Redox Biology 1 (2013) 304–312

Contents lists available at SciVerse ScienceDirect

Redox Biology
journal homepage: www.elsevier.com/locate/redox

Research Paper

Sites of reactive oxygen species generation by mitochondria

oxidizing different substrates
Casey L. Quinlan a,n, Irina V. Perevoshchikova a, Martin Hey-Mogensen a,b,
Adam L. Orr a, Martin D. Brand a
a
The Buck Institute for Research on Aging, Novato, CA 94945, USA
b
Department of Biomedical Sciences, Center for Healthy Aging, Copenhagen University, Denmark

art ic l e i nf o a b s t r a c t

Article history: Mitochondrial radical production is important in redox signaling, aging and disease, but the relative
Received 22 March 2013 contributions of different production sites are poorly understood. We analyzed the rates of superoxide/
Received in revised form H2O2 production from different defined sites in rat skeletal muscle mitochondria oxidizing a variety of
4 April 2013
conventional substrates in the absence of added inhibitors: succinate; glycerol 3-phosphate; palmitoyl-
Accepted 5 April 2013
carnitine plus carnitine; or glutamate plus malate. In all cases, the sum of the estimated rates accounted
fully for the measured overall rates. There were two striking results. First, the overall rates differed by an
Keywords: order of magnitude between substrates. Second, the relative contribution of each site was very different
Superoxide with different substrates. During succinate oxidation, most of the superoxide production was from the
Hydrogen peroxide site of quinone reduction in complex I (site IQ), with small contributions from the flavin site in complex I
Respiratory complexes
(site IF) and the quinol oxidation site in complex III (site IIIQo). However, with glutamate plus malate as
NADH
substrate, site IQ made little or no contribution, and production was shared between site IF, site IIIQo and
Ubiquinone
Cytochrome b
2-oxoglutarate dehydrogenase. With palmitoylcarnitine as substrate, the flavin site in complex II (site IIF)
was a major contributor (together with sites IF and IIIQo), and with glycerol 3-phosphate as substrate, five
different sites all contributed, including glycerol 3-phosphate dehydrogenase. Thus, the relative and
absolute contributions of specific sites to the production of reactive oxygen species in isolated
mitochondria depend very strongly on the substrates being oxidized, and the same is likely true in cells
and in vivo.
& 2013 The Authors. Published by Elsevier B.V. Open access under CC BY-NC-SA license.

Introduction It is well established that isolated mitochondria can produce

hydrogen peroxide (H2O2) in vitro [5,6]. Since the earliest observa-
An increasing number of hypotheses propose that production tions by Chance and colleagues [7], this field has expanded
of mitochondrial reactive oxygen species (ROS) plays a crucial role considerably and many characteristics of mitochondrial super-
in different areas of physiology and pathology [1–4]. Despite this, oxide/H2O2 production have been revealed. ROS are produced by
we know very little about which mitochondrial sites in the the leak of electrons from donor redox centers to molecular
electron transport chain and associated metabolic enzymes are oxygen. The mitochondrial electron transport chain (Fig. 1) con-
responsible for physiological or pathological ROS production under sists of several complexes containing multiple redox centers that
native conditions (i.e. in the absence of added inhibitors). normally facilitate transfer of electrons to their final acceptor,
molecular oxygen, which is reduced by four electrons to water at
complex IV. Premature single electron reduction of molecular
oxygen earlier in the chain forms the superoxide radical (O2
−)
and divalent reduction forms H2O2. Superoxide dismutase-2 in the
matrix converts superoxide to H2O2, which can escape and be
Abbreviations: IF, flavin site of complex I; IQ, quinone-binding site of complex I; assayed in the surrounding medium. The general term ‘ROS’ can
IIF, flavin site of complex II; IIIQo, quinol oxidation site of complex III; CDNB, refer to several different species, but in this context we use it to
1-chloro-2,4-dinitrobenzene; Eh, redox potential; ETF, electron transferring flavo- refer only to superoxide or H2O2.
protein; ETF:QOR, ETF:ubiquinone oxidoreductase; mGPDH, mitochondrial glycerol
Using inhibitors to manipulate the redox states of particular
3-phosphate dehydrogenase; OGDH, 2-oxoglutarate dehydrogenase; PDH, pyruvate
dehydrogenase; Q, ubiquinone; QH2, ubiquinol; ROS, reactive oxygen species. sites and prevent superoxide generation from others, at least ten
n different sites of superoxide/H2O2 production in the electron
Correspondence to: The Buck Institute for Research on Aging, 8001 Redwood
Blvd., Novato, CA 94945, USA. Tel.: +1 415 493 3676. transport chain and associated enzymes (Krebs cycle, β-oxidation
E-mail address: [email protected] (C.L. Quinlan). etc.) have been identified in mammalian mitochondria (Fig. 1).

2213-2317 & 2013 The Authors. Published by Elsevier B.V. Open access under CC BY-NC-SA license.
http://dx.doi.org/10.1016/j.redox.2013.04.005
 C.L. Quinlan et al. / Redox Biology 1 (2013) 304–312 305

Fig. 1. Isopotential groups, electron flow, and endogenous reporters of superoxide/H2O2 production at different sites in mitochondria. The three planes represent different
isopotential groups of redox centers, each operating at about the same redox potential (Eh): NADH/NAD+ at Eh∼−280 mV, QH2/Q at Eh∼+20 mV, and cytochrome c at
Eh∼+320 mV [31]. The normal flow of electrons from substrate dehydrogenases through NADH to oxygen is indicated by the large green arrows dropping down through the
isopotential planes. The enzymes that feed electrons into each isopotential group are represented as ovals, and relevant inhibitors are drawn with blunted arrows. Electrons
from NAD-linked substrates enter the NADH/NAD+ pool at Eh∼−280 mV through appropriate NAD oxidoreductases (NOR) including 2-oxoglutarate dehydrogenase (OGDH)
and malate dehydrogenase (MDH) and flow into complex I (site IF). They then drop down via site IQ to QH2/Q at Eh∼+20 mV in the next isopotential pool, providing the
energy to pump protons and generate protonmotive force (pmf). Q oxidoreductases (QOR) including complex II and mitochondrial glycerol 3-phosphate dehydrogenase
(mGPDH) can also pass electrons into the Q pool. Electrons flow from QH2 through complex III to cytochrome c at Eh∼+320 mV in the next isopotential pool, again pumping
protons and generating pmf. Ascorbate plus N,N,N′,N′-tetramethyl-phenylenediamine (TMPD) and cytochrome c oxidoreductases including sulfite oxidase can also pass
electrons to cytochrome c. Finally, electrons flow through complex IV to oxygen at Eh∼+600 mV, generating pmf. The redox state of NADH (outlined in blue), measured as
NAD(P)H using autofluorescence, reports the redox state of the first isopotential group. The redox state of cytochrome b566 (outlined in blue), measured using absorbance
spectroscopy, reports the redox state of the second isopotential group [21]. Red dots are sites of superoxide/H2O2 production. Named sites are the flavin/lipoate of
2-oxoglutarate dehydrogenase (OF/L), the complex I flavin and Q-binding sites (IF and IQ, respectively), the flavin site of complex II (IIF), the flavin/quinone site of mGPDH (GF/
Q), and the outer quinol-binding site of complex III (IIIQo). Other sites, both known and unknown, are indicated by unlabeled red dots. These include less well-described sites
such as pyruvate dehydrogenase, the ETF/ETF:QOR system, proline dehydrogenase, and dihydroorotate dehydrogenase.

