Summary
The electron transport chain (ETC) of mitochondria, bacteria, and archaea couples electron flow to proton pumping and is adapted to diverse oxygen environments. Remarkably, in mice neurological disease due to ETC complex I dysfunction is rescued by hypoxia through unknown mechanisms. Here, we show that hypoxia rescue and hyperoxia sensitivity of complex I deficiency is evolutionarily conserved to C. elegans and is specific to mutants that compromise the electron-conducting matrix arm. We show that hypoxia rescue does not involve the hypoxia-inducible factor pathway or attenuation of reactive oxygen species. To discover the mechanism, we use C. elegans genetic screens to identify suppressor mutations in the complex I accessory subunit NDUFA6/nuo-3 that phenocopy hypoxia rescue. We show that NDUFA6/nuo-3(G60D) or hypoxia directly restore complex I forward activity, with downstream rescue of ETC flux and, in some cases, complex I levels. Additional screens identify residues within the ubiquinone binding pocket as being required for the rescue by NDUFA6/nuo-3(G60D) or hypoxia. This reveals oxygen-sensitive coupling between an accessory subunit and the quinone binding pocket of complex I that can restore forward activity in the same manner as hypoxia.
In Brief
C. elegans mutants harboring a defective complex I are rescued by hypoxia or intra-complex genetic suppressor mutation, achieved by increasing forward flow of electrons via structural or chemical changes occurring in the CoQ binding pocket.
Graphical Abstract
Introduction
Complex I (NADH:ubiquinone oxidoreductase) is the primary entry point of electrons into the ETC, using NADH to reduce flavin and iron-sulfur clusters in proteins along a path to ubiquinone, and harnessing this electromotive force to pump protons across the mitochondrial inner membrane1,2. Complex I consists of 14 core subunits evolutionarily conserved from bacteria to mammals as well as dozens of accessory (or “supernumerary”) subunits that perform structural and/or regulatory roles, organized into functional modules (N, Q, Pp and Pd) corresponding to assembly intermediates3,4. Loss of complex I activity has numerous consequences for the cell, including decreased ATP production, loss of mitochondrial membrane potential, increased NADH/NAD+ ratio, and accumulation of excess, unused molecular oxygen. Deficiencies in complex I of the ETC underlie approximately 30% of mitochondrial diseases including Leigh syndrome, Leber’s hereditary optic neuropathy, and MELAS – devastating human disorders for which there are no approved medicines5–7. Loss of complex I activity is also associated with more common forms of neurodegeneration and even certain rare cancers8–10.
A mouse model of Leigh syndrome caused by homozygous deletion of the complex I subunit Ndufs4 is rescued by environmental hypoxia (11% oxygen), which extends lifespan, prevents neurodegeneration, and reverses late-stage neurological disease11,12. Conversely, exposure of the Ndufs4 knockout mouse to moderate hyperoxia (55% oxygen) exacerbates the disease, causing rapid death. Ndufs4 mice experience tissue hyperoxia in the brain at 21% oxygen, presumably due to decreased oxygen consumption by the ETC; interventions that lower brain oxygen levels rescue neurological disease and extend lifespan13. These findings indicate that oxygen is central to the neuropathology, but the precise mechanism underlying the complex I rescue by hypoxia and sensitivity to hyperoxia has not been established. Moreover, whether these phenomena are evolutionarily conserved and translate to other in vivo models of complex I deficiency is unknown.
Caenorhabditis elegans is an ideal system to dissect the genetic-environmental interaction of complex I dysfunction and oxygen. Nematode and mammalian complex I are highly homologous, with at least 42 of the 45 mammalian complex I subunits conserved14. Years of C. elegans genetic analysis of energy metabolism, anesthetic sensitivity, statin resistance, and longevity have generated a collection of viable complex I mutants15–18. C. elegans is commonly isolated from rotting fruit and vegetation on which abundant and diverse oxygen-consuming bacteria flourish19. Wild-type C. elegans prefer oxygen levels of 7–8%20, perhaps because hypoxia is correlated with bacterial nutrition. However C. elegans are naturally tolerant of a wide range of oxygen tensions, capable of reproducing and maintaining their metabolic rate from 1% to 100% oxygen21. Some C. elegans mitochondrial mutants are hypersensitive to elevated oxygen levels22–24, consistent with the Ndufs4 mutant mouse findings.
Here, we show that hypoxia rescue of complex I deficiency is evolutionarily conserved to C. elegans, and report that only a subset of complex I mutations are rescued by hypoxia and sensitive to moderate hyperoxia. These mutations partially compromise the soluble matrix arm of complex I (N and Q modules) that normally transfers electrons from NADH to CoQ. The rescue by hypoxia is neither dependent on the canonical oxygen-sensing hypoxia-inducible factor (HIF) signaling pathway, nor do mitochondrial reactive oxygen species (ROS) underlie the rescue by hypoxia or sensitivity to hyperoxia. Using C. elegans forward genetic selections, we identify intra-complex amino acid substitution mutations in accessory subunits NDUFA6/nuo-3 and NDUFA5/ndua-5 that phenocopy the suppression of complex I mutants by hypoxia. Through biochemical studies we show that NDUFA6/nuo-3(G60D) or hypoxia partially restore complex I forward electron transport activity, with downstream rescue of ETC flux and, in some cases, complex I levels. Additional C. elegans genetic screens identify NDUFS7/NDUF7 and NDUFS2/GAS-1 amino acid residues surrounding the ubiquinone binding pocket that are necessary for the rescue of complex I mutants by NDUFA6/nuo-3(G60D) or hypoxia. These results suggest that mutants harboring a defective complex I are rescued by increasing forward flow of electrons to CoQ, achieved through structural changes in the CoQ binding pocket.
Results
Hypoxia rescues a subset of complex I mutations in C. elegans
To determine if C. elegans ETC mutants are rescued by hypoxia, we exposed wild-type or mutant L1 stage animals to 21% or 1% oxygen and measured their rate of growth and development. C. elegans homozygous for a deletion in NDUFS4/lpd-5 encoding a supernumerary subunit of complex I arrested development in normoxia at an early larval stage, but animals at 1% oxygen developed beyond this arrest point to sterile adulthood (Fig 1A). This is consistent with hypoxia rescue of the mouse Ndufs4 knockout. However complete loss of complex I activity cannot be rescued by hypoxia, as deletions in subunits NDUFS3/nuo-2, NDUFS1/nuo-5, NDUFS7/nduf-7, or NDUFA9/nduf-9 caused early developmental arrest at 21% oxygen with no improvement at 1% oxygen (Fig 1B). Chemical inhibition of complex I activity with the plant natural product rotenone also caused developmental delay that was not rescued by hypoxia (Fig S1A). These results demonstrate that rescue of particular complex I deficiencies by hypoxia is evolutionarily conserved but does not generalize to all complex I lesions.
Figure 1. A subset of complex I mutations are rescued by hypoxia independent of HIF.
A. Growth of animals for 2 days (left) and 5 days (right) at 21% or 1% oxygen at room temperature. B. Growth of wild type and nuo-6(qm200) animals for two days; growth of nuo-2(tm5258), nuo-5(tm2751), nduf-7(tm1436), and nduf-9(mg747) animals for 4 days at room temperature. C. Ovine NADH:ubiquinone oxidoreductase, or complex I (PDB: 6ZKC43) in closed conformation. Colored red (on right) are the 6 C-terminal amino acids deleted in NDUFS7/nduf-7(et19) that interact with NDUFA9/NDUF-9 (yellow), phospholipid headgroups (pink), and NADPH (purple). Colored red (on left) is NDUFS2/GAS-1(R290) which is substituted to K in NDUFS2/gas-1(fc21) and interacts with conserved residues E200 and R260 of ND1/NDUO-1 (grey). All NDUFS4/LPD-5 (orange) is lost in lpd-5(mg746). D. Growth of animals for 2 days at room temperature. E. Fluorescent images of age-matched L4 stage animals containing hsp-6::gfp grown at 21% or 1% oxygen for one generation (left). Images were acquired at 69x magnification with an exposure time of 100 ms. Mean intestinal fluorescence of hsp-6::gfp in pictured animals (right). F. Growth of wild type, mev-1(kn1), and clk-1(qm30) animals for 2 days; growth of isp-1(qm150) animals for 3 days. The same wild type controls are used in panel 1D as the data were collected in the same experiment. G. Growth of animals for 3 days at 20°C. H-J. Growth of animals for 2 days (H), 5 days (I), or for 4 days (J) at room temperature. Statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001. See also Figure S1 and Table S1.
To further understand which complex I mutants are amenable to rescue by hypoxia, we tested additional mutations in the electron-conducting matrix arm. Animals carrying a partial loss-of-function allele in the complex I core subunit NDUFS7/nduf-7(et19) are viable, have reduced complex I activity, and grow slowly at 21% oxygen18. The NDUFS7/nduf-7(et19) mutation eliminates the last 6 amino acids of NDUF-7 which lie on the surface of complex I and are likely to stabilize interactions with the supernumerary subunit NDUFA9/NDUF-9, phospholipid headgroups, and non-catalytic NADPH (Figs 1C & S1B). In contrast to an nduf-7 deletion allele, the nduf-7(et19) mutant is partially rescued by hypoxia, growing faster at 1% oxygen (Fig 1D). Expression of the mitochondrial stress reporter hsp-6::gfp – a fusion to the promoter of an HSP70 mitochondrial chaperone and readout of mitochondrial membrane potential and protein import efficiency25,26 – was elevated in nduf-7(et19) animals grown at 21% oxygen, but attenuated in animals grown at 1% oxygen, suggesting that hypoxia restores mitochondrial function (Fig 1E). Another hypomorphic mutation in the complex I Q module, NDUFS2/gas-1(fc21), is an R290K substitution that causes reduced complex I activity, hypersensitivity to volatile anesthetics, and low broodsize15,16. NDUFS2/GAS-1(R290) is a highly conserved residue that may form stabilizing interactions with the mitochondrial-genome-encoded membrane subunit ND1 (Figs 1C & S1B). While hypoxia did not significantly rescue gas-1(fc21) growth rate (data not shown), possibly due to a mild phenotype at 21% oxygen, gas-1(fc21) strongly induced hsp-6::gfp at 21% oxygen, and this induction was decreased in 1% oxygen (Figs 1E & S1C), suggesting that the gas-1(fc21) complex I mutant is also partially rescued by hypoxia.
We tested if the slow growth of complex I mutants in the proton-pumping membrane arm was also suppressed by hypoxia. A partial loss-of-function allele of the membrane supernumerary subunit NDUFB4/nuo-6(qm200) with low complex I activity17 did not grow faster at 1% oxygen (Fig 1B). nuo-6(qm200) also induced hsp-6::gfp expression and this was not altered by 1% oxygen, further suggesting that this mutant is not rescued by hypoxia (Fig S1C). This also demonstrates that the hsp-6::gfp reporter is not basally affected by 1% oxygen independent of complex I, consistent with decreased hsp-6::gfp fluorescence in gas-1(fc21) and nduf-7(et19) animals exposed to hypoxia reflecting increased complex I activity. Similarly, hsp-6::gfp levels were elevated in NDUFS4/lpd-5(mg746) and NDUFS3/nuo-2(tm5258) null mutants, but partially suppressed by hypoxia only in lpd-5(mg746) (Fig S1C). Viable, slow-growing C. elegans mutants in other components of the ETC including mev-1(kn1)/complex II, clk-1(qm30)/CoQ biosynthesis, and isp-1(qm150)/complex III also exhibited a growth delay that was not rescued by hypoxia (Fig 1F). Taken together, these results indicate that a subset of complex I mutants can be rescued by hypoxia – specifically those that partially compromise the soluble portion (N and Q modules) of complex I responsible for passing electrons from NADH to CoQ (Table S1).
A set of fungal species that includes S. cerevisiae have lost the multisubunit, proton pumping complex I and instead harbor NDI1, a mitochondrial inner membrane type II NADH dehydrogenase27. This single polypeptide catalyzes two-electron transfer from NADH to CoQ without pumping protons and can rescue the survival and growth defects of complex I mutant mammals or C. elegans28–30. Expression of yeast NDI1 rescued the slow growth of lpd-5(mg746), nduf-7(et19), and gas-1(fc21) mutant C. elegans at 21% oxygen (Figs 1G–H). NDI1 could also suppress the growth arrest of complex I mutants nduf-9(mg747) and nuo-6(qm200) which were not rescued by hypoxia, demonstrating that their growth defect is indeed due to complex I deficiency (Fig S1D). Interestingly, while hypoxia improved the growth of lpd-5(mg746) and nduf-7(et19) mutants, it did not further benefit these complex I mutants when NDI1 was expressed despite a dynamic range sufficient for additivity (Figs 1H & S1E). This is consistent with NDI1 activity covering the same specific defect rescued by hypoxia (i.e. passing electrons from NADH to CoQ); alternatively hypoxia may have no effect in these genetic backgrounds due to NDI1-expressing animals having a liability in hypoxia.
Hypoxia rescue of C. elegans complex I mutants is independent of the HIF transcriptional response
To determine if hypoxia rescues through the canonical HIF/EGLN transcription factor-mediated oxygen-sensing pathway, we constructed double mutants with complex I subunits and components of the HIF signaling pathway. VHL-1 is an E3 ubiquitin ligase that targets EGLN-mediated hydroxylated HIF for degradation31. VHL-1 null mutants constitutively activate the hypoxia response even at normoxia. Neither lpd-5(mg746) nor nduf-7(et19) – mutants whose growth is rescued by hypoxia – were improved by loss of vhl-1 at 21% oxygen (Figs 1I & S1F). In fact, the nduf-7; vhl-1 and lpd-5; vhl-1 double mutants grew more slowly than the complex I single mutants. This is consistent with observations in the Ndufs4 mouse13, and suggests activation of the HIF hypoxia response is not sufficient to rescue the slow growth of complex I mutants. We wondered if activation of HIF may be beneficial in the context of hypoxia where its response is physiologically adaptive, but at 1% oxygen VHL-1 loss was also detrimental to the lpd-5 and nduf-7 mutants (Figs 1I & S1F). C. elegans contains only one HIF homolog, HIF-1, and HIF-1 null mutants are viable. This allowed us to ask if the HIF hypoxia response was necessary for the rescue by hypoxia. Despite hif-1 mutants themselves being sensitive to hypoxia, the nduf-7; hif-1 and lpd-5; hif-1 double mutants were both partially rescued by 1% oxygen (Figs 1J & S1G), demonstrating that the HIF transcriptional response to hypoxia is not necessary for the hypoxia rescue of complex I mutants.
Complex I mutants rescued by hypoxia are also sensitive to 50% oxygen
To further characterize the interaction between complex I deficiency and oxygen tension, we exposed viable, slow-growing ETC mutants to 50% and 100% oxygen (hyperoxia). Multiple C. elegans mitochondrial mutants were sensitive to 100% oxygen, including mutants in complex I, complex II, CoQ biosynthesis, and complex III (Figs 2A & S2A), as observed previously22–24. 100% oxygen also caused wild type animals to develop slowly and induced the hsp-6::gfp mitochondrial stress reporter, indicating mitochondrial dysfunction (Figs 2A–B & S2A). However at 50% oxygen the growth of wild type animals was not delayed, the hsp-6::gfp mitochondrial stress reporter was not induced, and ETC mutants in complex II, CoQ biosynthesis, and complex III were able to develop to fertile adulthood (Figs 2A–B & S2A). The only viable mutants that arrested development at 50% oxygen were the complex I mutants NDUFS7/nduf-7(et19) and NDUFS2/gas-1(fc21), which arrested at the L1 stage and L2 stage, respectively (Fig 2A). nduf-7(et19) and gas-1(fc21) mutants recovered from at least six days exposure to 50% oxygen when shifted to 21% oxygen (Fig S2B), suggestive of a reversible developmental arrest and reminiscent of the ability of advanced brain disease in Ndufs4 knockout mice to be reversed by hypoxia12. The sensitivity of nduf-7(et19) and gas-1(fc21) to 50% oxygen was rescued by expression of yeast NDI1 (Fig S2C), indicating that the vulnerability to hyperoxia arises from impairment of complex I oxidation of NADH or reduction of ubiquinone. Notably, the slow-growing complex I mutant NDUFB4/nuo-6(qm200) which was not rescued by hypoxia is also not sensitive to hyperoxia (Fig S2A). Taken together, these results suggest a mechanistic link between growth rescue by hypoxia and sensitivity to moderate hyperoxia (Table S1).
Figure 2. Complex I mutants rescued by hypoxia are sensitive to moderate hyperoxia and suppressed by intra-complex mutations in NDUFA6 or NDUFA5.
