Journal of Inorganic Biochemistry 130 (2014) 84–91

Contents lists available at ScienceDirect

Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio

(−)-Epicatechin regenerates the chlorinating activity of myeloperoxidase in vitro and in neutrophil granulocytes Jörg Flemmig a,b,⁎,1, Johannes Remmler a,1, Fiete Röhring a, Jürgen Arnhold a,b a b

Institute for Medical Physics and Biophysics, Medical Faculty, University of Leipzig, Haertelstrasse 16–18, 04107 Leipzig, Germany Translational Centre for Regenerative Medicine, University of Leipzig, Philipp-Rosenthal-Strasse 55, 04103 Leipzig, Germany

a r t i c l e

i n f o

Article history: Received 20 July 2013 Received in revised form 2 October 2013 Accepted 2 October 2013 Available online 11 October 2013 Keywords: Aminophenyl fluorescein Peroxidases Compound II accumulation Myeloperoxidase Neutrophils (−)-epicatechin

a b s t r a c t The heme-containing enzyme myeloperoxidase (MPO) is mainly expressed in polymorphonuclear leukocytes (PMNs), the most abundant immune cell type in the blood. Accordingly, MPO is classically attributed to the innate immune response against pathogens. Yet, new results also show immune-regulatory functions of the halogenating MPO activity including the formation of anti-inflammatory mediators. In this work we tested the ability of the flavonoid (−)-epicatechin to regenerate this enzymatic activity both in vitro at the isolated MPO–H2O2–Cl− system and ex vivo in human PMNs. For all experiments the non-fluorescent dye aminophenyl fluorescein (APF) was used. Upon oxidation by the MPO, the halogenation product hypochlorous acid (HOCl) fluorescein is formed which can be detected e.g. by flow cytometry. The in vitro- and ex vivo-results concordantly show that (−)-epicatechin is a suitable substrate to overcome a compound II accumulation of MPO which was experimentally forced by applying excess hydrogen peroxide. Thereby concentration-dependent effects of the flavan-3-ol were found in both cases and confirmed the proposed mode of action of (−)-epicatechin. The results are in accordance with previous stopped-flow kinetic studies which showed a high reactivity of the polyphenol with MPO compound II. The obtained data may contribute to the explanation of the well-known health promoting effects of (−)-epicatechin. Moreover, the presented study provides new insights into the role of MPO during inflammation. © 2013 Published by Elsevier Inc.

1. Introduction The function of myeloperoxidase (MPO) is classically attributed to the innate host response against pathogens as this heme peroxidase is mainly expressed in myeloid immune cells, namely polymorphonuclear leukocytes (PMNs, also called neutrophils) and, to a minor degree, in monocytes [1,2]. Furthermore, at neutral pH values MPO twoelectronically oxidizes chloride to form the bactericidal product hypochlorous acid (HOCl), a strong oxidant which is also responsible for MPO-derived tissue modifications at chronic and/or severe inflammatory conditions [1,3,4]. Yet, more recent studies also indicate immune-regulatory functions of HOCl like the limiting of the bactericidal PMN activity, apoptosis induction in these cells and the inhibition of pro-inflammatory pathways [5–7]. Thus it can be assumed that the HOCl-producing activity of MPO during acute inflammation helps to limit pathological side effects and

⁎ Corresponding author at: Institute for Medical Physics and Biophysics, Medical Faculty, University of Leipzig, Haertelstrasse 16–18, 04107 Leipzig, Germany. Tel.: +49 341 9715772; fax: +49 341 9718959. E-mail address: joerg.fl[email protected] (J. Flemmig). 1 These authors contributed equally to this work. 0162-0134/$ – see front matter © 2013 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jinorgbio.2013.10.002

to down-regulate the immune response [8]. In contrast, under chronic inflammatory conditions the ongoing PMN recruitment/activation and subsequent cell necrosis lead to MPO-catalyzed pro-inflammatory processes and HOCl-derived tissue destructions [1]. The formation of hypochlorous acid from chloride (halogenation cycle) is catalyzed by the MPO redox intermediate compound I which is formed upon reaction of the resting enzyme with hydrogen peroxide [9]. As a strong oxidant compound I also one-electronically oxidizes several substrates [8]. Thereby the MPO redox intermediate compound II is formed which is converted back to the native enzyme by a second one-electron step (peroxidase cycle) [10,11]. The enzymatic cycles of MPO as well as the APF detection system used in this work are shown in Fig. 1. Yet, as compound II is a much weaker oxidant than compound I many substrates readily react with the latter but much slower with the former. This leads to an accumulation of compound II which is unable to oxidize chloride [12,13]. The half-life time of this MPO redox intermediate is estimated to be about 1 h [14]. An example of this mechanism is H2O2 which, at high concentration, forces a compound II accumulation [15]. An elevated production of hydrogen peroxide occurs upon pro-inflammatory PMN activation [16]. Another pathway for the compound II accumulation is the direct conversion of the native enzyme by peroxynitrite (ONOO−), which is also formed at sites of inflammation [17,18]. It can be hypothesized that especially at inflammatory conditions the halogenating MPO

J. Flemmig et al. / Journal of Inorganic Biochemistry 130 (2014) 84–91

85

Fig. 1. Enzymatic activity of MPO. The enzyme can be activated from its native ferric form by hydrogen peroxide. Thereby compound I is formed which can subsequently undergo both oneand two-electron redox reactions. In the halogenation cycle compound I is directly reduced back to native MPO by two electron oxidation of (pseudo-)halides. Upon oxidation of chloride, hypochlorous acid (HOCl) is formed which is able to oxidize the non-fluorescent dye APF to fluorescein. Yet, compound I is also able to oxidize excess hydrogen peroxide by abstracting one electron. By adding low amounts of (−)-epicatechin compound II can be one electronically reduced back to the native enzyme which restores the ability of the enzyme to produce HOCl.

