1

A Drosophila model of mitochondrial disease caused by a complex I mutation that uncouples proton

pumping from electron transfer

Jonathon L. Burman*#, Leslie S. Itsara*§#, Ernst-Bernhard Kayser†, Wichit Suthammarak†1,

Adrienne M. Wang^, Matt Kaeberlein^, Margaret M. Sedensky†, Philip G. Morgan†, Leo J.

Pallanck*

*Department of Genome Sciences, University of Washington, Seattle, WA 98195, and §Molecular and

Cellular Biology Program, University of Washington, Seattle, WA 98195, ^Department of Pathology,

University of Washington, Seattle, WA 98195, and †Center for Developmental Therapeutics, Seattle

Children’s Research Institute, Seattle, WA 98101, 1Current affiliation: the Department of Biochemistry

and Faculty of Medicine Siriraj Hospital at Mahidol University in Bangkok, Thailand.

# Author contributions: J.LB. and L.S.I contributed equally to this work.

© 2014. Published by The Company of Biologists Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

http://dmm.biologists.org/lookup/doi/10.1242/dmm.015321Access the most recent version at DMM Advance Online Articles. Posted 1 August 2014 as doi: 10.1242/dmm.015321

http://dmm.biologists.org/lookup/doi/10.1242/dmm.015321Access the most recent version at First posted online on 1 August 2014 as 10.1242/dmm.015321

2

Running title: A fly model of complex I deficiency

Keywords: Mitochondria; Drosophila; Mitochondrial Disease, Respiratory Chain; Leigh syndrome;

Neurodegeneration

To whom correspondence should be addressed: Leo Pallanck: pallanck@u.washington.edu, University

of Washington, Department of Genome Sciences, W.H. Foege Building S-443, Box 355065, 3720 15th

Ave. NE, Seattle, WA 98195-5065. Office: (206) 616-5997; Fax: (206) 685-7301

Phil Morgan: pgm4@uw.edu Seattle Children's Research Institute, Rm 923, 1900 9th Ave., Seattle, WA

98105. Office: (206) 884-1102; Fax: (206)884-7175

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

3

ABSTRACT: Mutations affecting mitochondrial complex I, a multi-subunit assembly that couples

electron transfer to proton pumping, are the most frequent cause of heritable mitochondrial diseases.

However, the mechanisms by which complex I dysfunction results in disease remain unclear. Here, we

describe a Drosophila model of complex I deficiency caused by a homoplasmic mutation in the

mitochondrial-encoded NADH dehydrogenase subunit 2 (ND2) gene. We show that ND2 mutants exhibit

phenotypes that resemble symptoms of mitochondrial disease, including shortened lifespan, progressive

neurodegeneration, diminished neural mitochondrial membrane potential, and lower levels of neural ATP.

Our biochemical studies of ND2 mutants reveal that complex I is unable to efficiently couple electron

transfer to proton pumping. Thus, our study provides evidence that the ND2 subunit participates directly

in the proton pumping mechanism of complex I. Together, our findings support the model that diminished

respiratory chain activity, and consequent energy deficiency, are responsible for the pathogenesis of

complex I-associated neurodegeneration.

BACKGROUND: Mitochondria perform numerous vital cellular functions, including producing the

majority of ATP through oxidative phosphorylation via mitochondrial respiratory chain complexes.

NADH ubiquinone oxidoreductase, or complex I, is the largest complex of the mitochondrial respiratory

chain. It consists of ~45 subunits, including 7 that are encoded by the mitochondrial genome: (ND1-6 and

ND4L) (Efremov et al., 2010). All mitochondrial DNA (mtDNA) encoded subunits of complex I span the

mitochondrial inner membrane, where a subset of them may function in proton pumping given their

homology to characterized bacterial complex I subunits (Amarneh and Vik, 2003; Dieteren et al., 2008;

Efremov et al., 2010; Roberts and Hirst, 2012).

Mutations in genes encoding complex I subunits are a common cause of early-onset

mitochondrial diseases such as Leigh syndrome and mitochondrial myopathy, encephalomyopathy, lactic

acidosis, and stroke-like symptoms (MELAS) (Triepels et al., 2001). These diseases are highly

debilitating, and are typically characterized by progressive neurodegeneration, seizures and shortened

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

4

lifespan (Schon and Manfredi, 2003). However, the molecular mechanisms underlying the pathogenesis

of diseases associated with complex I deficiency remain unclear (Torraco et al., 2009).

Recently, Xu and colleagues developed a novel technology for creating targeted mtDNA

mutations in Drosophila melanogaster, and used this technology to generate strains that bear homoplasmic

mtDNA mutations (Xu et al., 2008). Two of the strains that were generated using this approach harbor

mutations in the ND2 gene (Xu et al., 2008). Because mutations in human ND2 have been demonstrated

to cause Leigh syndrome (Ugalde et al., 2007), Leber’s hereditary optic neuropathy (Brown et al., 1992),

and exercise intolerance (Schwartz and Vissing, 2002), we hypothesized that Drosophila ND2 mutants

might serve as an animal model of complex I disease. Here, we show that ND2 mutants exhibit a variety

of phenotypes that parallel symptoms of complex I deficiency in humans, including stress-induced

seizures, progressive neurodegeneration and shortened lifespan. Moreover, our biochemical studies of

ND2 mutants reveal that complex I inefficiently couples electron transfer to proton pumping, resulting in

decreased mitochondrial membrane potential and diminished energy production. Our findings provide the

first evidence that the ND2 subunit participates in the proton pumping activity of complex I, and suggest

that the symptoms of mitochondrial complex I deficiency derive from an energy deficit.

RESULTS: Drosophila ND2 mutants– In previous work, Xu and colleagues expressed a

mitochondrially targeted restriction enzyme in the germline of Drosophila females to select for surviving

offspring that bear a mtDNA mutation in the corresponding restriction site. Two of the strains that were

generated using this approach harbor mutations in the ND2 gene. One of these two mutants, mt: ND2ins1

(ND2ins1), bears a three nucleotide insertion in the ND2 coding sequence and the other, mt: ND2del1

(ND2del1), bears a nine nucleotide deletion in the ND2 coding sequence (Xu et al., 2008). The ND2ins1

mutation adds a serine at position 189, while the ND2del1 mutation removes three amino acids at positions

186-188 of the ND2 protein (Fig. S1). These mutations reside in evolutionarily conserved sequences (Fig.

S1), thus raising the possibility that flies bearing these mutations might serve as models of mitochondrial

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

5

disease. Because the phenotypes of these fly strains were not characterized in previous work, we

proceeded to test this hypothesis.

Behavioral analyses of ND2 mutants– Mitochondrial encephalopathies are typically characterized

by neurological symptoms such as seizures (Pieczenik and Neustadt, 2007). Similarly, Drosophila strains

with mutations affecting mitochondrial function often show an analogous seizure-like paralytic phenotype

caused by mechanical-stress, termed “bang sensitivity” (Celotto et al., 2011; Fergestad et al., 2006;

Ganetzky and Wu, 1982). To determine if ND2 mutants exhibited bang sensitivity we subjected control,

ND2del1 and ND2ins1 mutants to mechanical stress, and measured the time it took for them to recover from

paralysis. While ND2ins1 mutants were unaffected by mechanical stress, ND2del1 mutants displayed bang

sensitivity that progressively worsened with age (Fig. 1A; Movie S1). Our experiments also revealed that

ND2del1 mutant females displayed substantial bang-sensitive paralysis by 10-days of age, whereas ND2del1

mutant males did not exhibit significant bang-sensitivity until 26-days of age (Fig. 1A). Because ND2ins1

mutants lacked a bang sensitive phenotype, only ND2del1 flies (hereafter referred to as ND2 mutants) were

used in our subsequent experiments.

