Platforms for Fluorescent Sensor Design
EXPERIMENTAL SECTION
■
Synthesis of Compound 1a. Resorcin 3a (86.0 mg, 0.78 mmol)
and NaH (60% in mineral oil, 19.1 mg, 0.78 mmol) or triethylamine
(0.3 mL) were placed in a flask containing DMF (2 mL), and the
mixture was stirred at room temperature under nitrogen atmosphere
for 10 min. Compound 2a (200.1 mg, 0.31 mmol) in DMF (1.0 mL)
was introduced to the mixture via a syringe, and the reaction mixture
was heated at 50 °C for 4 h. The solution was then removed under
reduced pressure. The crude product was purified by silica gel flash
chromatography using petroleum ether/CH2Cl2/MeOH (25: 25: 1) as
eluent to give compound 1a as a blue-green solid (90.4 mg, yield
a
Notably, the relationship between the new NIR and CS NIR dyes is
just like that between fluorescein/7-hydroxycoumarin and rhodamine.
They are essentially complementary but not replaceable.
Whereas, just like the fluorescein/7-hydroxycoumarin plat-
form,5l,10,20d the new NIR dyes are primarily applicable for the
design of fluorescent sensors based on the hydroxyl function-
alized reactive sites (Scheme 7). For instance, the CS NIR dyes
could be employed to design a NIR fluorescent sensor for
HOCl based on thiosemicarbazide chemistry (a carboxylic acid
functionalized reactive site),8n but they are not suitable for
development of NIR fluorescent sensors for H2O2 and thiols
based on the borate chemistry20d and 2,4-dinitrobenzenesulfo-
nate chemistry27 (hydroxyl functionalized reactive sites). On
the other hand, the new NIR dyes could be used to construct
NIR fluorescent sensors for H2O2 and thiols based on borate
chemistry20d and 2,4-dinitrobenzenesulfonate chemistry27
(hydroxyl functionalized reactive sites) as described in this
work, but they are not applicable for development of an NIR
fluorescent sensor for HOCl based on thiosemicarbazide
chemistry (a carboxylic acid functionalized reactive site).8n In
other words, from the fluorescent sensor design point of view,
the novel type of NIR and the CS NIR dyes have different
scopes and limitations. They are essentially complementary but
not replaceable, just like the relationship between fluorescein/
7-hydroxycoumarin and rhodamine.
1
55.0%). H NMR (400 MHz, CDCl3): δ 1.41−1.45 (t, J = 7.2 Hz,
3H), 1.69 (s, 6H), 1.88−1.91 (m, 2H), 2.63−2.66 (t, J = 5.6 Hz, 2H),
2.69−2.72 (t, J = 5.6 Hz, 2H), 4.07−4.12 (q, J = 6.8 Hz, 2H), 5.92−
5.96 (d, J = 14.0 Hz, 1H), 6.97−7.00 (2H), 7.06−7.08 (d, J = 8.0 Hz,
1H), 7.19−7.23 (t, J = 7.6 Hz, 1H), 7.24−7.26 (d, J = 8.8 Hz, 1H),
7.35−7.39 (3H), 8.37−8.40 (d, J = 14.0 Hz, 1H). 13C NMR (100
MHz, CDCl3): δ 12.09, 20.75, 24.35, 28.39, 39.29, 48.92, 97.80,
103.01, 107.87, 109.87, 114.85, 115.04, 121.25, 121.45, 122.29, 124.57,
128.63, 129.35, 139.01, 139.79, 140.58, 141.89, 156.87, 162.32, 170.76.
HRMS (ESI) m/z calcd for C27H28NO2 (M+): 398.2115. Found:
398.2112.
Synthesis of Compound 1b. To a solution of compound 1a
(200.0 mg, 0.38 mmol) and iodoethane (592.8 mg, 3.80 mmol) in dry
DMF (2.0 mL), Cs2 CO3 (248.0 mg, 0.76 mmol) was added. The
mixture was stirred at room temperature for 1 h, and then the solvent
was evaporated under reduced pressure to give the crude product,
which was purified by silica gel flash chromatography using petroleum
ethyl acetate/acetone (3:1) as eluent to give compound 1b as a blue
solid (165.9 mg, 79.5%). 1H NMR (400 MHz, CDCl3): 1.40−1.48 (m,
6H), 1.74 (s, 6H), 1.87 (bs, 2H), 2.69 (bs, 4H), 4.11- 4.14 (q, J = 7.2
Hz, 2H), 4.40- 4.43 (q, J = 6.8 Hz, 2H), 6.30- 6.40 (dd, J = 14.8, 5.0
Hz, 1H), 6.76 (s, 1H), 6.77−6.84 (dd, 1H), 7.30−7.41 (m, 5H), 7.47−
7.49 (dd, J = 7.2, 2.0 Hz, 1H), 8.55−8.57 (d, J = 14.2 Hz, 1H). 13C
NMR (100 MHz, CDCl3): δ 11.94, 13.65, 19.31, 23.67, 27.30, 28.16,
40.39, 49.56, 63.81, 100.21, 102.19, 111.42, 112.69, 113.92, 114.72,
121.69, 126.11, 126.21, 128.07, 128.26, 133.56, 140.02, 140.71, 144.59,
153.51, 161.09, 161.60, 175.38. HRMS (ESI) m/z calcd for
CONCLUSION
■
Fluorescence microscopy with synthetic sensors offers a
powerful technique for studying biomolecules in their native
environment. NIR fluorescent sensors are advantageous over
visible light fluorescent sensors as NIR light leads to minimum
photodamage to biological samples, deeper tissue penetration,
and minimum interference from background autofluorescence.
