Fluorophotometric determination of hydrogen peroxide and other reactive oxygen species with fluorescein hydrazide (FH) and its crystal structure

Article abstract of DOI:10.1248/cpb.56.977

Methods for the fluorophotometric determination of hydrogen peroxide (H₂O₂) and other reactive oxygen species (ROS) were proposed by using the fluorescence reaction between H₂O₂ or other ROS and fluorescein hydrazide (FH). In the determination of H ₂O₂, the calibration curve exhibited linearity over the H₂O₂ concentration range of 2.1-460 ng ml^-1 at an emission wavelength of 527 nm with an excitation of 460 nm and with the relative standard deviations (n=6) of 4.06%, 1.78%, and 2.21% for 3.1 ng ml ^-1, 30.8 ng ml^-1, and for 308 ng ml^-1 of H ₂O₂, respectively. The detection limit for H ₂O₂ was 0.7 ng ml^-1 due to three blank determinations (p=3). The calibration curves for ROS-related compounds were also constructed under the optimum conditions. This method was successfully applied in the assay of H₂O₂ in human urine. In addition, we performed the characterization of FH, and interesting information was obtained with regard to the relationship between the chemical structure and fluorescence.

Full text of DOI:10.1248/cpb.56.977

July 2008
Chem. Pharm. Bull. 56(7) 977—981 (2008)
977
Fluorophotometric Determination of Hydrogen Peroxide and Other
Reactive Oxygen Species with Fluorescein Hydrazide (FH) and Its
Crystal Structure
Ryosuke NAKAHARA,* Tsuyoshi FUJIMOTO, Mitsunobu DOI, Kanako MORITA, Takako YAMAGUCHI, and
Yoshikazu FUJITA
Osaka University of Pharmaceutical Sciences; 4–20–1 Nasahara, Takatsuki, Osaka 569–1094, Japan.
Received March 18, 2008; accepted April 11, 2008; published online April 17, 2008
Methods for the fluorophotometric determination of hydrogen peroxide (H2O2) and other reactive oxygen
species (ROS) were proposed by using the fluorescence reaction between H2O2 or other ROS and fluorescein hy-
drazide (FH). In the determination of H2O2, the calibration curve exhibited linearity over the H2O2 concentration
range of 2.1—460 ng mlꢀ1 at an emission wavelength of 527 nm with an excitation of 460 nm and with the relative
standard deviations (nꢁ6) of 4.06%, 1.78%, and 2.21% for 3.1 ng mlꢀ1, 30.8 ng mlꢀ1, and for 308 ng mlꢀ1 of
H2O2, respectively. The detection limit for H2O2 was 0.7 ng mlꢀ1 due to three blank determinations (rꢁ3). The
calibration curves for ROS-related compounds were also constructed under the optimum conditions. This
method was successfully applied in the assay of H2O2 in human urine. In addition, we performed the characteri-
zation of FH, and interesting information was obtained with regard to the relationship between the chemical
structure and fluorescence.
Key words hydrogen peroxide; reactive oxygen species; fluorescein hydrazide; human urine; characterization
Reactive oxygen species (ROS) such as hydrogen peroxide tor, became a colorless, non-fluorescent substance.
(H2O2), superoxide (O2ꢀ), hydroxyl radical (·OH), nitric
Experimental
oxide (NO), peroxynitrite (ONOOꢀ), and peroxide (R–OOH)
Synthesis of Fluorescein Hydrazide (FH) A modification of the proce-
dure of Baeyer and Akita.22,23) for the synthesis of fluoescein and FH was
employed. Resorcinol 2 mol and phthalic anhydride 1 mol were combined
with methanesulfonic acid. The mixture was heated and dissolved in 5%
sodium hydroxide solution. The solution was poured into 30% acetic acid;
the resulting fluorescein, a yellow precipitate, was collected with a yield of
92.5%. Subsequently, fluorescein was added to a solution of excess 98% an-
hydrous hydrazine and heated at 80 °C in a warm water bath for 3 h. Upon
cooling to room temperature and with the addition of strong hydrochloric
acid into the solution, the light yellow substance was deposited. Then, the
substance obtained was dissolved in ethanol and treated with active carbon.
