Bioimaging of nitric oxide with fluorescent indicators based on the rhodamine chromophore

Article abstract of DOI:10.1021/ac001136i

Diaminofluoresceins are widely used for detection and imaging of nitric oxide (NO), but for biological applications, they have the disadvantages that the fluorescence of the fluorescein chromophore is pH-sensitive and overlaps the autofluorescence of cells. We have developed a membrane-permeable fluorescent indicator for NO based on the rhodamine chromophore, DAR-4M AM, which can be excited with 550-nm light. The fluorescence quantum yield of the product after reaction with NO is 840 times higher than that of DAR-4M. The detection limit of NO was 7 nM, and the fluorescence showed no pH dependency above pH 4. DAR-4M AM was successfully applied to practical bioimaging of NO produced in bovine aortic endothelial cells.

Full text of DOI:10.1021/ac001136i

Anal. Chem. 2001, 73, 1967-1973
Bioimaging of Nitric Oxide with Fluorescent
Indicators Based on the Rhodamine Chromophore
Hirotatsu Kojima,† Miki Hirotani,† Naoki Nakatsubo,† Kazuya Kikuchi,† Yasuteru Urano,†
Tsunehiko Higuchi,† Yasunobu Hirata,‡ and Tetsuo Nagano*,†
Graduate School of Pharmaceutical Sciences, and Second Department of Internal Medicine, Faculty of Medicine,
the University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
indicators include highly sensitive detection with fine temporal
and spatial resolution under microscopic observation. In bioim-
aging studies, longer wavelength excitation is desirable owing to
the fluorescence of various cell components (autofluorescence)
and the cytotoxicity caused by UV excitation. That was why we
chose fluorescein as the fluorophore of our diamine indicators.
Although the practical usefulness of DAFs for functional studies
of NO was confirmed, DAFs lack some other desirable charac-
teristics of NO indicators. To obtain higher photostability, longer
excitation wavelength, and applicability over a wider pH range,
we have examined the suitability of rhodamine, which is also
widely used in biological studies, as the fluorophore. It is
considered that the quenching mechanism of DAFs is based on
photoinduced electron transfer from the diaminobenzene moiety
to the xanthene moiety. We anticipated that this mechanism would
also be applicable to the rhodamine fluorophore.
Diaminofluoresceins are widely used for detection and
imaging of nitric oxide (NO), but for biological applica-
tions, they have the disadvantages that the fluorescence
of the fluorescein chromophore is pH-sensitive and over-
laps the autofluorescence of cells. We have developed a
membrane-permeable fluorescent indicator for NO based
on the rhodamine chromophore, DAR-4 M AM, which can
be excited with 5 5 0 -nm light. The fluorescence quantum
yield of the product after reaction with NO is 8 4 0 times
higher than that of DAR-4 M. The detection limit of NO
was 7 nM, and the fluorescence showed no pH depen-
dency above pH 4 . DAR-4 M AM was successfully applied
to practical bioimaging of NO produced in bovine aortic
endothelial cells.
Although nitric oxide (NO) is well established to be a signal
transmitter in vivo,1-5 many of the details remain controversial.
Several methods of detecting NO have been developed in order
to investigate its diverse physiological and pathophysiological roles
in vivo.6-9 We have been developing fluorescent indicators for NO,
such as the diaminofluoresceins (DAFs).10 The vicinal aromatic
diamine on fluorescein is transformed to the corresponding highly
fluorescent triazole by reaction with NO. The actual reaction
species are thought to be NO+ equivalents, such as nitric
anhydride (N2O3), which are generated by rapid autoxidation of
NO. The diacetyl derivatives of these indicators, which can
permeate into cells, where they are hydrolyzed by intracellular
esterases, enabled bioimaging of NO production in living cells
and tissues. The advantages of bioimaging with fluorescent
EXPERIMENTAL SECTION
Synthesis of DARs. 1H NMR spectra were recorded on a
JEOL JNM-LA300 instrument at 300 MHz; δ values are given in
ppm relative to tetramethylsilane. Electron ionization mass spec-
trometry (EI-MS) spectra were measured with a JEOL JMS-DX
300 mass spectrometer. Fast atom bombardment mass spectrom-
etry (FAB-MS) spectra were measured with a JEOL SX-102A mass
spectrometer. The synthetic pathways employed to obtain the new
dyes are outlined in Figure 1.
1 f 3 . A mixture of 1 (2.95 g, 10 mmol)10 and N,N-diethyl-
3-aminophenol (3.33 g, 20 mmol) was stirred at 180 °C for 2 h.
