Microcin J25 triggers cytochrome c release through irreversible damage of mitochondrial proteins and lipids

The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281

Contents lists available at ScienceDirect

The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel

Microcin J25 triggers cytochrome c release through irreversible damage of mitochondrial proteins and lipids María V. Niklison-Chirou a , Fernando Dupuy a , Liliana B. Pena b , Susana M. Gallego b , Maria Laura Barreiro-Arcos c , Cesar Avila a , Clarisa Torres-Bugeau a , Beatriz E. Arcuri a , Augusto Bellomio a , Carlos Minahk a , Roberto D. Morero a,∗ a Departamento de Bioquímica de la Nutrición, Instituto Superior de Investigaciones Biológicas (Consejo Nacional de Investigaciones Científicas y Técnicas— Universidad Nacional de Tucumán), Instituto de Química Biológica “Dr. Bernabe Bloj”, Chacabuco 461, San Miguel de Tucumán 4000, Argentina b Departamento de Química Biológica, Facultad de Farmacia y Bioquímica, Universidad de Buenos Aires, Buenos Aires, Argentina c Centro de Estudios Farmacológicos y Botánicos CEFYBO-CONICET, Facultad de Medicina, Universidad de Buenos Aires, Buenos Aires, Argentina

a r t i c l e

i n f o

Article history: Received 25 February 2009 Received in revised form 23 September 2009 Accepted 5 November 2009 Available online 13 November 2009 Keywords: Mitochondria Reactive oxygen species Microcin Cardiolipin Cytochrome c

a b s t r a c t We previously showed that the antimicrobial peptide microcin J25 induced the over-production of reactive oxygen species with the concomitant release of cytochrome c from rat heart mitochondria via the opening of the mitochondrial permeability transition pore. Here, we were able to demonstrate that indeed, as a consequence of the oxidative burst, MccJ25 induces carbonylation of mitochondrial proteins, which may explain the irreversible inhibition of complex III and the partial inhibition of superoxide dismutase and catalase. Moreover, the peptide raised the levels of oxidized membrane lipids, which triggers the release of cytochrome c. From in silico analysis, we hypothesize that microcin would elicit these effects through interaction with heme c1 at mitochondrial complex III. On the other hand, under an excess of l-arginine, MccJ25 caused nitric oxide overproduction with no oxidative damage and a marked inhibition in oxygen consumption. Therefore, a beneficial anti-oxidative activity could be favored by the addition of l-arginine. Conversely, MccJ25 pro-oxidative–apoptotic effect can be unleashed in either an arginine-free medium or by suppressing the nitric oxide synthase activity. © 2009 Elsevier Ltd. All rights reserved.

1. Introduction Microcin J25, a 21-amino acid antimicrobial peptide produced by an Escherichia coli strain, is active against close-related bacteria including certain human pathogens as some Salmonella and Shigella strains (Salomon and Farias, 1992). The peptide has an unusual lasso distinctive structure (Bayro et al., 2003; Rosengren et al., 2003; Wilson et al., 2003) and it is bacteriostatic agent in E. coli by inhibiting RNA polymerase causing an impaired transcription of genes encoding cell division proteins (Delgado et al., 2001). An alternative mechanism of action has been described on Salmonella serovar Newport cells (Rintoul et al., 2001). It was demonstrated that this

Abbreviations: MTP, mitochondrial transition pore; MccJ25, microcin J25; ROS, reactive oxygen species; RNS, reactive nitrogen species; HbO2 , oxyhemoglobin; BSA, bovine serum albumin; TCA, trichloroacetic acid; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; l-NMMA, l-NG-monomethyll-arginine; RCR, respiratory control rate;  , transmembrane electrical potential; ATP, adenosine-5 -triphosphate; ADP, adenosine-5 -diphosphate; SMP, submitochondrial particles; Cu–Zn SOD, copper–zinc superoxide dismutase; 2,4 DNPH, 2,4 dinitrophenyl hydrazine. ∗ Corresponding author. Tel.: +54 0381 4248921; fax: +54 0381 4248025. E-mail address: [email protected] (R.D. Morero). 1357-2725/$ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2009.11.002

antibiotic disrupts membrane integrity and therefore causes a dissipation of the membrane electrical potential in Salmonella serovar Newport cells. Furthermore, MccJ25 inhibits NADH and succinate dehydrogenase and alters the oxygen consumption rate (Rintoul et al., 2001). As a result, MccJ25 is a bactericidal peptide rather than a bacteriostatic one in these bacteria. In addition, an superoxide anion production was described in E. coli (Bellomio et al., 2007). It was recently shown that ROS was generated as a result of MccJ25–plasma membrane interaction. The ability of MccJ25 to interact with bacterial membranes is supported by studies carried out in liposomes (Rintoul et al., 2000). Moreover, MccJ25 is able to penetrate phospholipid monolayers at air–water interface in the absence of energy driven transport mechanism (Bellomio et al., 2005). Recently, the effect of MccJ25 on intact rat heart mitochondria was explored (Niklison Chirou et al., 2004). The peptide displays a potent effect as inhibitor of the complex III, disrupts the  and drastically diminishes the internal ATP level. Moreover, we confirmed that MccJ25 induces superoxide overproduction, thus increasing the mitochondrial inner membrane permeability and activating the mitochondrial transition pore, resulting in swelling and cytochrome c release (Niklison Chirou et al., 2008).

