Myeloperoxidase impairs the contractile function in isolated human cardiomyocytes

Abbreviations: ACS, acute coronary syndrome; ADHP, 10-acetyl-3,7-dihydroxyphenoxazine; APF, 2- (6-(4-aminophenoxy)-3-oxo-3H-xanten-9-yl)-benzoic acid; BSA, bovine serum albumin; CAD, coronary artery disease; CI, carbonylation index; CV, cardiovascular; DNPH, 2,4dinitrophenylhydrazine; DMF, dimethylformamide; DTDP, dithiodipyridine; DTNB, 5,5’-dithiobis(2-nitrobenzoic acid); DTT, dithiotreitol; ECL, enhanced chemiluminescence; EGTA, ethyleneglycoltetraacetic acid; Factive, cardiomyocyte active force; Fpassive, cardiomyocyte passive force; HDL, high-density lipoprotein; HF, heart failure; HOCl, hypochlorous acid; H2O2, hydrogen peroxide; Iso, isolating solution; LDL, low-density lipoprotein; LV, left ventricular; MetSO, methionine sulfoxide; MHC, myosin heavy chain; MI, myocardial infarction; MLC-1, myosin light chain–1; MPO, myeloperoxidase; MPO-I, MPO inhibitor (4-aminobenzhydrazide); MyBP-C, myosinbinding protein C; N2B, stiff titin isoform; N2BA, compliant titin isoform; NAC, N-acetyl-Lcysteine; NO, nitric oxide; NOS, nitric oxide synthase; NTB, 2‐nitro‐5‐thiobenzoic acid; PBS, phosphate-buffered saline; pCa50, measure of calcium sensitivity; PMSF, phenylmethylsulfonyl fluoride; ROS, reactive oxygen species; SDS, sodium dodecyl sulphate; SH, sulfhydryl; Tm, tropomyosin. *Corresponding author at: Division of Clinical Physiology, Institute of Cardiology, Faculty of Medicine, University of Debrecen, H-4032 Debrecen, Móricz Zsigmond krt. 22., Hungary, Telephone/Fax: +36 52 255928, E-mail: [email protected] Original Contribution 1 2 Myeloperoxidase impairs the contractile function in isolated human cardiomyocytes 3 4 Judit Kalász, Enikő Pásztorné Tóth, Miklós Fagyas, Ágnes Balogh, Attila Tóth, Viktória 5 Csató, István Édes, Zoltán Papp and Attila Borbély* 6 7 Division of Clinical Physiology, Institute of Cardiology, Faculty of Medicine, University of 8 Debrecen, Debrecen, Hungary 9 10 11 2 Abstract 1 Purpose: We set out to characterize the mechanical effects of myeloperoxidase (MPO) in 2 isolated left ventricular human cardiomyocytes. Oxidative myofilament protein modifications 3 (sulfhydryl (SH) group oxidation and carbonylation) induced by the peroxidase and 4 chlorinating activities of MPO were additionally identified. The specificity of the MPO-5 evoked functional alterations was tested with an MPO inhibitor (MPO-I) and the antioxidant 6 amino acid Met. 7 Results: The combined application of MPO and its substrate, hydrogen peroxide (H2O2), 8 largely reduced the active force (Factive), increased the passive force (Fpassive) and decreased 9 the Ca2+ sensitivity of force production (pCa50) in permeabilized cardiomyocytes. H2O2 alone 10 had significantly smaller effects on Factive and Fpassive and did not alter pCa50. The MPO-I 11 blocked both the peroxidase and chlorinating activities, while Met selectively inhibited the 12 chlorinating activity of MPO. All of the MPO-induced functional effects could be prevented 13 by the MPO-I and Met. Both H2O2 alone and MPO+H2O2 reduced the SH content of actin 14 and increased the carbonylation of actin and myosin-binding protein C to the same extent. 15 Neither the SH-oxidation nor the carbonylation of the giant sarcomeric protein titin was 16 affected by these treatments. 17 Conclusions: MPO activation induces a cardiomyocyte dysfunction by affecting Ca2+-18 regulated active and Ca2+-independent passive force production and myofilament Ca2+ 19 sensitivity, independently of protein SH oxidation and carbonylation. The MPO-induced 20 deleterious functional alterations can be prevented by the MPO-I and Met. Inhibition of MPO 21 may be a promising therapeutic target to limit myocardial contractile dysfunction during 22 inflammation. 23 Keywords: cardiomyocyte contractile function, myeloperoxidase, hydrogen peroxide, 24 oxidative post-translational protein modifications, antioxidants 25 3 Introduction 1 Oxidative stress-related myofilament protein alterations have been shown to play key roles in 2 the impaired cardiomyocyte contractility in response to myocardial inflammation, ischemia-3 reperfusion injury and left ventricular (LV) remodeling following a myocardial infarction 4 (MI) [1, 2]. In particular, reactive oxygen species (ROS) oxidize cellular components [3], 5 leading to cardiomyocyte contractile dysfunction, myocyte apoptosis or cardiac hypertrophy 6 [4, 5]. 7 Myeloperoxidase (MPO; EC 1.11.2.2) is a member of the heme peroxidase 8 superfamily, synthesized by neutrophils, monocytes and macrophages, stored in their 9 azurophilic granules and released in substantial amount upon leukocyte activation [6]. MPO 10 has beneficial effects in the innate host defense mechanisms [7]. Considerable evidence has 11 emerged to suggest, that ROS formation by MPO promotes various deleterious action in the 12 cardiovascular (CV) system and contributes to the development of CV diseases [6]. 13 Individuals with a total or subtotal MPO deficiency (a defect with a frequency of ≈1 in every 14 2000 to 4000 Caucasians) are protected from CV diseases [6]. An elevated level of 15 circulating MPO is a prognostic marker of mortality and predicts the risks of subsequent 16 major adverse cardiac events in patients with acute coronary syndrome (ACS) [8], 17 particularly in association with a low LV ejection fraction [9]. MPO also contributes to 18 adverse LV remodeling after a MI [10]. MPO exerts adverse effects on the vasculature, 19 oxidizes low-density lipoprotein (LDL) [11], impairs the high-density lipoprotein (HDL) 20 function [12] and reduces the bioavailability of nitric oxide (NO) [13]. MPO can therefore 21 serve as a valuable biomarker of inflammation in coronary artery disease (CAD) and ACS 22 [14]. The serum level of MPO correlates positively with the severity of the LV dysfunction 23 and seems to be an essential factor in the development and exacerbation of heart failure (HF) 24 [15, 16]. Interestingly, the MPO concentration was earlier found not to differ in ischemic and 25 4 non-ischemic cardiomyopathy, suggesting that MPO has an independent pathogenic role in 1 the LV dysfunction [17]. 