After cooling, the black amorphous solid was refluxed in 2 M HCl
(200 mL) for 2 h. The mixture was neutralized with sodium
hydroxide, and the precipitate was collected and dried. Extraction
with dichloromethane could also be used for collection. The
products were purified by silica gel chromatography with methanol/
dichloromethane (5:95 v/ v) to afford 3 (2.97 g, 59% yield).
3 f DAR-1 . 3 (2.97 g, 5.9 mmol) was reduced with sodium
sulfide and sodium hydrosulfide as previously described,10 to give
DAR-1 (0.51 g, 18%), with recovery of the starting material (2.22
g): mp 145 °C; 1H NMR (dimethyl sulfoxide (DMSO)-d6) δ 1.09
(t, 12H, CH3, J ) 6.8 Hz), 3.33 (m, 8H, CH2), 4.98 (s, 2H, NH2),
5.86 (s, 2H, NH2), 6.06 (d, 1H, ArH, J ) 7.7 Hz), 6.37-6.41 (m,
4H, ArH), 6.55 (d, 2H, ArH, J ) 8.6 Hz), 6.78 (d, 1H, ArH, J ) 7.7
Hz); EI-MS m/ z 472 (M+). Elemental Anal. Calcd for C28H32N4O3‚
0.5H2O: C, 69.83; H, 6.91; N, 11.64. Found: C, 69.85; H, 6.83; N,
11.44.
* Corresponding author: (e-mail) tlong@mol.f.u-tokyo.ac.jp; (fax) 81-3-5841-
4855.
† Graduate School of Pharmaceutical Sciences.
‡ Second Department of Internal Medicine, Faculty of Medicine.
(1) Murad, F. Angew. Chem., Int. Ed. Engl. 1 9 9 9 , 38, 1856-1868.
(2) Furchgott, R. F. Angew. Chem., Int. Ed. Engl. 1 9 9 9 , 38, 1870-1880.
(3) Ignarro, L. J. Angew. Chem., Int. Ed. Engl. 1 9 9 9 , 38, 1882-1892.
(4) Nathan, C.; Xie, Q. Cell 1 9 9 4 , 78, 915-918.
(5) Blottner, D. J. Neurosci. Res. 1 9 9 9 , 58, 139-151.
(6) Akaike, T.; Maeda, H. Methods Enzymol. 1 9 9 6 , 268, 211-221.
(7) Yoshimura, T.; Yokoyama, H.; Fujii, S.; Takayama, F.; Oikawa, K.; Kamada,
H. Nat. Biotechnol. 1 9 9 6 , 14, 992-994.
(8) Barker, S. L. R.; Zhao, Y.; Marletta, M. A.; Kopelman, R. Anal. Chem. 1 9 9 9 ,
71, 2071-2075.
(9) Meineke, P.; Rauen, U.; de Groot, H.; Korth, H.-G.; Sustmann, R. Chem.
Eur. J. 1 9 9 9 , 5, 1738-1747.
(10) Kojima, H.; Nakatsubo, N.; Kikuchi, K.; Kawahara, S.; Kirino, Y.; Nagoshi,
H.; Hirata, Y.; Nagano, T. Anal. Chem. 1 9 9 8 , 70, 2446-2453.
10.1021/ac001136i CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/24/2001
Analytical Chemistry, Vol. 73, No. 9, May 1, 2001 1967

Figure 1. Synthetic scheme of DARs: Ac, acetyl; Et, ethyl; AM, acetoxymethyl; Me, methyl; DIEA, diisopropylethylamine.
1968 Analytical Chemistry, Vol. 73, No. 9, May 1, 2001

DAR-1 f DAR-M. Methyl iodide (100 µL) was added in
several portions to a solution of DAR-1 (231 mg, 0.489 mmol) in
ethanol (10 mL) at 80 °C over 6.5 h. The product was purified by
silica gel chromatography with methanol/ dichloromethane (1:9
v/ v) and by preparative TLC to afford DAR-M (25 mg, 11%), with
recovery of the starting material: mp 150-154 °C; 1H NMR
(CDCl3) δ 1.13 (t, 12H, CH3, J ) 7.0 Hz), 2.86 (s, 3H, NCH3), 3.33
(q, 8H, CH2, J ) 7.0 Hz), 6.37-6.43 (m, 5H, ArH), 6.75 (d, 1H,
ArH, J ) 7.9 Hz), 6.81 (d, 2H, ArH, J ) 9.0 Hz); FAB-MS m/ z 487
((M + H)+).
