Expression of alternative oxidase in Drosophila ameliorates diverse phenotypes due to cytochrome oxidase deficiency

Mitochondrial dysfunction is a significant factor in human disease, ranging from systemic disorders of childhood to cardiomyopathy, ischaemia and neurodegeneration. Cytochrome oxidase, the terminal enzyme of the mitochondrial respiratory chain, is a frequent target. Lower eukaryotes possess alternative respiratory-chain enzymes that provide non-proton-translocating bypasses for respiratory complexes I (single-subunit reduced nicotinamide adenine dinucleotide dehydrogenases, e.g. Ndi1 from yeast) or III + IV [alternative oxidase (AOX)], under conditions of respiratory stress or overload. In previous studies, it was shown that transfer of yeast Ndi1 or Ciona intestinalis AOX to Drosophila was able to overcome the lethality produced by toxins or partial knockdown of complex I or IV. Here, we show that AOX can provide a complete or substantial rescue of a range of phenotypes induced by global or tissue-specific knockdown of different cIV subunits, including integral subunits required for catalysis, as well as peripheral subunits required for multimerization and assembly. AOX was also able to overcome the pupal lethality produced by muscle-specific knockdown of subunit CoVb, although the rescued flies were short lived and had a motility defect. cIV knockdown in neurons was not lethal during development but produced a rapidly progressing locomotor and seizure-sensitivity phenotype, which was substantially alleviated by AOX. Expression of Ndi1 exacerbated the neuronal phenotype produced by cIV knockdown. Ndi1 expressed in place of essential cI subunits produced a distinct residual phenotype of delayed development, bang sensitivity and male sterility. These findings confirm the potential utility of alternative respiratory chain enzymes as tools to combat mitochondrial disease, while indicating important limitations thereof.

3 mapped and sequenced to verify the construction and the absence of mutations. Following microinjection into w 1118 embryos (VANEDIS Drosophila Injection Service, Oslo, Norway), transgenic progeny were established as independent lines in the w 1118 background. Insertion sites were determined by inverse PCR as described (8). Lines tub-AOX 35 (intergenic insertion on X-chromosome), tub-AOX 50 and tub-AOX 112 (both on chromosome 2), and tub-AOX 7 (chromosome 3) were retained for further study, maintained as homozygotes, and used to generate lines with multiple insertions on different chromosomes where needed for specific experiments. For further details on the insertions see Figure S1A-D.

Experimental crosses and developmental time
Crosses were conducted in triplicate, typically using 12 virgin females and 6 males, mated for 3 days before tipping to fresh vials for a further 2 days. Mean developmental time to eclosion at 25 °C was measured as described in (73). Tests of the effects of different RNAi constructs using various drivers were generally conducted using parental flies with balancer markers, except where behavioural tests were to be conducted, in which case appropriate parental flies were themselves constructed from balanced stocks, in order to avoid possible confounding effects of the balancers.

Behavioral, toxin resistance and lifespan assays
Bang-sensitivity was measured as described previously (73). For studying temperaturedependent bang-sensitivity, flies were maintained at 29 °C after eclosion, for the times indicated in figure legends. Climbing ability was assayed essentially as described in (8), with minor modifications: virgin females and males were separated into groups of 5 flies in test vials with a line marked at 6 cm. After tapping the flies to the bottom of the vial, the number of flies reaching the line in 10 s was recorded. This was repeated 3 times for each group of 5 flies, giving an average score between 0-5 (the climbing index). Ten sets of flies of each genotype and sex were tested in each experiment. Resistance to cyanide was tested as 4 described previously (8). Groups of 80-100 flies of each indicated genotype were tested in batches of 10 flies per vial, and the time to complete paralysis of all 10 flies was recorded.
For lifespan curves, virgin females and males were collected in sets of 10 flies per vial, and maintained at 25 °C (or 29 °C for levy 1 and levy-KD) in standard medium. Flies were transferred to fresh food vials three times a week, counting the number of dead flies with each transfer. At least 10 vials of each sex and genotype were used in life-span experiments. Shortterm survival assays were conducted similarly, but using only 3 vials of each sex and genotype.

Protein analysis
Mitochondria from adult flies and pupae were isolated as described (74), from batches of 80-100 flies ground in. 0.5 ml of isolation medium, with final resuspension in 50 l of isolation medium without BSA. For isolation of mitochondria from L3 larvae, 0.01 M freshly neutralized cysteine hydrochloride was added to the isolation buffer. Protein concentrations were measured using the Bradford assay. SDS-PAGE and Western blotting using antibodies against AOX, ATP synthase subunit , NDUFS3 and GAPDH were as described previously (8,10). Blue native polyacrylamide gel electrophoresis (BNE) and in-gel activity staining of mitochondrial enzymes were performed essentially as described in (54), using NativePAGE Novex Bis-Tris Gel System (Invitrogen) and batches of 100 µg of isolated mitochondria.
Staining for cI activity was for 20 min, for cIV activity for the times indicated in figure legends.

RNA extraction and Q-RT-PCR
RNA from Drosophila adults and larvae was extracted as described (8). cDNA synthesis, qRT-PCR and data analysis were performed as described for Surf1 mRNA in (8), using the following primer sets (all denoted 5´ to 3´): AOX S and AOX R for AOX (8)

Metabolic analyses
Cytochrome c oxidase activity was measured using the CYTOCOX1 kit (Sigma), according to manufacturer s recommended conditions. Polarography was performed using a highresolution Oroboros 2 K respirometer for whole-fly homogenates, as described previously (75), sequentially using the following substrate mixes (Sigma): for cI-driven substrate oxidation, 5 mM each sodium pyruvate and proline, plus 1 mM ADP; for G3PDH-driven substrate oxidation: 0.5 M rotenone followed by 20 mM glycerol-3-phosphate; for cIV-driven substrate oxidation: 2.5 M antimycin followed by 2 mM ascorbate plus 400 M TMPD,

Imaging
Imaging of the head and thorax was carried out on flies from which abdomen and wings had been removed prior to fixation in 4% formaldehyde for 3-5 h and paraffin embedding (Sakura Tissue-Tek VIP 4). 4 m sections (Microm HM310 rotation microtome) were attached to Thermo Scientific SUPERFROST Plus microscope slides at 37 °C overnight. Paraffin was removed using three xylene washes, followed by a descending ethanol series. Sections were analysed both by immunocytochemistry and by haematoxylin and eosin (H&E) to distinguish morphological differences in the brain. For H&E staining slides were first incubated in haematoxylin (Mayers, Merck, diluted 1:4 with distilled water) for 3 min, washed with running tap water for 10 min, then counter-stained in 0.1 % Eosin Y (Histola, with glacial acetic acid added) for 40s. Excess colour was rinsed off with distilled water and slides.
Stained sections were dehydrated in an ascending alcohol series, and cleared with xylene before mounting with organic mounting medium (EUKITT) and viewing with an Olympus BX 51 microscope and Olympus ColorView IIIu camera operated by Olympus Cell B software v. 2.4. For immunocytochemistry, slides were rinsed with TBS-Tween (TBST) before incubation for 30 min at 95 °C in 10 mM citrate buffer, 2.5 mM EDTA, 0.05% Tween 20, pH 6.2. The solution was cooled on ice to room temperature, after which slides were rinsed with TBST, incubated in 0.3 % Triton X-100 in TBST for 30 min, and washed 3 times for 5 min in TBST, all at room temperature. After blocking with 5 % bovine serum albumin (BSA) in TBST overnight at 4° C, primary antibodies were added in TBST containing 5 % BSA, and incubated for 1 h at room temperature. Primary antibodies were against COXIV (Abcam, ab16056, rabbit polyclonal, 1:200), ATP5A (Abcam, ab14748, mouse monoclonal, 1:1000) or AOX (21 st Century Biologicals, custom-made rabbit polyclonal PI047AB, 1:1000). Following 3 further 10 min washes in TBST, secondary antibodies (Invitrogen Goat-anti-rabbit Alexa fluor 568, A-11011 or Goat-anti-mouse Alexa fluor 488, A-10680, both at 1:1000 in TBST plus 5% BSA) were added for a further 1 h incubation at room temperature. Slides were washed again with TBST (3 times, 10 min) and a glass coverslip was mounted with ProLong Gold with DAPI (Invitrogen). Slides were imaged using a Spinning Disc Confocal microscope (Nikon Eclipse Ti, Wallac-Perkin Elmer Ultraview system), Andor EMCCD camera and iQ software.
COX and SDH staining (76) was carried out on cryosections, prepared by casting flies anaesthetized with CO 2 into Tissue-Tek O.C.T. mount (Sakura) and snap freezing in isopentane surrounded by liquid N 2 . Cryoblocks were cut into 5 µm sections with a Leica CM3050S cryo-microtome at -16-18 °C) and attached to Polylysine slides (Thermo Scientific), which were stored at -20°C then thawed for staining at room temperature. COX activity was visualized by incubating slides for 15 min in 9 ml of 0.05 M phosphate buffer pH 7.4, containing 5 mg 3,3´diaminobenzidine, 20 mg catalase, 10 mg cytochrome c and 750 mg of sucrose, followed by three washes with distilled water. SDH activity was visualized by incubating slides either prestained or not for COX, at 37 °C for 5 min in 10 ml 0.05 M phosphate buffer pH 7.4 containing 10 mg nitroblue tetrazolium and 540 mg sodium succinate, followed by three washes with distilled water. Stained sections were dehydrated in an ascending alcohol series, and cleared with xylene slides, before mounting and visualization as described above for H&E-staining. GFP expression patterns generated by various GAL4 drivers were determined by crossing homozygous driver and GFP reporter lines. Flies were left to mate in the dark for 1 d in mating chambers, in order to collect L1, L2 and L3 larvae. Larvae were rinsed in PBS, dried on tissue, then placed on microscope slides cooled to 4 ºC (Thermo-Scientific, Superfrost plus). A drop of 70 % glycerol was added, and larvae were frozen at -20°C 10 (L1) or 20 min (L2/L3). Whole larvae on frozen slides were imaged with an Olympus SZX16 microscope and ColorView IIIu camera, controlled by Olympus Cell B software (v. 2.4), under u.v. light (fluorescent lamp Olympus U-RFL-T). L3 larvae were dissected in PBS using Dumont #55 INOX forceps, on microscope slides with a recessed well and individual organs and tissues were removed for separate imaging. Pupal dissections were performed between stages 8-10, when there was visible eye pigment. Pupae were immobilized on a microscope slide with double-sided tape and the pupal case was gently removed using dissection forceps, beginning from the anterior and lifting the case away from the dorsal side, after which 70% glycerol was added and the whole pupa imaged. Pupae were dissected in PBS to isolate thoracic muscle and the CNS (brain and ventral nerve cord) for imaging. Adult flies were anaesthetized for 10-20 min in otherwise empty vials containing a cotton swab dipped in FlyNap (Carolina Biological Supply Company) Whole flies were imaged under fluorescent and bright light and then dissected. Before dissection flies were dipped into 96 % EtOH for 10 s to soften the cuticle for easier dissection, which was done in PBS, to isolate thoracic muscle, CNS and abdominal organs for individual imaging. Dissection of testes for imaging by phase-contrast microscopy was conducted on 2 day-old males anaesthetized on ice. For video imaging of sperm motility, dissected testes were punctured in PBS containing 1% BSA.
No image manipulation was done, other than standard brightness and contrast optimization.