2 MPO is known to generate numerous reactive oxidants and diffusible radical species 3 via its peroxidase and chlorinating activities, which are capable of promoting an array of 4 reversible and irreversible post-translational protein modifications [18, 19]. The relative 5 concentrations of chloride and the reducing substrate determine whether MPO uses its 6 substrate hydrogen peroxide (H2O2) for peroxidation or chlorination. MPO amplifies the 7 oxidative potential of H2O2 [20-22], which may originate from a number of sources in vivo, 8 including leukocyte NADPH oxidases, xanthine oxidase and uncoupled NO synthase (NOS) 9 [23, 24]. The perfusion of isolated rat hearts with H2O2 led to disulfide cross-bridge 10 formation in actin and tropomyosin (Tm) [25]. In one of our previous studies, the sulfhydryl 11 (SH) oxidation of actin and myosin light chain-1 (MLC-1) was suggested as the mechanism 12 in the H2O2-evoked depressed cardiomyocyte contractility [26]. 13 MPO is unique in its ability to create hypochlorous acid (HOCl, a potent antimicrobial 14 agent) through its chlorinating activity [22]. Interestingly, the cardiac tissue is highly 15 susceptible to oxidation even by physiological concentrations of HOCl [27]. Importantly, 16 HOCl is much more effective than H2O2 in oxidizing proteins in the myocardium [27], it 17 causes SH oxidation [28] and carbonylation in myofilament proteins [29], it disturbs Ca2+ 18 homeostasis and Ca2+ handling [30], it increases the intracellular Ca2+ concentration in 19 isolated rat [31] and rabbit [32] ventricular cardiomyocytes, and it induces cardiomyocyte 20 death in rats [33]. It is also very important to consider, how far H2O2 or HOCl can diffuse on 21 the cellular scale and whether these substances are capable to penetrate the cell membranes. 22 H2O2 is stable [34], membrane permeable [35], although, in vivo concentration of H2O2 23 highly depends on its generation and consumption rates [36, 37]. HOCl appears to be more 24 toxic and reactive and can also penetrate through cell membranes, but has a much shorter 25 5 lifespan. An in vitro study revealed that HOCl production by neutrophils can be as high as 1 450 mM/h, which was shown to be less in an in vivo model [38]. MPO generates HOCl in 2 micro-molar concentration [39], but in inflammatic tissue it is estimated to be as high as 5 3 mM [40]. 4 The antioxidant amino acid Met acts as a scavenger of HOCl and has been shown to 5 prevent the HOCl-induced morphological changes and contractile dysfunction in murine 6 myocytes [41]. Moreover, the fact that MPO-derived chlorinating compounds can serve as 7 specific biomarkers for disease progression has attracted considerable interest in the 8 development of therapeutically useful MPO inhibitors (MPO-Is) [42]. 9 Although the role of MPO-derived oxidants in the pathogenesis of myocardial 10 ischemia and HF is relatively well established, only limited data are available as concerns the 11 exact cellular and subcellular mechanisms through which MPO could directly affect the 12 contractility of the myocardial cells, especially at the level of the myofibrillar proteins. In this 13 study, therefore, we set out (1) to characterize the functional effects of MPO and its substrate 14 H2O2 on single, permeabilized human cardiomyocytes; (2) to identify the biochemical 15 alterations induced by the peroxidase and chlorinating activities of MPO; (3) to investigate 16 the specificity of the MPO-induced contractile changes by using the MPO inhibitor (MPO-I) 17 4-aminobenzhydrazide and the antioxidant amino acid Met; and (4) to explore the MPO-18 related reversible and irreversible oxidative myofilament protein modifications in the human 19 LV myocardium. 20 21 22 23 6 Materials and methods 1 2 I. Human myocardial samples 3 LV myocardial tissue was obtained from the hearts of four general organ-donor patients (414 and 46-year-old women, and 53and 57-year-old men). All of these patients were free of any 5 cardiac abnormalities and had not received any medication except for plasma volume 6 expanders, dobutamine and furosemide. The cause of death included cerebral contusion, 7 cerebral hemorrhage and subarachnoidal hemorrhage. All biopsies were transported in 8 cardioplegic solution (pH 7.4; in mM: NaCl 110, KCl 16, MgCl2 1.6, CaCl2 1.2, NaHCO3 5) 9 and were frozen in liquid nitrogen and stored at -80 °C at the laboratory. The experiments on 10 human tissues complied in full with the Helsinki Declaration of the World Medical 11 Association and were approved by the Hungarian Ministry of Health (No. 323-8/2005-12 1018EKU) and by the Institutional Ethical Committee at the University of Debrecen, 13 Hungary. 