After removal of solvent, the residual was further purified by recrystalliza-
tion in ethanol, acetic acid and acetonitrile, providing pure FH as a white
board-shaped crystal with a yield of 89%. The structure of FH was verified
by MS, NMR, and X-ray crystallography. MS (EIꢁ): m/z 346 (Mꢁ). 1H-
NMR (500 MHz, DMSO) d: 4.37 (2H, s), d: 6.39 (2H, d, Jꢂ8.5 Hz), d: 6.45
(2H, dd, Jꢂ8.5, 2.5 Hz), d: 6.59 (2H, d, Jꢂ2.2 Hz), d: 6.99 (1H, m), d: 7.49
(2H, m), d: 7.77 (1H, m), d: 9.80 (2H, s).
affect the living body and cause various types of diseases
such as cancer,1) cardiovascular disorders,2) and neurodegen-
erative diseases.3,4) On the other hand, it is accepted that
some ROS control disinfection in cells and work as a biofac-
tor for signal transmission of insulin.5) In addition, H2O2 is a
major by-product of ROS in living organisms and a common
marker for oxidative stress in the field of clinical inspec-
tion.6—8) The development of methods for the determination
of these substances is significant in the fields of clinical and
biological studies. Several methods for the determination
of the spectrophotometry,9—11) fluorophotometry,12—18) and
19,20)
chemiluminescence
of H2O2 and other ROS have al-
ready been reported. These methods focus on the determina-
tion of individual ROS. However, it appears that most of
ROS have only minute life times under biological environ-
ments. Therefore, we considered that it is more practicable to
measure the total rather than individual ROS for a clinical
purpose. Hence, we developed a handy and highly sensitivity
method for the determination of ROS, including H2O2. We
have already developed methods for the fluorometric deter-
mination of cobalt(II) and H2O2 with fluorescein hydrazide
(FH),21) however, the method adopted for the determination
of H2O2 exhibited remarkably poor reproducibility. In this
study, method to obtain a more reproducible and sensitive
spectrofluorimetric determination of H2O2 and other ROS
were examined. Furthermore, the applicability of the method
for measuring H2O2 in urine samples was examined. More-
over, because an excellent crystal of FH was obtained, struc-
tural analysis by X-ray diffractometry was performed. As a
result, this FH crystal was confirmed to have a five-mem-
bered spirolactam structure. The relationship between fluo-
rescence and the chemical structure was studied; the fluores-
cein, which was the strong fluorescent substance that caused
Apparatus The fluorescence measurements were performed by using a
Hitachi model F-2500 spectrofluorophotometer equipped with an Usio
150 W Xenon lamp and 10ꢃ10 mm quartz cells. A Horiba model F-22 pH
meter equipped with a glass combined electrode was used for all the pH
measurements.
Reagent and Solutions An FH solution (1.0ꢃ10ꢀ4 M) was prepared by
dissolving FH (highly purified by recrystallization) in ethanol. A cobalt(II)
solution (1.0ꢃ10ꢀ4 M) was prepared by dissolving cobalt(II) chloride, 6-hy-
drate (Kishida Chemical Co., Ltd.). A 0.2 M Na2CO3/NaHCO3 buffer (pH
9.6) was used for pH adjustments. A 1.0% dodecyltrimethylammonium
chloride (DTAC) solution was prepared as a cationic surfactant by dissolv-
ing DTAC (Kishida Chemical Co., Ltd.) in water. A stock solution of
1.0ꢃ10ꢀ2 M H2O2 was prepared by the dilution of 30% H2O2 solution
(Kishida Chemical Co., Ltd.) and corrected by permanganometry, and the
working solution was prepared by the suitable dilution of this stock solution,
as required. Superoxide (O2ꢀ) was added as solid KO2. Hydroxyl radicals
(·OH) were generated by the reaction of Fe2ꢁ with H2O2. Nitric oxide (NO)
was delivered using 1-hydroxy-2-oxo-3-(N-methyl-3-aminoethyl)-3-methyl-
1-triazene (NOC7) (Dojindo Chemical Co.). A standard solution of NOC7
was freshly prepared in 0.1 M NaOH solution. Peroxynitrite (ONOOꢀ) was
prepared by using peroxynitrite solution (Dojindo Chemical Co.). tert-Butyl
hydroperoxide (TBHP) was delivered from 70% aqueous solution.