A. Growth of animals following L1 synchronization at 21% oxygen (black), 50% oxygen (orange), or 100% oxygen (red) incubated at 20°C. B. Mean intestinal fluorescence of hsp-6::gfp in L4 stage animals incubated at 21%, 50%, or 100% oxygen for 1 day at 20°C. Exposure time = 100 ms, magnification = 69x. C. gas-1(fc21) or nduf-7(et19) P0 animals were randomly mutagenized with EMS at 21% oxygen. F2 progeny were transferred to the non-permissive 50% oxygen tension and selected for growth. NDUFS4/lpd-5 null C. elegans are not fertile at any oxygen tension, making a forward genetic screen challenging. D. Ovine complex I (PDB: 6ZKC43) in closed conformation. Colored red is the ovine residue (K57) corresponding to C. elegans suppressor mutation NDUFA6/nuo-3(G60D) which lies in the LYRM domain responsible for binding the acyl chain (purple) of NDUFAB1. E. Growth of animals for 4 days at 50% oxygen followed by 1 day at 21% oxygen. F. Growth of animals for 2 days (graph, left) and 4 days (images, right) at room temperature. G. Mean intestinal fluorescence of hsp-6::gfp in L4 stage animals incubated at continuous 21% oxygen at 20°C. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001. See also Figure S2 and Table S2.
Oxygen-sensitive complex I mutants are suppressed by intra-complex mutations in NDUFA6 and NDUFA5
Based on these C. elegans genetic results and the extreme sensitivity to moderate hyperoxia of the Ndufs4 mouse11, we sought to identify C. elegans genetic suppressor mutations of the hyperoxia sensitivity in complex I mutants. To isolate such mutations, we performed two parallel forward genetic screens for suppressors of the nduf-7(et19) or gas-1(fc21) growth arrest at 50% oxygen. C. elegans animals were mutagenized with a DNA alkylating agent and grown at 21% oxygen for two generations to generate hundreds of thousands of randomly distributed new mutant alleles for genetic selection. This large collection of F2 generation animals was transferred to the non-permissive 50% oxygen tension as synchronized L1 animals and screened for rare mutants that could grow to adulthood. The nuclear and mitochondrial genomes of these suppressor mutants were then deep-sequenced (Fig 2C). As an endorsement of the depth of the selection, each of these screens identified intragenic amino acid substitutions in nduf-7 or gas-1 which acted as revertants of the original nduf-7(et19) or gas-1(fc21) mutations (Table S2). More importantly, the screens identified three independent dominant alleles of the complex I subunit NDUFA6/nuo-3, all of which encode G60D missense mutations, and one allele in the complex I subunit NDUFA5/ndua-5 that encodes an R126Q missense mutation (Table S2). The nuo-3(G60D) and ndua-5(R126Q) intra-complex suppressors lie in accessory subunits on opposite sides of the interface between the matrix and membrane domains of complex I (Figs 2D & S2D).
To prove that these candidate lesions caused the suppression of hyperoxia arrest, we used CRISPR/Cas9 to generate a nuo-3(G60D) allele in a wild-type genetic background, and then introduced this mutation into complex I mutants. The CRISPR-generated nuo-3(G60D) allele suppressed the nduf-7(et19) and gas-1(fc21) sensitivity to hyperoxia, confirming nuo-3(G60D) as the causative mutation in our screen (Fig 2E). We also found that nuo-3(G60D) is an excellent suppressor of the NDUFS4/lpd-5 null mutant, allowing the animals to develop to sterile adulthood at 21% oxygen much like the rescue by hypoxia (Fig 2F). Consistent with nuo-3(G60D) acting dominantly, the effects of hypoxia and nuo-3(G60D) on the NDUFS4/lpd-5(mg746) mutant were additive with respect to growth rate (Fig 2F), and more strikingly the lpd-5(mg746); nuo-3(G60D) double mutant in 1% oxygen was a non-sterile and viable strain able to generate progeny (Fig S2E). A similar CRISPR-based approach confirmed NDUFA5/ndua-5(R126Q) as a genuine complex I suppressor of nduf-7(et19) and gas-1(fc21) at 50% oxygen, and lpd-5(mg746) at 21% and 1% oxygen (Figs S2F–H). These results show that mutations in two distinct complex I genes can each suppress three distinct complex I lesions, suggesting they exert a general protective effect on complex I. We focus on the nuo-3(G60D) mutation that conferred the stronger suppression phenotype for the remainder of this study.
nuo-3(G60D) caused a partial reduction of hsp-6::gfp induction in the nduf-7(et19) and gas-1(fc21) mutants, suggesting an alleviation of mitochondrial stress through restored ETC activity (Fig 2G). Notably nuo-3(G60D) did not reduce hsp-6::gfp expression in wild type animals exposed to 100% oxygen, arguing against a protective effect on oxygen toxicity more broadly, and pointing to suppression of specific complex I mutants (Fig 2B). Although nuo-3(G60D) was able to suppress nduf-7(et19), gas-1(fc21), and lpd-5(mg746) animals, nuo-3(G60D) did not suppress other complex I mutants such as NDUFA9/nduf-9(mg747) and NDUFB4/nuo-6(qm200) at any oxygen tension (Figs S2I–J). nuo-3(G60D) also did not suppress slow-growing ETC mutants in complex II, CoQ biosynthesis, and complex III at any oxygen tension (Figs S2K). The emerging pattern is that the genetic requirements for rescue by 1% oxygen, sensitivity to 50% oxygen, and rescue by nuo-3(G60D) are identical, and we hypothesize that common mechanisms underlie these phenomena (Table S1).
Mitochondrial ROS does not underlie the mutant rescue by hypoxia or nuo-3(G60D)
Excess molecular oxygen can undergo partial reduction to form reactive oxygen species (ROS), and C. elegans mitochondrial mutants are sensitive to exogenous forms of oxidative stress23,32. Therefore, we measured steady-state levels of mitochondrial ROS in our mutants using MitoSOX, a dye targeted to the mitochondria that is oxidized by superoxide to produce red fluorescence. gas-1(fc21) and nduf-7(et19) displayed wild-type or decreased MitoSOX staining at 21% and 50% oxygen (Fig 3A), consistent with prior reports33,34. However, because mitochondrial uptake of MitoSOX is dependent on mitochondrial membrane potential, which is compromised in complex I mutants, we sought to validate this finding with orthogonal approaches. First, we analyzed expression of gst-4::gfp, a reporter for the NRF2-mediated antioxidant response, and found that its expression was not activated in the nduf-7(et19) and gas-1(fc21) mutants (Fig S3A). Exposure to 100% oxygen did activate the gst-4::gfp reporter in wildtype, but induction in the complex I mutant backgrounds was comparable. Second, we isolated mitochondria and performed western blots with an antibody against 4-hydroxynonenal (4-HNE), a product of lipid peroxidation that can bond covalently to lysine residues through Schiff base formation35. Consistent with the MitoSOX results, nduf-7(et19) displayed dramatically decreased 4-HNE staining, which was normalized by nuo-3(G60D) at both 21% and 1% oxygen (Fig 3B). Third, we directly measured ROS produced from complex I in purified mitochondrial membranes. In this in vitro assay NADH-driven superoxide production from complex I is converted to hydrogen peroxide which is used to oxidize Amplex Red. Both nduf-7(et19) and gas-1(fc21) displayed decreased ROS production by complex I, and nuo-3(G60D) increased ROS production in the gas-1(fc21) background (Fig 3C). Finally, we introduced transgenes over-expressing mitochondrial-localized superoxide dismutase SOD-2, which detoxifies superoxide into hydrogen peroxide, and observed no rescue of nduf-7(et19) or gas-1(fc21) slow growth rate at 21% or 50% oxygen (Fig 3D). Over-expression of SOD-2 in combination with mitochondrial-localized catalase, which can detoxify SOD-generated hydrogen peroxide, also produced no growth benefit in the gas-1(fc21) mutant (Fig S3B). These results argue against decreased mitochondrial ROS underlying the rescue by hypoxia, because mitochondrial ROS levels were not elevated in the complex I mutants and over-expression of ROS-detoxifying enzymes had no benefit.
Figure 3. Rescue of complex I mutants by hypoxia and nuo-3(G60D) is not due to alleviation of mitochondrial ROS.
A. MitoSOX fluorescence quantified by measuring the mean fluorescence in the posterior bulb of the pharynx after 1 day. Images were taken with an exposure time of 1 second at 63x magnification. B. SDS-PAGE western blot of isolated mitochondria purified from animals grown at 21% or 1% oxygen for 4 days. The mitochondrial proteins modified by 4-HNE in this experiment are unidentified. C. NADH-driven complex I-dependent superoxide production in the absence of piericidin by isolated mitochondrial membranes at 21% oxygen. Superoxide is converted by SOD to hydrogen peroxide, which then oxidizes AmplexRed to Resorufin via HRP. Resorufin absorbs light at 557 nm. D. Growth of animals for 2 days at room temperature at 21% or 50% oxygen. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001. See also Figure S3.
Complex I levels are compromised in oxygen-sensitive mutants but not necessarily rescued by hypoxia or nuo-3(G60D)
We sought to understand what biochemical features are shared between the complex I mutants amenable to rescue by hypoxia or nuo-3(G60D), and whether such features underlie this rescue. First, we purified mitochondria from nduf-7(et19) and gas-1(fc21) mutants and profiled their OXPHOS complexes using Blue Native PAGE (BN-PAGE). We observed a dramatic loss of assembled complex I in the nduf-7(et19) mutant at all oxygen tensions, which was not rescued by nuo-3(G60D) or hypoxia (Figs 4A–B & S4A). A faint new band running below the complex V dimer also appears in nduf-7(et19) (Fig S4A). Native gel followed by western blot for HA-tagged NDUFAB1/NDAB-1 revealed subcomplexes in nduf-7(et19) that may correspond to complex I degradation products (Fig S4B). Similarly, loss of Ndufs4 in mice causes a fragile complex I prone to degradation36,37. The gas-1(fc21) mutant also showed a pronounced loss of assembled complex I which was exacerbated at 50% oxygen, and nuo-3(G60D) rescued levels of assembled complex I in this mutant (Figs 4A–B & S4C). These results suggest that complex I instability may be a common trait in mutants whose growth is modified by environmental oxygen, but because nduf-7(et19) growth rate is rescued by hypoxia and nuo-3(G60D) without restoring the levels of assembled complex I, we further investigated the specific effects of nuo-3(G60D) on complex I levels and activity.
Figure 4. Complex I levels are compromised in oxygen-sensitive mutants but their restoration does not underlie the rescue by hypoxia or nuo-3(G60D).
A. BN-PAGE of isolated mitochondria followed by western blot from animals grown continuously at 21% oxygen. B. BN-PAGE of isolated mitochondria followed by complex I in-gel dehydrogenase activity assay from animals exposed to 50% oxygen for 1 day. C-D. TMT quantitative proteomics from animals exposed to 50% oxygen for 2 days. Plotted are log 2-fold ratios of all proteins from which at least two peptides were quantified. Complex I subunits are colored according to structural module4. E. TMT quantitative proteomics from animals exposed to 50% oxygen for 2 days. Plotted are log 2-fold ratios of all complex I N module (orange) and Q module (yellow) subunits from which at least two peptides were quantified. F-G. SDS-PAGE followed by western blot of whole worm lysate from animals exposed to 21%, 50%, or 1% oxygen. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001. See also Figure S4 and Table S3.
Concomitant with assembly defects, loss of complex I accessory subunits can lead to degradation of other subunits within the same structural module3. We performed quantitative proteomics using isobaric tags to assess how the C. elegans complex I mutations and the nuo-3(G60D) suppressor mutation affect levels of individual complex I proteins. We quantified over 7,000 proteins, including 36/42 subunits of C. elegans complex I. In concordance with our BN-PAGE observations (Figs 4A–B), multiple complex I subunits were lost in nduf-7(et19) and gas-1(fc21) at 50% oxygen, while subunits of ETC complexes II-V were unchanged (Figs 4C–D & S4D–E). In particular, Q module and N module complex I subunits were depleted, consistent with structural defects in the soluble arm of complex I. Introduction of the nuo-3(G60D) suppressor mutation rescued the loss of Q and N module proteins in the gas-1(fc21) mutant at 50% oxygen, but not in the nduf-7(et19) mutant, in line with BN-PAGE results (Fig 4E). We validated these high throughput results with SDS-PAGE western blots against Q module subunits NDUFS3/NUO-2 and NDUFS6/NDUF-6. At 21% and 50% oxygen we observed loss of these complex I subunits in nduf-7(et19) and gas-1(fc21) which was rescued by nuo-3(G60D) only in gas-1(fc21) (Figs 4F–G). Notably, the effects of hypoxia treatment on complex I mutants nduf-7(et19) and gas-1(fc21) phenocopied the nuo-3(G60D) mutant: NDUFS3/NUO-2 and NDUFS6/NDUF-6 levels were rescued by 1% oxygen in the gas-1(fc21) mutant but not in nduf-7(et19) (Fig 4G). The identical effects on complex I levels by hypoxia and nuo-3(G60D) further support a shared mechanism of action. Given that hypoxia and nuo-3(G60D) improve the growth and development of nduf-7(et19) without a restoration of complex I levels, the underlying rescue mechanism is unlikely to be a restoration of complex I stability.
Loss of complex I levels may reflect a general dysfunction in iron-sulfur (Fe-S) cluster synthesis or instability of Fe-S containing proteins, two phenomena that are sensitive to excess molecular oxygen38,39. To address this possibility, we measured steady-state Fe-S clusters in C. elegans complex I mutants. Lipoic acid is a protein modification generated by lipoic acid synthetase which requires mitochondrially-produced Fe-S clusters. Neither gas-1(fc21) nor nduf-7(et19) showed any deficit in lipoic acid levels at 1%, 21%, or 50% oxygen (Fig S4F), indicating that steady-state Fe-S levels were not compromised in these mutants. We also analyzed the levels of 50 Fe-S-containing proteins in our TMT proteomics data and observed no general decrease of these proteins in the gas-1(fc21) and nduf-7(et19) mutants, apart from complex I subunits (Figs S4G–H). This data argue against hypoxia rescuing complex I mutants through a general effect on Fe-S cluster synthesis or stability, and support our model that hypoxia acts specifically on complex I.
NDUFA6/nuo-3(G60D) and hypoxia rescue complex I forward activity in oxygen-sensitive mutants
To determine if nuo-3(G60D) restores forward electron flow through the ETC, we directly measured the redox state of C. elegans coenzyme Q9 using mass spectrometry. Validating this approach, complex I mutants showed diminished QH2/Q ratios, while the complex III (CoQ:Cytochrome c oxidoreductase) mutant isp-1(qm150) showed a ratio of QH2/Q that trended higher (Figs 5A & S5A). Additionally, the endogenous CoQ biosynthesis mutant clk-1(qm30) produced no detectable Q9 or Q9H2 (Fig S5B), demonstrating the specificity of this assay. The nuo-3(G60D) suppressor mutation rescued CoQ redox potential in complex I mutants nduf-7(et19) and gas-1(fc21) (Fig 5A). We also measured mitochondrial membrane potential using the fluorescent dye TMRM. nduf-7(et19) and gas-1(fc21) showed diminished TMRM fluorescence, which was partially rescued by the nuo-3(G60D) suppressor, consistent with an increase in proton pumping by the ETC (Fig 5B). Taken together, along with the rescue of hsp-6::gfp induction (Fig 2G) which is a readout of mitochondrial membrane potential and protein import efficiency, these experiments demonstrate rescue of forward electron flow through the ETC by nuo-3(G60D), supporting the hypothesis that complex I activity is restored by this intra-complex mutation.
Figure 5. NDUFA6/nuo-3(G60D) and hypoxia rescue complex I forward activity in oxygen-sensitive mutants.
A. Ratio of reduced CoQ9H2 to oxidized CoQ9 as determined by mass spectrometry. Samples were extracted from whole worms grown continuously at 21% oxygen. B. TMRM fluorescence quantified by measuring the mean fluorescence in the posterior bulb of the pharynx after 1 day at 21% oxygen. Images were taken with an exposure time of 20 ms at 63x magnification. C-E. Complex I-dependent oxidation of NADH by isolated mitochondrial membranes in vitro. NADH absorbs light at 340 nm. Plotted (at right) are rates of absorbance at 340 nm (minus 380 nm) per minute from the first 20 minutes of linear slopes. F. SDS-PAGE followed by western blot of whole worm lysate from animals exposed to 21% oxygen. Ti[NDI1] animals contain a single-copy integrated transgene expressing mito-targeted NDI1. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001. See also Figure S5.