Each of these sites has been at least partially characterized. The respectively), then measuring the redox state of the reporter under
sites that are often invoked as the most important mitochondrial native conditions in the absence of added inhibitors to predict the
superoxide producers are in respiratory complexes I and III [5,6]. contribution of the reported site to overall H2O2 production [21].
In complex I there are two sites: the flavin in the NADH-oxidizing In the present study, we extend this approach of using
site (site IF) and the ubiquinone-reducing site (site IQ) [8]. endogenous reporters under native conditions to encompass many
In complex III, the superoxide is thought to arise from the quinol more superoxide/H2O2-producing sites and a greater variety of
oxidizing site (site IIIQo) [9–11]. However, other sites of superoxide/ substrates. We determine the contributions of each site to overall
H2O2 production have also been defined, including 2-oxoglutarate H2O2 production by isolated skeletal muscle mitochondria oxidizing
dehydrogenase (OGDH) [12,13]; pyruvate dehydrogenase (PDH) four different substrate combinations in the absence of inhibitors:
[14]; complex II (site IIF) [15]; and glycerol 3-phosphate dehydro- (a) succinate, (b) glycerol 3-phosphate, (c) palmitoylcarnitine plus
genase (mGPDH) [16]. In addition, there are suggestions that other carnitine, and (d) glutamate plus malate. The results show that the
less well-described sites may also be involved in H2O2 production: absolute and relative contribution of each site differs greatly with
the electron transferring flavoprotein/ETF:Q oxidoreductase (ETF/ different substrates.
ETF:QOR) system of fatty acid β-oxidation [17,18]; proline dehy-
drogenase [19]; and dihydroorotate dehydrogenase [16,20].
Despite our understanding of the superoxide/H2O2-producing Materials and methods
capacities of mitochondrial enzymes in vitro, we know very little
about the native ROS-producing behavior of mitochondria in vitro or Animals, mitochondrial preparation, and reagents
in situ. This is because the standard way to implicate a specific ROS-
producing site in a particular phenotype is to inhibit or genetically Female Wister rats (Harlan Laboratories), age 5–8 weeks, were
modify the site, and observe the change in ROS signal or in the fed chow ad libitum and given free access to water. Mitochondria
downstream phenotype. However, this approach is fundamentally from hind limb skeletal muscle were isolated at 4 1C in Chappell–
flawed, because blocking a site of electron transport will invariably Perry buffer (CP1; 100 mM KCl, 50 mM Tris, 2 mM EGTA, pH 7.1 at
interrupt normal electron flow and alter the redox states of other 25 1C) by standard procedures [22]. The animal protocol was
sites in the electron flow pathway, which can dramatically alter their approved by the Buck Institute Animal Care and Use Committee,
rates of ROS production, leading to unreliable or incorrect conclu- in accordance with IACUC standards. All reagents were from Sigma
sions. This raises the question: how can the individual contributions except Amplex UltraRed, which was from Invitrogen.
from a complex suite of superoxide/H2O2-producing sites be assessed
within intact mitochondria under native conditions? To address this Superoxide/H2O2 production
question, we developed a novel method of estimating the rates
of superoxide generation from two specific sites (IF and IIIQo) by Rates of superoxide/H2O2 production were measured collec-
determining the dependence of superoxide production from each tively as rates of H2O2 production, as two superoxide molecules are
site (defined using inhibitors) on the redox state of its electron donor dismutated by endogenous or exogenous superoxide dismutase to
(reported by the redox states of NAD(P)H and cytochrome b566, yield one H2O2. H2O2 was detected using the horseradish peroxidase
306 C.L. Quinlan et al. / Redox Biology 1 (2013) 304–312

