METHYLENE BLUE DOES NOT BYPASS COMPLEX III ANTIMYCIN BLOCK IN MOUSE BRAIN MITOCHONDRIA

Abstract

Methylene blue (MB) is a promising prodrug to treat mitochondrial dysfunctions that is currently being used in clinical trials for Alzheimer’s disease. MB can penetrate the blood brain barrier, accumulating in brain mitochondria where it acts as a redox mediator in the electron transfer chain (ETC). Mitochondrial flavins are thought to reduce MB, which is then oxidized by cytochrome c, thereby bypassing inhibited Complex I of ETC. We found that in mouse brain mitochondria, MB fails to restore the membrane potential and respiration inhibited by antimycin. Furthermore, antimycin inhibits MB-induced H₂O₂ generation. Our data suggest that the acceptor of electrons from MB is a Qo ubiquinol-binding site of Complex III; thus, MB-based drugs might not be helpful in mitochondrial dysfunctions involving Complex III inhibition.

1. Introduction

Methylene blue (MB) is a thiazine dye that has been widely used as a treatment for malaria, methemoglobinemia, and cyanide poisoning for more than a century (Schirmer et al., 2011). However, over the last two decades MB has emerged as a promising and safe potential treatment for neurodegenerative diseases (Stack et al., 2014) and a “rejuvenating” drug, at least in cell culture (Atamna et al., 2008) and mice (Gureev et al., 2016). MB is amphiphilic, which allows it to penetrate the blood–brain barrier and the membranes of mitochondria (Rojas et al., 2012). It is also a redox-mediator capable of oxidizing intramitochondrial NADH and transferring the electrons to the downstream components of ETC. This effect was termed “alternative electron transport” (Wen et al., 2011). Because of that, MB can regulate mitochondrial metabolism and homeostasis of mitochondria-produced ROS, which play an important role in neurodegenerative disorder pathophysiology and aging (Harman, 2009).

Tretter et al. (2014), have reported MB-caused increase in the rate of H₂O₂ production in guinea pig brain mitochondria. According to these authors, MB can be reduced by NADH, FADH₂ and α-glycerophosphate to leucomethylene blue (MBH₂) which is then primarily oxidized by cytochrome c. The MBH₂ –mediated H₂O₂ generation, according to these authors, is caused by a non-enzymatic reaction of MBH₂ with O₂ (Tretter et al., 2014). There is a strong increase in interest to MB as a neuroprotective compound and the mechanism of its interaction with mitochondria, especially because MB is currently being used in clinical trials for the treatment of Alzheimer’s disease (Heard et al., 2018). However, guinea pigs is not a commonly used animal model in aging and neurological diseases studies; mice are. Therefore, we thought it would be interesting to investigate how MB affects the respiration, ΔΨm, and H₂O₂ generation in mouse brain mitochondria. We found that in mouse mitochondria, the mechanism of MB-mediated redox shuttling appears to be principally different from that in guinea pig brain mitochondria.

2. Methods

2.1. Animals.

Three months-old males and females C57BL/6 mice were obtained from the Stolbovaya Nursery (Scientific Center for Biomedical Technology, Russia). The animals were housed under standard conditions (25 °C, 12-h light/dark cycle, relative humidity, >40%) with ad libitumaccess to water and food (type ssniff Spezialdiäten GmbH, Germany). Animal maintenance and sacrifice conformed the rules set by Institutional Animal Care and Use Committee of Voronezh State University, which correspond to EU Directive 2010/63/EU for animal experiments.

2.2. Isolation of mouse brain mitochondria.

Mice were sacrificed by cervical dislocation followed by decapitation. The brain dissection and extraction of the cortex were carried according to Chinopoulos et al. (2011). Mitochondria isolation was performed by digitonin-based method described by Rosenthal et al. (1987). The homogenizing buffer (HB) comprised 200 mM mannitol, 75 mM sucrose, 20 mM HEPES (pH 7.4), 1 mM EGTA, and 2 mg/ml fatty acid free BSA. The washing buffer (WB) had the same composition except that BSA was omitted.

Mouse brain cortex was homogenized with a Dounce-type homogenizer (glass body – glass pestle). The homogenate was centrifuged 5 min × 900 g. The supernatant was transferred to the clean tubes and centrifuged for 10 min at 14,000 g. After this step, the supernatant was removed and the pellet was resuspended in WB. 0.2% (v/v) digitonin was added to the tubes for 2 min in ice. The tubes were centrifuged for 15 min at 14,000 g. The supernatant was removed and the pellet washed in WB twice by centrifuging 10 min × 14,000 g. The final pellet was resuspended in 100 μl of WB.