14 15 II. Force measurements in permeabilized cardiomyocyte preparations 16 Force measurements were performed as described previously [43]. In brief, frozen tissue 17 samples were first defrosted and mechanically disrupted in cell isolation solution (Iso) (in 18 mM: KCl 100, ethyleneglycoltetraacetic acid (EGTA) 2, MgCl2 1, Na2ATP 4, imidazole 10; 19 pH 7.0) containing phenylmethylsulfonyl fluoride (PMSF, 0.5 mM, Sigma-Aldrich, St. Louis, 20 MO, USA), leupeptin (40 μM, Sigma, St. Louis, MO, USA) and E-64 (10 μM, Sigma-21 Aldrich, St. Louis, MO, USA) protease inhibitors. The mechanically isolated cells were 22 skinned by incubation in Iso supplemented with 0.5% (v/v) Triton X-100 (Sigma-Aldrich, St. 23 Louis, MO, USA) for 5 min. Triton-X-100 was removed by washing at least three times in 24 Iso (1 ml in each washing step) and the skinned myocytes were kept in cell Iso on ice until 25 7 the measurements. A skinned single cardiomyocyte was mounted between two thin needles, 1 which were attached to a force transducer element (SensoNor, Horten, Norway) and an 2 electromagnetic motor (Aurora Scientific Inc., Aurora, Canada) through the use of silicone 3 adhesive (DAP, Baltimore, MD, USA) for determination of the mechanical parameters. The 4 measurements were performed at 15°C on the stage of a light microscope. The average 5 sarcomere length was adjusted to 2.3 μm. 6 The compositions of the relaxing and activating solutions used during force 7 measurements were calculated as described previously [43]. Both solutions were 8 supplemented with protease inhibitors: leupeptin (40 μM) and E-64 (10 μM). The pCa, i.e. 9 the -log10[Ca2+] values of the relaxing and activating solutions (pH 7.2), were 9.0 and 4.75, 10 respectively. Solutions with intermediate free [Ca2+] levels were obtained by mixing 11 activating and relaxing solutions [44]. Isometric force production was measured after the 12 preparation had been transferred from the relaxing solution to a set of Ca2+-containing 13 solutions. When a steady force level had been reached, the length of the myocyte was 14 reduced by 20% within 2 ms, and the myocyte was then quickly restretched (release-restretch 15 maneuver). As a result, the force first dropped from the peak isometric level to zero 16 (difference = total peak isometric force, Ftotal) and then started to redevelop. About 6 s after 17 the onset of force redevelopment, the cardiomyocyte was returned to the relaxing solution, 18 where the length of the myocyte was again reduced by 20% for 8 s to determine the Ca2+-19 independent passive force component (Fpassive). The Ca2+-activated isometric force (Factive) 20 was calculated by subtracting Fpassive from Ftotal. Factive at submaximal levels of activation was 21 normalized to that at maximal activation (pCa 4.75). Thereafter, the normalized force values 22 were plotted against the Ca2+ concentration of the activating solutions to create a sigmoidal 23 curve, in order to determine the Ca2+ sensitivity of force production (pCa50). Maximal active 24 8 force was also tested at the end of the experiments at pCa 4.75. Experiments that yielded a 1 value below 80% of the initial value were discarded. 2 To determine the mechanical consequences of myofilament protein oxidation, 3 cardiomyocytes were exposed to Iso supplemented with H2O2 (30 μM, Sigma-Aldrich, St. 4 Louis, MO, USA) for 15 min; MPO+H2O2 (8 U/l, Abcam, Cambridge, UK) for 15 min; 5 MPO+H2O2+MPO-I 4-aminobenzhydrazide (50 µM, Cayman Chemicals, Ann Arbor, MI, 6 USA) for 15 min; or MPO+H2O2+Met (10 mM, Sigma-Aldrich, St. Louis, MO, USA) for 15 7 min at 15 °C. The reversibility of MPO+H2O2 evoked effects were examined by the 8 application of the reducing agent dithiotreitol (DTT, Sigma-Aldrich, St. Louis, MO, USA, 10 9 mM, 30 min) to MPO+H2O2-treated cardiomyocytes. Force-pCa relationships and pCa50 10 values were determined before and after the application of these agents. The effects of the 11 applied agents on Factive and Fpassive were expressed relative to their control (untreated, before 12 application of the agent at pCa 4.75 and pCa 9.0, respectively). Changes in Factive and Fpassive 13 upon application of the agents were compared with the force values measured after 14 incubation of the cardiomyocytes in Iso for 15 min (time control). 15 16 III. Measurements of MPO activities 17 MPO chlorination and peroxidation assay kits (Cayman Chemicals, Ann Arbor, MI, USA) 18 were used. The chlorination activity assay utilizes a nonfluorescent substrate (APF, 2-(6-(4-19 aminophenoxy)-3-oxo-3H-xanthen-9-yl)benzoic acid), which is cleaved by the MPO20 generated hypochlorite (OCl-) to produce highly fluorescent fluorescein. The peroxidase 21 activity assay uses a nonfluorescent substrate (ADHP, 10-acetyl-3,7-dihydroxyphenoxazine) 22 which is converted by MPO to the fluorescent resorufin. Fluorescence was detected with a 23 NovoStar Microplate Reader (BMG Labtech, Ortenberg, Germany) at λex 485 nm, λem 520 nm 24 in the chlorination assay, and at λex 544 nm, λem 590 nm in the peroxidase assay. The reaction 25 9 solution contained the nonfluorescent substrate (APF (18 µM) or ADHP (45 µM)), assay 1 buffer (phosphate-buffered saline (PBS), pH 7.4) and H2O2 (30 µM), or MPO+H2O2 (38 U/l), 2 or MPO+H2O2+MPO-I (50 µM) or MPO+H2O2+Met (10 mM). Activities were measured for 3 5 min at 24-s intervals. Fluorescence intensities were fitted by linear regression analysis 4 (before saturation) and the slope of this relation was used to calculate MPO activities. Values 5 were corrected for the background (the activity determined in the absence of MPO). 6 7 IV. Biochemical assays for the identification of oxidative protein modifications 8 1. Ellman’s reaction 9 Overall myofilament SH group content was determined by Ellman’s reaction. Skinned 10 cardiomyocytes were treated with Iso (time control) or with Iso supplemented with H2O2 and 11 MPO as described for the mechanical experiments. Washing steps followed the treatments 12 and the cardiomyocytes were then incubated for 15 min in Ellman’s reagent (5,5’-dithio-13 bis(2-nitrobenzoic acid), DTNB; Sigma-Aldrich, St. Louis, MO, USA), which reacts with 14 myofilament SH groups and produces the yellow 2‐nitro‐5‐thiobenzoic acid (NTB). The 15 absorbance of NTB was measured with NovoStar Microplate Reader at 412 nm. N-Acetyl-L-16 cysteine (NAC, Sigma-Aldrich, St. Louis, MO, USA) was used to calibrate the NTB 17 absorbance in relation to the amount of SH groups. A known concentration of NAC was 18 reacted with Ellman’s reagent and the absorbance at 412 nm, fitted with a single exponential, 19 served as calibration curve. The SH contents in 1-mg lyophilized myocardial samples were 20 calculated from the measured absorbance, the tissue weight and the calibration curve. 