and Amplex UltraRed detection system [22]. Mitochondria (0.3 mg experiments, but for full disclosure we use “NAD(P)H”) [21,23] using a
protein ml−1) were suspended under non-phosphorylating condi- Shimadzu RF5301-PC or Varian Cary Eclipse spectrofluorometer at
tions in medium at 37 1C containing 120 mM KCl, 5 mM Hepes, λexcitation ¼365 nm, λemission ¼450 nm. NAD(P)H was assumed to be 0%
5 mM K2HPO4, 2.5 mM MgCl2, 1 mM EGTA, and 0.3% (w/v) bovine reduced after 5 min without added substrate and 100% reduced with
serum albumin (pH 7.0 at 37 1C), together with 5 U ml−1 horseradish 5 mM malate plus 5 mM glutamate and 4 mM rotenone. Intermediate
peroxidase, 25 U ml−1 superoxide dismutase, 50 mM Amplex UltraRed values were determined as % reduced NAD(P)H relative to the 0% and
and, except for experiments with palmitoylcarnitine, 1 mg ml−1 oli- 100% values.
gomycin. Reactions were monitored fluorometrically in a Shimadzu
RF5301-PC or Varian Cary Eclipse spectrofluorometer (λexcitation ¼
Cytochrome b566 redox state
560 nm, λemission ¼ 590 nm) with constant stirring, and calibrated
with known amounts of H2O2 [22].
Experiments were performed at 1.5 mg mitochondrial pro-
To correct for losses of H2O2 caused by peroxidase activity in
tein ml−1 in parallel with measurements of H2O2 production and
the matrix and give a better estimate of superoxide/H2O2 produc-
NAD(P)H redox state in the same medium. The reduction state of
tion rates, all H2O2 production rates were mathematically cor-
endogenous cytochrome b566 was measured with constant stirring
rected to the rates that would have been observed in these
at 37 1C in an Olis DW-2 dual wavelength spectrophotometer as
mitochondria after pre-treatment with 1-chloro-2,4-dinitroben-
the absorbance difference at 566–575 nm [11,21]. Cytochrome b566
zene (CDNB) to deplete glutathione and decrease glutathione
was assumed to be 0% reduced after 5 min without added
peroxidase and peroxiredoxin activity, as described in Refs.
substrate and 100% reduced with saturating substrates plus anti-
[21,23], using an empirical equation
mycin A. Intermediate values were determined as % reduced b566
vCDNB ¼ vcontrol þ ð100  vcontrol Þ=ð72:6 þ vcontrol Þ ð1Þ relative to the 0% and 100% values.
−1 −1
(rates in pmol H2O2 min mg protein ).
Definition of sites and calibration of endogenous reporters
NAD(P)H redox state
Site IF was defined as the site producing superoxide (measured
Experiments were performed using 0.3 mg mitochondrial protein as H2O2) in the presence of malate to reduce NAD to NADH, and
ml−1 at 37 1C in parallel with measurements of H2O2 production and rotenone to inhibit reoxidation of complex I by the Q-pool. Any
cytochrome b566 redox state in the same medium with the same H2O2 arising from reverse flow from the NADH pool into NAD
additions. The reduction state of endogenous NAD(P)H was deter- oxidoreductases such as OGDH or PDH will appear in the analysis
mined by autofluorescence (most of the autofluorescence signal is as a component of site IF (Fig. 2a). To decrease the contributions of
from NADH bound in the matrix, and NADPH hardly changes in our forward electron flow at OGDH and PDH to H2O2 production

Fig. 2. Calibration of endogenous reporters in isolated rat skeletal muscle mitochondria. (a) Definition of site IF (red dot) using malate as substrate in the presence of
rotenone to inhibit site IQ and cause reduction of upstream sites and oxidation of all downstream sites. The definition includes any small contribution from backflow of
electrons from NADH into other sites, e.g. OGDH and PDH, shown in pink. PF/L denotes the flavin/lipoate of pyruvate dehydrogenase. (b) Dependence of superoxide
production (measured as extramitochondrial appearance of H2O2) from site IF on the redox state of NAD(P)H (measured by autofluorescence). Malate was titrated from 20 μM
−5 mM in the presence of 4 μM rotenone, 1.5 mM aspartate and 2.5 mM ATP. (c) Definition of site IIIQo (red dot) using succinate as substrate, rotenone to inhibit site IQ, and
sufficient succinate+malonate (5 mM) to suppress site IIF. The rate of superoxide/H2O2 production from site IIIQo was defined as the rate that was inhibited by 2 mM
myxothiazol (after correction for changes at site IF using the calibration curve in (b)). The definition includes any small contribution from backflow of electrons from QH2 into
other sites, e.g. mGPDH or other Q-oxidoreductases, shown in pink. (d) Dependence of myxothiazol-sensitive superoxide production from site IIIQo on the redox state of
cytochrome b566 (measured by the absorbance spectroscopy). Electron input was titrated by adding different ratios of succinate:malonate (sum ¼5 mM) in the presence of
4 mM rotenone. Data are means 7 SEM (n≥4) recalculated from [21] using the new calibration in (b). For curve-fitting see Materials and methods.
 C.L. Quinlan et al. / Redox Biology 1 (2013) 304–312 307