2.3. Experimental conditions.

The incubation buffer comprised 200 mM mannitol, 75 mM sucrose, 20 mM HEPES, 4 mM K₂HPO₄, 1mM EGTA, and 2 mg/ml fatty acid free BSA (pH 7.4). Mitochondria were supported by NADH-generating substrates (5 mM malate + 5 mM pyruvate), FADH₂-generating substrate (10 mM succinate + 2 mM glutamate), or a mix of both (10 mM L-proline, which generates both FADH₂ and NADH). Measurements of membrane potential (ΔΨm) and rate of H₂O₂ production were performed at 37°C with 0.1 mg/ml mitochondrial protein using the Hitachi F-7000 fluorescence spectrophotometer (Hitachi High Technologies, UK). ΔΨm measuring was performed using the 1 μM of cationic dyes safranine O (Sigma-Aldrich, USA). The excitation wavelength was 495 nm. The emission wavelength was 586 nm.

H₂O₂ production was performed using the 1U Amplex Ultra Red (Invitrogen, USA) and 4U Horseradish Peroxidase (ThermoFisher Scientific, USA). Concentration of H₂O₂ was measured as the fluorescence intensity of the resorufin formed during the reaction. The excitation wavelength was 568 nm. The emission wavelength was 581 nm.

The rate of mitochondrial respiration was measured using an “Oxygraph” system (Hansatech Instruments, UK). Oxygen consumption was calculated as the negative time derivative of the oxygen concentration. All experiments included at least four independent measuring. Protein concentration was measured using Pierce™ BCA Protein Assay Kit (ThermoFisher Scientific, USA).

2.4. Statistical analysis.

Statistical analysis was performed using Microsoft Excel 2013 software. The results were expressed as means ± SEM. The differences were analyzed for statistical significance by using Student’s t-test. p < 0.05 was considered to be statistically significant.

3. Results and discussion

The antioxidant properties of MB are still open to debate. There are a number of studies demonstrating the antioxidant effect of MB in cell cultures exposed to exogenous H₂O₂ or rotenone (Atamna et al., 2008; Wen at al., 2011; Poteet et al., 2012). Other studies showed that MB increased the production of H₂O₂ in isolated brain mitochondria (Tretter et al., 2014; Gureev et al., 2016; Vekaria et al., 2017). In our experiments, 0.5 μM MB increased the rate of H₂O₂ production from 76.3 ± 16.6 pmol/min/mg to 1170 ± 20 pmol/min/mg in isolated brain mitochondria oxidizing pyruvate + malate. 1 μM MB increased the rate of H₂O₂ production to 1460 ± 48 pmol/min/mg, 2 μM MB to 1945 ± 44 pmol/min/mg (Fig 1). Thesdata are in line with earlier published results obtained with guinea pig brain mitochondria by Tretter et al (2014), who also reported about prooxidant properties of MB.

Fig. 1.

The effects of MB on H₂O₂ production in pyruvate + malate-supported mitochondria. In four independent experiments different MB concentrations (0 μM, 0.5 μM, 1 μM, 2 μM) were added to mitochondria. After that antimycin (1 μg/ml), rotenone (1 μM) were added. The adding of 0.5 μM, 1 μM, 2 μM of MB significantly increased the rate of H₂O₂ production (* – p<0.001); the adding antimycin decreased the rate of H₂O₂ production (# – p<0.001). The following adding of rotenone also decreased the rate of H₂O₂ production († – p<0.01, †† – p<0.001).

Complex III inhibitor antimycin increased the rate of H₂O₂ production by 7.3 times in mitochondria in the absence of MB. Surprisingly, antimycin decreased the rate of MB-induced H₂O₂ production by 25.4% in the presence of 0.5 μM MB, by 87.6% in the presence of 1μM MB, and by 95.9% in the presence of 2 μM MB. Further addition of rotenone also resulted in a reduction of the rate of H₂O₂ production (Fig 1). These data do not agree with the hypothesis of Tretter et al. (2014) that MB-induced H₂O₂ production happens because a direct electron donation by the reduced MB to O₂. Would it be so, blocking Complex III with antimycin should not decrease the rate of H₂O₂ emission. In fact, these data strongly suggest that reduced MB is oxidized at Qo site of complex III, not by the cytochrome c.