The bang sensitive phenotype of ND2 mutants may be a consequence of reduced ND2 activity, or

the acquisition of a novel activity. To distinguish between these possibilities, we tested whether restoring

NADH dehydrogenase activity would rescue the bang sensitive phenotype of ND2 mutants. Previous

studies have shown that ectopic expression of the yeast NADH dehydrogenase (Ndi1), which is composed

of a single subunit, can rescue complex I deficiency in both mammalian cells and Drosophila (Cho et al.,

2012; Seo et al., 1998). Thus, we tested whether expression of Ndi1 could rescue the bang sensitivity of

ND2 mutants. Ndi1 expression partially suppressed the bang sensitive phenotype of ND2 mutants,

demonstrating that the ND2del1 mutation results in a loss of ND2 activity (Fig. S2A).

Mitochondrial diseases are often associated with dramatically shortened lifespan (Distelmaier et

al., 2009; Ugalde et al., 2007). To determine if this was true for ND2 mutants we compared the lifespan of

ND2 mutants and controls. We found that the median lifespan of ND2 mutants was reduced compared to

wild-type controls, and that female ND2 mutants were shorter lived than male ND2 mutants (Fig. 1B).

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

6

Because female ND2 mutants exhibited stronger phenotypes than males, only females were used in our

subsequent experiments.

Another common symptom of mitochondrial disease is heat intolerance (Pieczenik and Neustadt,

2007). To test whether ND2 mutants were sensitive to heat-induced paralysis we exposed ND2 mutants

and controls to a temperature of 39oC, a condition previously shown to result in premature heat-induced

paralysis in mitochondrial mutants (Ganetzky and Wu, 1982). We found that ND2 mutants fully paralyzed

following exposure to 39oC for 6 minutes, and required a prolonged recovery period after being returned

to room temperature. In contrast, age-matched controls failed to undergo heat-induced paralysis under

these conditions (Fig. 1C).

Mitochondrial diseases preferentially affect tissues with high energetic demands, such as muscle

and brain (Chinnery, 2000). To test for impaired function in muscle and/or neural tissue, we measured the

flight performance of young ND2 mutants using an established flight assay (Benzer, 1973; Greene et al.,

2003). Young ND2 mutants displayed significantly reduced flying ability compared to age-matched

controls (Fig. 1D). In total these experiments support the hypothesis that the ND2del1 mutation affects

energy demanding muscle and/or brain tissue(s), mirroring deficits frequently seen in human

mitochondrial disease.

Mitochondrial impairment has been shown to sensitize Drosophila to hypercarbia- (Whelan et

al., 2010) and hypoxia-induced paralysis (Feala et al., 2007), so we tested whether ND2 mutants were also

sensitive to these conditions. To induce paralysis we exposed young and middle-aged ND2 mutants and

age-matched controls to carbon dioxide or nitrogen, respectively. We then measured the time required for

flies to recover from paralysis upon return to ambient conditions. Both young and middle-aged ND2

mutants took longer to recover from exposure to hypercarbic and hypoxic conditions than controls (Fig.

1E, F). Conversely, hyperoxic conditions have been shown to shorten wild type Drosophila lifespan, and

mitochondrial impairment enhances this effect (Whelan et al., 2010). To determine if ND2 mutants were

sensitive to hyperoxic conditions we housed young ND2 mutants and controls in 100% oxygen and

monitored their lifespan. As expected ND2 mutants and controls both exhibited a dramatically shortened

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

7

lifespan under these conditions, but the ND2 mutants had a significantly shorter lifespan relative to

controls (Fig. S2B).

Tissue integrity of ND2 mutants– A hallmark of mitochondrial disease is the progressive

degeneration of muscle and/or neural tissue(s) (Finsterer, 2008). We therefore analyzed muscle and brain

integrity by hematoxylin and eosin staining of tissue from middle-aged and old ND2 mutants and

controls. These analyses revealed no abnormalities in the flight muscles of middle-aged or old ND2

mutants or in the brains of middle-aged ND2 mutants (Fig. 2A, B). However, neurodegenerative vacuoles

observed in the brains of old ND2 mutants were both significantly larger and more numerous than those

seen in age-matched controls (Fig. 2B, C, D), indicating that the mutants have a progressive

neurodegenerative phenotype.

To test whether ND2 mutants have subtle defects in muscle integrity that were not evident from

our histological analyses, we used confocal microscopy to compare cytoskeletal architecture,

mitochondrial morphology, and apoptotic cell death in the indirect flight muscles of ND2 mutants and

controls by staining for actin, cytochrome C and cleaved caspase 3, respectively. We also included parkin

null animals in these studies as a positive control because they have previously been shown to exhibit

apoptotic degeneration of indirect flight muscles, including disruption of actin organization, and altered

mitochondrial morphology (Greene et al., 2003; Klein et al., 2014). We did not detect significant

differences in actin organization, mitochondrial size or cleaved caspase 3 staining between ND2 mutants

and aged-matched controls (Fig. S3), thus providing further evidence that the ND2del1 mutation does not

affect the musculature. However, in agreement with previous work, we observed disruption of the actin

cytoskeleton, enlarged mitochondria, and cleaved caspase 3 staining in parkin null flight muscle

preparations (Fig. S3).

Functional analyses of mitochondria isolated from ND2 mutants– To explore the biochemical

mechanisms underlying the ND2 mutant phenotypes we isolated mitochondria from middle-aged and old

ND2 mutants and controls, and used complex I–specific substrates to measure mitochondrial respiration.

We performed these experiments using both limiting and saturating amounts of ADP to measure state 3,

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

8

and maximal state 3, complex I-dependent respiratory rates, respectively (Fig. S4). As previously

reported, state 3 and maximal state 3 rates decreased with age in control flies (Ferguson et al., 2005), and

ND2 mutants displayed similar kinetics of decreased activity with age (Fig. 3A). Mitochondria isolated

from either middle-aged or old ND2 mutants showed unimpaired state 3 respiration, but maximal state 3

values were significantly decreased in ND2 mutants relative to age-matched controls (Fig. 3A and Table

1). The ND2del1 mutation thus causes a complex I–dependent respiratory defect specifically under

maximally demanding conditions. To ensure there was no functional variability between mitochondrial

preparations caused by our isolation method, we measured complex II- and complex IV-dependent

respiration in ND2 mutants and controls. We found no significant difference in complex II-dependent

respiration, complex II-dependent ADP/O ratios or complex IV-dependent respiration between ND2

mutant and control mitochondria (Table 1). These results confirm the integrity of our mitochondrial

preparations, and the respiratory comparisons made between ND2 mutants and controls (Fig 3A).

Oxidative phosphorylation coupling– The respiratory defect of ND2 mutants offered the

opportunity to investigate the functional role of the ND2 subunit in complex I activity. To test whether

ND2 plays a role in the proton pumping activity of complex I we measured the linkage of electron

transport to ATP production by determining the ADP/O ratio. Since electron transport is mechanistically

coupled to ADP phosphorylation via the proton circuit across the inner membrane, any defect that reduces

complex I's efficiency to pump protons should lead to a decrease in ADP/O. We found a significant

decrease in complex I–dependent ADP/O ratios in both middle-aged and old ND2 mutants (Fig. 3B and

Table 1). This decrease does not appear to be explained by a change in the efficiency of ATP production

by complex V, nor a proton leak in the mitochondrial membrane, as ND2 mutant ADP/O ratios were

normal when respiration was dependent on complex II (Table 1). Moreover, the mitochondrial respiratory

rate following depletion of ADP (state 4), an indicator of proton leak across the mitochondrial inner

membrane, was unaffected by the ND2del1 mutation (Fig. 3A and Table 1). Thus, our results demonstrate

that electron flow through complex I is inefficiently coupled to proton pumping in ND2 mutants,

indicating that the ND2 subunit functions in the proton pumping mechanism of complex I.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

9

Complex I integrity– While the decrease in ADP/O in ND2 mutants indicates that ND2 is required

for efficient coupling of electron transfer to proton pumping, it does not explain the decrease of maximal

complex I-dependant state 3 respiration in ND2 mutants. One possible explanation for this decrease is a

reduction in complex I abundance. To test this model we examined mitochondria isolated from ND2

mutants and controls on blue native gels. This analysis revealed that the abundance of fully assembled

complex I was decreased in ND2 mutants relative to controls, without an accompanying decrease in

complex V abundance (Fig. 3C). Western blot analysis also revealed a decrease in the abundance of the

complex I subunit NDUFS3 in ND2 mutants, but little to no change in the abundance of the β-subunit of

complex V relative to age-matched controls (Fig. 3D, E and Fig. S5). These findings suggest that ND2 is

also required for the structural integrity of complex I.