274

M.V. Niklison-Chirou et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281

In the present work, we were able to show that MccJ25 induced carbonylation of mitochondrial proteins. Moreover, the peptide was able to oxidize membrane lipids, in particular cardiolipin, which would trigger the release of cytochrome c. A preliminary approach using molecular docking let us speculate that microcin may locate in close contact to heme c1 unleashing the oxidative damage as well as preventing cytochrome c reduction. However, under an excess of l-arginine, MccJ25 switched targets and stimulated mtNOS with no mitochondrial damage. 2. Materials and methods 2.1. Materials MccJ25 was routinely purified from the supernatants of E. coli MC4100 cultures as described elsewhere (Bellomio et al., 2003). Rabbit anti-cytochrome c, ADP and ATP were purchased from Sigma (St. Louis, MO, USA). Cardiolipin was purchased from Avanti Polar Lipids (Alabaster, AL, USA). All other reagents used were of analytical grade or the purest available commercial form. 2.2. Mitochondrial isolation and preparation of submitochondrial membranes Wistar rats (250–300 g) were killed by CO2 inhalation, in accordance with the European directive for protection of vertebrate animals for scientific research. Rat hearts were homogenized in an ice-cold homogenization medium consisting of 0.23 M mannitol, 70 mM sucrose, 1 mM EGTA and 10 mM Tris–HCl (pH 7.4) with an Omni-Mixer (Sorvall, Norwalk, CT, USA). The homogenate was centrifuged at 900 × g for 10 min to discard nuclei and cell debris, and the supernatant was centrifuged at 17,000 × g for 10 min. The pellet, was washed and resuspended in the isolation buffer without EGTA (Niklison Chirou et al., 2008). In order to prepare submitochondrial particles, the mitochondrial pellet was resuspended (20 mg/ml) in 50 mM Tris–HCl (pH 7.6), 230 mM mannitol, 70 mM sucrose and sonicated three times, each consisting of a 30-s pulse burst, with 1-min intervals at 4 ◦ C. The sonicated mitochondria were centrifuged at 8500 × g for 10 min to remove the unbroken organelles. The supernatant was centrifuged again at 100,000 × g for 60 min, and the resulting pellet washed and resuspended in the same buffer (Boveris, 1984). Protein concentration was assayed by the Folin reagent with bovine serum albumin as standard (Lowry et al., 1951). 2.3. Mitochondrial oxygen consumption Oxygen uptake was determined with a Clark electrode in a 2 ml chamber at 30 ◦ C, in an air-saturated reaction medium consisting of 0.23 M mannitol, 70 mM sucrose, 20 mM Tris–HCl, pH 7.4, 5 mM potassium phosphate, 4 mM MgCl2 and 0.5 mg mitochondrial protein/ml. Respiratory rates were determined with 10 mM succinate as substrate for state 4, whereas state 3 was established by addition of 10 mM succinate and 1 mM ADP. Oxygen uptake was expressed as nanograms of oxygen per minute per milligram of membrane protein (Navarro et al., 2005).

SOD activity was analyzed using nitroblue tetrazolium and riboflavin (Beauchamp and Fridovich, 1971). Catalase activity was tested using Amplex red reagent. Briefly, mitochondria were treated with or without MccJ25 as described above, then organelles were sonicated and incubated with 10 ␮M H2 O2 for 10 min at 37 ◦ C. After incubation, 200 ␮M Amplex red and 0.4 U/ml horseradish peroxidase were added and incubated 5 min. The resofurin fluorescence produced by remaining H2 O2 was detected at 590 nm. Excitation wavelength was set at 530 nm. 2.5. Biochemical markers of oxidative stress Carbonylated proteins were determined in mitochondrial membranes as originally described by Reinheckel et al. (2000). Briefly, mitochondria (2 mg protein/ml) suspended in 100 mM potassium phosphate buffer containing 10 mM succinate were supplemented with 50 ␮l of 10% trichloroacetic acid; the precipitated proteins were suspended in 50 ␮l of 0.2% 2,4-dinitrophenyl hydrazine, incubated 1 h at 37 ◦ C, precipitated again with TCA. The precipitated was washed with ethanol:ethyl acetate (50:50), dissolved in 6 M guanidine hydrochloride in phosphate buffer (pH 6.5), and the absorbance determined at 370 nm (ε: 21 mM−1 cm−1 ). Protein carbonyls were expressed as picomoles per milligram of mitochondrial protein. For Western blot detection of carbonylated proteins, mitochondrial extracts (250 ␮g protein) that were previously derivatized with 2,4-dinitrophenylhydrazine (DNPH) (Levine et al., 1994) were separated by SDS-PAGE using 12% (w/v) running and 4% (w/v) stacking polyacrylamide gels, respectively (Laemmli, 1970). Two gels were run simultaneously: one for protein staining with Coomassie Brilliant Blue R-250 and the other for immunodetection. Derivatized proteins were transferred onto nitrocellulose membranes and were detected with rabbit anti-DNP primary antibody from Sigma–Aldrich (St. Louis, USA). Bands corresponding to oxidized proteins were visualized by secondary goat anti-rabbit immunoglobulins conjugated with horseradish peroxidase (DAKO), using 3,3 -diaminobenzidine (DAB) as substrate. Gels and membranes were photographed with Fotodyne equipment. Carbonylated cytochrome c was investigated as follows: mitochondrial proteins (250 ␮g) derivatized with DNPH as mentioned above, were separated by affinity chromatography. Antibodies antiDNP (50 ␮l) were linked to cyanogen bromide activated Sepharose 4% agarose matrix (100 ␮g) from Sigma–Aldrich. Samples were incubated overnight at 4 ◦ C with an excess of anti-DNP-agarose resin and then centrifuged for 5 min at 10,000 × g. Resin beads were washed 3 times with Tris-buffered saline and finally re-suspended in 100 mM glycine–HCl (50 ␮l, pH 2.5). After centrifugation, the pellets were discarded, the supernatants adjusted to pH 6.8 with 0.5 M Tris–HCl buffer (5 ␮l, pH 8.8) and used for immunodetection of the cytochrome c. DNPH derivatized proteins were separated by 15% (w/v) SDS-PAGE. The proteins were electrotransfered onto polyvinylidene difluoride membranes and the cytochrome c was detected using anti-cytochrome c primary antibodies (Santa Cruz) and goat anti-rabbit immunoglobulins horseradish peroxidase conjugate (DAKO), with DAB as substrate. Membranes were photographed with Fotodyne equipment. 2.6. Determination of NADPH oxidation

2.4. Enzyme assays Complex III activity was spectrophotometrically assayed following the changes of cytochrome c absorption at 550 nm (Niklison Chirou et al., 2004). Mitochondria were incubated with or without MccJ25 during 60 min at 37 ◦ C with constant stirring. Then activities were measured before and after washing the organelles in order to determine the reversibility of the inhibition.