To confirm this suggestion, we’ve assayed the rate of mitochondrial respiration. Rotenone completely inhibited State 3 mitochondrial respiration induced by ADP addition. The rate of oxygen consumption induced by rotenone addition decreased from 121.4 ± 7.2 nmol/min/mg to 1.95 ± 0.6 nmol/min/mg. Adding 1 μM MB partially restored the rate of mitochondrial respiration to 21 ± 3.2 nmol/min/mg. (Fig 2A). Hence, it is clear that MB can bypass Complex I inhibition. When antimycin was used to block ETC, it decreased the rate of oxygen consumption to 15.6 ± 4.9 nmol/min/mg. In this case, 1 μM MB had very little effect on the rate of mitochondria respiration (20.2 ± 6.1 nmol/min/mg) (Fig 2B), which suggest that MB cannot bypass complex III blockage.

Fig. 2.

The rate of oxygen consumption in pyruvate + malate-supported mitochondria with adding ADP (2mM). A. Rotenone (1 μM) inhibited ETC and decreased mitochondrial respiration (* – p<0.001). MB (1 μM) slightly restored respiration (# – p<0.05). B. Аntimycin (1 μg/ml) inhibited ETC and decreased mitochondrial respiration (* – p<0.001). MB did not restore the rate of mitochondrial respiration. The rate of the oxygen consumption is given as mean ± SEM (n=4).

Measuring ΔΨm changes further confirms that MB could act as an alternative electron acceptor bypassing Complex I inhibition, as it was described earlier (Tretter et al., 2014). 1 μM MB almost completely restored the ΔΨm after inhibiting the electron flow at Complex I by rotenone. This was observed in mitochondria oxidizing NADH-generating substrates pyruvate + malate and L-proline (Fig 3Adding 1 μM MB to mitochondria oxidizing these substrates largely prevented a loss of ΔΨm induced by Complex I inhibition with rotenone (Fig 3D).

Fig. 3.

The effects of MB on ΔΨm in mitochondria. (A) Mitochondria supported by pyruvate + malate. ΔΨm is inhibiting by 1 μM of rotenone (black curve) and 1 μg/ml of antimycin (gray curve). 1 μM of MB restored ΔΨm in mitochondria inhibiting by rotenone and MB did not restore ΔΨm in mitochondria inhibiting by antimycin. (B) MB did not restore ΔΨm in succinate-supported mitochondria inhibited by antimycin. (C) Mitochondria supported by L-proline. ΔΨm was inhibited by 1 μM of rotenone (black curve) and 1 μg/ml of antimycin (gray curve). 1 μM of MB restored ΔΨm in mitochondria inhibiting by rotenone and did not restore ΔΨm in mitochondria inhibiting by antimycin. (D) Mitochondria supported by pyruvate + malate. Pre-addition of 1 μM of MB partly prevented rotenone-induced inhibiting of ΔΨm. The following addition of 1 μM of MB restored ΔΨm.

As expected, blocking the electron flow in Complex III by antimycin have resulted in a loss of ΔΨm. To our surprise and in difference with the data published by Tretter et al, MB did not restore the ΔΨm in the mitochondria oxidizing either pyruvate + malate (Fig 3A), succinate (Fig 3B), or L-proline (Fig 3C). This was observed at a broad range of MB concentration (1 – 5 μM; data are not presented). Considering the reproducibility of these data in mouse brain mitochondria (n=8), we think that well-described “hormesis effect” of MB has nothing to do with that, and the issue is that in mouse brain mitochondria MB does not behave the same way as in guinea pig brain mitochondria (as reported by Tretter et al). That is, the effect of MB is species - specific.

In summary, our data confirm that MB can function as a prooxidant increasing the rate of H₂O₂ production in intact brain mitochondria. Our data also confirm that MB can be a useful compound restoring mitochondria respiration in case of Complex I inhibition (Fig 4). However, our data do not support the idea that MB is a universally efficient mitochondria-targeted neuroprotector that can restore mitochondria functions inhibited by multiple agents/conditions (Poteet et al., 2012; Atamna et al., 2008; Atamna et al., 2015) and improve memory and cognitive parameters (Callaway et al., 2002; Gonzalez-Lima and Bruchey, 2004; Wrubel et al., 2007). According to our data, MB is only efficient under circumstances of Complex I failure, but not when Complex III is inhibited.

Fig. 4.

The hypothetic scheme of alternative electron transport induced by methylene blue.