Comparative reactivity of the myeloperoxidase-derived oxidants hypochlorous acid and hypothiocyanous acid with human coronary artery endothelial cells

Free Radical Biology and Medicine 65 (2013) 1352–1362

Contents lists available at ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Original Contribution

Comparative reactivity of the myeloperoxidase-derived oxidants hypochlorous acid and hypothiocyanous acid with human coronary artery endothelial cells Mitchell M. Lloyd a, Michael A. Grima a, Benjamin S. Rayner a,b, Katrina A. Hadfield a, Michael J. Davies a,b, Clare L. Hawkins a,b,n a b

The Heart Research Institute, Newtown, Sydney, NSW 2042, Australia Sydney Medical School, University of Sydney, Sydney, NSW 2006, Australia

art ic l e i nf o

a b s t r a c t

Article history: Received 5 April 2013 Received in revised form 22 September 2013 Accepted 4 October 2013 Available online 10 October 2013

In the immune response, hypohalous acids are generated by activated leukocytes via the release of myeloperoxidase and the formation of H2O2. Although these oxidants have important bactericidal properties, they have also been implicated in causing tissue damage in inflammatory diseases, including atherosclerosis. Hypochlorous acid (HOCl) and hypothiocyanous acid (HOSCN) are the major oxidants formed by myeloperoxidase under physiological conditions, with the ratio of these oxidants dependent on diet and smoking status. HOCl is highly reactive and causes marked cellular damage, but few data are available on the effects of HOSCN on mammalian cells. In this study, we have compared the actions of HOCl and HOSCN on human coronary artery endothelial cells (HCAEC). HOCl reacts rapidly with the cells, resulting in extensive cell death by both apoptosis and necrosis, with necrosis dominating at higher oxidant doses. In contrast, HOSCN is consumed more slowly, with cell death occurring only by apoptosis. Exposure of HCAEC to HOCl and HOSCN induces changes in mitochondrial membrane permeability, which, in the case of HOSCN, is associated with mitochondrial release of proapoptotic factors, including cytochrome c, apoptosis-inducing factor, and endonuclease G. With each oxidant, apoptosis appears to be caspase-independent, with the inactivation of caspases 3/7 observed, and pretreatment of the cells with the caspase inhibitor Z-VAD-fmk having no effect on the extent of cell death. Loss of cellular thiols, depletion of glutathione, and the inactivation of thiol-dependent enzymes, including glyceraldehyde-3phosphate dehydrogenase, were seen with both oxidants, though to a much greater extent with HOCl. The ability of myeloperoxidase-derived oxidants to induce endothelial cell apoptosis may contribute to the formation of unstable lesions in atherosclerosis. The results with HOSCN may be particularly significant for smokers, who have elevated plasma levels of SCN
, the precursor of this oxidant. & 2013 Elsevier Inc. All rights reserved.

Keywords: Myeloperoxidase Hypochlorous acid Hypothiocyanous acid Endothelial cell Apoptosis Free radicals

In the immune response, the activation of leukocytes results in the release of a range of enzymes from intracellular granules and the production of hydrogen peroxide (H2O2) and superoxide radicals (O2d
) [1,2]. Myeloperoxidase (MPO)1 is released extracellularly and

Abbreviations: Ac-DEVD-MCA, acetyl-l-aspartyl-l-glutamyl-l-valyl-l-aspartic acid

α-(2-methylcoumaryl-7-amide; AIF, apoptosis-inducing factor; EndoG, endonu-

clease G; EtBr, ethidium bromide; GSH, reduced glutathione; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HBSS, Hanks’ buffered salt solution; HCAEC, human coronary artery endothelial cells; HUVEC, human umbilical vein endothelial cells; IAF, 5-iodoacetamidofluorescein; MPO, myeloperoxidase; TNB, 5-thio-2-nitrobenzoic acid; TNFR, tumor necrosis factor receptor; Z-VAD-fmk, benzoyl oxycarbonyl-Val-Ala-dl-Asp fluoromethyl ketone n Corresponding author at: The Heart Research Institute, 7 Eliza Street, Newtown, Sydney, NSW 2042, Australia. Fax: þ 61 2 9565 5584. E-mail addresses: [email protected], [email protected] (C.L. Hawkins). 0891-5849/$ - see front matter & 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.freeradbiomed.2013.10.007

into phagosomal compartments by neutrophils, monocytes, and some macrophages. MPO catalyzes the reaction of H2O2 with chloride ions (Cl
) to generate hypochlorous acid (HOCl), an oxidant with potent bactericidal properties [3]. MPO can also utilize the pseudo-halide thiocyanate (SCN
) at physiological halide ion concentrations (100–140 mM Cl
, 20–100 μM Br
, o1 μM I
, and r120 μM SCN
) to generate hypothiocyanous acid (HOSCN) [4]. Similar concentrations of HOSCN and HOCl are postulated to be generated under physiological conditions, owing to the high specificity constant of MPO for SCN
compared to Cl
[4]. HOCl and HOSCN play important roles in antibacterial defense mechanisms [3,5]. However, these oxidants can also induce host tissue damage, which is implicated in the progression of a number of inflammatory diseases [2,5,6]. The evidence linking MPO and cardiovascular disease is particularly compelling, with enzymatically active MPO protein present throughout the intima in all

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

grades of human atherosclerotic lesions [7]. Marked increases in the levels of the biomarkers 3-chloro-tyrosine and 5-chloro-uracil in human atherosclerotic lesions [8,9], together with increased staining of a monoclonal antibody specific for HOCl-damaged proteins in diseased tissue, are consistent with HOCl production [10,11]. Indirect evidence supports a role for SCN
-derived oxidants, with the observation of a significant correlation between early markers of disease (fatty streaks and lipid-laden macrophages) and serum SCN
levels in the aortae of young people [12,13]. In addition, cyanate (OCN
) production via MPO-catalyzed oxidation of SCN
has been implicated as the major pathway responsible for protein carbamylation in human atherosclerotic lesions, with higher levels of homocitrulline detected than in normal arterial tissue [14]. Exposure of many different cell types to physiologically and pathologically relevant doses of HOCl results in the depletion of glutathione (GSH), inactivation of key intracellular enzymes, activation of transcription factors, and cell death [15–21]. Whereas higher oxidant doses induce necrotic cell death, lower doses have been observed to induce apoptosis in many cell types, including endothelial, tumor, and cartilage cells, though there is a lack of information regarding the specific mechanism(s) involved [22–25]. In contrast, the role of HOSCN in the induction of cellular damage is controversial, with evidence for both a protective and a damaging role of this oxidant in biological systems (reviewed in [5]). Thus, HOSCN has limited toxicity in cells associated with the oral cavity and airway (e.g., [26,27]). However, treatment of erythrocytes, macrophages, and endothelial cells with HOSCN leads to the depletion of cellular thiols and the inactivation of thiol-dependent enzymes including GAPDH, glutathione transferases, membrane ATPases, protein tyrosine phosphatases, and caspases [15–17,28–30]. Exposure of cells to HOSCN, like HOCl, can result in the activation of transcription factors and the perturbation of signaling pathways [16,31,32]. The selectivity of HOSCN for cellular thiols can result in enhanced damage compared to HOCl [15,16,28]. For example, in macrophages, HOSCN induces apoptosis by a caspase-independent pathway, with greater efficacy and at lower doses compared to HOCl [15]. However, in human umbilical vein endothelial cells (HUVEC), the targeting and inactivation of caspases by HOSCN appears to prevent cell death by apoptosis [30]. Similarly, SCN
inhibits H2O2-induced apoptosis in HL-60 cells [33] and promotes the survival of eosinophils [34]. In this study, we have compared the effects of HOCl and HOSCN on the function of human coronary artery endothelial cells (HCAEC), using concentrations from 10 to 200 mM, which are based on estimates of the amounts of these oxidants produced in vivo, from calculating the concentration of oxidant produced when circulating concentrations of neutrophils (5  106 cells/ml) are activated (250–425 mM HOCl/h [35,36]). This is significant because dysfunction of arterial endothelial cells is a key event in the pathogenesis of atherosclerosis, a disease associated with the elevated production of MPO-derived oxidants [6,37]. Such data are also important in light of the known differences between HCAEC and HUVEC responses to other inflammatory stimuli [38,39].