assigned to site IF, we added 1.5 mM aspartate to remove endo- different redox potentials (Eh): at the isopotential group of redox
genous 2-oxoglutarate by transamination and 2.5 mM ATP to carriers around the NADH/NAD+ pool at Eh∼−280 mV and at the
decrease carbon flows at various points in the Krebs cycle, isopotential group of redox carriers around the QH2/Q pool at
particularly succinate thiokinase. This change to our previous Eh∼+20 mV (Fig. 1). At each isopotential group, an important
protocol [18,21] strongly decreased our earlier estimates of the determinant of the rate of superoxide or H2O2 production is the
rates from site IF. The rate of superoxide/H2O2 production from site redox state: a more reduced carrier generally leaks electrons to
IF defined by inhibitors in this way was measured as a function of oxygen at a faster rate. We exploited this relationship to create
the redox state of NAD(P)H at different malate concentrations assays of the rate of superoxide/H2O2 production from different
between 0.02 mM and 5 mM (Fig. 2b). The data were arbitrarily sites, either by defining the site precisely with inhibitors and
fitted by non-linear regression to a single exponential, to give the determining its dependence on the redox state of the appropriate
parameter values in the following equation: pool as described in Materials and methods (Fig. 2), or by
inhibiting the site and determining the change in superoxide/
vH2 O2 ð%NADðPÞHÞ ¼ 0:88  eð0:04ð%NADðPÞHÞÞ þ 26:5 ð2Þ
H2O2 production after correction for secondary changes in the
where vH2 O2 is the rate of H2O2 production. This equation was used redox states of the two pools. Specifically, to predict rates from
to predict the rate of superoxide/H2O2 production from site IF at each site with different respiratory substrates in the absence of
any observed NAD(P)H redox state in the absence of inhibitors in inhibitors, we measured the redox states of the endogenous
subsequent experiments. reporters NAD(P)H and cytochrome b566 as proxies of the redox
Superoxide production by site IIIQo was measured as H2O2 states of the two isopotential groups, and determined the rates of
production in the presence of succinate to reduce the Q pool, superoxide/H2O2 production from sites IF and IIIQo, respectively,
malonate to keep total succinate plus malonate at 5 mM and using the calibration curves in Fig. 2. Assessing the contribution of
inhibit superoxide/H2O2 formation from site IIF without fully other sites was a little more complicated, so we first provide an
inhibiting succinate oxidation [15], and rotenone to inhibit super- example of how the endogenous reporter calibration curves and
oxide formation from site IQ (Fig. 2c). Site IIIQo was defined as the the careful use of inhibitors can be used to quantify the sites of
component of the observed H2O2 production under these condi- H2O2 production during oxidation of the substrate succinate.
tions that was sensitive to myxothiazol (a specific inhibitor of site
IIIQo) after correction for the small difference in rates from site IF Sites of superoxide/H2O2 production during oxidation
before and after myxothiazol addition (calculated from parallel of succinate: a worked example
measurements of NAD(P)H and application of the calibration curve
in Fig. 2b). The rate of superoxide/H2O2 production from site IIIQo When succinate is oxidized in the absence of rotenone and other
defined by inhibitors in this way was measured as a function of the electron transport chain inhibitors, electrons flow through complex II
redox state of cytochrome b566 at different succinate: malonate into the Q pool. Next, there are two options (Fig. 3a). The electrons
ratios ranging from 75% to 100% succinate at 5% increments (total can flow forward to complex III and thermodynamically downhill to
dicarboxylate concentration 5 mM) (Fig. 2d). The data were more oxidized isopotential groups at cytochrome c and H2O/O2,
arbitrarily fitted in the same way as for site IF to give the pumping protons at complexes III and IV and generating protonmo-
parameter values using the following equation: tive force. Alternatively, they can flow in reverse to complex I, driven
thermodynamically uphill to the more negative isopotential group by
vH2 O2 ð%b566 Þ ¼ 5:22  eð0:064ð%b566 ÞÞ −5:22 ð3Þ
reversal of the proton pumps in complex I driven by the proto-
This equation was used to predict the rate of superoxide/H2O2 nmotive force generated by proton pumping at complexes III and IV.
production from site IIIQo at any observed cytochrome b566 redox In the process, a number of possible sites of superoxide/H2O2
state in the absence of inhibitors in subsequent experiments. production may be engaged, particularly sites IIF, IIIQo, IQ and IF
See Ref. [21] for more extensive discussion and descriptions. (Fig. 3a). However, it is generally agreed that the primary mechanism
of H2O2 production during succinate oxidation is reverse electron
Statistics transport into complex I, because this H2O2 production is very
sensitive to rotenone, the classic Q-site inhibitor of complex I
When using the calibration curves in Fig. 2 to calculate rates of [24,25]. What is not usually discussed is that when an inhibitor such
H2O2 production at a given reduction state of the reporter, the as rotenone is added there are subsequent shifts in electron
error in the measurements during calibration was taken into distribution that invariably change the rates of superoxide/H2O2
account. This error was calculated by standard methods of error production by sites both upstream and downstream of the inhibition
propagation through all the steps, as detailed in [21]. site [21]. Therefore, the decrease in rate observed after the addition
The significance of differences between reported and experi- of an inhibitor is not an accurate indication of the rate from its target
mentally measured rates of H2O2 production in each experimental site before inhibition. The endogenous reporter method described
condition was tested using Welch's t-test. Because error propaga- here circumvents this problem: by monitoring the changes in the
tion was used to capture uncertainty in the calibration curves, NADH isopotential group (through NAD(P)H) and the Q isopotential
individual data-points could not be used for statistical analysis. group (through cytochrome b566), the changes in electron distribu-
Instead we used the traits describing the population of data tion after inhibitors are added can be quantified and corrected for.
(mean, SEM based on error propagation and number of observa- In this worked example, we quantified the sites of H2O2
tions) to calculate if differences were significant (po 0.05). production during succinate oxidation by measuring the redox
states of the endogenous reporters, and the changes in these redox
states after addition of rotenone. From this information, we could
Results and discussion predict the contribution of each site to the total H2O2 produc-
tion observed during succinate oxidation. To begin, we measured
The aim of the present study was to determine the contribu- the rate of H2O2 production during succinate oxidation in the
tions of different sites of superoxide/H2O2 production to the total absence of rotenone (891 pmol H2O2 min−1 mg protein−1) (Fig. 4a).
observed H2O2 generation by mitochondria during oxidation of In parallel, in different cuvettes, we measured the reduction states
different substrates in the absence of inhibitors. Electrons leak of the reporters. We observed that NAD(P)H was 89% reduced and
from the respiratory chain to generate superoxide or H2O2 at two cytochrome b566 was 46% reduced (Fig. 4b, Table 1). From
308 C.L. Quinlan et al. / Redox Biology 1 (2013) 304–312

Fig. 3. Electron flows and sites of superoxide/H2O2 production during oxidation of different substrates by rat skeletal muscle mitochondria. Electron flows and postulated
(red and pink dots) and observed (red dots) sites of superoxide/H2O2 production during oxidation of (a) succinate, (b) glycerol 3-phosphate, (c) palmitoylcarnitine plus
carnitine and (d) glutamate plus malate.