Mitochondrial membrane potential and ATP production– The decreased maximal complex I-

dependent state 3 activity and ADP/O ratio of ND2 mutants suggested that the mutants might have

diminished mitochondrial membrane potential and consequently reduced ATP synthesis. To test these

possibilities, we measured mitochondrial membrane potential in dissociated neurons from old ND2

mutants and controls, and we compared ATP levels in the thoraces and heads of old ND2 mutants and

controls. We detected a decrease in the fraction of cells with polarized mitochondria in the brains of old

ND2 mutants relative to controls, and reduced ATP abundance in ND2 mutants that was specific to heads

(Fig. 4A-C). These findings suggest that the ND2del1 mutation affects the proton pumping activity of

complex I, which reduces the proton gradient available for ATP generation, leading to energy deficiency

in the brains of ND2 mutants.

Reactive oxygen species damage– Mitochondrial complex I deficiency is believed to trigger the

formation of reactive oxygen species (ROS), and increased ROS production is a widely believed cause of

the symptoms accompanying complex I-associated diseases (Lin et al., 2012). To test whether the

phenotypes of ND2 mutants could derive from excessive ROS production, we measured the abundance of

a common byproduct of oxidative stress: 4-hydroxynonenal (4-HNE) containing protein adducts. 4-HNE

protein adducts are elevated under conditions of increased oxidative stress (Poli and Schaur, 2000), and

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

10

their abundance can be easily measured using an antiserum that detects 4-HNE adducts, as has been

previously shown in Drosophila (Sun et al., 2014; Tsai et al., 1998). Using this antiserum on a western

blot of protein extracts from ND2 mutants and age-matched controls revealed no significant alteration in

the abundance of 4-HNE protein adducts between these genotypes in young, middle-aged or old animals

(Fig. S6). These data argue that the phenotypes of ND2 mutants are not a consequence of increased ROS-

mediated damage.

DISCUSSION: A number of hypotheses have been proposed to explain the pathological consequences of

complex I deficiency. Amongst the most enduring of these models are that complex I deficiency results in

the elevated production of toxic ROS and/or an energy deficit (Torraco et al. 2009; Lin et al. 2012). To

address these and other possible models of pathogenesis, we studied a Drosophila strain with a mutation

in the mitochondrial-encoded respiratory chain complex I subunit ND2. We show that ND2 mutants

exhibit many of the behavioral and histological characteristics of mitochondrial disease, suggesting that

ND2 mutants represent a valid model of this spectrum of disorders. Moreover, our work shows that ND2

mutants have diminished ATP abundance, mitochondrial membrane potential, complex I abundance, and

complex I activity. However, we detected no evidence for increased ROS in ND2 mutants. These findings

support a model in which the pathogenesis accompanying complex I deficiency is caused by diminished

energy production.

Several of our findings contrast with previously published work using other models of complex I

deficiency. For example, studies of Ndufs4 knockout mice, and transgenic mice harboring a pathogenic

point mutation in ND6 revealed decreased maximal complex I activity, and increased levels of ROS-

mediated damage (Quintana et al., 2010; Lin et al., 2012). These studies suggest that ROS is a contributor

to the pathogenesis of complex I deficiency. However, ROS levels are thought to increase as a result of

decreased basal complex I activity (Turrens, 2003). Therefore, it is possible that ND2 mutants do not

show changes in ROS-mediated damage due to their normal basal complex I activity, while the Ndufs4

and ND6 mouse models may exhibit increased ROS-mediated damage as a result of decreased basal

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

11

complex I activity. Further analyses of the basal activity of complex I in these mouse models of complex I

deficiency will be required to test this hypothesis. Another discrepancy of our work concerns the mild

phenotypes of ND2 mutants relative to other fly and mouse models of complex I deficiency. We believe

this discrepancy is best explained by differences in the severity of the mutations analyzed in each

respective study. The ND2del1 allele is an in-frame deletion predicted to make a full length protein lacking

3 amino acids. This mutation may therefore represent a hypomorphic allele of ND2. By contrast, many

animal models of complex I deficiency, with phenotypes more severe than those of ND2 mutants, harbor

severe loss-of-function or null alleles (Cho et al., 2012; Kruse et al., 2008; Zhang et al., 2013). In further

support of this model, the magnitudes of the decreases in complex I activity and abundance in ND2del1

mutants are similar to those seen in Caenorhabditis elegans upon partial inactivation of complex I

subunits (Falk et al., 2009). The relatively mild phenotypes of ND2 mutants potentially make them well

suited to genetic modifier screens aimed at the identification of novel genetic factors that influence the

phenotypes associated with complex I deficiency.

In addition to exploring the pathogenic mechanisms underlying complex I deficiency, our work

provides insight into the role of ND2 in complex I function. While the ND2 subunit has been

hypothesized to pump protons based on its membrane localization and sequence homology to a family of

cation antiporters (Efremov et al., 2010), only a study of the bacterial ND2 homolog: NuoN has provided

support for this model (Amarneh and Vik, 2003). Our finding that the ND2del1 mutation results in

uncoupling of proton pumping from electron transfer (i.e. decreased oxidative phosphorylation efficiency

(ADP/O) with normal basal complex I-mediated electron flux) strongly supports a direct role for ND2 in

the proton pumping mechanism of eukaryotic complex I.

While our work advances the understanding of the consequences of ND2 deficiency, and the

functional role of the ND2 subunit in eukaryotes, it also raises several questions. One important question

raised by our work concerns the mechanisms underlying the progressive nature of the ND2 mutant

phenotypes. One possible explanation of the progressive phenotypes of ND2 mutants is that energy

deficiency does not occur until a critical threshold of complex I activity is crossed. As previously reported

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

12

(Ferguson et al., 2005), and confirmed in our current study, complex I activity naturally declines with age

in Drosophila. This decline in complex I activity may not reach the point where it results in an energy

deficiency over the natural lifespan of Drosophila. However, the ND2 mutation exacerbates this natural

decline in complex I activity, and thus may account for the progressive nature of the ND2 mutant

phenotypes.

Another important question raised by our study is why a systemic deficit in complex I function

results in selective neurodegeneration. One possible explanation may be that neurons are more sensitive

to loss of mitochondrial function because they rely primarily on oxidative phosphorylation for ATP, while

other tissues, such as muscle, are capable of using both oxidative phosphorylation and glycolysis (Chan et

al., 2009). While neural selective vulnerability is not uncommon in mitochondrial-associated diseases

including Leigh Syndrome, Leber’s hereditary optic neuropathy and Alpers’ syndrome (Bianchi et al.,

2011; Koopman et al., 2012), further studies will be required to understand the mechanisms underlying

the progressive tissue-specific nature of mitochondrial disease.

In summary, our work suggests an important role for ND2 in the proton pumping mechanism of

complex I, and provides a genetically tractable animal model of complex I deficiency. This model should

prove a valuable tool in which to further explore the pathological mechanisms underlying human

mitochondrial disease.

MATERIALS AND METHODS: Drosophila strains and maintenance– All Drosophila

strains were maintained on standard cornmeal/molasses medium at 25°C with a 12-hour light-

dark cycle. The ND2del1 and ND2ins1 stock was obtained from the laboratory of Dr. Patrick

O'Farrell (University of California, San Francisco) (Xu et al., 2008). The isogenic w1118 stock was

obtained from the Bloomington Drosophila Stock Center at Indiana University. To control for

differences in nuclear genetic background, we outcrossed ND2 mutants to the w1118 strain. F1

offspring derived from crossing ND2 mutant females to w1118 males were used as the

experimental group, given that they inherit mtDNA from the ND2 mutant strain; F1 offspring

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

13

derived from crossing ND2 mutant males to w1118 females were used as the control group, given

that they inherit mtDNA from the w1118 strain. The homoplasmic status of the ND2del1 mutation

from the outcrossed ND2 mutant strain was reconfirmed by PCR and restriction digest analysis

(data not shown). Parkin null animals harbored the previously described park25 null allele

(Greene et al., 2003), and were composed of the genotype CyO/IF; park25/park25. Previously

described, transgenic flies capable of expressing yeast Ndi1 were obtained from Dr. David

Walker (Cho et al., 2012).

