hydrogen peroxide production

Accepted Manuscript Partial loss of complex I due to NDUFS4 deficiency augments myocardial reperfusion damage by increasing mitochondrial superoxide/hydrogen peroxide production Nidhi Kuksal, Danielle Gardiner, Dake Qi, Ryan J. Mailloux PII:

S0006-291X(18)30464-9

DOI:

10.1016/j.bbrc.2018.02.208

Reference:

YBBRC 39573

To appear in:

Biochemical and Biophysical Research Communications

Received Date: 23 February 2018 Accepted Date: 28 February 2018

Please cite this article as: N. Kuksal, D. Gardiner, D. Qi, R.J. Mailloux, Partial loss of complex I due to NDUFS4 deficiency augments myocardial reperfusion damage by increasing mitochondrial superoxide/ hydrogen peroxide production, Biochemical and Biophysical Research Communications (2018), doi: 10.1016/j.bbrc.2018.02.208. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Partial loss of complex I due to NDUFS4 deficiency augments myocardial reperfusion damage by increasing mitochondrial superoxide/hydrogen peroxide production

*corresponding author: [email protected] 1

RI PT

Nidhi Kuksal1,2, Danielle Gardiner1, Dake Qi2, and Ryan J. Mailloux1*

Department of Biochemistry, Faculty of Science, Memorial University of Newfoundland, St. John’s, Newfoundland and Labrador, Canada.

2

AC C

EP

TE D

M AN U

SC

Biomedical Sciences, Faculty of Medicine, Memorial University of Newfoundland, St. John’s, Newfoundland and Labrador, Canada

1

ACCEPTED MANUSCRIPT

Recent work has found that complex I is the sole source of reactive oxygen species (ROS) during myocardial ischemia-reperfusion (IR) injury. However, it has also been reported that heart mitochondria can also generate ROS from other sources in the respiratory chain and Krebs cycle.

RI PT

This study examined the impact of partial complex I deficiency due to selective loss of the Ndufs4 gene on IR injury to heart tissue. Mice heterozygous for NDUFS4 (NDUFS4+/-) did not display any significant changes in overall body or organ weight when compared to wild-type

SC

(WT) littermates. There were no changes in superoxide (O2●-)/hydrogen peroxide (H2O2) release from cardiac or liver mitochondria isolated from NDUFS4+/- mice. Using selective ROS release

M AN U

inhibitors, we found that complex III serves a major source for ROS production in WT and NDUFS4+/- cardiac mitochondria respiring under state 4 conditions. Subjecting hearts from NDUFS4+/- mice to reperfusion injury revealed that the partial loss of complex I decreases contractile recovery and increases myocardial infarct size. These results correlated with a

TE D

significant increase in O2●-/H2O2 release rates in mitochondria isolated from NDUFS4+/- hearts subjected to an IR challenge. Taken together, these results demonstrate that the partial absence of complex I sensitizes the myocardium towards IR injury and that the main source of ROS

AC C

EP

following reperfusion is complex III.

2

ACCEPTED MANUSCRIPT

1. Introduction Complex I is the main entry point for electrons from NADH into the electron transport

RI PT

chain (ETC) and is thus vital for oxidative phosphorylation and respiration. Electron movement through complex I to the ubiquinone binding pocket is coupled to the extrusion of protons and the establishment of a protonmotive force (PMF). Both the flavin mononucleotide (FMN) prosthetic group and the ubiquinone binding pocket in complex I are also sites for O2●- release

SC

[1]. Overproduction of ROS by complex I has been implicated in the development of several

M AN U

myocardial diseases [2]. Increased ROS release due to disruption of complex I is either associated with variants in mitochondrial or nuclear genes encoding complex I subunits or environmental factors that disrupt or block electron flow [3,4]. Disruption of normal mitochondrial function or signaling pathways that control O2●- release from complex I have also

TE D

been implicated in the development of heart disease and other disorders [5,6]. It is generally accepted that mitochondrial dysfunction plays a role in myocardial damage following reperfusion. Mitochondrial calcium overload, permeability transition pore opening,

EP

and the release of pro-apoptotic factors are important inducers of myocardial IR injury [7]. Induction of myocardial tissue damage and apoptosis is also triggered, in part, by the over

AC C

production of mitochondrial ROS following reperfusion [8]. The absence of O2 during ischemia results in the over reduction of electron donating centers and the buildup of metabolites in cardiomyocytes. Reperfusion with O2-saturated blood results in a burst in ROS production by mitochondria overwhelming antioxidant defenses and inducing oxidative distress and cell death [9]. The notion that the burst of ROS release following reperfusion is a major contributor to myocardial damage is supported by studies showing that ischemic preconditioning or pre-

3

ACCEPTED MANUSCRIPT

treatment with mitochondria-targeted antioxidants or electron transport chain blockers can curtail IR injury [10] [11].

RI PT

Accumulation of succinate during ischemia was recently shown to be a driving force behind reperfusion injury [12]. In this scenario, reintroduction of O2 induces the rapid oxidation of succinate resulting in reverse electron flow to complex I, stimulating a massive burst in ROS release culminating with myocardial damage [12]. It was also suggested that most, if not all, of

SC

the electrons from succinate produce ROS from reverse electron transfer (RET) to complex I [8].