Thus, robust strategies for design of NIR fluorescent sensors
are highly sought after. In this study, we have successfully
developed a new approach for ready construction of NIR
fluorescent sensors based on a novel class of NIR dyes
represented by 1a, 1c, and 1e, which are superior to the
traditional 7-hydroxycoumarin and fluorescein while retaining
an optically tunable hydroxyl group. In addition, we have
+
C29H32NO2 (M+): 426.2428. Found: 426.2436.
Synthesis of Compound 1c. Route A (shown in Scheme 1):
Resorcin 3a (275.6 mg, 2.48 mmol) and triethylamine (0.3 mL) were
placed in a flask containing DMF (2 mL), and the mixture was stirred
at room temperature under nitrogen atmosphere for 10 min.
Compound 2b (400.0 mg, 0.83 mmol) in DMF (1.0 mL) was
introduced to the mixture via a syringe, and the reaction mixture was
heated at 50 °C for 5 h. The solution was then removed under reduced
pressure. The crude product was purified by silica gel flash
chromatography using petroleum ether/ethyl acetate (2:1) to
CH2Cl2/ethanol (1:0 to 50:3) as eluent to give compound 1c as a
blue-green solid (252.4 mg, yield 59.5%).
13520
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
Route B (shown in Scheme 3): Compound 1d (200.2 mg, 0.44
mmol) was placed in a flask containing anhydrous CH2Cl2 (3 mL),
and the solution was stirred at 0 °C under nitrogen atmosphere for 30
min. BBr3 (1 mL BBr3 dissolved in 3 mL CH2Cl2) was slowly
introduced to the mixture via a syringe, and the reaction mixture was
warmed to room temperature and stirred at room temperature for 24
h. The mixture was added to a saturated NaHCO3 solution (50 mL)
and was extracted with CH2Cl2 (50 mL) three times. The organic
phases were combined and dried by anhydrous MgSO4. The solution
was then removed under reduced pressure. The crude product was
purified by silica gel flash chromatography using petroleum ether/ethyl
acetate (2:1) to CH2Cl2/ethanol (15:1) as eluent to give compound
1c as a blue-green solid (182.6 mg, yield 89.7%).
139. 97, 142.70, 157.42, 160.57, 166.39. HRMS (ESI) m/z calcd for
C27H27NO2Cl (M+): 432.1725. Found: 432.1742.
Synthesis of Compound 1f. 3-aminophenol 3c (338.0 mg, 3.11
mmol) and compound 2a (197.8 mg, 0.31 mmol) were placed in a
flask containing DMF (3.0 mL). After heating for 4 h at 80 °C under
nitrogen, the solution was concentrated under reduced pressure. The
resulting crude product was purified by silica gel flash chromatography
using petroleum ether/CH2Cl2/MeOH (25:25:1) as eluent to give
1
compound 1f as a red solid (105.6 mg, yield 65.2%). H NMR (400
MHz, CDCl3): δ 8.33−8.36 (d, J = 12.8 Hz, 1H), 7.63 (s, 1H), 7.49−
7.51 (d, J = 8.8 Hz, 1H), 7.41 (d, J = 2.4 Hz, 1H), 7.15−7.18 (m, 2H),
6.96−6.98 (dd, J = 8.8, 2.4 Hz, 1H), 6.84−6.88 (t, J = 7.2 Hz, 1H),
6.61−6.63 (d, J = 7.6 Hz, 1H), 5.54−5.58 (d, J = 12.8 Hz, 1H), 3.69−
3.74 (q, J = 7.2 Hz, 2H), 2.87−2.90 (t, J = 6.0 Hz, 2H), 2.76−2.78 (t, J
= 5.6 Hz, 2H), 1.91−1.97 (m, 2 H), 1.71 (s, 6H), 1.24−1.28 (t, J = 6.8
Hz, 3H). 13C NMR (100 MHz, CDCl3) 158.65, 157.89, 155.47,
147.82, 144.06, 139.59, 134.28, 128.44, 127.93, 127.51, 127.15, 124.07,
121.68, 121.43, 119.66, 117.79, 109.29, 105.88, 92.75, 46.00, 36.83,
30.48, 28.46, 26.51, 22.41, 10.99; HRMS (EI) m/z calcd for
1H NMR (500 MHz, CDCl3): δ 1.67 (s, 6H), 1.87−1.92 (m, 2H),
2.62 (t, J = 6.0 Hz, 2H), 2.67 (t, J = 6.0 Hz, 2H), 3.36 (s, 3H), 4.07−
4.12 (q, J = 7.2 Hz, 2H), 5.60 (d, J = 13.5 Hz, 1H), 6.58 (s, 1H), 6.78
(dd, J = 9.0, 1.5 Hz, 1H), 6.84 (d, J = 8.0 Hz, 1H), 7.05 (t, J = 8.0 Hz,
1H), 7.21 (d, J = 9.0 Hz, 1H), 7.26−7.29 (2H), 7.32(s, 1H), 8.08 (d, J
= 13.0 Hz, 1H). 13C NMR (125 MHz, CDCl3): δ 21.34, 24.52, 27.96,
28.59, 29.70, 30.08, 47.98, 94.47, 103.61, 107.82, 116.36, 116.51,
116.71, 122.02, 122.29, 127.37, 128.13, 129.79, 132.43, 139.59, 140.06,
143.87, 159.02, 159.91, 166.07. HRMS (ESI) m/z calcd for
+
C27H28N2O1 [M−H] : 396.2196. Found: 396.2193.
Synthesis of Compound 1g. 3-methoxyaniline 3d (381.2 mg,
3.10 mmol) and compound 2a (200.5 mg, 0.31 mmol) were placed in
a flask containing DMF (3.0 mL). After heating for 4 h at 150 °C
under nitrogen, the solution was concentrated under reduced pressure.