this information, we could predict using the equations describing rate of H2O2 production while site IF accounted for 7% and site IIIQo
the calibration curves in Fig. 2 that site IF was producing accounted for 10%. Since there was no unaccounted-for rate of
59 pmol H2O2 min−1 mg protein−1 (7% of the total), and site IIIQo H2O2 production, either before (891 pmol H2O2 min−1 mg protein−1
was producing 91 pmol H2O2 min−1 mg protein−1 (10%) (Fig. 4c, observed versus 886 pmol H2O2 min−1 mg protein−1 assigned, N.S.)
Table 1), with the remaining 741 pmol H2O2 min−1 mg protein−1 or after addition of rotenone (131 pmol H2O2 min−1 mg protein−1
coming from other sites. observed versus 171 pmol H2O2 min−1 mg protein−1 assigned, N.S.),
To identify and quantify these other sites, we added rotenone to no other site made a substantial contribution to the total observed
inhibit site IQ, and measured the changes in H2O2 production and the rate. The reason site IIF did not contribute is that the conventional
changes in the reduction state of the reporters (Figs. 4a and b, Table 1). succinate concentration used (5 mM) was sufficiently high to effec-
After rotenone addition, the overall rate of H2O2 production dropped tively abolish production of superoxide/H2O2 by site IIF [15].
to 131 pmol H2O2 min−1 mg protein−1, NAD(P)H became significantly There are no surprises in these conclusions (although they are
more oxidized and cytochrome b566 appeared to become slightly more more accurate than simply assigning the rotenone-sensitive signal
reduced (Fig. 4 a and b, Table 1). Again, using the calibration curves in to complex I): site IQ is the dominant superoxide/H2O2 producer
Fig. 2 we determined that site IF now appeared to produce slightly less when succinate is oxidized in the absence of rotenone. However,
superoxide/H2O2 (49 pmol H2O2 min−1 mg protein−1) and site IIIQo this example illustrates clearly that during oxidation of a single
appeared to produce more (124 pmol H2O2 min−1 mg protein−1) than substrate, superoxide/H2O2 are produced from multiple sites
before the addition of rotenone (Fig. 4c, Table 1). When rotenone was (Fig. 4d). It also clearly illustrates that site-specific inhibitors not
added, not only was reverse electron flow abolished, leading to partial only affect superoxide/H2O2 production at the site that they
oxidation of NAD(P)H as the electron supply from succinate through inhibit, but also (through changes in electron distribution) change
complex I failed, but there appeared to be a small increase in the superoxide/H2O2 production at other sites.
reduction level of cytochrome b566 as all flow was diverted through
complex III. The total change in superoxide/H2O2 production at site IF The contributions of specific sites to superoxide/H2O2 production
was calculated by subtracting the rate before rotenone addition during oxidation of different substrates
(59 pmol H2O2 min−1 mg protein−1) from the rate after rotenone
addition (49 pmol H2O2 min−1 mg protein−1) to give a final change in The approach described above as a worked example using
rate from site IF of −10 pmol H2O2 min−1 mg protein−1. The equivalent succinate as substrate was used to define the contributions of
calculation performed for site IIIQo gave a total change of +33 pmol specific sites to superoxide/H2O2 production during oxidation of
H2O2 min−1 mg protein−1. The rotenone-sensitive rate (759 pmol three additional important conventional substrates: glycerol
H2O2 min−1 mg protein−1) was then corrected for the changes in 3-phosphate, palmitoylcarnitine plus carnitine, and glutamate plus
superoxide/H2O2 production from sites IF and IIIQo to give a final malate. In each condition, the reduction state of the reporters was
adjusted rate of 736 pmol H2O2 min−1 mg protein−1, all assigned to site measured in parallel with H2O2 production. The changes in
IQ (it was not from site IF, as has been suggested [26], because site IF reduction state of the reporters after addition of inhibitors of
superoxide production was fully corrected for using NAD(P)H as the complex I or complex II were measured and used to correct for
endogenous reporter). This painted a final picture (Fig. 4d) showing consequent changes in electron distribution and superoxide/H2O2
that during succinate oxidation, site IQ contributed 83% of the observed production at other sites. Table 1 details the observed redox states
 C.L. Quinlan et al. / Redox Biology 1 (2013) 304–312 309

Fig. 4. Sites of superoxide/H2O2 production during succinate oxidation: a worked example. (a) Representative trace of Amplex UltraRed/resorufin fluorescence during
oxidation of 5 mM succinate by rat skeletal muscle mitochondria before and after addition of 4 mM rotenone. Numbers by the traces represent average calibrated rates in
pmol H2O2 min−1 mg protein−1. (b) Reduction states of the endogenous reporters NAD(P)H and cytochrome b566 before and after rotenone addition. The difference in
reduction state before and after rotenone was significant for NAD(P)H (p ¼ 0.01), but not for cytochrome b566 (p ¼ 0.11) by Student's t-test. (c) Rates of superoxide/H2O2
production from sites IF and IIIQo before and after rotenone addition, predicted from (b) using the calibration curves in Fig. 2. Results are means 7 SEM (n ¼4). To account for
error in both the measurements and the calibration curves, a Welch's t-test was used. The difference in rates before and after rotenone did not fall within the 95% confidence
interval (p ¼0.09 for site IF and 0.15 for IIIQo). (d) Observed total rate of H2O2 production during succinate oxidation (before addition of rotenone) and assigned rates of
superoxide/H2O2 production from sites IF, IIIQo and IQ during succinate oxidation. Results are means 7SEM (n ¼4); the error bars on the sum column show the combined
propagated errors in the total sum value. There was no significant difference between the observed rate and the sum of the assigned rates (Welch's t-test; p o0.05).

of the reporters with each substrate before and after addition of calibration curve in Fig. 2c). With glycerol 3-phosphate, palmi-
the inhibitors rotenone or malonate. toylcarnitine plus carnitine, and glutamate plus malate, the con-
Fig. 5 shows the central result of the present paper: the contribu- tribution of site IIF was estimated from the decrease in rate of H2O2
tions of each of the individual sites of superoxide/H2O2 production to production after addition of malonate followed by correction for
overall H2O2 generation by isolated skeletal muscle mitochondria with changes in the rate from site IF (calculated from the redox state of
each of the four substrate combinations under native conditions in the NAD(P)H before and after malonate addition) and from site IIIQo
absence of added inhibitors. The first striking result is that the overall (calculated from the redox state of cytochrome b566 before and
rates of H2O2 generation were very different with different substrates, after malonate addition). The contributions of mGPDH (with
as previously observed by [27] and others [28–30]. The second, crucial, glycerol 3-phosphate as substrate) and OGDH (with glutamate
observation is that the relative contribution of each site was very plus malate as substrate) were estimated as the difference between
different with different substrates. Thus, the relative and absolute the sum of the rates estimated for the other sites and the observed
contributions of specific sites to the production of reactive oxygen total rate.
species in isolated mitochondria depend very strongly on the sub-
strates being oxidized.
To determine the contributions of individual sites to overall Sites of superoxide/H2O2 production during oxidation
H2O2 generation with each substrate, the rates of H2O2 production of glycerol 3-phosphate
at site IF and IIIQo were first estimated from the redox states of
NAD(P)H and cytochrome b566 using the calibration curves in Glycerol 3-phosphate oxidation by mGPDH results directly in Q
Fig. 2. With succinate and glycerol 3-phosphate, the contribution pool reduction (Fig. 3b). From the Q pool, electrons may flow both
of site IQ was then estimated from the decrease in the rate of H2O2 forward (to complex III and beyond, generating protonmotive
production after addition of rotenone followed by correction for force) and in reverse (to complex I and complex II). In the process,
changes in the rate from site IF (calculated from the redox state of a number of possible sites of superoxide/H2O2 production may be
NAD(P)H before and after rotenone addition using the calibration engaged, particularly mGPDH and sites IIF, IIIQo, IQ and IF (Fig. 3b).
curve in Fig. 2a) and from site IIIQo (calculated from the redox state Protonmotive force is required for reverse electron flow to com-
of cytochrome b566 before and after rotenone addition using the plex I [25], but not for reversal into complex II [16]. Complex II has
310 C.L. Quinlan et al. / Redox Biology 1 (2013) 304–312