M AN U

However, mitochondria can contain up to 12 sites for ROS release and pyruvate dehydrogenase complex (PDHC), α-ketoglutarate dehydrogenase (KGDHC), and complexes II and III have been documented to be significant sources of O2●- and H2O2 in cardiac mitochondria [6,13]. Furthermore, recent work has demonstrated that conditions are unfavorable for the production of ROS by RET from succinate during reperfusion [10]. We examined the response of hearts from

TE D

mice heterozygous for the Ndufs4 gene (NDUFS4+/-), an accessory protein that is required for the assembly of complex I, towards an IR challenge. It was found that NDUFS4+/- cardiac mitochondria display similar O2●-/H2O2 release rates when compared to mitochondria isolated

EP

from WT mice. Furthermore, using ROS release inhibitors, it was found that complex III

AC C

accounts for a significant amount of the O2●-/H2O2 released by WT and NDUFS4+/mitochondria oxidizing succinate under state 4 respiratory conditions. However, challenging complex I deficient mice with reperfusion injury induced a ~2-fold increase in ROS release which correlated with increased myocardial damage and diminished recovery from IR. Overall, these findings demonstrate that the full expression of complex I is integral for protecting mice from an IR challenge and that complex III serves as the major source of ROS during succinate oxidation in mitochondria respiring under state 4 conditions.

4

ACCEPTED MANUSCRIPT

2. Materials and Methods Chemicals: H2O2 (30% solution), pyruvate, malate, α-ketoglutarate, mannitol, Hepes, sucrose,

RI PT

EGTA, fatty acid free bovine serum albumin, Bradford reagent, superoxide dismutase (SOD), myxothiazol, 3-methyl-2-oxovaleric acid (KMV), rotenone, and horse radish peroxidase (HRP) were purchased from Sigma. Amplex Ultra Red (AUR) reagent was acquired from Invitrogen. Anti-OXPHOS cocktail and anti-NDUFS4 were purchased from Abcam. Anti-GPX1, anti-

SC

TRX2, anti-SDHA, anti-MnSOD, anti-mouse and anti-rabbit horseradish peroxidase secondary

M AN U

antibodies, atpenin A5, and CPI-613 were purchased from Santa Cruz.

Animals: All experiments were approved by Memorial University’s Animal Care and Use committee and conducted according to institutional and Federal animal care guidelines. Male and female mice heterozygous for the Ndufs4 gene (Ndufs4+/-) were purchased from Jackson

TE D

Laboratory. Mice were housed in ventilated cages and fed a standard chow diet (44.2% carbohydrate, 6.2% fat, 18.6% crude protein; diet T.2018, Harlan, Indianapolis, IN) ad libitum and given free access to water. Mice were bred, and, upon weaning, new litters were ear notched

EP

and genotyped according to the supplier’s instructions. WT and NDUFS4+/- mice were weighed on a weekly basis and euthanized for either reperfusion injury experiments or isolation of

AC C

mitochondria from liver or cardiac tissue. Tissues were weighed before isolating mitochondria. Note that livers from 4-week old NDUFS4-/- mice were also extracted for mitochondrial isolation for immunoblots to confirm partial or complete loss of the Ndufs4 gene. Mouse heart perfusions and estimation of infarct size: Krebs-Henseleit buffer (KH buffer; 7 mM glucose, 0.4 mmol/l oleate, 1% (w/v) BSA, and 10 εU/ml insulin) was prepared fresh the day of experiments. Eight to ten-week-old WT and NDUFS4+/- were heparinized and then anesthetized with pentobarbital sodium (60 mg/kg). Hearts were then quickly excised and placed in the 5

ACCEPTED MANUSCRIPT

Langendorff setup. The aorta was then cannulated and hearts perfused with KH buffer. Left ventricular developed pressure (LVDP) and heart rate (HR) during the perfusion. Flow was then adjusted to 4 mL/min and hearts perfused for 30 minutes followed by 15 minutes of stop-flow

RI PT

ischemia and 30 minutes of reperfusion. Once completed, hearts were immediately placed in icecold mitochondrial isolation buffer or sectioned for triphenyltetrazolium chloride (TTC) staining to estimate infarct size.

SC

Preparation of mitochondria: All steps were performed on ice or at 4 °C. Liver and cardiac

M AN U

tissue were exciced and immediately placed in mitochondrial isolation buffer (220 mM mannitol, 70 mM sucrose, 1 mM EGTA, 20 mM HEPES, pH 7.4). Livers and hearts were cut into smaller pieces and washed thrice in isolation buffer. Next, pieces were minced on Teflon watch glasses using steel razors. For liver, the minced tissue was placed in isolation buffer containing 0.5% (w/v) defatted BSA. Samples were then homogenized using the Potter-Elvejham method.

TE D

Homogenates were then placed in centrifuge tubes and spun at 800 xg for 9 min. The supernatant was then collected and centrifuged again at 12,000 xg for 9 min. The supernatant was decanted and the sides of the tube carefully wiped clean of any fat. The resulting pellet was resuspended in

EP

15 mL of isolation buffer containing 0.5% (w/v) BSA and centrifuged at 12,000 xg for 9 min.