The resulting crude product was purified by silica gel flash
chromatography using petroleum ether/CH2Cl2/MeOH (25:25:1) as
eluent to give compound 1g as a red solid (81.6 mg; yield 48.9%). 1H
NMR (400 MHz, CDCl3, TMS) δ 8.44 (d, J = 12.8 Hz, 1H), 7.58 (s,
1H), 7.47−7.49 (d, J = 8.8 Hz, 1H), 7.32 (s, 1H), 7.13−7.17 (m, 2H),
6.99−7.02 (m, 1H), 6.82−6.86 (m, 1H), 6.58−6.60 (d, J = 8.0 Hz,
1H), 5.56−5.59 (d, J = 13.2 Hz, 1H), 3.94 (s, 3H), 3.66−3.72 (m,
2H), 2.85 (t, J = 4.2 Hz, 2H), 2.76 (m, 2H), 1.87−1.92 (m, 2H), 1.77
(s, 6H), 1.24 (t, J = 4.2 Hz, 3H). 13C NMR (100 MHz, CDCl3)
159.65, 157.48 155.47, 144.11, 139.16, 133.32, 129.90, 128.50, 127.51,
127.48, 125.24, 125.08, 121.80, 121.50, 119.29, 117.35, 107.69, 106.81,
+
C26H26NO2 (M+): 384.1958. Found: 384.1971.
Synthesis of Compound 1d. Route A (shown in Scheme 1): To
a solution of compound 1c (200.0 mg, 0.39 mmol) and iodomethane
(110.9 mg, 0.80 mmol) in dry DMF (2.0 mL), Cs2 CO3 (248.0 mg,
0.76 mmol) was added. The mixture was stirred at room temperature
for 1 h, and then the solvent was evaporated under reduced pressure to
give the crude product, which was purified by silica gel flash
chromatography using petroleum/ethyl acetate (1:1) to CH2Cl2/
ethanol (15:1) as eluent to give compound 1d as a blue solid (167.5
mg, 81.6%).
Route B (shown in Scheme 3): To a solution of 2-hydroxy-4-
methoxybenzaldehyde 6 (1.0 g, 6.6 mmol) and cyclohexanone 7 (1.3
g, 13.2 mmol) in ethanol (50.0 mL) was added piperidine (0.6 mL)
and acetic acid (0.2 mL). The mixture was heated and stirred at reflux
condition for 12 h, and then the solvent was evaporated under reduced
pressure to give a residue. The resulting residue was treated with
Fisher aldehyde 5 (2.8 g, 14.0 mmol) in acetic anhydride (10 mL), and
the mixture was heated and stirred at 50 °C for 8 h. Subsequently, the
solvent was evaporated under reduced pressure to give the crude
product which was further purified by silica gel flash chromatography
using petroleum/ethyl acetate (3:1) to CH2Cl2/ethanol (1: 0 to 15:1)
as eluent to afford compound 1d as a blue solid (547.6 mg, 19.5%).
1H NMR (400 MHz, CDCl3): 1.82 (s, 6H), 1.91−1.97 (m, 2H)
2.74−2.77 (t, J = 6.0 Hz, 2H), 2.79−2.82 (t, J = 6.0 Hz, 2H), 3.98 (s,
3H), 4.04 (s, 3H), 6.49−6.53 (d, J = 14.8 Hz, 1H), 6.88−6.91 (2H),
7.29 (1H), 7.37−7.43 (3H), 7.46−7.51 (2H), 8.63−8.67 (d, J = 15.2
Hz, 1H). 13C NMR (100 MHz, CDCl3): δ 20.41, 24.85, 28.36, 29.32,
29.78, 34.50, 50.61, 56.60, 100.90, 104.36, 112.69, 113.49, 115.33,
115.89, 122.51, 127.32, 127.44, 128.99, 129.30, 134,19, 141.55, 142.27,
145.73, 154.62, 162.02, 163.24, 177.52; HRMS (ESI) m/z calcd for
105.62, 103.64, 100.76, 92.53, 55.26, 45.76, 36.61, 30.37, 28.60, 26.23,
+
22.32, 10.86. HRMS (EI) m/z calcd for C28H31N2O+ [M−H]
410.2353. Found: 410.2344.
:
Synthesis of Compound NIR-H2O2. Compound 1e (110 mg,
0.226 mmol) and CsCO3 (130 mg,0.678 mmol) were dissolved in
CH2Cl2. The mixture was stirred at room temperature for 10 min, and
then compound 9 (180 mg, 0.678 mmol) was added. After overnight
reaction, the solvent was removed under reduced pressure. The
resulting residue was purified by a silica gel column (CH2Cl2/
C2H5OH = 10:1) to afford compound NIR-H2O2 as a blue solid
1
(102.9 mg, isolated yield: 65.7%): H NMR (500 Hz, MeOD): δ =
1.37−1.40 (t, J = 7.5 Hz, 3H), 1.7 (s, 6H), 1.8−1.82 (t, J = 6.0 Hz,
2H), 2.58−2.6 (t, J = 6 Hz, 2H), 2.62 −2.65 (t, J = 6 Hz, 2H), 4.30−
4.34 (d, J = 7 Hz, 2H), 5.25 (s, 2H), 6.41−6.44 (d, J = 15 Hz, 1H), 7.0
(s, 1H), 7.11 (s, 1H), 7.37−7.50 (6H), 7.59−7.6 (d, J = 7 Hz, 2H),
7.7−7.72(d, J = 7.5 Hz, 1H),8.54−8.57 (d, J = 14.0 Hz,1H). 13C NMR
(125 Hz, MeOD): 10.22, 13.97, 19.6, 22.52, 26.00, 29.15, 31.14, 42.65,
53.24, 59.33, 71.78, 73.71, 103.89, 106.37, 112.19, 114.98, 116.88,
118.50, 122.60, 124.86, 128.32, 129.78, 130.12, 130.90, 131.34, 133.90,
136.45, 143.43, 144.84, 148.41, 155.06, 159.16, 163.10, 180.33. HRMS
(ESI) m/z calcd for C34H34*B1Cl1N1O4 ([M]): 565.2300. Found:
565.2317.