Mitochondrial NADPH was monitored at 25 ◦ C by measuring its intrinsic fluorescence at 450 nm (exc 340 nm), in an ISS PC1 spectrofluorometer (Niklison Chirou et al., 2008). Mitochondria were suspended in Tris–potassium phosphate (pH 7.4), 150 mM sucrose, 50 mM KCl, 10 mM succinate, 1 ␮M rotenone and preincubated for 1 min in the presence or absence of 0.5 ␮M stigmatellin. Oxidation of NADPH was started by adding 20 ␮M MccJ25.

M.V. Niklison-Chirou et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281

2.7. Spectrophotometric determination of NO Mitochondrial NO production was determined by the oxyhemoglobin oxidation assay as described elsewhere (Navarro et al., 2005). SMP (0.5 mg protein/ml) were resuspended in 50 mM phosphate buffer, pH 7.2 supplemented with 0.1 mM NADPH, 0.2 mM l-arginine, 1 mM CaCl2 , 4 ␮M Cu, Zn-SOD, 0.1 ␮M catalase and 25 ␮M HbO2 heme, at 37 ◦ C. A diode array spectrophotometer (DU7500, Beckman, Fullerton, CA, USA) was used to follow the absorbance change at 577 nm with a reference wavelength at the isosbestic point of 591 nm (ε577–591 : 11.2 mM/cm). Production of NO was calculated from the absorbance change and expressed in nanomoles NO per minute per milligram of protein. As a control, the NO production was inhibited by 2 mM l-NG-monomethyl-larginine. For Western blot detection of nitrotyrosine-containing proteins, mitochondrial extracts (30 ␮g protein) from MccJ25-treated and untreated samples were separated by SDS-PAGE using 10% (w/v) running and 4% (w/v) stacking polyacrylamide gels, respectively (Laemmli, 1970). Two gels were run simultaneously: one for protein staining with Coomassie Brilliant Blue R-250 and the other for immunodetection. Proteins were transferred onto PVDF membranes and were detected with polyclonal anti-nitrotyrosine antibody from Sigma–Aldrich. Bands were visualized with a secondary monoclonal antibody conjugated with horseradish peroxidase (Sigma–Aldrich) by chemiluminescence. Gels and membranes were photographed with Fotodyne equipment. 2.8. Conjugated dienes Mitochondria were resuspended in the presence or the absence of MccJ25 in 0.23 M mannitol, 70 mM sucrose, 5 mM potassium phosphate, 4 mM MgCl2 , 0.2% BSA, 20 mM Tris–HCl, pH 7.4, at a final concentration of 1 mg/ml. Aliquots were taken at different time points upon addition of 10 mM succinate. Lipids were extracted (Folch et al., 1957) and dried under nitrogen. The total mitochondrial lipids were dissolved in chloroform and separated by thin-layer chromatography (chloroform/acetone/methanol/acetic

275

acid/water, 6:8:2:2:1). Following iodine visualization, cardiolipin band was scraped out. The total lipids or cardiolipins were dissolved in n-hexane and absorbance measured at 233 nm in order to determine total dienes (Petrosillo et al., 2003). 2.9. Molecular docking study of cytochrome bc1–MccJ25 interaction The tridimensional structure of bovine cytochrome bc1 resolved by X-ray diffraction at 2.1 Å, PDB accession 1PPJ (Huang et al., 2005) as well as the NMR-solution structure of MccJ25, 1PP5 accession code (Bayro et al., 2003) were used for docking calculations. As a first approach, a rigid-body docking was performed using PatchDock server (Schneidman-Duhovny et al., 2008), based on shape complementarity principles. The structure was then refined by side-chain rearrangement and soft rigid-body optimization using FireDock server (Andrusier et al., 2007; Mashiach et al., 2008). The figure and the calculation of distances were prepared using VMD 1.8.7 (Humphrey et al., 1996). MccJ25–cytochrome bc1 contacts and surfaces were estimated with PISA server (Krissinel and Henrick, 2007). 3. Results 3.1. Microcin J25 uncouples mitochondria Since it was previously shown that MccJ25 greatly inhibited ATP production in mitochondria (Niklison Chirou et al., 2004), we decided to study the respiratory control rate. As it can be seen in Fig. 1, MccJ25 decreased RCR in a concentration-dependent manner, which leads to an uncoupling between ATP synthesis and respiration (Faramarzi-Roques et al., 2004). Actually, both state 3 and 4 respiration rate were increased in the presence of 20 ␮M MccJ25. However, there was a 3-fold increase for state 4, while state 3 was increased by 30% (Fig. 1) suggesting that mitochondria were under an oxidative stress. This result is in agreement with our previous finding of ROS production induced by MccJ25 (Niklison Chirou et al., 2008). This effect was transient though. Indeed, the respiration was greatly reduced after 1 h of incubation (data not shown). The reactive oxygen species overproduction reported previously upon addition of MccJ25 (Niklison Chirou et al., 2008) is likely connected to this oxidative burst. 3.2. Carbonylated proteins are produced by MccJ25

Fig. 1. MccJ25 increases oxygen consumption in isolated mitochondria. Energized mitochondria (0.5 mg mitochondrial protein/ml) were suspended in buffer 0.23 M mannitol, 70 mM sucrose, 20 mM Tris–HCl, pH 7.4, 5 mM potassium phosphate, 4 mM MgCl2 , and incubated with different concentrations of MccJ25 at 30 ◦ C. State 3 () was induced by the addition of 10 mM succinate and 1 mM ADP and state 4 () by the addition of 10 mM succinate only. Oxygen consumption rate was determined as described in Section 2, and expressed as nanograms of oxygen per minute per milligram protein membrane. RCR () was calculated as the state 3/state 4 ratio. Results are expressed as mean ± SD of five independent experiments.