been shown to generate superoxide in both the forward and uncharacterized sites, since such sites would have to be engaged
reverse reactions [15]. during oxidation of glycerol 3-phosphate but not during oxidation of
With glycerol 3-phosphate as substrate, the total observed rate of succinate (where there was no unassigned H2O2 production, Fig. 5),
H2O2 production was 622 pmol H2O2 min−1 mg protein−1. Sites IF despite the less extensive reduction of the two isopotential groups
(6%) and IIIQo (4%) made only modest contributions to this total rate with glycerol 3-phosphate as substrate (Table 1).
(Fig. 5). The contributions of sites IQ (33%) and IIF (26%) (after
correction for changes in sites IF and IIIQo following the addition of Sites of superoxide/H2O2 production during oxidation
inhibitors) were more substantial. The absolute rates from sites IF, of palmitoylcarnitine plus carnitine
IIIQo and IQ were lower with glycerol 3-phosphate than with
succinate, because the two isopotential pools were less reduced Carnitine enhances oxidation of palmitoylcarnitine by remov-
(Table 1), presumably because succinate was a better substrate than ing inhibitory acetyl-CoA in the form of acetylcarnitine and by
glycerol 3-phosphate in this experiment, which had sub-optimal promoting entry of palmitoylcarnitine into the mitochondrial
glycerol 3-phosphate and calcium concentrations. After these matrix [18]. When palmitoylcarnitine is metabolized by the
assignments, 192 pmol H2O2 min−1 mg protein−1 (31% of the total) β-oxidation pathway, electrons enter the respiratory chain at two
remained unaccounted for. mGPDH itself is known to generate ROS sites: from acyl-CoA dehydrogenase through the electron transfer-
(predominantly superoxide), which are released to both the matrix ring flavoprotein (ETF) and ETF:ubiquinone oxidoreductase (ETF:
and the intermembrane space [16]. In this case, we assume that the QOR) to the Q-pool, and from 3-hydroxyacyl-CoA dehydrogenase
remaining observed rates of H2O2 production must have arisen from to NADH (Fig. 3c). Oxidation of the end product, acetyl-CoA, in the
mGPDH, since all other known sites were already accounted for, and citric acid cycle leads to further electron input through the NAD-
we assign this site by difference (31%). It is unlikely that any of the linked dehydrogenases and through complex II during oxidation of
unassigned H2O2 generation in this condition was from succinate. Electron entry by more than one route suggests that

Table 1
Reduction levels of reporters and corresponding calculated rates of H2O2 production from site IF and site IIIQo in the presence of different substrates and inhibitors. With each
substrate, the redox states of NAD(P)H and cytochrome b566 were measured in parallel with the measurements of overall rates of H2O2 production reported in Fig. 5. Eqs.
(2) and (3) describing the calibration curves in Fig. 2 were used to calculate the rates of superoxide/H2O2 production from site IF and site IIIQo from these redox
measurements. Concentrations of substrates and inhibitors are given in Fig. 5. All rates are in pmol H2O2 min−1 mg protein−1. Data are means 7SEM (n≥4). Error values on the
calculated rates of H2O2 production represent the propagated error as described in Materials and methods and Ref. [21].

Substrates and inhibitors Redox state of site IF reporter and calculated rate of Redox state of site IIIQo reporter and calculated rate of
superoxide/H2O2 production by site IF superoxide/H2O2 production by site IIIQo

Reduced Reported rate of Reduced Reported rate of

NAD(P)H (%) H2O2 production cytochrome b566 (%) H2O2 production

Succinate 897 1 59 74 467 2 917 18

Succinate+rotenone 807 2 49 74 507 1 1247 15
Glycerol 3-phosphate 59 7 3 34 76 297 2 277 7
Glycerol 3-phosphate+malonate 597 3 34 76 297 2 277 7
Glycerol 3-phosphate+malonate+rotenone 397 2 31 75 317 3 337 9
Palmitoylcarnitine+carnitine 807 1 49 74 427 3 727 18
Palmitoylcarnitine+carnitine+malonate 797 1 48 74 447 2 807 17
Glutamate+malate 857 7 54 79 387 5 547 18

Fig. 5. Native rates of superoxide/H2O2 production by mitochondria oxidizing different substrates. Observed total rate of H2O2 production (gray bars) and sum of assigned
rates of superoxide/H2O2 production from different sites (colored stacked bars) in the presence of different substrates as indicated: 5 mM succinate; 27 mM disodium rac-α/
β-glycerol phosphate (25% active optical isomer sn-glycerol 3-phosphate); 15 mM palmitoyl-L-carnitine plus 2 mM L-carnitine; and 5 mM L-glutamate plus 5 mM L-malate.
With each substrate, the reduction states of NAD(P)H and cytochrome b566 were measured in parallel with H2O2 production and the calibration curves in Fig. 2 were used to
predict the rates of production from sites IF and IIIQo. With succinate, 4 mM rotenone was subsequently added to allow calculation of the rate from site IQ as described in the
text (data from Fig. 4d). With glycerol 3-phosphate, 1 mM malonate and 4 mM rotenone were subsequently added to allow calculation of the rates from sites IIF and IQ,
respectively. The rate assigned to mGPDH was calculated by difference. With palmitoylcarnitine plus carnitine, 1 mM malonate was subsequently added to allow calculation
of the rate from site IIF (data recalculated from [18]). With glutamate plus malate, the rate assigned to OGDH was calculated by difference (data recalculated from [21]).
Results are means 7 SEM (n¼ 4–6); the error bars on the sum columns show the combined propagated errors in the total sum value. There was no significant difference
between observed and assigned rates with succinate or with palmitoylcarnitine plus carnitine (Welch's t-test; po 0.05).
 C.L. Quinlan et al. / Redox Biology 1 (2013) 304–312 311