Synthesis of Compound NIR-thiol. To a solution of compound
1a (100.0 mg, 0.19 mmol) and 2,4-dinitrobenzenesulfonyl chloride
(203.4 mg, 0.76 mmol) in dry CH2Cl2 (10 mL), Cs2 CO3 (30.1 mg,
0.091 mmol) was added. The mixture was stirred at room temperature
for 30 min, and then the solvent was evaporated under reduced
pressure to give the crude product, which was purified on a basic Al2O3
column eluted with CH2Cl2/C2H5OH (100:1 to 30:1) to yield
compound NIR-thiol as a blue solid (85.9 mg, yield 60.0%). 1H NMR
(400 MHz, CDCl3): δ 1.54−1.57 (t, 3H), 1.81 (s, 6H), 1.90 (2H),
2.72 (4H), 4.54−4.56 (q, J = 7.2 Hz, 2H), 6.64−6.68 (d, J = 15.2 Hz,
1H), 7.02−7.04 (2H), 7.21 (s, 1H), 7.37−7.39 (d, J = 8.4 Hz, 1H),
7.46−7.50 (2H), 7.51−7.57 (2H), 8.23 (d, J = 2.0 Hz, 1H), 8.26−8.29
(1H), 8.32−8.35 (d, J = 8.8 Hz, 1H), 8.38−8.40 (d, J = 8.8 Hz, 1H),
+
C27H28NO2 (M+): 398.2115. Found: 398.2127.
Synthesis of Compound 1e. 4-chloro resorcin 3b (134.1 mg,
0.93 mmol) and triethylamine (0.3 mL) were placed in a flask
containing DMF (2.0 mL), and the mixture was stirred at room
temperature under nitrogen atmosphere for 10 min. Compound 2a
(200.0 mg, 0.31 mmol) in DMF (1.0 mL) was introduced to the
mixture via a syringe, and the reaction mixture was heated at 75 °C for
4 h. The solution was then removed under reduced pressure. The
crude product was purified by silica gel flash chromatography using
CH2Cl2/EtOH (50:1) as eluent to give compound 1e as a blue-green
1
solid (60.7 mg, yield 35.0%). H NMR (500 MHz, CDCl3): δ 1.36−
1.39 (t, J = 7.2 Hz, 3H), 1.67 (s, 6H), 1.89−1.92 (2H), 2.61−2.64 (t, J
= 6.0 Hz, 2H), 2.68−2.70 (t, J = 6.0 Hz, 2H), 3.89−3.93 (q, J = 7.2
Hz, 2H), 5.67−5.70 (d, J = 13.5 Hz, 1H), 6.77 (s, 1H), 6.88−6.89 (d, J
= 8.0 Hz, 1H), 7.08−7.11 (t, J = 7.5 Hz, 1H), 7.29−7.33 (3H), 7.41 (s,
1H), 8.13−8.16 (d, J = 13.5 Hz, 1H). 13C NMR (125 MHz, CDCl3): δ
11.66, 21.19, 24.44, 28.07, 28.57, 38.24, 47.90, 94.74, 103.82, 108.23,
115.22, 115.72, 117.54, 122.24, 122.87, 127.26, 128.27, 134.45, 139.19,
13521
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
8.60−8.66 2H), 8.68 (s, 1H). 13C NMR (125 MHz, CDCl3): δ 13.16,
20.07, 24.11, 27.78, 28.93, 29.48, 41.69, 51.36, 106.68, 110.10, 113.31,
115.72, 118.23, 118.69, 120.52, 121.68, 122.74, 125.29, 127.42, 128.39,
128.81, 128.94, 129.46, 129.66, 130.91, 131.56, 131.86, 133.05, 134.39,
140.79, 142.48, 145.36, 146.60, 147.45, 148.21, 148.83, 149.49, 151.14,
152.92, 158.79, 167.79, 178.65. HRMS (ESI) m/z calcd for
C33H30N3O8S+ (M+): 628.1748. Found: 628.1749.
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures and some spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
weiyinglin@hnu.edu.cn
■
Spectroscopic Studies of the Sensor NIR-H2O2. Superoxide
(O2· −) was added as solid KO2. Hydroxyl radical and tert-butoxy
radical (·OtBu) were generated by reaction of Fe2+ with H2O2 or
TBHP, respectively.21b Nitric oxide (NO) was generated from DEA/
NONOate (stock solution 1 mM in 0.01 M NaOH).29 Singlet oxygen
(1O2) was generated from ClO− and H2O2.30 A stock solution of
DEA/NONOate was prepared in 0.01 M NaOH solution. Various
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
−
−
−
1
Funding was partially provided by NSFC (20872032,
20972044, 21172063), NCET (08-0175), the Doctoral Fund
of Chinese Ministry of Education (20100161110008), and the
Fundamental Research Funds for the Central Universities,
Hunan University.
analytes represented by H2O2, NO3 , NO2 , HClO, O2 ·, ·OH, O2,
CH3COOOH, NO, Fe3+, GSH, TBHP, vitamin C,·OtBu were added
to the solution of NIR-H2O2 (final concentration, 1 μM) in 0.1 M
PBS, pH 7.4 (containing 1% DMSO). The resulting solution was kept
at ambient temperature for 30 min, and then the fluorescence
intensities were recorded with excitation at 690 nm.