ROS produced during mitochondrial oxidative stress can target proteins, which in turn may end up heavily oxidized. The carbonylation of proteins is a marker of a severe and irreversible damage (Nystrom, 2005). Carbonylation of proteins could be detected when mitochondria were treated with 20 ␮M MccJ25. The oxidation of proteins had no lag phase and was a concentration-dependent process (Fig. 2A and B). Control showed negligible carbonyls production. In order to confirm these findings, DNPH-derivatized proteins were resolved by SDS-PAGE and visualized after Western blot using a polyclonal anti-DNP antibody as described in Section 2. Once again, we observed a concentration-dependent effect. It is important to note that many proteins were carbonylated after incubation with MccJ25, indicating a non-specific protein damage. The strongest signals could be found in proteins raging from 45 to 70 kDa as well as proteins with a relative mass of 30 kDa (Fig. 3A). Not noticeable band appeared in the range of cytochrome c when total proteins were run. A Coomassie staining was performed as a loading control (Fig. 3B). Because this negative result could be due to a low cytochrome c concentration relative to total carbonylated proteins, we decided

276

M.V. Niklison-Chirou et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281

Fig. 2. Protein oxidation in intact mitochondria produced by MccJ25. Activated mitochondria (1 mg/ml) were resuspended in 0.23 M mannitol, 70 mM sucrose, 5 mM potassium phosphate, 4 mM MgCl2 , 0.2% BSA, 20 mM Tris–HCl, pH 7.4 and incubated with 20 ␮M MccJ25 at 37 ◦ C. (A) Aliquots were precipitated with TCA at appropriate times and incubated with 2,4 DNPH. Proteins were resuspended in 6 M guanidine and carbonyls were determined by absorbance at 370 nm. Control (), 20 ␮M MccJ25 (䊉), 40 ␮M MccJ25 () and 60 ␮M MccJ25 (). (B) carbonyls were analyzed at different concentrations of MccJ25 following 30 min of incubation at 37 ◦ C.

to pull down total oxidized proteins first and then visualize cytochrome c by a specific antibody. After this treatment, we were able to confirm that MccJ25 was able to induce carbonylation of cytochrome c (Fig. 3C). Interestingly, stable oligomers of cytochrome c were also observed. Moreover, these oligomers

represent more than 90% of the signal in MccJ25-treated samples, confirming that cytochrome c underwent oxidative damage upon incubation with MccJ25 (Kim et al., 2006). The intensity of oligomeric bands was proportional to the concentration of peptide.

Fig. 3. Effect of MccJ25 on patterns of total oxidized proteins. (A) Western blotting with anti-DNP antibody. (B) Coomassie staining. Extracts (250 ␮g total protein) were subjected to SDS-PAGE (12%, w/v, polyacrylamide), with bands visualized as described in Section 2. Gels and membranes were photographed with a Fotodyn. Lane A: Molecular mass markers (in kDa), lane B: control, lane C: 20 ␮M MccJ25 and lane D: 60 ␮M MccJ25. Electrophoresis and Western blot data shown are representative of several experiments. (C) Proteins were immunoseparated by anti-DNP beads after derivatization with 2,4 DNPH and resolved by SDS-PAGE (15%, w/v). Anti-cytochrome c antibody was used to identify carbonylated cytochrome c, visualized as described in Section 2. Bands were photographed with a Fotodyn. Lane A: Molecular mass marker, lane B: control, lane C: 20 ␮M MccJ25 and lane D: 60 ␮M MccJ25. Western blot data shown are representative of several experiments.

M.V. Niklison-Chirou et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281

277

Table 1 NO production in intact mitochondria induced by 20 ␮M MccJ25. MccJ25 (20 ␮M)

Control State 4 State 3 RCR State 3 + l-arginine State 3 + l-NMMA

110.6 ± 397.8 ± 3.55 223.1 ± 400.1 ±

9.7 12.3 10.1 11.4

244.7 ± 466.9 ± 1.69 114.3 ± 302.2 ±

8.1 19.7 9.2 8.9

of concentrations as high as 100 ␮M of MccJ25 (data not shown). Therefore, MccJ25 damages proteins mainly by MccJ25-induced ROS rather than RNS-mediated modifications. 3.4. MccJ25 induces the formation of conjugated dienes Fig. 4. MccJ25 enhances NO production in SMP. SMP (0.1 mg protein/ml final concentration) was diluted in 0.1 mM NADPH, 0.2 mM l-arginine, 1 mM CaCl2 , 4 ␮M Cu, Zn-SOD, 0.1 ␮M catalase, and 25 ␮M HbO2 heme, in 50 mM phosphate buffer, pH 7.2 at 37 ◦ C. Production of NO was determined spectrophometrically as described in Section 2 upon addition of different concentrations of MccJ25 (black bars). As a control, MccJ25-treated SMP were coincubated with 2 mM l-NMMA (grey bars).