activity is impaired by an increased accumulation of compound II. Consequently less HOCl is produced which may lead to impairments regarding the termination of acute inflammatory reactions. A further hint for the immune regulatory functions of the halogenating MPO activity comes from the fact that certain substances with well-known anti-inflammatory properties are also good substrates for compound II and thus regenerate native MPO [8]. In fact, for the reaction between (−)-epicatechin and compound II the highest second order rate constant known so far was determined (4.5 × 106 M−1 s−1 at pH 7.0 and 25 °C) suggesting this flavonoid as a good candidate for a MPO compound II resolution [19]. (−)-Epicatechin can be found e.g. in green tea and was shown to exhibit numerous beneficial health effects [20–22]. Subsequent in vivo-studies clearly showed regulatory properties of this flavonoid on the immunological activity of PMNs [23]. Accordingly, our in vitro-kinetic studies on isolated MPO clearly showed an enhancing effect of (−)-epicatechin on the halogenating enzyme activity [24]. Thereby we applied monochlorodimedon (MCD), a well-established dye for the spectrophotometric detection of HOCl [24,25]. Yet for HOCl detection in PMNs we used aminophenyl fluorescein (APF) which is well applicable for flow cytometric measurements [26,27]. In this paper we now used APF to investigate the effect of (−)-epicatechin on the isolated MPO–H2O2–Cl− system as well as on the chlorinating MPO activity in isolated human PMNs. Reactivating effects of this flavonoid regarding the HOCl-producing MPO activity would give new insights into the mode of action of this well-known anti-inflammatory polyphenol.

2. Material and methods 2.1. Materials Neutrophil MPO (E.C 1.11.2.2) was purchased from Planta GmbH, Vienna, Austria. APF was obtained from Biomol GmbH, Hamburg, Germany. All other chemicals were obtained from Sigma-Aldrich, Taufkirchen, Germany. Working solutions of H2O2 were prepared by dilution of a corresponding 30% stock solution. Their concentrations were determined using ε240 = 43.6 M−1 cm−1 [28]. The solutions were stored on ice and used within 3 h.

For the in vitro-experiments stock solutions of 100mM (−)-epicatechin (E4018, 98% purity, Sigma Aldrich) were prepared in DMSO/water (50/50 v/v) and further diluted in water. For the experiments on isolated PMNs 1 M (−)-epicatechin stock solutions were prepared in 100% DMSO. In both cases the final DMSO concentration did not exceed 0.013%. 2.2. Kinetic studies of the APF oxidation by MPO If not stated otherwise, the measurements were performed in 10mM phosphate buffered saline (PBS) at pH 7.4. For each measurement 10 nM MPO were pre-incubated with 10 μM aminophenyl fluorescein (APF) either in the presence or absence of (−)-epicatechin (0 to 32 μM) at 37 °C. To start the chlorinating activity H2O2 (0 to 250 μM) was added to the samples via an injection device and the formation of fluorescein from APF was detected within the next 15 to 30 min. All measurements were performed using a fluorescence microplate reader Tecan Infinite 200 PRO, Männedorf, Switzerland. The fluorescence intensity was monitored at 522 nm using an excitation at 488 nm. Control experiments were performed either in the absence of MPO, H2O2, APF, or chloride or in the additional presence of 4-aminobenzoic acid hydrazide (4-ABAH). This compound is a well-known MPO inhibitor [29]. All measurements were performed at least in triplicate. From the obtained kinetic data an averaged curve was created for each sample composition. Afterwards these data were fitted to an exponential function using the equation y = A1 − A2e−kx (exponential function “MnMolecular1”, Origin Pro 8G SR2, OriginLab Corporation, Northampton, MA, USA). Thereby y corresponds to the observed fluorescence intensity value (arbitrary units, a.u.) at time point x (s). If not stated otherwise, the data used for the curve fitting correspond to the data recorded between 0.5 and 6.5 min after the addition of H2O2. The value for the observed rate constant (kobs, s−1) obtained from the curve fits reflects the rate of the APF oxidation by HOCl and was used as a parameter for the chlorinating activity of MPO. The standard error (SE) given in the corresponding results section reflects the deviation of the fitted exponential function from the averaged kinetic curve of the experimental data. In the supplemental tables with the complete fitting parameters the number of iterations (I) used for the curve fittings is also given.