AC C

The pellet was then resuspended in isolation buffer free of BSA. The final concentration of the liver mitochondrial pellet was ~12-16 mg/mL. For cardiac mitochondria, minced tissue was added to isolation buffer containing 0.5%

(w/v) defatted BSA and 1 U subtilisin A. Tissue was homogenized as described above and the homogenate was centrifuged at 10,000 xg for 9 min. The pellet was then washed in 20 mL isolation buffer to removal excess subtilisin A and centrifuged again at 800 xg for 9 min. The supernatant was then collected and centrifuged at 12,000 xg for 9 min. The resulting pellet was 6

ACCEPTED MANUSCRIPT

resuspended in isolation buffer. The final concentration of the cardiac mitochondria pellet was ~6-8 mg/mL.

RI PT

Amplex UltraRed assays: Rates of O2•-/H2O2 production were estimated using the Amplex UltraRead assay. Briefly, liver and mitochondria were diluted to 3 mg/mL and 1 mg/mL, respectively, in mitochondrial isolation buffer. Samples were then loaded into individual wells of a 96-well black plate containing mitochondrial isolation buffer. The final concentration of

SC

mitochondria per well was 0.3 mg/mL and 0.1 mg/mL for liver and cardiac samples respectively.

M AN U

For experiments that utilized inhibitors, samples were pre-incubated in different inhibitors for ROS release; CPI-613 (250 µM), KMV (10 mM), rotenone (4 µM), atpenin A5 (40 µM) or myxothiazol (4 µM) for 5 min at 25 °C (Figure 3A). Reaction mixtures were then supplemented with horseradish peroxidase (3 U/mL), superoxide dismutase (25 U/mL), and Amplex UltraRed reagent (10 µM). Reactions were then initiated by the addition of the following substrates;

TE D

pyruvate and malate, α-ketoglutarate and malate, or succinate. The final concentration of all substrates in each reaction mixture was 50 µM. Changes in resorufin fluorescence were monitored every 30 seconds for 5 minutes at excitation:emission wavelengths 565 nm:600 nm

EP

using a SpectraMax5 plate reader (Molecular Devices). Rates of ROS release were estimated

AC C

using standard curves and results were normalized to protein content per well. To estimate the contribution of individual sites of production towards overall ROS release, rates were normalized to the control.

NAD(P)H autofluorescence: The redox state of the mitochondrial nicotinamide pool was estimated by measuring its autofluorescence at the excitation:emission wavelengths of 376 nm: 420 nm. Briefly, mitochondria collected from hearts subjected to IR injury were diluted to 0.1

7

ACCEPTED MANUSCRIPT

mg/mL in isolation buffer and the redox state of the pool was measures using a SpectraMax5 plate reader (Molecular Devices).

RI PT

Immunoblot: Isolated mitochondria were diluted to 2 mg/mL in Laemmli buffer containg 2% (v/v) 2-mercaptoethanol and then heated at 100 °C for 10 min. Samples were then loaded in the wells of a 10% isocratic denaturing gel and electrophoresed at 80 V. Once the running front penetrated the resolving gel, the voltage was increased to 200 V. Once complete, proteins in the

SC

gel were electroblotted onto the surface of nitrocellulose membranes by wet-tank transfer.

M AN U

Successful transfer was confirmed by Ponceau S staining. Nonspecific binding sites were blocked with 5% (w/v) non-fat skim milk dissolved in tris-buffered saline/0.1% (v/v) tween-20 (TBS-T). Membranes were then washed and probed overnight with OXPHOS cocktail (1/2000; Abcam), NDUFS4 (1/2000; Abcam), SDHA (1/3000; Santa Cruz), MnSOD (1/3000; Santa Cruz), GPX1 (1/2000; Santa Cruz), and TRX2 (1/2000; Santa Cruz) primary antibodies under

TE D

constant agitation and at 4 °C. Membranes were then washed with TBS-T and probed for one hour with anti-mouse (1/3000; Abcam) or anti-rabbit (1/3000; Santa Cruz) diluted in blocking solution at room temperature under constant agitation. Bands were visualized using WestPico

EP

Super Signal Chemiluminescent substrate and ImageQuant LAS 4000. Band intensities were

AC C

quantified using ImageJ software.