Standard Procedure for the Determination of H2O2 De-ionized water
the spirolactam formation by acting as the hydrazine conduc- was used throughout the experiment. All the materials and reagents were of
∗ To whom correspondence should be addressed. e-mail: d06001@gly.oups.ac.jp © 2008 Pharmaceutical Society of Japan

978
Vol. 56, No. 7
the analytical grade and were used without further purification. An H2O2
sample (2.1—460 ng) was placed in a 10 ml calibrate flask, and then 2.5 ml
of 1.0% DTAC solution, 2.0 ml of the Na2CO3/NaHCO3 buffer solution (pH
HCl/tris(hydroxymethyl)aminomethane. The maximum, con-
stant value of RFI was observed in the at pH range 9.6—9.8
with 0.2 M Na2CO3/NaHCO3 buffer solution; thus, 2.0 ml of
the buffer solution (pH 9.6) was used for the pH adjustments
9.6), 1.0 ml of 1.0ꢃ10ꢀ4 M cobalt(II) solution, and 0.4 ml of a 1.0ꢃ10ꢀ4
M
FH solution were added. The solution was diluted to 10 ml with water, trans-
ferred into a test tube, mixed well, and maintained at 80 °C for 20 min. After in the final volume of 10 ml.
cooling at room temperature in water for 5 min, the difference between the
An addition of various surfactants in the coloring or fluo-
relative fluorescence intensities {RFIꢂ(AꢀB)/B} of solution A and reagent
blank (solution B), which were prepared under the same conditions, was
measured by the excitation/emission wavelength at 460/527 nm.
rescence reactions between various reagents and metal ions
has already offered many advantages in comparison to the
absence of surfactants.24—29) Accordingly, in order to devel-
op the fluorescence and enhance the sensitivity, we examin-
ed the effects of different surfactants: Cationic [DTAC,
Results and Discussion
Spectral Properties FH is a colorless, non-fluorescent cetyltrimethyl ammonium chloride (CTAC), stearyltrimethyl
spirolactam hydrazide. Figure 1 shows the absorption spectra ammonium chloride (STAC), cetylpyridinium chloride
of fluorescein, FH and its reaction solution with H2O2 in a (CPC), cetylpyridinium bromide (CPB), benzyldimethyl-
10 ml calibrate flask, 2.5 ml of 1.0% DTAC solution, 2.0 ml tetradecyl ammonium chloride (Zephiramine)], Anionic
of the Na2CO3/NaHCO3 buffer solution (pH 9.6), and 1.0 ml [sodium dodecyl sulfate (SDS), di-(2-ethylhexyl) sodium sul-
of 1.0ꢃ10ꢀ4 M cobalt(II) solution. It can be observed that FH fosuccinate (AOT)], Nonionic [polyoxyethylene sorbitan
is not absorbed in the visible region (curve C); the molar ab- monolaurate (Tween 20), Triton X-405, polyvinylpyrrolidone
sorptivity of FH at the characteristic absorption (494 nm, (PVP), polyvinylalchol (PVA nꢂ500)], and Amphoteric
curve A) of fluorescein is only 4.0ꢃ10ꢀ3 ꢀ1 cmꢀ1. This is [Swanol AM-301]. The maximum, constant value of RFI was
M
attributed to its closed spirolactam form. However, upon re- obtained over the range of 1.0—2.5 ml of 1.0% DTAC solu-
action with H2O2, the green color (curve B), which indicates tion in a final volume of 10 ml.
fluorescein, is noticeably restored. When H2O2 was intro-
The metal ions reacted catalytically when small amounts
duced into a solution of FH, a fluorescence emission identi- of the metal ions coexisted in various redox reactions. The
cal to that of fluorescein was observed with the maximum various metal ions were tested for the assay of H2O2.
fluorescence emission at 527 nm. Further, the fluorescence Cobalt(II) is superior to the various metal ions tested:
development reaction was significantly facilitated when it cobalt(II), zinc(II), iron(II), nickel(II), copper(II), palla-
was carried out in the presence of cobalt(II) and cationic sur- dium(II), manganese(II), platinum(II), iron(III), alu-
factants. In order to explore the possible reaction products, minum(III), gold(III), yttrium(III), lanthanum(III), rhodi-
the fluorescence emission spectra of the reaction solution um(III), terbium(III), zirconium(IV), tin(IV), tantalum(V),
were compared with that of fluorescein (Fig. 2, G). Figure 2 molybdenum(VI), and tungsten(VI). The maximum, constant
shows the fluorescence emission spectra of the mixture of FH value of RFI was obtained in the fluorescence reaction with
with cobalt(II) and DTAC as blank solution (Fig. 2, A), and 1.0ꢃ10ꢀ5 M cobalt(II) in the final concentration. In this ex-
obtained after H2O2 was added to the blank solution as a amination condition, Cu(II) was not reacted with FH, though
sample solution (Fig. 2, B—F). The excitation and emission it was reported that FH provided fluorescent responses to
spectra of the sample solution were observed at pH 9.6. The Cu(II) itself.30,31) The effect of concentration of FH was ex-
results indicate that these spectra are identical and both have amined by varying the amounts of the FH solution, while
the same values of the maximum emission wavelengths. maintaining a fixed final concentration of H2O2 (ng mlꢀ1).