21 Measurements were performed in triplicates. 22 23 24 25 16 Effects of MPO+H2O2 on the SH oxidation and carbonylation of myofilament proteins 1 Attempts were made to identify the changes in the oxidative status of myofilament proteins 2 contributing to the MPO-induced cardiomyocyte dysfunction in parallel with the functional 3 measurements. Relative SH contents were determined in human LV skinned cardiomyocytes. 4 The baseline SH content of myofilament proteins in the donor heart samples varied between 5 98.04.6% and 104.13.9% (p=0.35). Ellman’s reaction revealed a small, but significant 6 decrease in the overall amount of SH groups in response to H2O2 (to 90.4±1.5%, p<0.05, 7 n=3) or MPO+H2O2 treatments (to 86.7±4.0%, p<0.01, n=3) (Fig. 4A). An SH group 8 biotinylation assay was applied to identify individual myofibrillar proteins affected by MPO-9 mediated SH oxidation. Samples treated with the oxidative agent DTDP were used as positive 10 controls. H2O2 and MPO+H2O2 lowered the SH content of actin to similar extents (to 11 75.9±7.1%, p<0.01, n=4, and 84.2±4.4%, p<0.05 vs. time control, respectively, n=9) (Fig. 12 4B). In contrast, the SH contents of myosin-binding protein C (MyBP-C, Fig. 4C) and the 13 more compliant (N2BA) and stiffer (N2B) isoforms of the giant sarcomeric protein titin were 14 not affected by these treatments (Fig. 4D-F). Using immunoblots a Tm and an actin 15 containing complex was observed at on approximately 90 kDa molecular weight level under 16 non-reducing conditions (in a buffer not containing β-ME), however, no increase in its 17 intensity and that of Tm and actin could be detected after H2O2 and MPO+H2O2 treatments 18 (Fig. 5). 19 Protein carbonylation assays revealed a modest, but significant increase in the 20 carbonylation of actin upon H2O2 treatment (CI=1.1±0.05, p<0.05 vs. the time control, n=5), 21 which was not further affected by the addition of MPO (CI=1.1±0.05, p=0.1 vs. the time 22 control, n=11) (Fig. 6A). Similarly as for actin, a slight, but significant increase in the 23 carbonyl content of MyBP-C was observed both after H2O2 (CI=1.5±0.2, p<0.05 vs. the time 24 control, n=2) and after MPO+H2O2 application (CI=1.4±0.2, p<0.05 vs. the time control, 25 17 n=4) (Fig. 6B). The extent of carbonyl group formation in the N2BA and N2B titin isoforms 1 remained unaltered after H2O2 or MPO+H2O2 treatment (CI=0.9±0.2 and CI=1.0±0.2 for 2 N2BA; CI=1.0±0.1 and CI=0.9±0.1 for N2B, respectively) (Fig. 6C-E). 3 18 Discussion 1 2 This is the first reported investigation of the direct effects of MPO on the contractile function 3 of single, isolated human myocardial cells. The in vitro model experiments revealed that (1) 4 MPO impairs Ca2+-dependent isometric force generation, increases the Ca2+-independent 5 Fpassive and decreases the Ca2+ sensitivity of force production; (2) the MPO-induced functional 6 changes can be prevented by an MPO-I and the antioxidant Met; (3) the levels of SH 7 oxidation in actin and of carbonylation in actin and MyBP-C are increased by the application 8 of MPO+H2O2 or H2O2 alone; (4) the MPO-evoked functional effects are probably mediated 9 by the chlorinating activity of MPO. 10 Myocardial inflammation and ischemia-reperfusion injury are characterized by 11 enhanced extents of oxidative stress and contractile dysfunction [46]. The application of 12 MPO+H2O2 to human cardiomyocytes appreciably reduced the Ca2+-activated Factive and 13 markedly decreased pCa50. In contrast, H2O2 (30 μM) alone induced a smaller decrease in 14 Factive. Consistent with our findings, a lower concentration of H2O2 (10 μM) did not result in a 15 decrease in the maximal Ca2+-activated force in skinned rat heart preparations [47, 48]. This 16 suggests that the action of H2O2 on contractile force generation is concentration-dependent. 17 Lower concentrations have no measurable effects, whereas higher concentrations affect the 18 cardiomyocyte contractility. The deleterious effect on Factive can be explained by the MPO-19 mediated H2O2-derived production of HOCl. In a previous study, HOCl treatment alone (10 20 μM and 50 μM for 1 min) evoked a significant decrease in the maximum Ca2+-activated force 21 [47], similarly to the result of MPO+H2O2 treatment in the present study. Interestingly, 22 neither the H2O2nor the MPO-induced functional changes were related to any deterioration 23 in the cross-striation pattern of the cardiomyocytes under the light microscope. It is important 24 19 to note, however, that electron microscopy has revealed a myofilament lattice disruption after 1 HOCl treatment [47]. 2 The subtle increase after H2O2 application and the marked elevation in the Ca2+-3 independent Fpassive upon MPO+H2O2 treatment in the present study are consistent with the 4 observations that H2O2 at low (<10 µM) concentration did not alter Fpassive, while HOCl (10 5 μM and 50 μM) induced a significant rise in Fpassive of skinned rat trabeculae [48]. It is well 6 established that the giant sarcomeric protein titin plays a key role in the development of 7 Fpassive in permeabilized cardiomyocytes by acting as a molecular spring in the sarcomere 8 [49]. The cardiomyocyte Fpassive can be modulated by the titin isoform switch (between the 9 short and stiff N2B and the longer and more compliant N2BA isoforms [50]) and by several 10 post-translational modifications, including phosphorylation [51], SH oxidation [52] and 11 potentially carbonylation. One elegant study demonstrated that the oxidative stress-induced 12 formation of disulfide bridges within the titin molecule (N2B unique sequence, N2B-Us) 13 reduced the contour length of the N2B-Us, leading to stiffening of the whole titin molecule 14 [52]. In the present study, neither SH oxidation nor carbonylation of the N2B and N2BA titin 15 isoforms was found to be affected by MPO or H2O2 treatment. This may be explained by the 16 distinct sensitivities of the titin N2B isoform, actin and MyBP-C to oxidative changes based 17 on the differences in their ultrastructures and SH group contents. Our results indicate that 18 modifications other than titin SH oxidation or carbonylation might be responsible for the 19 marked elevation in Fpassive after MPO treatment in human cardiomyocytes. 