Because complex I levels in nduf-7(et19) are not restored by nuo-3(G60D) or hypoxia, despite overall growth and ETC activity being rescued, we used this mutant for further biochemical studies of complex I activity. We isolated mitochondrial membranes and performed in vitro assays using absorbance spectrophotometry to monitor the consumption of NADH. Wild type NADH oxidase activity was 90% inhibited by rotenone and piericidin (complex I inhibitors), and 75% inhibited by antimycin and sodium azide (which target complexes III and IV, respectively) (Fig 5C), confirming the activity was complex I and ETC dependent. We measured NADH oxidase activity from nduf-7(et19) mutant membranes and observed a 50% reduction in complex I activity which was partially rescued by the nuo-3(G60D) suppressor (Fig 5D). We confirmed through western blot of the same mitochondrial membranes that complex I levels were not changed in the nduf-7(et19); nuo-3(G60D) double mutant (Fig S5C), despite the increase in NADH oxidation rate. This data demonstrates that nuo-3(G60D) can restore mutant complex I forward activity independent of complex I levels, explaining the rescue of downstream ETC flux and growth. We then performed the same complex I activity assay using wild type and nduf-7(et19) membranes in a hypoxia cabinet. Strikingly, we observed that acute incubation at 1% oxygen was also able to increase the NADH oxidation rate from nduf-7(et19) mutant membranes in vitro (Fig 5E), showing that hypoxia, like nuo-3(G60D), can directly restore the forward activity of complex I.
Restored ETC flux is upstream of complex I protein level rescue
To untangle the effects of nuo-3(G60D) in the gas-1(fc21) mutant, where complex I levels are restored (Figs 4A–B), we took advantage of the yeast NDI1 bypass of complex I deficiency which restores electron flow from NADH to CoQ. nduf-7(et19) and gas-1(fc21) mutants rescued by single-copy integrated NDI1 were tested for complex I subunit levels. While NDI1 rescued the slow growth of both mutants, complex I levels were partially restored only in gas-1(fc21) (Figs 5F & S5D). This pattern matches exactly the effects of hypoxia and nuo-3(G60D), and proves that restoration of ETC flux is sufficient (i.e. upstream) of complex I level rescue in the gas-1(fc21) mutant. Our model is that nuo-3(G60D) and hypoxia restore complex I forward activity, with many downstream consequences such as increased proton pumping, oxygen consumption, ATP production, and in the case of some mutants, complex I levels.
Complex I subunit levels are decreased in the brains of Ndufs4 knockout mice36,37, and we measured the effects of two hypoxia regimes that improve the disease phenotype in these animals. We observed a modest increase in the levels of NDUFS1, NDUFS6, NDUFS3, and NDUFA9 in both a “prevention” paradigm11 where 11% oxygen treatment began at weaning (day 25–30) and a “rescue” paradigm12 in which mice were exposed to 21% oxygen for 55 days until they developed advanced disease symptoms and then transferred to 11% oxygen (Fig S5E). Similarly in C. elegans, the NDUFS4lpd-5(mg746); nuo-3(G60D) double mutant, which is fertile at 1% oxygen (Fig S2E) and thus can be probed by western blot, displayed complex I levels that were sensitive to increasing amounts of oxygen (Fig S5F). In concordance, the lpd-5(mg746); nuo-3(G60D) double mutant was sensitive to 50% oxygen (Fig S5G). Taken together, these results show that hypoxia restores complex I levels in NDUFS4 mutants as it does in gas-1(fc21).
Complex I rescue by NDUFA6/nuo-3(G60D) requires LYRM domain activity
To better understand the mechanism of how nuo-3(G60D) leads to increased complex I forward activity in complex I mutants, we generated transgenic animals expressing nuo-3 variants with a wild-type nuo-3 gene at the endogenous locus. Over-expression of nuo-3(G60D) but not wild-type nuo-3 suppressed gas-1(fc21) and nduf-7(et19) sensitivity to hyperoxia (Figs 6A–B), supporting a dominant gain-of-function mode of action for the G60D mutation (as opposed to increased wild-type function). Transgenic rescue of gas-1(fc21) by nuo-3(G60D) was not as effective as mutating the endogenous nuo-3 locus, suggesting that competition with wild type NUO-3 for inclusion in complex I was detrimental to growth in hyperoxia (Fig S6A). To confirm that both transgenes were functional, we generated a nuo-3 null allele using CRISPR that we found arrested development at an early larval stage in 21% and 1% oxygen (Fig S6B). Either nuo-3(wt) or nuo-3(G60D) transgenes could rescue the lethality of the nuo-3(null) (Fig S6C), confirming their functionality.
Figure 6. Complex I rescue by NDUFA6/nuo-3(G60D) requires LYRM domain activity.
A. Growth of animals for 5 days at 50% oxygen followed by 1 day at 21% oxygen. B. Growth of animals for 5 days at 50% oxygen followed by 3 days at 21% oxygen. C. Ovine complex I (PDB: 6ZKC43) in closed conformation. Colored red is the residue corresponding to C. elegans nuo-3(G60D) based on homology modelling. Colored red and represented as sticks are residues corresponding to C. elegans LYRM residues Y38, K39, and F70. D-E. Growth of animals for 6 days at 50% oxygen. F. Growth of animals for 5 days at 21% oxygen at 20°C. G-H. Growth of animals for 3 days at 21% oxygen incubated at 20°C. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test or Dunnett’s Multiple Comparison Test (G-H). Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001. See also Figure S6.
Glycine 60 of NDUFA6/NUO-3 lies within the conserved LYRM domain responsible for binding the acyl carrier protein (NDUFAB1 or ACP) to complex I40 (Fig 6C). The motif is characterized by a L-Y-R tripeptide followed by a downstream F, and mutation of these residues in the yeast Yarrowia lipolytica partially reduces ACP binding and complex I activity41. To test if altering LYRM domain activity would affect the function of NUO-3 in the context of complex I rescue, we generated nuo-3 transgenes containing the F70A mutation. Neither nuo-3(F70A) nor nuo-3(G60D F70A) could suppress gas-1(fc21) sensitivity to hyperoxia, demonstrating that LYRM domain activity is required for the nuo-3(G60D) suppression (Fig 6D). In C. elegans and other nematodes the L-Y-R sequence is replaced by the functionally adjacent A-Y-K (Fig S6D). We introduced Y38A K39A substitutions into nuo-3 transgenes and again found the rescue of gas-1(fc21) by nuo-3(G60D) to be abolished when LYRM domain activity was compromised (Fig 6E). Importantly, nuo-3(G60D F70A) and nuo-3(Y38A K39A G60D) transgenes were both able to partially rescue the growth and fertility defects of the nuo-3(null) mutant, demonstrating that these protein variants were still functional (Fig 6F). Taken together, this argues against nuo-3(G60D) acting as a dominant negative, as weakening the LYRM domain is insufficient to rescue complex I mutants, and in fact shows a requirement for LYRM domain activity (i.e. ACP binding) in the action of the nuo-3(G60D) suppressor.
To further explore how the G60D substitution affects the function of NUO-3 we used CRISPR/Cas9 editing followed by homologous repair with degenerate oligos to generate all possible G60 missense mutations at the chromosomal nuo-3 locus in a single experiment. In a wild-type background all G60 substitutions are tolerated with no discernable effects on growth (Fig S6E). To our surprise, many G60 amino acid substitutions were able to rescue the developmental arrest of NDUFS4/lpd-5 null animals at 21% oxygen, including R, K, D, H, T, V, I, M, L, F, and W (Figs 6G–H). However, the sterically small amino acids with side chains of 0–1 carbons (e.g. G, A, or S) as well as proline did not rescue complex I mutants (Figs 6G–H). These results are in concordance with our genetic screen, which used a mutagen that favors G-to-A or C-to-T transitions and could thus only isolate G60D or G60S substitutions. Interestingly when comparing NUO-3 sequences across nematode evolution we found that NUO-3 amino acid 60 is conserved as either a G, S, or P (Fig S6D), suggesting selection for these small side-chain amino acids in normal complex I function. NUO-3(G60) lies near the end of the alpha helix that surrounds the NDUFAB1 acyl chain, and residues such as G and P tend to break alpha helices42. Outside essential features of the LYRM domain, the primary sequence of NDUFA6/NUO-3 has diverged across eukaryotic evolution, and nuo-3(G60) may correspond to a lysine residue in mammals or be functionally analogous to NDUFA6(G59) (Fig S6D). An intriguing possibility is that in the context of NDUFS4 loss, mutating this residue to a larger amino acid may stabilize the alpha helix and enhance binding to NDUFAB1.
Residues surrounding the CoQ binding cavity are required for complex I rescue by NDUFA6/nuo-3(G60D) or hypoxia
Complex I mutants nduf-7(et19) and gas-1(fc21) are sensitive to low doses of the complex I inhibitor rotenone (0.1μM) – a concentration that does not affect the growth of wild type animals – likely due to additive inhibitory effects on complex I (Fig S7A). Surprisingly, nuo-3(G60D) as a single mutant was hypersensitive to rotenone, which was especially evident at higher doses of 0.5 and 1.0 μM (Figs 7A & S7A). Rotenone is a plant-derived inhibitor of complex I that occupies the quinone binding pocket43. Expression of the yeast complex I bypass protein NDI1 rendered wild-type C. elegans rotenone-resistant, proving that rotenone inhibits C. elegans growth rate by the inhibition of complex I (Fig S7B). Sensitivity of nuo-3(G60D) to rotenone may imply a change in conformation near the rotenone/CoQ binding site, which is consistent with NUO-3/NUYM mutations in Yarrowia having long-range effects on complex I structure that disrupt the formation of loops required for CoQ redox chemistry44. These results revealed that the nuo-3(G60D) mutant harbors a liability to growth in rotenone, potentially explaining the presence of G, S, and P residues at this site in evolution, and presenting an opportunity to screen for rare suppressor mutations.
Figure 7. Mutations surrounding the CoQ binding pocket block the ability of NDUFA6/nuo-3(G60D) or hypoxia to rescue complex I.
A. Growth of animals for 3 days at 21% oxygen with 0, 0.5, or 1.0 μM rotenone. B. nuo-3(G60D) P0 animals were randomly mutagenized with EMS in the absence of rotenone. F2 progeny were transferred to rotenone-containing plates and selected for growth. C. Growth of animals for 2 days at 21% oxygen incubated at room temperature with 0 or 1.0 μM rotenone. D. Ovine complex I (PDB: 6ZKC43) in closed conformation. Labelled and colored red are residues corresponding to C. elegans nuo-3(G60D) suppressor mutations NDUFS2/GAS-1(V161I), NDUFS7/NDUF-7(A119T or A119V), and NDUFS7/NDUF-7(M80I). Decylubiquinone is colored magenta. E-F. Growth of animals for 3 days at 21% oxygen incubated at 20°C. G. Growth of animals for 4 days at 21% oxygen incubated at room temperature. H. NADH-driven complex I-dependent superoxide production in the presence or absence of piericidin by isolated mitochondrial membranes at 21% oxygen. Resorufin absorbs light at 557 nm. I. Growth of animals for 3 days (left) and 5 days (right) at 21% or 1% oxygen incubated at room temperature. J. Complex I mutations result in decreased flux through the ETC, leading to local hyperoxia which can further inhibit complex I activity. Hypoxia or NDUFA6/nuo-3(G60D) mutation restore complex I forward activity in a manner dependent on residues surrounding the CoQ binding cavity. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001. See also Figure S7.
To understand the downstream effectors of nuo-3(G60D) rescue, we performed a forward genetic screen for suppressors of NDUFA6/nuo-3(G60D) growth arrest on rotenone (Fig 7B). Following sequencing of the nuclear genome and mtDNA we identified four independent alleles of NDUFS7/nduf-7, one allele of NDUFS2/gas-1, and eight alleles of MED15/mdt-15 (Table S2). CRISPR-Cas9 generated alleles in a nuo-3(G60D) background suppressed the rotenone hypersensitivity phenotype, confirming the genetic identity of these suppressor mutations (Figs 7C & S7C). Based on high resolution structures, the CoQ binding pocket within complex I is defined by mobile loops emanating from NDUFS2, NDUFS7, ND1, and ND345–47. Strikingly, the mutations we isolated in complex I subunits were amino acid substitutions in three highly conserved residues surrounding the CoQ/rotenone binding cavity: NDUFS2/GAS-1(V161I), NDUFS7/NDUF-7(A119T or A119V), and NDUFS7/NDUF-7(M80I) (Figs 7D & S7D). Unlike the complex I mutants gas-1(fc21) and nduf-7(et19) used for much of the analysis above, the nduf-7 and gas-1 amino acid substitution mutations isolated as nuo-3(G60D) suppressors do not grow slowly at 21% oxygen, nor are they sensitive to 50% oxygen (Figs S7E–F). The multiple mdt-15 alleles that emerged from the rotenone resistance screen may activate rotenone detoxification. mdt-15 encodes a component of the Mediator transcriptional regulatory complex that controls drug detoxification; gain-of-function mutations in mdt-15, including the same P117L substitution mutation isolated in our screen, activate C. elegans xenobiotic detoxification pathways48. All mutations isolated for NDUFA6/nuo-3(G60D) rotenone resistance confer partial resistance to rotenone in a NDUFA6/nuo-3 wild-type background (Fig S7B). Thus, complex I structural changes that prevent rotenone binding (in the case of the nduf-7 and gas-1 amino acid substitution mutations) or activation of a rotenone detoxifying response (in the case of mdt-15) mediate the nuo-3(G60D) suppression. Taken together, these results suggest that nuo-3(G60D) causes long-range conformational changes around the CoQ binding site that render animals hypersensitive to rotenone.
We tested whether the same suppressor mutations that rescue the hypersensitivity of nuo-3(G60D) to rotenone would also block the beneficial effects nuo-3(G60D) has on oxygen-sensitive complex I mutants. Indeed, introduction of gas-1(V161I) or nduf-7(M80I) mutation completely suppressed the rescue of NDUFS4/lpd-5(mg746) by nuo-3(G60D) (Figs 7E–F), identifying a downstream genetic requirement for the restoration of complex I activity. nduf-7(A119V) mutation partially suppressed the rescue of lpd-5(mg746) by nuo-3(G60D) (Fig 7E), consistent with its weaker rotenone resistance phenotype. mdt-15(P117L) had no effect on rescue of lpd-5(mg746) by nuo-3(G60D) (Fig 7G), consistent with its mechanism of action being complex I-independent activation of rotenone detoxification. Importantly, the gas-1(V161I) mutation did not affect the growth rates of the lpd-5(mg746) or nuo-3(G60D) single mutants (Figs 7F & S7G), demonstrating no synthetic sickness but rather a specific activity for this mutation in blocking rescue of complex I activity by nuo-3(G60D). The overlapping genetic requirements for the beneficial and harmful effects of nuo-3(G60D) suggest that the same structural changes that produce rotenone sensitivity also underlie the rescue of NDUFS4/lpd-5(mg746), perhaps due to increasing accessibility of rotenone and CoQ, respectively. Conversely, the nuo-3(G60D) suppressor mutations (i.e. gas-1(V161I)) may decrease the accessibility of rotenone and CoQ to the cavity, leading to their rotenone-resistance and blockage of complex I rescue.
To further elucidate the effects of gas-1(V161I) and nduf-7(A119V) suppressor mutations on complex I, we isolated mitochondrial membranes from these mutants (in an otherwise wild-type background) and performed in vitro experiments using absorbance spectrophotometry. Western blotting revealed that complex I levels were not decreased in these mutant membranes, but rather slightly increased (Fig S7H). NADH oxidase activity was also not compromised in these mutants (Fig S7I), indicating that the flow of electrons from NADH to CoQ remains unchanged. However NADH-driven ROS production (without piericidin) was decreased in gas-1(V161I) and nduf-7(A119V) (Fig 7H), consistent with these mutations decreasing accessibility of the CoQ binding pocket, reducing the leak of electrons to molecular oxygen. In contrast, ROS production from complex I in the presence of piericidin, which blocks electron flow and is a readout of the flavin site reducing molecular oxygen, was unchanged in the mutants (Fig 7H), consistent with overall complex I levels being unaffected.
Finally, we asked if the genetic requirements for complex I rescue by NDUFA6/nuo-3(G60D) were also necessary for the beneficial effects of hypoxia on complex I mutants. The NDUFS2/gas-1(V161I) mutation significantly blunted the rescue of NDUFS4/lpd-5(mg746) by hypoxia in a nuo-3(wt) background (Fig 7I), identifying a “hypoxia-resistant” mutation that blocks hypoxia rescue of complex I dysfunction, and further suggesting the rescue by hypoxia and nuo-3(G60D) may share a common mechanism. nduf-7(A119V) and mdt-15(P117L) had no effect on lpd-5(mg746) growth in hypoxia (Fig 7I), consistent with the former’s weaker nuo-3(G60D) suppression phenotype and the latter’s complex I-independent mode of action. Unlike hif-1(ia4) null animals, none of the nuo-3(G60D) suppressors were sensitive to hypoxia on their own (Fig S7J), supporting a specific effect of gas-1(V161I) on the NDUFS4/lpd-5-hypoxia interaction. This data suggests that hypoxia is beneficial to complex I mutants in a manner dependent on residues in the CoQ binding pocket, supporting a model in which hypoxia promotes forward flow through complex I (Fig 7J).