Raw 264.7 Murine Macrophages Culture and Imaging Using
NIR-H2O2. Raw 264.7 murine macrophages were obtained from a
Xiangya hospital and cultured in Dulbecco’s modified eagle medium
(DMEM) supplemented with 10% FBS (fetal bovine serum) in an
atmosphere of 5% CO2 and 95% air at 37 °C. For detection of
endogenously produced H2O2, the living RAW 264.7 macrophages
were coincubated with PMA (1 μg/mL), NIR-H2O2 (1 μM), Hoechst
33258 (4.5 μM), and mito tracker green (1 μM) for 60 min. Prior to
imaging, the cells were washed 3× with 1 mL of PBS, and the
fluorescence images were acquired through an Olympus fluorescence
microscopy equipped with a cooled CCD camera.
Fluorescent Imaging in Living Mice Using NIR-H2O2. ICR
mice, 20−25 g, were given an ip injection of LPS (1 mg in 400 μL
saline). After 4 h, the mice were anesthetized by an ip injection of
xylazine (10 mg/kg) and ketamine (80 mg/kg), and the abdominal fur
was removed with an electric shaver. Then, the mice were
intraperitoneally injected with NIR-H2O2 (400 μM in 400 μL of 1:9
= DMSO:PBS). As a control, untreated (neither treated with LPS nor
NIR-H2O2) or unstimulated mice ip injected only with NIR-H2O2
(400 μM in 400 μL of 1: 9 = DMSO: PBS) were also prepared. The
mice were then imaged (30 min after the injection of NIR-H2O2) by
using a FMT 2500 LX quantitative tomography in vivo imaging system,
with an excitation filter of 670 nm and an emission filter of 690−740
nm.
Bel 7702 Cells Incubation and Fluorescence Imaging Using
NIR-thiol. Bel 7702 cells were cultured in DMEM supplemented with
10% FBS in an atmosphere of 5% CO2 and 95% air at 37 °C. One day
before imaging, the cells were seeded in 24-well flat-bottomed plates in
an atmosphere of 5% CO2, 95% air at 37 °C for 24 h. The cells were
incubated with the sensor NIR-thiol (5 μM) and Hoechst 33258 (4.5
μM) for 30 min at 37 °C in PBS buffer (containing 1% DMSO) and
then washed with PBS buffer 3×. For the control experiment, the cells
were pretreated with PBS solution containing N-ethylmaleimide (1
mM) for 30 min, and then incubated with the sensor NIR-thiol (5
μM) in PBS (containing 1% DMSO) for 30 min at 37 °C.
Fluorescence imaging was then carried out after washing cells with
PBS. Fluorescence imaging of intracelluar thiols was conducted using a
Nikon eclipase TE300 inverted fluorescence microscope.
Fluorescent Imaging in Living Mice Using NIR-thiol. ICR
mice, 20−25 g, were ip injected with a different amount of NIR-thiol
(0, 20, 40, or 160 nanomoles) in 100 μL of 1:3 = DMSO:PBS.
Subsequently, the mice were anesthetized by ip injection of xylazine
(10 mg/kg) and ketamine (80 mg/kg), and their abdominal fur was
removed with an electric shaver. The mice were then imaged by using
a FMT 2500 LX quantitative tomography in vivo imaging system, with
an excitation filter of 670 nm and an emission filter of 690−740 nm.
REFERENCES
■
(1) (a) Lichtman, J. W.; Conchello, J. A. Nat. Methods 2005, 2, 910−
919. (b) Ghoroghchian, P. P.; Therien, M. J.; Hammer, D. A.
Nanomed. Nanobiotechnol. 2009, 1, 156−167. (c) Valeur, B. Molecular
Fluorescence: Principles and Applications; Wiley-VCH: Weinheim,
Germany, 2001.
(2) For some reviews, see: (a) Loudet., A.; Burgess, K. Chem. Rev.
2007, 107, 4891−4932. (b) Goncalves., M. S. T. Chem. Rev. 2009, 109,
190−212. (c) Qian, X.; Xiao, Y.; Xu, Y.; Guo, X.; Qian, J.; Zhu, W.
Chem. Commun. 2010, 46, 6418−6436.
(3) For some recent examples, see: (a) Qian, Y.; Karpus, J.; Kabil, O.;
Zhang, S. Y.; Zhu, H. L.; Banerjee, R.; Zhao, J.; He, C. Nat. Commun.
2011, 2, 495. (b) Yang, Y.; Lowry, M.; Xu, X.; Escobedo, J. O.; Sibrian-
Vazquez, M.; Wong, L.; Schowalter, C. M.; Jensen, T. J.; Fronczek, F.
R.; Warner, I. M.; Strongin, R. M. Proc. Natl. Acad. Sci. U.S.A. 2008,
105, 8829−8834. (c) Palama, I.; DiMaria, F.; Viola, I.; Fabiano, E.;
Gigli, G.; Bettini, C.; Barbarella, G. J. Am. Chem. Soc. 2011, 133,
17777−17785. (d) Liu, C.; Pan, J.; Li, S.; Zhao, Y.; Wu, L. Y.;
Berkman, C. E.; Whorton, A. R.; Xian, M. Angew. Chem., Int. Ed. 2011,
50, 10327−10329. (e) Peng, X.; Yang, Z.; Wang, J.; Fan, J.; He, Y.;
Song, F.; Wang, B.; Sun, S.; Qu, J.; Qi, J.; Yan, M. J. Am. Chem. Soc.