Materials and methods

1353

292 nm and pH 11 was measured using a molar extinction coefficient of 350 M
1 cm
1 [40]. HOSCN was prepared enzymatically using lactoperoxidase (from bovine milk; Calbiochem, Kilsyth, VIC, Australia), as previously described [15,16,41]. HOSCN concentration and cellular oxidant consumption were assessed by quantifying the loss of 5-thio-2-nitrobenzoic acid (TNB) at 412 nm [15,41] using an extinction coefficient of 14,150 M
1 cm
1 [42]. Cell culture and treatment HCAEC were obtained from Cell Applications (San Diego, CA, USA) and cultured in HCAEC MesoEndo growth medium (Cell Applications) in 75-cm2 tissue culture flasks at 37 1C in a humidified atmosphere of 5% CO2. Cells were harvested with trypsin/ EDTA (1:250) and centrifuged at 800 g for 5 min before being plated at a concentration of 0.2  106 cells/ml in 6-, 12-, or 24well plates at a volume of 2, 1, or 0.5 ml, respectively, and adhering overnight at 37 1C and 5% CO2. In each case, cells were washed before treatment with warm (37 1C) Hanks’ buffered salt solution (HBSS; Sigma–Aldrich) to prevent confounding reactions of the hypohalous acids with cell medium components. Cellular lysis assay Cell lysis was assessed by measuring DNA release into the cell medium using ethidium bromide (EtBr; Roche, Mannheim, Germany), as previously described [15]. Results are expressed as a percentage of the DNA in the same number of untreated control cells after lysis with 0.1% (v/v) Triton X-100 (Sigma–Aldrich). Fluorescence changes due to EtBr binding to released DNA were measured using λex 360 nm and λem 580 nm. Cell viability Cell viability was assessed using a commercial MTT (3-(4,5dimethyl-2-yl)-2,5-diphenyltetrazolium bromide) assay kit (Jomar Bioscience, Kensington, SA, Australia). HCAEC (2  104) were plated in 96-well tissue culture plates and incubated for 15 min to 2 h with 100 ml HOCl or HOSCN (10–400 mM; 50 nmol–2 mmol/ 106 cells), before being washed and incubated for 4 h with cell medium (100 ml) containing MTT (10 ml). Measurement of apoptosis and necrosis Apoptosis and necrosis were quantified using the annexin V–FITC Apoptosis Detection Kit (BD, North Ryde, NSW, Australia). After oxidant treatment, cells were detached using 2 mM EDTA (Sigma–Aldrich) in phosphate-buffered saline (PBS; Amresco, Solon, OH, USA), resuspended in binding buffer (10 mM Hepes/ NaOH (pH 7.4), 140 mM NaCl, and 2.5 mM CaCl2), and labeled with annexin V–FITC and propidium iodide. The staining was quantified using a Cytomics FC 500 flow cytometer (Beckman Coulter, Gladesville, NSW, Australia). DNA fragmentation in apoptotic cells was examined using the APO-BrdU TUNEL assay kit (Invitrogen), with cells detached after oxidant treatment using trypsin. Cells were fixed by addition of the cell suspension (500 ml) to 5 ml of 1% (w/v) paraformaldehyde, followed by centrifugation and washing with PBS. Cells were then stored in 70% (v/v) ethanol for 48 h at
20 1C before flow cytometry analysis.

Reagents and preparation and quantification of hypohalous acids Assessment of mitochondrial membrane permeability All aqueous solutions and buffers were prepared using Nanopure H2O filtered through a four-stage Milli Q system (MilliporeWater, Lane Cove, NSW, Australia). HOCl was prepared by dilution of a concentrated stock solution of NaOCl (BDH, Poole, Dorset, UK). To determine HOCl concentrations, the optical absorbance at

Mitochondrial membrane potential was assessed using the MitoProbe JC-1 assay kit for flow cytometry (Molecular Probes, Eugene, OR, USA). After hypohalous acid treatment, cells were detached using trypsin/EDTA (Thermotrace, Melbourne, VIC, Australia) and

1354

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

resuspended in PBS before being labeled with JC-1, a dye that exhibits potential-dependent accumulation in mitochondria. Flow cytometric analysis was then carried out using a Cytomics FC 500 flow cytometer (Beckman Coulter). Caspase activity assay After oxidant treatment, cells (0.2  106) were lysed in H2O and added to a buffer containing 100 mM Hepes (Sigma– Aldrich), 10% (w/v) sucrose (Sigma–Aldrich), 0.1% (w/v) Chaps (Sigma–Aldrich), 0.1% (v/v) Nonidet P-40 (Calbiochem), 5 mM dithiothreitol (Sigma–Aldrich), and 50 μM Ac-DEVD-MCA (acetylL-aspartyl-L-glutamyl-L-valyl-L-aspartic acid α-(2-methylcoumaryl7-amide; Peptide Institute, Osaka, Japan) as previously described [43]. The release of the fluorescent MCA subunit was measured over 60 min at 37 1C at λex 360 nm and λem 460 nm using a FlexStation 3 microplate reader (Molecular Devices, Sunnyvale, CA, USA). Quantification of total cell thiols Cells (0.2  106 seeded in a 12-well plate) were washed with HBSS (warmed to 37 1C) and lysed in 200 ml of cold (4 1C) sterile H2O (Baxter) after treatment with HOCl or HOSCN. ThioGlo 1 (Calbiochem) was added to a final concentration of 13 μM, before incubation in the dark for 5 min and recording of the fluorescence at λex 360 nm and λem 530 nm, as previously described [44]. Cellular thiols were quantified using a standard curve constructed with GSH.

Determination of cellular GSH Cellular GSH was quantified using monobromobimane with HPLC separation as described previously [45]. Briefly, HCAEC (0.2  106) were treated with oxidant and washed with HBSS before lysis in 150 ml KPBS buffer (50 mM potassium phosphate buffer, 17.5 mM EDTA, 50 mM serine, 50 mM boric acid, pH 7.4), addition of 10 ml monobromobimane (3 mM in acetonitrile), and incubation for 30 min in the dark. Perchloric acid was added (10 ml, 70% v/v) to stop the reaction, before the protein and cellular debris were pelleted by centrifugation (10 min, 4 1C, 10,000 g) and the supernatant was filtered through 0.2-mm centrifugal filters (Millipore). GSH was quantified after separation using a Shimadzu HPLC system equipped with a fluorescence detector (RF-20AXs; Shimadzu, Rydalmere, NSW, Australia), with a Synergi 4-mm Hydro-RP C-18 column (150  4.6 mm; Phenomenex, Lane Cove, NSW, Australia) and a flow rate of 1 ml min
1 at 30 1C. Mobile phase A consisted of 1% (v/v) acetic acid and 5% (v/v) acetonitrile and mobile phase B consisted of 1% (v/v) acetic acid and 20% (v/v) acetonitrile, adjusted to a pH of 4.25 using ammonium hydroxide in each case. The detector λex and λem were set at 390 and 480 nm, respectively.