Discussion
Complex I, the main entry point of energy into the ETC, is an ancient complex whose 14 core subunits are evolutionarily conserved across all major domains of life49. In recent years advances in structural biology have deepened our understanding of the macromolecular organization of complex I, revealing an Fe-S cluster electron transfer path in the soluble arm of complex I that ultimately reduces CoQ and induces proton pumping across the membrane arm to store energy45–47. But exactly how CoQ reduction transduces proton pumping across the membrane arm is actively debated, and what role eukaryotic-specific accessory subunits may have in this mechanism is unknown. The current genetic study has revealed the existence of a dense, oxygen-sensitive allosteric network within complex I (involving accessory subunits NDUFA6 and NDUFAB1) with effects centered on the CoQ binding pocket (Figs 7D & 7J). These genetic interactions likely reflect allosteric coupling between an accessory subunit and energy conversion centers within complex I that arose through co-evolution of this complex with fluctuations of ambient oxygen that occur over evolutionary, developmental, and physiological timescales.
Our mechanistic study was initially motivated by the discovery that low oxygen can rescue phenotypes of the Ndufs4 knockout mouse11. We now find that that the ability of hypoxia to rescue NDUFS4 knockout is evolutionarily conserved to C. elegans. We tested a total of nine C. elegans complex I mutants for rescue by hypoxia and found that only those mutations that partially compromise the matrix arm of complex I could be suppressed by hypoxia. Our in vivo results from C. elegans are distinct from a genome-wide CRISPR screen performed in mammalian K562 cells at differing oxygen tensions which found nearly all complex I subunit knockouts were rescued at 1% oxygen50. This discordance may reflect the fact that cell culture models are more tolerant of mitochondrial dysfunction. In fact rho0 cells altogether lacking mtDNA and the respiratory chain can be grown if supplemented with glucose, pyruvate, and uridine51. Given that human disease-causing mutations in complex I are widely distributed throughout the complex and give rise to heterogeneous phenotypes5, these results may help predict which human mutations are “oxygen-sensitive” and hence candidates for hypoxia-based therapy.
To further explore the genetics of C. elegans complex I mutations that are suppressed by hypoxia, we took advantage of the sensitivity of complex I mutants to 50% oxygen. The use of this moderate hyperoxia regime has physiological relevance, as brains of Ndufs4 knockout mice experience local hyperoxia, likely due to reduced ETC activity and less associated oxygen consumption13. We identified a C. elegans intra-complex I mutation, NDUFA6/nuo-3(G60D), capable of rescuing all three complex I mutants rescued by hypoxia. These mutations are distributed across the soluble matrix arm, leading us to hypothesize that NDUFA6/nuo-3(G60D) is generally protective to this specific class of complex I mutants. Using in vitro biochemical assays, we showed that NDUFA6/nuo-3(G60D) rescues complex I forward activity - the flow of electrons from NADH to CoQ - while not necessarily restoring complex I protein levels. This data is consistent with a model in which nuo-3(G60D) mutation alters complex I structure around the CoQ binding pocket to create a more active enzyme in these mutant backgrounds. This change in activity will lead to more oxygen consumption at complex IV, lowering local oxygen levels which may further stabilize the fragile matrix arm, consistent with proposed models39 (Fig 7J). The downstream consequences on the cell include more mitochondrial ATP production, increased membrane potential, and normalized NADH/NAD+ balance. These factors individually or in combination likely contribute to the growth benefit of the complex I mutants.
To discover the downstream effectors of the rescue by NDUFA6/nuo-3(G60D), we took advantage of the hypersensitivity of this mutant to rotenone – a feature consistent with increased accessibility of the rotenone/CoQ binding pocket. We performed a forward genetic screen to identify nuo-3(G60D) suppressor mutations and identified conserved residues in NDUFS2/GAS-1 and NDUFS7/NDUF-7 surrounding the CoQ binding pocket (Fig 7D) that are necessary for the effects of NDUFA6/nuo-3(G60D) on rotenone sensitivity and complex I rescue. Two of these residues, NDUFS2/GAS-1(V161) and NDUFS7/NDUF-7(M80), were previously found through mutational studies in Yarrowia to define the CoQ binding pocket and be important for complex I activity but not complex I abundance52,53. Our combined genetics and biochemistry support a model in which nuo-3(G60D) alters complex I conformation to enhance binding of both the natural electron acceptor ubiquinone and the ubiquinone competitive inhibitor rotenone, and mutations identified in our suppressor screen such as NDUFS2/gas-1(V161I) revert these effects. This data is consistent with work from Yarrowia demonstrating NDUFA6 mutations trigger long-range conformational changes that impact the Q binding site44 and supports a model in which increased forward activity of complex I precedes the restoration of complex I stability in oxygen-sensitive mutants.
Four lines of evidence support the model that NDUFA6/nuo-3(G60D) is a “hypoxia-mimetic” mutation that rescues complex I deficiency by a mechanism that is similar to hypoxia: (1) the subset of complex I lesions rescued by nuo-3(G60D) and hypoxia are identical, (2) nuo-3(G60D) protects these complex I mutants against moderate hyperoxia, (3) nuo-3(G60D) and hypoxia have identical effects on complex I protein levels and complex I activity, and (4) the same mutation, NDUFS2/gas-1(V161I), blocks the rescue by nuo-3(G60D) or hypoxia, jointly pointing to a shared rescue mechanism. The acute rescuing effects of hypoxia on complex I activity in vitro, combined with the identification of NDUFS2/gas-1(V161I) as a “drug-resistant” mutant for hypoxia rescue, strongly support the model that hypoxia acts like nuo-3(G60D) to rescue complex I activity either through triggering a structural change near the CoQ binding pocket, or alternatively by altering the local chemical environment.
The NDUFA6/NUO-3(G60D) residue lies in the conserved LYRM domain of NDUFA6, responsible for tethering the acyl carrier protein NDUFAB1/ACP to complex I40. We show through structure-function studies that LYRM domain activity is required for nuo-3(G60D) suppression activity. NDUFAB1 is one of the last accessory subunits to be incorporated into complex I and is essential for its assembly and function3,4,54. Additionally, NDUFAB1 can bind to other LYRM domain-containing proteins and is required for at least two more essential processes in the mitochondria: it is the scaffold upon which type II fatty acid synthesis proceeds55, an upstream requirement for protein lipoylation, and it is an essential component of the Fe-S cluster biosynthesis machinery56. C. elegans and mammalian mutants in Fe-S cluster biosynthesis are rescued by hypoxia38, and all the above NDUFAB1-dependent processes were found to be selectively essential at 21% oxygen (i.e. rescued by hypoxia) in a genome wide CRISPR screen50. We therefore hypothesize that NDUFAB1 may coordinate a general response to oxygen in the mitochondria. Our mutational studies at the NUO-3(G60) site revealed that many amino acid substitutions can rescue NDUFS4/lpd-5 loss, arguing against a strong structural role for the G60 residue (e.g. salt bridge, polar contacts), and suggesting that the local dynamics of this region near the end of the NDUFAB1 acyl chain have important consequences for complex I activity. Interestingly there is precedent for NDUFAB1 over-expression in mice increasing complex I formation and providing protection from ischemia-reperfusion injury57.
Most components of the mitochondrial electron transport chain, including many complex I proteins, are conserved across major domains of life including bacteria and archaea49. Hence, these ETC components evolved in the last universal common ancestor (LUCA) prior to the great oxygenation event almost 3 billion years ago58 by using terminal electron acceptors other than oxygen. Many ecological niches today continue to be anaerobic, where microbes couple their ETC to terminal electron acceptors such as nitrate, sulfate, or ferric iron. Tantalizingly, the NDUFS7/NDUF-7(M80I) mutation is present in many bacterial species that grow in anaerobic environments, and the NDUFS7/NDUF-7(A119V) mutation is present in yeast (Fig S7D), showing that our mutants reflect existing natural variations that have been selected for in microbial evolution. Even among diverse eukaryotes that have evolved during the past billion years, hypoxia is often encountered in their normal life cycles59, for example in deep water, crowded ecosystems, or poorly vascularized mammalian tumors. Thus, hypoxia treatment of mitochondrial disease marshals for medicine ancient and highly evolved adaptations of the electron transport chain to natural oxygen tension changes.
Limitations of the Study
This work demonstrates that some, but not all, C. elegans complex I mutants are rescued by hypoxia and sensitive to high oxygen. For these mutants the underlying mechanism of hypoxia rescue is a restoration of complex I activity. In some but not all cases, a restoration of ETC flux leads to an increase in complex I protein levels, likely due to decreased molecular oxygen. In agreement, we observed a small increase in complex I levels in the brains of Ndufs4 mice exposed to hypoxia. Future work in mammals is required to determine if the rescue of complex I mutants by hypoxia is also due to an increase in complex I activity. Our “hypoxia-mimetic” and “hypoxia-resistant” mutations point to the CoQ binding pocket as a site of action of hypoxia – either through inducing allosteric structural changes or by altering quinone chemistry. Structural studies of C. elegans complex I in our mutant backgrounds, or mammalian complex I in hypoxia, may help to reveal chemical or conformation changes that restore the flow of electrons from NADH to ubiquinone.
STAR Methods
RESOURCE AVAILABILITY
Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Vamsi K. Mootha (vamsi_mootha@hms.harvard.edu).
Materials availability
C. elegans strains and plasmids generated in this study are available upon request from the lead contact.
Data and code availability
Proteomics datasets are available in Table S3. All data reported in this paper will be shared by the lead contact upon request.
This paper does not report original code.
Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.
EXPERIMENTAL MODELS AND STUDY PARTICIPANT DETAILS
C. elegans strain maintenance
C. elegans were propagated on NGM plates seeded with E. coli strain OP5061. A complete list of strains used in this study can be found in the Key Resources Table. Some strains were provided by the CGC, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440). Some strains were provided by the Mitani Lab through the National Bio-Resource Project of the MEXT, Japan. In concordance with C. elegans nomenclature and with approval from wormbase.org, C33A12.1 (ortholog of NDUFA5) has been renamed ndua-5, Y56A3A.19 (ortholog of NDUFAB1) has been renamed ndab-1, and F37C12.3 (ortholog of NDUFAB1) has been renamed ndab-2.
Key Resources table
| REAGENT or RESOURCE | SOURCE | IDENTIFIER |
|---|---|---|
| Antibodies | ||
| Anti-NDUFS3 (mouse) | Abcam | ab14711 |
| Anti-NDUFS6 (rabbit) | Abcam | ab195807 |
| Anti-NDUFS1 (rabbit) | Abcam | ab185733 |
| Anti-NDUFA9 (mouse) | Abcam | ab14713 |
| Anti-ATP5A (mouse) | Abcam | ab14748 |
| Anti-actin (rabbit) | Abcam | ab179467 |
| Anti-HA (rabbit) | Abcam | ab236632 |
| Anti alpha-tubulin (11H10) (rabbit) | Cell signaling | 2125S |
| Anti-Lipoic acid (rabbit) | Sigma (Calbiochem) | 437695 |
| Anti-HNE (rabbit) | Sigma (Calbiochem) | 393206 |
| Bacterial and virus strains | ||
| E. coli OP50 | CGC | N/A |
| Chemicals, peptides, and recombinant proteins | ||
| Rotenone | Cayman chemical | 13995 |
| MitoSOX Red | ThermoFisher | M36008 |
| Ethyl methanesulfonate | Sigma | M0880 |
| Subtilisin A protease | Sigma | P5380 |
| Digitonin | Sigma | 300410 |
| Cytochrome c | Sigma | C7752 |
| Alamethicin | Sigma | A4665 |
| Iodonitrotetrazolium chloride | Sigma | I8377 |
| NADH | Sigma | N8129 |
| HRP | Sigma | 516531 |
| SOD | Sigma | S5395 |
| Amplex Red | ThermoFisher | A12222 |
| TMRM | ThermoFisher | T668 |
| Antimycin A | Sigma | A8674 |
| Sodium azide | Sigma | 71289 |
| Experimental models: Organisms/strains | ||
| C. elegans strain CB5602: vhl-1(ok161) X | CGC | CB5602 |
| C. elegans strain CW152: gas-1(fc21) X | CGC | CW152 |
| C. elegans strain MQ130 : clk-1(qm30) III | CGC | MQ130 |
| C. elegans strain MQ1333: nuo-6(qm200) I | CGC | MQ1333 |
| C. elegans strain MQ887: isp-1(qm150) IV | CGC | MQ887 |
| C. elegans strain N2: wild type | CGC | N2 |
| C. elegans strain SJ4100: zcIs13[hsp-6::GFP] V | CGC | SJ4100 |
| C. elegans strain TK22: mev-1(kn1) III | CGC | TK22 |
| C. elegans strain ZG31: hif-1(ia4) V | CGC | ZG31 |
| C. elegans strain FX16526: nduf-7(tm1436)/hT2 I | Mitani | FX16526 |
| C. elegans strain FX18578: nuo-5(tm2751)/nT1 V | Mitani | FX18578 |
| C. elegans strain FX19115: nuo-2(tm5258)/hT2 I | Mitani | FX19115 |
| C. elegans strain GR3406: lpd-5(mg746[354bp DEL])/hT2 I | This study | GR3406 |
| C. elegans strain GR3407: nduf-9(mg747[578bp DEL])/qC1 III | This study | GR3407 |
| C. elegans strain GR3409: lpd-5(mg746[354bp DEL])/hT2; vhl-1(ok161) | This study | GR3409 |
| C. elegans strain GR3410: lpd-5(mg746[354bp DEL])/hT2; hif-1(ia4) | This study | GR3410 |
| C. elegans strain GR3411: mgEx865[Prpl-28::NDI1 + ofm-1::gfp] | This study | GR3411 |
| C. elegans strain GR3412: lpd-5(mg746[354bp DEL])/hT2; mgEx865[Prpl-28::NDI1 + ofm-1::gfp] | This study | GR3412 |
| C. elegans strain GR3413: nduf-7(et19); zcIs13[hsp-6::GFP] | This study | GR3413 |
| C. elegans strain GR3414: zcIs13[hsp-6::GFP]; gas-1(fc21) | This study | GR3414 |
| C. elegans strain GR3415: nuo-6(qm200); zcIs13[hsp-6::GFP] | This study | GR3415 |
| C. elegans strain GR3416: lpd-5(mg746[354bp DEL])/hT2; zcIs13[hsp-6::GFP] | This study | GR3416 |
| C. elegans strain GR3417: nuo-2(tm5258)/hT2; zcIs13[hsp-6::GFP] | This study | GR3417 |
