A

orlp©

K

1

iigr[Dtacotktia

t

0d

Talanta 72 (2007) 1283–1287

A novel fluorescent method for determinationof peroxynitrite using folic acid as a probe

Jun-Chao Huang a, De-Jia Li b, Jun-Chen Diao a,Jie Hou a, Jiang-Lan Yuan a, Guo-Lin Zou a,∗

a State Key Laboratory of Virology, College of Life Sciences, Wuhan University, Wuhan, Hubei 430072, Chinab Medical College, Zhengzhou University, Zhengzhou, Henan 450052, China

Received 10 May 2006; received in revised form 14 January 2007; accepted 14 January 2007Available online 20 January 2007

bstract

A novel method for the determination of peroxynitrite using folic acid as a fluorescent probe is described. The method is based on the oxidationf the reduced, low-fluorescent folic acid by peroxynitrite to produce a high-fluorescent emission product. The fluorescence increase is linearly

elated to the concentration of peroxynitrite in the range of 3 × 10−8 to 5.0 × 10−6 mol L−1 with a correlation coefficient of 0.998, and the detectionimit is 1 × 10−8 mol L−1. Interferences from some metal ions normally seen in biological samples, and also some anions structurally similar toeroxynitrite were studied. The optimal conditions for the detection of peroxynitrite were evaluated.

2007 Elsevier B.V. All rights reserved.

pldefdlbm[[

abiod

eywords: Peroxynitrite; Folic acid; Fluorimetry; Probe

. Introduction

Peroxynitrite chemistry is of remarkable interest withncreasing evidences showing the importance of peroxynitriten the development of oxidative damage in various patholo-ies [1,2]. Peroxynitrite, generated from the diffusion-controlledeaction between the nitrogen monoxide and superoxide radicals3,4], can cause lipid peroxidation [5], chemical cleavage ofNA [6,7], inactivation of key metabolic enzymes (e.g., aconi-

ase [8,9], succinate dehydrogenase, ribonucleotide reductase,nd cytochrome oxidase of the mitochondrial electron transporthain [10,11]), and reduction in cellular antioxidant defenses byxidation of thiol pools [12]. Peroxynitrite can also nitrate pro-ein tyrosine residues, possibly leading to inactivation of tyrosineinase [13] or the disruption of key cytoskeletal componentshat may contribute to the pathogenesis of diseases, includ-ng inflammatory processes, ischemia-reperfusion, septic shock,

nd neurodegenerative disorders [14–16].

Due to the great importance of the research mentioned above,he point of interest was concentrated upon the measurement of

∗ Corresponding author. Tel.: +86 27 87645674; fax: +86 27 87669560.E-mail address: jchuang77@163.com (G.-L. Zou).

dimtpai

039-9140/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.talanta.2007.01.033

eroxynitrite in both pathological and normal conditions in bio-ogical systems. However, the peroxynitrite assay is extremelyifficult because of the low concentration, high activity, andlusive natures of peroxynitrite. Currently, most of techniquesor peroxynitrite measurement are indirectly based on chemicaletection of the decomposition products removed from bio-ogical systems. Peroxynitrite generation is usually measuredy ultraviolet–visible (UV–vis) spectroscopy [17], chemilu-inescence [18], immunohistochemistry [19], amperometry

20,21], electron spin resonance (ESR) [22,23], and fluorescence24–26].

Two fluorogenic probes, dihydrodichlorofluorescein (DCFH)nd dihydrorhodamine-123 (DHR-123), which are considered toe ideal, have been widely employed to monitor peroxynitriten various systems [27,28]. The above methods are based on thexidation of the reduced, non-fluorescent forms of fluorescentyes such as fluorescein and rhodamine by peroxynitrite to pro-uce the parent dye molecule, resulting in a dramatic increasen fluorescence response. However, the synthesis of these probe

olecules is rather difficult and inconvenient [29]. Additionally,

he use of organic dyes is very likely to result in environmentalollution, which should be avoided as possible as we can. Thus,cheap, fast, and simple method to determine the peroxynitrite

s needed.