DAR-1 f DAR-1 EE. Concentrated H2SO4 (0.1 mL) was
added to an ethanol solution of DAR-1 (113 mg, 2.39 mmol), and
the solution was refluxed for 2 h. After cooling, ethanol was
removed in vacuo and the residue was neutralized with sodium
hydroxide to collect the resulting precipitate. After drying, it was
purified by silica gel chromatography with methanol/ dichlo-
romethane (8:92 v/ v) to give DAR-1 EE (26 mg, 20%), with
recovery of the starting material (62 mg): 1H NMR (CDCl3) δ
0.66 (t, 3H, CH3, J ) 7.1 Hz), 1.33 (t, 12H, CH3, J ) 7.1 Hz), 3.61
(q, 8H, NCH2, J ) 7.1 Hz), 4.19 (q, 2H, CH2, J ) 7.1 Hz), 4.55 (s,
2H, NH2), 6.05 (s, 2H, NH2), 6.33 (d, 1H, ArH, J ) 7.9 Hz), 6.74
(d, 2H, ArH, J ) 2.2 Hz), 6.83 (dd, 2H, ArH, J ) 9.7, 2.2 Hz), 6.94
(d, 1H, ArH, J ) 7.8 Hz), 7.45 (d, 1H, ArH, J ) 9.7 Hz); EI-MS
m/ z 501 (M+).
yield): mp 158-163 °C; 1H NMR (DMSO-d6) δ 1.08 (t, 12H, CH3,
J ) 6.8 Hz), 4.95 (br, 2H, NH2), 5.52 (br, 2H, NH2), 6.09 (s, 1H,
ArH), 6.37 (d, 2H, ArH, J ) 2.4 Hz), 6.41 (dd, 2H, ArH, J ) 8.8,
2.4 Hz), 6.51 (d, 2H, ArH, J ) 8.8 Hz), 6.89 (s, 1H, ArH); FAB-MS
m/ z 472 (M+).
1 f 6 . Compound 6 was synthesized from 1 and N,N-
dimethyl-3-aminophenol by a method similar to that used to obtain
3 (yield 66.4%).
6 f DAR-4 . Compound 6 was reduced by a similar method
(solvent: dichloromethane/ ethanol 7:13 v/ v) to that used to obtain
DAR-1 AM, to afford DAR-4 (65.4%): 1H NMR (CDCl3) δ 2.97 (s,
12H, CH3), 5.03 (br, 2H, NH2), 6.34 (d, 1H, ArH, J ) 7.7 Hz), 6.40
(dd, 2H, ArH, J ) 8.8, 2.6 Hz), 6.46 (d, 2H, ArH, J ) 2.6 Hz), 6.75
(d, 2H, ArH, J ) 8.8 Hz), 6.90 (d, 1H, ArH, J ) 7.7 Hz); FAB-MS
m/ z 417 ((M + H)+).
DAR-4 f DAR-4 M. A methyl group was introduced into
DAR-4 (0.55 g, 1.3 mmol) by a method similar to that used to
obtain DAR-M, to afford DAR-4M (0.14 g, 24.5%) with recovery of
DAR-4 (0.25 g): mp 219 °C; 1H NMR (CDCl3) δ 2.92 (s, 3H, CH3),
2.97 (s, 12H, CH3), 4.99 (br, 2H, NH2), 6.40 (dd, 2H, ArH, J ) 8.6,
2.6 Hz), 6.47 (d, 2H, ArH, J ) 2.6 Hz), 6.47 (d, 1H, ArH, J ) 8.0
Hz), 6.75 (d, 2H, ArH, J ) 8.6 Hz), 6.85 (d, 1H, ArH, J ) 8.0 Hz);
FAB-MS, m/ z 431 ((M + H)+).
6 f 7 . An acetoxymethyl group was introduced into 6 (2.44
g, 5.46 mmol) by a method similar to that used to obtain 4 , to
afford 7 (0.445 g, 13.6%) with recovery of 6 (1.70 g).
7 f DAR-4 AM. Compound 7 was reduced by a similar
method (solvent: dichloromethane/ ethanol 1:1 v/ v) to that used
to obtain DAR-1 AM, to afford DAR-4 AM (50.9%):1H NMR (CDCl3)
δ 1.69 (s, 3H, COCH3), 3.33 (s, 12H, CH3), 4.81 (br, 2H, NH2),
5.32 (s, 2H, OCH2O), 6.29 (d, 1H, ArH, J ) 7.9 Hz), 6.47 (br, 2H,
NH2), 6.78 (d, 2H, ArH, J ) 2.6 Hz), 6.91 (dd, 2H, ArH, J ) 9.5,
2.6 Hz), 6.94 (d, 1H, ArH, J ) 7.9 Hz), 7.45 (d, 2H, ArH, J ) 9.5
Hz).