86

J. Flemmig et al. / Journal of Inorganic Biochemistry 130 (2014) 84–91

2.3. Kinetic studies on the reaction between APF and HOCl Additional experiments were performed for determining the second order rate constant for the oxidation of APF by HOCl. Briefly, in 10 mM PBS final concentrations of 200–1000 nM HOCl were added to 10 nM APF at 37 °C. Afterwards the fluorescence intensity increase at 522 nm (excitation at 488 nm) was monitored for 1 min at time intervals of 2 s. From the observed curves kobs values were calculated and plotted again the HOCl concentration. The concentration of the HOCl stock solution was determined spectroscopically immediately before the experiment by using ε290 = 350 M−1 cm−1 for −OCl [30]. 2.4. Isolation of human neutrophils Neutrophils were isolated from heparinized peripheral human blood of healthy volunteers [31]. Briefly, dextran-enhanced sedimentation, Ficoll Histopaque density gradient centrifugation, and hypotonic lysis of residual erythrocytes were subsequently applied. Cell purity was tested by applying CD16-FITC antibodies and flow cytometry. Cell vitality was detected by applying the dye 5,5′,6,6′-tetrachloro1,1′,3,3′-tetraethylbenzimidazolylcarbo-cyanine iodide (JC-1) and flow cytometry [32,33]. In further experiments only cell cultures with purity as well as vitality values higher than 90% were used. Cells were incubated at 37 °C, 95% humidity, and 5% CO2 content in Ca2+-containing HBSS. 2.5. Flow cytometry of APF-stained neutrophils For the experiments 106 PMNs/ml (in Hanks balanced salt solution, HBSS, supplemented with Ca2 + and adjusted to pH 7.4) were first pre-incubated with 5 μM APF for 30min at 37 °C. Afterwards H2O2 (0–5mM) was added and the cells were incubated another 60min at 37 °C. In control experiments the catalase inhibitor 1,2,4-triazole (10 mM) or the MPO inhibitor 4-ABAH (500 μM) was added to the samples 15 min before APF addition. If (−)-epicatechin was included in the experiment, the polyphenol (0–128 μM) was added to the samples 15min after H2O2 stimulation. All concentrations are final ones. Flow cytometry measurements were performed using a FACSCalibur flow cytometer, BD Biosciences, Franklin Lakes, NY, USA supplied with a 488 nm argon laser. For the detection of the APF fluorescence channel 1 (530 ± 15 nm) was used. In each measurement 104 events were detected. The software Flowing Software 2.4.1 by Perttu Terho was used for data representation. For statistical analysis of the obtained fluorescence intensity distributions (histograms) the geometric mean was determined following the suggestions of comparable works [34,35]. At least three independent experiments were performed under identical experimental conditions and mean and standard deviation (s.d.) of the obtained values were calculated. In the graphs one asterisk (*) indicates weak significant effects (p b 0.05) as compared to the control. Two asterisks (**) show significant effects (p b 0.01) and three asterisks (***) indicate strong significant (p b 0.001) changes versus control. Significance was tested using the program Origin Pro 8G SR2 by applying a paired t-test. 3. Results 3.1. Hydrogen peroxide inhibits the chlorinating activity of myeloperoxidase Upon addition of increasing concentrations of H2O2 the initial rate of APF oxidation by the MPO–H2O2–Cl− system continuously decreased (Fig. 2). In Fig. 2A representative examples for the kinetics of the fluorescence intensity increase at 522 nm in the presence of different H2O2 concentrations are shown. By increasing the amount of H2O2 from 20 μM (black line) to 250 μM (lightest grey line), a unidirectional decrease in the rate of fluorescein formation was observed.

Fig. 2. Kinetics of the APF oxidation by MPO in the presence of H2O2. 10 μM APF were preincubated with 10 nM MPO and 140 mM Cl− at 37 °C and pH 7.4. After H2O2 addition (20–250 μM), the fluorescence intensity at 522 nm was recorded. In (A) examples for the time-dependent changes in the fluorescence intensity at different H2O2 concentrations are shown. At 20 μM H2O2 (black line), APF was completely and quickly oxidized leading to a final fluorescence intensity of about 26,500 a.u. after about 2.2 min. With increasing H2O2 concentrations (shown in brighter shades of grey), the APF oxidation rate continuously decreased. At H2O2 concentrations of 200 μM or higher also an incomplete APF oxidation was observed. The shown data are representative examples from three independent measurements. For every H2O2 concentration averaged kinetic curves were fitted to the exponential function y = A1 − A2e−kx. The plot of the observed k values against the H2O2 concentration (B) confirmed a H2O2-mediated inhibition of the chlorinating MPO activity. The SE as a sign for the deviation of the fitted functions from the observed kinetic data are too small to be visible.

By using H2O2 concentrations of 200 μM or higher, a significant decrease in the final fluorescence intensity (about 26,500 a.u. at 20 μM H2O2 and 19,100 a.u. at 250 μM H2O2, respectively) was also observed as well, suggesting an incomplete APF oxidation. In some cases (e.g. at 40 μM or 60 μM H2O2) the maximum fluorescence value reached after about 5 min considerably decreased till the end of the measurement after 30 min. Therefore, only the first 6.5 min of the measurement data was used for detailed kinetic analysis. For each hydrogen peroxide concentration at least three independent measurements were performed. The obtained kinetic curves were averaged and fitted to an exponential function (y = A1 − A2e−kx). As shown in Fig. 2B, the obtained kobs values confirmed a continuous decrease in the APF oxidation rate with increasing H2O2 concentrations. While at 20μM hydrogen peroxide a kobs value of 0.027s−1 was observed, at 150 μM H2O2 this value was about 4.5 times lower (0.006 s−1). With even higher concentrations of hydrogen peroxide the APF oxidation rate slightly slowed further down (kobs = 0.003 s−1 at 250 μM H2O2). As can be seen in Table S1 where the complete data from the described curve fits are given at hydrogen peroxide concentrations of 150 μM or higher, also the values A1 and A2 significantly decreased. This reflects well the already stated incomplete APF oxidation at very high H2O2 concentrations. As revealed by control experiments (not shown) in the absence of the enzyme, H2O2 or chloride as well as in the additional presence of 500 μM 4-ABAH (MPO inhibitor), no APF oxidation was observed. In the absence of APF no significant fluorescence signals were detectable at all (not shown). In conclusion, H2O2 is a suitable tool to inhibit the chlorinating MPO activity. To study whether this effect can be ascribed to a compound II

J. Flemmig et al. / Journal of Inorganic Biochemistry 130 (2014) 84–91

accumulation of the MPO, which may be resolved by (−)-epicatechin, all further in vitro kinetic studies were performed using H2O2 concentrations of 150 μM or 20 μM, respectively (highlighted in grey in Fig. 2B). 3.2. (−)-Epicatechin regenerates the chlorinating myeloperoxidase activity by resolving compound II In the presence of 150 μM H2O2 the slow oxidation of APF by the MPO–H2O2–Cl− system was strongly influenced by the additional presence of (−)-epicatechin (Fig. 3A). Thereby three phases can be distinguished: The addition of up to 3 μM of the polyphenol increased the rate of fluorescein formation by about factor 4.8 (kobs = 0.005 s−1 for 0 μM (−)-epicatechin and 0.024 s−1 for 3 μM (−)-epicatechin). With further increasing (−)-epicatechin concentrations up to 10 μM (kobs = 0.006 s−1) this effect was completely reversed. By applying even higher polyphenol concentrations (up to 32 μM were tested), only a moderate further decrease in the APF oxidation rate was observed. Yet, as can be seen in Table S2 (complete data from the applied curve fits regarding the effect of (−)-epicatechin at 150 μM H2O2), at 20 μM and higher polyphenol concentration the values for A1 and A2 decreased considerably. This reflects the observed incomplete APF oxidation observed under these conditions (lower final fluorescence intensity values). In contrast to this, the high oxidation rate of APF by the MPO–H2O2–Cl− system observed in the presence of 20 μM H2O2 (kobs = 0.029 s− 1) continuously decreased with increasing (−)-epicatechin concentrations (Fig. 3B). By applying 10 μM of the polyphenol a 4.1-fold lower kobs value (0.007 s− 1) was observed. With further increasing (−)-epicatechin concentrations this value did not decrease further.