2011, 133, 6626−6635. (f) Sotgiu, G.; Galeotti, M.; Samor, C.;
Bongini, A.; Mazzanti, A. Chem.Eur. J. 2011, 17, 7947−7952.
(g) Kim, E.; Koh, M.; Ryu, J.; Park, S. B. J. Am. Chem. Soc. 2008, 130,
12206−12207. (h) Peng, T.; Yang, D. Org. Lett. 2010, 12, 496−499.
(i) Leevy, W. M.; Gammon, S. T.; Jiang, H.; Johnson, J. R.; Maxwell,
D. J.; Jackson, E. N.; Marquez, M.; Piwnica-Worms, D.; Smith, B. D. J.
Am. Chem. Soc. 2006, 128, 16476−16477. (j) Li, J.; Yao, S. Q. Org.
Lett. 2009, 11, 405−408. (k) Hapuarachchige, S.; Montano, G.;
Ramesh, C.; Rodriguez, D.; Henson, L. H.; Williams, C. C.;
Kadavakkollu, S.; Johnson, D. L.; Shuster, C. B.; Arterburn, J. B. J.
Am. Chem. Soc. 2011, 133, 6780−6790. (l) Yuan, L.; Lin, W.; Xie, Y.;
Chen, B.; Zhu, S. J. Am. Chem. Soc. 2012, 134, 1305−1315.
(4) Ueno, T.; Nagano, T. Nat. Meth. 2011, 8, 642−645.
(5) For some reviews, see: (a) de Silva, A. P.; Gunaratne, H. Q. N.;
Gunnlaugsson, T.; Huxley, A. J. M.; McCoy, C. P.; Rademacher, J. T.;
Rice, T. E. Chem. Rev. 1997, 97, 1515−1566. (b) Nolan, E. M.;
Lippard, S. J. Chem. Rev. 2008, 108, 3443−3480. (c) Que, E. L.;
Domaille, D. W.; Chang, C. J. Chem. Rev. 2008, 108, 1517−1549.
(d) Quang, D. T.; Kim, J. S. Chem. Rev. 2010, 110, 6280−6301.
(e) Thomas, S. W., III; Joly, G. D.; Swager, T. M. Chem. Rev. 2007,
107, 1339−1386. (f) Han, J.; Burgess, K. Chem. Rev. 2010, 110, 2709−
2728. (g) Kobayashi, H.; Ogawa, M.; Alford, R.; Choyke, P. L.; Urano,
Y. Chem. Rev. 2010, 110, 2620−2640. (h) Chen, X.; Tian, X.; Shin, I.;
Yoon, J. Chem. Soc. Rev. 2011, 40, 4783−4804. (i) Cho, D.-G.; Sessler,
J. L. Chem. Soc. Rev. 2009, 38, 1647−1662. (j) Kim, H. N.; Lee, M. H.;
Kim, H. J.; Kim, J. S.; Yoon, J. Chem. Soc. Rev. 2008, 37, 1465−1472.
(k) Beija, M.; Afonso, C. A. M.; Martinho, J. M. G. Chem. Soc. Rev.
13522
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523

Journal of the American Chemical Society
Article
2009, 38, 2410−2433. (l) Chen, X.; Pradhan, T.; Wang, F.; Kim, J. S.;
Yoon, J. Chem. Rev. 2012, 112, 1910−1956.
(6) For some books, see: (a) Demchenko, A. P. Introduction to
Fluorescence Sensing; Springer: New York, 2008. (b) Wang, B.; Anslyn.,
E. V. Chemosensors: Principles, Strategies, and Applications; Wiley:
Hoboken, NJ, 2011.
(7) For some books and reviews, see: (a) Near-Infrared Applications
in Biotechnology;Raghavachari, R., Ed.; Marcel Dekker: New York,
2001. (b) Weissleder, R. Nat. Biotechnol. 2001, 19, 316−317.
(c) Frangioni., J. V. Curr. Opin. Chem. Biol. 2003, 7, 626−634.
(g) Escobedo, J. O.; Rusin, O.; Lim, S.; Strongin, R. M. Curr. Opin.
Chem. Biol. 2010, 14, 64−70. (h) Hilderbrand, S. A.; Weissleder, R.
Curr. Opin. Chem. Biol. 2010, 14, 71−79.
(15) (a) Champagne, B.; Guillaume, M.; Zutterman, F. Chem. Phys.
Lett. 2006, 425, 105−109. (b) Benzi, C.; Bertolino, C. A.; Miletto, I.;
Ponzio, P.; Barolo, C.; Viscardi, G.; Coluccia, S.; Caput, G. Dyes and
Pigments 2009, 83, 111−120.
(16) (a) Bashan, N.; Kovsan, J.; Kachko, I.; Ovadia., H.; Rudich, A.
Physiol. Rev. 2009, 89, 27−71. (b) Kuznetsov, A. V.; Kehrer, I.; Kozlov,
A. V.; Haller, M.; Redl, H.; Hermann, M.; Grimm, M.; Troppmair, J.
Anal. Bioanal. Chem. 2011, 400, 2383−2390. (c) Pandey, A. N.;
Tripathi, A.; PremKumar, K. V.; Shrivastav, T. G.; Chaube, S. K. J. Cell.
Biochem. 2010, 111, 521−528. (d) Peyrot, F.; Ducrocq, C. J. Pineal Res.
2008, 45, 235−246. (e) Rhee, S. G. Science. 2006, 312, 1882−1883.
(f) Reth, M. Nat. Immun. 2002, 3, 1129−1134. (g) Rhee, S. G. Exp.