Data analysis: All experiments were performed twice and at least in quadruplicate. Data was analyzed using Graph Pad Prism 6 software. Using paired two-tailed Student T-tests or 2-way ANOVA with a Tukey’s post-hoc test. * or #; p ≤ 0.05, ** or ##; p ≤ 0.01, *** or ###; p ≤ 0.001. * represents genotype-associated statistical differences and # indicates statistical differences when compared to control with a genotype. 3. Results 8

ACCEPTED MANUSCRIPT

3.1 Partial loss of complex I does not decrease mitochondrial ROS release but increases antioxidant enzyme levels

RI PT

Mice homozygous for NDUFS4 develop Leigh’s syndrome like symptoms and chronic encephalomyopathy early on after weaning [14]. Mice homozygous for NDUFS4 are also much smaller and weigh significantly less when compared to WT or NDUFS4+/- littermates [14]. Therefore, we chose to conduct our studies on NDUFS4+/- mice since they do not display any

SC

aberrant phenotypes that would otherwise compromise measurements of ROS release from

M AN U

individual sites of production in mitochondria [14]. Mice heterozygous or homozygous for NDUFS4 were identified by genotyping and the absence or ~50% loss in NDUFS4 protein was confirmed by immunoblot (Figure 1A). Partial loss of complex I was confirmed using OXPHOS antibody cocktail. Immunoblotting for different subunits of the respiratory complexes showed that subunits corresponding to complex V and III did not display any different in protein

TE D

expression (Figure 1B). However, MTCO1 subunit for complex IV was decreased in NDUFS4+/- cardiac mitochondria when compared to WT littermates. In addition, NDUFB8, a complex I subunit, displayed a ~50% decrease in protein expression in NDUFS4+/- liver

EP

mitochondria (Figure 1B). This confirms that the partial ablation of the Ndufs4 gene results in a

AC C

~50% decrease in complex I levels. NDUFS4+/- mice also did not display any aberrant changes in overall body weight or liver and heart weight (Figure 1C and 1D). Next, we examined ROS release from liver and cardiac mitochondria isolated from WT

and NDUFS4+/- mice. Measurements were conducted under state 4 respiratory conditions using different substrates that feed electrons into different parts of the respiratory chain (Figure 3A). Liver mitochondria enriched from NDUFS4+/- mice did not display any significant change in O2●-/H2O2 release when oxidizing Krebs cycle linked substrates, pyruvate and 2-oxoglutarate, 9

ACCEPTED MANUSCRIPT

when compared to WT littermates (Figure 2A). No differences in ROS production were also observed when succinate, which drives O2●-/H2O2 release exclusively from the respiratory chain, served as the substrate (Figure 2A). We also observed no changes in O2●-/H2O2 production

RI PT

cardiac mitochondria isolated from NDUFS4+/- mice oxidizing Krebs cycle-linked substrates or succinate (Figure 2B). Overall, these results demonstrate that a ~50% decrease in complex I availability does not alter ROS release from cardiac or liver mitochondria respiring under state 4

SC

conditions.

M AN U

We reasoned that the lack of a difference in ROS release between WT and NDUFS4+/mitochondria may be due to an increase in the availability of H2O2 quenching enzymes. Thus, we decided to immunoblot for mitochondrial enzymes involved in the two major H2O2 clearing pathways; glutathione peroxidase-1 (GPX1) and thioredoxin-2 (TRX2) (Figure 2C). Cardiac mitochondria enriched from NDUFS4+/- mice displayed a two-fold increase in the expression of

TE D

both GPX1 and TRX2 (Figure 2C). This indicates that partial loss of complex I augments mitochondrial H2O2 clearing capacity through increased expression of antioxidant enzymes involved in the glutathione (GSH) and thioredoxin (TRX) systems which could account for the

AC C

EP

lack of difference in rates of O2●-/H2O2 production in WT and NDUFS4+/- mice.

3.2 Complex III is also an important source of ROS in cardiac mitochondria To ascertain if partial loss of complex I alters ROS release from mitochondria, we used

several selective inhibitors for O2●-/H2O2 production. We first decided to examine ROS release from individual sites of production in liver mitochondria since it was recently shown that αketoglutarate dehydrogenase complex (KGDHC) and complex III serve as the major release sites

10

ACCEPTED MANUSCRIPT

during pyruvate and α-ketoglutarate oxidation [15]. It was found that 3-methyl-2-oxo valeric acid (KMV) and CPI-613, selective inhibitors for ROS production from KGDHC and 2-oxoacid dehydrogenases (Figure 3A), induced a >90% decrease in O2●-/H2O2 release from liver

RI PT

mitochondria oxidizing pyruvate or 2-oxoglutarate (Figure 3B). However, no differences were noted between mitochondria enriched from the livers of WT and NDUFS4+/- mice. KMV and CPI-613 also block NADH production, a key factor required for ROS release from the

SC

respiratory chain. Therefore, we also tested the impact of atpenin A5 and myxothiazol on ROS release. Atpenin A5 inhibits ROS release from complex II during the oxidation of NADH by

M AN U

complex I by blocking the ubiquinone binding site (Figure 3A). Myxothiazol impedes ROS release from complex III by preventing semiquinone radical formation (Figure 3A). Atpenin A5 induced >80% decrease in O2●-/H2O2 production and myxothiazol lowered ROS release by ~50% (Figure 3B). No differences in ROS release were observed between mitochondria isolated from

TE D

WT or NDUFS4+/- mice. Overall, these results demonstrate that the partial loss of complex I does not alter the rate of ROS release from major sites of production in liver mitochondria. In addition, the major sites for production in liver mitochondria are complexes II and III and not

EP

PDHC and KGDHC.