3.3. Nitric oxide production In order to know whether MccJ25 was able to induce the production of reactive nitrogen species as well, l-arginine was added to the reaction; we measured NO production in SMP (0.5 mg protein/ml) by a colorimetric assay. Indeed, MccJ25 was able to stimulate mtNOS activity in a concentration dependent manner (Fig. 4). As we expected, l-NMMA completely inhibited NO production produced by MccJ25. It is important to note that the assay was carried out in the presence of SOD and catalase, therefore there was no interference with ROS. As it was previously reported, NO inhibits mitochondrial respiration (Brown, 1997), therefore we tested the ability of MccJ25 to influence respiration when l-arginine, the substrate of nitric oxide synthase, was present. In fact, a 50% inhibition of O2 consumption was achieved in state 3 when 20 ␮M MccJ25 was added. This finding is in sharp contrast with the effect of MccJ25 alone (see above). l-NMMA prevented this inhibition, confirming that NO was responsible for this effect (Table 1). As expected, the preincubation of isolated mitochondria with 100 ␮M EGTA prevented NO-induced inhibition of oxygen consumption triggered by 20 ␮M MccJ25 (data not shown). In spite of NO production in the presence of l-arginine, we were unable to find nitrated and nitrosylated proteins upon addition

Because ROS were involved with mitochondrial protein damage, we decided to study possible alterations of mitochondrial lipids as well induced by MccJ25, since membrane phospholipids are important targets of these reactive species (Rubbo and Radi, 2008). As it is shown in Fig. 5A, MccJ25 triggered the peroxidation of lipids after a lag phase of approximately 10 min. On the other hand, dienes were not generated in a significant extension up to 80 min of incubation in controls. MccJ25-derived dienes were formed at a greater extent in state 3 as compared to state 4 (Fig. 5B), likely due to the higher rates of respiration in state 3. It is important to note that the addition of l-arginine and MccJ25, i.e. mitochondria with stimulated NO production, induced no peroxidation of lipids (data not shown), indicating that lipid oxidation would depend mainly on MccJ25-induced ROS. Cardiolipin is one of the major phospholipid in the mitochondrial inner membrane. It is prone to oxidation under oxidative stress since it is highly unsaturated. We purified by TLC cardiolipin from both MccJ25-treated and untreated mitochondria and production of dienes was measured in these fractions. As we expected, cardiolipin was oxidized upon addition of 20 ␮M MccJ25. Although some oxidation can be detected in controls, owing the polyunsaturation of cardiolipin, the level of oxidized cardiolipin in MccJ25-treated mitochondria was 2.5 higher than in the controls (Fig. 6). 3.5. MccJ25 inhibits mitochondrial enzymes Since we previously showed that MccJ25 was able to inhibit mitochondrial complex III in vitro (Niklison Chirou et al., 2004), we decided to test whether this effect was reversible or not in intact

Fig. 5. MccJ25 induces lipid peroxidation in mitochondria. Activated mitochondria (1 mg/ml) were resuspended in 0.23 M mannitol, 70 mM sucrose, 5 mM potassium phosphate, 4 mM MgCl2 , 0.2% BSA, 20 mM Tris–HCl, pH 7.4. (A) Time course: mitochondria incubated with () or without () 20 ␮M MccJ25. (B) Lipid peroxidation at different concentrations of MccJ25. Mitochondria in either state 3 () or state 4 () were incubated 30 min at 37 ◦ C with different concentrations of MccJ25. Lipids were extracted and absorbance was measured at 233 nm.

278

M.V. Niklison-Chirou et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281 Table 2 Inhibition of SOD and catalse activity mediated by MccJ25.

State 3 State 3 + 20 ␮M MccJ25

SOD inhibition (%)

Catalase inhibition (%)

0 74.7 ± 8.3

0 27.9 ± 2.7

Table 3 Inhibition of MccJ25-induced NADPH depletion. Mitochondrial NAOPH levels (%) Control 20 ␮M MccJ25 Stigmatellin Stigmatellin + 20 ␮M MccJ25 Fig. 6. Cardiolipin oxidation induced by MccJ25. Activated mitochondria (1 mg/ml) were resuspended in 0.23 M mannitol, 70 mM sucrose, 5 mM potassium phosphate, 4 mM MgCl2 , 0.2% BSA, 20 mM Tris–HCl, pH 7.4 and incubated 30 min at 37 ◦ C with 20 ␮M MccJ25. Lipids were extracted, separated by TLC and cardiolipin spot was scraped, dissolved and dienes were determined by absorbance at 233 nm.

mitochondria. For this purpose, mitochondria were washed after treatment with MccJ25 in order to get rid of the peptide, and then complex III activity was measured. A 29 ± 4% inhibition was consistently found (data not shown), strongly suggesting that MccJ25 might cause a dual inhibition i.e. an irreversible damage of complex III as well as a reversible inhibition. On the other hand, we also identified other mitochondrial enzymes inhibited by MccJ25 such as superoxide dismutase and catalase, which are key players in the mitochondrial antioxidant defense system (Table 2). From in silico analysis, we predict that MccJ25 would interact with mitochondrial bc1 complex, mainly at cytochrome c1 region. As can be seen in Fig. 7A, MccJ25 would locate in close proximity to heme c1, with both tyrosines 9 and 20 oriented toward the heme molecule. As a consequence, MccJ25 would block the electron flow from FeS center to heme c1. Interestingly, microcin sits in the interface between the two bc1 monomers (Fig. 7B). The interface between MccJ25 and neighboring complex III subunits in the putative docking solution model involves a contact of approximately

100 41.93 96.31 77.07

± ± ± ±

1.9 2.3 3.1 2.7

1250 Å2 (G −11 kcal/mol) where 9 hydrogen bonds and various hydrophobic interactions are established. The distances between the calculated MccJ25 center of mass and the heme c1 iron center and Rieske center are 16.74 Å and 20.7 Å respectively. On the other hand, the distance from the phenol moiety of tyrosine 9 and the iron center was found to be 15.4 Å, whereas the distance from tyrosine 20 to the heme c1 was calculated to be 13.16 Å. A part of bc1 monomer is represented in Fig. 7A, showing the heme groups, the Rieske center, the docked MccJ25 and stigmatellin. This result not only supports our hypothesis that MccJ25 would interfere with electron flow at complex III, unleashing the overproduction of ROS but also it sheds light on how MccJ25 reversible inhibit cytochrome c reductase activity beyond the irreversible inhibition already discussed. In order to confirm that heme c1-FeS region of complex III would interact with microcin triggering the oxidative burst, we used stigmatellin, a well-known inhibitor of complex III at FeS center. Stigmatellin did inhibit MccJ25 pro-oxidative activity as measured by the consumption of mitochondrial NAPH pool (Table 3), indicating that iron sulfur protein or heme c1 might be targets at complex III.