3 f 4 . Acetoxymethyl bromide (AMBr; 100 µL) was added
to a 1,2-dichloroethane solution (5 mL) of 3 (0.49 g, 0.98 mmol)
and diisopropylethylamine (DIEA; 200 µL). The mixture was
stirred in a stoppered vessel at room temperature. After 24 h, 100
µL of AMBr and 200 µL of DIEA were added and stirring was
continued for 18 h. Dichloromethane was added and the solution
was washed with water. After removal of the solvent, the product
was purified by silica gel chromatography with methanol/
dichloromethane (6:94 v/ v) to give 4 (124 mg, 19%), with recovery
of the starting material (359 mg).
4 f DAR-1 AM. Compound 4 (0.38 g, 0.58 mmol) was
reduced with Pd/ C (10% Pd, 98 mg) in EtOH (90 mL) under
hydrogen (1 atm) at room temperature for 2 h. The product was
purified by silica gel chromatography with methanol/ dichlo-
romethane (1:9 v/ v) to give DAR-1 AM (0.29 mg, 80%): 1H NMR
(CDCl3) δ 1.31 (t, 12H, CH3, J ) 7.0 Hz), 1.67 (s, 3H, COCH3),
3.59 (q, 8H, CH2, J ) 7.0 Hz), 5.34 (s, 2H, OCH2O), 6.24 (d, 1H,
ArH, J ) 7.9 Hz), 6.71 (d, 2H, ArH, J ) 2.6 Hz), 6.82 (dd, 2H,
ArH, J ) 9.5, 2.6 Hz), 6.93 (d, 1H, ArH, J ) 7.9 Hz), 7.41 (d, 2H,
ArH, J ) 9.5 Hz); FAB-MS m/ z 545 (M+).
DAR-4 AM f DAR-4 M AM. A methyl group was introduced
into DAR-4 AM (133 mg, 0.234 mmol) by a method similar to that
used to obtain DAR-M, to afford DAR-4M AM (40.5 mg, 27.5%)
with recovery of DAR-4 AM (iodide 70 mg): mp 177-178 °C; 1H
NMR (CDCl3) δ 1.69 (s, 3H, COCH3), 2.99 (d, 3H, NCH3, J ) 4.6
Hz), 3.36 (s, 12H, CH3), 5.06 (q, 1H, NH, J ) 4.6 Hz), 5.33 (s, 2H,
OCH2O), 6.34 (s, 2H, NH2), 6.46 (d, 1H, ArH, J ) 7.9 Hz), 6.69
(d, 1H, ArH, J ) 7.9 Hz), 6.79 (d, 2H, ArH, J ) 2.6 Hz), 6.91 (dd,
2H, ArH, J ) 9.5, 2.6 Hz), 7.48 (d, 2H, ArH, J ) 9.5 Hz); FAB-MS
m/ z 503 (M+).
DAR-1 AM f DAR-M AM. A methyl group was introduced
into DAR-1 AM (300 mg, 0.48 mmol) by a method similar to that
used to obtain DAR-M, to afford DAR-M AM (68 mg, 21%) with
recovery of DAR-1 AM (bromide 87 mg and iodide 67 mg): mp
Triazole Compounds of DARs. NO gas was introduced into
a methanol solution of each DAR (a mixture of methanol and
dichloromethane was used as the solvent only in the case of DAR-
4M) to yield the corresponding triazole compound (DAR-1 T,
DAR-2 T, DAR-M T, DAR-4M T). Each triazole compound was
purified by silica gel chromatography with methanol/ dichlo-
romethane (1:9 v/ v, a small amount of acetic acid (0.02% v/ v) was
added only in the case of purification of DAR-4M T): (DAR-1 T)
1
147-150 °C; H NMR (CDCl3) δ 1.34 (t, 12H, CH3, J ) 7.0 Hz),
1.67 (s, 3H, COCH3), 2.98 (s, 3H, NCH3), 3.63 (q, 8H, CH2, J )
7.0 Hz), 5.36 (s, 2H, OCH2O), 6.42 (d, 1H, ArH, J ) 8.0 Hz), 6.66
(d, 1H, ArH, J ) 8.0 Hz), 6.75 (d, 2H, ArH, J ) 2.4 Hz), 6.84 (dd,
2H, ArH, J ) 9.3, 2.4 Hz), 7.43 (d, 2H, ArH, J ) 9.3 Hz); FAB-MS
m/ z 559 (M+).