From the above results, it was assumed that FH was oxidized The maximum, almost constant value of RFI was obtained by
by H2O2 to yield the highly fluorescent product, fluorescein.
Optimization of Reaction Variables The effect of pH
on the fluorescence investigated by using various buffer
solutions and pH such as 0.2 M Na2CO3/NaHCO3, 0.05 M
borax/0.1 M sodium hydroxide, 0.2 M NH3/NH4Cl and, 0.2 M
Fig. 2. Fluorescence Emission Spectra of FH in the Absence and Presence
of H2O2 with a Standard Solution
(A) 4ꢃ10ꢀ6 M FH; (B)—(F) reaction solution of 4ꢃ10ꢀ6 M FH with H2O2 (3.0, 30,
Fig. 1. Absorption Spectra of the Probe FH and Its Reaction Solution with
H2O2 as well as Fluorescein in the Standard Procedure Solution against the
300, 3000, 30000 ng mlꢀ1, respectively); (G) 4ꢃ10ꢀ6 M fluorescein. The conditions
were 2.5 ml of 1.0% DTAC solution, 2.0 ml of the Na2CO3/NaHCO3 buffer solution
Corresponding Reagent Blank
(pH 9.6), and 1.0 ml of 1.0ꢃ10ꢀ4 M cobalt(II) solution. The excitation and emission val-
ues were selected as 494 nm and 527 nm, respectively. Slit band: excitation/emis-
sionꢂ5.0/5.0 nm.
(A) 4ꢃ10ꢀ6 M FH; (B) reaction solution of 4ꢃ10ꢀ6 M FH with H2O2 (300 ng mlꢀ1);
and (C) 4ꢃ10ꢀ6 M fluorescein.

July 2008
979
using the final concentration of FH at 4.0ꢃ10ꢀ6 M for the de- present method were not different from those obtained by the
termination of H2O2. comparison procedure, the Cayman Clinical H2O2 Assay Kit
The fluorescence development in this reaction system did (the Wolff’s FOX Assays), at a 95% confidence level: the cal-
not occur instantaneously at room temperature. Thus, the ef- culated tꢂ0.584 was lower than the critical t value (pꢂ0.05,
fects of the incubation temperature and time were investi- nꢂ6) of 2.447. These results are given in Table 3.
gated by heating for 10—60 min at 50, 60, 70, and 80 °C.
Characterization of FH Xanthene dyes such as fluores-
The maximum, constant value of RFI was obtained at 80 °C cein have been commonly used in various fields for a long
for 15—40 min, followed by cooling in water for 5 min. The time. Though, the relationship between a chemical structure
RFI value remained constant for at least 2 h after the solution and fluorescence remains almost unknown. Recently, it was
was cooled to room temperature.
reported that the lacton form of fluorescein is non or almost
Calibration Curve and Reproducibility A calibration no fluorescent, while its carboxyl form is fluorescent.32)
curve for H2O2 was constructed by the standard procedure. A However, there are not too a lot of examples of the crystal
good linear relationship and wide dynamic range were ob- structure where these ideas can be proven. The factors re-
served over 2.1—460 ng mlꢀ1 of H2O2. The limit of detec- sponsible for this are considered to be the difficulty in the
tion, defined as (3.3ꢃS.D. of blank)/(slope of analytical cali- crystallization and the logical design of a fluorescent group.