87

Yet as can be seen in Table S3 (complete data from the curve fits) starting at 2 μM (−)-epicatechin, again the values for A1 and A2 continuously decreased reflecting lower final fluorescence intensity values of the kinetic curves. Therefore it can be concluded that in the presence of 150 μM H2O2 a strong compound II accumulation takes place which can be resolved by the addition of 3 μM (−)-epicatechin. By applying higher polyphenol concentrations, this promotion of the chlorinating MPO activity is reversed. This may be explained by an increasing reaction between (−)-epicatechin and compound I. In fact, at polyphenol concentrations of 15μM or higher an accelerated peroxidase cycle leads to a consumption of H2O2 without the formation of HOCl. In contrast at 20 μM H2O2 no significant compound II accumulation takes place. Therefore upon addition of increasing amounts of (−)-epicatechin the chlorinating activity of MPO compound I is increasingly disturbed leading to a continuous decrease in the APF oxidation rate. As a result a progressive drop in final fluorescence intensity was observed confirming a shift from the halogenation to the peroxidase activity of MPO. 3.3. Hypochlorous acid quickly converts APF to fluorescein We also conducted control experiments to verify that HOCl formed during the halogenating MPO activity instantly converts APF to fluorescein. As shown in Fig. S1A the addition of increasing amounts of HOCl to 10 nM APF led to accelerated increases in the fluorescence intensity. The observed kinetic curves were fitted to monoexponential functions. As shown in Fig. S1B the replot of the obtained kobs values against the actual HOCl concentration (200 nM to 1000 nM) led to a straight line (R2 = 0.93). From the slope a second order rate constant of 1.41 × 105 M−1 s−1 for the reaction between APF and HOCl was calculated. The slight deviation from the linear dependence observed at 1000 nM HOCl may be attributed to the fact, that under these conditions, the applied curve fits correspond to less data points due to the fast reached maximum under these conditions. Moreover at high excess of HOCl chlorination of fluorescein may occur, diminishing the observed fluorescence intensity [26]. 3.4. Hydrogen peroxide-induced myeloperoxidase activity in neutrophils

Fig. 3. Kinetics of the APF oxidation by MPO at 20 μM or 150 μM H2O2 in the presence of (−)-epicatechin. All other experimental conditions as well as the method for kobs value determination are as given in Fig. 1. For 150 μM H2O2 (A) the plot of the kobs values against the (−)-epicatechin concentration led to a strong increase in the APF oxidation rate up to 3 μM. Higher (−)-epicatechin concentrations (up to 10 μM) turned back this effect completely. With further increasing (−)-epicatechin concentrations, the calculated k values did not further drop. By applying 20 μM H2O2 (B) with increasing (−)-epicatechin concentrations, the initially high APF oxidation rate (kobs value) continuously decreased. The SE as a sign for the deviation of the fitted functions from the observed kinetic data are too small to be visible.

We next incubated freshly isolated PMNs with different hydrogen peroxide concentrations in order to evaluate conditions, where a strong compound II accumulation can be assumed while the cell vitality still remains unaltered. In the absence of hydrogen peroxide the APF-stained neutrophils exhibited a narrow fluorescence intensity distribution at low values (Fig. 4A). The single sharp peak at a fluorescence intensity of about 60 a.u. can be attributed to a basal HOCl producing MPO activity driven by intracellularly generated H2O2. The very small and broad fluorescence intensity maximum at about 8 a.u. can be assigned to the autofluorescence of the PMNs. The selected histogram shown in (Fig. 4B) refers to a sample where 1 mM H2O2 was added to the cells. Now, a broader fluorescence intensity distribution including considerably higher values was observed. This distribution reflects the differences of individual cells in responding to H2O2. In samples treated additionally with the MPO inhibitor 4-ABAH (500 μM) only the auto-fluorescence of the cells was observed, despite the presence of 1 mM H2O2 (Fig. 4C). This confirms a specific APF oxidation by MPO-derived HOCl. The histogram plot given in (Fig. 4D) shows the APF-derived fluorescence of PMNs treated with only 200μM H2O2 but in the additional presence of the catalase inhibitor 1,2,4-triazole (10 mM). The inhibition of catalase caused enhanced fluorescence values. This is quite comparable to the results with 1 mM H2O2 in the absence of 1,2,4-triazole. The representative examples shown in Fig. 4A–D correspond to a single PMN isolation. For statistical analysis of all samples the geometric mean values were determined

88

J. Flemmig et al. / Journal of Inorganic Biochemistry 130 (2014) 84–91

Fig. 4. Stimulation of APF-stained PMNs with H2O2. 106 cells were preincubated with 5 μM APF. After stimulation with H2O2 the cells were analyzed by flow cytometry. While in the absence of additional H2O2 (A) only a basal HOCl production occurred, with increasing H2O2 concentrations up to 1 mM (B), an increasing APF oxidation by activated cells was observed. The usage of 4-ABAH (C) confirmed the MPO dependence of this reaction as in the presence of the MPO-inhibitor the increased APF-oxidation was inhibited by about 92%. In the presence of the catalase inhibitor 1,2,4-triazole (D), a maximal HOCl production was observed at 200 μM H2O2. The shown histogram plots correspond to one representative example. For the analysis, the obtained geometric means for the fluorescence intensities were plotted against the H2O2 concentration (E). Mean and standard deviation of three independent experiments are displayed. While in the sole presence of H2O2 (squares) a maximal cell stimulation was observed using 1 mM H2O2, in the presence of the catalase inhibitor 1,2,4-triazole (circles) this value was reached at 200 μM H2O2. Upon inhibition of the MPO with 4-ABAH (triangles) no APF oxidation was observed.