Fig. 7. MccJ25 locates to heme c1 in cytochrome bc1 complex. (A) Close up view of cytochrome bc1. The three heme groups (bH, bL, and c1) appear in orange. The complex III specific mitochondrial inhibitor stigmatellin and the FeS center are represented in lime sticks and magenta spheres respectively. Microcin backbone is represented in blue whereas the MccJ25 tyrosines are depicted in red. (B) Representation of the MccJ25 binding site at the monomer–monomer cytochrome bc1 interface. The surface of the cytochrome bc1 is shown displaying one monomer in green and the other in grey whereas microcin is represented in red. These images were made with VMD 1.8.7 software support. VMD is developed with NIH support by the Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

M.V. Niklison-Chirou et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281

279

Fig. 8. Dual mechanism of action of MccJ25 in intact mitochondria. (A) MccJ25 induces membrane permeabilization with concomitant H+ release and Ca2+ uptake. Then MccJ25 can interact with complex III in order to induce ROS overproduction at complex III, which may lead to protein and lipid oxidation. Protein damage strengthens mitochondrial oxidative damage while cardiolipin oxidation leads to cytochrome c release and aggregation. Calcium, may enhance ROS production upon inhibition of complex III. (B) Ca2+ influx induced by MccJ25 can activate mtNOS in the presence of an excess of l-arginine. The activation of this enzyme inhibits not only oxidative damage of lipids and proteins but also oxygen consumption.

4. Discussion 4.1. MccJ25 induces ROS-related damage via interaction with complex III In the present paper we showed that MccJ25 produced an important decrease of RCR, an useful parameter for the evaluation of mitochondrial function (Lobao-Soares et al., 2005). It was previously reported that mitochondria under oxidative stress usually display a decreased RCR such as mitochondria under severe hypoxia (Magalhaes et al., 2005), paraquat treatment (Yamamoto et al., 1987) and superoxide dismutase deficiency (Kokoszka et al., 2001). We found a significant protein oxidation upon addition of MccJ25 as a consequence of the oxidative stress induced by this peptide. Notably, MccJ25 induced oligomerization of cytochrome c. On this regard, it was shown that carbonyl derivatives as well as dityrosines formation were stimulated upon treatment with H2 O2 , which ultimate led to cytochrome c oligomerization (Kim et al., 2006). Since carbonylation of proteins means an irreversible damage, we searched for some key enzymes involved in the antioxidant mitochondrial defenses. It was found that mainly superoxide dismutase and to a lesser extent catalase were inhibited by MccJ25, presumably by oxidative inactivation. The inhibition of these enzymes might launch a vicious cycle of ROS production and ROSinduced cellular damage. In the same regard, it was proposed that nitroglycerin may trigger an oxidative damage in heterozygous deficiency of Mn-SOD in mice (Daiber et al., 2005). Furthermore, we found that MccJ25 irreversibly inhibited the cytochrome c reductase activity by 30%, which is lower than the inhibition that we reported previously (Niklison Chirou et al., 2004). Since a covalent interaction of MccJ25 with complex III is unlikely, we hypothesize that MccJ25 might irreversible inhibit, at least partially, complex III through carbonylation of cytochrome c and other proteins of that complex. It was found in the present work that stigmatellin, an inhibitor of the iron sulfur protein (Starkov and Fiskum, 2001) was able to inhibit MccJ25-induced depletion of NADPH, which is a good marker of mitochondrial oxidative stress defenses. It is tempting to speculate that beyond the irreversible indirect inhibition of complex III by MccJ25, this peptide might interact with complex III at either iron sulfur protein or heme c1, which is located downstream the Rieske center, enhancing the inhibition of complex III activity. In order to clarify which part of complex III would interact with microcin, a molecular docking was performed as a preliminary approach. Interestingly, the docking results indicated that microcin may situate in close contact to heme c1 interfer-

ing with the electron flow from the Rieske center, most likely thorough tyrosines residues which might trap electrons coming to the heme c1. Interestingly, MccJ25 Tyr9 was recently found to be associated to superoxide production in bacteria and being essential for activity on respiratory chain in E. coli (Chalon et al., 2009). It can be suggested that stigmatellin might interfere the electron flow from FeS to microcin tyrosines, by inhibiting the Rieske center, thus preventing the MccJ25-induced ROS overproduction. As we found oxidized proteins, we decided to study in detail whether mitochondrial lipids were also targeted by MccJ25induced ROS. Interestingly, a lag phase was observed in lipid oxidation upon addition of MccJ25 whereas protein oxidation had no lag phase, suggesting that proteins might be closer to the ROS generation site(s). Oxidized lipids are usually associated with perturbations in membrane permeability (Brookes et al., 2004; Catala, 2009; Nigam and Schewe, 2000). Therefore, mitochondrial membrane alteration induced by MccJ25 observed previously (Niklison Chirou et al., 2004, 2008) may be the result of both a direct interaction of MccJ25 with membranes and an indirect effect due to oxidation of lipids. Interestingly, MccJ25 induced the oxidation of cardiolipin. It is well know that in order to achieve complete release of cytochrome c, structural and functional changes of the inner membrane may be needed (Petrosillo et al., 2001; Scorrano et al., 2002) because cytochrome c is strongly bound to the inner membrane through cardiolipin (Gottlieb, 2000). In fact, the disruption of the mitochondrial outer membrane is not sufficient to accomplish total cytochrome c release (Gottlieb et al., 2002). A probable way to detach cytochrome c from the inner membrane would be via peroxidation of cardiolipin by reactive oxygen species (Petrosillo et al., 2001). Therefore, we hypothesize that MccJ25 might induce cytochrome c release not only because of the disruption of the outer membrane but also through peroxidation of cardiolipin. 4.2. Excess of l-arginine switches mitochondrial fate by inducing overproduction of NO It was shown that MccJ25 induced the production of nitric oxide in both SMP and intact mitochondria when l-arginine was present in the medium. Because MccJ25 was already shown to stimulate Ca2+ influx (Niklison Chirou et al., 2008), it is proposed that MccJ25 would indirectly activate mtNOS in intact mitochondria through the induction of Ca2+ transport into these organelles. However, since MccJ25 induced NO production in SMP where Ca2+ was already present, we cannot rule out a direct interaction between MccJ25 and mtNOS.