Mechanical stress-induced paralysis– Flies were assayed for bang sensitivity using a

modification of a previously published protocol (Ganetzky and Wu, 1982). Briefly, flies were

vortexed for 10 seconds in inverted glass vials containing cotton stoppers, and the time required

for each individual animal to right itself was recorded (Movie S1).

Heat induced paralysis– 14-day-old flies were assayed for heat-induced paralysis by

placing groups of 5-10 animals into pre-warmed vials maintained at 39oC. The time for the flies

to become completely paralyzed was recorded. After exposure to 39oC for 6 minutes the animals

were then placed in new room temperature vials (~20oC), and the recovery time of ND2 mutants

from paralysis recorded.

Lifespan– Lifespan assays using vials containing populations of 12-20 flies were

transferred to new vials every two days throughout the assay period. At least 60 flies were

monitored for each experiment. For sex specific lifespan assays, ND2 mutant males and females

or gender- and age-matched control flies were separated within 1 day of eclosion, and the assay

conducted as described above.

Flight– Flight assays were performed as previously described (Benzer, 1973; Greene et

al., 2003). Briefly, an acetate sheet was divided into five parts, coated with vacuum grease, and

inserted into a 1 L graduated cylinder. 7-day-old flies were gently tapped into a funnel at the top

of the cylinder, and became stuck to the vacuum grease where they alighted. The acetate sheet

was removed, and the number of flies in each section was counted. Flies that alighted in a higher

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

14

section of the acetate sheet received a correspondingly higher value in scoring. The number of

flies alighting in the top section was summed and multiplied by four, the number of flies alighting

in the second highest section was summed and multiplied by three, and so on. Finally, the

weighted sum for the entire group of flies was determined, and normalized to four times the total

number of flies used in the assay (the maximum possible score), and this normalized value was

reported as the flight index. Flies that alighted at the top of the cylinder received a flight index

score of 1, whereas flies that fell to the bottom of the cylinder received a flight index score of 0.

At least six groups of 11-20 flies per genotype were tested.

Sensitivity to CO2 exposure- Young (3-day-old) or middle-aged (16-day-old) control or

ND2 mutants were exposed to 100% CO2 (Praxair) on a gas permeable mesh pad for 1 minute,

rendering them completely immobile. The gas was then shut off, leaving the flies exposed only to

room air, and the time to recover from paralysis measured.

Hypoxia Sensitivity– Three separate sets of 10-15 age-matched young (3-day-old) and

middle-aged (16-day-old) flies were placed into 70 mL chambers and 100% N2 (Praxair) certified

at less than 10 ppm O2 was flowed through the chamber at 350 mL/min for 1.5 min to induce

paralysis. After 30 seconds, flies were exposed to room air at the same rate, and the time from the

introduction of room air to recovery from paralysis recorded.

Hyperoxia Sensitivity- Female flies were collected within 48 hours of eclosion, and

placed into 4 vials of 25 flies per genotype. Vials were then placed into a temperature controlled

hyperoxia chamber and maintained in a 100% O2 environment at 25oC. Flies were transferred to

new vials every 24 hours and scored for survival.

Confocal microscopy- Indirect flight muscles were dissected from 40-day-old control or

ND2 mutant animals, or from 7 day-old parkin null animals in phosphate buffered saline (PBS).

Dissected muscles were placed in 0.3% Triton X-100 (Sigma), 4% paraformaldehyde (Ted Pella

Inc.) in PBS, and rotated at room temperature for 40 minutes. Muscles were then washed twice in

PBS and placed in blocking buffer (0.3% Triton X-100, 5% Normal Goat Serum (Fisher

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

15

Scientific) in PBS) for 45 minutes. Muscles were then incubated with a mouse primary antibody

raised against cytochrome C (BD Labs # 556433) (1/1000), and a rabbit primary antibody raised

against cleaved caspase 3 (Cell Signaling #9661) (1/400) for 2 days at 4oC in 0.3% Triton X-100

and 2.5% normal goat serum in PBS. Tissue was then washed twice in 0.3% Triton X-100 in

PBS, for 20 minutes each, at room temperature. The tissue was then placed in secondary

antibodies: mouse-Alexa 647 and rabbit-Alexa-488 (Molecular Probes), and incubated for 1 hour

rotating at room temperature. 20 minutes prior to removal from secondary antibodies, Phalloidin-

568 (Life Technologies) was added at a concentration of 1/400. Tissue was removed and washed

two times for 10 minutes each in 0.3% Triton X-100 in PBS. The muscles were then placed in

PBS containing DAPI (Sigma-Aldrich) at 2 μg/ml for 5 minutes. Cells were washed in PBS for

10 minutes, and mounted between two glass slides with Fluoromount (Sigma-Aldrich), and

imaged using an Olympus FV-1000 with a 60x lens and a 4X digital zoom.

Image quantification- Images of muscles stained with cytochrome C were opened in

Image J, converted to 8-bit, thresholded, watershed, and analyzed using the analyze particle tool.

The average perimeter of distinct mitochondria was used as an indicator of mitochondrial

morphology and mass. For cleaved caspase 3 staining the number of cleaved caspase 3 positive

myocytes, as indicated by DAPI positive/cytochrome C/cleaved caspase 3 positive staining, were

manually counted.

Tissue preparation and sectioning– Middle-aged (14-day-old) or old (42-day-old) flies

were anesthetized, mounted in fixing collars and placed in Carnoy's fixation solution (10% acetic

acid; 30% chloroform; 60% absolute ethanol) for 3.5 hours. Fixed flies were then placed in 95%

ethanol twice for 30 minutes each, in 100% ethanol for 45 minutes, then in methyl benzoate

overnight. Flies were then incubated in a 1:1 mixture of methyl benzoate:paraffin for 1 hour at

60°C, transferred to 100% liquid paraffin and incubated at 60°C for 30 minutes, then transferred

to fresh 100% liquid paraffin four times for 15-30 minutes each. The flies were next transferred to

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

16

100% liquid paraffin in a plastic mold, placed at 60°C for 15 minutes, and the paraffin was

allowed to harden at room temperature. Five μm thick coronal histological sections were cut on a

Shandon Finesse 325 Microtome, and processed for hematoxylin and eosin staining. Images were

collected on a Nikon Optiphot-2 using a 20x objective. Vacuole size was quantified in Image J

(Abramoff, 2004) by converting pixel values into micrometers (0.439 μm/pixel). Only vacuoles

present in at least two consecutive sections were quantified to avoid potential artifacts derived

from tissue sectioning.

Mitochondrial Preparations– Mitochondria were obtained from 0.25-0.5 g of 14-day-old

or 30 to 35-day-old flies. Flies were chilled on ice, and then homogenized using a chilled

Potter/Elvehjem homogenizer with 10 manual strokes in 20 mL of homogenization buffer (200

mM mannitol, 70 mM sucrose, 5 mM MOPS, 2 mM EDTA; 0.4% defatted BSA; pH 7.4),

avoiding shearing. The homogenate was then filtered through cotton gauze, and spun at 300 x g

for 4 minutes at 4°C. The supernatant was then filtered through gauze and spun at 10,000 x g for

10 minutes at 4°C. The pellet was resuspended in 10 ml of resuspension buffer (200 mM

mannitol, 70 mM sucrose, 5 mM MOPS, 2 mM EDTA; pH 7.4), and spun at 10,000 x g for 10

minutes at 4°C. The mitochondrial pellet was then resuspended in 10-50 µl of final buffer (200

mM mannitol, 70 mM sucrose, 5 mM MOPS; pH 7.4), to a final concentration of ~50 μg

protein/μL. 150-300 μg of mitochondria was used for oxidative phosphorylation assays, and 250

μg for blue native gel analysis. All reagents were obtained from Sigma (St. Louis, MO).