20 The significant decrease observed in pCa50 after MPO+H2O2 in this study is in marked 21 contrast with the previous finding of an increase in pCa50 in skinned rat trabeculae in 22 response to HOCl treatment [48]. This apparently conflicting result might be explained by (1) 23 the different concentration of HOCl produced by the MPO under our experimental 24 conditions; (2) a difference in susceptibility of the myofilaments to HOCl between the two 25 20 species; and (3) the difference in the experimental setting, permeabilized, single 1 cardiomyocytes presenting a negligible diffusion obstacle in comparison with trabeculae. 2 Further, the pronounced MPO-induced decrease in pCa50 suggests that different myofilament 3 protein modifications occur and contribute to pCa50 in the course of MPO and H2O2 4 treatments. Under these experimental conditions H2O2 more probably induced a structural, 5 rather than a regulatory alteration in the contractile apparatus because pCa50 was not affected. 6 The deleterious effect on the maximal Factive and the modest increase in Fpassive upon H2O2 7 administration implies that the H2O2-induced contractile alterations could be explained by a 8 reduction in the number of force-generating cross-bridges due to the diminished longitudinal 9 transmission of force along the sarcomeres. These findings are consistent with the 10 observations of MacFarlane et al., who exposed the superoxide anion (from which H2O2 11 formed endogenously through spontaneous or superoxide dismutase-catalyzed dismutation) 12 to chemically skinned rat cardiac muscles. They also found a dose-dependent reduction in the 13 maximal Factive without any alteration in the pCa50 and concluded that some aspect of the 14 cross-bridge behavior is particularly vulnerable to superoxide [53]. 15 A substantial number of data indicate that the inhibition of MPO may well be useful 16 in CV pathologies characterized by elevated MPO levels (myocardial inflammation, 17 ischemia-reperfusion injury and acute MI). Thus, despite the fact that MPO-Is may have 18 adverse effects on the function of MPO in the innate host-defense mechanisms, potential 19 therapeutic interventions through which to inhibit MPO have aroused considerable interest 20 [42]. In the present study, both the MPO-I 4-aminobenzhydrazide (50 µM) and the 21 antioxidant amino acid Met (10 mM) were equally able to prevent all of the MPO-evoked 22 deleterious contractile effects in skinned human cardiomyocytes, the latter potentially by 23 scavenging the HOCl generated by MPO. MPO activity assays suggested that the Met-24 inhibited chlorinating activity is responsible for the MPO-evoked functional changes. HOCl 25 21 reacts most rapidly with the sulfur-containing residues (Met and Cys) [54]. It is likely, 1 therefore, that the high concentration of Met used in this study diminished the HOCl-evoked 2 oxidative capacity. The oxidation of Met residues results in the generation of Met-sulfoxide 3 (MetSO), a process that may be reversed by MetSO reductase [55]. Met is therefore 4 considered to play a protective role against the deleterious effects of protein oxidation [28]. 5 Interestingly, the incomplete reversion and oxidation of physiologically relevant Met residues 6 has been shown to contribute to the impaired function of proteins [56], including actin [57]. It 7 is important to note, that other HOCl scavenging substances than Met (e.g. glutathione, 8 taurine and L-ascorbic acid) were also tested recently in HOCl scavenging assays [58]. Given 9 the rapid reaction rates of HOCl with biological materials, however, much higher doses of L-10 ascorbic acid and thiols were required to effectively protect against the direct oxidative 11 damage induced by HOCl. This latter suggests that inhibiting the generation of HOCl may be 12 a better choice than scavenging HOCl after its generation, for amelioration of HOCl induced 13 biological damage. 14 The distinct effect of the reducing agent DTT on Factive and Fpassive after MPO+H2O2 15 treatment found in this study might be explained by different modifications on the structural 16 conformation or functional activity of the contractile and regulatory myofilament proteins. 17 The precise nature of the redox-dependent functional changes upon H2O2 and MPO+H2O2 18 treatment is complex and determined also by the type and site of the induced post-19 translational modifications on individual proteins within the sarcomere [59]. SH residues of 20 Cys can undergo both reversible and irreversible modifications. The reaction between the Cys 21 thiolate anion and H2O2 results in formation of intraor intermolecular disulfide bonds, which 22 is reversible, but further oxidation can generate sulfinic or sulfonic acid, which are 23 considered irreversible alterations [60]. The HOCl-induced protein carbonylation is thought 24 22 to be irreversible, while methionine oxidation can be reversed by MetSO-reductase [28] or 1 can lead to an irreversible product (methionine-sulfone) [55]. 