1
mp >300 °C; H NMR (DMSO-d6) δ 1.10 (t, 12H, CH3, J ) 6.4
Hz), 3.34 (m, 8H, CH2), 6.50-6.71 (m, 7H, ArH), 8.01 (d, 1H, ArH,
J ) 8.1 Hz); FAB-MS m/ z 484 ((M+H)+). (DAR-2 T) mp >300
2 f 5 . Compound 5 was synthesized from 2 by a method
similar to that used to obtain 3 (yield 7.7%).
1
°C; H NMR (DMSO-d6) δ 1.08 (t, 12H, CH3, J ) 7.0 Hz), 6.38
5 f DAR-2 . Compound 5 was reduced by a method similar
to that used to obtain DAR-1 AM, to afford DAR-2 (quantitative
(dd, 2H, ArH, J ) 8.2, 2.2 Hz), 6.44 (d, 2H, ArH, J ) 2.2 Hz), 6.49
(d, 2H, ArH, J ) 8.6 Hz), 7.63 (s, 1H, ArH), 8.57 (s, 1H, ArH);
Analytical Chemistry, Vol. 73, No. 9, May 1, 2001 1969

EI-MS m/ z 483 (M+). (DAR-M T) mp 155-160 °C; 1H NMR
(CDCl3) δ 1.12 (t, 12H, CH3, J ) 7.1 Hz), 3.32 (q, 8H, CH2, J ) 7.1
Hz), 4.37 (s, 3H, NCH3), 6.31 (dd, 2H, ArH, J ) 9.0, 2.5 Hz), 6.43
(d, 2H, ArH, J ) 2.5 Hz), 6.58 (d, 2H, ArH, J ) 9.0 Hz), 7.26 (d,
1H, ArH, J ) 8.6 Hz), 7.83 (d, 1H, ArH, J ) 8.6 Hz); FAB-MS m/ z
498 ((M + H)+). (DAR-4M T) mp 242-245 °C; 1H NMR (CDCl3)
δ 2.98 (s, 12H, CH3), 4.41 (s, 3H, CH3), 6.36 (dd, 2H, ArH, J )
9.0, 2.6 Hz), 6.50 (d, 2H, ArH, J ) 2.6 Hz), 6.62 (d, 2H, ArH, J )
9.0 Hz), 7.24 (d, 1H, ArH, J ) 8.5 Hz), 7.77 (d, 1H, ArH, J ) 8.5
Hz); FAB-MS m/ z 442 ((M + H)+).
Fluorometric Analysis. A fluorescence spectrophotometer
(F4500, Hitachi, Tokyo, Japan) was used. The slit width was 2.5
nm for both excitation and emission. The photomultiplier voltage
was 950 V. DARs were dissolved in DMSO (for fluorometric
analysis; Dojin, Kumamoto, Japan) to obtain 10 mM stock
solutions. Relative quantum efficiencies of fluorescence of DARs
and their triazole forms were obtained by comparing the area
under the corrected emission spectrum of the test sample at 535-
nm excitation with that of a solution of rhodamine B in ethanol,
which has a quantum efficiency of 0.97 according to the litera-
ture.11 NO solution was prepared as described before.10
HP LC Analysis. The HPLC unit consisted of a pump (PU980,
Jasco, Tokyo, Japan), a multiwavelength detector (MD910, Jasco),
and a column (Inertsil ODS-2, 4.0 × 150 mm, GL Sciences Inc.,
Tokyo, Japan). Gradient elution was employed. (DAR-1 EE:
sodium phosphate buffer (pH 7.4, 10 mM)/ acetonitrile 25:75 v/ v
f 0:100 for 15 min, DAR-1 AM: 0.1% H3PO4/ MeOH 60:40 v/ v f
40:60 for 10 min). The flow rate was 1.0 mL/ min. Absorbance
was monitored in the range of 500-600 nm. Aliquots of 10 µL
were injected for analysis.
Hydrolysis by Rat Brain Homogenate. Male Sprague-
Dawley rats (8 weeks old) were decapitated with a guillotine, and
the whole brains were removed. A 5-fold excess (by weight) of
tris-HCl buffer (0.1 M, pH 7.6) containing 0.1% Triton X-100 was
added, and homogenization was carried out. The supernatant after
centrifugation of the homogenate at 40000g for 10 min was used
for hydrolysis of the ester derivatives of dyes (10 µM). The
incubation was carried out at 37 °C.
50 (Hamamatsu Photonics, Shizuoka, Japan), which is an imaging
system including a cooled charge-coupled device (CCD) camera.