1 lanta

fafbasltopoedsb

2

2

w0tsHub

2

dnHqsias

2

safltta

2

ba

wsce

2

iitttotBtut

3

3

obepscc

3.2. Optimization of the general procedure

The effect of pH on the fluorogenic reaction was studied inthe range of 7–9.4 in barbital buffer solution, and the results are

284 J.-C. Huang et al. / Ta

Folic acid is made up of a pterin moiety (purine and pyrazineused together) that is linked to the side chain containing p-minobenzoic acid (pteroic acid) and glutamic acid. Folic acidunctions as a cofactor in the transfer and utilization of one car-on groups, which plays a key role in the biosynthesis of purinesnd pyrimidines and regeneration of methionine [30]. Recenttudies showed that the pathogenesis of cardiovascular, hemato-ogical and neurological diseases and cancer are associated withhe antioxidant activity of folic acid. Folic acid can act as a per-xynitrite scanvenger [31,32] due to its high reaction rate witheroxynitrite. Folic acid is a low-fluorescent substance, but thexidation folic acid by peroxynitrite can give high-fluorescentmission product. Comparing with the above mentioned twoyes, folic acid is relative inexpensive, not toxic to biologicalystem and stable in solution. The present study was designed touild a new fluorometric method for peroxynitrite determination.

. Experimental

.1. Chemicals

Folic acid (Shanghai Chem. Agent, Inc., Shanghai, China)as prepared by dissolving appropriate amount of folic acid in.001 mol L−1 NaOH and kept frozen, and the working barbi-al buffer solution was prepared by dissolving 4.125 g barbitalodium in 500 mL distilled water and add 0.7 mL 1.0 mol L−1

Cl. All the reagents were of analytical reagent grade and weresed without further purification, unless stated otherwise. Dou-ly distilled water was used throughout.

.2. Synthesis of peroxynitrite

Peroxynitrite was synthesized according to the previousescription [33]. An aqueous solution of 0.6 mol L−1 sodiumitrite was rapidly mixed with an equal volume of 0.7 mol L−1

2O2 containing 0.6 mol L−1 HCl and then immediatelyuenched with the same volume of 1.5 mol L−1 NaOH. Thenome MnO2 powder was added to the mixture solution to elim-nate the excess H2O2, then the mixture was filtered and storedt −18 ◦C. Peroxynitrite concentration was determined by UVpectrometry at 302 nm (ε = 1670 L mol−1 cm−1) [29].

.3. Apparatus

Absorption spectra were obtained on a cary-100 UV–visiblepectrophotometer (Varian, USA). The fluorescence spectrand relative fluorescence intensity were measured with auorescence-4500 (Hitachi, Japan) with a 10 mm quartz cuvette,

he excitation and the emission wavelength slits were respec-ively set at 5.0 and 10.0 nm. All pH values were measured withpH S-301 digital ion meter.

.4. Procedure

In a set of 10 mL-volumetric tubes containing pH 9.4 bar-ital buffer solution, 1.0 mL of folic acid (1.0 × 10−4 mol L−1)nd different amount of peroxynitrite were added. The tubes

Fam2

72 (2007) 1283–1287

ere closed and then quickly and carefully shaken. The reactionolution was kept at room temperature for 5 min. The fluores-ence intensity of the solution was recorded at 460 nm with thexcitation wavelength set at 380 nm.

.5. Detection of peroxynitrite in biological samples

At first, the Hela cells were maintained in Dulbecco’s mod-fied Eagle’s medium (Gibco-BRL) which contains 10% heat-nactivated fetal calf serum, penicillin (100 U mL−1) and strep-omycin (50 �g mL−1) in a 5% CO2 environment at 37 ◦C. Next,hese cells were plated on a six-well chamber. When growing upo 80% confluency, they were treated with various concentrationf adriamycin for 6 h. Then the adherent cells were detached byrypsin treatment and washed twice with PBS containing 1%SA. After that, these cells were added to barbital buffer solu-

ion containing 1.5 × 10−5 mol L−1 folic acid and were brokensing an ultrasonic disintegrator. Finally, the suspension wasaken for analysis under the optimally experimental conditions.

. Results and discussion

.1. Spectra characteristics

Fig. 1 shows the fluorescence excitation and emission spectraf folic acid and the mixture of folic acid with peroxynitrite inarbital buffer solution (pH 9.4). Folic acid has low fluorescencexcitation and emission spectra. However, high fluorescenceroduct generated by the introduction of peroxynitrite into theolution of folic acid, resulting in dramatic increase in spectraharacteristics with excitation maximum at 380 nm and fluores-ence emission maximum at 460 nm.

ig. 1. Fluorescence excitation and emission spectra of the system. (And a) Folic acid, 1.5 × 10−5 mol L−1; (B and b) folic acid, 1.5 × 10−5

ol L−1 + peroxynitrite, 2.0 × 10−6 mol L−1; pH 9.4 barbital buffer solution,5 ◦C.