Enzyme activity assays GAPDH activity was measured by monitoring the formation of NADH at 340 nm after addition of glyceraldehyde 3-phosphate as described previously [17].

Fig. 1. Cell lysis, viability, and oxidant consumption observed on exposure of HCAEC to HOCl and HOSCN. The rate and extent of cell lysis induced by (A) HOCl and (B) HOSCN on incubation with HCAEC (0.2  106 cells) were assessed by measuring the change in EtBr fluorescence on DNA release. Cells were exposed to 0 (●), 50 (■), 100 (♦), 200 (m), and, in (B), (○) 400 μM oxidant. Data represent means 7 SEM (n Z 3). *p o 0.05, **p o 0.01, and ***p o 0.001, significant increase in the extent of cell lysis with oxidant exposure compared to untreated, incubation control cells by two-way ANOVA with Bonferroni post hoc testing. Data in (C) show the change in cellular viability compared to control cells assessed by MTT assay after treatment of HCAEC with HOCl (black bars) or HOSCN (white bars) for 15 min, followed by incubation in cell medium for 4 h at 37 1C. Data represent means 7 SEM (n Z 3); ***p o 0.001, significant difference between HOCl and HOSCN treatment by two-way ANOVA with Bonferroni post hoc testing. a and b show a significant (p o 0.05) decrease compared to control cells with HOCl and HOSCN treatment, respectively, by one-way ANOVA with Dunnett's post hoc testing. Data in (D) show the loss of HOCl (200 mM, open circles) and HOSCN (200 mM, closed circles) from the cellular supernatant on exposure to HCAEC (0.2  106 cells) for 15–240 min by TNB assay. Data are expressed as a percentage of the initial oxidant added to the cells as means 7 SEM (n ¼ 3). The loss of oxidant was significant (p o 0.001) for HOCl at all time points and at 415 min with HOSCN by one-way ANOVA with Dunnett's post hoc testing.

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

Oxidation of thiol-containing proteins After hypohalous acid treatment, cells (0.2  106) were lysed in H2O and the protein concentration in each sample was determined using the bicinchoninic acid protein assay (Pierce, Rockford, IL, USA). The amount of protein in each sample was standardized to ensure equal loading onto each gel, and thiols were labeled with 200 mM IAF (5-iodoacetamidofluorescein; Invitrogen, Eugene, OR, USA) for 15 min as described previously [15,44]. Samples were reduced by incubation with equal volumes of loading buffer (2% (w/v) SDS, 10% (v/v) glycerol (Amersham), saturated bromophenol blue (ICN Biomedicals, Aurora, OH, USA), 5% (v/v) β-mercaptoethanol (Sigma–Aldrich) in 0.3 M Tris (pH 6.8, Amresco)) at 95 1C for 5 min, before separation of cellular proteins by SDS–PAGE [46]. Gels were run in the dark for 15 h at a constant current of 17 mA per gel and then scanned using a Bio-Rad Molecular Imager PharosFX fluorescence scanner with λex 488 nm and λem 530 nm to visualize IAF binding. Gels were then stained with silver nitrate to visualize total protein [47]. Band density analysis was carried out using the Bio-Rad Quantity One software.

1355

in the cytoplasmic or mitochondrial extracts were resuspended in 5  SDS loading buffer and reduced as described above. Proteins were separated by electrophoresis using SDS–PAGE with a 4% stacking gel and a 12% resolving gel (Bio-Rad). Membranes were blocked for 1 h at 22 1C using 5% (w/v) skim milk powder in Tris-buffered saline/Tween 20 (TBST; 0.1% (v/v) Tween 20), before incubation with the following primary antibodies: cytochrome c (BD Biosciences; dilution 1:1000), apoptosis-inducing factor (AIF;

Western blotting experiments HCAEC (0.2  106 cells/ml) were treated with HOCl or HOSCN (100 mM) or HBSS for 2 h before being washed (with HBSS) to remove any residual oxidant, and cell lysis and subcellular fractionation were done using a mitochondrial isolation kit, according to the manufacturer's instructions (No. 89874, Pierce). Staurosporin (1 mM) was used as a positive control to induce apoptosis. Proteins

Fig. 3. Increased TUNEL staining observed on treatment of HCAEC with HOCl and HOSCN. Cells (0.2  106) were exposed to HOCl (white bars) or HOSCN (hatched bars) or incubated in HBSS as a control before flow cytometry assessment of DNA fragmentation in apoptotic cells after 2 h of oxidant treatment using the APO-BrdU TUNEL assay kit. Data are presented as means 7 SEM (n Z 3); *p o 0.05, significant increase compared to untreated control cells by one-way ANOVA with Dunnett's post hoc testing.

Fig. 2. Necrotic and apoptotic cell death is observed in HCAEC on exposure to HOCl and HOSCN. Cells (0.2  106) were exposed to HOCl (white bars) or HOSCN (hatched bars), or incubated in HBSS as a control (black bars), for 1 or 2 h, and the percentages of healthy, apoptotic, and necrotic cells in the population were assessed using the annexin V–FITC apoptosis detection kit with flow cytometric analysis. Results are expressed as the percentage of apoptotic or necrotic cells in the total cell population. Data are presented as means 7 SEM (n Z 3). Experiments were not performed with o25 mM HOSCN, represented as “ND”. (A) Assessment of necrosis (propidium iodide uptake) after 1 h incubation of cells with oxidant; (B) assessment of apoptosis (annexin V staining) after 1 h incubation of cells with oxidant; (C) as in (A) except after 2 h; (D) as in (B) except after 2 h. *p o 0.05, **p o 0.01, and ***p o 0.001, significant increase in the percentage of apoptotic or necrotic cells compared to control, according to two-way repeated-measures ANOVA with Bonferroni post hoc testing.

1356

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

Santa Cruz; dilution 1:500), endonuclease G (EndoG; Santa Cruz; dilution 1:500), β-actin (Santa Cruz; dilution 1:1000). Primary antibodies were diluted in TBST (0.2% v/v Tween 20) containing 3% skim milk powder and incubated overnight at 4 1C. Membranes were washed 3  5 min in wash buffer (TBST; 0.1% Tween 20) and incubated for 1 h at 22 1C with either horse anti-mouse–horseradish peroxidase (HRP) or goat anti-rabbit–HRP (Cell Signaling) secondary antibodies diluted in TBST (0.2% (v/v) Tween 20) containing 3% skim milk powder. Membranes were then washed 3  5 min in wash buffer before exposure to Western Lighting chemiluminescence reagent as per the manufacturer's instructions (PerkinElmer). Results were visualized using a ChemiDoc XRS (Bio-Rad) and densitometry was performed using Image Lab (Bio-Rad). Statistical analyses All statistical analyses were carried out using GraphPad Prism 5 (GraphPad Software, San Diego, CA, USA), with p o 0.05 taken as significant. Details of specific tests performed in each case are given in the relevant figure legends.

Results

respectively, after 60 min incubation (Supplementary Table 1). The loss of HOCl and HOSCN was greater in the presence of the cells than with HBSS alone (data not shown). This assay may underestimate the extent of oxidant consumption, particularly with HOCl, as secondary oxidants, e.g., N-chloramines, formed by this oxidant also react with TNB. Changes in HCAEC viability on treatment with HOCl compared to HOSCN were investigated further by labeling cells exposed to each oxidant (25–100 mM; 125–500 nmol/106 cells for 1 or 2 h) with annexin V–FITC and propidium iodide and analysis by flow cytometry. Exposure of the cells to HOCl under these conditions resulted in extensive necrosis in a time- and dose-dependent manner, as assessed by a significant increase in the cellular uptake of propidium iodide (Fig. 2A and C, with raw flow cytometry data presented in Supplementary Fig. 1). Under these conditions, there was no significant increase in annexin V binding in the absence of propidium iodide, consistent with late-stage apoptosis/necrosis (Fig. 2B and D). In contrast, treatment of HCAEC with HOSCN resulted in a significant degree of apoptosis (annexin V binding) in the absence of significant necrosis (uptake of propidium iodide; Fig. 2). The extent of apoptosis observed in control cells incubated with HBSS increased over time, which prevented examination of longer time points (Fig. 2D).