280

M.V. Niklison-Chirou et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281

Because superoxide anion production is induced by MccJ25 (Niklison Chirou et al., 2008), the highly reactive peroxynitrite is likely to be produced (Huie and Padmaja, 1993). However, we were unable to demonstrate any protein damage induced by those species. Furthermore, NO overproduction triggered by the combination of both l-arginine and MccJ25 did not enhance lipid peroxidation neither, indicating that damage of lipid and protein would depend only on MccJ25-induced ROS production. On the other hand, mitochondria under overproduction of NO induced by MccJ25 not only showed no oxidative damage but also displayed a slower respiration rate, which could be ascribed to the well-known inhibitory effect of NO on cytochrome c oxidase activity (Cleeter et al., 1994). In conclusion, we propose that mitochondria incubated with MccJ25 may have two different fates: on one hand, calcium influx as well as interaction with complex III and membranes can lead to the overproduction of ROS with the concomitant lipid and protein damage and cytochrome c release (Fig. 8A). On the other, calcium influx as well as a direct interaction of MccJ25 with mtNOS under an excess of l-arginine would trigger NO overproduction with no oxidative damage and a marked inhibition in oxygen consumption (Fig. 8B). As corollary, MccJ25 activity on mitochondria could be modulated by the environment i.e. either beneficial anti-oxidative or pro-oxidative–apoptotic activities could be favored depending on the use of an excess of l-arginine or by suppressing the nitric oxide synthase activity respectively. Acknowledgements Financial support was provided by CONICET (Grant PIP 4996) and CIUNT (Grant 26/D228) and the Agencia Nacional de Promoción Científica y Técnica (PICTO 843, PAE 22642). M.V.N., F.D., C.A., C.T.B. and L.P. are recipient of a CONICET fellowship. R.D.M., C.M., A.B., S.G. and M.L.B.A. are career investigator of CONICET. We are deeply indebted to Lici Schurig-Briccio for her helpful assistance in the determination of carbonylated proteins. References Andrusier N, Nussinov R, Wolfson HJ. FireDock: fast interaction refinement in molecular docking. Proteins 2007;69:139–59. Bayro MJ, Mukhopadhyay J, Swapna GV, Huang JY, Ma LC, Sineva E, et al. Structure of antibacterial peptide microcin J25: a 21-residue lariat protoknot. J Am Chem Soc 2003;125:12382–3. Beauchamp C, Fridovich I. Superoxide dismutase: improved assays and an assay applicable to acrylamide gels. Anal Biochem 1971;44:276–87. Bellomio A, Rintoul MR, Morero RD. Chemical modification of microcin J25 with diethylpyrocarbonate and carbodiimide: evidence for essential histidyl and carboxyl residues. Biochem Biophys Res Commun 2003;303:458–62. Bellomio A, Oliveira RG, Maggio B, Morero RD. Penetration and interactions of the antimicrobial peptide, microcin J25, into uncharged phospholipid monolayers. J Colloid Interface Sci 2005;285:118–24. Bellomio A, Vincent PA, de Arcuri BF, Farias RN, Morero RD. Microcin J25 has dual and independent mechanisms of action in Escherichia coli: RNA polymerase inhibition and increased superoxide production. J Bacteriol 2007;189: 4180–6. Boveris A. Determination of the production of superoxide radicals and hydrogen peroxide in mitochondria. Methods Enzymol 1984;105:429–35. Brookes PS, Yoon Y, Robotham JL, Anders MW, Sheu SS. Calcium, ATP, and ROS: a mitochondrial love-hate triangle. Am J Physiol Cell Physiol 2004;287: C817–33. Brown GC. Nitric oxide inhibition of cytochrome oxidase and mitochondrial respiration: implications for inflammatory, neurodegenerative and ischaemic pathologies. Mol Cell Biochem 1997;174:189–92. Catala A. Lipid peroxidation of membrane phospholipids generates hydroxyalkenals and oxidized phospholipids active in physiological and/or pathological conditions. Chem Phys Lipids 2009;157:1–11. Chalon MC, Bellomio A, Solbiati JO, Morero RD, Farias RN, Vincent PA. Tyrosine 9 is the key amino acid in microcin J25 superoxide overproduction. FEMS Microbiol Lett 2009. Cleeter MW, Cooper JM, Darley-Usmar VM, Moncada S, Schapira AH. Reversible inhibition of cytochrome c oxidase, the terminal enzyme of the mitochondrial respiratory chain, by nitric oxide. Implications for neurodegenerative diseases. FEBS Lett 1994;345:50–4.

