Exp ession o al e na i e oxidase in D osophila
amelio a es di e se pheno ypes due
o cy och ome oxidase de iciency
Kia K. Kemppainen1, Juho Rinne1, Ashwin S i am1, Ma i Lakanmaa1, Akba Zeb1, Tea Tuomela1,
Anna Popples one1, Sa pal Singh2, Albe o Sanz1, Pie e Rus in3and Howa d T. Jacobs1,4,∗
1
Ins i u e o Biomedical Technology and Tampe e Uni e si y Hospi al, Uni e si y o Tampe e, FI-33014 Tampe e,
Finland,
2
School o Medicine and Biomedical Sciences, S a e Uni e si y o New Yo k a Bu alo, 206 Ca y Hall, Bu alo, NY
14214, USA,
3
INSERM UMR 676, Ho
ˆpi al Robe Deb e
´,48BdSe
´ u ie , 75019 Pa is, F ance and
4
Molecula Neu ology
Resea ch P og am, Uni e si y o Helsinki, FI-00014 Helsinki, Finland
Recei ed Augus 20, 2013; Re ised and Accep ed No embe 22, 2013
Mi ochond ial dys unc ion is a signi ican ac o in human disease, anging om sys emic diso de s o childhood
o ca diomyopa hy, ischaemia and neu odegene a ion. Cy och ome oxidase, he e minal enzyme o he mi o-
chond ial espi a o ychain, is a equen a ge . Lowe euka yo es possess al e na i e espi a o y-chain enzymes
ha p o idenon-p o on- ansloca ingbypasses o espi a o ycomplexesI(single-subuni educednico inamide
adenine dinucleo ide dehyd ogenases, e.g. Ndi1 om yeas ) o III 1IV [al e na i e oxidase (AOX)], unde condi-
ions o espi a o y s ess o o e load. In p e ious s udies, i was shown ha ans e o yeas Ndi1 o Ciona in es-
inalis AOX o D osophila was able o o e come he le hali y p oduced by oxinso pa ialknockdown o complex I
o IV. He e, we show ha AOX can p o ide a comple e o subs an ial escue o a ange o pheno ypes induced by
global o issue-speci ic knockdowno di e en cIV subuni s, includingin eg alsubuni s equi ed o ca alysis,as
well as pe iphe al subuni s equi ed o mul ime iza ion and assembly. AOX was also able o o e come he pupal
le hali y p oduced by muscle-speci ic knockdown o subuni CoVb, al hough he escued lies we e sho li ed
and had a mo ili y de ec . cIV knockdown in neu ons was no le hal du ing de elopmen bu p oduced a apidly
p og essing locomo o and seizu e-sensi i i y pheno ype, which was subs an ially alle ia ed by AOX.
Exp ession o Ndi1 exace ba ed he neu onal pheno ype p oduced by cIV knockdown. Ndi1 exp essed in place
o essen ial cI subuni s p oduced a dis inc esidual pheno ype o delayed de elopmen , bang sensi i i y and
male s e ili y. These indings con i m he po en ial u ili y o al e na i e espi a o y chain enzymes as ools o
comba mi ochond ial disease, while indica ing impo an limi a ions he eo .
INTRODUCTION
Mi ochond ial diseases a ec ing he espi a o y complexes o
he oxida i e phospho yla ion (OXPHOS) sys em a e a di e se
collec ion o pa hologies, which can a ec almos any issue,
a any age (1,2). They a e ypically p og essi e in na u e, and
no e ec i e ea men s a e cu en ly a ailable. Whe e gene ic
causes a e known, hey can include lesions in any o hund eds
o genes whose p oduc s a e needed o he biosyn hesis o
unc ion o he espi a o y complexes. These genes, mo eo e ,
a e dis ibu ed be ween nuclea and mi ochond ial DNA
(m DNA), and a subse o mi ochond ial diseases also esul s
om de ec i e communica ion be ween he cell’s wo genomes.
In o de o unde s and be e he pa hophysiological mechan-
isms o mi ochond ial disease, and de elop a possible s a egy
o e en ual he apy, we ha e exploi ed he ac ha lowe euka -
yo es, including plan s and many in e eb a es, possess an al e -
na i e, non-p o on-pumping espi a o y chain in mi ochond ia,
whose componen s can ac as a bypass o he OXPHOS sys em
unde condi ions o espi a o y s ess o o e load (3,4). Al e na-
i e educed nico inamide adenine dinucleo ide(NADH) dehy-
d ogenases such as yeas Ndi1 can eplace complex I (cI),
∗
To whom co espondence should be add essed a : Ins i u e o Biomedical Technology, FI-33014 Uni e si y o Tampe e, Finland. Tel: +358-3-3551-7731;
Fax: +358-3-3551-7710; Email: [email p o ec ed]
#The Au ho 2013. Published by Ox o d Uni e si y P ess.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/3.0/),
which pe mi s un es ic ed euse, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.
Human Molecula Gene ics, 2014, Vol. 23, No. 8 2078–2093
doi:10.1093/hmg/dd 601
Ad ance Access published on No embe 29, 2013
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om
while al e na i e oxidases (AOXs) can eplace complexes III +
IV (cIII +IV). The al e na i e enzymes a e each composed o a
single gene p oduc , which oge he conduc elec ons om
NADH o molecula oxygen ia ubiquinone as an in e media e
elec on accep o . AOX becomes enzyma ically ac i e only
when elec on ans e ia he OXPHOS sys em becomes inhib-
i ed beyond ubiquinone. The main mechanism o achie e his is
belie ed o be ha AOX has a much highe K
m
o i s subs a e,
ubiquinol, han does cIII. In con as , Ndi1 may be cons i u i ely
ac i e, bu i s ole in elec on low unde no mal physiological
condi ions may be limi ed by he igh coupling be ween cI and
cIII in supe complexes.
Like Ndi1, AOX is absen om e eb a es, as well as a h o-
pods (5), bu we easoned ha i he gene p oduc we e ans-
e ed o highe me azoans i migh be able o unc ionally
eplace he co esponding OXPHOS complexes unde condi-
ions esembling hose pe aining in mi ochond ial disease.
When exp essed ansgenically in human cells (6,7), D osophila
(8) o mos ecen ly he mouse (9), AOX om he u ocho da e
Ciona in es inalis was shown o be enzyma ically unc ional
when cIII o cIV was inhibi ed by speci ic oxins, o gene ic mu-
a ion. Simila ly, when yeas Ndi1 was ans e ed o mamma-
lian cells o o D osophila, i was unc ional, enabling NADH
oxida ion when cI was inhibi ed by o enone (10). Rema kably,
he ubiqui ous p esence o hese al e na i e enzymes is ole a ed
by he whole o ganism, o which i can con e oxin esis ance
in i o (8–10). In wild- ype lies, no dele e ious pheno ype is
p oduced by cons i u i e AOX (o Ndi1) exp ession (8,10).
Howe e , he al e na i e enzymes can o e come he le hali y
p oduced by pa ial knockdown o a leas some subuni s o
he co esponding OXPHOS complexes, including wo subuni s
o he memb ane po ion o cI (10), a nuclea -coded ‘supe nume -
a y’ subuni o cIV [Cox6c, he cyclope gene p oduc (8)] and a
cIV assembly ac o , Su 1 (8). No e, howe e , ha AOX
canno complemen he o al loss o cIV, e.g. ia a null mu a ion
in Cox6c (9).