AC C

Next, we conducted the same experiments with cardiac mitochondria. KMV and CPI-613 augmented ROS release when pyruvate and malate served as substrates and no differences were detected between the two genotypes (Figure 3C). This is attributed to the accumulation of NADH and increased ROS release by the respiratory chain, an observation that is consistent with previous findings [6]. KMV and CPI-613 also did not alter O2●-/H2O2 release in mitochondria oxidizing α-ketoglutarate (Figure 3C). Unlike other Krebs cycle-linked metabolites, succinate donates electrons directly to the UQ pool through its oxidation by complex II, which allows one

11

ACCEPTED MANUSCRIPT

to examine rates of production from the respiratory chain only (Figure 3C). Succinate is also often characterized as a driving force behind most of the ROS release from cardiac mitochondria, especially in reperfusion injury [16]. We observed no genotype differences between cardiac

RI PT

mitochondria enriched from WT and NDUFS4+/- mice when succinate served as the substrate (Figure 3C). However, we noted that myxothiazol decreased ROS release from both WT and NDUFS4+/- cardiac mitochondria by ~50% (Figure 3C). Rotenone, which inhibits O2●-/H2O2

SC

production by preventing reverse electron flow to complex I, lowered ROS release by ~35% and atpenin A5 almost abolished succinate-driven ROS production (Figure 3C). Taken together,

M AN U

these results demonstrate that under state 4 respiration conditions, complexes I and III are the major ROS sources in cardiac mitochondria, with the latter producing more than the former. 3.3 Partial loss of NDUFS4 compromises heart recovery and augments damage following

TE D

reperfusion

The results collected above indicate that the partial loss of complex I does not affect ROS release from cardiac mitochondria and that complex III is a major site of production. Next, we

EP

tested the effect of reperfusion injury on the myocardium of NDUFS4+/- mice. Hearts were excised and perfused using a Langendorff constant pressure system. Hearts were initially

AC C

perfused for 30 minutes then subjected to no-flow ischemia for 15 minutes followed by 30 minutes of reperfusion (Figure 4A). Cardiac function and recovery from ischemia was assessed by monitoring left ventricular developed pressure (LVDP) and heart rate (HR). It was observed that partial loss of complex I induced a significant decrease in cardiac rate of recovery from ischemia during reperfusion when compared to WT mice (Figure 4A). Following reperfusion, hearts were collected and infarct size was measured by TTC staining. It was found that the

12

ACCEPTED MANUSCRIPT

infarct size comprised ~60% of the total area of the myocardium in NDUFS4+/- hearts (Figure 4B). By contrast, the infarct size in WT hearts was ~39% of the overall myocardial area.

RI PT

3.4 Complex I deficiency augments ROS release following reperfusion injury In Chouchani et al, it was found that complex I served as the sole source of ROS in the myocardium following reperfusion injury [12]. If this is true, then even a partial loss in complex

SC

I should hypothetically protect the myocardium from IR injury by decreasing mitochondrial ROS release. Based on our results so far, we have found that complex III is a major source of ROS in

M AN U

cardiac mitochondria, even under conditions that favor reverse electron flow, and that NDUFS+/hearts display increased sensitivity towards reperfusion injury. We followed this up by testing ROS release rates from mitochondria isolated from hearts subjected to reperfusion injury. Following reperfusion injury, cardiac mitochondria enriched from NDUFS4+/- mice displayed a significant increase in ROS release when compared to WT controls (Figure 4C). Indeed, cardiac

TE D

mitochondria from complex I deficient mice produced ~2-fold more O2●-/H2O2 when oxidizing either Krebs cycle linked substrates (pyruvate and 2-oxoglutarate) or succinate when respiring

EP

under state 4 conditions (Figure 4C). We also observed no difference in NADH availability in mitochondria collected from WT and NDUFS4+/- mice following reperfusion (Figure 4D).

AC C

4. Discussion

It has been documented that mitochondria can contain up to 12 distinct sites for ROS

release associated with nutrient metabolism [17]. In cardiac mitochondria, some important sites for production include complexes I, II, and III and PDHC and KGDHC [6,13]. However, in a recent study it was demonstrated that complex I is the sole source of ROS during reperfusion injury [8,12]. This is associated with the accumulation of succinate and RET from complex II to

13

ACCEPTED MANUSCRIPT

complex I, which induces a burst of O2●- release following the reintroduction of oxygen [8]. The authors also contend that this burst in ROS release is very rapid, happening within seconds, and is responsible for the opening of the mitochondrial permeability transition pore (MPTP) [16].

RI PT

The caveat to this suggested mechanism is that prolonged ischemia results in the depolarization of the transmembrane potential of protons (∆ΨM) across the mitochondrial inner membrane, a vital driving force for RET to complex I [18]. In addition, recovery of the ∆ΨM after reperfusion

SC

is slow due to the opening of MPTP, which happens within ~15 seconds of reperfusion [18]. Furthermore, it is well accepted that MPTP opening occurs in response to mitochondrial calcium

M AN U

overload and that the burst of ROS release happens after pore opening [10]. Finally, reperfusion induces a rapid increase in respiration and substrate oxidation in an effort to re-establish a protonmotive force and replenish cellular ATP levels [10]. Overall, these conditions would favor forward electron flow to complex IV thus mitigating RET from succinate to complex I.