Oxidative Phosphorylation Assays– Oxygen consumption of isolated mitochondria was

monitored using a Clark-type electrode (Oxytherm with analysis software Oxyg32, Hansatech

Instruments, Pentney, Norfolk, Great Britain) by an adaptation of (Kayser et al., 2001). Briefly,

150-300 μg of purified mitochondria were stirred in 500 μL of assay buffer (100 mM KCl, 50

mM MOPS, 1 mM EGTA, 5mM potassium phosphate, 1 mg/ml defatted BSA; pH 7.4) at 30°C.

Ten mM malate and 20 mM pyruvate were added as complex I specific electron donor substrates,

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

17

20 mM succinate was added as a Complex II specific electron donor substrate or 25 mM TMPD

and 250 mM ascorbate pH 7.0 were added as Complex IV specific electron donor substrates. A

low dose of ADP (90.7 nmol) was added to determine state 3 mitochondrial respiration rates, and

following depletion of ADP, ensuing state 4 respiration rates. A saturating dose of ADP (1 μmol)

was used to assess maximal complex I-dependent electron transport capacity under

phosphorylating conditions (maximal state 3). Saturating amounts of TMPD/ascorbate were used

to measure Complex IV activity. ADP/O was defined as the number of ADP molecules

phosphorylated per oxygen atom reduced to water, and ADP/O was calculated by dividing the

amount of ADP added (90.7 nmol) by the amount of oxygen consumed between the time of ADP

addition and the onset of state 4 (Figure S4).

ATP determination– Heads or thoraces were sectioned from five 35-day-old flies and

homogenized in 6 M guanidine HCl (Sigma); 10 mM TRIS (Sigma) pH 7.3 as previously

described (Xu et al., 2008). ATP content was measured using an ATP determination kit

(Molecular Probes, Eugene, OR). Protein abundance was measured using a Bradford protein

assay, and ATP abundance was normalized to total protein abundance as previously described

(Xu et al., 2008).

Mitochondrial Membrane Potential– Measurement of neural mitochondrial membrane

potential from 30 to 35-day-old flies was conducted as previously described (Burman et al.,

2012), except that dissections of animals were performed in supplemented DME/Ham’s F-12

High Glucose media lacking phenol red (Sigma), and all incubation steps were carried out at

25°C. In addition, 10 nM TMRE was utilized throughout the protocol. The % of cells harboring

polarized mitochondria was determined using FloJo software (Treestar Inc., Ashland, OR), and

defined as the % of cells exhibiting relative TMRE fluorescence above 103 arbitrary fluorescence

units, as previously experimentally determined (Burman et al., 2012). All flow cytometry

measurements were performed on a Becton-Dickinson LSR II flow cytometer.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

18

Blue Native Gel Electrophoresis and In-gel Activity Staining– Assays were carried out

using mitochondrial preparations from 14-day-old flies as previously described (Suthammarak et

al., 2009; Wittig et al., 2006). All reagents were purchased from Sigma.

Western blots– 14-day-old whole flies were flash frozen in liquid nitrogen and

homogenized with a pestle in 2x lysis buffer (50 mM TRIS pH 8.0; 300 mM NaCl; 2mM EDTA;

1% SDS; 2% Triton X-100) with protease inhibitor cocktail (Sigma). Homogenates were

centrifuged at 21,000 x g for 5 minutes, and the supernatant subjected to Western blot processing

and analysis. Blots were labeled with monoclonal antibodies to actin (Millipore #MAB1501)

diluted 1/25,000, the β-subunit of ATP Synthase (Invitrogen #A21351) diluted 1/1,500, the

NDUFS3 subunit of complex I (Abcam #17D95) diluted 1/800, and an anti-HNE fluorophore

antibody (Calbiochem 393206) diluted 1/2500. This antibody detects a stable, crosslinking

product between two lysyl-residues and 4-HNE: 2:1 Nα-acetyllysine-HNE fluorophore. The

antigen is taken as a proxy for accumulated 4-HNE damage, or accumulated ROS damage in

general (Sun et al., 2014; Tsai et al., 1998). All reagents were purchased from Sigma unless

otherwise noted. Western blot band intensities were quantified using Image J gel analysis tools,

and the intensities were expressed as the % of control values following normalization to actin

(Abramoff, 2004). Complete western blots are shown in Figure S5 as a demonstration of samples

run in parallel.

Statistics– Unless otherwise stated statistical significance tests were calculated using an

unpaired two-tailed Student’s t-test.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

19

Acknowledgements – We thank Dr. Patrick H. O'Farrell (University of California, San

Francisco) for providing the Drosophila ND2 mutant stocks (Xu et al., 2008); Dr. Ying-Tzang

Tien (Department of Pathology, University of Washington) for histological staining of Drosophila

tissue sections; Dr. Renata de Lima Sales Goncalves (Buck Institiute of Aging, Novato, USA),

Laurie Andrews and Selina Yu (University of Washington) for technical assistance; Tarannum

and Afshan Dhillon, Stephanie N. Yu and Ryan Hau (University of Washington) for maintaining

Drosophila stocks. We thank Dr. Evelyn Vincow for critical reading of the manuscript.

Competing interests statement: The authors declare no competing financial interests.

Author Contributions: J.L.B., L.S.I., A.M.W., M.K, E.B.K, P.G.M., M.M.S., and L.J.P.

designed research; J.L.B., L.S.I., A.M.W, W.S., and E.B.K. performed research; J.L.B., L.S.I.,

W.S., E.B.K., P.G.M., M.M.S., and L.J.P. analyzed data; J.L.B., L.S.I., E.B.K., M.M.S., P.G.M.

and L.J.P. wrote the paper. These studies were also supported by the University of Washington

School of Medicine and Department of Pathology (M.K.) and NIH training grant T32AG000057

(A.M.W).

Funding – This work was supported by a Parkinson Society Canada Research Fellowship

(to J.L.B.), and a subsequent Canadian Institute of Health Research Fellowship (to J.L.B),

National Institute of Neurological Disorders and Stroke Fellowship F31NS071857 (to L.S.I.),

National Institute of Health grant 1RO1GM086394 (to L.P.) and a grant from the Muscular

Dystrophy Association (to L.P.). LSI and AMW were supported on the Genetic Approaches to

Aging Training Grant T32AG00005 (to Peter Rabinovitch). WS was supported by a NW

Mitochondrial Guild Postdoctoral Fellowship.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