The s uc u e o me azoan cIV is ela i ely well unde s ood
(11). I comp ises a ca aly ic co e, composed o he h ee
m DNA-encoded subuni s, whose unc ional assembly is de-
penden on a se o phylogene ically conse ed nuclea -coded
subuni s (Cox4, Cox5a, Cox5b, Cox6b, Cox6c, p obably Cox7c,
Cox8, using he mouse nomencla u e) equi ed o p oduce, ia a
s epwise assembly pa hway, a memb ane-bound sub-complex
wi h enzyma ic ac i i y (12,13). The inco po a ion o p os he ic
g oups (a- ype haemes and coppe ) is also in ol ed in his
p ocess. The o he nuclea -coded subuni s (Cox6a and Cox7a)
a e hen inco po a ed o o m he ully unc ional cIV, which
also con ains a ecen ly desc ibed subuni (14) o me ly belie ed
o be a cons i uen o cI. Some o hese subuni s, no ably Cox6a,
a e belie ed o be impo an o dime iza ion o he complex
and, oge he wi h o he speci ic p o eins such as Rc 1
(HIG2A) and Cox7RP, may also be equi ed o he o ma ion
o supe complexes con ainingcIV,as well ascI and/o cIII indi -
e en s oichiome ies (15–18). The mo e speci ic oles o he
nuclea -coded subuni s o cIV (o en e med ‘supe nume a y’
since hey a e absen in bac e ia), o example in physiological
egula ion, a e less clea . Null mu a ions in mos o he abo e
subuni s a e le hal in D osophila, whe eas a splice-si e mu a ion
in he le y gene, encoding subuni Cox6a, p oduces an
adul -onse neu odegene a i e pheno ype (19).
Cy och omecoxidase (COX) de iciency inhumans has a wide
a ie y o pa hological mani es a ions and a di e si y o gene ic
causes (20,21). O gans a ec ed can be he cen al ne ous
sys em, skele al o hea muscle, he li e o a combina ion o
hese and o he o gans, and issue-speci ici y is poo ly unde -
s ood. Unde lying gene ic de ec s commonly impac accesso y
ac o s o cIV biosyn hesis, including assembly ac o s such
as SURF1 (22,23) o C2ORF64 (24) o p o eins in ol ed in co-
ac o syn hesis o anspo , such as SCO2 (25) and COX10
(26). Compa a i ely a e mu a ions a e also ound in genes o
s uc u al subuni s o cIV, such as COX6B1 (27)o COX7B
(28), as well as hose encoded in m DNA (29–31), plus genes
equi ed mainly o exclusi ely o he biosyn hesis o
m DNA-encoded cIV subuni s [e.g. LRPPRC (32), TACO1
(33)]. Pa hological COX de iciency has been epo ed in many
o he diso de s (34,35), including neu odegene a i e condi ions
such as Alzheime demen ia (36–38) and Hun ing on’s disease
(39) al hough i s ae iological signi icance emains unclea . COX
is also a pa hological a ge in ischaemia, sepsis and o he ypes
o oxic inju y (40).
Some cIV subuni s a e encoded by isogenes di e en ially
exp essed be ween issues, whe e hey a e adap ed o speci ic
physiological condi ions (35,41). Tissues also a y in hei sub-
s a e dependence, e lec ed in he deg ees o which he di e en
OXPHOS complexes con ibu e o h eshold e ec s o espi -
a ion (42). Fu he mo e, supe complexes, whose o ma ion may
also a y be ween issues, en ain a g ea e o lesse deg ee o
channelling o espi a o y elec on low, and may also limi he
deg ee o which exogenously in oduced al e na i e espi a o y
enzymes can con ibu e o espi a ion (18). In a gene al sense,
hese phenomena a e belie ed o con ibu e o he bewilde ing
issue-di e si y o mi ochond ial disease, hough he e emain
ew conc e e mechanis ic explana ions o his. The ela i e
impo ance in COX-associa ed pa hology o dis u bed edox
homeos asis, apop osis induc ion, de anged cell signalling, ad-
enosine iphospha e (ATP) de iciency, p o eo oxic s ess and
o he ypes o me abolic dis u bance emains a opic o in ense
deba e.
Thepossibleuseo AOXin u u e he apies o COXde iciency
will ob iously be limi ed by hese conside a ions, as well as by he
ac ha he al e na i e espi a o y chain enzymes do no con ib-
u edi ec ly oATP p oduc ion. Comple ely eplacing he unc ion
o one o mo e OXPHOS complexes canno comple ely es o e
ATP p oduc ion capaci y o wild- ype, e en hough i can acili-
a e p o on-pumping linked o elec on ans e a OXPHOS
complexes o he han he one(s) i is bypassing. Because AOX
bypasses wo o he h ee p o on- ansloca ing s eps, i should
no es o e ATP p oduc ion o hose subs a es whose oxida ion
supplies elec ons di ec ly o ubiquinone ia cII o o he
dehyd ogenases. E en i AOX can es o e edox balance, mi i-
ga e mi ochond ial eac i e oxygen species (ROS) p oduc ion
(6–9) and limi sys emic lac ic acidosis (6), issues ha
depend on such subs a es may s ill su e a subs an ial loss o
unc ion ha AOX canno co ec . On he o he hand, whe e
COXac i i y is limi ing o espi a ion, AOXmay p omo easig-
ni ican es o a ion o ATP p oduc ion ( ia p o on pumping a
o he si es), as well as o he me abolic bene i s.
D osophila o e s a con enien model o human diseases, in-
cluding hose a ec ing mi ochond ial unc ions. The compos-
i ion o he OXPHOS complexes, he o e all s uc u e and
Human Molecula Gene ics, 2014, Vol. 23, No. 8 2079
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om
gene con en o m DNA and he ene ge ic physiology o he main
o gan sys ems a e all simila o hose o humans. The ac ha
AOX exp ession in he ly is able o con e cyanide esis ance
and o e come he le hali y o pa ial knockdown o Cox6c o
Su 1 aises he ques ion o how a i is able o eplace cIV
h oughou he li e cycle. We hus se ou o use he ly as
a model sys em o add ess his issue, ocusing on h ee speci ic
ques ions ele an o unde s anding he pa hophysiology o
mi ochond ial disease and he possible u u e use o AOX in
he apy. Fi s , wha pheno ypes esul om de iciencies in di e -
en classes o cIV subuni s, and can AOX alle ia e hese pheno-
ypes? Secondly, in which issues is loss o COX unc ion c ucial
o p oducing such pheno ypes, and hei escue by AOX?
Thi dly, wha esidual pheno ypes esul om eplacing cIV unc-
ion wi h AOX? The many gene ic ools a ailable in D osophila
(including e ec i e and speci ic RNAi in i o)makei anideal
ool o such a s udy, in ad ance o ansla ing he mos p ominen
indings o he mouse, which is less lexible, and also akes much
longe and is subs an ially mo e expensi e o manipula e.