Mol. Med. 1999, 31, 53−59. (h) Stone, J. R.; Yang, S. Antioxid. Redox
Signaling 2006, 8, 243−270.
(8) For some examples, see: (a) Fischer, G. M.; Daltrozzo, E.;
Zumbusch, A. Angew. Chem., Int. Ed. 2011, 50, 1406−1409. (b) Fischer,
G. M.; Ehlers, A. P.; Zumbusch, A.; Daltrozzo, E. Angew. Chem., Int.
Ed. 2007, 46, 3750−3753. (c) Qin, W.; Ding, D.; Liu, J.; Yuan, W. Z.;
Hu, Y.; Liu, B.; Tan, B. Z. Adv. Funct. Mater. 2012, 22, 771−779.
(d) Umezawa, K.; Nakamura, Y.; Makino, H.; Citterio, D.; Suzuki, K. J.
Am. Chem. Soc. 2008, 130, 1550−1551. (e) Baumes, J. M.;
Gassensmith, J. J.; Giblin, J.; Lee, J.-J.; White, A. G.; Culligan, W. J.;
Leevy, W. M.; Kuno, M.; Smith, B. D. Nat. Chem 2010, 2, 1025−1030.
(17) (a) Winterbourn, C. C. Nat. Chem. Biol. 2008, 4, 278−286.
(b) Paulsen, C. E.; Carroll, K. S. ACS Chem. Biol 2010, 5, 47−62.
(c) Miller, E. W.; Dickinson, B. C.; Chang, C. J. Proc. Natl. Acad. Sci.
U.S.A. 2010, 107 (36), 15681−15686. (d) D’Autrea
M. B.. Nat. Rev. Mol. Cell. Biol. 2007, 8, 813−824.
(18) Galaris, D.; Skiada, V.; Barbouti, A. Cancer Lett. 2008, 266, 21−
29.
́
ux, B.; Toledano,
(19) Lin, M. T.; Beal, M. F. Nature 2006, 443, 787−795.
(20) For some reviews, see: (a) Miller, E. W.; Chang, C. J. Curr.
Opin. Chem. Biol. 2007, 11, 620−625. (b) Chen, X.; Tian, X.; Shin, I.;
Yoon, J. Chem. Soc. Rev. 2011, 40, 4783−4804. (c) McQuade, L. E.;
Lippard, S. J. Curr. Opin. Chem. Biol. 2010, 14, 43−49. (d) Lippert, A.
R.; Van De Bittner, G. C.; Chang, C. J. Acc. Chem. Res. 2011, 44, 793−
804.
(21) For some recent examples, see: (a) Dickinson, B. C.; Huynh, C.;
Chang, C. J. J. Am. Chem. Soc. 2010, 132, 5906−5915. (b) Srikun, D.;
Miller, E. W.; Domaille, D. W.; Chang, C. J. J. Am. Chem. Soc. 2008,
130, 4596−4597. (c) Abo, M.; Urano, Y.; Hanaoka, K.; Terai, T.;
Komatsu, T.; Nagano, T. J. Am. Chem. Soc. 2011, 133, 10629−10637.
(d) Du, L.; Ni, N.; Li, M.; Wang, B. Tetrahedron Lett. 2010, 51, 1152−
1154. (e) Xu, K. H.; Liu, F.; Wang, H. X.; Wang, S. S.; Wang, L. L.;
Tang, B. Sci China Ser B-Chem 2009, 52, 734−740. (f) Karton-Lifshin,
N.; Segal, E.; Omer, L.; Portnoy, M.; Satchi-Fainaro, R.; Shabat, D. J.
Am. Chem. Soc. 2011, 133, 10960−10965. (g) Chung, C.; Srikun, D.;
Lim, C. S.; Chang, C. J.; Cho, B. R. Chem. Commun 2011, 47, 9618−
9620.
(f) Burckstummer, H.; Kronenberg, N. M.; Meerholz, K.; Wurthner, F.
̈
̈
̈
Org. Lett. 2010, 12, 3666−3669. (g) Bouit, P.; Piazza, E. D.; Rigaut, S.;
Guennic, B. L.; Aronica, C.; Toupet, L.; Andraud, C.; Maury, O. Org.
Lett. 2008, 10, 4159−4162. (h) Toutchkine, A.; Nguyen, D.-V.; Hahn,
K. M. Bioconjugate Chem. 2007, 18, 1344−1348. (i) McDonnell, S. O.;
O’Shea, D. F. Org. Lett. 2006, 8, 3493−3496. (j) Kundu, K.; Knight, S.
F.; Willett, N.; Lee, S.; Taylor, W. R.; Murthy, N. Angew. Chem., Int.
Ed. 2009, 48, 299−303. (k) Descalzo, A. B.; Xu, H.-J.; Xue, Z.-L.;
Hoffmann, K.; Shen, Z.; Weller, M. G.; You, X.-Z.; Rurack, K. Org.
Lett. 2008, 10, 1581−1584. (l) Bai, M.; Achilefu, S. Bioorg. Med. Chem.
Lett. 2011, 21, 280−284. (m) Zhang, Z.; Berezin, M. Y.; Kao, J. L. F.;
d’Avignon, A.; Bai, M.; Achilefu, S. A. Angew. Chem., Int. Ed. 2008, 47,
3584−3587. (n) Yuan, L.; Lin, W.; Yang, Y.; Chen, H. J. Am. Chem.
Soc. 2012, 134, 1200−1211. (o) Olivier, J.-H.; Widmaier, J.; Ziessel, R.
Chem.Eur. J. 2011, 17, 11709−11714.
(9) For some examples, see: (a) Sasaki, E.; Kojima, H.; Nishimatsu,
H.; Urano, Y.; Kikuchi, K.; Hirata, Y.; Nagano, T. J. Am. Chem. Soc.