Cellular viability and oxidant consumption Initial experiments were performed to determine the extent of cell lysis observed on oxidant treatment by assessing cellular DNA release using EtBr. HCAEC (0.2  106 cells) were exposed to HOCl or HOSCN (50–200 μM; 250 nmol–1 mmol/106 cells) for 15– 240 min. HOCl induced DNA release from HCAEC in a time- and concentration-dependent manner, with significant cell lysis observed after exposure of the cells to Z100 μM oxidant for 15 min (Fig. 1A). In contrast, with HOSCN there was no evidence for cell lysis with up to 400 μM HOSCN (2 mmol/106 cells) with incubation times of 15–240 min (Fig. 1B). However, some lysis was apparent with Z200 μM HOSCN treatment on increasing the incubation time to 360 min. The effect of HOCl and HOSCN on HCAEC viability was examined further by measuring the conversion of MTT to formazan by metabolically active cells over 4 h. A significant loss in cellular viability reflected by a loss in metabolic activity was observed on exposure of HCAEC to HOCl and HOSCN for 15 min (Fig. 1C). As in the lysis experiments, HOCl was more potent compared to HOSCN at reducing cellular viability at equivalent oxidant doses. These data may also reflect the ability of HOCl (and HOSCN) to induce mitochondrial damage rather than loss of viability, which is a limitation of the MTT assay. In each case, a greater extent of cell death was observed on increased incubation time of the oxidant with the cells (data not shown). To determine whether the differing degree of cell lysis and loss of viability induced by HOCl and HOSCN was associated with varying rates of reaction of each oxidant with HCAEC, the consumption of the hypohalous acid was investigated using the TNB assay. Exposure of the cells to HOCl (200 mM, 1 mmol/106 cells) resulted in 470% loss of the initial oxidant added by 15 min, with a 490% loss observed at longer incubation times (Fig. 1D). A similar, rapid, almost complete consumption of HOCl was seen on addition of lower concentrations of oxidant (25–100 mM, data not shown). In contrast, a much lesser extent of oxidant consumption was observed on exposure of cells to HOSCN, with only 30% of the initial oxidant lost by 240 min (Fig. 1D). The rate and extent of HOSCN consumption appeared to be independent of the initial concentration of oxidant added and, in each case, reached a plateau at 60 min. Exposure of HCAEC to 50, 100, and 200 mM HOSCN resulted in the consumption of 20, 35, and 60 mM HOSCN,

Fig. 4. Increased mitochondrial membrane depolarization observed on treatment of HCAEC with HOCl and HOSCN. Cells (0.2  106) were exposed to (A) HOCl or (B) HOSCN, compared to HBSS-treated control cells, before flow cytometry assessment of mitochondrial membrane potential after 1 (black bars) or 2 h (white bars) using the MitoProbe JC-1 assay kit. Mitochondrial membrane depolarization was also assessed in HCAEC exposed to either oligomycin or CCCP as positive control. Data are presented as means 7 SEM (n Z 9). *p o 0.05, **p o 0.01, and ***p o 0.001, significant increase compared to untreated control cells by one-way ANOVA with Dunnett's post hoc testing.

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

Given the difference in the extent of HCAEC consumption of HOSCN compared to HOCl, additional experiments were performed with lower concentrations of HOCl (5–10 mM, 25– 50 nmol/106 cells) to facilitate the comparison of the pathway responsible for cell death at lower oxidant doses and consumption. After 1 h treatment, a significant increase in annexin V binding was observed in the HOCl-treated compared to control cells (Fig. 2B). However, this induction of apoptosis was not apparent on further incubation of the cells for 2 h, in which an increase in apoptosis in control cells incubated with HBSS was also observed (Fig. 2D). No significant necrosis was observed on exposure of HCAEC to o 25 mM HOCl (Fig. 2A and C). The induction of apoptosis by HOCl and HOSCN was examined further by using 5-bromo-2′-deoxyuridine 5′-triphosphate (BrdU) to label cleavage sites resulting from DNA fragmentation (TUNEL method). An increase in TUNEL-positive cells was observed on treatment of HCAEC with both HOCl and HOSCN (25–50 mM) for 2 h (Fig. 3). In this case, there was no significant difference observed between the oxidants. Further characterization of the mechanisms involved in cell death To characterize the pathways involved in HOCl- and HOSCNinduced cell death in HCAEC, the effect of oxidant treatment on mitochondrial membrane permeability was examined. HCAEC

1357

(0.2  106 cells) were exposed to HOCl or HOSCN (25–100 μM; 125 nmol–500 mmol/106 cells) for 1 or 2 h before being washed to remove residual oxidant and labeled with JC-1, which exhibits potential-dependent accumulation in mitochondria. Flow cytometry analysis revealed a significant, dose-dependent, loss of red JC-1 aggregates in the mitochondria and increasing green JC-1 within the cytosol, consistent with depolarization and loss of mitochondrial membrane potential on exposure of HCAEC to HOCl (Fig. 4B). A decrease in the JC-1 red:green ratio was also seen with HOSCN, though in this case, only on addition of 100 mM oxidant, where 35 mM HOSCN was consumed. With HOCl, the change in JC-1 fluorescence at 100 mM oxidant is similar to that observed on treatment of HCAEC with oligomycin or Carbonyl cyanide 3-chlorophenylhydrazone as positive control (Fig. 4A). With HOSCN, mitochondrial membrane depolarization was associated with an increase in cytosolic levels of the proapoptotic factors cytochrome c, AIF, and EndoG, as detected by Western blotting experiments of cell extracts after subcellular fractionation and mitochondrial isolation (Fig. 5). The extent of release of cytochrome c, AIF, and EndoG observed in HCAEC (0.2  106 cells) exposed to HOSCN (100 mM) for 2 h was similar to that observed in cells exposed to staurosporin (1 mM), as a positive control for the induction of apoptosis (Fig. 5). In contrast, no evidence was obtained for increased release of cytochrome c, AIF, or EndoG in the corresponding experiments with HOCl (Supplementary Fig. 2).

Fig. 5. Mitochondrial release of proapoptotic factors on treatment of HCAEC with HOSCN. Cells (0.2  106) were treated with HOSCN or incubated in HBSS as a control for 2 h before washing, cell lysis, and mitochondrial isolation. The release of (A) cytochrome c, (B) AIF, and (C) EndoG into the cytoplasmic fraction was assessed by Western blotting, with lane 1 showing untreated control cells, lane 2 cells treated with HOSCN (100 mM), and lane 3 cells exposed to staurosporin (1 mM) as a positive control. The fold change in each apoptotic factor compared to untreated cells was determined after normalization to β-actin concentration. Data represent means 7 SEM (n ¼ 3); *p o 0.05, significant increase compared to control by repeated-measures one-way ANOVA with Dunnett's post hoc test.