from each recorded fluorescence intensity distribution and plotted against the hydrogen peroxide concentration (Fig. 4E). With increasing H2O2 concentrations (squares) a continuous rise of the APF-derived fluorescence intensity was observed. Yet, while between 0 μM and 100 μM H2O2 the geometric mean increased about 9.6-fold (from 16.54 ± 8.76 a.u. to 159.56 ± 52.92 a.u.), at 1000 μM H2O2 the value (393.1 ± 78.97 a.u.) was only about 2.5-fold higher as compared to 100 μM. The mean fluorescence values slowly decreased at H2O2 concentrations higher than 1 mM (not shown).

In the additional presence of 1,2,4-triazole (circles) a maximum fluorescence value was already observed at 200 μM H2O2 (380.02 ± 117.02 a.u.). Using a higher hydrogen peroxide concentration, the 1,2,4-triazole containing samples showed a decrease in the HOCl producing MPO activity. As has already been mentioned, no significant APF oxidation was observed at any H2O2 concentration in the presence of the MPO inhibitor 4-ABAH (triangles). As the usage of H2O2 concentrations higher than 1 mM (up to 5 mM were tested) led to a significant decrease in PMN vitality we activated

J. Flemmig et al. / Journal of Inorganic Biochemistry 130 (2014) 84–91

89

PMNs with 1 mM H2O2 in further experiments regarding the modulation of the chlorinating activity of MPO by (−)-epicatechin. 3.5. (−)-Epicatechin overcomes the compound II accumulation in neutrophils The addition of (−)-epicatechin to hydrogen peroxide-treated neutrophils increased the chlorinating activity of intracellular MPO (Fig. 5). Again for each sample from the obtained APF-derived fluorescence intensity distribution the geometric mean was calculated as a parameter for the HOCl-producing activity of the cells. The value calculated for the APF-stained PMNs in the absence of (−)-epicatechin was set to 100%. In Fig. 5A, the data from all analyzed samples are summarized. Briefly, the displayed means and standard deviations reflect the data from 14 (0–32 μM (−)-epicatechin) or 9 (64–128 μM (−)-epicatechin) independent cell preparations. Despite the huge standard deviations the plot of mean values against the polyphenol concentration clearly shows an about 1.25-fold increase of the APF-oxidation in the presence of 16–64μM (−)-epicatechin as compared to the control. In the presence of 128μM polyphenol this effect seems to be partially reversed (only 13% increase in comparison to the sample without (−)-epicatechin). Yet, a more detailed analysis of samples summarized in (Fig. 5A) revealed two separate groups regarding the MPO activity-enhancing effect of (−)-epicatechin. As can be seen in Fig. 5B, about half of the samples (7/14 samples for 0–32 μM, 5/9 samples for 64–128 μM) showed a strong concentrationdependent effect of (−)-epicatechin on the chlorinating MPO activity in PMNs stimulated with 1 mM H2O2. Even at 1 μM of the polyphenol, a significant increase in the geometric mean of about 16% was observed. At 16 μM (−)-epicatechin the HOCl production was increased by about 48%. Again, this effect diminished at higher polyphenol concentrations (28% increase at 128 μM (−)-epicatechin). Regarding the other sample group (Fig. 5C), no significant effect of (−)-epicatechin on the chlorinating MPO activity in the H2O2-stimulated PMNs was observed. In Fig. 5D and Fig. 5E two representative examples for the analysis of hydrogen peroxide-stimulated APF-stained cells in the absence (grey) or presence (black) of 16 μM (−)-epicatechin are shown. In Fig. 5D a sample from the first group is displayed where in the presence of polyphenol considerably higher fluorescence values were observed (83% increased geometric mean). In contrast, in Fig. 5E a sample from the second group is given where (−)-epicatechin had no stimulating effect on the chlorinating MPO activity. In conclusion, (−)-epicatechin overcomes a hydrogen peroxideinduced compound II accumulation of MPO in PMNs. Thereby the concentration dependence resembles a bell-shaped course which was also observed in the in vitro experiments with isolated MPO. Although the effect was also visible from the global analysis of all samples, in about half of the probes the described impact of (−)-epicatechin was quite strong (about 48% higher APF oxidation) while in the other probes no significant effect was observed. 4. Discussion The in vitro studies with isolated MPO clearly showed that the application of excess hydrogen peroxide is a suitable method to force a compound II accumulation in this enzyme (see Fig. 1). With increasing H2O2 concentrations a continuous decrease in the APF oxidation rate was observed reflecting a slower production of HOCl. Moreover, at very high H2O2 concentrations (over 150 μM) a decrease in the final fluorescence intensity was observed as well. This results from an incomplete APF oxidation due to the consumption of hydrogen peroxide in the peroxidase cycle. At low H2O2 concentrations (e.g. 40 μM), after reaching the maximum fluorescence as a sign for complete APF oxidation, sometimes a considerable decrease in the fluorescence was observed at longer