bration), was 0.7 ng mlꢀ1. The correlation coefficient was This time, we were able to obtain the crystal of FH from the
0.999. The relative standard deviations (nꢂ6) were 4.06% ethanol–acetonitrile solution after purifying it several times.
for 3.1 ng mlꢀ1 of H2O2 1.78% for 30.8 ng mlꢀ1 of H2O2, and Further the structure was confirmed by MS, NMR, and X-ray
2.21% for 308 ng mlꢀ1 of H2O2. The sensitivity of the pro- crystallography. As shown in Table 4 and Fig. 3, FH was
posed method is almost doubled and the reproducibility ob-
tained is excellent. When the other ROS, superoxide (O2ꢀ),
Table 2. Effect of Foreign Substances on the Determination of H2O2
hydroxyl radical (·OH), nitric oxide (NO), peroxynitrite
Added
RFI
Recovery
%
(ONOOꢀ), and peroxide (R–OOH) are examined, an excel-
lent result is obtained, as shown in Table 1.
Substances
None
Cu(II)
Ni(II)
Mg(II)
Zn(II)
Ca(II)
Added as
ng mlꢀ1
Mole ratio (SꢀB)/B
Interference from Foreign Substances For the assess-
ment of the advantages of our method in environmental and
clinical assays, the influences of various foreign substances
on the determination of H2O2 were examined. Inorganic ions
such as magnesium(II), zinc(II), calcium(II), molybde-
num(VI), sodium, chloride, fluoride, potassium, bromide,
cyanide, phosphate, ammonium, nitrate, and sulfate did not
noticeably affect the accuracy of the determination of H2O2,
even when these ions were present in excessively large
amounts as compared to that of H2O2. Among foreign metal
ions, the presence of copper(II), nickel(II), iron(II), and
iron(III) resulted in a slight increase or decrease in determi-
nation of H2O2. The presence of organic substances such as
ascorbic acid, caffeine, glucose, HSA, urea, uric acid, creati-
nine, and alanine caused considerably less interference. The
results are summarized in Table 2.
Application The proposed method was used for the de-
termination of H2O2 in human urine. Spot human urine sam-
ples were collected from healthy human volunteers and as-
sayed immediately. Preliminary treatments were not con-
ducted to remove co-existing substances and the urine was
merely diluted 100 times. Neat standards were prepared
under the same conditions. The average recovery of H2O2
from samples was 96.7%, with RSD values of less than
5.8%. A T-test demonstrated that the results obtained by the
—
Nitrate
Nitrate
Chloride
Chloride
Chloride
Sulfate
Alum
Sodium
—
—
—
1
1/10
100
100
10
9.3
9.8
11.6
9.3
9.2
9.7
10.2
9.1
9.3
10.1
9.3
9.3
8.5
9.3
9.2
9.3
9.3
9.3
8.3
10.8
9.3
9.8
9.3
9.3
10.6
9.3
100
105.4
124.4
100
98.4
104
110
98.1
100
109
100
100
91.3
100
99
100
100
100
89.2
116.1
100
105.4
100
100
114
100
6.4ꢃ103
2.9ꢃ102
2.4ꢃ103
6.5ꢃ103
1.1ꢃ104
1.4ꢃ104
11
Fe(II)
5
Fe(III)
Mo(VI)
NaCl
NaF
KBr
1/5
100
50
100
100
100
10
9.6ꢃ103
2.9ꢃ103
4.3ꢃ103
1.2ꢃ104
6.5ꢃ103
1.4ꢃ103
5.4ꢃ102
1.0ꢃ104
1.7ꢃ104
1.7ꢃ103
1.8ꢃ103
1.9ꢃ104
1.8ꢃ105
10
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
KCN
Na2HPO4
NH4Cl
KNO3
K2SO4
NH3
Ascorbic acid
Caffeine
Glucose
HSA
10
100
100
1000
10
100
1000
—
100
100
50
Urea
6.0ꢃ103
1.7ꢃ105
6.6ꢃ104
8.9ꢃ104
Uric acid
Creatinine
Alanine
1000
H2O2, 30.8 ng mlꢀ1; FH, 4.0ꢃ10ꢀ6 M; DTAC, 0.25%; Co(II), 1.0ꢃ10ꢀ5 M; pH, 9.6;
excitation/emission, 460/527 nm.