2 The extent of overall SH oxidation observed after MPO treatment in this study was 3 comparable to that in heart tissue slices exposed to high-dose HOCl [27]. There is 4 biochemical evidence that oxidative modifications modulate the architecture of the 5 myofilament protein actin [61] and myosin [62]. In vitro exposure of permeabilized human 6 LV cardiomyocytes to the oxidative agent DTDP resulted in a decrease in maximal Ca2+-7 activated force production with a parallel reduction in the SH content of actin and MLC-1 8 [26]. Consistent with this, in the present study H2O2 decreased the SH content of actin. 9 However, despite the marked reduction in Factive, no additional decrease in this parameter was 10 detected after MPO+H2O2 application, suggesting that SH oxidation may not be the main 11 contributor to the MPO-evoked decrease in Factive under these experimental conditions. 12 Moreover, formation of an actin and a Tm containing protein complex observed in this study 13 is also unlikely to be responsible for the contractile changes observed in the cardiomyocytes 14 after H2O2 and MPO+H2O2 administration. The possible functional consequences of the 15 observed protein complexes require further examinations. 16 In a mouse model of experimental MI, we recently identified the increased 17 carbonylation of actin and myosin heavy chain (MHC) in the infarcted area [2]. Similarly to 18 MPO, in vitro Fenton-based myofilament carbonylation decreased pCa50, irrespectively of the 19 phosphorylation status of the myofilaments. Moreover, pCa50 correlated strongly with the 20 myofilament carbonylation levels. In accord with this, a marked (3-fold) increase in carbonyl 21 group formation in actin was observed after 1 mM, but not after 0.1 mM H2O2 treatment [25]. 22 The application of H2O2 to cardiomyocytes at a concentration higher than 0.1 mM was 23 hindered by its inhibitory effect on the activity of MPO [42]. 30 μM H2O2 lowered Factive in 24 parallel with a slight, but significant increase in the carbonylation of actin and MyBP-C. 25 23 Similarly to SH oxidation, carbonylation of these myofilament proteins was not further 1 affected by the addition of MPO, despite its noteworthy effects on cardiomyocyte active and 2 passive force production. This implies that the physiological effects of MPO-catalyzed 3 oxidative processes are independent of SH group oxidation or carbonylation of human 4 myocardial proteins. 5 Oxidative modifications in the myocardium primarily have been considered to result 6 in reduced force generation, as also demonstrated in the present study. However, recent 7 evidence suggests a more complex picture. Reactive oxygen and nitrogen species can activate 8 protective mechanisms and signaling pathways (redox regulation) [60] or even increase 9 cardiac performance [63]. Mild oxidative stress induced S-nitrosylation at specific Cys 10 residues was shown to be cardioprotective [64]. Subtle increases in ROS production may 11 even enhance cardiac contractility under physiological conditions [65]. Indeed, certain 12 oxidative myofilament modifications can lead to positive functional consequences, such as 13 nitroxyl (HNO), a reactive nitrogen species related to nitric oxide, induces formation of actin-14 Tm heterodimers, which correlates with the increase in Ca2+ sensitivity and dimeric forms of 15 MHC and MLC-1, which are associated with increased force generation [63]. HNO was also 16 shown to increase maximum tension and Ca2+ sensitivity of trabeculae sarcomeres 17 functioning in situ [66]. These results strongly suggest that the beneficial or deleterious 18 functional outcome is likely dictated by the strength and the nature of the oxidizing agent and 19 the redox milieu of the myofilament compartment. 20 Since isolation of cardiomyocytes and assessment of myofilament properties was 21 performed on LV biopsies of unused donor hearts, possible changes in the phosphorylation 22 and oxidative status of the myofilament proteins occurring before or during tissue sampling 23 may have been interfered with the results of this study. In addition, activation of the β-24 adrenergic signaling and various oxidative pathways might also influence the baseline 25 24 mechanical and biochemical characteristics of the cardiomyocytes. We have checked the 1 baseline functional parameters of the cells in the study and found no major differences in the 2 cardiomyocyte mechanical properties. Moreover, the baseline myofilament SH contents were 3 also similar in the LV samples used for the cardiomyocyte isolation. These observations are 4 in line with those found in our previous study, in which the reducing agent DTT did not affect 5 Factive and pCa50 of cardiomyocytes derived from human donor hearts [26]. 6 In this study LV heart samples were frozen and their functional and biochemical 7 properties were evaluated upon thawing. To validate the use of defrosted biopsy samples, in 8 one of our previous studies [67] force recordings of cardiomyocytes isolated from a biopsy 9 sample immediately after procurement were compared to those of cardiomyocytes isolated 10 from a defrosted biopsy of the same patient. These force recordings yielded identical results. 11 In addition, the extent of tissue heterogeneity was also addressed in previous studies using 12 explanted hearts [68, 69] or surgically procured biopsies [70]. In these studies the variability 13 of force measurements of cardiomyocytes isolated from different portions of the heart was 14 always less than 5%. 15 It is also important to note that several additional MPO-sensitive processes, such as 16 protein halogenation [71], protein nitration [72], Met oxidation, sulfonic acid generation 17 (Cys), [73] or protein degradation [28], might be responsible for the observed functional 18 alterations. Further studies are clearly required to elucidate the relative contributions of these 19 processes to the overall pump function during human cardiac pathologies associated with 20 elevated MPO levels. 