Fig. 5. Regeneration of HOCl producing activity in H2O2-stimulated PMNs. 106 cells were preincubated with 5 μM APF. Afterwards, the samples were stimulated with 1 mM H2O2 either in the presence or in the absence of (−)-epicatechin and, finally, analyzed by flow cytometry. For each polyphenol concentration from the obtained APF-derived fluorescence intensity distributions, geometric means were determined and related to the corresponding sample in the absence of (−)-epicatechin. The summarized data (mean and standard deviation) were always plotted against the polyphenol concentration. The summary of all samples (A) already indicates a concentration-dependent effect of (−)-epicatechin on the chlorinating activity of the MPO in H2O2-stimulated cells. Up to about 32 μM, a significant increase of the APF-derived fluorescence by about 25% was observed. This effect diminished at higher polyphenol concentrations. In (B) and (C), the samples were split into two groups. In about one half of the samples (B) a strong effect of (−)-epicatechin (48% increased chlorinating activity at 16 μM) was observed. In the other samples (C) a weak significant effect (13% higher values) was only observed at 4 μM of the polyphenol. In (D) a representative example from the sample group analyzed in (B) is given. The addition of 16 μM (−)-epicatechin (black) led to an increase in the APF-derived fluorescence as compared to the control (grey). In (E) an example for the samples analyzed in (C) is given (no effect of (−)-epicatechin).

incubation times. This may be attributed to the formation of fluorescein aggregates which leads to a bathochromic shift of the fluorescence maximum and causes a decrease in the fluorescence intensity at 522 nm [26]. Therefore, for the kinetic analysis we only used the first phase of the APF oxidation where previous measurements revealed a

90

J. Flemmig et al. / Journal of Inorganic Biochemistry 130 (2014) 84–91

linear dependence on the HOCl concentration. Further control experiments, where single components of the MPO–H2O2–Cl− system were omitted or the MPO inhibitor 4-ABAH was included, confirmed a HOCl-dependent APF oxidation. Although a reaction of the dye with compound I and/or compound II cannot be excluded no fluorescein seems to be formed in this reaction. Besides its higher sensitivity due to the detection method (fluorescence versus absorbance) this fact revealed a main advantage of APF over monochlorodimedon (MCD) for detection of the chlorinating MPO activity. Although the latter species is highly sensitive to HOCl, it is also oxidized with a high rate by MPO compound I [24]. A further advantage of the applied APF detection system is the high reaction rate between the dye and HOCl. As shown in this work the second order rate constant for the oxidation of APF by HOCl is about 1.41 × 105 M−1 cm−1 was calculated. Nevertheless regarding the used experimental setup it has to be stated that a compound II accumulation of MPO by excess H2O2 is unlikely to occur under in vivo conditions and therefore only represents a model system. Especially under inflammatory conditions e.g. nitrosative stress and the subsequent formation of peroxynitrite (ONOO−) are more likely pathways [17,18]. The in vitro-results regarding the effect of (−)-epicatechin on the halogenating MPO activity nicely reflect the assumption that in the presence of 150 μM H2O2 a compound II accumulation takes place. Upon addition of 3 μM of the flavonoid an about 20-fold faster APF oxidation was observed as a sign for the regeneration of the HOClproducing activity (Fig. 1). At higher (−)-epicatechin concentrations, this effect was reversed as the polyphenol increasingly competes with Cl− for compound I. Accordingly, at very high (−)-epicatechin concentrations also an incomplete APF oxidation was observed as a result of the consumption of H2O2 in the peroxidase cycle. Comparable measurements were made in the presence of 20 μM H2O2 where no compound II accumulation was expected. In fact, in these experiments increasing amounts of (−)-epicatechin led to a continuous decrease in the APF oxidation rate as a sign for a shift of the enzymatic activity from the halogenation to the peroxidase cycle. From stopped flow-measurements it is known that at pH 7.0 and 25 °C the second order rate constant for the oxidation of (−)-epicatechin by compound II of MPO is 4.5 × 106 M−1 s−1 while the value for the reaction with compound I is only about 4 times higher(1.9 × 107 M−1 s−1) [19]. Therefore this polyphenol is a good candidate for the resolution of accumulated compound II. In contrast, at pH 7.0 and 15 °C H2O2 reacts about 560 times faster with compound I (4.4 × 104 M−1 s−1) than with compound II (7.8 × 101 M−1 s−1) thus leading to an accumulation of this MPO redox intermediate [15,36]. In fact, in vitro-studies revealed that at neutral pH, 140 mM Cl− and 100 μM H2O2 the halogenating MPO activity rapidly drops to about 5% of its initial value [25]. A significant formation of MPO compound III as a reaction product from compound II and H2O2 is only expected at millimolar hydrogen peroxide concentrations [36]. Furthermore in the in vitro-experiments at 150 μM H2O2 the optimal (−)-epicatechin concentration (3 μM) restored about 83% of the original halogenating MPO activity at 25 μM H2O2 (kobs values of 0.024 s−1 and 0.029 s−1 were observed, respectively). Therefore it can be assumed that the addition of excess hydrogen peroxide only leads to a reversible MPO inhibition by compound II accumulation and not to protein destruction. Unlike the in vitro-results, the experiments with isolated PMNs showed continuously rising fluorescence values upon addition of increasing amounts of up to 1 mM H2O2 suggesting an activation of the cells. It is well known that hydrogen peroxide is a pro-inflammatory PMN-stimulator [37]. At even higher H2O2 concentrations the APFderived fluorescence slightly decreased. Yet as shown by control measurements at concentrations over 1 mM a significant drop in the cell vitality was observed, thus limiting the amount of hydrogen peroxide applicable. A strong pro-apoptotic effect of such high H2O2 concentrations has already been reported [38]. Again control measurements with