RESULTS AND DISCUSSION
We designed DAR-1 and DAR-2, based on the highly fluores-
cent Rhodamine B fluorophore, and synthesized them from
phthalic anhydride derivatives (1, 2) and N,N-diethyl-3-aminophe-
nol (Figure 1). The same alkaline reduction system of a nitro
group with sulfide and hydrosulfide ions that we had employed
to synthesize DAFs was used.10 However, the yield was low due
to the poor solubility of 3 in water. We found that catalytic
hydrogenation with Pd in an organic solvent afforded higher yields
of the desired diamines.
It was confirmed that the reaction of DARs and NO affords
the corresponding fluorescent triazole, as in the case of DAFs.
To image NO in living cells, DARs should be membrane-
permeable. We therefore introduced an ethyl group into the
carboxyl group in DAR-1, anticipating that the ethyl ester, DAR-1
EE, would be hydrolyzed by cytosolic esterases after permeation
through the cell membrane. Then, we applied this dye to imaging
of NO produced by activated rat aortic smooth muscle cells, using
a procedure similar to that reported in the previous paper.10 There
was almost no autofluorescence of the cells due to the longer
wavelength excitation, as expected. Although the NO imaging with
DAR-1 EE was successful, we encountered a difficulty in loading.
It took a longer time to obtain a stable fluorescence intensity in
the cells than in the case of DAF derivatives. This presumably
occurred because the ethyl ester, which was attached to a bulky
acid, was not readily hydrolyzed. DAR-1 EE remaining in the
cytosol might be distributed to the fatty membrane in cells, and
fluorescent rhodamines generally show higher fluorescence
intensity in organic solvents, such as ethanol, than in water. We
therefore examined the hydrolysis of DAR-1 EE with rat brain
homogenate as an esterase preparation, by means of HPLC.
During a 30-min incubation, DAR-1 EE was not hydrolyzed at all.
Therefore, most of the loaded DAR-1 EE was probably localized
on membranes in the smooth muscle cells, in unhydrolyzed form.
Next, we synthesized and examined the acetoxymethyl ester
of DAR-1 (DAR-1 AM; Figure 1), because acetoxymethyl esters
were reported to be easily hydrolyzed by intracellular esterases.12
The esterification was carried out before reduction of the nitro
group, because diamine is reactive to acetoxymethyl bromide. We
found that 27% of DAR-1 AM was hydrolyzed in 30 min by rat
brain homogenate. So, we decided to utilize this ester.
Imaging P rocedures. Primary cultured endothelial cells from
bovine aorta were passaged in Dulbecco’s modified Eagle’s
medium (DMEM) containing 10% fetal bovine serum and antibiot-
ics. Cells between the 14th and 16th passages on glass-bottomed
culture dishes (P35G-0-1-C; MatTek Corp., Ashland, MA) were
used for these experiments. The cells were incubated for 30 min
at 37 °C in DMEM containing 10 µM DAR-4M AM (serum-free,
final 0.2% DMSO) for loading, and postincubation with DMEM
was carried out for 1 h. Then, the medium was switched to
phosphate-buffered saline (PBS)(+). The cells were mounted on
an inverted fluorescence microscope (IX70: Olympus, Tokyo,
Japan) equipped with an objective lens (×20), an excitation filter
(520-550 nm), a dichroic mirror (565 nm), and a long-pass
emission filter (580 nm). The air temperature was maintained at
37 °C with a warming box (IX-IBM; Olympus). Bradykinin and
We expected that the fluorescence of DAR-1 T would be
independent of physiological pH change, because the rhodamine
fluorophore of DAR-1 T does not have phenolic hydroxyl groups.
However, the triazole proton of DAR-1 T affected the fluorescence
due to its pKa value, 6.69.13 Therefore, we introduced a methyl
group into DAR-1.13 The fluorescence intensity of the triazole form
of DAR-M (DAR-M T; Figure 1) was stable above pH 4, while
that of DAR-1 T or DAR-2 T was unstable around pH 7 (Figure
2).
For bioimaging applications, an acetoxymethyl ester was
introduced into DAR-M. DAR-M AM (Figure 1) was applied to
NO imaging in bovine aortic endothelial cells. However, the
L-NAME were purchased from Sigma (St. Louis, MO). Bradykinin
(1 mM, 1 µL) was carefully added to 1 mL of PBS(+) over the
cells. Optical signals were recorded at 10-s intervals with an Argus
(12) Tsien, R. Y. Nature 1 9 8 1 , 290, 527-528.
(11) Nishikawa, Y.; Hiraki, K. Analytical Methods of Fluorescence and Phospho-
rescence; Kyoritsu Publishing Co.: Tokyo, 1984; pp 76-80.