TE D

Based on the information above we decided to examine the impact of the partial loss of complex I on reperfusion injury. For this, we used mice heterozygous for the Ndufs4 gene, a subunit for complex I that is required for its assembly and stabilization. If complex I is the sole

EP

source for ROS release during reperfusion injury, then even the partial loss of this respiratory

AC C

complex should diminish myocardial damage by limiting O2●-/H2O2 production. However, we found that hearts from mice heterozygous for NDUFS4 displayed a significant increase infarct size following reperfusion. The increased sensitivity towards myocardial damage correlated with an elevation in ROS release during the oxidation of Krebs cycle-linked substrates or succinate. Our results suggest that fully functional complex I may be required to protect the myocardium from reperfusion injury by maintaining mitochondrial electron flow. This may prevent the full depolarization of the mitochondrial inner membrane following MPTP lessening reperfusion

14

ACCEPTED MANUSCRIPT

injury. The increase in ROS production by mitochondria deficient in complex I could be due to the redistribution of electrons flow in the respiratory chain resulting in augmented O2●-/H2O2 formation by the other sites of production like complex III. Recent work has demonstrated that

RI PT

the association of complex I and III is essential for maintaining a functional respirasome in cardiac mitochondria, limiting ROS release [19]. Although speculative, it is possible that loss of respirasome stability due to a deficiency in complex I may induce a redistribution of electrons in

SC

the electron transport chain augmenting ROS release and increasing myocardial damage.

M AN U

In the present study, it was also found that complex I can produce ROS during RET in mitochondria not subjected to reperfusion injury. This does suggest that complex I is a source of ROS in cardiac mitochondria. Indeed, rotenone and malonate have been found to protect from reperfusion injury by inhibiting ROS release [11]. Studies have suggested that this cardioprotective effect is related to the inhibition of RET-driven ROS release following succinate

TE D

oxidation [11,12]. However, as mentioned above, the depolarization of the mitochondrial inner membrane and depletion of ATP during ischemia would actually favor rapid O2 consumption and forward electron flow to re-establish the mitochondrial membrane potential. In addition, if

EP

complex I was the ROS source in cardiac mitochondria, then even its partial loss would have

AC C

diminished mitochondrial O2●-/H2O2 production. However, in our hands it was observed that complex I deficiency did not alter ROS release in liver or cardiac mitochondria oxidizing Krebs cycle-linked substrates or succinate and the rate of O2●-/H2O2 production only increased after reperfusion challenge. Moreover, it was found that complex III, rather than complex I, was the major source of ROS during state 4 respiration, conditions that actually favor RET to complex I. In liver, complexes II and III, rather than complex I, were the major sources of ROS.

15

ACCEPTED MANUSCRIPT

In conclusion, our results demonstrate that the partial loss of complex I due to targeted deletion of the Ndufs4 gene augments myocardial reperfusion damage, which is also associated with an increase in mitochondrial ROS release. These findings also show that mitochondrial

RI PT

ROS sources other than complex I can contribute to myocardial reperfusion damage. Acknowledgements

SC

This work was funded by the Memorial University of Newfoundland Institutional MultiDisciplinary, Seed, and Bridge Fund (RJM) and The Research Development Corporation (RDC)

M AN U

of Newfoundland (DQ). Nidhi Kuksal was funded by the Queen Elizabeth Diamond Jubilee Scholarship.

References

AC C

EP

TE D

[1] J.R. Treberg, C.L. Quinlan, M.D. Brand, Evidence for two sites of superoxide production by mitochondrial NADH-ubiquinone oxidoreductase (complex I), J Biol Chem 286 (2011) 2710327110. [2] L.K. Russell, B.N. Finck, D.P. Kelly, Mouse models of mitochondrial dysfunction and heart failure, J Mol Cell Cardiol 38 (2005) 81-91. [3] Y. Ichikawa, M. Ghanefar, M. Bayeva, R. Wu, A. Khechaduri, S.V. Naga Prasad, R.K. Mutharasan, T.J. Naik, H. Ardehali, Cardiotoxicity of doxorubicin is mediated through mitochondrial iron accumulation, J Clin Invest 124 (2014) 617-630. [4] G. Karamanlidis, C.F. Lee, L. Garcia-Menendez, S.C. Kolwicz, Jr., W. Suthammarak, G. Gong, M.M. Sedensky, P.G. Morgan, W. Wang, R. Tian, Mitochondrial complex I deficiency increases protein acetylation and accelerates heart failure, Cell Metab 18 (2013) 239-250. [5] R.J. Mailloux, J.Y. Xuan, S. McBride, W. Maharsy, S. Thorn, C.E. Holterman, C.R. Kennedy, P. Rippstein, R. deKemp, J. da Silva, M. Nemer, M. Lou, M.E. Harper, Glutaredoxin-2 is required to control oxidative phosphorylation in cardiac muscle by mediating deglutathionylation reactions, J Biol Chem 289 (2014) 14812-14828. [6] J. Chalker, D. Gardiner, N. Kuksal, R.J. Mailloux, Characterization of the impact of glutaredoxin-2 (GRX2) deficiency on superoxide/hydrogen peroxide release from cardiac and liver mitochondria, Redox Biol 15 (2017) 216-227. [7] A.P. Halestrap, Calcium, mitochondria and reperfusion injury: a pore way to die, Biochem Soc Trans 34 (2006) 232-237.