4-ABAH confirmed a MPO-dependent APF oxidation. Briefly, the analysis of the geometric means indicates that in the presence of 500 μM 4-ABAH about 92% of the HOCl-derived APF-oxidation was suppressed. In PMNs H2O2 has a dual effect on the halogenating MPO activity, depending on its final cellular concentration. Up to 1 mM (or 200 μM at catalase depression by 1,2,4-triazole) it leads to a pro-inflammatory cell activation and serves as a primary MPO substrate while the cell vitality is not significantly influenced. In contrast, upon addition of higher amounts of H2O2 a drop in the APF oxidation occurs suggesting an inhibition of the halogenating MPO activity. In the absence of catalase inhibitor, this effect is accompanied by the increasing cytotoxic effect of hydrogen peroxide concentrations higher than 1 mM. The addition of (−)-epicatechin to the PMNs (pre-treated with 1 mM H2O2) led to a significant promotion of the halogenating MPO activity. Comparable results were obtained at 150 μM H2O2 in the presence of 1,2,4-triazole (not shown). These data confirm a compound II accumulation under the chosen experimental conditions. Moreover, the observed bell shaped concentration dependence of the (−)-epicatechin effect (increasing APF oxidation at low concentrations, reversal at higher ones) strongly resembles the actual and previous in vitro-results [24]. A more detailed analysis of the data revealed that about half of the tested cell preparations showed quite strong effects while in the other samples (−)-epicatechin did not influence the halogenating MPO activity at all. These donor-specific properties may be attributed to different factors including the varying presence of other compound II-resolving substances, the actual status of MPO in the PMN preparations, the putative presence of MPO inhibiting substances and, last but not least, the pre-stimulation of PMNs in donor's blood. Among the substances known to convert compound II of MPO into the ferric enzyme form with sufficiently high rate are serotonin, ascorbic acid, tyrosine, urate, drugs like acetaminophen and 5-aminosalicylic acid and, most probably, a number of polyphenols other than (−)-epicatechin taken up with the food [8]. In fact, the intracellular concentration (PMN) of the compound II resolving substance ascorbic acid strongly varies between 0.7 and 2.3 mM [14,39]. Although we did not find any information about the exact mechanism of the uptake of flavonoids by PMNs we suggest a quick phagocytosis of the (−)-epicatechin. In fact, a study on the MPO-stimulating effect of (−)-epicatechin gallate revealed an uptake of the flavonoid in less than 60 min [40]. Of course, especially at higher (−)-epicatechin concentration, we cannot exclude a certain HOCl-scavenging effect of the flavonoid which would interfere with the APF oxidation. Yet, for the reaction between (−)-epicatechin and HOCl, second order rate constants in the range of 105 M−1 s−1 were determined [41]. Therefore the reaction between the flavonoid and MPO compound II (4.5 × 106 M−1 s−1) most likely prevails, especially under the chosen experimental conditions (strong compound II accumulation). In fact, as stated above both with isolated MPO and in the experiments with PMNs we observed a bell shaped dependence of APF fluorescence on (−)-epicatechin concentration. A simple HOCl-scavenging effect of the flavonoid would have led to a continuous decrease in the APF-derived fluorescence intensity. In conclusion, we were able to show that the dietary flavonoid (−)-epicatechin both in vitro and in vital human neutrophils successfully overcomes a H2O2-derived accumulation of MPO compound II and, thus, promotes the halogenating activity of this heme peroxidase. As this enzymatic activity contributes to the regulation and termination of acute inflammatory reactions, the reported results may contribute to the explanation of well-known anti-inflammatory and healthpromoting properties of this polyphenol [24]. In fact, e.g. the related epigallocatechin gallate (EGCG) was shown to exhibit protective effects against chronic inflammatory and autoimmune diseases both in man and mouse [42–44]. These effects may be related to an enhanced halogenating MPO activity.

J. Flemmig et al. / Journal of Inorganic Biochemistry 130 (2014) 84–91

Yet as already stated the physiological role of the HOCl-producing MPO activity is ambivalent and still under discussion: While there are clear indications for immune-regulatory functions of HOCl at local sites of acute inflammation [7,8] the halogenating MPO activity can also lead to inflammatory lesions and resulting pathologies, especially under severe and/or chronic inflammatory conditions [3,4]. Therefore health-promoting effects of (−)-epicatechin as a result of the MPO compound II-resolving properties of this flavonoid may only play a role under conditions of limited inflammatory reactions. Authorship JF and JA initiated the study, JF, JR and JA analyzed the data and wrote the manuscript. JF, JR, and FR performed the experiments. Disclosure The authors have no conflict of interest to disclose. Abbreviations

4-ABAH APF EGCG FITC HBSS JC-1 MCD MPO PBS PMNs ROS

4-aminobenzoic acid hydrazide aminophenyl fluorescein epigallocatechin gallate fluorescein isothiocyanate Hanks balanced salt solution 5,5′,6,6′-tetrachloro-1,1′,3,3′tetraethylbenzimidazolylcarbocyanine iodide monochlorodimedon myeloperoxidase phosphate buffered saline polymorphonuclear leukocytes reactive oxygen species

Acknowledgment This work was made possible by funding from the German Federal Ministry of Education and Research (BMBF 1315883). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jinorgbio.2013.10.002. References [1] [2] [3] [4]

S.J. Klebanoff, J. Leukoc. Biol. 77 (2005) 598–625. E. Malle, P.G. Furtmuller, W. Sattler, C. Obinger, Br. J. Pharmacol. 152 (2007) 838–854. T. Finkel, Curr. Opin. Cell Biol. 15 (2003) 247–254. L. Zheng, B. Nukuna, M.L. Brennan, M. Sun, M. Goormastic, M. Settle, D. Schmitt, X. Fu, L. Thomson, P.L. Fox, H. Ischiropoulos, J.D. Smith, M. Kinter, S.L. Hazen, J. Clin. Invest. 114 (2004) 529–541.