fatty acid oxidation may generate superoxide/H2O2 from several calibration curves measured in the presence of inhibitors and
different sites, particularly ETF/ETF:QOR and sites IIIQo, IQ and IF particular substrates apply under native conditions using other
(Fig. 3c). Site IIF may also contribute by reverse flow from the Q substrates. The good agreement between calculated and measured
pool, or by forward flow from succinate. Importantly, under this total rates (Fig. 5) supports the validity of this assumption.
condition the acetyl-CoA generated by β-oxidation will tend to
deplete inhibitory oxaloacetate through the action of citrate
synthase, making site IIF more prone to generate superoxide/ Potential pathological and physiological implications
H2O2 [18].
With palmitoylcarnitine plus carnitine as substrate, the total The results obtained using the four different substrates
observed rate of H2O2 production was 199 pmol H2O2 min−1 mg pro- described in Fig. 5 provide a very clear illustration of the remark-
tein−1. From the redox states of the reporters and the corrected effects ably different H2O2-producing profiles that can be attained by
of malonate on H2O2 production, we were able to account for the isolated muscle mitochondria oxidizing conventional substrates.
entirety of this observed H2O2 production. There were approximately When taken as a whole, two results are striking. (i) The native
equal contributions from sites IIF (38%) and IIIQo (36%), with site IF rates differed greatly between substrates. This implies that mito-
producing slightly less (25%) (Fig. 5). The absence of unaccounted-for chondrial superoxide/H2O2 production rates in vivo likely depend
H2O2 production indicated that neither site IQ nor the ETF/ETF:QOR critically on the substrate being oxidized, so physiological or
system generated substantial superoxide/H2O2. Indeed, the ETF/ETF: pathological changes in substrate may be very important determi-
QOR system does not generate measureable superoxide/H2O2 except nants of rates of radical production, even at the same overall rate
under very specific conditions and in the presence of respiratory of oxygen consumption. (ii) The contribution of each site differed
chain inhibitors [18]. markedly between substrates. With succinate, site IQ dominated,
with relatively small contributions from sites IF and IIIQo. However,
Sites of superoxide/H2O2 production during oxidation with palmitoylcarnitine plus carnitine, site IIF was an important
of glutamate plus malate contributor, and with glycerol 3-phosphate, five sites contributed,
including site IIF and mGPDH. Thus, which sites contribute to
In the final condition, the substrate pair glutamate plus malate superoxide and H2O2 production in mitochondria, in cells, and
was used to generate NADH (Fig. 3d). During oxidation of this in vivo under both physiological and pathological conditions likely
substrate combination, malate is oxidized to oxaloacetate by malate depends critically on the substrates being utilized.
dehydrogenase and NADH is generated. Glutamate is used to Notably, the sites are known to differ markedly in the topology
transaminate the oxaloacetate to form aspartate and 2-oxoglutarate of superoxide production [6,16,27,32,33]. Essentially all super-
(and some glutamate may be oxidized to 2-oxoglutarate by gluta- oxide/H2O2 from sites IF, IQ and IIF is directed to the matrix, but
mate dehydrogenase). The aspartate is exchanged for glutamate on about half the superoxide from site IIIQo and mGPDH appears in
the glutamate–aspartate antiporter, and much of the 2-oxoglutarate the intermembrane space. Thus, the strength of mitochondrial
is exchanged for malate on the oxoglutarate–malate antiporter superoxide signaling in the cytosol (and also the amount of
(although some may be oxidized by 2-oxoglutarate dehydrogenase). oxidative damage caused by superoxide and H2O2 in the matrix)
In this way, removal of oxaloacetate maximizes NADH generation by will differ substantially between substrates, even at identical total
malate dehydrogenase [31]. rates of mitochondrial superoxide/H2O2 production.
With glutamate plus malate as substrate, the total observed rate of
H2O2 production was 182 pmol H2O2 min−1 mg protein−1. Under this
condition, sites IF and IIIQo contributed equally (30% each) [21] (Fig. 5). Acknowledgments
Malonate did not significantly inhibit the rate (3% inhibition75%), so
site IIF was not a significant contributor, presumably because it was Supported by National Institutes of Health grants P01 AG025901,
inhibited by oxaloacetate or malate under these conditions [15]. PL1 AG032118, R01 AG03354 and TL1 AG032116 (M.D.B., C.L.Q.) and
A significant proportion (41%) of the total H2O2 production was The Ellison Medical Foundation, grant AG-SS-2288-09 (M.D.B., I.V.P.).
unassigned, indicating that another site also contributed under this Fellowship support was from The Glenn Foundation (I.V.P) and The
condition. We assume this was from forward flow through OGDH Carlsberg Foundation (M.H-M).
(Fig. 5) because some 2-oxoglutarate was likely formed during
transamination of glutamate, but we cannot exclude a contribution References
from site IQ or uncharacterized other sites in this complex metabolic
condition. In principle the contribution of site IQ could be assessed by [1] S.K. Powers, J. Duarte, A.N. Kavazis, E.E. Talbert, Reactive oxygen species are
the decrease in rate observed following addition of rotenone after signalling molecules for skeletal muscle adaptation, Experimental Physiology
correction for secondary changes in other defined sites, but definitive 95 (2010) 1–9.
[2] S.J. Ralph, J. Neuzil, Mitochondria as targets for cancer therapy, Molecular
conclusions could not be drawn because of the relatively small rates Nutrition and Food Research 53 (2009) 9–28.
of unassigned H2O2 production, large changes in redox states of the [3] M. Sundaresan, Z.X. Yu, V.J. Ferrans, K. Irani, T. Finkel, Requirement for
reporters following addition of rotenone, relatively large errors generation of H2O2 for platelet-derived growth factor signal transduction,
Science 270 (1995) 296–299.
involved, and possible changes in other uncharacterized sites follow- [4] M.E. Witte, J.J. Geurts, H.E. de Vries, P. van der Valk, J. van Horssen,
ing addition of rotenone. Mitochondrial dysfunction: a potential link between neuroinflammation and
neurodegeneration? Mitochondrion 10 (2010) 411–418.
[5] M.P. Murphy, How mitochondria produce reactive oxygen species, Biochemical
Advantages and limitations of the approach used Journal 417 (2009) 1–13.
[6] M.D. Brand, The sites and topology of mitochondrial superoxide production,
The great strength of the approach we have used here to Experimental Gerontology 45 (2010) 466–472.
[7] A. Boveris, N. Oshino, B. Chance, The cellular production of hydrogen peroxide,
measure the contributions of different sites to overall mitochon-
Biochemical Journal 128 (1972) 617–630.
drial H2O2 production [21] is that, unlike all previous approaches, [8] J.R. Treberg, C.L. Quinlan, M.D. Brand, Evidence for two sites of superoxide
it reports rates under native conditions in the absence of added production by mitochondrial NADH-ubiquinone oxidoreductase (complex I),
inhibitors. This gives it great potential for future studies with Journal of Biological Chemistry 286 (2011) 27103–27110.
[9] D.M. Kramer, A.G. Roberts, F. Muller, J.L. Cape, M.K. Bowman, Q-cycle bypass
physiological substrate mixes in mitochondria and in intact cells reactions at the Qo site of the cytochrome bc1 (and related) complexes,
and organisms. Its main limitation is the assumption that the Methods in Enzymology 382 (2004) 21–45.
312 C.L. Quinlan et al. / Redox Biology 1 (2013) 304–312