1358

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

The release of cytochrome c observed in HCAEC treated with HOSCN was not associated with an increase in caspase 3/7 activity, as assessed by the release of MCA from the caspase-specific substrate Ac-DEVD-MCA. Evidence was obtained for a decrease in caspase activity, which was particularly marked in cells exposed to HOCl for both 1 and 2 h (Fig. 6). In each case, the extent of caspase inactivation observed was greater in cells treated with HOCl compared to HOSCN (Fig. 6). The role of caspases in apoptosis observed in HCAEC exposed to HOCl or HOSCN was examined further in cells pretreated with the caspase inhibitor Z-VAD-fmk (benzoyloxycarbonyl-Val-Ala-DL-Asp fluoromethyl ketone). No significant difference was seen in the extent of apoptosis induced by either HOCl (10 mM) or HOSCN (100 mM) after 1 h treatment in cells pretreated with Z-VAD-fmk compared to untreated cells (Fig. 7A). In contrast, pretreatment with Z-VAD-fmk dramatically reduced the extent of apoptosis in cells exposed to the potent apoptosis inducer staurosporin. This confirms that under the conditions employed, Z-VAD-fmk is an effective caspase inhibitor and can prevent caspase-dependent apoptosis in HCAEC (Fig. 7B). Hypohalous acid induces oxidation of cellular thiols Previous studies have suggested that the difference in the ability of HOCl compared to HOSCN to induce apoptosis may be related to targeting of intracellular thiols [15]. Thus, depletion of total (low-molecular-mass and protein-bound) cell thiols was assessed using ThioGlo 1, under conditions where minimal cell

Fig. 7. Pretreatment of HCAEC with the caspase inhibitor Z-VAD-fmk does not affect the induction of apoptosis by HOCl or HOSCN. (A) Cells (0.2  106) were pretreated with the caspase inhibitor Z-VAD-fmk (100 mM for 1 h) before addition of HOCl (white bars; 10 mM) or HOSCN (hatched bars; 100 mM) and further incubation for 1 h, with the % of apoptotic cells in the total cell population determined by annexin-V staining and flow cytometry. (B) The change in caspase 3/7 activity observed on staurosporin treatment (1 mM, 1 h) is reversed by pretreatment of the cells with Z-VAD-fmk (100 mM for 1 h). In each case, data represent means 7 SEM (n ¼ 3). *p o 0.05, significant increase in apoptosis compared to control by one-way ANOVA with Dunnett's post hoc test; a and b show a significant (p o 0.05) change compared to control or staurosporin in the absence of Z-VAD-fmk, respectively.

Fig. 6. HOCl and HOSCN induce caspase 3/7 inactivation. Cells (0.2  106) were treated with HOCl (white bars; 25–100 mM) or HOSCN (hatched bars; 25–100 mM) or HBSS (black bars) and incubated for (A) 1 or (B) 2 h before caspase 3/7 activity was assessed by monitoring the release of the fluorescent MCA subunit from the substrate Ac-DEVD-MCA using a plate reader. Data represent means 7 SEM (n ¼ 3). *p o 0.05 and ***p o 0.001, significant decrease in caspase activity compared to control by one-way ANOVA with Dunnett's post hoc test. #p o 0.05, significant difference in the caspase inactivation on exposure of cells to HOCl or HOSCN by two-way ANOVA with Bonferroni's post hoc test.

lysis was observed. Reaction of HCAEC (0.2  106 cells) with HOCl (50–200 μM; 250 nmol–1 mmol/106 cells) for 15 min resulted in significant thiol loss compared to the untreated, control cells with concentrations Z100 μM (Fig. 8A). In contrast, no change in total cell thiol concentration was observed in analogous experiments with HOSCN (Fig. 8B). This may reflect the low consumption of HOSCN by the HCAEC over 15 min (Fig. 1D), as significant thiol oxidation was seen on treatment of HCAEC with HOSCN concentrations 4300 μM (Fig. 8B) and with lower levels of oxidant at longer incubation times (data not shown). HOCl induced rapid depletion of cellular GSH (Fig. 8C), whereas no significant loss of GSH was seen with HOSCN under identical conditions (Fig. 8D). Similarly, a significant difference in the extent of inactivation of the thiol-dependent enzyme GAPDH induced by the two oxidants was detected, with decreased GAPDH activity observed at 25 mM HOCl and 200 mM HOSCN (Fig. 8E and F). The loss in GAPDH activity is attributed to the reactivity of HOCl and HOSCN with the critical active-site thiol residue [17,18].

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

The sensitivity of other thiol-containing proteins to HOCl and HOSCN was investigated using a thiol-specific, fluorescent labeling approach combined with SDS–PAGE. HCAEC (0.2  106 cells) were exposed to HOCl or HOSCN (100 or 200 μM) for 15 min, before removal of residual oxidant and labeling with IAF. Both oxidants decreased the fluorescent staining intensity of multiple protein bands compared to the untreated cells (Fig. 9A). These changes are attributed to oxidant-induced thiol oxidation reducing the extent of IAF labeling, as no changes were observed in the total protein levels (Fig. 9B). Control experiments showed no significant association of the IAF with nonthiol protein residues, as evidenced by a lack of fluorescent bands on pretreatment of the cell lysates with the thiol-alkylating agent N-ethylmaleimide before labeling with IAF (results not shown), consistent with previous studies [48]. Densitometry was performed on 14 IAF-labeled protein bands, indicated in Fig. 9C. HOCl induced a significant decrease in the density of all 14 IAF-labeled bands compared to control cells (Table 1). With HOSCN, no significant differences were detected even at longer incubation times (Table 1).

1359

Discussion In this study, the abilities of the major MPO-derived oxidants HOCl and HOSCN to induce damage to primary endothelial cells isolated from human coronary arteries have been directly compared. Both HOCl and HOSCN induce significant endothelial cell death, with the cell death induced by HOSCN occurring primarily by apoptosis, whereas evidence for apoptosis and cellular necrosis was observed with HOCl on HCAEC consumption of comparable concentrations of oxidant. Exposure of HCAEC to HOCl resulted in rapid consumption of the oxidant and cell lysis in a time- and concentration-dependent manner, consistent with previous studies with related cell types, including HUVEC (e.g., [18]). HCAEC are highly permeable to HOCl (or oxidant species derived from it), with evidence obtained for rapid consumption of cellular thiols, including GSH and protein thiols, and the inactivation of GAPDH, as reported previously for HUVEC (e.g., [18,49]). Evidence was also obtained in this study for the rapid inactivation of the thiol-dependent caspases 3/7 in

Fig. 8. Effect of HOCl and HOSCN on cellular thiol levels and GAPDH activity. Cells (0.2  106) were exposed to (A, C, E) HOCl or (B, D, F) HOSCN for 15 min before the determination of (A, B) total cell thiol concentration using the ThioGlo assay, (C, D) GSH concentration using monobromobimane with HPLC separation, and (E, F) GAPDH activity as described under Materials and methods. Data represent the means 7 SEM (n Z 3); np o 0.05 and nnnp o 0.001, significant decrease in thiol and GSH concentration or GAPDH activity compared to control cells, by one-way repeated-measures ANOVA with Dunnett's post hoc testing.

1360

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

Fig. 9. SDS–PAGE separation of IAF-labeled proteins in hypohalous acid-treated cells. Cells (0.2  106) were exposed to HOCl or HOSCN (100 or 200 μM) for 15 min, before removal of residual oxidant and labeling with IAF and separation by SDS–PAGE. In (A) and (B) lane 1 shows control cells; lane 2 shows HOCl-treated cells, and lane 3 shows HOSCN-treated cells. (A) IAF-labeled, thiol-containing proteins, visualized using a fluorescence scanner with λex 488 nm and λem 530 nm. (B) Total protein visualized by silver staining. (C) A section of the gel is shown representing the 14 IAF-labeled, thiol-containing proteins from untreated control cells selected for average density calculations and data comparison (Table 1). Representative gels from four replicates are shown (40 μg protein loaded per lane).