Table 3. Determination of H2O2 in Human Urine
Found
Table 1. Determination of ROS
RSDb)
(%)
Recoveryc)
(%)
Concentration
RSD
(%)
DL
Sample
Sample
range (ng mlꢀ1
)
(ꢂ3.3 s/slope)
Present method Other methoda)
H2O2
O2ꢀ
·OH
2.1—460
1.2—120
1.7—170
6.0—300
5.6—580
30—900
1.8
5.2
2.9
3.5
3.8
5.3
0.7
9.2
2.1
1.2
1.3
10.2
A
B
C
D
6.3
19.3
4.8
6.1
10.9
2.9
3.1
2.2
5.8
4.3
103.3
92.2
95.3
96.1
NO
8.6
10.1
ONOOꢀ
R–OOH
The calculated tꢂ0.584 was lower than the critical t value (pꢂ0.05, nꢂ6) of 2.447.
a) Other method; The Cayman Chemical Hydrogen Peroxide Assay Kit. b) Average of
5 determinations. c) H2O2 taken; 30.8 ng mlꢀ1. FH, 4.0ꢃ10ꢀ6 M; DTAC, 0.25%; Co(II),
1ꢃ10ꢀ5 M; pH, 9.6; excitation/emission, 460/527 nm.
Conditions: H2O2, 30.8 ng; FH, 4.0ꢃ10ꢀ6 M; DTAC, 0.25%; Co(II), 1ꢃ10ꢀ5 M; pH,
9.6; excitation/emission, 460/527 nm; correlation coefficient is all 0.999.

980
Vol. 56, No. 7
Table 4. Crystal and Experimental Data for FH
based on a redox reaction among H2O2, cobalt(II), and FH.
The proposed method is more sensitive and reproducible than
the previous method. On the occasion of characterization of
FH, the xanthene scaffold and benzene ring have two-faced-
ness and an almost formed spirolactam round form at right
angles to each other. As a result, it hardly probable that the
total fluorescence may have disappeared because an elec-
tronic conjugated system of the xanthene scaffold that was a
fluorescent characteristic of fluorescein collapsed by the for-
mation of a spirolactam ring. Although further investigations
are necessary for the elucidation of this reaction mechanism,
the developed procedure is suitable for the analysis of ROS
in clinical studies. In conclusion, the improved method offers
many advantages for H2O2 determination. First, the method
is remarkably improved with respect to reproducibility and
sensitivity in comparison with the former method. Second,
there is a remarkable improvement in the influences of for-
eign substances by this method. Finally, the new method is
superior to the former method with regard to the wide dy-
namic range of the calibration curve.
Formula
C20H14N2O4, CH3CN, 0.5(CH3OH)
Formula weight
403.41
Crystal system
Space group
Monoclinic
P21/n
A
B
C
b
V
Z
9.854(1) Å
8.643(1) Å
23.298(3) Å
91.687(2)°
1983.5(4) Å3
4
T
120(2) K
0.35ꢃ0.35ꢃ0.05 mm3
1.351 g cmꢀ3
844
Crystal size
Dx
F(000)
m(MoKa)
No. of reflections (obs)
0.096 mmꢀ1
21332
Rint
0.0275
qmax
27.10°
4361
No. of reflections (refine)
No. of reflections (Iꢄ2s(I))
No. of parameters
R
3478
296
0.0549
wR
0.16
Goodness of fit
(D/s)max
Fraction for qmax
Drmax
0.973
0.008
0.998
Acknowledgments The authors thank Dr. K. Minoura and Mrs. M. Fu-
jitake of Osaka University of Pharmaceutical Sciences for help with meas-
urements of NMR and MS. This study was supported by a Grant-in-Aid for
High Technology Research from Ministry of Education, Culture, Sports,
Science and Technology of Japan.
0.918 e Åꢀ3
ꢀ0.310 e Åꢀ3
Drmin
References
1) Ohshima H., Tatemichi M., Sawa T., Arch. Biochem. Biophys., 417,
3—11 (2003).
2) Shah M. A., Channon M. A., Heart, 90, 486—487 (2004).
3) Barnham J. K., Masters L. C., Bush I. A., Nat. Rev. Drug Discov., 3,
205—214 (2004).
4) “Metals and Oxidative Damage in Neurological Disorders,” ed. by
Connor R. J., Plenum Press, New York, 1997.
5) Kojou S., Vitamine, 79, 334—337 (2005).
6) “Molecular Biology of Free Radicals in Human Diseases,” ed. by
Aruoma I. O., Halliwell B., OICA International, Micoud St. Lucia,
1998.
7) Balaban S. R., Nemoto S., Finkel T., Cell, 120, 483—495 (2005).
8) Beckman B. K., Ames N. B., Physiol. Rev., 78, 547—581 (1998).