| C. elegans strain GR3418: nduf-7(et19); vhl-1(ok161) | This study | GR3418 |
| C. elegans strain GR3419: nduf-7(et19); hif-1(ia4) | This study | GR3419 |
| C. elegans strain GR3420: nduf-7(et19); mgEx865[Prpl-28::NDI1 + ofm-1::gfp] | This study | GR3420 |
| C. elegans strain GR3421: gas-1(fc21); mgEx865[Prpl-28::NDI1 + ofm-1::gfp] | This study | GR3421 |
| C. elegans strain GR3422: nduf-9(mg747[578bp DEL])/qC1; mgEx865[Prpl-28::NDI1 + ofm-1::gfp] | This study | GR3422 |
| C. elegans strain GR3423: nuo-6(qm200); mgEx865[Prpl-28::NDI1 + ofm-1::gfp] | This study | GR3423 |
| C. elegans strain GR3408: nduf-7(et19) I | This study, derived from QC134 (CGC) | GR3408 |
| C. elegans strain GR3425: nuo-3(mg748[G60D]); zcIs13[hsp-6::GFP] | This study | GR3425 |
| C. elegans strain GR3426: nduf-7(et19); nuo-3(mg748[G60D]); zcIs13[hsp-6::GFP] | This study | GR3426 |
| C. elegans strain GR3427: nuo-3(mg748[G60D]); zcIs13[hsp-6::GFP]; gas-1(fc21) | This study | GR3427 |
| C. elegans strain GR3428: nduf-7(et19); nuo-3(mg748[G60D]) | This study | GR3428 |
| C. elegans strain GR3429: nuo-3(mg748[G60D]); gas-1(fc21) | This study | GR3429 |
| C. elegans strain GR3430: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg748[G60D]) | This study | GR3430 |
| C. elegans strain GR3431: ndua-5(mg749[R126Q]) IV | This study | GR3431 |
| C. elegans strain GR3432: ndua-5(mg749[R126Q]); gas-1(fc21) | This study | GR3432 |
| C. elegans strain GR3433: lpd-5(mg746[354bp DEL])/hT2; ndua-5(mg749[R126Q]) | This study | GR3433 |
| C. elegans strain GR3434: nduf-7(et19); ndua-5(mg749[R126Q]) | This study | GR3434 |
| C. elegans strain GR3435: nduf-9(mg747[578bp DEL])/qC1; nuo-3(mg748[G60D]) | This study | GR3435 |
| C. elegans strain GR3436: nuo-6(qm200); nuo-3(mg748[G60D]) | This study | GR3436 |
| C. elegans strain GR3437: isp-1(qm150) nuo-3(mg748[G60D]) | This study | GR3437 |
| C. elegans strain GR3438: mev-1(kn1); nuo-3(mg748[G60D]) | This study | GR3438 |
| C. elegans strain GR3439: clk-1(qm30); nuo-3(mg748[G60D]) | This study | GR3439 |
| C. elegans strain GR3440: lpd-5(mg746[354bp DEL]); nuo-3(mg748[G60D]) | This study | GR3440 |
| C. elegans strain GR3424: nuo-3(mg748[G60D]) IV | This study | GR3424 |
| C. elegans strain GR3441: unc-119(ed3); mgTi62[Prpl-28::ndab-1 cDNA::HA + unc-119(+)] | This study | GR3441 |
| C. elegans strain GR3442: nduf-7(et19); unc-119(ed3); mgTi62[Prpl-28::ndab-1 cDNA::HA + unc-119(+)] | This study | GR3442 |
| C. elegans strain GR3443: gas-1(fc21); mgEx866[Prpl-28::nuo-3 cDNA + ofm-1::gfp] | This study | GR3443 |
| C. elegans strain GR3444: gas-1(fc21); mgEx867[Prpl-28::nuo-3 cDNA + ofm-1::gfp] | This study | GR3444 |
| C. elegans strain GR3445: gas-1(fc21); mgEx868[Prpl-28::nuo-3(G60D) cDNA + ofm-1::gfp] | This study | GR3445 |
| C. elegans strain GR3446: gas-1(fc21); mgEx869[Prpl-28::nuo-3(G60D) cDNA + ofm-1::gfp] | This study | GR3446 |
| C. elegans strain GR3447: nduf-7(et19); mgEx870[Prpl-28::nuo-3(G60D) cDNA + ofm-1::gfp] | This study | GR3447 |
| C. elegans strain GR3448: nduf-7(et19); mgEx871[Prpl-28::nuo-3(G60D) cDNA + ofm-1::gfp] | This study | GR3448 |
| C. elegans strain GR3449: gas-1(fc21); mgEx872[Prpl-28::nuo-3(F70A) cDNA + ofm-1::gfp] | This study | GR3449 |
| C. elegans strain GR3450: gas-1(fc21); mgEx873[Prpl-28::nuo-3(F70A) cDNA + ofm-1::gfp] | This study | GR3450 |
| C. elegans strain GR3451: gas-1(fc21); mgEx874[Prpl-28::nuo-3(G60D F70A) cDNA + ofm-1::gfp] | This study | GR3451 |
| C. elegans strain GR3452: gas-1(fc21); mgEx875[Prpl-28::nuo-3(G60D F70A) cDNA + ofm-1::gfp] | This study | GR3452 |
| C. elegans strain GR3453: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg750[G60R]) | This study | GR3453 |
| C. elegans strain GR3454: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg751[G60L]) | This study | GR3454 |
| C. elegans strain GR3455: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg752[G60K]) | This study | GR3455 |
| C. elegans strain GR3456: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg753[G60H]) | This study | GR3456 |
| C. elegans strain GR3457: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg754[G60F]) | This study | GR3457 |
| C. elegans strain GR3458: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg755[G60V]) | This study | GR3458 |
| C. elegans strain GR3459: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg756[G60T]) | This study | GR3459 |
| C. elegans strain GR3460: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg757[G60W]) | This study | GR3460 |
| C. elegans strain GR3461: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg758[G60N]) | This study | GR3461 |
| C. elegans strain GR3462: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg759[G60Q]) | This study | GR3462 |
| C. elegans strain GR3463: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg760[G60S]) | This study | GR3463 |
| C. elegans strain GR3464: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg761[G60C]) | This study | GR3464 |
| C. elegans strain GR3465: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg762[G60A]) | This study | GR3465 |
| C. elegans strain GR3466: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg763[G60Y]) | This study | GR3466 |
| C. elegans strain GR3467: lpd-5(mg746)/hT2; nuo-3(mg764[G60E]) | This study | GR3467 |
| C. elegans strain GR3468: lpd-5(mg746)/hT2; nuo-3(mg765[G60P]) | This study | GR3468 |
| C. elegans strain GR3469: lpd-5(mg746)/hT2; nuo-3(mg766[G60I]) | This study | GR3469 |
| C. elegans strain GR3470: lpd-5(mg746)/hT2; nuo-3(mg767[G60M]) | This study | GR3470 |
| C. elegans strain GR3471: gas-1(fc21); mgEx876[Prpl-28::nuo-3(Y38A K39A G60D) cDNA + ofm-1::gfp] | This study | GR3471 |
| C. elegans strain GR3472: gas-1(fc21); mgEx877[Prpl-28::nuo-3(Y38A K39A G60D) cDNA + ofm-1::gfp] | This study | GR3472 |
| C. elegans strain GR2251: mdt-15(mg640[P117L]) III | Ruvkun Lab (Mao et al 2019) | GR2251 |
| C. elegans strain GR3473: nuo-3(mg748[G60D]); gas-1(mg768[V161I]) | This study | GR3473 |
| C. elegans strain GR3474: nduf-7(mg769[A119T]); nuo-3(mg748[G60D]) | This study | GR3474 |
| C. elegans strain GR3475: nduf-7(mg770[A119V]); nuo-3(mg748[G60D]) | This study | GR3475 |
| C. elegans strain GR3476: mdt-15(mg640[P117L]); nuo-3(mg748[G60D]) | This study | GR3476 |
| C. elegans strain GR3477: nduf-7(mg771[M80I]); nuo-3(mg748[G60D]) | This study | GR3477 |
| C. elegans strain GR3478: lpd-5(mg746[354bp DEL])/hT2; gas-1(mg772[V161I]) | This study | GR3478 |
| C. elegans strain GR3479: lpd-5(mg746[354bp DEL]) nduf-7(mg773[M80I])/hT2[nduf-7(mg774[M80I])] | This study | GR3479 |
| C. elegans strain GR3480: lpd-5(mg746[354bp DEL])/hT2; nuo-3(mg748[G60D]); gas-1(mg772[V161I]) | This study | GR3480 |
| C. elegans strain GR3481: lpd-5(mg746[354bp DEL]) nduf-7(mg773[M80I])/hT2[nduf-7(mg774[M80I])]; nuo-3(mg748[G60D]) | This study | GR3481 |
| C. elegans strain GR3482: lpd-5(mg746[354bp DEL]) nduf-7(mg775[A119V])/hT2; nuo-3(mg748[G60D]) | This study | GR3482 |
| C. elegans strain GR3483: lpd-5(mg746[354bp DEL])/hT2; mdt-15(mg640[P117L])/hT2; nuo-3(mg748[G60D]) | This study | GR3483 |
| C. elegans strain GR3484: lpd-5(mg746[354bp DEL]) nduf-7(mg775[A119V])/hT2 | This study | GR3484 |
| C. elegans strain GR3485: lpd-5(mg746[354bp DEL])/hT2; mdt-15(mg640[P117L])/hT2 | This study | GR3485 |
| C. elegans strain GR3486: gas-1(mg776[V161I]) X | This study | GR3486 |
| C. elegans strain GR3487: nduf-7(mg777[A119T]) I | This study | GR3487 |
| C. elegans strain GR3488: nduf-7(mg778[A119V]) I | This study | GR3488 |
| C. elegans strain GR3540: gas-1(fc21); wuIs156[sod-2(genomic) + rol-6(su1006)] | This study | GR3540 |
| C. elegans strain GR3541: nduf-7(et19); wuIs156[sod-2(genomic) + rol-6(su1006)] | This study | GR3541 |
| C. elegans strain GR3542: gas-1(fc21); wuIs156[sod-2(genomic) + rol-6(su1006)]; mgEx892[Prpl-28::MTS::ctl-2 + ofm-1::gfp] | This study | GR3542 |
| C. elegans strain GR3543: gas-1(fc21); wuIs156[sod-2(genomic) + rol-6(su1006)]; mgEx893[Prpl-28::MTS::ctl-2 + ofm-1::gfp] | This study | GR3543 |
| C. elegans strain CL2166: dvIs19[gst-4::gfp] III | CGC | CL2166 |
| C. elegans strain GR3544: nduf-7(et19); dvIs19[gst-4::gfp] | This study | GR3544 |
| C. elegans strain GR3545: dvIs19[gst-4::gfp]; gas-1(fc21) | This study | GR3545 |
| C. elegans strain GR3546: nuo-3(mg787[G60STOP])/tmC9[F36H1.2 (tmIs1221)] IV +12.5 | This study | GR3546 |
| C. elegans strain GR3547: unc-119(ed3); mgTi69[Prpl-28::NDI + unc-119(+)] | This study | GR3547 |
| C. elegans strain GR3548: nduf-7(et19); unc-119(ed3); mgTi69[Prpl-28::NDI1 + unc-119(+)] | This study | GR3548 |
| C. elegans strain GR3549: unc-119(ed3); gas-1(fc21); mgTi69[Prpl-28::NDI1 + unc-119(+)] | This study | GR3549 |
| C. elegans strain GR3550: nuo-3(mg787[G60STOP])/tmC9; mgEx894[Prpl-28::nuo-3 cDNA + ofm-1::gfp] | This study | GR3550 |
| C. elegans strain GR3551: nuo-3(mg787[G60STOP])/tmC9; mgEx895[Prpl-28::nuo-3 cDNA + ofm-1::gfp] | This study | GR3551 |
| C. elegans strain GR3552: nuo-3(mg787[G60STOP])/tmC9; mgEx896[Prpl-28::nuo-3(G60D) cDNA + ofm-1::gfp] | This study | GR3552 |
| C. elegans strain GR3553: nuo-3(mg787[G60STOP])/tmC9; mgEx897[Prpl-28::nuo-3(G60D) cDNA + ofm-1::gfp] | This study | GR3553 |
| C. elegans strain GR3554: nuo-3(mg787[G60STOP])/tmC9; mgEx898[Prpl-28::nuo-3(G60D F70A) cDNA + ofm-1::gfp] | This study | GR3554 |
| C. elegans strain GR3555: nuo-3(mg787[G60STOP])/tmC9; mgEx876[Prpl-28::nuo-3(Y38A K39A G60D) cDNA + ofm-1::gfp] | This study | GR3555 |
| C. elegans strain GR3556: nuo-3(mg787[G60STOP])/tmC9; mgEx877[Prpl-28::nuo-3(Y38A K39A G60D) cDNA + ofm-1::gfp] | This study | GR3556 |
| Oligonucleotides | ||
| nuo-3(G60) CRISPR Guide: GGACTTCATGATATGCCGCT | This study | N/A |
| nuo-3(G60D) CRISPR Repair Ultramer: TGGTGGGATTTCGGACTTCATGATATGCCGCTCGACGTGTTCCGTGCTGTCATCAAGAAG | This study | N/A |
| nuo-3(G60?) CRISPR Repair Ultramer: TGGTGGGATTTCGGACTTCATGATATGCCGCTCnnnGTGTTCCGTGCTGTCATCAAGAAG | This study | N/A |
| nuo-3(G60STOP) CRISPR Repair Ultramer: TGGTGGGATTTCGGACTTCATGATATGCCGCTCTGAGTGTTCCGTGCTGTCATCAAGAAG | This study | N/A |
| ndua-5(R126) CRISPR Guide: attacagGCCGAATATGAAC | This study | N/A |
| ndua-5(R126Q) CRISPR Repair Ultramer: ttaaatgctcaataattacagGCCGAATATGAACTaGAAACTACTCAGGCAATTGTTGATTCAAAAGCATGGGAGCCTCTCGT | This study | N/A |
| nduf-7(A119) CRISPR Guide: TCATATATGCGGCGAAGTGC | This study | N/A |
| nduf-7(A119T) CRISPR Repair Ultramer: TCTGGCATTTGATCATATATtCttCtcaaTGtAGGTGCCATTTTGTTGGTTACTGTACCG | This study | N/A |
| nduf-7(A119V) CRISPR Repair Ultramer: TCTGGCATTTGATCATATATtCttCtcaaTaCAGGTGCCATTTTGTTGGTTACTGTACCG | This study | N/A |
| gas-1(V161) CRISPR Guide: AGCCTGCTCATTGCACATCA | This study | N/A |
| gas-1(V161I) CRISPR Repair Ultramer: GCCAAAGACCAAGCCTGCTCATTGCACATCATactGAtGTAATCGAGACGATCGAAGTAT | This study | N/A |
| nduf-7(M80) CRISPR Guide: TACCGATCCATATCATATCT | This study | N/A |
| nduf-7(M80I) CRISPR Repair Ultramer: GGCTCTGAAAACGACTCCGTACCGATCCATgTCgTAcCgaGGtGCgGCaAAaTGtATCATTTCGACAGCACAACATGCGAGACCGAA | This study | N/A |
| Recombinant DNA | ||
| Plasmid: pJDM34 [Prpl-28::NDI1::tbb-2 3’UTR] (Minimos) | This study | N/A |
| Plasmid: pJDM65 [Prpl-28::nuo-3 cDNA::tbb-2 3’UTR] (Minimos) | This study | N/A |
| Plasmid: pJDM66 [Prpl-28::nuo-3(G60D) cDNA::tbb-2 3’UTR] (Minimos) | This study | N/A |
| Plasmid: pJDM67 [Prpl-28::nuo-3(F70A) cDNA::tbb-2 3’UTR] (Minimos) | This study | N/A |
| Plasmid: pJDM68 [Prpl-28::nuo-3(G60D F70A) cDNA::tbb-2 3’UTR] (Minimos) | This study | N/A |
| Plasmid: pJDM76 [Prpl-28::nuo-3(Y38A K39A G60D) cDNA::tbb-2 3’UTR] (Minimos) | This study | N/A |
| Plasmid: pJDM86 [Prpl-28::ndab-1 cDNA::HA::tbb-2 3’UTR] (Minimos) | This study | N/A |
| Plasmid: pJDM117 [Prpl-28::NDI1-MTS::ctl-2::tbb-2 3’UTR] (Minimos) | This study | N/A |
Mouse strains, husbandry, and hypoxia exposure
Ndufs4 knockout (KO) mice were generously provided by the Palmiter laboratory (University of Washington). Pups were genotyped and weaned at 25–30 d of age. All cages were provided with food and water ad libitum and supporting Napa gel was provided as needed. Mice were maintained in a standard 12h light-dark cycle at a temperature between 20–25°C and humidity between 30% and 70%. Ndufs4 KO and wild-type (WT) controls were either housed in standard oxygen conditions (21% O2) until day 55 when Ndufs4 KO present advanced disease symptoms62 (“normoxia”), in hypoxia chambers (11% O2) starting at weaning until day 55 which prevents the disease11 (“hypoxia prevention”), or in hypoxia chambers (11% O2) starting at day 55 for 1 month which reverses the disease12 (“hypoxia rescue”), as previously described. Male and female mice were used as no sex-specific differences in disease progression have been reported or found in our experience working with these mice. For hypoxia treatments, mice were housed at ambient sea-level pressure in plexiglass chambers, and 11% O2 levels were maintained OxyCycler A84XOV Multi-Chamber Dynamic Oxygen Controller (BioSpherix Ltd., Parish, NY) using nitrogen gas (Airgas). The CO2 concentration in each chamber, as well as the temperature and humidity, were monitored continuously. Mice were exposed to gas treatment continuously for 24 hours per day, 7 days a week. The chambers were briefly opened three times per week to monitor health status, provide food and water, and change the cages. Mice were euthanized at the end of the experiment by harvesting vital organs under deep isoflurane anesthesia. The Massachusetts General Hospital Institutional Animal Care and Use Committee approved all animal work in this manuscript.
METHOD DETAILS
C. elegans strain generation
To generate mutants with CRISPR/Cas9, 30 pmol S. pyogenes Cas9 (IDT) was injected into C. elegans gonads along with 90 pmol tracrRNA (IDT), 95 pmol crRNA (IDT), ssODN repair template (when applicable), and 40 ng/μl PRF4::rol-6(su1006) plasmid was used as a marker of successful injections63. Alternatively, plasmids were injected encoding the Cas9 protein and guide RNA as described64. All guide RNAs and repair templates can be found in the Key Resources Table. Using CRISPR/Cas9, we generated a NDUFS4/LPD-5 null allele, mg746, that carries a 354 bp deletion resulting in a 103 amino acid deletion plus frameshift, which can be propagated as a balanced heterozygote. We also generated a NDUFA9/NDUF-9 null allele, mg747, using CRISPR that carries a 578 bp deletion resulting in a 112 amino acid deletion plus frameshift, which can be propagated as a balanced heterozygote.