16

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[8] E.T. Chouchani, V.R. Pell, A.M. James, L.M. Work, K. Saeb-Parsy, C. Frezza, T. Krieg, M.P. Murphy, A Unifying Mechanism for Mitochondrial Superoxide Production during Ischemia-Reperfusion Injury, Cell Metab 23 (2016) 254-263. [9] D.N. Granger, P.R. Kvietys, Reperfusion injury and reactive oxygen species: The evolution of a concept, Redox Biol 6 (2015) 524-551. [10] T. Andrienko, P. Pasdois, A. Rossbach, A.P. Halestrap, Real-Time Fluorescence Measurements of ROS and [Ca2+] in Ischemic / Reperfused Rat Hearts: Detectable Increases Occur only after Mitochondrial Pore Opening and Are Attenuated by Ischemic Preconditioning, PLoS One 11 (2016) e0167300. [11] E.J. Lesnefsky, Q. Chen, B. Tandler, C.L. Hoppel, Mitochondrial Dysfunction and Myocardial Ischemia-Reperfusion: Implications for Novel Therapies, Annu Rev Pharmacol Toxicol 57 (2017) 535-565. [12] E.T. Chouchani, V.R. Pell, E. Gaude, D. Aksentijevic, S.Y. Sundier, E.L. Robb, A. Logan, S.M. Nadtochiy, E.N.J. Ord, A.C. Smith, F. Eyassu, R. Shirley, C.H. Hu, A.J. Dare, A.M. James, S. Rogatti, R.C. Hartley, S. Eaton, A.S.H. Costa, P.S. Brookes, S.M. Davidson, M.R. Duchen, K. Saeb-Parsy, M.J. Shattock, A.J. Robinson, L.M. Work, C. Frezza, T. Krieg, M.P. Murphy, Ischaemic accumulation of succinate controls reperfusion injury through mitochondrial ROS, Nature 515 (2014) 431-435. [13] R.J. Mailloux, D. Gardiner, M. O'Brien, 2-Oxoglutarate dehydrogenase is a more significant source of O2(.-)/H2O2 than pyruvate dehydrogenase in cardiac and liver tissue, Free Radic Biol Med 97 (2016) 501-512. [14] S.E. Kruse, W.C. Watt, D.J. Marcinek, R.P. Kapur, K.A. Schenkman, R.D. Palmiter, Mice with mitochondrial complex I deficiency develop a fatal encephalomyopathy, Cell Metab 7 (2008) 312-320. [15] L. Slade, J. Chalker, N. Kuksal, A. Young, D. Gardiner, R.J. Mailloux, Examination of the superoxide/hydrogen peroxide forming and quenching potential of mouse liver mitochondria, Biochim Biophys Acta 1861 (2017) 1960-1969. [16] V.R. Pell, E.T. Chouchani, M.P. Murphy, P.S. Brookes, T. Krieg, Moving Forwards by Blocking BackFlow: The Yin and Yang of MI Therapy, Circ Res 118 (2016) 898-906. [17] M.D. Brand, Mitochondrial generation of superoxide and hydrogen peroxide as the source of mitochondrial redox signaling, Free Radic Biol Med 100 (2016) 14-31. [18] T.N. Andrienko, P. Pasdois, G.C. Pereira, M.J. Ovens, A.P. Halestrap, The role of succinate and ROS in reperfusion injury - A critical appraisal, J Mol Cell Cardiol 110 (2017) 1-14. [19] R. Moreno-Loshuertos, J.A. Enriquez, Respiratory supercomplexes and the functional segmentation of the CoQ pool, Free Radic Biol Med 100 (2016) 5-13.

Figure captions

Figure 1: Partial ablation of the Ndufs4 gene decreases complex I levels but does not alter body, liver, or heart weight. A) Immunoblot for NDUFS4 protein levels in liver mitochondria from WT, NDUFS4+/-, and NDUFS4-/- mice. SDHA served as the loading control. N=2. B) Immunoblot detection of respiratory complex subunit levels using the OXPHOS antibody cocktail in cardiac mitochondria isolated from WT and NDUFS4+/- mice. Bands corresponding 17

ACCEPTED MANUSCRIPT

to the complex I subunit (NDUFB8) and complex IV subunit (MTCO1) were quantified using Image J software. Membranes were then stripped and reprobed for MnSOD as the loading control. N=2. C) Body weight measurements for WT and NDUFS+/- mice. N=6, mean±SEM. D.

RI PT

WT and NDUFS4+/- liver and heart weight at 8 weeks of age. N=6, mean±SEM.