21 22 Conclusion 23 MPO-derived oxidants contribute to myocardial contractile dysfunction by decreasing 24 the cardiomyocyte force production and the myofilament Ca2+ sensitivity and increasing 25 25 Fpassive in human cardiomyocytes. These effects could be prevented by MPO inhibition and 1 the antioxidant Met. The associated functional and biochemical alterations may provide a 2 pharmacological tool for the prevention and/or reversion of MPO-induced contractile protein 3 alterations, which could have therapeutic implications in cardiac pathologies characterized by 4 elevated MPO levels. 5 6 7 8 9 10 11 32 [44] Fabiato, A.; Fabiato, F. Calculator programs for computing the composition of the 1 solutions containing multiple metals and ligands used for experiments in skinned muscle 2 cells. J Physiol (Paris) 75:463-505; 1979. 3 [45] Canton, M.; Menazza, S.; Sheeran, F. L.; Polverino de Laureto, P.; Di Lisa, F.; Pepe, 4 S. Oxidation of myofibrillar proteins in human heart failure. J Am Coll Cardiol 57:300-309; 5 2011. 6 [46] Vinten-Johansen, J. Involvement of neutrophils in the pathogenesis of lethal 7 myocardial reperfusion injury. Cardiovasc Res 61:481-497; 2004. 8 [47] Miller, D. J.; MacFarlane, N. G. Intracellular effects of free radicals and reactive 9 oxygen species in cardiac muscle. J Hum Hypertens 9:465-473; 1995. 10 [48] MacFarlane, N. G.; Miller, D. J. Effects of the reactive oxygen species hypochlorous 11 acid and hydrogen peroxide on force production and calcium sensitivity of rat cardiac 12 myofilaments. Pflugers Arch 428:561-568; 1994. 13 [49] van Heerebeek, L.; Franssen, C. P.; Hamdani, N.; Verheugt, F. W.; Somsen, G. A.; 14 Paulus, W. J. Molecular and cellular basis for diastolic dysfunction. Curr Heart Fail Rep 15 9:293-302; 2012. 16 [50] LeWinter, M. M.; Granzier, H. Cardiac titin: a multifunctional giant. Circulation 17 121:2137-2145; 2010. 18 [51] Borbely, A.; van Heerebeek, L.; Paulus, W. J. Transcriptional and posttranslational 19 modifications of titin: implications for diastole. Circ Res 104:12-14; 2009. 20 [52] Grutzner, A.; Garcia-Manyes, S.; Kotter, S.; Badilla, C. L.; Fernandez, J. M.; Linke, 21 W. A. Modulation of titin-based stiffness by disulfide bonding in the cardiac titin N2-B 22 unique sequence. Biophys J 97:825-834; 2009. 23 33 [53] MacFarlane, N. G.; Miller, D. J. Depression of peak force without altering calcium 1 sensitivity by the superoxide anion in chemically skinned cardiac muscle of rat. Circ Res 2 70:1217-1224; 1992. 3 [54] Szuchman-Sapir, A. J.; Pattison, D. I.; Ellis, N. A.; Hawkins, C. L.; Davies, M. J.; 4 Witting, P. K. Hypochlorous acid oxidizes methionine and tryptophan residues in myoglobin. 5 Free Radic Biol Med 45:789-798; 2008. 6 [55] Vogt, W. Oxidation of methionyl residues in proteins: tools, targets, and reversal. 7 Free Radic Biol Med 18:93-105; 1995. 8 [56] Oien, D. B.; Canello, T.; Gabizon, R.; Gasset, M.; Lundquist, B. L.; Burns, J. M.; 9 Moskovitz, J. Detection of oxidized methionine in selected proteins, cellular extracts and 10 blood serums by novel anti-methionine sulfoxide antibodies. Arch Biochem Biophys 485:35-11 40; 2009. 12 [57] Dalle-Donne, I.; Rossi, R.; Giustarini, D.; Gagliano, N.; Di Simplicio, P.; Colombo, 13 R.; Milzani, A. Methionine oxidation as a major cause of the functional impairment of 14 oxidized actin. Free Radic Biol Med 32:927-937; 2002. 15 [58] Zhongwei Liu, Y. Y., Suhua Wang, Wei-Yi Ong, Choon Nam Ong, Dejian Huang. 16 Assaying myeloperoxidase inhibitors and hypochlorous acid scavengers in HL60 cell line 17 using quantum dots. American Journal of Biomedical Sciences 5:140-153; 2013. 18 [59] Steinberg, S. F. Oxidative stress and sarcomeric proteins. Circ Res 112:393-405; 19 2013. 20 [60] Chung, H. S.; Wang, S. B.; Venkatraman, V.; Murray, C. I.; Van Eyk, J. E. Cysteine 21 oxidative posttranslational modifications: emerging regulation in the cardiovascular system. 22 Circ Res 112:382-392; 2013. 23 [61] Milzani, A.; DalleDonne, I.; Colombo, R. Prolonged oxidative stress on actin. Arch 24 Biochem Biophys 339:267-274; 1997. 25 34 [62] Wilson, G. J.; dos Remedios, C. G.; Stephenson, D. G.; Williams, D. A. Effects of 1 sulphydryl modification on skinned rat skeletal muscle fibres using 5,5'-dithiobis(2-2 nitrobenzoic acid). J Physiol 437:409-430; 1991. 3 [63] Gao, W. D.; Murray, C. I.; Tian, Y.; Zhong, X.; DuMond, J. F.; Shen, X.; Stanley, B. 4 A.; Foster, D. B.; Wink, D. A.; King, S. B.; Van Eyk, J. E.; Paolocci, N. Nitroxyl-mediated 5 disulfide bond formation between cardiac myofilament cysteines enhances contractile 6 function. Circ Res 111:1002-1011; 2012. 7 [64] Kohr, M. J.; Evangelista, A. M.; Ferlito, M.; Steenbergen, C.; Murphy, E. S-8 nitrosylation of TRIM72 at cysteine 144 is critical for protection against oxidation-induced 9 protein degradation and cell death. J Mol Cell Cardiol 69:67-74; 2014. 10 [65] Kubin, A. M.; Skoumal, R.; Tavi, P.; Konyi, A.; Perjes, A.; Leskinen, H.; Ruskoaho, 11 H.; Szokodi, I. Role of reactive oxygen species in the regulation of cardiac contractility. J 12 Mol Cell Cardiol 50:884-893; 2011. 13 [66] Dai, T.; Tian, Y.; Tocchetti, C. G.; Katori, T.; Murphy, A. M.; Kass, D. A.; Paolocci, 14 N.; Gao, W. D. Nitroxyl increases force development in rat cardiac muscle. J Physiol 15 580:951-960; 2007. 16 [67] Borbely, A.; van der Velden, J.; Papp, Z.; Bronzwaer, J. G.; Edes, I.; Stienen, G. J.; 17 Paulus, W. J. Cardiomyocyte stiffness in diastolic heart failure. Circulation 111:774-781; 18 2005. 19 [68] van Der Velden, J.; Klein, L. J.; Zaremba, R.; Boontje, N. M.; Huybregts, M. A.; 20 Stooker, W.; Eijsman, L.; de Jong, J. W.; Visser, C. A.; Visser, F. C.; Stienen, G. J. Effects of 21 calcium, inorganic phosphate, and pH on isometric force in single skinned cardiomyocytes 22 from donor and failing human hearts. Circulation 104:1140-1146; 2001. 