[10] F.L. Muller, A.G. Roberts, M.K. Bowman, D.M. Kramer, Architecture of the Qo measured using endogenous reporters, Free Radical Biology and Medicine 53
site of the cytochrome bc1 complex probed by superoxide production, (2012) 1807–1817.
Biochemistry 42 (2003) 6493–6499. [22] C. Affourtit, C.L. Quinlan, M.D. Brand, Measurement of proton leak and
[11] C.L. Quinlan, A.A. Gerencser, J.R. Treberg, M.D. Brand, The mechanism of electron leak in isolated mitochondria, Methods in Molecular Biology 810
superoxide production by the antimycin-inhibited mitochondrial Q-cycle, (2012) 165–182.
Journal of Biological Chemistry 286 (2011) 31361–31372. [23] J.R. Treberg, C.L. Quinlan, M.D. Brand, Hydrogen peroxide efflux from muscle
[12] V.I. Bunik, C. Sievers, Inactivation of the 2-oxo acid dehydrogenase complexes mitochondria underestimates matrix superoxide production—a correction
upon generation of intrinsic radical species, European Journal of Biochemistry using glutathione depletion, FEBS Journal 277 (2010) 2766–2778.
269 (2002) 5004–5015. [24] R.G. Hansford, B.A. Hogue, V. Mildaziene, Dependence of H2O2 formation by
[13] L. Tretter, V. Adam-Vizi, Generation of reactive oxygen species in the reaction rat heart mitochondria on substrate availability and donor age, Journal of
catalyzed by alpha-ketoglutarate dehydrogenase, Journal of Neuroscience 24 Bioenergetics and Biomembrane 29 (1997) 89–95.
(2004) 7771–7778. [25] A.J. Lambert, M.D. Brand, Superoxide production by NADH:ubiquinone oxidor-
[14] A.A. Starkov, G. Fiskum, C. Chinopoulos, B.J. Lorenzo, S.E. Browne, M.S. Patel, eductase (complex I) depends on the pH gradient across the mitochondrial
M.F. Beal, Mitochondrial alpha-ketoglutarate dehydrogenase complex gener- inner membrane, Biochemical Journal 382 (2004) 511–517.
ates reactive oxygen species, Journal of Neuroscience 24 (2004) 7779–7788. [26] L. Kussmaul, J. Hirst, The mechanism of superoxide production by NADH:
[15] C.L. Quinlan, A.L. Orr, I.V. Perevoshchikova, J.R. Treberg, B.A. Ackrell, ubiquinone oxidoreductase (complex I) from bovine heart mitochondria,
Proceedings of the National Academy of Sciences of the USA 103 (2006)
M.D. Brand, Mitochondrial complex II can generate reactive oxygen species
7607–7612.
at high rates in both the forward and reverse reactions, Journal of Biological
[27] J. St-Pierre, J.A. Buckingham, S.J. Roebuck, M.D. Brand, Topology of superoxide
Chemistry 287 (2012) 27255–27264.
production from different sites in the mitochondrial electron transport chain,
[16] A.L. Orr, C.L. Quinlan, I.V. Perevoshchikova, M.D. Brand, A refined analysis of
Journal of Biological Chemistry 277 (2002) 44784–44790.
superoxide production by mitochondrial sn-glycerol 3-phosphate dehydro-
[28] F.L. Muller, Y. Liu, M.A. Abdul-Ghani, M.S. Lustgarten, A. Bhattacharya,
genase, Journal of Biological Chemistry 287 (2012) 42921–42935. Y.C. Jang, H. Van Remmen, High rates of superoxide production in skeletal-
[17] E.L. Seifert, C. Estey, J.Y. Xuan, M.E. Harper, Electron transport chain-dependent muscle mitochondria respiring on both complex I- and complex II-linked
and -independent mechanisms of mitochondrial H2O2 Emission during long- substrates, Biochemical Journal 409 (2008) 491–499.
chain fatty acid oxidation, Journal of Biological Chemistry 285 (2010) [29] F. Zoccarato, L. Cavallini, A. Alexandre, Succinate is the controller of O2−/H2O2
5748–5758. release at mitochondrial complex I: negative modulation by malate, positive
[18] I.V. Perevoshchikova, C.L. Quinlan, A.L. Orr, M.D., Brand. Sites of superoxide by cyanide, Journal of Bioenergetics and Biomembranes 41 (2009) 387–393.
and hydrogen peroxide production during fatty acid oxidation in rat skeletal [30] F. Zoccarato, L. Cavallini, S. Bortolami, A. Alexandre, Succinate modulation of
muscle. Free Radical Biology and Medicine 61, 2013, 298–309. H2O2 release at NADH:ubiquinone oxidoreductase (Complex I) in brain
[19] T.A. White, N. Krishnan, D.F. Becker, J.J. Tanner, Structure and kinetics of mitochondria, Biochemical Journal 406 (2007) 125–129.
monofunctional proline dehydrogenase from Thermus thermophilus, Journal of [31] D.G. Nicholls, S.J. Ferguson, Bioenergetics 3, Academic Press, London , 2002.
Biological Chemistry 282 (2007) 14316–14327. [32] S. Miwa, M.D. Brand, The topology of superoxide production by complex III
[20] J.H. Forman, J. Kennedy, Superoxide production and electron transport in and glycerol 3-phosphate dehydrogenase in Drosophila mitochondria, Biochi-
mitochondrial oxidation of dihydroorotic acid, Journal of Biological Chemistry mica Biophysica Acta 1709 (2005) 214–219.
250 (1975) 4322–4326. [33] F.L. Muller, Y. Liu, H. Van Remmen, Complex III releases superoxide to both
[21] C.L. Quinlan, J.R. Treberg, I.V. Perevoshchikova, A.L. Orr, M.D. Brand, Native sides of the inner mitochondrial membrane, Journal of Biological Chemistry
rates of superoxide production from multiple sites in isolated mitochondria 279 (2004) 49064–49073.