Analysis of elav-GAL4-driven expression by immunocytochemistry and histochemistry
Comparable magnifications are shown, though some images have been resized and/or cropped for clarity. In some images signal has also been adjusted for brightness and contrast, but no gamma correction or any other manipulations were carried out.

B4
Ndi1 was able to complement the larval lethality of global knockdown of either of two singlecopy genes for cI subunits, namely Ndufa8 (mouse nomenclature: Drosophila gene CG3683) or Ndufa9 (Drosophila gene CG6020), driven by da-GAL4 (20). The rescued flies, in both cases, eclosed with a long developmental delay (3-5 d, panel A) and were mildly (but significantly) bang-sensitive when first tested at room temperature the day after eclosion (panel B). When mated with wild-type flies of the opposite sex, the rescued females were fertile, giving 24.5 + 5.1 and 17.8 + 6.1 progeny from individual females rescued from knockdown of CG3683 and CG6020, respectively. The resulting larvae appeared weak, and during wandering stage did not emerge more than 1-3 cm from the food before pupariation.
However, all pupae eclosed and the resulting adults appeared normal. In contrast, rescued males produced no progeny (>50 Ndi1-rescued males tested from knockdown of each gene).
Testis morphology of the rescued males appeared grossly normal (phase-contrast microscopy, panel C), although knockdown was clearly effective in testis, as judged by the levels of the cI marker protein NDUFS3 in Western blots (panel D). When testes from the Ndi1-rescued males were punctured by dissection needles, superficially mature sperm were seen, but were much less plentiful than from control males, and were immotile when exposed to standard buffer, whereas sperm from control males processed in parallel were motile. In (D) all flies carry the Ndi1 A46 transgene, but there is expression only where knockdown of the indicated genes is co-driven by da-GAL4. Molecular weights extrapolated from markers; the separate panels are from nonadjacent tracks of the same gel.