2005, 127, 3684−3685. (b) Kiyose, K.; Aizawa, S.; Sasaki, E.; Kojima,
H.; Hanaoka, K.; Terai, T.; Urano, Y.; Nagano, T. Chem.Eur. J.
2009, 15, 9191−9200.
(22) Li, H.; Li, Q.; Wang, X.; Xu, K.; Chen, Z.; Gong, X.; Liu, X.;
Tong, L.; Tang, B. Anal. Chem. 2009, 81, 2193−2198.
(23) Lee, D.; Khaja, S.; Velasquez-Castano, J. C.; Dasari, M.; Sun, C.;
Petros, J.; Taylor, W. R.; Murthy, N. Nat. Mater. 2007, 6, 765−769.
(24) Beiser, A.; Selhub, J.; Jacques, P. F.; Rosenberg, I. H.;
D’Agostino, R. B.; Wilson, P. W. F..; Wolf, P. A. N. Engl. J. Med.
2002, 346, 476−483.
(10) For some reviews and examples, see: (a) Reymond, J. L.; Fluxa,
̀
V. S.; Maillard, N. Chem. Commun. 2009, 34−46. (b) Song, F.; Garner,
A. L.; Koide, K. J. Am. Chem. Soc. 2007, 129, 12354−12355.
(c) Kamiya, M.; Kobayashi, H.; Hama, Y.; Koyama, Y.; Bernardo, M.;
Nagano, T.; Choyke, P. L.; Urano, Y. J. Am. Chem. Soc. 2007, 129,
3918−3929. (d) Lavis, L, D.; Raines, R. T. ACS Chem. Biol. 2008, 3,
142−155. (e) Huang, S.-T.; Teng, C.-J.; Lee, Y.-H.; Wu, J.-Y.; Wang,
K.-L.; Lin, C.-M. Anal. Chem. 2010, 82, 7329−7334. (f) Lavis, L. D.;
Chao, T.-Y.; Raines, R. T. Chem. Sci. 2011, 2, 521−530. (g) Taki, M.;
Iyoshi, S.; Ojida, A.; Hamachi, I.; Yamamoto, Y. J. Am. Chem. Soc.
2010, 132, 5938−5939. (h) Chen, X.; Sun, M.; Ma, H. Curr. Org.
Chem. 2006, 10, 477−489. (i) Razgulin, A.; Ma, N.; Rao, J. Chem. Soc.
Rev. 2011, 40, 4186−4216.
(25) Ball, R. O.; Courtney-Martin, G.; Pencharz, P. B. J. Nutr. 2006,
136, 1682S−1693S.
(26) For a review and some examples, see: (a) Chen, X.; Zhou, Y.;
Peng, X.; Yoon, J. Chem. Soc. Rev. 2010, 39, 2120−2135. (b) Sreejith,
S.; Divya, K. P.; Ajayaghosh, A. Angew. Chem., Int. Ed. 2008, 47, 7883.
(c) Zhang, M.; Yu, M.; Li, F.; Zhu, M.; Li, M.; Gao, Y.; Li, L.; Liu, Z.;
Zhang, J.; Zhang, D.; Yi, T.; Huang, C. J. Am. Chem. Soc. 2007, 129,
10322. (d) Jung, H. S.; Han, J. H.; Pradhan, T.; Kim, S.; Lee, S. W.;
Sessler, J. L.; Kim, T. W.; Kang, C.; Kim, J. S. Biomaterials 2012, 33,
945. (e) Chen, Y.; Zhao, J.; Guo, H.; Xie, L. J. Org. Chem. 2012, 77,
2192−2206.
(11) Wolinska, E.; Henary, M.; Paliakov, E.; Strekowski, L. J.
Heterocycl. Chem. 2009, 46, 925−930.
(12) The pKa was calculated according to the Henderson−
Hasselbach-type mass action equation (log[(Fmax − F)/(F − Fmin)]
= pKa − pH).
(27) Maeda, H.; Matsuno, H.; Ushida, M.; Katayama, K.; Saeki, K
Angew. Chem., Int. Ed. 2005, 44, 2922−2925.
(28) Mosmann, T. J. Immunol. Meth. 1983, 65, 55−63.
(29) Zhang, X.; Kim, W.-S.; Hatcher, N.; Potgieter, K.; Moroz, L. L.;
Gillette, R.; Sweedler, J. V. J. Biol. Chem. 2002, 277, 48472−48478.
(30) Tang, B.; Zhang, N.; Chen, Z.; Xu, K.; Zhuo, L.; An, L.; Yang, G.
Chem.Eur. J. 2008, 14, 522−528.
(13) (a) Bouit, P.-A.; Aronica, C.; Toupet, L.; Guennic, B. L.;
Andraud, C.; Maury, O. J. Am. Chem. Soc. 2010, 132, 4328−4335.
(b) Wurthner, F.; Archetti, G.; Schmidt, R.; Kuball, H.-G. Angew.
̈
Chem., Int. Ed. 2008, 47, 4529−4532.
(14) (a) Valeur, B. Molecular Fluorescence: Principles and Applications,
Wiley-VCH: Weinheim, Germany, 2001. (b) Magde, D.; Rojas, G. E.;
Seybold, P. Photochem. Photobiol. 1999, 70, 737−744. (c) Oushiki, D.;
Kojima, H.; Terai, T.; Arita, M.; Hanaoka, K.; Urano, Y.; Nagano, T. J.
Am. Chem. Soc. 2010, 132, 2795−2801.
13523
dx.doi.org/10.1021/ja305802v | J. Am. Chem. Soc. 2012, 134, 13510−13523