Table 1 Effect of HOCl and HOSCN on thiol-containing proteins in HCAEC. Band

IAF staining density HOCl nnn

1 2 3 4 5 6 7

52.0 63.8nnn 55.3nnn 51.6nnn 54.1nnn 56.5nnn 63.1nnn

Band

HOSCN 95.3 98.7 94.6 90.0 96.3 101.5 105.0

IAF staining density HOCl

8 9 10 11 12 13 14

nnn

59.9 54.5nnn 47.6nnn 50.3nnn 58.3nnn 70.1nnn 74.2nn

HOSCN 100.8 81.9 77.1 87.8 87.9 96.5 113.3

The relative average density ratios were standardized by dividing the average density for each band obtained from the fluorescence scans (Fig. 9C) by the average density for each lane obtained by analyzing the selected band on the corresponding total protein gel scan (stained by silver nitrate staining). Results are expressed as a percentage of the controls (n Z 8). Significant differences in relative average band density compared to control were calculated by one-way repeated-measures ANOVA with Dunnett's post hoc testing. nn

p o 0.01. p o 0.001.

nnn

HCAEC exposed to HOCl. Under the conditions employed in this study, significant cell death via apoptosis was apparent on treating the cells with low (o 25 mM) HOCl, consistent with previous studies with HUVEC [22,50] and human saphenous vein endothelial cells [23]. With HOSCN, significant cell lysis and reduced viability, as assessed by measuring cellular metabolism using MTT, was observed only with high oxidant doses ( 4200 mM) and after prolonged incubation of the cells (360 min). With this oxidant, a much slower rate and reduced extent of oxidant consumption was observed compared to HOCl, consistent with previous studies [15,30]. HOSCN was also able to induce apoptosis, as evidenced by the significant increase in the population of cells expressing phosphatidyl serine on the outer membrane that stain positively with annexin V–FITC and increased TUNEL staining. The potency of

HOSCN as an inducer of apoptosis is highlighted by the observation that significant apoptosis is seen on exposure of the cells to 25 mM oxidant, despite the cells consuming less than 25% of the added HOSCN. Although evidence was obtained to support apoptotic cell death with o100 mM HOSCN, a significant reduction in cellular viability assessed by MTT assay was apparent only with 4200 mM HOSCN. This discrepancy may reflect differences in the experimental protocol, as the ability of cells to metabolize MTT is assessed after oxidant treatment, over a 4-h incubation period in the presence of complete cell medium. This additional incubation step with medium was not included in the apoptosis studies and may allow repair of HOSCNinduced damage at low oxidant doses, hence influencing the extent of cell death. This is particularly relevant in light of recent studies showing formation of protein sulfenic acids in cells exposed to HOSCN, which are responsible for the reversible inactivation of a number of intracellular enzymes, including GAPDH and creatine kinase [17]. The potential for repair of HOSCN-induced damage may also rationalize the discrepancy between our data with HCAEC and related studies with HUVEC, in which HOSCN is reported to inhibit both caspase 3 activation and apoptosis [30]. Thus, in the HUVEC study, apoptosis was assessed after incubation of the cells for 2.5 h with cell medium after the initial oxidant treatment [30]. However, this behavior may also be related to differences between arterial and venous endothelial cells, particularly as there are key differences in the responses of each cell type to other inflammatory stimuli [38,39]. The data from this study are consistent with apoptotic cell death in each case via a caspase-independent pathway, with decreased caspase activity seen in HCAEC exposed to both HOCl and HOSCN. In addition, no reduction in cellular apoptosis was observed on pretreatment of the cells with the caspase inhibitor Z-VAD-fmk, although this inhibitor significantly reduced apoptosis induced by staurosporin. However, additional experiments are required to definitively exclude a role for caspases in HOCl- and HOSCN-mediated apoptosis, particularly as recent studies highlight the induction of programmed necrosis in cells exposed to the caspase inhibitor Z-VAD-fmk [51]. Changes in mitochondrial membrane permeability were also apparent in cells exposed to HOSCN and particularly HOCl. In the case of HOSCN, increased mitochondrial release of AIF and EndoG to the cytoplasm was also observed. Although some evidence for increased cytochrome c was obtained, this did not reach statistical significance, which further supports the induction of apoptosis by a non-caspase-dependent pathway, such as that involving AIF or EndoG. The greater extent of DNA fragmentation seen in HOSCNtreated HCAEC, reflected by the population of TUNEL-positive cells, is consistent with the increased release of AIF and EndoG. Apoptotic cell death by a caspase-independent pathway has been reported previously in other cell types exposed to hypohalous acids, in which inactivation of caspases 3/7 is also apparent [15,24]. The process(es) responsible for triggering apoptosis in HCAEC remains to be fully determined. In macrophages, HOSCN-induced apoptosis correlated with thiol loss, with increased oxidation of protein thiols observed, compared to HOCl [15]. However, this does not seem to be the case in HCAEC, in which the extent of thiol oxidation, reflected by GAPDH inactivation, GSH loss, and protein thiol modification, was much less marked with HOSCN compared to HOCl. It is therefore possible that HOSCN exerts its proapoptotic effects by interaction with membrane receptors, which is consistent with the effects of low levels of HOSCN and limited intracellular thiol loss. The tumor necrosis factor receptor (TNFR) superfamily members are characterized by Cys-rich extracellular domains, which are likely to be favorable targets of HOSCN oxidation. Recent studies show that TNFRs undergo a conformational change on oxidation of these Cysrich domains, which results in the activation of NF-κB [52]. Activation

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

of NF-κB has been implicated in detrimental inflammatory signaling cascades in HUVEC exposed to HOSCN [31,32]. It is not clear whether the potential activation of NF-κB by HOSCN also plays a role in apoptosis, as this transcription factor is known to have both pro- and antiapoptotic functions, depending on the nature of the stimulus [53,54]. Stimulation of TNFRs and other cell surface receptors, such as the epidermal growth factor receptor, has also been implicated in mitogen-activated protein kinase signaling (reviewed in [55]). This may be particularly relevant given that exposure of various cell types to HOSCN results in hyperphosphorylation of p38α and ERK1/2 [16,32,56]. It is well established that p38 plays a critical role in stress-induced apoptosis [16,55]. Similarly, although phosphorylation of ERK1/2 is more commonly associated with promotion of proliferation and cell survival, recent studies show that it may also have a proapoptotic role (reviewed in [55]). In conclusion, we have demonstrated that exposure of HCAEC to the MPO-derived oxidants HOCl and HOSCN induces cell death, with the likely involvement of different cellular pathways. The ability of these oxidants to induce apoptosis in HCAEC may have implications for the development of atherosclerosis, a disease in which MPO has been strongly linked to pathogenesis (reviewed in [6,37]). Apoptotic endothelial cells are a common feature of advanced human atherosclerotic lesions, in which they are believed to contribute to plaque erosion, leading to instability and possible thrombosis (e.g., [57]). These data provide a possible rationale for the induction of differential cellular effects in smokers compared to nonsmokers, given that SCN
, the precursor to HOSCN, is significantly elevated in the plasma of smokers [58], though smoking may also influence lesion development by other pathways unrelated to SCN
and the formation of HOSCN.

Acknowledgments The authors acknowledge funding support from the National Health and Medical Research Council (570829), National Heart Foundation of Australia (CR08S3959), and the Australian Research Council (FT120100682). We also thank Shirley Nakhla for technical assistance with the flow cytometry experiments and Dr. David van Reyk for helpful discussions.

Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.freeradbiomed. 2013.10.007. References [1] Weiss, S. J.; LoBuglio, A. F. Phagocyte-generated oxygen metabolites and cellular injury. Lab. Invest. 47:5–18; 1982. [2] Klebanoff, S. J. Myeloperoxidase: friend and foe. J. Leukocyte Biol. 77:598–625; 2005. [3] Hampton, M. B.; Kettle, A. J.; Winterbourn, C. C. Inside the neutrophil phagosome: oxidants, myeloperoxidase, and bacterial killing. Blood 92:3007–3017; 1998. [4] van Dalen, C. J.; Whitehouse, M. W.; Winterbourn, C. C.; Kettle, A. J. Thiocyanate and chloride as competing substrates for myeloperoxidase. Biochem. J. 327:487–492; 1997. [5] Barrett, T. J.; Hawkins, C. L. Hypothiocyanous acid: benign or deadly? Chem. Res. Toxicol. 25:263–273; 2012. [6] Davies, M. J.; Hawkins, C. L.; Pattison, D. I.; Rees, M. D. Mammalian heme peroxidases: from molecular mechanisms to health implications. Antioxid. Redox Signaling 10:1199–1234; 2008. [7] Daugherty, A.; Dunn, J. L.; Rateri, D. L.; Heinecke, J. W. Myeloperoxidase, a catalyst for lipoprotein oxidation, is expressed in human atherosclerotic lesions. J. Clin. Invest. 94:437–444; 1994.

1361

[8] Hazen, S. L.; Heinecke, J. W. 3-Chlorotyrosine, a specific marker of myeloperoxidase-catalysed oxidation, is markedly elevated in low density lipoprotein isolated from human atherosclerotic intima. J. Clin. Invest. 99:2075–2081; 1997. [9] Takeshita, J.; Byun, J.; Nhan, T. Q.; Pritchard, D. K.; Pennathur, S.; Schwartz, S. M.; Chait, A.; Heinecke, J. W. Myeloperoxidase generates 5-chlorouracil in human atherosclerotic tissue: a potential pathway for somatic mutagenesis by macrophages. J. Biol. Chem. 281:3096–3104; 2006. [10] Hazell, L. J.; Arnold, L.; Flowers, D.; Waeg, G.; Malle, E.; Stocker, R. Presence of hypochlorite-modified proteins in human atherosclerotic lesions. J. Clin. Invest. 97:1535–1544; 1996. [11] Hazell, L. J.; Baernthaler, G.; Stocker, R. Correlation between intima-to-media ratio, apolipoprotein b-100, myeloperoxidase, and hypochlorite-oxidized proteins in human atherosclerosis. Free Radic. Biol. Med. 31:1254–1262; 2001. [12] Scanlon, C. E. O.; Berger, B.; Malcom, G.; Wissler, R. W. Evidence for more extensive deposits of epitopes of oxidized low density lipoproteins in aortas of young people with elevated serum thiocyanate levels. Atherosclerosis 121:23–33; 1996. [13] Botti, T. P.; Amin, H.; Hiltscher, L.; Wissler, R. W. A comparison of the quantitation of macrophage foam cell populations and the extent of apolipoprotein E deposition in developing atherosclerotic lesions in young people: high and low serum thiocyanate groups as an indication of smoking. Atherosclerosis 124:191–202; 1996. [14] Wang, Z.; Nicholls, S. J.; Rodriguez, E. R.; Kummu, O.; Horkko, S.; Barnard, J.; Reynolds, W. F.; Topol, E. J.; DiDonato, J. A.; Hazen, S. L. Protein carbamylation links inflammation, smoking, uremia and atherogenesis. Nat. Med. 13:1176–1184; 2007. [15] Lloyd, M. M.; Van Reyk, D. M.; Davies, M. J.; Hawkins, C. L. HOSCN is a more potent inducer of apoptosis and protein thiol depletion in murine macrophage cells than HOCl or HOBr. Biochem. J. 414:271–280; 2008. [16] Lane, A. E.; Tan, J. T.; Hawkins, C. L.; Heather, A. K.; Davies, M. J. The myeloperoxidase-derived oxidant HOSCN inhibits protein tyrosine phosphatases and modulates cell signalling via the mitogen-activated protein kinase (MAPK) pathway in macrophages. Biochem. J. 430:161–169; 2010. [17] Barrett, T. J.; Pattison, D. I.; Leonard, S. E.; Carroll, K. S.; Davies, M. J.; Hawkins, C. L. Inactivation of thiol-dependent enzymes by hypothiocyanous acid: role of sulfenyl thiocyanate and sulfenic acid intermediates. Free Radic. Biol. Med. 52:1075–1085; 2012. [18] Pullar, J. M.; Winterbourn, C. C.; Vissers, M. C. M. Loss of GSH and thiol enzymes in endothelial cells exposed to sublethal concentrations of hypochlorous acid. Am. J. Physiol. 277:H1505–H1512; 1999. [19] Pullar, J. M.; Vissers, M. C.; Winterbourn, C. C. Living with a killer: the effects of hypochlorous acid on mammalian cells. IUBMB Life 50:259–266; 2000. [20] Schraufstatter, I. U.; Browne, K.; Harris, A.; Hyslop, P. A.; Jackson, J. H.; Quehenberger, O.; Cochrane, C. G. Mechanisms of hypochlorite injury of target cells. J. Clin. Invest. 85:554–562; 1990. [21] Schoonbroodt, S.; Legrand Poels, S.; Best Belpomme, M.; Piette, J. Activation of the NF-kappaB transcription factor in a T-lymphocytic cell line by hypochlorous acid. Biochem. J. 321:777–785; 1997. [22] Vissers, M. C. M.; Pullar, J. M.; Hampton, M. B. Hypochlorous acid causes caspase activation and apoptosis or growth arrest in human endothelial cells. Biochem. J. 344; 1999. (443–339). [23] Sugiyama, S.; Kugiyama, K.; Aikawa, M.; Nakamura, S.; Ogawa, H.; Libby, P. Hypochlorous acid, a macrophage product, induces endothelial apoptosis and tissue factor expression: involvement of myeloperoxidase-mediated oxidant in plaque erosion and thrombogenesis. Arterioscler. Thromb. Vasc. Biol. 24:1309–1314; 2004. [24] Whiteman, M.; Chu, S. H.; Siau, J. L.; Rose, P.; Sabapathy, K.; Schantz, J. T.; Cheung, N. S.; Spencer, J. P.; Armstrong, J. S. The pro-inflammatory oxidant hypochlorous acid induces Bax-dependent mitochondrial permeabilisation and cell death through AIF-/EndoG-dependent pathways. Cell. Signalling 19:705–714; 2007. [25] Whiteman, M.; Rose, P.; Siau, J. L.; Cheung, N. S.; Tan, G. S.; Halliwell, B.; Armstrong, J. S. Hypochlorous acid-mediated mitochondrial dysfunction and apoptosis in human hepatoma HepG2 and fetal liver cells: role of mitochondrial permeability transition. Free Radic. Biol. Med. 38:1571–1584; 2005. [26] Tenovuo, J.; Larjava, H. The protective effect of peroxidase and thiocyanate against hydrogen peroxide toxicity assessed by the uptake of [3H]-thymidine by human gingival fibroblasts cultured in vitro. Arch. Oral Biol. 29:445–451; 1984. [27] Gould, N. S.; Gauthier, S.; Kariya, C. T.; Min, E.; Huang, J.; Day, B. J. Hypertonic saline increases lung epithelial lining fluid glutathione and thiocyanate: two protective CFTR-dependent thiols against oxidative injury. Respir. Res. 11:119–128; 2010. [28] Arlandson, M.; Decker, T.; Roongta, V. A.; Bonilla, L.; Mayo, K. H.; MacPherson, J. C.; Hazen, S. L.; Slungaard, A. Eosinophil peroxidase oxidation of thiocyanate— characterization of major reaction products and a potential sulfhydryl-targeted cytotoxicity system. J. Biol. Chem. 276:215–224; 2001. [29] Grisham, M. B.; Ryan, E. M. Cytotoxic properties of salivary oxidants. Am. J. Physiol. 258:C115–C121; 1990. [30] Bozonet, S. M.; Scott-Thomas, A. P.; Nagy, P.; Vissers, M. C. Hypothiocyanous acid is a potent inhibitor of apoptosis and caspase 3 activation in endothelial cells. Free Radic. Biol. Med. 49:1054–1063; 2010. [31] Wang, J. G.; Mahmud, S. A.; Nguyen, J.; Slungaard, A. Thiocyanate-dependent induction of endothelial cell adhesion molecule expression by phagocyte peroxidases: a novel HOSCN-specific oxidant mechanism to amplify inflammation. J. Immunol. 177:8714–8722; 2006.