Daiber A, Oelze M, Sulyok S, Coldewey M, Schulz E, Treiber N, et al. Heterozygous deficiency of manganese superoxide dismutase in mice (Mn-SOD+/−): a novel approach to assess the role of oxidative stress for the development of nitrate tolerance. Mol Pharmacol 2005;68:579–88. Delgado MA, Rintoul MR, Farias RN, Salomon RA. Escherichia coli RNA polymerase is the target of the cyclopeptide antibiotic microcin J25. J Bacteriol 2001;183:4543–50. Faramarzi-Roques R, Malgat M, Desgrandchamps F, Ballanger P, Mazat JP. Mitochondrial metabolism in the rat during bladder regeneration induced by small intestinal submucosa. BJU Int 2004;94:419–23. Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 1957;226:497–509. Gottlieb RA. Mitochondria: execution central. FEBS Lett 2000;482:6–12. Gottlieb E, Armour SM, Thompson CB. Mitochondrial respiratory control is lost during growth factor deprivation. Proc Natl Acad Sci USA 2002;99:12801–6. Huang LS, Cobessi D, Tung EY, Berry EA. Binding of the respiratory chain inhibitor antimycin to the mitochondrial bc1 complex: a new crystal structure reveals an altered intramolecular hydrogen-bonding pattern. J Mol Biol 2005;351:573– 97. Huie RE, Padmaja S. The reaction of no with superoxide. Free Radic Res Commun 1993;18:195–9. Humphrey W, Dalke A, Schulten K. VMD: visual molecular dynamics. J Mol Graph 1996;14, 33–38, 27–38. Kim NH, Jeong MS, Choi SY, Kang JH. Oxidative modification of cytochrome c by hydrogen peroxide. Mol Cells 2006;22:220–7. Kokoszka JE, Coskun P, Esposito LA, Wallace DC. Increased mitochondrial oxidative stress in the Sod2 (+/−) mouse results in the age-related decline of mitochondrial function culminating in increased apoptosis. Proc Natl Acad Sci USA 2001;98:2278–83. Krissinel E, Henrick K. Inference of macromolecular assemblies from crystalline state. J Mol Biol 2007;372:774–97. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680–5. Levine RL, Williams JA, Stadtman ER, Shacter E. Carbonyl assays for determination of oxidatively modified proteins. Methods Enzymol 1994;233:346–57. Lobao-Soares B, Bianchin MM, Linhares MN, Carqueja CL, Tasca CI, Souza M, et al. Normal brain mitochondrial respiration in adult mice lacking cellular prion protein. Neurosci Lett 2005;375:203–6. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265–75. Magalhaes J, Ascensao A, Soares JM, Ferreira R, Neuparth MJ, Marques F, et al. Acute and severe hypobaric hypoxia increases oxidative stress and impairs mitochondrial function in mouse skeletal muscle. J Appl Physiol 2005;99:1247–53. Mashiach E, Schneidman-Duhovny D, Andrusier N, Nussinov R, Wolfson HJ. FireDock: a web server for fast interaction refinement in molecular docking. Nucleic Acids Res 2008;36:W229–32. Navarro A, Gomez C, Sanchez-Pino MJ, Gonzalez H, Bandez MJ, Boveris AD, et al. Vitamin E at high doses improves survival, neurological performance, and brain mitochondrial function in aging male mice. Am J Physiol Regul Integr Comp Physiol 2005;289:R1392–9. Nigam S, Schewe T. Phospholipase A(2)s and lipid peroxidation. Biochim Biophys Acta 2000;1488:167–81. Niklison Chirou MV, Minahk CJ, Morero RD. Antimitochondrial activity displayed by the antimicrobial peptide microcin J25. Biochem Biophys Res Commun 2004;317:882–6. Niklison Chirou M, Bellomio A, Dupuy F, Arcuri B, Minahk C, Morero R. Microcin J25 induces the opening of the mitochondrial transition pore and cytochrome c release through superoxide generation. FEBS J 2008;275:4088–96. Nystrom T. Role of oxidative carbonylation in protein quality control and senescence. EMBO J 2005;24:1311–7. Petrosillo G, Ruggiero FM, Pistolese M, Paradies G. Reactive oxygen species generated from the mitochondrial electron transport chain induce cytochrome c dissociation from beef-heart submitochondrial particles via cardiolipin peroxidation. Possible role in the apoptosis. FEBS Lett 2001;509:435–8. Petrosillo G, Ruggiero FM, Paradies G. Role of reactive oxygen species and cardiolipin in the release of cytochrome c from mitochondria. FASEB J 2003;17:2202–8. Reinheckel T, Korn S, Mohring S, Augustin W, Halangk W, Schild L. Adaptation of protein carbonyl detection to the requirements of proteome analysis demonstrated for hypoxia/reoxygenation in isolated rat liver mitochondria. Arch Biochem Biophys 2000;376:59–65. Rintoul MR, de Arcuri BF, Morero RD. Effects of the antibiotic peptide microcin J25 on liposomes: role of acyl chain length and negatively charged phospholipid. Biochim Biophys Acta 2000;1509:65–72. Rintoul MR, de Arcuri BF, Salomon RA, Farias RN, Morero RD. The antibacterial action of microcin J25: evidence for disruption of cytoplasmic membrane energization in Salmonella newport. FEMS Microbiol Lett 2001;204:265–70. Rosengren KJ, Clark RJ, Daly NL, Goransson U, Jones A, Craik DJ. Microcin J25 has a threaded sidechain-to-backbone ring structure and not a head-to-tail cyclized backbone. J Am Chem Soc 2003;125:12464–74. Rubbo H, Radi R. Protein and lipid nitration: Role in redox signaling and injury. Biochim Biophys Acta 2008;1780:1318–24. Salomon RA, Farias RN. Microcin 25, a novel antimicrobial peptide produced by Escherichia coli. J Bacteriol 1992;174:7428–35. Schneidman-Duhovny D, Dror O, Inbar Y, Nussinov R, Wolfson HJ. PharmaGist: a webserver for ligand-based pharmacophore detection. Nucleic Acids Res 2008;36:W223–8.

M.V. Niklison-Chirou et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 273–281 Scorrano L, Ashiya M, Buttle K, Weiler S, Oakes SA, Mannella CA, et al. A distinct pathway remodels mitochondrial cristae and mobilizes cytochrome c during apoptosis. Dev Cell 2002;2:55–67. Starkov AA, Fiskum G. Myxothiazol induces H(2)O(2) production from mitochondrial respiratory chain. Biochem Biophys Res Commun 2001;281:645–50.

281

Wilson KA, Kalkum M, Ottesen J, Yuzenkova J, Chait BT, Landick R, et al. Structure of microcin J25, a peptide inhibitor of bacterial RNA polymerase, is a lassoed tail. J Am Chem Soc 2003;125:12475–83. Yamamoto T, Anno M, Sato T. Effects of paraquat on mitochondria of rat skeletal muscle. Comp Biochem Physiol C 1987;86:375–8.