Figure 2: Partial ablation of the Ndufs4 gene does not alter the rate of ROS release from liver or cardiac mitochondria but increases the levels of H2O2 quenching enzymes in

SC

cardiac mitochondria. A) O2●-/H2O2 release from liver mitochondria prepared from WT and

M AN U

NDUFS4+/- mice oxidizing pyruvate (50 µM) and malate (50 µM), α-ketoglutarate (50 µM) and malate (50 µM), or succinate (50 µM). Rates for ROS release were estimated using Amplex Ultra Red (10 µM). B) O2●-/H2O2 release from cardiac mitochondria prepared from WT and NDUFS4+/- mice oxidizing pyruvate (50 µM) and malate (50 µM), α-ketoglutarate (50 µM) and malate (50 µM), or succinate (50 µM). Rates for ROS release were estimated using Amplex

TE D

Ultra Red (10 µM). N=5, mean±SEM. C) Cardiac mitochondria from WT and NDUFS4+/- mice at 8 weeks of age were immunoblotted for GPX1 and TRX2 to ascertain if the partial absence of NDUFS4 alters H2O2 quenching capacity. SDHA served as the loading control. Bands were

EP

quantified using ImageJ software. N=2.

AC C

Figure 3: Complex II and III are high capacity sites for ROS release from liver and cardiac mitochondria. A) Depiction of the two experimental systems utilized to estimate ROS release rates from individual sites of production in liver and cardiac mitochondria. Dotted arrows represent directions of electron flow in the two systems. Purples circles indicate sites of O2●/H2O2 release in the two systems. IF: complex I FMN, IIF: complex II FAD, QO: outer leaflet UQH2 binding pocket in complex III, KF: α-ketoglutarate dehydrogenase FAD, PF: pyruvate dehydrogenase FAD. B). O2●-/H2O2 release profile for liver mitochondria isolated from WT and 18

ACCEPTED MANUSCRIPT

NDUFS4+/- mice at 8 weeks of age oxidizing pyruvate (50 µM) and malate (50 µM), αketoglutarate (50 µM) and malate (50 µM) in absence or presence of CPI-613 (250 µM), 3methyl-2-oxovaleric acid (KMV; 10 mM), myxothiazol (Myxo; 4 µM), or atpenin A5 (AA5; 40

RI PT

µM). Rates of ROS release were measured using Amplex UltraRed (10 µM). Rates of O2●-/H2O2 release were normalized to rates measured in the absence of inhibitor to ascertain the overall contribution of the different sites towards ROS production. N=5, mean ±SEM, 2-way ANOVA

SC

with a Tukey’s post-hoc test. C) O2●-/H2O2 release profile for cardiac mitochondria isolated from WT and NDUFS4+/- mice at 8 weeks of age oxidizing pyruvate (50 µM) and malate (50 µM), α-

M AN U

ketoglutarate (50 µM) and malate (50 µM), or succinate (50 µM) in absence or presence of CPI613 (250 µM), 3-methyl-2-oxovaleric acid (KMV; 10 mM), (Rot; 4 µM), myxothiazol (Myxo; 4 µM), or atpenin A5 (AA5; 40 µM). Rates of ROS release were measured using Amplex UltraRed (10 µM). Rates of O2●-/H2O2 release were normalized to rates measured in the absence

TE D

of inhibitor to ascertain the overall contribution of the different sites towards ROS production. N=5, mean ±SEM, 2-way ANOVA with a Tukey’s post-hoc test. Figure 4: Partial ablation of Ndufs4 sensitizes the myocardium to reperfusion injury which

EP

correlates with increased ROS release. A). Hearts were excised, placed in the Langendorff

AC C

setup, and the aorta cannulated. Hearts were then perfused with Krebs-Henseleit buffer at a constant flow of 4 mL/min for 30 minutes followed by 15 minutes of stop-flow ischemia and then 30 minutes of reperfusion. Left ventricular developed pressure (LVDP) and heart rate (HR) were measured before, during, and after ischemia to ascertain heart function and recovery. N=5, mean±SEM, paired two-tailed Student T-Test. B) Determination of the extent of myocardial necrosis after heart reperfusion by TTC staining. Infarct size was quantified using ImageJ software. N=9, mean±SEM, paired two-tailed Student T-Test. C) Mitochondria were isolated

19

ACCEPTED MANUSCRIPT

from WT and NDUFS4+/- hearts subje

cted to 30 minutes of reperfusion and O2●-/H2O2

release rates were measured in samples oxidizing pyruvate (50 µM) and malate (50 µM), αketoglutarate (50 µM) and malate (50 µM), or succinate (50 µM). Rates of ROS release were

RI PT

estimated using Amplex UltraRed (10 µM). D) Estimation of NAD(P)H levels in mitochondria isolated from WT and NDUFS4+/- hearts subjected to reperfusion injury. N=4-5, mean±SEM,

AC C

EP

TE D

M AN U

SC

paired two-tailed Student T-Test.

20

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

21

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

22

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

23

ACCEPTED MANUSCRIPT

Highlights

EP

TE D

M AN U

SC

RI PT

Partial loss of complex I does not alter ROS production. Complex III is the major source of ROS in heart mitochondria. NDUFS4 hets are sensitized to IR injury. IR injury increases ROS production in NDUFS4 het mitochondria.

AC C

• • • •