J.-C. Huang et al. / Talanta 72 (2007) 1283–1287 1285

F −5 −1

+

srcttTtbtptshHr

tsitfliofamoh

tsFwtcfs

Fig. 3. The kinetic characteristics of the detection system. (A) Folica −5 −1 −6 −1

1s

3

ipwTitb

spthe detection data reveals a good correlation and no statisticaldifference compared with the standard data. It can be seen thatthe new method is comparatively precise.

ig. 2. Effect of pH on the system. Folic acid, 1.5 × 10 mol Lperoxynitrite, 2.0 × 10−6 mol L−1; barbital buffer solution, 25 ◦C.

hown in Fig. 2. The data shown are average values of threeepeated determinations. As can be seen from Fig. 2, fluores-ence intensity of the detective system increased greatly withhe pH value, and then leveled off as the pH value approachedo 9.4, which is near the upper limit pH value of barbital buffer.he fluorescence intensity is pH dependent, because peroxyni-

rite can only exist for a short moment in neutral environment,ut a much longer time in basic environment. For example,he half-life of peroxynitrite is less than 1 s at pH 7.4. Mosteroxynitrite decayed before reacting with folic acid in neu-ral environment, and with pH increased, it can become moretable and react with more folic acid. But when pH was tooigh, peroxynitrite became too stable to react with folic acid.ence, pH 9.4 of barbital buffer was chosen for the fluorogenic

eaction.The kinetic characteristics of the proposed detection sys-

em were studied. Upon the addition of peroxynitrite to theolution of folic acid in barbital buffer, the fluorescencentensity of the detective system was recorded as a func-ion of reaction time. From Fig. 3, we can see that theuorescence intensity of the detection system reached its max-

mum value in about 230 s, then the fluorescence intensityf the detective system almost remained unchanged. There-ore, to obtain a highly sensitive and reproducible results,

5-min reaction time was selected in the following experi-ent. Further results showed that the fluorescence intensity

f the detective system almost remained unchanged in a fewours.

The effect of the concentration of folic acid on �F of the sys-em was studied and the results were shown in Fig. 4. The datahown are average values of three repeated determinations. Fromig. 4, it can be seen that �F of the detection system increasedhen the concentration of folic acid increased from 5.0 × 10−8

o 1.5 × 10−5 mol L−1 and the increase slowed down when theoncentration of folic acid was up to 1.5 × 10−5 mol L−1, there-ore, 1.5 × 10−5 mol L−1 of folic acid was recommended for theubsequent experiment.

Fm(

cid, 1.5 × 10 mol L + peroxynitrite, 2.0 × 10 mol L ; (B) folic acid,.5 × 10−5 mol L−1 + peroxynitrite, 2.0 × 10−7 mol L−1; pH 9.4 barbital bufferolution, 25 ◦C.

.3. Analytical performance

Under the selected conditions given above, the fluorescencencrement shows a linear relationship with the concentration oferoxynitrite in the range of 3 × 10−8 to 5.0 × 10−6 mol L−1

ith a correlation coefficient 0.998, which is shown in Fig. 5.he detection limit, calculated according to the 3Sb/k criterion,

n which k is the slope of the range of the linearity used and Sb,he standard deviation (n = 9) of the blank solution, is found toe 1.0 × 10−8 mol L−1.

In order to demonstrate the accuracy of the new method, aeries of peroxynitrite standard solutions are determined by theroposed fluorometric method. The results are shown in Table 1,

ig. 4. Effect of the concentration of folic acid on the fluorescence incre-ent of the system. Folic acid (different concentrations) + peroxynitrite

1.0 × 10−5 mol L−1). pH 9.4 barbital buffer solution, 25 ◦C.

1286 J.-C. Huang et al. / Talanta 72 (2007) 1283–1287

Fig. 5. The calibration curve in the range of 3 × 10−8 to 5.0 × 10−6 mol L−1

peroxynitrite. Folic acid, 1.5 × 10−5 mol L−1; pH 9.4 barbital buffer solution,25 ◦C.

Table 1Comparison of determination results (mol L−1) between this method and stan-dard spectrophotometry

Sample no. Standard method This method Relative error (%)

1 5.0 × 10−7 4.76 × 10−7 5.02 8.0 × 10−7 7.88 × 10−7 1.53 1.6 × 10−6 1.61 × 10−6 −0.645

tatttb

TTo

I

PNNHNGMGGAUCFM

Table 3Standard additions and recovery data of cell samples

Sampleno.