1362

M.M. Lloyd et al. / Free Radical Biology and Medicine 65 (2013) 1352–1362

[32] Wang, J. G.; Mahmud, S. A.; Thompson, J. A.; Geng, J. G.; Key, N. S.; Slungaard, A. The principal eosinophil peroxidase product, HOSCN, is a uniquely potent phagocyte oxidant inducer of endothelial cell tissue factor activity: a potential mechanism for thrombosis in eosinophilic inflammatory states. Blood 107:558–565; 2006. [33] Wagner, B. A.; Reszka, K. J.; McCormick, M. L.; Britigan, B. E.; Evig, C. B.; Burns, C. P. Role of thiocyanate, bromide and hypobromous acid in hydrogen peroxideinduced apoptosis. Free Radic. Res. 38:167–175; 2004. [34] Mahmud, S. A.; Wang, J. G.; Slungaard, A. Thiocyanate-dependent inhibition of spontaneous and agonist-induced eosinophil apoptosis by eosinophil peroxidase (EPO): a potential physiologic role for endogenously generated HOSCN in maintaining eosinophil viability. Blood 108; 2006. (466A–466A). [35] Kettle, A. J.; Winterbourn, C. C. Assays for the chlorination activity of myeloperoxidase. Methods Enzymol. 233:502–512; 1994. [36] Weiss, S. J.; Klein, R.; Slivka, A.; Wei, M. Chlorination of taurine by humanneutrophils—evidence for hypochlorous acid generation. J. Clin. Invest. 70:598–607; 1982. [37] Nicholls, S. J.; Hazen, S. L. Myeloperoxidase and cardiovascular disease. Arterioscler. Thromb. Vasc. Biol. 25:1102–1111; 2005. [38] Johnson, A. R. Human pulmonary endothelial cells in culture: activities of cells from arteries and cells from veins. J. Clin. Invest. 65:841–850; 1980. [39] Hauser, I. A.; Johnson, D. R.; Madri, J. A. Differential induction of VCAM-1 on human iliac venous and arterial endothelial cells and its role in adhesion. J. Immunol. 151:5172–5185; 1993. [40] Morris, J. C. The acid ionization constant of HOCl from 5 1C to 35 1C. J. Phys. Chem. 70:3798–3805; 1966. [41] Hawkins, C. L.; Pattison, D. I.; Stanley, N. R.; Davies, M. J. Tryptophan residues are targets in hypothiocyanous acid-mediated protein oxidation. Biochem. J. 416:441–452; 2008. [42] Eyer, P.; Worek, F.; Kiderlen, D.; Sinko, G.; Stuglin, A.; Simeon-Rudolf, V.; Reiner, E. Molar absorption coefficients for the reduced Ellman reagent: reassessment. Anal. Biochem. 312:224–227; 2003. [43] Hampton, M. B.; Morgan, P. E.; Davies, M. J. Inactivation of cellular caspases by peptide-derived tryptophan and tyrosine peroxides. FEBS Lett. 527:289–292; 2002. [44] Hawkins, C. L.; Morgan, P. E.; Davies, M. J. Quantification of protein modification by oxidants. Free Radic. Biol. Med. 46:965–988; 2009. [45] Kariya, C.; Chu, H. W.; Huang, J.; Leitner, H.; Martin, R. J.; Day, B. J. Mycoplasma pneumoniae infection and environmental tobacco smoke inhibit lung

[46] [47]

[48]

[49]

[50]

[51] [52]

[53]

[54]

[55] [56]

[57] [58]

glutathione adaptive responses and increase oxidative stress. Infect. Immun. 76:4455–4462; 2008. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage t4. Nature 227:680–685; 1970. Shevchenko, A.; Wilm, M.; Vorm, O.; Mann, M. Mass spectrometric sequencing of proteins from silver stained polyacrylamide gels. Anal. Chem. 68:850–858; 1996. Summers, F. A.; Forsman Quigley, A.; Hawkins, C. L. Identification of proteins susceptible to thiol oxidation in endothelial cells exposed to hypochlorous acid and N-chloramines. Biochem. Biophys. Res. Commun. 425:157–161; 2012. Peskin, A. V.; Winterbourn, C. C. Taurine chloramine is more selective than hypochlorous acid at targeting critical cysteines and inactivating creatine kinase and glyceraldehyde-3-phosphate dehydrogenase. Free Radic. Biol. Med. 40:45–53; 2006. Vissers, M. C.; Lee, W. G.; Hampton, M. B. Regulation of apoptosis by vitamin C: specific protection of the apoptotic machinery against exposure to chlorinated oxidants. J. Biol. Chem. 276:46835–46840; 2001. Kim, S. J.; Li, J. Caspase blockade induces RIP3-mediated programmed necrosis in toll-like receptor-activated microglia. Cell Death Dis. 4:e716; 2013. Ozsoy, H. Z.; Sivasubramanian, N.; Wieder, E. D.; Pedersen, S.; Mann, D. L. Oxidative stress promotes ligand-independent and enhanced ligand-dependent tumor necrosis factor receptor signaling. J. Biol. Chem. 283:23419–23428; 2008. Kaltschmidt, B.; Kaltschmidt, C.; Hofmann, T. G.; Hehner, S. P.; Droge, W.; Schmitz, M. L. The pro- or anti-apoptotic function of NF-kappaB is determined by the nature of the apoptotic stimulus. Eur. J. Biochem. 267:3828–3835; 2000. Fan, C.; Yang, J.; Engelhardt, J. F. Temporal pattern of NF-kappaB activation influences apoptotic cell fate in a stimuli-dependent fashion. J. Cell Sci. 115:4843–4853; 2002. Runchel, C.; Matsuzawa, A.; Ichijo, H. Mitogen-activated protein kinases in mammalian oxidative stress responses. Antioxid. Redox Signaling 15:205–218; 2011. Rehani, T.; Jonas, A.; Slungaard, A. The major phagocyte peroxidase-derived oxidant HOSCN stimulates the gene specific p38 MAPK-dependent NF-kappaB mediated activation of tissue factor, VCAM-1 and ICAM-1 expression in endothelium. Blood 116:633; 2010. Tabas, I.; Ron, D. Integrating the mechanisms of apoptosis induced by endoplasmic reticulum stress. Nat. Cell Biol. 13:184–190; 2011. Morgan, P. E.; Pattison, D. I.; Talib, J.; Summers, F. A.; Harmer, J. A.; Celermajer, D. S.; Hawkins, C. L.; Davies, M. J. High plasma thiocyanate levels in smokers are a key determinant of thiol oxidation induced by myeloperoxidase. Free Radic. Biol. Med. 51:1815–1822; 2011.