20

REFERENCES

Abramoff, M. D., Magalhaes, P.J., Ram, S.J. (2004). Image Processing with ImageJ. Biophotonics International 11, 36-42. Amarneh, B. and Vik, S. B. (2003). Mutagenesis of subunit N of the Escherichia coli complex I. Identification of the initiation codon and the sensitivity of mutants to decylubiquinone. Biochemistry 42, 4800-8. Benzer, S. (1973). Genetic dissection of behavior. Sci Am 229, 24-37. Bianchi, M., Rizza, T., Verrigni, D., Martinelli, D., Tozzi, G., Torraco, A., Piemonte, F., Dionisi-Vici, C., Nobili, V., Francalanci, P. et al. (2011). Novel large-range mitochondrial DNA deletions and fatal multisystemic disorder with prominent hepatopathy. Biochem Biophys Res Commun 415, 300-4. Brown, M. D., Voljavec, A. S., Lott, M. T., Torroni, A., Yang, C. C. and Wallace, D. C. (1992). Mitochondrial DNA complex I and III mutations associated with Leber's hereditary optic neuropathy. Genetics 130, 163-73. Burman, J. L., Yu, S., Poole, A. C., Decal, R. B. and Pallanck, L. (2012). Analysis of neural subtypes reveals selective mitochondrial dysfunction in dopaminergic neurons from parkin mutants. Proc Natl Acad Sci U S A 109, 10438-43. Celotto, A. M., Chiu, W. K., Van Voorhies, W. and Palladino, M. J. (2011). Modes of metabolic compensation during mitochondrial disease using the Drosophila model of ATP6 dysfunction. PLoS One 6, e25823. Chan, C. S., Gertler, T. S. and Surmeier, D. J. (2009). Calcium homeostasis, selective vulnerability and Parkinson's disease. Trends Neurosci 32, 249-56. Chinnery, P. F. (2000). Mitochondrial Disorders Overview. GeneReviews June. Cho, J., Hur, J. H., Graniel, J., Benzer, S. and Walker, D. W. (2012). Expression of yeast NDI1 rescues a Drosophila complex I assembly defect. PLoS One 7, e50644. Diaz, F., Thomas, C. K., Garcia, S., Hernandez, D. and Moraes, C. T. (2005). Mice lacking COX10 in skeletal muscle recapitulate the phenotype of progressive mitochondrial myopathies associated with cytochrome c oxidase deficiency. Hum Mol Genet 14, 2737-48. Dieteren, C. E., Willems, P. H., Vogel, R. O., Swarts, H. G., Fransen, J., Roepman, R., Crienen, G., Smeitink, J. A., Nijtmans, L. G. and Koopman, W. J. (2008). Subunits of mitochondrial complex I exist as part of matrix- and membrane-associated subcomplexes in living cells. J Biol Chem 283, 34753-61. Distelmaier, F., Koopman, W. J., van den Heuvel, L. P., Rodenburg, R. J., Mayatepek, E., Willems, P. H. and Smeitink, J. A. (2009). Mitochondrial complex I deficiency: from organelle dysfunction to clinical disease. Brain 132, 833-42. Efremov, R. G., Baradaran, R. and Sazanov, L. A. (2010). The architecture of respiratory complex I. Nature 465, 441-5. Falk, M. J., Rosenjack, J. R., Polyak, E., Suthammarak, W., Chen, Z., Morgan, P. G. and Sedensky, M. M. (2009). Subcomplex Ilambda specifically controls integrated mitochondrial functions in Caenorhabditis elegans. PLoS One 4, e6607. Feala, J. D., Coquin, L., McCulloch, A. D. and Paternostro, G. (2007). Flexibility in energy metabolism supports hypoxia tolerance in Drosophila flight muscle: metabolomic and computational systems analysis. Mol Syst Biol 3, 99. Fergestad, T., Bostwick, B. and Ganetzky, B. (2006). Metabolic disruption in Drosophila bang-sensitive seizure mutants. Genetics 173, 1357-64. Ferguson, M., Mockett, R. J., Shen, Y., Orr, W. C. and Sohal, R. S. (2005). Age-associated decline in mitochondrial respiration and electron transport in Drosophila melanogaster. Biochem J 390, 501-11. Finsterer, J. (2008). Leigh and Leigh-like syndrome in children and adults. Pediatr Neurol 39, 223-35.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

21

Ganetzky, B. and Wu, C. F. (1982). Indirect Suppression Involving Behavioral Mutants with Altered Nerve Excitability in DROSOPHILA MELANOGASTER. Genetics 100, 597-614. Greene, J. C., Whitworth, A. J., Kuo, I., Andrews, L. A., Feany, M. B. and Pallanck, L. J. (2003). Mitochondrial pathology and apoptotic muscle degeneration in Drosophila parkin mutants. Proc Natl Acad Sci U S A 100, 4078-83. Kayser, E. B., Morgan, P. G., Hoppel, C. L. and Sedensky, M. M. (2001). Mitochondrial expression and function of GAS-1 in Caenorhabditis elegans. J Biol Chem 276, 20551-8. Klein, P., Muller-Rischart, A. K., Motori, E., Schonbauer, C., Schnorrer, F., Winklhofer, K. F. and Klein, R. (2014). Ret rescues mitochondrial morphology and muscle degeneration of Drosophila Pink1 mutants. Embo J 33, 779. Koopman, W. J., Distelmaier, F., Smeitink, J. A. and Willems, P. H. (2012). OXPHOS mutations and neurodegeneration. Embo J 9, 9-29. Kruse, S. E., Watt, W. C., Marcinek, D. J., Kapur, R. P., Schenkman, K. A. and Palmiter, R. D. (2008). Mice with mitochondrial complex I deficiency develop a fatal encephalomyopathy. Cell Metab 7, 312-20. Lin, C. S., Sharpley, M. S., Fan, W., Waymire, K. G., Sadun, A. A., Carelli, V., Ross-Cisneros, F. N., Baciu, P., Sung, E., McManus, M. J. et al. (2012). Mouse mtDNA mutant model of Leber hereditary optic neuropathy. Proc Natl Acad Sci U S A 109, 20065-70. Pieczenik, S. R. and Neustadt, J. (2007). Mitochondrial dysfunction and molecular pathways of disease. Exp Mol Pathol 83, 84-92. Poli, G. and Schaur, R. J. (2000). 4-Hydroxynonenal in the pathomechanisms of oxidative stress. IUBMB Life 50, 315-21. Quintana, A., Kruse, S. E., Kapur, R. P., Sanz, E. and Palmiter, R. D. (2010). Complex I deficiency due to loss of Ndufs4 in the brain results in progressive encephalopathy resembling Leigh syndrome. Proc Natl Acad Sci U S A 107, 10996-1001. Roberts, P. G. and Hirst, J. (2012). The deactive form of respiratory complex I from Mammalian mitochondria is a na+/h+ antiporter. J Biol Chem 287, 34743-51. Schon, E. A. and Manfredi, G. (2003). Neuronal degeneration and mitochondrial dysfunction. J Clin Invest 111, 303-12. Schwartz, M. and Vissing, J. (2002). Paternal inheritance of mitochondrial DNA. N Engl J Med 347, 576-80. Seo, B. B., Kitajima-Ihara, T., Chan, E. K., Scheffler, I. E., Matsuno-Yagi, A. and Yagi, T. (1998). Molecular remedy of complex I defects: rotenone-insensitive internal NADH-quinone oxidoreductase of Saccharomyces cerevisiae mitochondria restores the NADH oxidase activity of complex I-deficient mammalian cells. Proc Natl Acad Sci U S A 95, 9167-71. Sun, Y., Yolitz, J., Alberico, T., Sun, X. and Zou, S. (2014). Lifespan extension by cranberry supplementation partially requires SOD2 and is life stage independent. Exp Gerontol 50, 57-63. Suthammarak, W., Yang, Y. Y., Morgan, P. G. and Sedensky, M. M. (2009). Complex I function is defective in complex IV-deficient Caenorhabditis elegans. J Biol Chem 284, 6425-35. Torraco, A., Diaz, F., Vempati, U. D. and Moraes, C. T. (2009). Mouse models of oxidative phosphorylation defects: powerful tools to study the pathobiology of mitochondrial diseases. Biochim Biophys Acta 1793, 171-80. Triepels, R. H., Van Den Heuvel, L. P., Trijbels, J. M. and Smeitink, J. A. (2001). Respiratory chain complex I deficiency. Am J Med Genet 106, 37-45. Tsai, L., Szweda, P. A., Vinogradova, O. and Szweda, L. I. (1998). Structural characterization and immunochemical detection of a fluorophore derived from 4-hydroxy-2-nonenal and lysine. Proc Natl Acad Sci U S A 95, 7975-80.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

22

Turrens, J. F. (2003). Mitochondrial formation of reactive oxygen species. J Physiol 552, 335-44. Ugalde, C., Hinttala, R., Timal, S., Smeets, R., Rodenburg, R. J., Uusimaa, J., van Heuvel, L. P., Nijtmans, L. G., Majamaa, K. and Smeitink, J. A. (2007). Mutated ND2 impairs mitochondrial complex I assembly and leads to Leigh syndrome. Mol Genet Metab 90, 10-4. Whelan, J., Burke, B., Rice, A., Tong, M. and Kuebler, D. (2010). Sensitivity to seizure-like activity in Drosophila following acute hypoxia and hypercapnia. Brain Res 1316, 120-8. Wittig, I., Braun, H. P. and Schagger, H. (2006). Blue native PAGE. Nat Protoc 1, 418-28. Xu, H., DeLuca, S. Z. and O'Farrell, P. H. (2008). Manipulating the metazoan mitochondrial genome with targeted restriction enzymes. Science 321, 575-7. Zhang, K., Li, Z., Jaiswal, M., Bayat, V., Xiong, B., Sandoval, H., Charng, W. L., David, G., Haueter, C., Yamamoto, S. et al. (2013). The C8ORF38 homologue Sicily is a cytosolic chaperone for a mitochondrial complex I subunit. J Cell Biol 200, 807-20.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