To generate transgenic animals carrying extra-chromosomal arrays, a mix consisting of 50 ng/μl plasmid DNA of interest and 50 ng/μl plasmid DNA containing ofm-1::gfp was injected into C. elegans P0 gonads. F1 progeny displaying the co-injection marker were singled to new plates and screened for lines in which the array was inherited by the F2 generation; at least three independent lines were generated for each construct. To generate animals carrying single-copy integrated transgenes, a plasmid mix consisting of 50 ng/μl Mos1 transposase, 2 ng/μl myo-2::mCherry, 2 ng/μl myo-3::mCherry, and 12 ng/μl miniMos transgene was injected into unc-119(ed3) mutant C. elegans as described65. All plasmids can be found in the Key Resources Table.
Growth and fluorescent reporter assays
To measure C. elegans growth and development crowded plates of gravid animals were washed into tubes in M9 buffer [3 g KH2PO4, 6 g Na2HPO4, 5 g NaCl, 1 ml 1 M MgSO4, H2O to 1 liter] and incubated with 20% bleach and 10% 5M KOH for 5 minutes while vortexing. The resulting embryos were washed 3x in M9 buffer and allowed to hatch overnight while rocking in M9. The following day arrested L1 animals were dropped onto E. coli OP50 plates and incubated at 20°C. For assays in hyperoxia (50% or 100% oxygen) plates were sealed in a modular chamber (Stemcell Technologies #27310) and flushed for 3 minutes with either a 50:50 mixture of oxygen and nitrogen, or with pure oxygen gas. For assays in hypoxia (1% oxygen), animals were incubated in a Hypoxic in vitro cabinet (Coy Laboratory Products, Inc) at room temperature. To expose developing animals to the complex I inhibitor rotenone (Cayman chemical #13995) a stock solution of 1 mM rotenone was made in DMSO, diluted in water to the appropriate concentration, and then added to NGM plates that had been pre-seeded with E. coli. The plates were dried in a sterile hood after which arrested larvae were added; the final concentration of DMSO in the plates was less than 0.1%. To measure animal length, images were acquired using a ZEISS Axio Zoom V16 microscope with ZEN PRO software and the midline of individual animals was quantified in FIJI software.
To measure hsp-6::gfp or gst-4::gfp fluorescence, L4 animals were mounted on agar pads, immobilized in sodium azide, and imaged at 70x magnification using a ZEISS Axio Zoom V16 microscope with ZEN PRO software. Fluorescent images were quantified by calculating the mean fluorescence along the midline of the intestine using FIJI software. For TMRM and MitoSOX assays, stock solutions of 50 mM TMRM (ThermoFisher T668) or 5 mM MitoSOX Red (ThermoFisher M36008) were prepared in DMSO. A dilution was made in M9 Buffer and added to NGM plates pre-seeded with E. coli for a final concentration of 200 nM TMRM or 20 μM MitoSOX and dried in the dark. L4 animals were added to TMRM or MitoSOX containing plates and incubated in the dark for 24 hours at 21% or 50% oxygen. Animals were then picked to new plates and allowed to destain for 1 hour before being mounted on agar pads and immobilized in sodium azide (for MitoSOX) or 10 mg/ml levamisole (for TMRM). Images were acquired at 63x magnification for 200 ms (TMRM) or 1 second (MitoSOX) using a ZEISS Axio Imager Z1 microscope equipped with a Zeiss AxioCam HRc digital camera. Fluorescent images were quantified by calculating the mean fluorescence in the posterior bulb of the pharynx using FIJI software.
Genetic screens and sequence analysis
To screen for genetic suppressors of nduf-7(et19), gas-1(fc21), or nuo-3(G60D), thousands of L4 animals were exposed to 47 mM EMS (ethyl methanesulfonate) (Sigma M0880) for four hours while rocking. Animals were then washed twice with M9 buffer and allowed to recover on standard NGM plates. F1 animals were bleach prepped as described above to generate a synchronized L1 stage population of mutagenized F2 animals, which were then dropped onto standard NGM plates at 50% oxygen (for the nduf-7(et19) and gas-1(fc21) suppressor screens) or NGM plates at 21% oxygen containing 1 μM rotenone (for the nuo-3(G60D) suppressor screen). Plates were checked daily and F2 individuals capable of growing to adulthood were transferred onto new plates. Fertile isolates were retested using F3 or F4 progeny to confirm their phenotype and then genomic DNA for whole genome sequencing was isolated using Gentra Puregene Tissue Kit (Qiagen 158667).
To identify candidate suppressor mutations in screen isolates we sheared genomic DNA using a Covaris S2 sonicator and prepared libraries using the NEBNext DNA library prep kit for Illumina as described66. Libraries with unique barcodes were quantified using the Qubit dsDNA HS Assay Kit (Life Technologies Q32851) and pooled in sets of 24 and sequenced using Illumina HiSeq. Raw FASTQ files were analyzed on the Galaxy platform (usegalaxy.org) with the following workflow: TrimGalore! to trim reads, Map with BWA to align reads to the C. elegans reference genome (including the mtDNA), MiModD to call variants, and SnpEff to identify mutations that may affect protein function. Lists of protein-altering mutations from each suppressor strain were then compared to identify genes with multiple mutant alleles. These candidate genes were then verified using targeted CRISPR/Cas9-based editing.
Mitochondrial biochemistry
To isolate C. elegans mitochondria for downstream applications, at least four crowded 10 cm plates of mixed-stage animals were washed into M9 buffer and spun gently for 1 minute at 200g. Supernatant was removed, replaced with 10 ml fresh M9, and animals were rotated for 30–60 minutes to wash off contaminating E. coli. Animals were then washed twice with fresh M9, supernatant was removed, and the worm pellet placed on ice. If the mitochondria were to be used for Blue Native PAGE, 25 μl of 10 mg/ml subtilisin A protease (Sigma P5380) was added for 10 minutes. If mitochondria were to be used for SDS PAGE Western Blot or membrane isolation, no protease was added. Worms were resuspended in sucrose buffer [5 ml 1M Sucrose, 2.5 ml 0.1M Tris/MOPS pH 7.4, 0.25 ml 0.1M EGTA/Tris pH 7.4, 17.25 ml ddH2O, 1 protease inhibitor tablet], spun gently for 1 minute, supernatant removed, and resuspended in 1 ml sucrose buffer. Samples were transferred to a Dounce homogenizer fitted with the “tight” B pestle, and 40–50 strokes performed on ice. Samples were spun at 4°C for 5 minutes at 2000g, supernatants containing mitochondria transferred to new tubes, and spun again at 4°C for 10 minutes at 13,000g. Supernatant was removed, mitochondrial pellets were resuspended in sucrose buffer with pipetting, and protein concentration was measured using Bradford assay (Sigma B6916).
For assessment of ETC complexes and supercomplexes with Blue Native PAGE, 40 μl samples were prepared containing 25 μg mitochondria, 1% digitonin (Sigma 300410), NativePAGE Sample Buffer, and incubated on ice for 15 minutes. Samples were then spun at 4°C for 30 minutes at 20,000g, and supernatants were transferred to new tubes. To each sample G-250 was added to a final concentration of 0.1%, and then loaded into a 1mm thick 3–12% Native PAGE 10-well gel along with Native protein ladder (Thermo LC0725). The gel was run at 150V for 20 minutes in dark cathode buffer at 4°C, and then run at 250V for 90 minutes in light cathode buffer. If the gel was to be stained for ETC complexes, it was transferred to 40% methanol/10% acetic acid fixation solution and microwaved for 45 seconds at power level 10. The gel was then incubated on a rocker for 30 minutes and room temperature, rinsed with Milli-Q water, and stained with Imperial Protein Stain for 2 hours. The gel was de-stained with UltraPure water and imaged with a GE Amersham Imager. If the gel was to be used for the complex I in-gel activity assay, it was immediately transferred after the run to 5 mM Tris-HCl pH 7.4, 2.5 mg/ml iodonitrotetrazolium chloride (Sigma I8377), and 0.5 mg/ml NADH (Sigma N8129). The gel was incubated at room temperature for 30 minutes, the reaction terminated by addition of 10% acetic acid, rinsed with water and imaged with a GE Amersham Imager. If the native gel was instead used for western blotting, it was rinsed with ddH2O and transferred to PVDF membrane (Thermo LC2007) using NuPAGE Transfer Buffer and a semi-dry transfer apparatus for 20 minutes at 180 mA (0.18 A). The membrane was then incubated in 8% acetic acid for 5 minutes on a shaker, rinsed with ddH2O, rinsed in methanol for 1 minute, and then transferred to 0.5X TBST for 5 minutes. The membrane could then proceed to blocking and incubation with primary antibodies (see below).
SDS-PAGE western blots were performed with either isolated mitochondria (as above) or with whole worm lysate, prepared by snap-freezing worm pellets of equal volume in liquid nitrogen and then mixing with NuPAGE LDS sample buffer and boiling at 100°C for 20 minutes, followed by centrifugation at maximum speed for 10 minutes. Samples were run in 4–12% NuPAGE Bis-Tris gels for 2 hours at 100 V. Gels were transferred to nitrocellulose membranes using the iBlot Dry Blotting system, and then blocked in 5% milk in TBST for 1 hour. Membranes were incubated in primary antibodies overnight at 4°C in 5% milk/TBST. The following day membranes were washed for 1 hour in TBST, incubated with secondary antibodies (1:10,000 dilution) for 1 hour at room temperature in 5% milk/TBST, and then washed in TBST for another hour. Blots were developed using Pierce ECL Western Blotting Substrate (Fisher 32209) and imaged with a GE Amersham Imager.
For western blots of mouse brains, samples were rapidly harvested, flash-frozen in liquid nitrogen, and pulverized in liquid nitrogen using a mortar and pestle. Total proteins were isolated in RIPA buffer with EDTA-EGTA (Boston Bioproducts) supplemented with EDTA-free COmplete protease inhibitor cocktail (Roche) and quantified using the Pierce BCA protein assay (Thermofisher). 10 μg of total protein extracts in 1X Laemmli SDS-Sample buffer with beta-mercapto ethanol were separated by SDS-page using Novex WedgeWell 4–12% or 14% Tris-Glycine precast gels (Invitrogen) and transferred into polyvinylidene difluoride membranes (Biorad) using the Transblot Turbo blotting system (Biorad). Membranes were blocked in 5% milk in TBST for 1 hour at room temperature and incubated in primary antibodies overnight at 4°C in 3% milk/TBST. The following day membranes were washed for 30 minutes in TBST, incubated with secondary antibodies (1:10,000 dilution) for 1 hour at room temperature, and then washed for 30 min in TBST. Immunodetection was performed according to standard techniques using enhanced chemiluminescence (Western Lighting Plus, Perkin Elmer) captured in Amersham Hyperfilm.
To generate mitochondrial membranes for in vitro assays mitochondrial pellets were generated as above (from 10x starting material!) and resuspended in [20 mM Tris- HCl, 1 mM EDTA, 10% glycerol, pH 7.4] and stored at −80°C as described67. Mitochondrial suspensions were thawed on ice and sonicated in 1 mL volume in a round bottom 2 mL tube at 10% amplitude (2 seconds on, 1 second off) for 1 minute. Sonicated samples were ultra-centrifuged at 75,000g for 1 hour at 4°C. The pellets containing mitochondrial membranes were homogenized in resuspension buffer and stored at −80°C. All in vitro assays at 21% oxygen were performed in a PerkinElmer EnVision Plate Reader at room temperature using absorbance spectrophotometry. NADH oxidation assays were carried out in 10 mM Tris-SO4 (pH 7.4), 250 mM sucrose, 1.5 μM cytochrome c (Sigma C7752), 15 μg/ml alamethicin (Sigma A4665), 25 μg/ml mitochondrial membranes (protein), and 200 μM NADH (Sigma N8129), and consumption of NADH was monitored at 340 nm (minus 380 nm). Amplex red assays were carried out in 10 mM Tris-SO4 (pH 7.4), 250 mM sucrose, 1.5 μM cytochrome c, 15 μg/ml alamethicin, 80 μg/ml mitochondrial membranes (protein), 10 μM NADH, 1 μM or 0 μM piericidin, 2 U/ml HRP (Sigma 516531), 10 U/ml SOD (Sigma S5395), and 10 μM Amplex Red (ThermoFisher A12222) and production of resorufin was monitored at 557 nm (minus 620 nm). in vitro assays at 1% oxygen were performed in a Cytation 5 plate reader from Biotek, housed in a hypoxia glove box from Coy Lab Products. Normoxia-matched controls were performed on the same instrument immediately following runs in hypoxia.
Mass Spectrometry
Quantitative tandem mass tag proteomics was performed by the Thermo Fisher Center for Multiplexed Proteomics in the Department of Cell Biology at Harvard Medical School. From frozen worm pellets total protein quantification was performed using the micro-BCA assay by Pierce. Samples were reduced with DTT and alkylated with iodoacetimide. Proteins were precipitated using methanol/chloroform and the pellet resuspended in 200 mM EPPS, pH 8.0. Digestion was performed sequentially using LysC (1:50) and Trypsin (1:100) based on protease to protein ratio. Peptides were detected (MS1) in the Orbitrap; sequenced (MS2) in the ion trap; and quantified (MS3) in the Orbitrap. ~2 μl of each TMT-labelled sample was mixed to verify labelling success. Peptides were separated using a gradient of 3 to 27% 90% Acetonitrile in 0.1% formic acid over 180 minutes. MS2 spectra were searched using the Comet algorithm against a custom C. elegans + E. coli database containing its reversed complement and known contaminants. Peptide spectral matches were filtered to a 1% false discovery rate (FDR) using the target-decoy strategy combined with linear discriminant analysis. Proteins were quantified only from peptides with a summed SN threshold of >100. Raw data available in Table S3.
To determine the redox potential of the quinone pool, worms were washed in M9 buffer for 30 minutes, rinsed again in M9 to remove E. coli contamination, and incubated on ice. The worm pellet was extracted in 1 mL 95% methanol, 5% hexane. Samples were vortexed, bath sonicated for 1 minute, then spun down at 21.1 k x g for 20 minutes. 1 μL of each sample was analyzed using a Q Exactive Plus Orbitrap Mass Spectrometer with a DionexUltiMate 3000 UHPLC system (Thermo Fisher Scientific). Metabolites were separated on a Phenomenex Luna C8(2) column (2 X 30 mm, 3 μM particle size). Mobile phase A was 3% Methanol, 97 % water, 10 mM formic acid. Mobile phase B was 100% methanol with 2mM ammonium formate, and 0.15% formic acid. The gradient was: 90% B for 0–2 minutes, increased to 99% B from 2–8 mins, held at 99% B from 8–12 minutes, decreased to 90% B from 12–12.5 minutes, held at 90% B from 12.5–14 minutes. The flow rate was 300 μL/min. The MS data acquisition was positive ionization full scan mode in a range of 550–2000 m/z, with resolving power of 140,000, AGC target of 3E6, and maximum injection time of 80 msec. All LC-MS data was collected with samples injected in a randomized order. Data was processed with TraceFinder 4.1 software, quantifying the ammonium adducts of Q9 and Q9H2.
QUANTIFICATION AND STATISTICAL ANALYSYS
All statistical analyses were performed using GraphPad Prism software. Statistical tests are detailed in the Figure Legends. Typically, statistical significance was calculated using one-way ANOVA followed by correction for multiple hypothesis testing. Error bars represent standard deviation and ‘n’ refers to the number of animals tested in a single experiment. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
Supplementary Material
A. Growth of wild type animals for 2 days at room temperature exposed to differing concentrations of the complex I inhibitor rotenone. B. Multiple sequence alignment of NDUFS2/GAS-1 (C. elegans residues 284–299) and NDUFS7/NDUF-7 (C. elegans residues 182–199) including homologs from mammals, invertebrates, and fungi made using ClustalW. C. Mean intestinal fluorescence of hsp-6::gfp in age-matched animals grown at 21% or 1% oxygen for one generation. All genotypes were imaged at L4/young adult stage except nuo-2(tm5258) which arrest at L2 stage. D. Growth of animals for 2 days at 20°C. E. Growth of animals for 2 days at room temperature. F-G. Growth of animals for 2 days (left) and 3 days (right) at room temperature. In all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
A. Growth of animals following L1 synchronization at 21% oxygen (black), 50% oxygen (orange), or 100% oxygen (red) incubated at 20°C. ‘x’ indicates animals were dead. B. Growth of animals following L1 synchronization at 21% oxygen (black) or 50% oxygen for 2–6 days and then shifted to 21% oxygen (orange) at 20°C. Grey arrows indicate the day animals were shifted from 50% to 21% oxygen. C. Growth of animals for 3 days at 50% oxygen incubated at 20°C. D. Ovine complex I (PDB: 6ZKC43) in closed conformation. The NDUFA5/NDUA-5(R126Q) suppressor mutation lies at the interface of NDUFA5 and NDUFA10. E. Images of animals after 1 generation growth at 1% or 21% oxygen at room temperature. Magnification = 7x. F-G. Growth of animals for 3 days at room temperature. H. Growth of animals for 2 days at 50% oxygen followed by 1 day at 21% oxygen. I. Growth of animals for 5 days at room temperature. J. Growth of animals for 2 days at room temperature. K. Growth of animals at 50% oxygen for 4 days at room temperature, or growth at 100% oxygen for 2 days followed by 21% oxygen for 2 days at room temperature. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
A. Fluorescent images of L4 stage animals containing gst-4::gfp grown continuously at 21% oxygen or exposed to 100% oxygen for 1 day. Images were acquired at 69x magnification with an exposure time of 50 ms. B. Growth of animals for 3 days at 20°C at 21% or 50% oxygen. Statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. *** = p value <0.001.
A. Blue Native PAGE of isolated mitochondria purified from animals grown at 21% or 1% oxygen for 4 days. * denotes faint band that may correspond to CI subcomplex. B. BN-PAGE of isolated mitochondria followed by western blot from animals grown continuously at 21% oxygen. Black arrow indicates CI subcomplexes observed in nduf-7(et19) which run at the same size as complex V dimers. C. BN-PAGE of isolated mitochondria purified from animals grown at 21% or 50% oxygen for 2 days. D-E. TMT quantitative proteomics from animals exposed to 50% oxygen for 2 days. Plotted are log 2-fold ratios of all proteins from which at least two peptides were quantified. F. SDS-PAGE of whole worm lysate from animals exposed to 21% or 50% oxygen. Anti-lipoic acid antibodies recognize the modified E2 subunits of PDH (DLAT) and OGDH/KGDH (DLST). G-H. TMT quantitative proteomics from animals exposed to 50% oxygen for 2 days. Plotted are log 2-fold ratios of all proteins from which at least two peptides were quantified. Iron-sulfur cluster containing C. elegans proteins were identified by homology to a previously published list60.
A. Ratio of reduced CoQ9H2 to oxidized CoQ9 as determined by mass spectrometry. Samples were extracted from whole worms grown continuously at 21% oxygen. B. Mass spectrometry traces of CoQ9H2 and CoQ9 in wild type and clk-1(qm30) backgrounds. clk-1 mutants can survive using dietary CoQ8 from their bacterial food. C. SDS-PAGE followed by western blot of C. elegans mitochondrial membranes used in in vitro assays. D. Growth of animals for 3 days at room temperature at 21% oxygen. E. SDS-PAGE followed by western blot of brains from wild type or Ndufs4 mice exposed to 21% oxygen, 11% oxygen from weaning (day 25–30), or shifted from 21% to 11% oxygen after 55 days. F. SDS-PAGE followed by western blot of whole worm lysate from animals grown continuously at 1% oxygen and then exposed to 1%, 21%, or 50% oxygen for 3 days. G. Growth of animals for 3 days at room temperature. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
A. Growth of animals for 4 days at 50% oxygen. B. Growth of animals for 3 days at 21% or 1% oxygen at room temperature. C. Growth of animals for 4 days at 21% oxygen at 20°C. D. Multiple sequence alignment of NDUFA6/NUO-3 homologs from nematodes (above) and eukaryotes (below) made using ClustalW. The alignment corresponds to amino acids 33–83 of C. elegans NUO-3. E. Growth of animals for 3 days at 21% oxygen. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test or Dunnett’s Multiple Comparison Test (D). Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
A. Growth of animals for 3 days at 21% oxygen with 0, 0.1, or 1.0 μM rotenone. B. Growth of animals for 2 days at 21% oxygen with 0, 1.0, or 5.0 μM rotenone incubated at 20°C. C. Growth of animals for 2 days at 21% oxygen incubated at room temperature with 0 or 1.0 μM rotenone. D. Multiple sequence alignment of NDUFS2/GAS-1 (C. elegans residues 153–164) and NDUFS7/NDUF-7 (C. elegans residues 70–81…114–121) including homologs from animals, fungi, and bacteria made using ClustalW. E. Growth of animals for 2 days at 21% oxygen incubated at 20°C. F. Growth of animals for 2 days at 50% oxygen incubated at 20°C. G. Growth of animals for 2 days at 21% oxygen incubated at room temperature. H. SDS-PAGE followed by western blot of C. elegans mitochondrial membranes used in in vitro assays. I. Complex I-dependent oxidation of NADH by isolated mitochondrial membranes at 21% oxygen. NADH absorbs light at 340 nm. J. Growth of animals for 2 days at 21% or 1% oxygen incubated at room temperature. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test (A, D-F, H) or Dunnett’s Multiple Comparison Test (B). Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
Table S3: Quantitative TMT Proteomics from Complex I mutants at 1%, 21%, and 50% oxygen (raw data), related to Figure 4
Highlights.
Hypoxia rescue and hyperoxia sensitivity of complex I mutants is conserved in C. elegans
Hypoxia rescue is independent of HIF activation or ROS attenuation
NDUFA6/nuo-3(G60D) mimics acute hypoxia in restoring complex I forward activity
Residues in the CoQ binding pocket are required for rescue by nuo-3(G60D) or hypoxia
Acknowledgements
We thank Peter Breen, Connie Chan, Sarah Chang, Thomas Hercher, Sharon Kim, Andrew Markhard, and Jason McCoy for technical assistance. We thank Zenon Grabarek and all members of the Mootha and Ruvkun labs for helpful discussions. This work was supported in part by the National Institutes of Health (K99GM140217 to J.D.M., F32GM133047 to O.S.S., R01AG016636 to G.R., R01NS124679 to V.K.M.). J.D.M. was supported by The Jane Coffin Childs Memorial Fund for Medical Research. M.M. was supported by the Deutsche Forschungsgemeinschaft (431313887). V.K.M. is an Investigator of the Howard Hughes Medical Institute.
Footnotes
Declaration of Interests
V.K.M. is listed as an inventor on patents filed by MGH on therapeutic uses of hypoxia. V.K.M. is a paid advisor to 5am Ventures. O.S.S was a paid consultant for Proteinaceous, Inc. V.K.M. is on the Advisory Board at Cell.
Inclusion and Diversity
We support inclusive, diverse, and equitable conduct of research.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Hirst J. (2013). Mitochondrial Complex I. Annu. Rev. Biochem. 82, 551–575. 10.1146/annurev-biochem-070511-103700. [DOI] [PubMed] [Google Scholar]
- 2.Sazanov LA (2015). A giant molecular proton pump: structure and mechanism of respiratory complex I. Nat. Rev. Mol. Cell Biol. 16, 375–388. 10.1038/nrm3997. [DOI] [PubMed] [Google Scholar]
- 3.Stroud DA, Surgenor EE, Formosa LE, Reljic B, Frazier AE, Dibley MG, Osellame LD, Stait T, Beilharz TH, Thorburn DR, et al. (2016). Accessory subunits are integral for assembly and function of human mitochondrial complex I. Nature 538, 123–126. 10.1038/nature19754. [DOI] [PubMed] [Google Scholar]
- 4.Guerrero-Castillo S, Baertling F, Kownatzki D, Wessels HJ, Arnold S, Brandt U, and Nijtmans L. (2017). The Assembly Pathway of Mitochondrial Respiratory Chain Complex I. Cell Metab 25, 128–139. 10.1016/j.cmet.2016.09.002. [DOI] [PubMed] [Google Scholar]
- 5.Fiedorczuk K, and Sazanov LA (2018). Mammalian Mitochondrial Complex I Structure and Disease-Causing Mutations. Trends Cell Biol 28, 835–867. 10.1016/j.tcb.2018.06.006. [DOI] [PubMed] [Google Scholar]
- 6.Russell OM, Gorman GS, Lightowlers RN, and Turnbull DM (2020). Mitochondrial Diseases: Hope for the Future. Cell 181, 168–188. 10.1016/j.cell.2020.02.051. [DOI] [PubMed] [Google Scholar]
- 7.Frazier AE., Thorburn DR., and Compton AG. (2019). Mitochondrial energy generation disorders: genes, mechanisms, and clues to pathology. Journal of Biological Chemistry 294, 5386–5395. 10.1074/jbc.r117.809194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Schapira AHV, Cooper JM, Dexter D, Jenner P, Clark JB, and Marsden CD (1989). MITOCHONDRIAL COMPLEX I DEFICIENCY IN PARKINSON’S DISEASE. Lancet 333, 1269. 10.1016/s0140-6736(89)92366-0. [DOI] [PubMed] [Google Scholar]
- 9.Gopal RK, Calvo SE, Shih AR, Chaves FL, McGuone D, Mick E, Pierce KA, Li Y, Garofalo A, Allen EMV, et al. (2018). Early loss of mitochondrial complex I and rewiring of glutathione metabolism in renal oncocytoma. Proc National Acad Sci 115, E6283–E6290. 10.1073/pnas.1711888115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gopal RK, Kübler K, Calvo SE, Polak P, Livitz D, Rosebrock D, Sadow PM, Campbell B, Donovan SE, Amin S, et al. (2017). Widespread Chromosomal Losses and Mitochondrial DNA Alterations as Genetic Drivers in Hürthle Cell Carcinoma. Cancer Cell 34, 242–255.e5. 10.1016/j.ccell.2018.06.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jain IH, Zazzeron L, Goli R, Alexa K, Schatzman-Bone S, Dhillon H, Goldberger O, Peng J, Shalem O, Sanjana NE, et al. (2016). Hypoxia as a therapy for mitochondrial disease. Science 352, 54–61. 10.1126/science.aad9642. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ferrari M, Jain IH, Goldberger O, Rezoagli E, Thoonen R, Cheng K-H, Sosnovik DE, Scherrer-Crosbie M, Mootha VK, and Zapol WM (2017). Hypoxia treatment reverses neurodegenerative disease in a mouse model of Leigh syndrome. Proceedings of the National Academy of Sciences 114, E4241–E4250. 10.1073/pnas.1621511114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jain IH, Zazzeron L, Goldberger O, Marutani E, Wojtkiewicz GR, Ast T, Wang H, Schleifer G, Stepanova A, Brepoels K, et al. (2019). Leigh Syndrome Mouse Model Can Be Rescued by Interventions that Normalize Brain Hyperoxia, but Not HIF Activation. Cell Metabolism 30, 824–832.e3. 10.1016/j.cmet.2019.07.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Falk MJ, Rosenjack JR, Polyak E, Suthammarak W, Chen Z, Morgan PG, and Sedensky MM (2009). Subcomplex Iλ Specifically Controls Integrated Mitochondrial Functions in Caenorhabditis elegans. Plos One 4, e6607. 10.1371/journal.pone.0006607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kayser EB, Morgan PG, Hoppel CL, and Sedensky MM (2001). Mitochondrial expression and function of GAS-1 in Caenorhabditis elegans. J. Biol. Chem. 276, 20551–20558. 10.1074/jbc.m011066200. [DOI] [PubMed] [Google Scholar]
- 16.Kayser EB., Morgan PG., and Sedensky MM. (1999). GAS-1: a mitochondrial protein controls sensitivity to volatile anesthetics in the nematode Caenorhabditis elegans. Anesthesiology 90, 545–554. 10.1097/00000542-199902000-00031. [DOI] [PubMed] [Google Scholar]
- 17.Yang W, and Hekimi S. (2010). Two modes of mitochondrial dysfunction lead independently to lifespan extension in Caenorhabditis elegans. Aging Cell 9, 433–447. 10.1111/j.1474-9726.2010.00571.x. [DOI] [PubMed] [Google Scholar]
- 18.Rauthan M, Ranji P, Abukar R, and Pilon M. (2015). A Mutation in Caenorhabditis elegans NDUF-7 Activates the Mitochondrial Stress Response and Prolongs Lifespan via ROS and CED-4. G3 (Bethesda) 5, 1639–1648. 10.1534/g3.115.018598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Félix M-A, and Duveau F. (2012). Population dynamics and habitat sharing of natural populations of Caenorhabditis elegans and C. briggsae. BMC Biol. 10, 59. 10.1186/1741-7007-10-59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Gray JM, Karow DS, Lu H, Chang AJ, Chang JS, Ellis RE, Marletta MA, and Bargmann CI (2004). Oxygen sensation and social feeding mediated by a C. elegans guanylate cyclase homologue. Nature 430, 317–322. 10.1038/nature02714. [DOI] [PubMed] [Google Scholar]
- 21.Voorhies WAV, and Ward S. (2000). Broad oxygen tolerance in the nematode Caenorhabditis elegans. J. Exp. Biol. 203, 2467–2478. [DOI] [PubMed] [Google Scholar]
- 22.Jafari G, Wasko BM, Tonge A, Schurman N, Dong C, Li Z, Peters R, Kayser E-B, Pitt JN, Morgan PG, et al. (2015). Tether mutations that restore function and suppress pleiotropic phenotypes of the C. elegans isp-1(qm150) Rieske iron-sulfur protein. Proceedings of the National Academy of Sciences 112, E6148–57. 10.1073/pnas.1509416112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hartman PS, Ishii N, Kayser EB, Morgan PG, and Sedensky MM (2001). Mitochondrial mutations differentially affect aging, mutability and anesthetic sensitivity in Caenorhabditis elegans. Mechanisms of Ageing and Development 122, 1187–1201. [DOI] [PubMed] [Google Scholar]
- 24.Ishii N, Takahashi K, Tomita S, Keino T, Honda S, Yoshino K, and Suzuki K. (1990). A methyl viologen-sensitive mutant of the nematode Caenorhabditis elegans. Mutat Res Dnaging 237, 165–171. 10.1016/0921-8734(90)90022-j. [DOI] [PubMed] [Google Scholar]
- 25.Shpilka T, Du Y, Yang Q, Melber A, Naresh NU, Lavelle J, Kim S, Liu P, Weidberg H, Li R, et al. (2021). UPRmt scales mitochondrial network expansion with protein synthesis via mitochondrial import in Caenorhabditis elegans. Nat. Commun. 12, 479. 10.1038/s41467-020-20784-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Haynes CM, Yang Y, Blais SP, Neubert TA, and Ron D. (2010). The Matrix Peptide Exporter HAF-1 Signals a Mitochondrial UPR by Activating the Transcription Factor ZC376.7 in C. elegans. Mol Cell 37, 529–540. 10.1016/j.molcel.2010.01.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.MARRES CAM, VRIES S, and GRIVELL LA (1991). Isolation and inactivation of the nuclear gene encoding the rotenone-insensitive internal NADH: ubiquinone oxidoreductase of mitochondria from Saccharomyces cerevisiae. Eur J Biochem 195, 857–862. 10.1111/j.1432-1033.1991.tb15775.x. [DOI] [PubMed] [Google Scholar]
- 28.Cossard R., Esposito M., Sellem CH., Pitayu L., Vasnier C., Delahodde A., and Dassa EP. (2015). Caenorhabditis elegans expressing the Saccharomyces cerevisiae NADH alternative dehydrogenase Ndi1p, as a tool to identify new genes involved in complex I related diseases. Frontiers Genetics 6, 206. 10.3389/fgene.2015.00206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.DeCorby A, Gášková D, Sayles LC, and Lemire BD (2007). Expression of Ndi1p, an alternative NADH:ubiquinone oxidoreductase, increases mitochondrial membrane potential in a C. elegans model of mitochondrial disease. Biochimica Et Biophysica Acta Bba - Bioenergetics 1767, 1157–1163. 10.1016/j.bbabio.2007.07.003. [DOI] [PubMed] [Google Scholar]
- 30.McElroy GS, Reczek CR, Reyfman PA, Mithal DS, Horbinski CM, and Chandel NS (2020). NAD+ Regeneration Rescues Lifespan, but Not Ataxia, in a Mouse Model of Brain Mitochondrial Complex I Dysfunction. Cell Metab 32, 301–308.e6. 10.1016/j.cmet.2020.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ohh M, Park CW, Ivan M, Hoffman MA, Kim T-Y, Huang LE, Pavletich N, Chau V, and Kaelin WG (2000). Ubiquitination of hypoxia-inducible factor requires direct binding to the β-domain of the von Hippel–Lindau protein. Nat Cell Biol 2, 423–427. 10.1038/35017054. [DOI] [PubMed] [Google Scholar]
- 32.Govindan JA, Jayamani E, and Ruvkun G. (2019). ROS-based lethality of Caenorhabditis elegans mitochondrial electron transport mutants grown on Escherichia coli siderophore iron release mutants. P Natl Acad Sci Usa 116, 21651–21658. 10.1073/pnas.1912628116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Chen S, Luo S, Zhang Z, and Ma DK (2019). VHL-1 inactivation and mitochondrial antioxidants rescue C. elegans dopaminergic neurodegeneration. Protein Cell 10, 610–614. 10.1007/s13238-019-0621-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Dancy BM, Brockway N, Ramadasan-Nair R, Yang Y, Sedensky MM, and Morgan PG (2016). Glutathione S-transferase mediates an ageing response to mitochondrial dysfunction. Mech Ageing Dev 153, 14–21. 10.1016/j.mad.2015.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Dalleau S, Baradat M, Guéraud F, and Huc L. (2013). Cell death and diseases related to oxidative stress:4-hydroxynonenal (HNE) in the balance. Cell Death Differ 20, 1615–1630. 10.1038/cdd.2013.138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Kruse SE, Watt WC, Marcinek DJ, Kapur RP, Schenkman KA, and Palmiter RD (2008). Mice with mitochondrial complex I deficiency develop a fatal encephalomyopathy. Cell Metabolism 7, 312–320. 10.1016/j.cmet.2008.02.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Adjobo-Hermans MJW., Haas R. d e, Willems PHGM., Wojtal A., Vries S E. van E., Wagenaars JA, Brand M. van den, Rodenburg RJ, Smeitink JAM, Nijtmans LG, et al. (2020). NDUFS4 deletion triggers loss of NDUFA12 in Ndufs4 −/− mice and Leigh syndrome patients: A stabilizing role for NDUFAF2. Biochimica Et Biophysica Acta Bba - Bioenergetics 1861, 148213. 10.1016/j.bbabio.2020.148213. [DOI] [PubMed] [Google Scholar]