23 [69] van der Velden, J.; Papp, Z.; Zaremba, R.; Boontje, N. M.; de Jong, J. W.; Owen, V. 24 J.; Burton, P. B.; Goldmann, P.; Jaquet, K.; Stienen, G. J. Increased Ca2+-sensitivity of the 25 35 contractile apparatus in end-stage human heart failure results from altered phosphorylation of 1 contractile proteins. Cardiovasc Res 57:37-47; 2003. 2 [70] van der Velden, J.; Klein, L. J.; van der Bijl, M.; Huybregts, M. A.; Stooker, W.; 3 Witkop, J.; Eijsman, L.; Visser, C. A.; Visser, F. C.; Stienen, G. J. Isometric tension 4 development and its calcium sensitivity in skinned myocyte-sized preparations from different 5 regions of the human heart. Cardiovasc Res 42:706-719; 1999. 6 [71] Davies, M. J. Myeloperoxidase-derived oxidation: mechanisms of biological damage 7 and its prevention. J Clin Biochem Nutr 48:8-19; 2011. 8 [72] Yan, Z.; Liang, F.; Guo, L.; Wang, J.; Wang, X. L.; Cheng, X. L.; Ma, X. L.; Liu, H. 9 R. Myeloperoxidase increased cardiomyocyte protein nitration in mice subjected to nonlethal 10 mechanical trauma. Biochem Biophys Res Commun 393:531-535; 2010. 11 [73] Hawkins, C. L.; Morgan, P. E.; Davies, M. J. Quantification of protein modification 12 by oxidants. Free Radic Biol Med 46:965-988; 2009. 13 14 15 16 36 Figure captions 1 2 FIG. 1. Myeloperoxidase (MPO) and hydrogen peroxide (H2O2) impair the force 3 generation of human permeabilized cardiomyocytes. (A) A single cardiomyocyte (isolated 4 from a human left ventricle myocardium) mounted between a sensitive force transducer and 5 an electromagnetic motor. (B) Original force recordings of maximal Ca2+-activated active 6 (Factive) and Ca2+-independent passive (Fpassive) force components before (left panel) and after 7 MPO+H2O2 treatment (right panel) at pCa (i.e -log10[Ca2+]) 4.75 and pCa 9.0, respectively. 8 MPO + H2O2 were applied in Iso for 15 min. (C) pCa-force relationships determined before 9 and after H2O2 or MPO+H2O2 treatments (number of cardiomyocytes, n=7 and 12, 10 respectively). Force levels are expressed relative to the values measured before the 11 treatments. (* vs. Before H2O2, # vs. Before MPO+H2O2,  vs. After H2O2; *,#,p<0.05) (D) 12 Changes in Fpassive measured in the presence of Iso and after sequential applications of H2O2 13 or MPO+H2O2. (E) Significant rightward shift (i.e. decrease in the Ca2+ sensitivity of force 14 production (pCa50)) in the normalized pCa-force relationships in response to MPO+H2O2, but 15 no change after H2O2 treatment (F). (Data are expressed as mean±SEM.) 16 17 FIG. 2. Similar effects of the MPO inhibitor (MPO-I), but distinct actions of methionine 18 (Met) on the chlorinating and peroxidase activities of myeloperoxidase (MPO). Met 19 inhibits the chlorinating (A), but not the peroxidase (B) activity of MPO. Values are 20 expressed relative to the MPO activity measured in the presence of Iso and hydrogen 21 peroxide (H2O2). (Data are expressed as mean±SEM, *p<0.05). 22 23 FIG. 3. The myeloperoxidase inhibitor (MPO-I) and methionine (Met) prevent the 24 MPO-induced changes in isometric force production of human cardiomyocytes. 25 Maximal (pCa 4.75) Ca2+-dependent active (Factive) (A) and Ca2+-independent (pCa 9) passive 26 37 (Fpassive) force (B) in left ventricular cardiomyocytes treated in isolating solution (Iso) 1 supplemented with hydrogen peroxide (H2O2) or myeloperoxidase (MPO)+H2O2, MPO-I or 2 Met. Forces are expressed relative to the values measured before the subsequent treatments. 3 The MPO-I (C) and Met (D) prevent the MPO-evoked rightward shift in the normalized pCa-4 force relationships. Dashed lines indicate force-pCa relationships determined in Iso. (E) 5 Changes in the Ca2+ sensitivity of force production (pCa50) upon H2O2, MPO+H2O2, MPO-I 6 or Met treatments. (Data are expressed as mean±SEM, *p<0.05) 7 8 FIG. 4. Myeloperoxidase (MPO) and hydrogen peroxide (H2O2) similarly alter 9 sulfhydryl (SH) group oxidation in myofilament proteins. (A) SH group oxidation in a 10 cardiomyocyte suspension treated in isolating solution (Iso) supplemented with H2O2 and 11 MPO (Ellman’s reaction). (B-E) Representative examples of SH content determination in 12 actin (B), myosin-binding protein C (MyBP-C) (C), N2BA (D, E) and N2B (D, F) titin 13 isoforms after H2O2 or MPO+H2O2 treatments through use of a protein biotinylation assay. 14 T2 indicates the titin degradation product. Samples exposed to dithiodipyridine (DTDP, 2.5 15 mM, for 2 min) were used as positive control. Total protein amount was determined with the 16 Sypro Ruby Protein Blot Stain. Values are expressed relative to the SH group content 17 determined in Iso (time control). (Data are expressed as mean±SEM, *p<0.05 vs. Iso.) 18 19 FIG. 5. No additional disulfide cross-bridge formation after hydrogen-peroxide (H2O2) 20 and myeloperoxidase (MPO) treatment. Left ventricular myocardial samples solubilized in 21 non-reducing (-β-mercaptoethanol (β-ME)) or reducing (+β-ME) sample buffers and probed 22 with anti-tropomyosin (Tm) (left panel) and anti-actin (right panel) antibodies after 23 immunoblotting. (Protein amount was determined with the Sypro Ruby Protein Blot Stain, 24 MW - molecular weight.) 25 38 1 FIG. 6. Myeloperoxidase (MPO) and hydrogen peroxide (H2O2) increase the 2 carbonylation of actin and myosin-binding protein C (MyBP-C), but not that of titin. 3 Representative examples and measurement of carbonyl group formation in actin (A), MyBP-4 C (B), N2BA (C, D) and N2B (C, E) titin isoforms treated with isolating solution (Iso) 5 supplemented with H2O2 or MPO+H2O2. Left ventricular myocardial samples treated with 6 Fenton reagent (FeSO4, H2O2 and ascorbic acid) served as positive control. Protein 7 carbonylation is expressed as carbonylation index (CI) (CI=1, carbonyl group content 8 measured in Iso). Total protein amount was determined with the Sypro Ruby Protein Blot 9 Stain. (Data are expressed as mean±SEM, *p<0.05) 10 11