23

FIGURE LEGENDS:

Fig. 1. ND2 mutants exhibit behavioral abnormalities and shortened lifespan. (A) ND2 mutants

exhibit stress-induced paralysis. Both male and female ND2del1 and ND2ins1 mutants, and male

and female controls, were assayed for the length of time that flies remained paralyzed following

mechanical stress. Error bars represent standard error of the mean; n=3 independent groups of 6-

12 individual animals. Measurements for ND2ins1 mutants were taken at 2, 10, 26 and 46 days of

age. (B) Lifespan analysis depicting a decrease in the median lifespan of ND2 mutants. The

median lifespan was 37 days for control females, 47 days for control males, 23 days for ND2

mutant females, and 27 days for ND2 mutant males. Error bars represent standard error of the

mean (p<0.0001); n=10 independent groups of 12-20 animals for control and ND2 mutants. (C)

ND2 mutants exhibit heat-induced paralysis. 14-day-old ND2 mutants were incubated at 39oC,

and the time for each individual fly to paralyze recorded in seconds. After 6 minutes flies were

then placed in room temperature vials, and the time to recover from paralysis recorded in

seconds. Controls did not exhibit heat-sensitive paralysis over the time course of the assay (n/a).

Histograms depict the median time to paralyze and recover from paralysis (seconds); (n=3

independent groups of 7-9 animals); p=4.37x10-17 and p=8.7x10-11 for paralysis and recovery,

respectively; ***=p<0.001; Error bars represent standard error of the mean. (D) ND2 mutants

have reduced flight ability. 7-day-old ND2 mutants and controls were tested for their flying

ability by measuring the distance that flies alighted when dispensed into a cylinder (p=0.001).

Histograms represent the average flight index of the indicated genotypes, and error bars represent

the standard error of the mean; n=6 independent groups of 11-20 animals; **=p<0.01. (E) ND2

mutants are hypersensitive to CO2 exposure. Histograms depict the time to recover (seconds) for

control and ND2 mutant flies after hypercarbia induced paralysis in young (3 days) and middle-

aged (16 days of age) animals (n = 2 groups of N = 12 animals; p = 7.0 x 10-5 for 3-day-old

animals; n = 2; N = 11; p = 0.0013 for 16-day-old animals; ***=p ≤ 0.001; **=p<0.01); error

bars represent standard deviations. (F) ND2 mutants are hypersensitive to N2-induced hypoxia.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

24

Histograms depict the time to recover (seconds) for control and ND2 mutant flies after hypoxia

induced paralysis in young (3 days of age) and middle-aged (16 days of age) animals (n=3 groups

of 10-15 animals for each genotype); ***=p=0.0004; *=p=0.02; Error bars represent the standard

error of the mean.

Fig. 2. ND2 mutants exhibit progressive neurodegeneration, but lack muscle pathology. (A)

Thoracic muscle integrity was analyzed by coronal sectioning of the thorax of middle-aged (26-

day-old) or old (42-day-old) ND2 mutants and controls. Representative images of hematoxylin

and eosin stained thoracic muscle sections are depicted. (B) Brain integrity was analyzed by

coronal sectioning of the heads of middle-aged or old ND2 mutants and controls. Representative

images of hematoxylin and eosin stained coronal brain sections are depicted. Arrows demark

neurodegenerative vacuoles. Scale bars represent 200 μm. (C) Quantification of brain vacuole

size demonstrates an increase in vacuole size in ND2 mutants (p=4x10e-5). Vacuoles were rarely

detected in the brains of middle-aged ND2 mutants and were, therefore, not subjected to

quantification. Error bars represent standard error of the mean; n=26 animals for ND2 mutants;

n=21 animals for controls; ***=p<0.001. (D) Quantification of the number of brain vacuoles

demonstrates an increase in vacuole number in ND2 mutants (p=0.008). Vacuoles were rarely

detected in the brains of middle-aged control or ND2 mutants and were, therefore, not subjected

to quantification. Error bars represent standard error of the mean; n=26 animals for ND2 mutants;

n=21 animals for controls; **=p<0.01.

Fig. 3. ND2 mutants exhibit decreased mitochondrial function. (A) Mitochondrial complex I-

dependent respiration rates were measured in mitochondria isolated from middle-aged (14-day-

old) or old (30 to 35-day-old) ND2 mutant or controls, revealing age-related declines in both state

3 and maximal state 3 values within genotypes (p=0.08 and p=0.03 for comparison of middle-

aged and old control or ND2 mutant state 3 values, respectively; p=0.001 and p=0.003 for

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

25

comparison of middle-aged and old control or ND2 mutant maximal state 3 values, respectively).

State 4 respiration was similar under all conditions tested. No differences were found for state 3

values between control and ND2 mutants at either middle- or old ages. However, maximal state 3

values were decreased in both middle-aged and old ND2 mutants when compared to age-matched

controls (p=0.04 for middle-aged, and p=0.003 for old animals). Error bars represent standard

deviations; n=3 independent mitochondrial preparations; **=p<0.01; *=p<0.05; #=p<0.1). (B)

Oxidative phosphorylation efficiency (ADP/O) is decreased in ND2 mutants relative to controls at

both middle and old ages (p=0.02 and p=0.01, respectively). Error bars represent standard

deviations; n=3 independent mitochondrial preparations; *=p<0.05. (C) Blue native gel/Complex

I In-gel activity analysis of mitochondria isolated from middle-aged (14-day-old) ND2 mutants

and controls reveals similar levels of complex V in both genotypes, but decreased fully assembled

complex I abundance in ND2 mutants. (D) Western blot analysis of protein extracts from 14-day-

old ND2 mutants and controls reveals a specific decrease in the abundance of the complex I

subunit NDUFS3, with no decrease in the Complex Vβ subunit. (E) Quantification of Western

blot data for NDUFS3 (n=3 independent tissue extracts; p=0.003) and Complex Vβ (n=3

independent tissue extracts; p=0.46) levels normalized to actin, and expressed as the % of control

values. Error bars represent standard error of the mean; **=p<0.01.

Fig. 4. ND2 mutants exhibit decreased membrane potential and ATP production. (A) Neurons

from 30 to 35-day-old ND2 mutants and controls labeled with the mitochondrial membrane

potential dye TMRE were analyzed by flow cytometry. Histograms depict the relative

mitochondrial membrane potential of neurons from these preparations stained with TMRE. The

y-axis represents the percent maximal number of cells measured at each fluorescence intensity.

(B) Quantification of the average percent of isolated neurons exhibiting mitochondrial membrane

potential-dependent fluorescence above 103 arbitrary units (% Polarized mitochondria) are shown

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

26

for control and ND2 mutants (p=0.004). Error bars represent standard deviation; n=3 independent

brain preparations; **=p<0.01. (C) ATP levels were determined in heads of 35-day-old ND2

mutants and controls, and displayed relative to controls values (p=0.05). Error bars represent

standard error of the mean; n=3 groups of 5 individuals; *=p≤0.05 for a one-tailed t-test.

Table 1. Oxygen consumption measurements in middle-aged or old mitochondria isolated from

ND2 mutants or aged-matched controls. Respiratory rates are expressed in units of O2 consumed

nmol/min/mg (n=3 independent mitochondrial preparations). ADP/O ratios are unit-free (n=3

independent mitochondrial preparations). The corresponding standard deviations and p-values for

each measurement are indicated.

Fig. S1. ND2 sequence alignment. Primary amino acid sequence alignment of ND2 from Homo

sapiens, Mus musculus, Danio rerio, Drosophila melanogaster, Caenorhabditis elegans and

Escherichia coli are depicted. Identical residues are colored red, and similar residues colored blue.

The three amino acid deletion caused by the ND2del1 mutation is indicated with a green bar

(residues 186-188). An insertion of a serine residue caused by the ND2ins1 mutation at position

189 is depicted by a blue bar, and the L71P missense mutation found in a human Leigh syndrome

patient (Ugalde et al., 2007) is indicated by a red bar.