91

[5] T.O. Hirche, J.P. Gaut, J.W. Heinecke, A. Belaaouaj, J. Immunol. 174 (2005) 1557–1565. [6] M.M. Lloyd, D.M. van Reyk, M.J. Davies, C.L. Hawkins, Biochem. J. 414 (2008) 271–280. [7] T. Ogino, M. Hosako, K. Hiramatsu, M. Omori, M. Ozaki, S. Okada, Biochim. Biophys. Acta 1746 (2005) 135–142. [8] J. Arnhold, J. Flemmig, Arch. Biochem. Biophys. 500 (2010) 92–106. [9] A.N. Hiner, E.L. Raven, R.N. Thorneley, F. Garcia-Canovas, J.N. Rodriguez-Lopez, J. Inorg. Biochem. 91 (2002) 27–34. [10] P.G. Furtmuller, M. Zederbauer, W. Jantschko, J. Helm, M. Bogner, C. Jakopitsch, C. Obinger, Arch. Biochem. Biophys. 445 (2006) 199–213. [11] W. Jantschko, P.G. Furtmuller, M. Allegra, M.A. Livrea, C. Jakopitsch, G. Regelsberger, C. Obinger, Arch. Biochem. Biophys. 398 (2002) 12–22. [12] P.G. Furtmuller, J. Arnhold, W. Jantschko, H. Pichler, C. Obinger, Biochem. Biophys. Res. Commun. 301 (2003) 551–557. [13] J. Nève, N. Parij, N. Moguilevsky, Eur. J. Pharmacol. 417 (2001) 37–43. [14] B.G. Bolscher, G.R. Zoutberg, R.A. Cuperus, R. Wever, Biochim. Biophys. Acta 784 (1984) 189–191. [15] P.G. Furtmuller, U. Burner, W. Jantschko, G. Regelsberger, C. Obinger, FEBS Lett. 484 (2000) 139–143. [16] J. Dong, P. Chen, R. Wang, D. Yu, Y. Zhang, W. Xiao, Int. J. Biol. Sci. 7 (2011) 881–891. [17] P.G. Furtmuller, W. Jantschko, M. Zederbauer, M. Schwanninger, C. Jakopitsch, S. Herold, W.H. Koppenol, C. Obinger, Biochem. Biophys. Res. Commun. 337 (2005) 944–954. [18] R. Floris, S.R. Piersma, G. Yang, P. Jones, R. Wever, Eur. J. Biochem. 215 (1993) 767–775. [19] H. Spalteholz, P.G. Furtmuller, C. Jakopitsch, C. Obinger, T. Schewe, H. Sies, J. Arnhold, Biochem. Biophys. Res. Commun. 371 (2008) 810–813. [20] B. Buijsse, E.J. Feskens, F.J. Kok, D. Kromhout, Arch. Intern. Med. 166 (2006) 411–417. [21] C. Heiss, A. Dejam, P. Kleinbongard, T. Schewe, H. Sies, M. Kelm, JAMA 290 (2003) 1030–1031. [22] C. Heiss, P. Kleinbongard, A. Dejam, S. Perre, H. Schroeter, H. Sies, M. Kelm, J. Am. Coll. Cardiol. 46 (2005) 1276–1283. [23] S. Kori, H. Namiki, K. Suzuki, Int. Immunopharmacol. 9 (2009) 1159–1167. [24] T. Kirchner, J. Flemmig, P.G. Furtmuller, C. Obinger, J. Arnhold, Arch. Biochem. Biophys. 495 (2010) 21–27. [25] K.W. Zuurbier, A.R. Bakkenist, R. Wever, A.O. Muijsers, Biochim. Biophys. Acta 1037 (1990) 140–146. [26] J. Flemmig, J. Zschaler, J. Remmler, J. Arnhold, J. Biol. Chem. 287 (2012) 27913–27923. [27] K. Setsukinai, Y. Urano, K. Kakinuma, H.J. Majima, T. Nagano, J. Biol. Chem. 278 (2003) 3170–3175. [28] R.F. Beers Jr., I.W. Sizer, J. Biol. Chem. 195 (1952) 133–140. [29] A.J. Kettle, C.A. Gedye, C.C. Winterbourn, Biochem. J. 321 (1997) 503–508. [30] J.C. Morris, J. Phys. Chem. 70 (1966) 3798–3805. [31] J. Flemmig, J. Lessig, U. Reibetanz, P. Dautel, J. Arnhold, Cell. Physiol. Biochem. 21 (2008) 287–296. [32] M. Reers, T.W. Smith, L.B. Chen, Biochemistry 30 (1991) 4480–4486. [33] H. Kankaanranta, M.A. Lindsay, M.A. Giembycz, X. Zhang, E. Moilanen, P.J. Barnes, J. Allergy Clin. Immunol. 106 (2000) 77–83. [34] M. Kono, K. Saigo, Y. Takagi, S. Kawauchi, A. Wada, M. Hashimoto, T. Sugimoto, M. Takenokuchi, T. Morikawa, K. Funakoshi, Blood Transfus. 11 (2013) 53–60. [35] K. Temmerman, W. Nickel, J. Lipid Res. 50 (2009) 1245–1254. [36] W. Jantschko, F.P. Georg, M. Zederbauer, M. Lanz, C. Jakopitsch, C. Obinger, Biochem. Biophys. Res. Commun. 312 (2003) 292–298. [37] L. Pricop, J. Gokhale, P. Redecha, S.C. Ng, J.E. Salmon, J. Immunol. 162 (1999) 7041–7048. [38] Y. Kasahara, K. Iwai, A. Yachie, K. Ohta, A. Konno, H. Seki, T. Miyawaki, N. Taniguchi, Blood 89 (1997) 1748–1753. [39] W. Jin, L. Jiang, Electrophoresis 23 (2002) 2471–2476. [40] H. Sakagami, K. Asano, Y. Hara, T. Shimamura, J. Leukoc. Biol. 51 (1992) 478–483. [41] L. Gebicka, E. Banasiak, Toxicol. In Vitro 26 (2012) 924–929. [42] K. Gillespie, I. Kodani, D.P. Dickinson, K.U. Ogbureke, A.M. Camba, M. Wu, S. Looney, T.C. Chu, H. Qin, F. Bisch, M. Sharawy, G.S. Schuster, S.D. Hsu, Life Sci. 83 (2008) 581–588. [43] Y. Cai, T. Kurita-Ochiai, T. Hashizume, M. Yamamoto, FEMS Immunol. Med. Microbiol. (2012) 10–695x. [44] S.K. Lin, H.H. Chang, Y.J. Chen, C.C. Wang, D.L. Galson, C.Y. Hong, S.H. Kok, Arthritis Rheum. 58 (2008) 3145–3156.