9) Vieira I. C., Filho O. F., Analyst, 123, 1809—1812 (1998).
10) Chai X.-S., Hou Q. X., Luo Q., Zhu J. Y., Anal. Chim. Acta, 507,
285—288 (2004).
11) Deiana L., Carru C., Pes G., Tadolini B., Free Radic. Res., 31, 237—
244 (1999).
12) Lazrus A. L., Kok G. L., Gitlin S. N., Lind J. A., McLaren S. E., Anal.
Chem., 57, 917—922 (1985).
13) Liu Z., Cai R., Mao L., Huang H., Ma W., Analyst, 124, 173—176
(1999).
Fig. 3. OTEP of FH
14) Odo J., Matsumoto K., Shinmoto E., Hatae Y., Shiozaki A., Anal. Sci.,
20, 707—710 (2004).
15) Chang M. C. Y., Pralle A., Isacoff E. Y., Chang C. J., J. Am. Chem,
Soc., 126, 15392—15393 (2004).
16) Maeda H., Fukuyasu Y., Yoshida S., Fukuda M., Saeki K., Matsuno
H., Yamaguchi Y., Yoshida K., Hirata K., Miyamoto K., Angew.
Chem., Int. Ed., 43, 2389—2391 (2004).
17) Setsukinai K., Urano Y., Kakinuma K., Majima H. J., Nagano T., J.
Biol. Chem., 278, 3170—3175 (2003).
18) Maeda H., Yamamoto K., Nomura Y., Kohno I., Hafsi L., Ueda N.,
Yoshida S., Fukuda M., Fukuyasu Y., Yamaguchi Y., Itoh N., J. Am.
Chem. Soc., 127, 68—69 (2005).
19) Zappacosta B., Persichilli S., Mormile F., Minucci A., Russo A., Giar-
dina B., De Sole P., Clin. Chim. Acta, 310, 187—191 (2001).
20) Yamashiro N., Uchida S., Satoh Y., Morishima Y., Yokoyama H., Satoh
T., Sugama J., Yamada R., J. Nucl. Sci. Technol., 41, 890—897 (2004).
21) Mori I., Takasaki K., Fujita Y., Matsuo T., Talanta, 47, 631—637
(1998).
formed with the benzene ring through hydrazine for a xan-
thene scaffold to have a plane structure and a spirolactam
round was formed with a connection that was almost right
angled. We found that hydrazation fluorescein would force
this platform to adopt a closed, colorless, and non-fluorescent
spirolactam form. Oxidation with H2O2 and other ROS
would subsequently produce the open, colored, and fluores-
cent fluorescein product.
Conclusion
We have described the synthesis, properties, application,
and characterization of fluorescein-hydrazide, a new class of
fluorescent probes for detecting ROS. A simple, highly sensi-
tive, and large dynamic range fluorometric method for the
determination of H2O2, was established with FH and
22) Baeyer A., Ber. Deut. Chem. Ges., 4, 555—558 (1871).
cobalt(II) in a DTAC micellar medium. This procedure is 23) Akita T., Yakugaku Zasshi, 81, 518—521 (1961).

July 2008
981
24) Mori I., Fujita Y., Kato K., Toyoda M., Akagi M., Fresenius J. Anal.
Chem., 343, 902—904 (1992).
25) Mori I., Fujita Y., Kato K., Toyoda M., Akagi M., Bull. Chem. Soc.
Jpn., 62, 2536—2540 (1989).
26) Nonova D., Stoyaok K., Anal. Chim. Acta, 138, 321—328 (1982).
27) Mori I., Fujita Y., Kato K., Toyoda M., Kubo S., Fresenius J. Anal.
Chem., 342, 80—82 (1992).
Anal. Chem., 330, 619—623 (1988).
29) Mori I., Fujita Y., Kato K., Toyoda M., Akagi M., Talanta, 38, 683—
690 (1991).
30) Chen X., Ma H., Anal. Chim. Acta, 575, 217—222 (2006).
31) Dujols V., Ford F., Czarnik A. W., J. Am. Chem. Soc., 119, 7386—7387
(1997).
32) Sun W. C., Gee K. R., Klaubert D. H., Haugland R. P., J. Org. Chem.,
62, 6469—6475 (1997).
28) Mori I., Fujita Y., Nakahashi Y., Tanaka T., Ishihara S., Fresenius J.