In a i s s ep, o alida e use o he inducible (ups eam ac i-
a ing sequence) UAS-AOX ansgenic lines ha we c ea ed
p e iously, we i s conduc ed an impo an con ol o e i y
ha he escue seen p e iously was no subjec o any con ound-
ing, ‘p omo e -dilu ion’ e ec s om he simul aneous use o
ansgenes dependen on he same ansc ip ion ac o . We
hen p oceeded o es he pheno ypes gene a ed by knockdown
o di e en cIV subuni s in he whole ly and in speci ic issues,
and he abili y o AOX o escue hese pheno ypes, ocusing on
he majo di e en ia ed cell- ypes a ec ed by human mi ochon-
d ial disease (muscle and neu ons).
The esul s highligh he di e en pheno ypes p oduced by de-
iciencies o di e en subuni s, indica e a c ucial de elopmen al
ole o cIV in muscle and con i m COX de iciency as a cause
o adul -onse neu odegene a ion. AOX exp ession was able o
pa ially escue hese pheno ypes. In con as , loss o p o on-
pumping a cI p omo es a dis inc pheno ype o de elopmen al
delay, bang sensi i i y and male in e ili y.
RESULTS
Cons i u i e low-le el AOX exp ession con e s esis ance
o COX inhibi ion
In p e ious expe imen s we exp essed inducible AOX using
he UAS/Gal4 (yeas ansc ip ion ac i a o p o ein GAL4)
sys em, which pa ially escued he le hali y o cyanide ea -
men o he dis up ion o COX by knockdown o he Cox6c
subuni ( he cyclope gene: see Table 1) o he Su 1 assembly
ac o . Howe e , he UAS/Gal4 sys em esul s in a e y high
le el o ansgene exp ession. Fu he mo e, since COX knock-
down by RNAi also in ol es he same induc ion sys em, we
canno comple ely exclude a con ibu ion o he escue om p o-
mo e compe i ion, e en hough UAS-GFP (g een luo escen
p o ein) was unable o escue. We he e o e c ea ed ansgenic
lies bea ing he same AOX ansgene, bu unde he con ol o
he cons i u i e a- ubulin p omo e ( ub-AOX), wi h independ-
en single inse ions in o non-coding DNA on ch omosomes X, 2
and 3 (Supplemen a y Ma e ial, Fig. S1A–D). Exp ession o
hese ansgenes a he RNA le el was 20–50- old lowe han
om UAS-AOX d i en by da-GAL4 (Fig. 1A), hough s ill
much highe han UAS-AOX in he absence o a GAL4 d i e .
ub-AOX exp ession was subs an ially highe in la ae and
adul males han emales (Fig. 1A), and was main ained o e
he i s 2weeks o adul li e o a a iable ex en .
AOX exp ession a he p o ein le el was also less han when
d i en by GAL4, e en when we combined ub-AOX inse ions
Table 1. Nomencla u e, exp ession pa e ns and assembly o COX subuni s in D osophila
Subuni
name
a
O icial gene name(s)
and symbol(s)
b
Exp ession pa e n(s)
c
Commen s
d
Cox1 m CoI Ubiqui ous m DNA-encoded, pa o co e sub-complex S2
Cox2 m CoII Ubiqui ous m DNA-encoded, inco po a ed in o sub-complex S3
Cox3 m CoIII Ubiqui ous m DNA-encoded, inco po a ed in o sub-complex S3
Cox4 CoIV Ubiqui ous Two isogenes wi h di e en exp ession pa e ns, pa o co e sub-complex S2
CG10396 Tes is-speci ic
Cox5a CoVa Ubiqui ous Pa o co e sub-complex S2
Cox5b CoVb Ubiqui ous Inco po a ed in o sub-complex S3
Cox6a le y Ubiqui ous Inco po a ed a inal assembly s eps in o ma u e complex IV; le y
1
splice-si e
mu an mani es s adul -onse neu odegene a ion (19)
Cox6b CoVIb Ubiqui ous Inco po a ed in o sub-complex S3
Cox6c cyclope Ubiqui ous Null-mu an la al le hal, inco po a ed in o sub-complex S3
Cox7a CG9603 Ubiqui ous, lowes in es is Inco po a ed a inal assembly s eps in o ma u e complex IV; 99% o b ain
Cox7a exp ession con ibu ed by CG9603
CG34172 Mainly muscle-speci ic (hea c op,
hindgu , ca cass, lowe in head)
CG18193 Tes is-speci ic
Cox7b None iden i ied P oposed o be equi ed o an ea ly assembly s ep (28); clea o hologues
iden i ied only in e eb a es
Cox7c CoVIIc Ubiqui ous, bu low in es is Only ecen ly iden i ied in D osophila
Cox8 CoVIII Ubiqui ous, bu low in es is Inco po a ed in o sub-complex S3
a
Using mouse nomencla u e. No e ha , some subuni s a e encoded by gene amilies in mouse bu single genes in D osophila, and ice e sa.
b
F om www. lybase.o g.
c
F om www. lya las.o g.
d
Assembly p og am based on Re . (12).
2080 Human Molecula Gene ics, 2014, Vol. 23, No. 8
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om
on all h ee majo ch omosomes in o a single line o lies
(Fig. 1B). ub-AOX con e ed cyanide esis ance o subs a e
oxida ion by isola ed mi ochond ia (Supplemen a y Ma e ial,
Fig. S1E), and single ub-AOX inse ions we e able o con e
almos as g ea a ime o esis ance o cyanide as UAS-AOX
d i en by da-GAL4 (Fig. 1C, compa e wi h Fig. 5A o Re . 8).
A single copy o ub-AOX was also able o escue he semi-
le hali y o knockdown o ei he Cox6c (Fig. 2A) o Su 1
(Fig. 2B and C), when RNAi was induced wi h di e en doses
o RU486, in he p esence o ubiqui ously exp essed GeneS-
wi ch, a modi ied e sion o GAL4 ha is dependen on he
d ug o ac i i y. [No e ha , o cla i y, we use he mammalian
(mouse) names o all subuni s and he genes ha encode hem
h oughou he ex , indica ing he D osophila gene name only
upon i s men ion, in Table 1and in igu e legends.] Knockdown
o Cox6c by abou 80% a he RNA le el, using 10 mMRU486
(Supplemen a y Ma e ial, Fig. S2A), no only esul ed in
le hali y o abou h ee-qua e s o he p ogeny (Fig. 2A), bu
hose lies ha we e able o eclose did so wi h a 3-day delay
(Fig. 2D), a pheno ypic ea u e seen in many D osophila
OXPHOS mu an s. The p esence o ub-AOX co ec ed bo h o
hese pheno ypes (Fig. 2A and D). ub-AOX also enabled de el-
oping lies o each pupal s age a doses o RU486 ha induced
a deg ee o Su 1 knockdown ha p e en ed any pupa ia ion
o con ol lies (Fig. 2B). A 0.1 mMo he d ug, a subs an ial
numbe o ub-AOX exp esso s e en eached eclosion, whe eas
only a ew iable adul s we e p oduced a his dose by con ol
lies unde Su 1 knockdown alone (Fig. 2C).
Ex en o COX de iciency escued by AOX exp ession
Knockdown o subuni s Cox5a o Cox5b also p oduced an
RU486 dose-dependen le hali y (Fig. 2E and F). This enabled
us o es how escue is a ec ed by he numbe o cons i u i ely
Figu e 1. T ansgenic exp ession o ub-AOX.(A) Quan i a i e eal ime-polyme ase chain eac ion (qRT-PCR) analysis o AOX mRNA exp ession (no malized
agains RpL32) in ansgenic adul s and la ae, as indica ed. Means +SD o a leas h ee echnical eplica es o each o a leas h ee biological eplica es. Compa ing
(ups eam ac i a ing sequence) UAS-AOX exp ession d i en by da-GAL4 (yeas ansc ip ion ac i a o p o ein GAL4) wi h ha o ub-AOX,P,0.01 o each sex/
age analysed; simila ly, compa ing UAS-AOX exp ession in he absence o d i e wi h ha o ub-AOX,P,0.01 in each case, excep whe e indica ed (#), whe e P,
0.05 (S uden ’s es , wo- ailed, unequal a iances). (B) Wes e n blo o AOX p o ein and adenosine iphospha e (ATP) syn hase subuni a(loading con ol) in
1-day-old adul s o he ansgenic s ains indica ed. ub-AOX deno es lies homozygous o each o ub-AOX
7
, ub-AOX
35
and ub-AOX
50
ansgenes (males a e hemi-
zygous o ub-AOX
35
). Replica e ba ches o p o ein ex ac s om 30 emales o 40 males o each geno ype a e shown in adjacen lanes. (C) Su i al ime on
cyanide-imp egna ed aga o lies o he (homozygous) s ains indica ed. Means +SD o 80–100 lies o each g oup, in ba ches o 10 lies pe ial. P,0.01 in
each case, in compa ison wi h w
1118
con ol lies o same sex (S uden ’s es , wo- ailed, unequal a iances). See also Supplemen a y Ma e ial, Figu e S1.
Human Molecula Gene ics, 2014, Vol. 23, No. 8 2081
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om
exp essed copies o AOX, and o es ima e he amoun o knock-
down o COX equi ed o p oduce AOX- escuable le hali y.
A single copy o ub-AOX was su icien o o e come he le hal-
i y o knocking down Cox5a o Cox5b o abou hal o he
wild- ype le el a la al s age L3 (Supplemen a y Ma e ial,
Fig. S2A), which also p oduced a dec ease o abou 50% in he
enzyma ically measu ed COX ac i i y (Supplemen a y Ma e ial,
Fig. S2B). Flies escued om he le hali y o Cox5a o Cox5b
knockdown by wo copies o ub-AOX eclosed wi h a sligh , bu
s a is ically non-signi ican de elopmen al delay (Supplemen a y
Ma e ial, Fig. S2D). The escued lies we e e ile when ma ed o
wild- ype lies o he opposi e sex.
Flies in which Cox5b was knocked down du ing de elopmen
by a highe dose (3 mM) o RU486, ailed o de elop beyond he
la al s age, bu ou copies o ub-AOX enabled he lies o each
pupal s age, wi h 12%o he pupae eclosing.The lies we e es ed
o bang sensi i i y, a senso ineu al pheno ype associa ed wi h
mi ochond ial dys unc ion, bu no bang sensi i i y was ound
(Supplemen a y Ma e ial, Fig. S2E).
Ha ing excluded p omo e -dilu ion e ec s as a eason o AOX
escue o COX de iciency, we se ou o use he UAS-AOX lines, in
combina ion wi h di e en GAL4 d i e s, o analyse he de elop-
men al pheno ypes p oduced by knockdown o COX subuni s, and
he deg ee o which AOX exp ession is able o complemen hem.
AOX exp ession escues de elopmen al le hali y o COX
knockdown
The nuclea -coded subuni s o cIV a e belie ed o con e a ious
egula o y p ope ies on COX, as well as being equi ed o i s
p ope assembly and s abili y. De iciency o speci ic subuni s
leads o he accumula ion o a ious assembly in e media es
Figu e 2. Rescue o COX de iciency by ub-AOX.(A) ub-AOX escue o de elopmen al le hali y and (D) o de elopmen al delay, om pa ial knockdown o Cox6c
(D osophila gene cyclope) using ub-GS d i e . P opo ion o eclosing p ogeny o eclosion day o di e en geno ypes and concen a ions o RU486 as shown,
means +SD om 4 o mo e biological eplica es. (Band C) ub-AOX escue o de elopmen al le hali y om pa ial knockdown o Su 1 using ub-GS d i e .
The numbe o pupae o eclosing lies o di e en geno ypes and concen a ions o RU486 as shown, means+SD om 3 o mo e biological eplica es. (Eand
F) Rescue o de elopmen al le hali y ompa ial knockdown o Cox5a o Cox5b (D osophila genes CoVa,CoVb) using ub-GS d i e , as shown. Numbe o eclosing
p ogeny a di e en concen a ions o RU486, o he indica ed numbe s o ub-AOX ansgenes. Means +SD om 3 o mo e biological eplica es. P,0.01 (∗)o ,
0.05 (#), S uden ’s es , wo- ailed, unequal a iances, compa ing lies wi h and wi hou ub-AOX. See also Supplemen a y Ma e ial, Figu e S2. No e ha a 0 mM
RU486 all knockdown lines es ed we e indis inguishable om wild- ype lies in he assays shown, and ha wild- ype lies eclose on Days 10 and 11 a 258C. Two
copies o ub-AOX also escued he le hali y o Cox6b knockdown (Supplemen a y Ma e ial, Fig. S2C).
2082 Human Molecula Gene ics, 2014, Vol. 23, No. 8
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om

and o sub ly di e en pheno ypes. Mos nuclea -coded subuni s
o COX in D osophila a e encoded by single-copy nuclea
genes (Table 1). Two no able excep ions a e Cox4, encoded by
a speci ic isogene (CG10396) in es is, wi h a second isogene
(CoIV, CG10664) exp essed ubiqui ously, and subuni Cox7a,
also encoded by es is-speci ic (CG18193) and ubiqui ous
(CG9603) isogenes, plus an addi ional, widely exp essed
isogene (CG34172), p ominen ly exp essed in muscle (ca cass,
c op, hindgu and hea ).
Using da-GAL4 o d i e simul aneously he high-le el ex-
p ession o UAS-AOX and he knockdown o nuclea -coded
COX subuni s in he whole de eloping ly, we delinea ed
he limi s o AOX escue o COX de iciency. Knockdown o
he soma ic Cox4 isogene p oduced ea ly la al le hali y
(Fig. 3A), consis en wi h he in ol emen o his subuni in an
essen ial ea ly s ep in cIV assembly, and i s equi emen o
he p oduc ion o an enzyma ically unc ional complex. Knock-
down la ae su i ed o a leas 15 days, bu ne e de eloped
beyond a mo phologically abno mal L1 o L2 s age (Fig. 3A),
whe eas co-exp ession o UAS-AOX enabled de elopmen
o p oceed as a as he pupal s age (Fig. 3A), albei wi h
e y ew lies eclosing (Supplemen a y Ma e ial, Fig. S3A).
Blue na i e polyac ylamide gel elec opho esis (BNE) in-gel
his ochemis y con i med he unc ional knockdown (≥50% de-
c ease o COX ac i i y) a la al s age L3 in he ‘ escued’ lies
(Fig. 3E), as did pola og aphy using a cIV-speci ic subs a e
mix (Fig. 3F). In e es ingly, oxygen consump ion d i en by
cI- o G3PDH-linked subs a es was inc eased compa ed wi h
Figu e 3. Pa ial escue o global COX de iciency by UAS-AOX.(A) Mic og aphs illus a ing ypical pheno ypes p oduced by knockdown o Cox4 (D osophila gene
CoIV) d i en by da-GAL4. In he absence o ansgenic escue, p ogeny a es ed as abno mal L1/L2 la ae. Co-exp ession o UAS-AOX enabled de elopmen o
p oceed as a as la e pupa, al hough ew lies eclosed. See also Supplemen a y Ma e ial, Figu e S3A. (B) Mic og aph illus a ing ypical pupal-le hal pheno ype p o-
duced by knockdown o he majo Cox7a-encoding isogene CG9603, d i en by da-GAL4.(C) AOX escue o pupal le hali y om Cox7a (CG9603) knockdown.
P opo iono eclosingp ogeny o di e en geno ypes as shown,means +SD om3 o mo e biological eplica es.Flieswi h he da-GAL4d i e ha eno malb is les,
dis inguishing hem om hose wi h he Sb balance . Exp ession o UAS-AOX p oduced a ull escue whe eas UAS-Ndi1 exp ession p oduced none. (D) Asco ba e/
TMPD-d i enoxygen consump iono homogena es omUAS-AOXexp essing L3 la ae o adul lies, wi h o wi hou knockdown o Cox7a (CG9603).Means +SD
om 3 o mo e biological eplica es; as e isks indica e signi ican di e ences (P,0.01, S uden ’s es , wo- ailed, unequal a iances). (E) BNE gels o mi ochon-
d ial ex ac s (37.5 mg mi ochond ial p o ein pe lane) om L3 la ae o he geno ypes shown, s ained his ochemically o cI o cIV ac i i y (cIV ac i i y s aining
pe o med o he indica ed imes). See Re . 10, Figu e 3, o mig a ion o majo bands in ela ion o molecula weigh ma ke s on hese gels. As e isk indica es as-
sembly sub-complex S3 (12). (F) Oxygen consump ion o homogena es om L3 la ae o he indica ed geno ypes, d i en by di e en subs a e mixes (py u a e +
p oline, G3P and asco ba e +TMPD, espec i ely, as desc ibed in he Ma e ials and me hods sec ion o cI-, G3PDH- and cIV-linked espi a ion. Fo cla i y, oxygen
consump ion is exp essed as a pe cen age o co esponding alues o con olla ae exp essing AOX bu wi hou RNAi. Means +SD, ≥3 biological eplica es; as e -
isks indica e signi ican di e ences (P,0.01, S uden ’s es , wo- ailed, unequal a iances).
Human Molecula Gene ics, 2014, Vol. 23, No. 8 2083
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om
con ol la ae, sugges ing a compensa ion o missing p o on-
pumping ac i i y in esponse o AOX ‘ eplacemen ’ o he
deple ed cIV.
Exp ession o UAS-Ndi1 as a con ol p oduced no escue o
Cox4 knockdown (Supplemen a y Ma e ial, Fig. S3A). Knock-
down o Cox5b p oduced a simila la al-le hal pheno ype ha
was also pa ially escued by UAS-AOX, allowing lies o
de elop o pupal s age, wi h 5–10% o pupae eclosing as
iable lies (Supplemen a y Ma e ial, Fig. S3B).
Subuni Cox7a is equi ed o he assembly o he ully unc-
ional holoenzyme, bu no o he o ma ion o an enzyma ically
ac i e sub-complex (deno ed as assembly in e media e S3).
Knockdown o he majo isogene o Cox7a (CG9603) also
esul ed in le hali y, bu in con as o he ea ly la al le hali y
p oduced by knockdown o co e subuni s, Cox7a knockdown
p oduced le hali y only a he pupal s age (Fig. 3B). This was
comple ely escued by co-exp ession o UAS-AOX (Fig. 3C),
bu no by UAS-Ndi1 (Fig. 3C). Consis en wi h he inco po -
a ion o Cox7a a a la e s age in cIV assembly and he pa ial e-
dundancy o he majo Cox7a-encoding isogene CG9603 wi h
isogene CG34172 in some issues, knockdown o CG9603
esul ed in only a pa ial loss o espi a o y capaci y in he
whole ly, based on pola og aphy (Fig. 3D) o BNE in-gel his o-
chemis y (Fig. 3E), wi h a clea accumula ion o assembly sub-
complex S3 in L3 la ae.
AOX- escuable pupal le hali y due o COX knockdown is
p ima ily a muscle pheno ype
To es he issue(s) in which COX exp ession is c i ical o he
comple ion o de elopmen , we employed a se o issue-speci ic
d i e s o knockdown ei he o he enzyma ically c ucial subuni s
Cox5b (Fig. 4A) o Cox6b (Supplemen a y Ma e ial, Fig. S4A).
No le hali y was p oduced by Cox5b knockdown using any o
i e ne ous sys em-speci ic d i e s, wo o which a e ac i e in
all neu ons (ela -GAL4 loca ed on ch omosome 3, Blooming on
s ain 8760 and n 2-GAL4), no we e he p ogeny lies bang-
sensi i e (Supplemen a y Ma e ial, Fig. S4B). Le hali y was
also no p oduced wi h d i e BG57 (Fig. 4A), which exp esses
in la al muscles om L2 s age and in pupal and adul s ages
Figu e 4. AOX escue o issue- es ic ed cIV knockdown. (A–C) Knockdown o Cox5b (D osophila gene CoVb) using he d i e s indica ed. Flies wi h he CyO
balance ma ke a e p ogeny om he same c osses, bu wi hou d i e . (A and B) P opo ion o eclosing p ogeny o he geno ypes indica ed, means+SD o 3
o mo e biological eplica es. (C) Climbing index (de ined as in he Ma e ials and me hods sec ion) o 1-day-old UAS-AOX ansgenic lies bea ing he G14
d i e , wi h and wi hou knockdown o Cox5b. Means+SD o 10 ba ches o 5 lies, o each sex and geno ype. No e ha he AOX- escued Cox5b-knockdown
males we e unable o climb a all in his expe imen . (Dand E) Su i al cu es o lies o he indica ed geno ypes, ollowing Cox5b o Cox4 (D osophila gene
CoIV)knockdown and ansgene exp ession d i en by (D) ela -GAL4 s ain 458 o (E) 8760, as shown. Da a a e means om wo independen expe imen s. See
also Supplemen a y Ma e ial, Figu e S4.
2084 Human Molecula Gene ics, 2014, Vol. 23, No. 8
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om
p edominan ly in abdominal muscle (Supplemen a y Ma e ial,
Fig. S5A). Howe e , knockdown using d i e G14, which ex-
p esses mainly in he ho acic muscles (Supplemen a y Ma e ial,
Fig. S5B), p oduced le hali y (Fig. 4A and B), wi h lies dying a
he la e pupal s age o du ing a emp ed eclosion. Cox6b knock-
down ga e simila esul s (Supplemen a y Ma e ial, Fig. S4A).
Co-exp ession o AOX in he same issues ia UAS-AOX e-
scued he le hali y o Cox5b knockdown by he G14 d i e
(Fig. 4B). Mul iple copies o ub-AOX also escued his le hali y,
hough less comple ely (Supplemen a y Ma e ial, Fig. S4C).
Flies escued by UAS-AOX om he muscle-speci ic le hali y
o Cox5b knockdown ne e heless mani es ed a se e e loco-
mo o de ec (Fig. 4C), and all died wi hin 2 weeks, as a esul
o becoming apped in he ood.
AOX escue o pheno ypes p oduced by COX knockdown
using neu on-speci ic d i e s
The ailu e o induce le hali y by COX knockdown in neu ons is
supe icially su p ising. We i s e i ied his o knockdown
o a u he COX subuni (Cox4), which ga e a simila esul
(Supplemen a y Ma e ial, Fig. S4D). Nex , we es ed a second
ela -GAL4 d i e loca ed on ch omosome X (Blooming on
s ain 458). Di e en c osses we e implemen ed, so as o s udy
emales he e ozygous o he d i e , plus males hemizygous
o he d i e , o con ol males lacking he d i e comple ely
(Supplemen a y Ma e ial, Fig. S4E, Supplemen a y Ma e ial,
Table S1), as well as any escue p oduced by UAS-AOX (o
UAS-Ndi1). In gene al, ela -GAL4 d i e s ain 458 p oduced
s onge e ec s han d i e s ain 8760, and s onge pheno ypes
o knockdown o Cox4, which was now almos comple ely
le hal (Supplemen a y Ma e ial, Fig. S4E and F), han o Cox5b.
Males we e also mo e a ec ed han he e ozygous emales. Ex-
p ession o AOX, bu no Ndi1, imp o ed all pheno ypes. Viable
Cox5b-knockdown lies we e ans e ed o esh ials and hei
su i al acked o e 2 weeks (Fig. 4D). All geno ypes mani es ed
locomo o impai men a e se e al days, which wasmo e se e e in
males. Females wi h o wi hou co-exp ession o AOX su i ed
h oughou he expe imen , bu hal o he emales co-exp essing
Ndi1 died by 2 weeks. Cox5b-knockdown males had a mean li e-
span o jus 8 days, bu co-exp ession o AOX enabled mos lies
o su i e a leas 2 weeks. In con as , Cox5b-knockdown males
co-exp essing Ndi1 all died a e only a ew days.
When knockdown was d i en by n 2-GAL4 o he au oso-
mally loca ed ela -GAL4 d i e (Blooming on s ain 8760)
he ou come o simila su i al expe imen s was much milde .
Cox4 knockdown again ga e a s onge pheno ype han ha o
Cox5b, and was again s onge in males han emales. A e 2
weeks, some o he lies wi h Cox4 knocked down by he
n 2-GAL4 d i e had died ( hough his a ied be ween expe i-
men s), whe eas Cox5b knockdown using he n 2-d i e did no
impai su i al o e 2 weeks a all. Cox4 knockdown by he au o-
somal ela -GAL4 d i e (Fig. 4E) p oduced a simila e ec
as Cox5b knockdown by he s onge , X-ch omosomal ela -
GAL4 d i e (Fig. 4D). Knockdown lies we e able o climb,
and we e obse ed eeding, al hough mos ly emained mo ion-
less, whe eas lies co-exp essing AOX we e no mally ac i e
and able o ly. Mos knockdown males died wi hin a ew days,
a pheno ype exace ba ed by co-exp ession o Ndi1, whe eas
80% o males co-exp essing AOX su i ed a leas 2 weeks.
Females we e less a ec ed, hough ollowed a simila pa e n
as wi h Cox5b knockdown d i en by X-ch omosomal ela -
GAL4, wi h AOX a o ding p o ec ion agains ea ly dea h.
To accoun o he inconsis ency be ween he e ec s o knock-
down using di e en neu on-speci ic d i e s we e-examined he
issue-speci ici y o hese d i e s, using UAS-GFP (S inge o
mCD8) as a epo e (Supplemen a y Ma e ial, Fig. S5) and by
immunocy ochemis y and his ochemis y (Supplemen a y Ma-
e ial, Fig. S6). These expe imen s e ealed ha exp ession
d i en by n 2-GAL4 d i e was indeed much weake han ha
d i en by ela -GAL4, while he la e ga e also a e y weak ex-
p ession in ho acic muscles a pupal s age (see legends o Sup-
plemen a y Ma e ial, Figs S5 and S6 o de ailed explana ions).
Se ial sec ions o he b ain and ho acic muscle e ealed no g oss
ana omical de ec s om Cox4 knockdown using e en he s on-
ges ela -GAL4 d i e (Supplemen a y Ma e ial, Fig. S6G).
AOX pa ially escues adul neu odegene a ion caused
by COX de iciency
In o de o con i m ha he degene a i e pheno ypes p oduced
by ela -GAL4 d i en COX knockdown and pa ially escued
by AOX exp ession we e indeed neu onal, we adop ed wo s a -
egies. Fi s , we ook ad an age o he ac ha Cox7a in muscle is
p edominan ly encoded by isogene CG34172, allowing us o
knock down he ubiqui ously exp essed isogene CG9603 speci -
ically in neu ons using ela -GAL4, wi h minimal e ec s on
muscle. Secondly, we analysed he e ec s o AOX exp ession
in he le y
1
mu an in he gene encoding Cox6a, a subuni e-
qui ed o dime iza ion o cIV. The splice-si e mu a ion esul s
in a ameshi ea ly in he polypep ide sequence, and p oduces
an adul -onse neu odegene a i e pheno ype (19).
Neu onal deple ion o Cox7a d i en by ela -GAL4 esul ed
in locomo o dys unc ion (Fig. 5A) and seizu e sensi i i y a
298C (Fig. 5B), bo h o which we e pa ially escued by co-
exp ession o AOX. Nei he pheno ype was p oduced by
muscle-speci ic knockdown using he G14 d i e (Fig. 5A and
B), as expec ed, gi en ha Cox7a exp ession in muscle elies
mainly on isogene CG34172. Cox7a unc ion is also equi ed
du ing de elopmen (Fig. 3C), bu AOX- escue o he le hali y
ga e an adul pheno ype simila o ha seen in lies speci ically
knocked down o Cox7a and escued by AOX co-exp ession in
neu ons, namely a mild bu p og essi e locomo o de ec
(Fig. 5C) and seizu e sensi i i y a 298C, which was alsop og es-
si e (Fig. 5D). Howe e , he escued lies showed no mal su -
i al a 2 weeks. When ma ed o wild- ype lies o he opposi e
sex, males and emales we e bo h e ile and hei p ogeny
appea ed no mal. This con as s wi h he pheno ype exhibi ed
by lies knocked down o subuni s o cI, bu escued by Ndi1
(Supplemen a y Ma e ial, Fig. S7), which includes p onounced
de elopmen al delay, bang sensi i i y a oom empe a u e im-
media ely upon eclosion, and male s e ili y.
Ubiqui ous AOX exp ession using ei he ub-AOX (Fig. 6A)
o UAS-AOX d i en by da-GAL4 (Fig. 6B) also alle ia ed
he empe a u e-dependen seizu e sensi i i y exhibi ed by
le y
1
mu an lies, al hough e ec s on li espan a 298C we e
minimal (Supplemen a y Ma e ial, Fig. S8A–C). Exp ession
o UAS-Ndi1 in place o UAS-AOX did no alle ia e he seizu e
sensi i i y o le y
1
lies (Supplemen a y Ma e ial, Fig. S8D),
ins ead mildly exace ba ed i , simila o he obse a ion ha
Human Molecula Gene ics, 2014, Vol. 23, No. 8 2085
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om
Ndi1 exp ession enhanced he degene a i e e ec s o Cox5b
knockdown by ela -GAL4 (Fig. 6D and E).
Al hough le y
1
was p e iously sugges ed o be a null mu a ion
(o Cox6a), we also es ed he e ec s o Cox6a knockdown
by RNAi. Ubiqui ous knockdown was semi-le hal o emales
a pupal s age (Fig. 6C), al hough he deg ee o male le hali y
a ied be ween expe imen s. Eclosing lies ailed o in la e
hei wings and died in he ood sho ly a e eclosion. Co-
exp ession o UAS-AOX pa ially escued emale le hali y
(Fig. 6C, Supplemen a y Ma e ial, Fig. S8E), bu he esul ing
lies we e s ill weak. Knockdown o Cox6a using he muscle-
speci ic d i e G14 did no p oduce he unin la ed wings pheno-
ype (0 o 143 lies analysed), whe eas ela -GAL4 did so (223 o
237 lies). Co-exp ession o AOX escued his pheno ype, wi h
only 9 o 129 lies ailing o in la e wings a eclosion (Supple-
men a y Ma e ial, Fig. S8E). The Cox6a knockdown pheno ype
hus appea s o be neu onal and again pa ially escued by AOX.
The mo e se e e o ganismal pheno ype p oduced by Cox6a
knockdown han by he le y
1
splice-si e mu a ion in Cox6a
was e lec ed in a mo e se e e biochemical pheno ype as
de ec ed by BNE in-gel his ochemis y (Fig. 6D). BNE gels
e ealed a mul iplici y o complexes showing COX ac i i y, co -
esponding o monome ic and dime ic cIV, as well as supe com-
plexes ha we e no ully cha ac e ized. Knockdown o Cox6a
esul ed in a clea and ep oducible inc ease in he mobili y o
he monome ic complex, and a subs an ial dec ease in he abun-
dance o all mul ime ic and supe complexes. One o hese
comig a ed on gels wi h he majo complex exhibi ing cI ac i -
i y, and cI ac i i y based on his assay was also clea ly dec eased
in Cox6a knockdown lies, hough was ha dly a ec ed by he
le y
1
mu a ion.
Table 2shows a summa y o he di e en pheno ypes p o-
duced by knockdown o COX subuni s using a ious d i e s,
and hei alle ia ion by ub-AOX and/o UAS-AOX in he
expe imen s desc ibed abo e.
DISCUSSION
AOX can pa ially eplace he unc ions o COX in i o
Despi e a much lowe le el o exp ession han wi h he GAL4
sys em, AOX exp essed cons i u i ely unde he a- ubulin p o-
mo e was e ec i e in comba ing di e se insul s a ec ing cIV
in eg i y o ac i i y. Flies we e p o ec ed om cyanide oxici y
and om gene ic manipula ions o Cox6c (cyclope) o he cIV
assembly ac o Su 1, o app oxima ely he same ex en as
when AOX exp ession was d i en by ubiqui ously exp essed
GAL4 (8). These indings indica e, u he mo e, ha AOX
escue d i en by GAL4 o GeneSwi ch is no due o p omo e di-
lu ion e ec s, and hus alida e he use o GAL4 d i e s in he
emainde o he expe imen s epo ed he e.
The pheno ypes associa ed wi h knockdown o mos nuclea -
coded subuni s o cIV we e signi ican ly alle ia ed by ub-AOX
exp ession, including Cox4, Cox5a, Cox5b, Cox6a, Cox 6b and
Figu e5. AOXpa ially escuesneu onal pheno ypes esul ing om Cox7aknockdown.(AandC) Climbingindex and(Band D) bangsensi i i ya 298C o lieso he
indica ed geno ypes, ages and sex. Means +SD gene a ed by analysis o ba ches o .50 indi idual lies o each class. Whe e, indica ed, P,0.001, S uden ’s es ,
wo- ailed, unequal a iances, (A and B) compa ing lies knocked down o Cox7a (isogene CG9603) wi h and wi hou co-exp ession o AOXo (C and D) compa ing
lies o a gi en sex exp essing AOX, wi h o wi hou concomi an knockdown o Cox7a.
2086 Human Molecula Gene ics, 2014, Vol. 23, No. 8
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om
45. Dell’agnello, C., Leo, S., Agos ino, A., Szabadkai, G., Ti e on, C., Zulian,
A., P elle, A., Roube oux, P., Rizzu o, R. and Ze iani, M. (2007) Inc eased
longe i y and e ac o iness o Ca(2+)-dependen neu odegene a ion in
Su 1 knockou mice. Hum. Mol. Gene .,16, 431–444.
46. So¨ ensen, L., Eks and, M., Sil a, J.P., Lindq is , E., Xu, B., Rus in, P.,
Olson, L. and La sson, N.G. (2001) La e-onse co icohippocampal
neu odeple ion a ibu able o ca as ophic ailu e o oxida i e
phospho yla ion in MILON mice. J. Neu osci.,21, 8082–8090.
47. Du ou , E., Te zioglu, M., S e ky, F.H., So¨ ensen, L., Gal e , D., Olson, L.,
Wilbe z, J. and La sson, N.G. (2008) Age-associa ed mosaic espi a o y
chain de iciency causes ans-neu onal degene a ion. Hum. Mol. Gene .,17,
1418–1426.
48. Bu man, J.L., Yu, S., Poole, A.C., Decal, R.B. and Pallanck, L. (2012)
Analysis o neu al sub ypes e eals selec i e mi ochond ial dys unc ion in
dopamine gicneu ons om pa kinmu an s. P oc. Na l. Acad. Sci. USA,109,
10438–19443.
49. Humph ey, D.M., Pa sons, R.B., Ludlow, Z.N., Riemenspe ge , T.,
Esposi o, G., Ve s eken, P., Jacobs, H.T., Bi man, S. and Hi h, F. (2012)
Al e na i e oxidase escues mi ochond ia-media ed dopamine gic cell loss
in D osophila.Hum. Mol. Gene .,21, 2698–2712.
50. Massie, R., Wong, L.J. and Milone, M. (2010) Exe cise in ole ance due o
cy och ome b mu a ion. Muscle Ne e,42, 136–140.
51. Sanz, A., Fe na´ndez-Ayala, D.J., S e ana os, R.K. and Jacobs, H.T. (2010)
Mi ochond ial ROS p oduc ion co ela es wi h bu does no di ec ly egula e
li espan in D osophila.Aging,2, 200–223.
52. D’Au elio, M., Gajewski, C.D., Lenaz, G. and Man edi, G. (2006)
Respi a o y chain supe complexes se he h eshold o espi a ion de ec s in
human m DNA mu an cyb ids. Hum. Mol. Gene .,15, 2157–2169.
53. S e ana os, R. and Sanz, A. (2011) Mi ochond ial complex I: a cen al
egula o o he aging p ocess. Cell Cycle,10, 1528–1532.
54. Bahado ani, S., Cho, J., Lo, T., Con e as, H., Lawal, H.O., K an z, D.E.,
B adley, T.J. and Walke , D.W. (2010) Neu onal exp ession o a
single-subuni yeas NADH-ubiquinone oxido educ ase (Ndi1) ex ends
D osophila li espan. Aging Cell,9, 191–202.
55. O’B ien, S.J. and MacIn y e, R.J. (1972) The a-glyce ophospha e cycle in
D osophila melanogas e . I. Biochemical and de elopmen al aspec s.
Biochem Gene .,7, 141–161.
56. Da is, M.B. and MacIn y e, R.J. (1988) A Gene ic analysis o he
a-Glyce ophospha e oxidase locus in D osophila melanogas e .Gene ics,
120, 755–766.
57. Zhang, Y.Q., Roo e, J.,B ogna, S.,Da is, A.W., Ba bash, D.A., Nash, D. and
Ashbu ne , M. (1999) s ess sensi i e B encodes an adenine nucleo ide
anslocase in D osophila melanogas e .Gene ics,153, 891–903.
58. Toi onen, J.M., O’Dell, K.M., Pe i , N., I ine, S.C., Knigh , G.K.,
Leh onen, M., Longmui , M., Luo o, K., Tou aille, S., Wang, Z. e al. (2001)
echnical knockou ,aD osophila model o mi ochond ial dea ness.
Gene ics,159, 241–254.
59. Toi onen, J.M., Manji y, S., Tou aille, S., Alzia i, S., O’Dell, K.M. and
Jacobs,H.T. (2003)Genedosageand selec i eexp essionmodi ypheno ype
in a D osophila model o human mi ochond ial disease. Mi ochond ion,3,
83–96.
60. Jacobs, H.T., Fe na´ndez-Ayala, D.J., Manji y, S., Kemppainen, E.,
Toi onen, J.M. and O’Dell, K.M. (2004) Mi ochond ial disease in lies.
Biochim. Biophys. Ac a,1659, 190–196.
61. G aham, B.H., Li, Z., Alesii, E.P., Ve s eken, P., Lee, C., Wang, J. and
C aigen, W.J. (2010) Neu ologic dys unc ion and male in e ili y in
D osophila po in mu an s: a new model o mi ochond ial dys unc ion and
disease. J. Biol. Chem.,285, 11143–11153.
62. Oli a, M., De Pin o, V., Ba san i, P. and Caggese, C. (2002) A gene ic
analysis o he po in gene encoding a ol age-dependen anion channel
p o ein in D osophila melanogas e .Mol. Gene . Genomics,267, 746–756.
63. Ro io, A.T., Ma ching on, D.R.,Dona , S.,Schuppe, H.C.,Abel, J., F i sche,
E., Ellio , D.J., Laippala, P., Ahola, A.L., McNay, D. e al. (2001) Mu a ions
a he mi ochond ial DNA polyme ase (POLG) locus associa ed wi h male
in e ili y. Na . Gene .,29, 261–262.
64. Baklou i-Ga gou i, S., Gho bel, M., Chak oun, N., Sellami,A., Fakh akh, F.
and Amma -Keskes, L. (2012) The CAG epea polymo phism o
mi ochond ial polyme ase gamma (POLG) is associa ed wi h male
in e ili y in Tunisia. And ologia,44(Suppl. 1), 68–73.
65. Ruiz-Pesini, E., Lapen
˜a, A.C., Dı´ez-Sa´nchez, C., Pe´ ez-Ma os, A.,
Mon oya, J., Al a ez, E., Dı´az, M., U ie´s, A., Mon o o, L., Lo´pez-Pe´ ez,
M.J. and En ı´quez, J.A. (2000) Human m DNA haplog oups associa ed wi h
high o educed spe ma ozoa mo ili y. Am. J. Hum. Gene .,67, 682–689.
66. Piomboni, P., Foca elli, R., S enda di, A., Fe amosca, A. and Za a, V.
(2012) The ole o mi ochond ia in ene gy p oduc ion o human spe m
mo ili y. In . J. And ol.,35, 109–124.
67. Williams, S.L., Valno , I., Rus in, P. and Taanman, J.-W. (2004)
Cy och ome c oxidase subassemblies in ib oblas cul u es om pa ien s
ca ying mu a ions in COX10, SCO1, o SURF1. J. Biol. Chem.,279, 7462–
7469.
68. Shishido, E., Takeichi, M. and Nose, A. (1998) D osophila synapse
o ma ion: egula ion by ansmemb ane p o ein wi h Leu- ich epea s,
CAPRICIOUS. Science,280, 2118–2121.
69. Sykio is, G.P. and Bohmann, D. (2008) Keap1/N 2 signaling egula es
oxida i e s ess ole ance and li espan in D osophila.De . Cell,14, 76–85.
70. Zo dan, M.A., Ciso o, P., Benna, C., Agos ino, A., Rizzo, G., Piccin, A.,
Pego a o, M., Sand elli, F., Pe ini, G., Tognon, G. e al. (2006)
Pos - ansc ip ional silencing and unc ional cha ac e iza ion o he
D osophila melanogas e homolog o human Su 1. Gene ics,172, 229–
241.
71. Die zl, G., Chen, D., Schno e , F., Su, K.C., Ba ino a, Y., Fellne , M.,
Gasse , B., Kinsey, K., Oppel, S., Scheiblaue , S. e al. (2007) A
genome-wide ansgenic RNAi lib a y o condi ional gene inac i a ion in
D osophila.Na u e,448, 151–156.
72. Ba olo, S., Ca e , L.A. and Posakony, J.W. (2000) GFP and
be a-galac osidase ans o ma ion ec o s o p omo e /enhance analysis in
D osophila.Bio echniques,29, 726–732.
73. Kemppainen, E., Fe na´ndez-Ayala, D.J., Galb ai h, L.C., O’Dell, K.M. and
Jacobs, H.T. (2009) Pheno ypic supp ession o he D osophila
mi ochond ialdisease-likemu an ko(25 ) byduplica iono hemu an gene
in i s na u al ch omosomal con ex . Mi ochond ion,9, 353–363.
74. Miwa, S., S -Pie e, J., Pa idge, L. and B and, M.D. (2003) Supe oxide and
hyd ogen pe oxide p oduc ion by D osophila mi ochond ia. F ee Radic.
Biol. Med.,35, 938–948.
75. S e ana os, R., S i am, A., Ki i an a, E., Mohan,A., Ayala, V., Jacobs, H.T.,
Pamplona, R. and Sanz, A. (2012) dj-1b egula es oxida i e s ess,
insulin-like signaling and de elopmen in D osophila melanogas e .Cell
Cycle,11, 3876–3886.
76. Ross, J.M. (2011) Visualiza ion o mi ochond ial espi a o y unc ion using
cy och ome c oxidase/succina e dehyd ogenase (COX/SDH)
double-labeling his ochemis y. J. Vis. Exp. 2011;(57):e3266.
Human Molecula Gene ics, 2014, Vol. 23, No. 8 2093
a Tampe e Uni e si y Lib a y. Depa men o Heal h Sciences on Sep embe 27, 2016h p://hmg.ox o djou nals.o g/Downloaded om