Peroxynitritein samples(10−7 mol L−1)

Peroxynitriteadded(10−7 mol L−1)

Peroxynitritefound(10−7 mol L−1)

Recovery(%)

1 3.36 1.0 3 98 9110.0 11.15 83

2 2.62 1.0 3 22 8810.0 11.33 90

3 1.26 1.0 1.91 8510.0 9.68 86

Tr

ilticbd

3

ciysdtop

4

3.6 × 10−6 3.68 × 10−6 −2.24.6 × 10−6 4.75 × 10−6 −3

The interferences of the proposed probe for peroxyni-rite were studied. A variety of interfering agents from somemino acids, metal ions normally seen in biological sampleso anions structurally similar to peroxynitrite were added to

he detected system containing 2.0 × 10−6 mol L−1 peroxyni-rite and 1.5 × 10−5 mol L−1 folic acid in pH 9.4 of barbitaluffer. Results in Table 2 show a fairly satisfactory selectiv-

able 2he response for a variety of potential interferences expressed as a percentagef the �F of peroxynitrite in pH 9.4 barbital buffer solution, 25 ◦C

nterference Concentration (10−6 mol L−1) Peroxynitrite (%)

eroxynitrite 2 100 (F = 906)O3

− 1000 0.54O2

− 100 −0.29

2O2 100 0.5H2OH 1000 1.1lucose 100 −0.2ethionine 1000 −0.19lycine 1000 −4.2lutathione 1000 −5.0scorbic acid 1000 −2.2ric acid 100 −2.3u2+ 20 −3.8e3+ 20 −1.9n2+ 20 −1.8

ofiotppTripDiTHtspr

he final adriamycin concentration for samples 1–3 are 50, 30 and 15 �mol L−1,espectively.

ty of the method. Especially the interference of H2O2 is quiteow. Because the reactivity of peroxynitrite is 2000 times greaterhan H2O2 [34]. But the interference of Cu2+, Mn2+ and Fe3+

s a little high, because these metal ions or their compoundsan react with peroxynitrite [35–37]. But these metal ion cane inactivated by the addition of a chelated reagent such asiethylenetriaminepentaacetic acid (DTPA).

.4. Analysis of biological samples

As is well known that peroxynitrite can be generated fromell injury induced by adriamycin. Results of the fluorescencentensity and the recovery data of addition of external perox-nitrite standard solutions to cells treated with adriamycin, arehown in Table 3, which are average values of three repeatedeterminations. The recovery data is not good enough, becausehe substances of broken cells in suspension can react with per-xynitrite, which inhibits the reaction between folic acid anderoxynitrite.

. Conclusion

In summary, a novel fluorescent probe for the determinationf peroxynitrite was proposed in this paper. As an indicator,olic acid is easily available, reasonably inexpensive, and stablen solution. The present method for the determination of per-xynitrite is cheap, simple and sensitive. The response time ofhe proposed probe for peroxynitrite is less than 5 min. Com-ared with the commonly used probe DHR-123, the fluorescentrobe reported here has some advantages. (I) Higher sensitivity.he oxidation of folic acid by peroxynitrite is linear over the

ange of 3 × 10−8 to 5.0 × 10−6 mol L−1, while the linear ranges within 1.0 × 10−6 mol L−1 for DHR-123. In addition, the pro-osed probe has a detection limit of 1.0 × 10−8 mol L−1 whereasHR-123 has a detection limit of 1.0 × 10−7 mol L−1 [38]. This

s desirable for probing of peroxynitrite in lower concentrations.hese make it ideal for detecting peroxynitrite formation. (II)igher stability. Folic acid demonstrates greater photostability

han those of DCFH and DHR-123, both of which are extremelyensitive to light induced oxidation [39]. (III) The fluorescentrobe reported here is commercially available, and less likely toesult in environmental pollution. Since folic acid is much likely

lanta

troa

A

rS

R

[[[

[[

[

[[

[[

[[[[

[[

[

[

[[[[[

[

[[

[mun. 317 (2004) 873.

J.-C. Huang et al. / Ta

o coexist with peroxynitrite within living beings, the oxidationeaction might probably be another pathway of less injury to cellsr organisms. Hence, we believe that the proposed probe can bepplicable to the study of peroxynitrite in biological systems.

cknowledgement

The authors gratefully acknowledge the support of thisesearch by a grant (no. 30370366) from the National Naturalcience Foundation of China.

eferences

[1] J.S. Beckman, W.H. Koppenol, Am. J. Physiol. Cell Physiol. 271 (1996)C1424.