Fig. S2. Rescue of stress-induced paralysis by overexpression of yeast Ndi1, and increased

hyperoxia sensitivity in ND2 mutants. (A) 20 to 23-day-old female ND2 mutants expressing Ndi1

under the control of a Hsp70-GAL4 driver, and age-matched sibling ND2 mutants lacking Ndi1

expression (control) were maintained at 25oC, and assayed for the length of time (seconds) that

flies remained paralyzed following mechanical stress. Error bars represent standard error of the

mean; n=3 independent groups of 15 individual animals. p=0.044; *=p<0.05. (B) Survival of

ND2 mutants and controls in 100% oxygen. Histograms depict the median survival of control

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

27

(120 hours ± 1.533) and ND2 mutants (96 hours ± 1.625) (n=4 groups of 25 animals for each

genotype); ***=p<0.0001; Error bars represent the standard error of the mean.

Fig. S3. ND2 mutants do not exhibit degenerative changes in muscle fibers. (A) Confocal images

of indirect flight muscles from 40-42 day-old controls, ND2 mutants, and 10-day-old parkin

mutants immunostained for actin (Phalloidin (red)), cytochrome C (cyt C) (cyan), and cleaved

caspase 3 (CC3) (green). Scale bar= 20 μm (B) Quantification of the number of cleaved caspase 3

positive muscle cells from images of controls (n=11 animals), ND2 mutants (n=8 animals), and

parkin mutants (n=9 animals). p=0.049 for parkin mutants vs. w1118, and p=0.26 for ND2

mutants vs. w1118. ns= not significant; *=p<0.05. Error bars represent standard error of the

mean. (C) Quantification of the average mitochondrial perimeter (μm) from images of control

(n=5 animals; N=3369 mitochondria), ND2 mutant (n=6 animals; N=2456 mitochondria) and

parkin mutants (n=5 animals; N=880 mitochondria) muscle preparations. p=0.03 for parkin

mutants vs. w1118, and p=0.61 for ND2 mutants vs. w1118. ns= not significant; *=p<0.05. Error

bars represent standard error of the mean.

Fig. S4: Representative oxygen electrode traces obtained using mitochondria isolated from

control or ND2 mutants: Mitochondria were isolated from 14-day-old control or ND2 mutants,

and respiratory traces measured using a Clark-type electrode. 300 µg of protein was added to 0.5

ml respiration buffer at point A. At point B the complex I-linked substrates malate and pyruvate

were added to a initial concentration of 10 and 20 mM, respectively. At point C, 90 pmol ADP

was added to an initial concentration of 180 μM (“low ADP”). At point D, ADP was added to an

initial concentration of 2 mM (“high ADP”). The rates of oxygen consumption were measured

from the slopes of the oxygen traces for state 3, state 4, and maximal state 3, as indicated by the

arrows. ADP/O ratios were measured as the ratio of the pmol ADP added at point C and the pmol

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

28

oxygen atoms consumed between point C, and the transition between state 3 and state 4

respiration.

Fig. S5. Decreased complex I abundance in ND2 mutants. Western blot analysis of protein

extracts from 14-day-old ND2 mutants and controls reveals a decrease in the abundance of

mitochondrial complex I, as indicated by (A, B) decreased levels of the complex I subunit

NDUFS3, with (B) no decrease in the Complex Vβ subunit. For clarity, whole blots are shown

with their respective molecular weight markers.

Fig. S6. ND2 mutants do not exhibit changes in ROS-mediated protein damage: A) Western blot

analysis of protein extracts from ND2 mutants and controls throughout an aging time course

reveals no increase in ROS-mediated protein damage in ND2 mutants relative to controls, as

indicated by 4-HNE protein adduct levels at 1, 14, 21 or 35 days of age. B) Quantification of

Western blot data from (A), expressed as the % of control values following normalization to

actin. n=3 independent western blot experiments for each age group. Error bars represent standard

error of the mean.

Movie S1. Mechanical stress-induced paralysis of ND2 mutants. Middle-aged female ND2

mutants and controls were vortexed in inverted vials containing cotton stoppers for 10 seconds

and then filmed in real time. Controls (upper frame) were repeatedly vortexed throughout the

movie, but displayed no movement defects. However, ND2 mutants (lower frame) remained

paralyzed for >50 seconds before righting themselves.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

29

30

31

32

33

Table 1: Complex I (CI), Complex II (CII) and Complex (IV) oxygen consumption rates and ratios from middle-aged and old animals

Genotype Age (days)

Measurement Respiration Rate (nmol O2 /min/mg)

or ratio

Standard Deviation

P value

Control 14 CI:State 3 86.2 ±27.2 0.50 ND2del1 14 CI:State 3 99 ±14.6 Control 14 CI:State 4 27.2 ±5.3 0.32 ND2del1 14 CI:State 4 32.5 ±6.4 Control 14 CI:Max.State3 180.2 ±13.6 0.04 ND2del1 14 CI:Max.State3 150.5 ±13.4 Control 14 CI:ADP/O 2.7 ±0.2 0.02 ND2del1 14 CI:ADP/O 1.9 ±0.1 Control 14 CII:State 3 42 ±1.95 0.19 ND2del1 14 CII:State 3 49.9 ±6.2 Control 14 CII:Max.State3 43.2 ±4 0.16 ND2del1 14 CII:Max.State3 52.2 ±7.2 Control 14 CII:ADP/O 1.7 ±0.1 0.42 ND2del1 14 CII:ADP/O 1.6 ±0.1 Control 14 CIV 426.2 ±100 0.39 ND2del1 14 CIV 546.2 ±198.25 Control 35 CI:State 3 48.4 ±5.3 0.45 ND2del1 35 CI:State 3 54.5 ±12.5 Control 35 CI:State 4 19.6 ±9.85 0.39 ND2del1 35 CI:State 4 25.7 ±5.65 Control 35 CI:Max.State 3 119.5 ±25.2 0.003 ND2del1 35 CI:Max.State3 82.6 ±8.8 Control 35 CI:ADP/O 2.9 ±0.1 0.01 ND2del1 35 CI:ADP/O 1.8 ±0.3 Control 35 CIV 313 ±75.4 0.32 ND2del1 35 CIV 406 ±62.2

34

35

36

37

38

39

40

TRANSLATIONAL IMPACT: Clinical issue: Electron transport chain (ETC) complex I

dysfunction is the most common cause of primary mitochondrial diseases. For example,

mutations in the mitochondrial encoded complex I subunit NADH dehydrogenase subunit 2

(ND2) have been shown to be a cause of Leigh syndrome. Leigh syndrome patients can suffer

from numerous symptoms that include decreased lifespan, seizures, heat intolerance,

neurodegeneration, and decreased complex I activity and abundance. However, the molecular

mechanisms leading from ETC dysfunction to disease pathogenesis remain unclear.

Results: In this study the authors characterize the first Drosophila melanogaster model of

complex I deficiency caused by a mutation in a mitochondrial-encoded complex I subunit. The

authors present biochemical and behavioral data showing that ND2 mutants exhibit phenotypes

that resemble Leigh syndrome. These phenotypes include decreased lifespan, seizures, heat

intolerance, movement deficits, neurodegeneration, and decreased complex I activity and

abundance. In addition, the authors uncover a defect in the ability of ND2 mutants to pump

protons through the ETC. This is demonstrated by the observation of a decrease in the efficiency

of complex I, but not complex II, to power ADP conversion to ATP in ND2 mutants. The authors

hypothesize that the pathogenesis of ND2 mutants results from decreased efficiency of proton

pumping by complex I, leading to energy deficits, and age-related neurodegeneration. Supporting

this hypothesis the authors demonstrate decreased ATP levels, and decreased mitochondrial

membrane potential in the brains of old ND2 mutants.

Implications and future directions: Composed of 45 subunits, complex I is the most

intricate complex of the ETC, but the function of many complex I subunits remain unclear. The

author's show that ND2 has a critical role in the proton pumping mechanism of complex I, and

point towards deficits in energy production as etiological factors underlying the pathogenesis of

complex I deficiency. This Drosophila model exhibits a number of phenotypes amenable to

genetic screening methodologies, and therefore provides a powerful genetic system in which to

further dissect the mechanisms underlying mitochondrial disease pathogenesis.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt