Forum: Oxidative Stress Status

SENSITIVE AND NONENZYMATIC MEASUREMENT OF HYDROGENPEROXIDE IN BIOLOGICAL SYSTEMS

SEBASTIAN MUELLER

Department of Internal Medicine IV, University of Heidelberg, Heidelberg, Germany

(Received3 August1999;Revised7 February2000;Accepted15 March 2000)

Abstract—The increasing demand in detecting H2O2 under various experimental conditions is only partly fulfilled bymost conventional peroxidase-based assays. This article describes a sensitive and nonenzymatic H2O2 assay that is basedon the chemiluminescence reaction of luminol with hypochlorite. It allows the determination of H2O2 down tonanomolar concentrations. Actual H2O2 concentrations rather than a turnover of H2O2 can be determined in monolayercultures, perfusates, suspensions of intact cells, organelles, and crude homogenates. One of the strengths of this assayis that it may be used to assess fast enzyme kinetics (catalase, glutathione peroxidase, oxidases) at very low H2O2

concentrations. Its use together with a glucose oxidase/catalase system appears to be a powerful tool in studying signalfunctions of H2O2 in various biological systems on a quantitative basis. Several applications are discussed in detail todemonstrate the technical requirements and analytical potentials. © 2000 Elsevier Science Inc.

Keywords—Hydrogen peroxide, Free radicals, Catalase, Oxidase, Glutathione peroxidase, Chemiluminescence, Lu-minol, Glucose oxidase

INTRODUCTION

H2O2 is a central oxygen metabolite, produced in severalcellular compartments and the source of other reactiveoxygen species, e.g., the highly reactive hydroxyl radi-cal. Besides its role in cellular toxicity, H2O2 has re-cently gained much attention as a possible signalingmolecule involved in signal transduction pathways [1–3].The increasing need to detect H2O2 under several exper-imental conditions is only partly fulfilled by conven-tional peroxidase-based assays using different probes aselectron donors. These assays may remove or generate

reactive oxygen species themselves and, therefore, giverise to severe artifacts [4,5]. This article describes anonenzymatic H2O2 assay that has been recently appliedto various biological systems [5]. Advantages of thisassay are that it may be used (i) to assess fast enzymatickinetics entailed in catalase-, glutathione peroxidase–,and oxidase-catalyzed reactions at low H2O2 concentra-tions; (ii) to measure actual H2O2 concentrations inmonolayer cultures and perfusates; and (iii) to estimatediffusion barriers for H2O2 [5–9]. A computer-drivenchemiluminometer, an apparatus routinely found in bio-medical laboratories, is required for these measurements.

PRINCIPLE

The assay is based on the oxidation of luminol (5-amino-2,3-dihydro-1,4-phthalazinedione) by sodium hy-pochlorite (NaOCl) [5]. Luminol is oxidized by NaOClto diazaquinone in a two-electron oxidation, which isfurther specifically converted by H2O2 to an excitedaminophthalate via ana-hydroxy-hydroperoxide [5,10,11]. The short luminescence signal (less than 2 s) of thisreaction has a maximum wavelengh at 431 nm; it islinearly dependent on H2O2 down to the 1029 M range.One-electron transfers inherent in peroxidase-based as-

Sebastian Mueller: Born 1967 in Dresden (Germany). Medical stud-ies in Leipzig and Strasbourg. MD-PhD 1994 in Leipzig. From 1994–1997 clinical specialization and research at Department of Gastroen-terology/University of Heidelberg with W. Stremmel. Visiting scholar1992 at Chemistry Department/University of Denver with G. Eaton and1997–1998 at Department of Molecular Pharmacology and Toxicolo-gy/ University of Southern California in Los Angeles with E. Cadenasas Feodor-Lynen fellow of the Alexander-von-Humboldt foundation.S.M. is now at the Dept. of Gastroenterology/University of Heidelberg.Research interests: biological functions of hydrogen peroxide, hydro-gen peroxide metabolism, iron regulation and oxidative stress, oxida-tive metabolism of peroxisomes.

Address correspondence to: Dr. Sebastian Mueller, University ofHeidelberg, Department of Internal Medicine IV, Bergheimer Strasse58, 69115 Heidelberg, Germany; Tel:149 (6221) 56 8612; Fax:149(6221) 40 8366; E-Mail: sebastian.mueller@urz.uni-heidelberg.de.

Free Radical Biology & Medicine, Vol. 29, No. 5, pp. 410–415, 2000Copyright © 2000 Elsevier Science Inc.Printed in the USA. All rights reserved

0891-5849/00/$–see front matter

PII S0891-5849(00)00261-6

410

says, which may cause redox-cycling reactions or reduc-tion of oxygen to superoxide [11–13], are avoided by thismethod. The luminol/hypochlorite-dependent chemilu-minescence exceeds by far the unspecific so-called lu-minol-dependent chemiluminescence (generated byother oxidation pathways) [14]. Additionally, the assayprocedure permits a simple subtraction of unspecificsignals [5]. When used as a flow system, rapid kinetics ofH2O2-removing or -generating enzymes (catalase, GPO,oxidases) can be studied at physiologically low H2O2

concentrations. Due to the short time of measurement, itactually determines the H2O2 (M) concentration ratherthan an H2O2 generation rate (mol per time).

DETERMINATION OF FAST H 2O2 KINETICS IN

ENZYMATIC REACTIONS USING A FLOW SYSTEM

The sample is continuously pumped out from a reac-tion reservoir and luminol and NaOCl are continuouslyadded to the sample allowing a real-time registration ofH2O2, e.g., during fast enzyme kinetics. One of theadvantages is that the sample in the reservoir is not incontact with any of the reagents. The procedure requiresa large sample volume (up to 100 ml). This is usually nota problem since enzyme solutions can be highly diluteddue to the sensitivity of the assay.

Equipment

The procedure requires the following equipment:

1) chemiluminometer—any luminometer that allows theinstallation of a flow cell in front of the photomulti-plier (e.g., the AutoLumat LB 953 from BertholdEG&G; Wildbad, Germany) can be used; the lumi-nometer is controlled by a computer equipped withsoftware for further processing of time/luminescenceintensity data;

2) perfusion pump for NaOCl and luminol;3) peristaltic pump for sample aspiration;4) flow cell—a flow cell that allows separate and contin-

uous addition of luminol and NaOCl buffered in PBS atpH 7.4, and the continuous addition of the sample isrequired; in the author’s laboratory a peristaltic pump isused with a 3 mmpolyethylene pipeline to continuouslyaspirate the sample solution (ca. 4 ml/min); black 50 mlplastic syringes are loaded with luminol and NaOClwork solutions and both reactions are continuouslypumped via the same perfusion pump into the polyeth-ylene pipeline (ca. 12 ml/h);

5) graduated cylinder of 100 ml for sample solution;6) magnetic stirrer to continuously mix sample solution;

and7) temperature control unit (if necessary).

Reagents

The procedure requires the following reagents:

1) 50 ml 1024 M luminol in 10 mM PBS at pH 7.4(working solution);

2) 50 ml 1024 M NaOCl in tridistilled water (workingsolution); and

3) 100 ml 1022 M H2O2 in tridistilled water for calibra-tion.

Procedure

The syringes are loaded with the working solutions ofNaOCl and luminol and 50–100 ml PBS is added intothe graduated cylinder. All pumps are switched on andthe system is allowed to equilibrate for about 5 min. Theoptimal measuring range is found by adjusting the per-fusion pump and calibrated by addition of 1025 M H2O2

and catalase, respectively.

Example 1: determination of catalase activity atphysiological H2O2 concentrations [6]

The sample (tissue homogenate, intact cells, purifiedenzyme) is added into 50 ml 10 mM PBS (graduatedcylinder) containing 1025 H2O2. A magnetic stirrer isused to ensure a rapid mixing of the enzyme substratesolution. The exponential, catalase-mediated decomposi-tion of H2O2 can be followed down to 1029 M H2O2 (seeFig. 1). Catalase activity is described by the rate constantk 5 ln (S1/S2)/dt where dt is the measured time interval,and S1 and S2 are H2O2 concentrations at time t1 and t2,

Fig. 1. Parallel determination of catalase and glutathione peroxidase(GPO) activities in a hemolysate at very low H2O2 concentrations.Addition of highly diluted hemolysate (1:1000) is followed by thecatalase-mediated exponential degradation of H2O2. Catalase is subse-quently inhibited by sodium azide (1 mM), which can be demonstratedby further incubation with H2O2. Finally, glutathione (2 mM) is added.This step requires a recalibration of the system. After a new bolus ofH2O2 (1024 M) the GPO-mediated decay of H2O2 is visible. Both,catalase and GPO activity can be calculated from this experiment.

411Sensitive H2O2 measurement

respectively. The first-order decay of H2O2 can be fol-lowed over three orders of magnitude and k is subse-quently calculated by linear regression analysis (using asimple curve fitting program). In the above equation, theratio (S1/S2) rather than absolute values of H2O2 concen-trations is important so that k can be calculated directlyfrom the luminescence intensities k5 ln (I1/I2)/dt wheredt is the measured time interval, and I1 and I2 are lumi-nescence integrals at time t1 and t2. The constant k can beused as a direct measure of catalase concentration. Thespecific catalase activity k’1 is obtained by dividing k bythe molar concentration of catalase (e): k’1 5 k/e. k’1 isknown for many catalases from different cell types. Thevalue k’1 for purified catalase from human erythrocytesis 3.43 107 M21s21. This value is used to calculate theabsolute concentration of enzyme in blood and tissues[15].

The hypochlorite/luminol technique provides severaladvantages in comparison to conventional spectrophoto-metric and titrimetric catalase assays: (i) due to the lowH2O2 concentrations used, molecular oxygen is com-pletely dissolved and not liberated in gaseous form; (ii)since maximal extracellular H2O2 concentrations areknown to reach only micromolar levels, determinationsof catalase activity at submicromolar concentrationsmuch better reflect physiological conditions; and (iii)repetitive measurements for more than 30 min are pos-sible without loss of enzyme activity and cell viability.The assay has been successfully used to compare cata-lase activity of intact and homogenated cells/organelles[6]. From these data, the diffusion coefficient for H2O2

can be calculated with respect to different membranes[16].

Example 2: parallel determination of catalase andglutathione peroxidase activity in human erythrocytes[8]

This procedure allows the determination of both en-zyme activities at low, noninactivating H2O2 concentra-tions and can be applied to all crude cell and tissuehomogenates. Highly diluted fresh hemolysate (1:1000)is added into a graduated cylinder containing 50 ml of1025 M H2O2 in PBS (Fig. 1). The solution is perma-nently stirred. The high dilution insures a decrease in theconcentration of glutathione to undetectable concentra-tions. After addition of the diluted hemolysate, an expo-nential decay of H2O2 is observed corresponding tocatalase activity. Catalase activity is then calculated asthe rate constant k of the exponential decay of H2O2 bylinear regression analysis as described above [6,15]. Inthe next step, catalase is inhibited by addition of 1 mMNaN3 and 2 mM GSH is added. This amount representsthe intracellular GSH concentration in erythrocytes. The

system is recalibrated by addition of 1024 M H2O2 andGPO-mediated H2O2 decay is observed. Based on theping-pong kinetics with infinite limiting maximum ve-locities and Michaelis-Menton constants established forGPO, the maximum velocity needs to be determined forall individual conditions [17].

Example 3: steady state generation of H2O2 with aglucose/glucose oxidase/catalase system to studyH2O2-dependent signaling pathways [7,9]

The glucose/glucose oxidase system appears to be apowerful tool in studying signal functions of H2O2 on aquantitative basis. During the oxidation of glucose byglucose oxidase, H2O2 is generated following a zero-order kinetic with dH2O2/dt 5 kGOX (kGOX 5 rate con-stant) if dioxygen and glucose are maintained at a con-stant concentration. Accumulation of H2O2 can becontrolled by adding appropriated amounts of catalase.H2O2 degradation rate by catalase is described bydH2O2/dt 5 kCAT 3 [H2O2]. Thus, steady state levels ofH2O2 are generated when kGOX 5 kCAT 3 [H2O2], andat a constant glucose and dioxygen concentration [H2O2]5 kGOX/kCAT. By varying the enzyme activities, theH2O2 concentration can be adjusted and maintained overhours. The luminol/hypochlorite assay assists by mea-suring this steady state as shown in Fig. 2. Steady stategeneration in turn allows exact time and dose-dependentstudies on redox-sensitive signaling pathways instead ofsimply adding H2O2 as bolus. The GOX/catalase systemwas successfully used to study the regulation of ironprotein 1 (IRP-1) by H2O2. It was shown that 10mMH2O2 (steady state) suffice to activate IRP-1 within 20min by a still unknown signaling cascade [7,9,18].

Fig. 2. Generation of H2O2 steady state concentrations with catalaseand glucose/glucose oxidase (GOX). The steady state concentration ofH2O2 is determined by the GOX/catalase ratio and can be maintainedover hours. This tool appears to be very useful in studying signalfunctions of H2O2 on a quantitative basis in various biological systems.

412 S. MUELLER

SINGLE TIME POINT DETERMINATION OF H 2O2 USING

AN INJECTION SYSTEM

In this procedure, luminol is premixed with the sam-ple (e.g., culture medium or perfusate sample). At theappropriate time, NaOCl is added and the luminescenceintensity is measured immediately. The measurement isfast and only a small sample volume is needed. Aninjection device in measuring position is a requisite con-dition because the luminescence reaction reaches com-pletion within less than 2 s. As an advantage, samplematerials are saved and the procedure can be fully auto-mated.

Equipment

The procedure requires the following equipment:

1) chemiluminometer with injection device in measur-ing position (e.g., AutoLumat LB 953 from BertholdEG&G; Wildbad, Germany); other injection devicesare helpful for complete automatization of the exper-iment (e.g., addition of cell stimulators); the lumi-nometer should be controlled by a computer equippedwith software allowing automated performance; and

2) polystyrene tubes for chemiluminometer.

Reagents

The procedure requires the following reagents:

1) stock solution of 1023 M luminol in 10 mM PBS atpH 7. 4 (final concentration of luminol between 1025

and 1024 M);2) stock solution of 1024 M NaOCl in tridistilled water

(final concentration of NaOCl between 1026 and1025 M); and

3) 1023 M H2O2 in tridistilled water for calibration.

Procedure

The injector in measuring position is loaded withNaOCl solution and washed. For optimal measuringrange, samples with PBS and luminol are loaded con-taining catalase (e.g., 1027 M final concentration) and1025 or 1026 M H2O2. If necessary, the NaOCl concen-tration needs to be adjusted. In a typical experiment, theinjection device adds 50ml of NaOCl (1026–1025 Mfinal concentration) into 950ml sample with luminol(5 3 1025 M final concentration). Usually, samples aremeasured together with an H2O2 calibration solution atthe beginning and the end of the batch.

Determination of H2O2 release by stimulatedneutrophils [5]

In the experiment shown in Fig. 3, 10ml of luminol

stock solution was added into 20 polystyrene tubes con-taining 1 ml of 105 PMN/ml in HANK’s buffer. Everysecond sample contains sodium azide (1 mM final con-centration). In a fully automated experiment, the neutro-phils are stimulated by addition of 10mM FMLP. Atdifferent time points, 50ml NaOCl is added (53 1026 Mfinal concentration). In parallel, the luminescence inten-sity is recorded for 2 s. The system is calibrated usingknown amounts of H2O2. A rapid increase of H2O2 isobserved after stimulation of neutrophils. H2O2 is laterremoved by myeloperoxidase and catalase, which can beinhibited by sodium azide. Using this method, changes ofH2O2 concentration could be detected with less than3000 neutrophils/ml.

H2O2 removal from culture medium of B6 fibroblasts[7,9]

Figure 4 shows the removal of H2O2 from culturemedium (three different volumes) by fibroblasts growingin monolayer culture. B6 fibroblasts were cultured inRPMI medium in 10 cm culture dishes at 37°C. 1024 MH2O2 (final concentration) is added to the medium and500 ml of the medium is transferred at different timepoints into polystyrene tubes. 10ml luminol stock solu-tion is added. After the addition of 50ml NaOCl (5 31026 M final concentration), luminescence intensity isrecorded for 2 s. The culture medium should be usedwithout serum addition. The determination of H2O2 re-moval in cell culture medium is required at all conditionswhere H2O2 is applied to cells to study its signal func-tions.

GENERAL COMMENTS

The assay is very sensitive and detects H2O2 at con-centrations as low as 1029 M. It should be noted that

Fig. 3. Increase of H2O2 concentration in suspension of neutrophils(oxygen burst) after stimulation with 1mM FMLP (arrow). Addition ofsodium azide inhibits cellular myeloperoxidase and catalase and sub-sequently leads to H2O2 accumulation.

413Sensitive H2O2 measurement

normal tridistilled water already contains traces of H2O2

sometimes as high as up to 1027 M. This is one of thereasons to calibrate the assay with at least 1026 M H2O2

and to add catalase. The samples should be kept in thedark since photoreactions generate H2O2 (e.g., UV light).Any unspecific chemiluminescence can be detected uponH2O2 removal with catalase. The sample volume shouldbe at least 10 times higher than the volume of reagents,because small contamination of reagents with H2O2 cansignificantly decrease the sensitivity of the assay.

NaOCl working solutions should be prepared withtridistilled water to minimize its degradation and theyshould be freshly prepared and kept in the dark. For eachsystem, the optimal NaOCl concentration should be de-tected separately. Too high NaOCl concentrations favorunspecific oxidation reactions of luminol. Therefore,NaOCl concentration should be chosen as low as possi-ble (usually 1026–1025 M). Any compound containing,for example, sulfhydryl- or aminogroups will competewith luminol for NaOCl leading to a decrease in sensi-tivity [10]. In these cases, samples are diluted (e.g.,whole blood 1:100 to 1:10,000) and/or higher concentra-tions of NaOCl should be used. pH is critical for thereaction and should be kept stable at a value of 7.4 withany appropriate buffer. No luminescence develops atvalues below 6.5. The chemiluminescence duration in-creases with pH above 7.4.

Troubleshooting should include: (i) correct concentra-tions of reactants, (ii) proper installation of the flow cellin front of the photomultiplier, (iii) proper injection ofNaOCl in front of the photomultiplier, (iv) low content ofNaOCl-reactive compounds (e.g., solutions of sulfhydryl

group–containing proteins should be diluted at least toless than 1026 M), and (v) stability of NaOCl solutions.

CALIBRATION

The luminol/hypochlorite assay is calibrated withknown concentrations of H2O2 usually between 1023–1025 M taken from stock solutions. Commercial stocksolutions of 30% H2O2 are stable for many weeks oncekept at 4°C and in the dark. Commercial stock solutionsof NaOCl are also stable at 4°C and in the dark. Stocksolutions of NaOCl and H2O2 can be determined spec-trophotometrically ate2905 350 M21cm21 at pH 12 ande230 5 74 M21cm21, respectively [19,20]. A routinecalibration for established conditions requires at leastone sample with a known H2O2 concentration and asample without H2O2 upon removal by catalase. Thesemeasurements provide the available measuring range. Ifnecessary, the hypochlorite/luminol assay allows thesubtraction of the unspecific luminol-dependent chemi-luminescence. It simply needs to be measured prior toNaOCl addition and subtracted from the overall lumines-cence intensity. The same is valid for any non-H2O2-related luminescence that is detected after addition ofcatalase.

SUMMARY

This article describes a sensitive and nonenzymicH2O2 assay that is based on the chemiluminescencereaction of luminol with hypochlorite. Actual H2O2 con-centrations can be measured in monolayer cultures, per-fusates, suspensions of intact cells, and crude homoge-nates. One of the strengths of this assay is that it may beused to assess fast enzyme kinetics at very low H2O2

concentrations. The luminol/hypochlorite assay opensnew niches in studying the functions of H2O2 in biolog-ical systems and its metabolism.

Acknowledgements— This work was supported financially by a Fe-odor-Lynen fellowship of the Alexander-von-Humboldt foundation andby a grant from the Deutsche Forschungsgemeinschaft (SFB601/C2D.10049500).

REFERENCES

[1] Sen, C. K.; Sies, H.; Baeurle, P. A., eds.Antioxidant and redoxregulation of genes.San Diego: Academic Press; 2000.

[2] Chance, B.; Sies, H.; Boveris, A. Hydroxyperoxide metabolism inmammalian organs.Physiol. Rev.59:527–605; 1979.

[3] Khan, A. U.; Wilson, T. Reactive oxygen species as cellularmessengers.Chem. Biol.2:437–445; 1995.

[4] Misra, H. P.; Squatrito, P. M. The role of superoxide anion inperoxidase-catalyzed chemiluminescence luminol.Arch. Bio-chem. Biophys.215:59–65; 1982.

[5] Mueller, S.; Arnhold, J. Fast and sensitive chemiluminescence

Fig. 4. H2O2 removal by cultured B6 fibroblasts (monolayer) in differ-ent volumes of culture medium. After bolus addition, H2O2 is rapidlydecomposed by the cells within minutes. This instability needs to beconsidered when studying the effects of H2O2 on cellular functions.

414 S. MUELLER

determination of H2O2 concentration in stimulated human neu-trophils.J. Biolumin. Chemilumin.10:229–237; 1995.

[6] Mueller, S.; Riedel, D. H.; Stremmel, W. Determination of cata-lase activity at physiological H2O2 concentrations.Anal. Bio-chem.245:55–60; 1997.

[7] Mueller, S.; Pantopoulos, K.; Hentze, M. W.; Stremmel, W. Achemiluminescence approach to study the regulation of iron me-tabolism by oxidative stress. In: Hastings, J. W.; Kricka, L. J.;Stanley, P. E., eds.Bioluminescence and chemiluminescence:molecular reporting with photons.Baffins Lane, Chichester, Sus-sex: John Wiley & Sons Ltd.; 1997:338–341.

[8] Mueller, S.; Riedel, H. D.; Stremmel, W. Direct evidence forcatalase as the predominant H2O2-removing enzyme in humanerythrocytes.Blood 90:4973–4978; 1997.

[9] Pantopoulos, K.; Mueller, S.; Atzberger, A.; Ansorge, W.; Strem-mel, W.; Hentze, M. W. Differences in the regulation of IRP-1(iron regulatory protein 1) by extra- and intracelullar oxidativestress.J. Biol. Chem.272:9802–9808; 1997.

[10] Arnhold, J.; Mueller, S.; Arnold, K.; Sonntag, K. Mechanisms ofinhibition of chemiluminescence in the oxidation of luminol bysodium hypochlorite.J. Biolumin. Chemilumin.8:307–313; 1993.

[11] Merenyi, G.; Lind, J.; Eriksen, T. E. Luminol chemiluminescence:chemistry, excitation, emitter.J. Biolumin. Chemilumin.5:53–56;1990.

[12] Merenyi, G.; Lind, J.; Eriksen, T. E. The reactivity of superoxide(O2

2) and its ability to induce chemiluminescence and luminol.Photochem. Photobiol.41:203–208; 1985.

[13] Faulkner, K.; Fridovich, I. Luminol and lucigenin as detectors forO2

2. Free Radic. Biol. Med.15:447–451; 1993.

[14] Allen, R. C.; Loose, L. D. Phagocytic activation of a luminol-dependent chemiluminescence in rabbit alveolar and peritonealmacrophages.Biochem. Biophys. Res. Commun.69:245–252;1976.

[15] Aebi, H. Catalase in vitro.Methods Enzymol.105:121–126; 1984.[16] Nicholls, P. Activity of catalase in the red cell.Biochim. Biophys.

Acta 99:286–297; 1965.[17] Flohe, L.; Gunzler, W. Assays of glutathione peroxidase.Methods

Enzymol.105:114–121; 1984.[18] Pantopoulos, K.; Hentze, M. W. Activation of iron regulatory

protein 1 by oxidative stress in vitro.Proc. Natl. Acad. Sci. USA95:10559–10563; 1998.

[19] Beers, R. F.; Sizer, I. W. A spectrometric method for measuringthe breakdown of hydrogen peroxide by catalase.J. Biol. Chem.195:133–140; 1952.

[20] Morris, J. C. The acid ionization constant of HOCl from 5 to 35°.J. Phys. Chem.70:2798–3806; 1966.

ABBREVIATIONS

GSH—glutathioneGPO—glutathione peroxidaseNaOCl—sodium hypochloriteIRP-1—iron regulatory protein-1FMLP—N-formyl-methionine-leucine-phenylalanine

(chemotactic tripeptide)

415Sensitive H2O2 measurement