[2] C. Ducrocq, B. Blanchard, B. Pignatelli, Cell. Mol. Life Sci. 55 (1999)1068.

[3] R. Kissner, T. Nauser, P. Bugnon, Chem. Res. Toxicol. 10 (1997) 1285.[4] T. Nauser, W.H. Koppenol, J. Phys. Chem. A 106 (2002) 4084.[5] R. Radi, J.S. Beckman, K.M. Bush, Arch. Biochem. Biophys. 288 (1991)

481.[6] T. Douki, J. Cadet, B.N. Ames, Chem. Res. Toxicol. 9 (1996) 3.[7] V. Yermiloy, J. Rubio, M. Becchi, Carcinogenesis 16 (1995) 2045.[8] L. Castro, M. Rodriguez, R. Radi, J. Biol. Chem. 269 (1994) 29409.[9] A. Hausladen, I. Fridovich, J. Biol. Chem. 269 (1994) 29405.10] R. Ridi, M. Rodriguez, L. Castor, Arch. Biochem. Biophys. 308 (1994) 89.11] J.P. Bolanos, S.J.R. Heals, J.M. Land, J. Neurochem. 64 (1995) 1965.12] M.G. Salgo, E. Bermudez, G.L. Squadrito, Arch. Biochem. Biophys. 322

(1995) 500.13] B.L. Martin, D. Wu, S. Jack, J. Biol. Chem. 265 (1990) 7108.14] N. Fukuyama, Y. Takebayashi, M. Hida, Free Radic. Biol. Med. 22 (1997)

771.15] C. Szado, Shock 6 (1996) 79.

[

[[

72 (2007) 1283–1287 1287

16] M.A. Packer, J.L. Scarlett, S.W. Martin, Gen. Pharm. 31 (1998) 179.17] A. Vandervlite, J.P. Eiserich, C.A. Oneill, Arch. Biochem. Biophys. 319

(1995) 341.18] K. Dai, A.G. Vlessidis, N.P. Evmiridis, Talanta 59 (2003) 55.19] S.B. Digerness, K.D. Harris, J.W. Kirklin, Free Radic. Biol. Med. 27 (1999)

1386.20] J. Xue, X. Ying, J. Chin, Anal. Chem. 72 (2000) 5313.21] P. Wang, J.L. Zweier, J. Biol. Chem. 271 (1996) 29223.22] S. Dikalov, M. Skatchkov, B. Fink, Nitric Oxide 1 (1997) 423.23] S. Dikalov, M. Skatchkov, E. Basseneg, Biochem. Biophys. Res. Commun.

230 (1997) 54.24] J. Glebska, W.H. Knoppenol, Free Radic. Biol. Med. 35 (2003) 678.25] F.J. Martin-Romero, Y.G. Martin, F. Henao, J. Fluorescence 14 (2004)

17.26] S.L. Hempel, G.R. Buettner, D.A. Wessels, Free Radic. Biol. Med. 27

(1999) 146.27] N.W. Kooy, J.A. Royall, H. Ischiropoulos, Free Radic. Biol. Med. 16 (1994)

149.28] H. Possel, H. Noack, W. Augustin, FEBS Lett. 416 (1997) 175.29] X.F. Yang, X.Q. Guo, Y.B. Zhao, Talanta 57 (2002) 883.30] O. Stanger, Curr. Drug Metab. 3 (2002) 211.31] M. Basshir, FEBS Lett. 555 (2003) 601.32] R. Joshi, S. Adhikari, B.S. Patro, Free Radic. Biol. Med. 30 (2001)

1390.33] J.S. Beckman, T.W. Beckman, J. Chen, Proc. Natl. Acad. Sci. U.S.A. 87

(1990) 1620.34] G.T. Vatassery, Biochem. Pharmacol. 52 (1996) 579.35] J.M. Souza, E. Daikhin, M. Yudkoff, Arch. Biochem. Biophys. 371 (1999)

169.36] A. Daiber, M. Bachschmid, J.S. Beckman, Biochem. Biophys. Res. Com-

37] P. Salvemini, Z.Q. Wang, M.K. Stern, Proc. Natl. Acad. Sci. U.S.A. 95(1998) 2659.

38] H. Ischiropoulos, J.S. Beckman, J.P. Crow, Methods 7 (1995) 109.39] J. Liang, Z.H. Liu, R.X. Cai, Anal. Chim. Acta 530 (2005) 317.