(13) Kojima, H.; Hirotani, M.; Urano, Y.; Kikuchi, K.; Higuchi, T.; Nagano, T.
Tetrahedron Lett. 2 0 0 0 , 41, 69-72.
1970 Analytical Chemistry, Vol. 73, No. 9, May 1, 2001

Table 1. Absorbance and Fluorescence Properties of Dyesa
extinction coeff and absorption max
(× 104 M-1 cm-1, nm)
rel quantum efficiencies
fluoresc max of
triazole form (nm)
dyeb
DAN
DAN-1
DAF-2
DAF-5
DAF-FM
DAR-1
diamine
triazole
diamine
triazole
0.5, 336
0.6, 340
7.9, 486
9.8, 499
8.4, 487
12, 550
10, 549
9.8, 550
7.8, 543
0.6, 357
0.6, 360
7.3, 491
7.9, 503
7.3, 495
8.7, 556
7.1, 552
7.5, 558
7.6, 554
433
447
513
523
515
575
571
574
572
0.51
0.83
0.63
0.92
0.70
0.81
0.25
0.34
0.29
0.42
0.002
0.005
0.007
0.005
0.004
0.006
0.007
0.0005
DAR-2
DAR-M
DAR-4M
a
All data were obtained at 20 °C in 0.1 M sodium phosphate buffer, pH 7.4. b DAN, 2,3-diaminonaphthalene; DAN-1, 4-[[(3-amino-2-
naphthalenyl)amino]methyl]-benzoic acid.15
Figure 2. pH profiles of the indicators after reaction with NO.
Triazole forms of DAR-1, DAR-2, DAR-M, DAF-2, DAF-5, and DAF-
FM (final 1 µM) were added to sodium phosphate solution adjusted
to various pH values. The pH was measured after mixing. The
fluorescence intensities of DAR-1 T (closed triangle), DAR-2 T (closed
square), DAR-M T (closed circle), DAF-2 T (open triangle), DAF-5 T
(open square), and DAF-FM T (open circle) were determined at 580,
575, 575, 515, 520, and 515 nm with excitation at 565, 565, 560,
495, 505, and 495 nm, respectively. The photomultiplier voltage was
950 (DARs) or 400 V (DAFs). The curves were fitted to the following
equations: (DAR-1 T) intensity ) 783/(1 + 102.13-pH) + 42.7/(1 +
106.69-pH), R ) 0.956; (DAR-2 T) intensity ) 899/(1 + 101.62-pH) -
218/(1 + 107.27-pH), R ) 0.987; (DAR-M T) intensity ) 1000/(1 +
102.39-pH), R ) 0.951; (DAF-2 T) intensity ) 460/(1 + 106.27-pH) -
55.6/(1 + 107.94-pH), R ) 1.00; (DAF-5 T) intensity ) 505/(1 +
104.59-pH) - 119/(1 + 107.41-pH), R ) 1.00; (DAF-FM T) intensity )
518/(1 + 104.38-pH), R ) 0.994.
Figure 3. Excitation (A) and emission (B) spectra for DAR-4M at
37 °C in sodium phosphate buffer (0.1 M, pH 7.4) with NO values
ranging from 0 to 1.22 µM. NO solution was added to 10 µM DAR-
4M solution under aerobic conditions. The spectra were obtained from
an average of five accumulations after addition of NO solution.
difficulty in loading was hardly improved, compared to DAR-1 EE.
We thought that this might be because DAR-M AM, which has
four ethyl groups, is too hydrophobic to be stably distributed to
the cytosol. Therefore, we changed the fluorophore to tetrameth-
ylrhodamine and examined the suitability of DAR-4M AM for
bioimaging (Figure 1). The fluorescence spectra of DAR-4M are
shown in Figure 3 and the properties of the dyes that have been
developed are summarized in Table 1. It was found that the
quantum efficiency of DAR-4M T is the highest of all DARs
examined. Judging from NO standard curves calibrated with the
indicators, the sensitivity of DAR-4M to NO is twice that of DAR-
1. The detection limit of NO by DAR-4M was 7 nM. DAR-4M
should offer a good signal-to-noise ratio in the examination of
biological samples because of the low background fluorescence
with the longer wavelength excitation, even though the quantum
yield of DAR-4M T is lower than that of fluorescein dyes.
Moreover, as we mentioned above, DAR-4M permitted NO
detection above pH 4, while DAF-FM, a fluorescein dye, can only
be applied above pH 5.8.14
We applied DAR-4M AM to the imaging of NO in bovine aortic
endothelial cells, for which a high sensitivity to NO is required.