- 38.Ast T, Meisel JD, Patra S, Wang H, Grange RMH, Kim SH, Calvo SE, Orefice LL, Nagashima F, Ichinose F, et al. (2019). Hypoxia Rescues Frataxin Loss by Restoring Iron Sulfur Cluster Biogenesis. Cell 177, 1507–1521.e16. 10.1016/j.cell.2019.03.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Baik AH, Haribowo AG, Chen X, Queliconi BB, Barrios AM, Garg A, Maishan M, Campos AR, Matthay MA, and Jain IH (2023). Oxygen toxicity causes cyclic damage by destabilizing specific Fe-S cluster-containing protein complexes. Mol. Cell 83, 942–960.e9. 10.1016/j.molcel.2023.02.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Angerer H. (2015). Eukaryotic LYR Proteins Interact with Mitochondrial Protein Complexes. Biology 4, 133–150. 10.3390/biology4010133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Angerer H, Radermacher M, Mańkowska M, Steger M, Zwicker K, Heide H, Wittig I, Brandt U, and Zickermann V. (2014). The LYR protein subunit NB4M/NDUFA6 of mitochondrial complex I anchors an acyl carrier protein and is essential for catalytic activity. Proceedings of the National Academy of Sciences 111, 5207–5212. 10.1073/pnas.1322438111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Chou PY, and Fasman GD (1979). Advances in Enzymology and Related Areas of Molecular Biology. John Wiley & Sons, Inc. “Prediction of the Secondary Structure of Proteins from their Amino Acid Sequence” Editor: Alton Meister. 10.1002/9780470122921.ch2. [DOI] [PubMed] [Google Scholar]
- 43.Kampjut D, and Sazanov LA (2020). The coupling mechanism of mammalian respiratory complex I. Sci New York N Y 370. 10.1126/science.abc4209. [DOI] [PubMed] [Google Scholar]
- 44.Yoga EG, Parey K, Djurabekova A, Haapanen O, Siegmund K, Zwicker K, Sharma V, Zickermann V, and Angerer H. (2020). Essential role of accessory subunit LYRM6 in the mechanism of mitochondrial complex I. Nat Commun 11, 6008. 10.1038/s41467-020-19778-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Zhu J, Vinothkumar KR, and Hirst J. (2016). Structure of mammalian respiratory complex I. Nature 536, 354–358. 10.1038/nature19095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Fiedorczuk K, Letts JA, Degliesposti G, Kaszuba K, Skehel M, and Sazanov LA (2016). Atomic structure of the entire mammalian mitochondrial complex I. Nature 538, 406–410. 10.1038/nature19794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Zickermann V, Wirth C, Nasiri H, Siegmund K, Schwalbe H, Hunte C, and Brandt U. (2015). Mechanistic insight from the crystal structure of mitochondrial complex I. Science 347, 44–49. 10.1126/science.1259859. [DOI] [PubMed] [Google Scholar]
- 48.Mao K, Ji F, Breen P, Sewell A, Han M, Sadreyev R, and Ruvkun G. (2019). Mitochondrial Dysfunction in C. elegans Activates Mitochondrial Relocalization and Nuclear Hormone Receptor-Dependent Detoxification Genes. Cell Metabolism 29, 1182–1191.e4. 10.1016/j.cmet.2019.01.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Gabaldón T, Rainey D, and Huynen MA (2005). Tracing the Evolution of a Large Protein Complex in the Eukaryotes, NADH:Ubiquinone Oxidoreductase (Complex I). J. Mol. Biol. 348, 857–870. 10.1016/j.jmb.2005.02.067. [DOI] [PubMed] [Google Scholar]
- 50.Jain IH., Calvo SE., Markhard AL., Skinner OS., To T-L., Ast T., and Mootha VK. (2019). Genetic Screen for Cell Fitness in High or Low Oxygen Highlights Mitochondrial and Lipid Metabolism. Cell 181, 716–727.e11. 10.1016/j.cell.2020.03.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.King MP, and Attardi G. (1989). Human cells lacking mtDNA: repopulation with exogenous mitochondria by complementation. Science 246, 500–503. 10.1126/science.2814477. [DOI] [PubMed] [Google Scholar]
- 52.Tocilescu MA, Fendel U, Zwicker K, Kerscher S, and Brandt U. (2007). Exploring the Ubiquinone Binding Cavity of Respiratory Complex I*. J Biol Chem 282, 29514–29520. 10.1074/jbc.m704519200. [DOI] [PubMed] [Google Scholar]
- 53.Angerer H, Nasiri HR, Niedergesäß V, Kerscher S, Schwalbe H, and Brandt U. (2012). Tracing the tail of ubiquinone in mitochondrial complex I. Biochimica Et Biophysica Acta Bba - Bioenergetics 1817, 1776–1784. 10.1016/j.bbabio.2012.03.021. [DOI] [PubMed] [Google Scholar]
- 54.Sackmann U, Zensen R, Röhlen D, Jahnke U, and Weiss H. (1991). The acyl-carrier protein in Neurospora crassa mitochondria is a subunit of NADH:ubiquinone reductase (complex I). Eur. J. Biochem. 200, 463–469. 10.1111/j.1432-1033.1991.tb16205.x. [DOI] [PubMed] [Google Scholar]
- 55.Mikolajczyk S, and Brody S. (1990). De novo fatty acid synthesis mediated by acyl-carrier protein in Neurospora crassa mitochondria. Eur. J. Biochem. 187, 431–437. 10.1111/j.1432-1033.1990.tb15322.x. [DOI] [PubMed] [Google Scholar]
- 56.Vranken JGV, Jeong M-Y, Wei P, Chen Y-C, Gygi SP, Winge DR, and Rutter J. (2016). The mitochondrial acyl carrier protein (ACP) coordinates mitochondrial fatty acid synthesis with iron sulfur cluster biogenesis. Elife 5, 174. 10.7554/elife.17828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Hou T, Zhang R, Jian C, Ding W, Wang Y, Ling S, Ma Q, Hu X, Cheng H, and Wang X. (2019). NDUFAB1 confers cardio-protection by enhancing mitochondrial bioenergetics through coordination of respiratory complex and supercomplex assembly. Cell Res 29, 754–766. 10.1038/s41422-019-0208-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Canfield DE (2005). THE EARLY HISTORY OF ATMOSPHERIC OXYGEN: Homage to Robert M. Garrels. Annu. Rev. Earth Planet. Sci. 33, 1–36. 10.1146/annurev.earth.33.092203.122711. [DOI] [Google Scholar]
- 59.Müller M., Mentel M., Hellemond JJ. van Henze K, Woehle C, Gould SB, Yu R-Y, Giezen M. van der, Tielens AGM, and Martin WF (2012). Biochemistry and evolution of anaerobic energy metabolism in eukaryotes. Microbiol. Mol. Biol. Rev. : MMBR 76, 444–495. 10.1128/mmbr.05024-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Andreini C, Banci L, and Rosato A. (2016). Exploiting Bacterial Operons To Illuminate Human Iron-Sulfur Proteins. J. Proteome Res. 15, 1308–1322. 10.1021/acs.jproteome.6b00045. [DOI] [PubMed] [Google Scholar]
- 61.Brenner S. (1974). The genetics of Caenorhabditis elegans. Genetics 77, 71–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Quintana A, Kruse SE, Kapur RP, Sanz E, and Palmiter RD (2010). Complex I deficiency due to loss of Ndufs4 in the brain results in progressive encephalopathy resembling Leigh syndrome. Proc. Natl. Acad. Sci. 107, 10996–11001. 10.1073/pnas.1006214107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Ghanta KS, and Mello CC (2020). Melting dsDNA Donor Molecules Greatly Improves Precision Genome Editing in Caenorhabditis elegans. Genetics 216, 643–650. 10.1534/genetics.120.303564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64.Arribere JA, Bell RT, Fu BXH, Artiles KL, Hartman PS, and Fire AZ (2014). Efficient marker-free recovery of custom genetic modifications with CRISPR/Cas9 in Caenorhabditis elegans. Genetics 198, 837–846. 10.1534/genetics.114.169730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65.Frøkjær-Jensen C, Davis MW, Sarov M, Taylor J, Flibotte S, LaBella M, Pozniakovsky A, Moerman DG, and Jorgensen EM (2014). Random and targeted transgene insertion in Caenorhabditis elegans using a modified Mos1 transposon. Nat. Methods 11, 529–534. 10.1038/nmeth.2889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Lehrbach NJ, Ji F, and Sadreyev R. (2017). Next-Generation Sequencing for Identification of EMS-Induced Mutations in Caenorhabditis elegans. Curr Protoc Mol Biology 117, 7.29.1–7.29.12. 10.1002/cpmb.27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Yin Z, Burger N, Kula-Alwar D, Aksentijević D, Bridges HR, Prag HA, Grba DN, Viscomi C, James AM, Mottahedin A, et al. (2021). Structural basis for a complex I mutation that blocks pathological ROS production. Nat Commun 12, 707. 10.1038/s41467-021-20942-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
A. Growth of wild type animals for 2 days at room temperature exposed to differing concentrations of the complex I inhibitor rotenone. B. Multiple sequence alignment of NDUFS2/GAS-1 (C. elegans residues 284–299) and NDUFS7/NDUF-7 (C. elegans residues 182–199) including homologs from mammals, invertebrates, and fungi made using ClustalW. C. Mean intestinal fluorescence of hsp-6::gfp in age-matched animals grown at 21% or 1% oxygen for one generation. All genotypes were imaged at L4/young adult stage except nuo-2(tm5258) which arrest at L2 stage. D. Growth of animals for 2 days at 20°C. E. Growth of animals for 2 days at room temperature. F-G. Growth of animals for 2 days (left) and 3 days (right) at room temperature. In all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
A. Growth of animals following L1 synchronization at 21% oxygen (black), 50% oxygen (orange), or 100% oxygen (red) incubated at 20°C. ‘x’ indicates animals were dead. B. Growth of animals following L1 synchronization at 21% oxygen (black) or 50% oxygen for 2–6 days and then shifted to 21% oxygen (orange) at 20°C. Grey arrows indicate the day animals were shifted from 50% to 21% oxygen. C. Growth of animals for 3 days at 50% oxygen incubated at 20°C. D. Ovine complex I (PDB: 6ZKC43) in closed conformation. The NDUFA5/NDUA-5(R126Q) suppressor mutation lies at the interface of NDUFA5 and NDUFA10. E. Images of animals after 1 generation growth at 1% or 21% oxygen at room temperature. Magnification = 7x. F-G. Growth of animals for 3 days at room temperature. H. Growth of animals for 2 days at 50% oxygen followed by 1 day at 21% oxygen. I. Growth of animals for 5 days at room temperature. J. Growth of animals for 2 days at room temperature. K. Growth of animals at 50% oxygen for 4 days at room temperature, or growth at 100% oxygen for 2 days followed by 21% oxygen for 2 days at room temperature. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
A. Fluorescent images of L4 stage animals containing gst-4::gfp grown continuously at 21% oxygen or exposed to 100% oxygen for 1 day. Images were acquired at 69x magnification with an exposure time of 50 ms. B. Growth of animals for 3 days at 20°C at 21% or 50% oxygen. Statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. *** = p value <0.001.
A. Blue Native PAGE of isolated mitochondria purified from animals grown at 21% or 1% oxygen for 4 days. * denotes faint band that may correspond to CI subcomplex. B. BN-PAGE of isolated mitochondria followed by western blot from animals grown continuously at 21% oxygen. Black arrow indicates CI subcomplexes observed in nduf-7(et19) which run at the same size as complex V dimers. C. BN-PAGE of isolated mitochondria purified from animals grown at 21% or 50% oxygen for 2 days. D-E. TMT quantitative proteomics from animals exposed to 50% oxygen for 2 days. Plotted are log 2-fold ratios of all proteins from which at least two peptides were quantified. F. SDS-PAGE of whole worm lysate from animals exposed to 21% or 50% oxygen. Anti-lipoic acid antibodies recognize the modified E2 subunits of PDH (DLAT) and OGDH/KGDH (DLST). G-H. TMT quantitative proteomics from animals exposed to 50% oxygen for 2 days. Plotted are log 2-fold ratios of all proteins from which at least two peptides were quantified. Iron-sulfur cluster containing C. elegans proteins were identified by homology to a previously published list60.
A. Ratio of reduced CoQ9H2 to oxidized CoQ9 as determined by mass spectrometry. Samples were extracted from whole worms grown continuously at 21% oxygen. B. Mass spectrometry traces of CoQ9H2 and CoQ9 in wild type and clk-1(qm30) backgrounds. clk-1 mutants can survive using dietary CoQ8 from their bacterial food. C. SDS-PAGE followed by western blot of C. elegans mitochondrial membranes used in in vitro assays. D. Growth of animals for 3 days at room temperature at 21% oxygen. E. SDS-PAGE followed by western blot of brains from wild type or Ndufs4 mice exposed to 21% oxygen, 11% oxygen from weaning (day 25–30), or shifted from 21% to 11% oxygen after 55 days. F. SDS-PAGE followed by western blot of whole worm lysate from animals grown continuously at 1% oxygen and then exposed to 1%, 21%, or 50% oxygen for 3 days. G. Growth of animals for 3 days at room temperature. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test. Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
A. Growth of animals for 4 days at 50% oxygen. B. Growth of animals for 3 days at 21% or 1% oxygen at room temperature. C. Growth of animals for 4 days at 21% oxygen at 20°C. D. Multiple sequence alignment of NDUFA6/NUO-3 homologs from nematodes (above) and eukaryotes (below) made using ClustalW. The alignment corresponds to amino acids 33–83 of C. elegans NUO-3. E. Growth of animals for 3 days at 21% oxygen. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test or Dunnett’s Multiple Comparison Test (D). Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
A. Growth of animals for 3 days at 21% oxygen with 0, 0.1, or 1.0 μM rotenone. B. Growth of animals for 2 days at 21% oxygen with 0, 1.0, or 5.0 μM rotenone incubated at 20°C. C. Growth of animals for 2 days at 21% oxygen incubated at room temperature with 0 or 1.0 μM rotenone. D. Multiple sequence alignment of NDUFS2/GAS-1 (C. elegans residues 153–164) and NDUFS7/NDUF-7 (C. elegans residues 70–81…114–121) including homologs from animals, fungi, and bacteria made using ClustalW. E. Growth of animals for 2 days at 21% oxygen incubated at 20°C. F. Growth of animals for 2 days at 50% oxygen incubated at 20°C. G. Growth of animals for 2 days at 21% oxygen incubated at room temperature. H. SDS-PAGE followed by western blot of C. elegans mitochondrial membranes used in in vitro assays. I. Complex I-dependent oxidation of NADH by isolated mitochondrial membranes at 21% oxygen. NADH absorbs light at 340 nm. J. Growth of animals for 2 days at 21% or 1% oxygen incubated at room temperature. For all panels statistical significance was calculated using one-way ANOVA followed by Tukey’s Multiple Comparison Test (A, D-F, H) or Dunnett’s Multiple Comparison Test (B). Error bars represent standard deviation. n.s. = not significant, * = p value <0.05, ** = p value <0.01, *** = p value <0.001.
Table S3: Quantitative TMT Proteomics from Complex I mutants at 1%, 21%, and 50% oxygen (raw data), related to Figure 4
Data Availability Statement
Proteomics datasets are available in Table S3. All data reported in this paper will be shared by the lead contact upon request.
This paper does not report original code.
Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.