The loading was easier than with any other DAR examined. The
cells were incubated in DMEM containing 10 µM DAR-4M AM
(14) Kojima, H.; Urano, Y.; Kikuchi, K.; Higuchi, T.; Hirata, Y.; Nagano, T. Angew.
Chem., Int. Ed. Engl. 1 9 9 9 , 38, 3209-3212.
(15) Kojima, H.; Sakurai, K.; Kikuchi, K.; Kawahara, S.; Kirino, Y.; Nagoshi, H.;
Hirata, Y.; Akaike, T.; Maeda, H.; Nagano, T. Biol. Pharm. Bull. 1 9 9 7 , 20,
1229-1232.
Analytical Chemistry, Vol. 73, No. 9, May 1, 2001 1971

Figure 4. Bright-field and fluorescence images of cultured bovine aortic endothelial cells loaded with DAR-4M AM. The upper images are the
bright-field, the fluorescence, and the fluorescence ratio images at the start of measurement, respectively. The middle and the lower images are
fluorescence ratio images at the indicated times after the start of measurement. The fluorescence ratio images indicate the ratio of the intensity
to the initial intensity at the start. However, areas where the fluorescence intensity was low were excluded and are shown in white. These
images correspond to the fluorescence intensity data in Figure 5A.
at 37 °C for 30 min. After this incubation, they were postincubated
with DMEM without the dye for 1 h to allow stabilization of the
fluorescence intensity due to hydrolysis and distribution of the
dye.
Our diamine indicators, before trapping NO, have weak
fluorescence. When using DAFs, it is sometimes difficult to tell
whether the cells are loaded with the dye, because autofluores-
cence of the cells overlaps the fluorescence of DAFs. On the other
hand, we could easily distinguish the cells loaded with DAR-4M,
because the unloaded cells had almost no fluorescence. Moreover,
DARs are more photostable than DAFs, as we mentioned previ-
ously.13
Figure 5. Fluorescence response of DAR-4M AM-loaded cultured
The fluorescence image shows that the dye was successfully
loaded into cells and was distributed to the cytosol, although it
was hardly distributed to the nucleus, as shown in “Fluo” in Figure
4. The fluorescence changes are shown in pseudocolored pictures
(Figure 4) and graphs (Figure 5). The fluorescence intensity of
the cells is shown as the ratio of the intensity to the initial intensity
bovine aortic endothelial cells, reflecting NO production in response
to the addition of 1 µM bradykinin at 150 s. (A) The data corresponding
to the images in Figure 4. (B) Control data in the presence of a NOS
inhibitor, L-NAME (1 mM). The medium was switched to medium
containing L-NAME just before measurement. The line shows mean
values of the average fluorescence intensity in seven areas containing
a cell.
1972 Analytical Chemistry, Vol. 73, No. 9, May 1, 2001

cence intensity of all cells was increased, but it was observed that
the NO productivity varied from cell to cell (Figure 4). This fine
spatial resolution is one of the major advantages of our method.
Thus, we succeeded in highly sensitive imaging of NO with a
Rhodamine-type indicator. The characteristics of DAF and DAR
are compared in Table 2. DAR-4M AM should be useful for
bioimaging in samples that have strong autofluorescence in the
case of 490-nm excitation or whose intracellular pH may fall below
6.
Table 2. Comparison of DAF- and DAR-Type Indicators
DAF
DAR
suitable excitation range
applicable range of
intracellular pH
blue (∼490 nm)
green (∼550 nm)
above pH 6
above pH 4
detection limit of NO
photostability of the
corresponding triazole
3 nM (DAF-FM)
severely
photobleached
(more than 72%)
equal
7 nM (DAR-4M)
not photobleached
a
distribution to nucleus
and cytosol
cytosol > nucleus
a
Determined by measuring the fluorescence intensity after
exposure to sunlight for 3 h.13,14
ACKNOWLEDGMENT
This work was supported by Grants-in-Aid for Scientific
Research on Priority Areas, 09304061, 10169215, and 10557243,
from the Ministry of Education, Science, Sport, and Culture of
Japan.
before stimulation for ease of comparison. A fluorescence increase
in cells was observed after stimulation with 1 µM bradykinin,
which raises the intracellular Ca2+ level and thereby activates
NOS, and this increase was suppressed by addition of a NOS
Received for review September 22, 2000. Accepted
February 18, 2001.
inhibitor (1 mM
L-NAME; Figure 5). Therefore, this fluorescence
increase was caused by NO production in the cells. The fluores-
AC001136I
Analytical Chemistry, Vol. 73, No. 9, May 1, 2001 1973