• neurodegenerative disease;
  • Leigh syndrome;
  • fibroblasts;
  • knockout mouse;
  • validity


  1. Top of page
  2. Abstract
  3. Introduction
  4. Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts
  5. Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models
  6. The Validity of Cellular and Animal Models
  7. Acknowledgements
  8. References

To allow the rational design of effective treatment strategies for human mitochondrial disorders, a proper understanding of their biochemical and pathophysiological aspects is required. The development and evaluation of these strategies require suitable model systems. In humans, inherited complex I (CI) deficiency is one of the most common deficiencies of the mitochondrial oxidative phosphorylation system. During the last decade, various cellular and animal models of CI deficiency have been presented involving mutations and/or deletion of the Ndufs4 gene, which encodes the NDUFS4 subunit of CI. In this review, we discuss these models and their validity for studying human CI deficiency. © 2013 IUBMB Life, 65(3):202–208, 2013


  1. Top of page
  2. Abstract
  3. Introduction
  4. Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts
  5. Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models
  6. The Validity of Cellular and Animal Models
  7. Acknowledgements
  8. References

Understanding the cellular and biological consequences of mutations in mitochondrial protein-encoding genes is of prime importance for the development of rational treatment strategies for mitochondrial disorders. This requires proper cell and animal models to examine the genetic, biochemical, and pathophysiological aspects of the disease. During the last decade, advances made in mammalian genetics have not only shed more light on the pathophysiology of mitochondrial disorders but also increased the complexity of treatment development. In this sense, it is a challenge to select a model that is optimally compatible to the human situation with respect to cellular and biological pathophysiology. This review focuses on cellular and animal models of (isolated) complex I (CI) deficiency, which is the most common disorder of the mitochondrial oxidative phosphorylation (OXPHOS) system in humans (1, 2). Furthermore, we limit ourselves to models that involve mutations in, or deletion of, the Ndufs4 gene. The latter encodes an 18-kDa protein, the NDUFS4 (NADH dehydrogenase [ubiquinone] iron-sulfur protein 4) accessory subunit of CI (OMIM 602694) (3–5). In humans, CI consists of 44 different subunits that combine into the CI holo-complex to form a ∼1 MDa assembly (2, 6). Seven CI subunits originate from the mitochondrial DNA (mtDNA) and the remainder from the nuclear DNA (nDNA). Functionally, CI constitutes part of the electron transport chain by extracting electrons from NADH and donating them to ubiquinone (1, 2). The energy released by this reaction is used to expel protons from the mitochondrial matrix across the mitochondrial inner membrane. In this way, CI contributes to the maintenance of the matrix-directed proton-motive force, which fuels mitochondrial ATP production and many other mitochondrial functions (1, 2). According to the most recent CI assembly model, subunits of the dehydrogenase, hydrogenase, and proton-translocation module assemble via several intermediates into the L-shaped holo enzyme (see ref.7 and the references therein). During the CI assembly process, the NDUFS4 subunit appears to be incorporated a relatively late stage (8, 9). Evidence was provided that several CI subunits can be phosphorylated by cyclic AMP (cAMP)-dependent protein kinase A (PKA) (10). In case of NDUFS4, its phosphorylation site has been localized at the C-terminal end of the protein (amino acids 171–173: RVSTK) in human fibroblasts (11). Phosphorylation of RVSTK appears to promote import and maturation of the precursor protein (12), thereby affecting CI assembly and activity (5). Below, we first summarize the results obtained with patient-derived fibroblasts ensuing from Ndufs4 mutations and/or deletion (Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts section). Next, we provide an overview of the current animal models (Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models section). Finally, we discuss the validity of the cell and animal models using established criteria (The Validity of Cellular and Animal Models section).

Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts

  1. Top of page
  2. Abstract
  3. Introduction
  4. Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts
  5. Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models
  6. The Validity of Cellular and Animal Models
  7. Acknowledgements
  8. References

In humans, the first nuclear-encoded pathogenic mutation in CI (i.e., an AAGTC duplication at position 466–470 in exon 5) was identified in the NDUFS4 subunit (13). Since then, several other recessive Ndufs4 mutations have been identified resulting in CI deficiency (OMIM 252010): a single base deletion at position 289/290 in exon 3, a C316T transition in the same exon, a G44A nonsense mutation in exon 1, and a nonsense mutation (IVS1nt -1) in intron 1 of the NDUFS4 gene, all of which result in the absence of the NDUFS4 protein. In total, 14 NDUFS4 patients have currently been described in the literature (14, 15). Clinically, these patients develop Leigh(-like) syndrome (LS) (6, 15, 16). LS is a multisystemic progressive neurodegenerative disorder of infancy. The disease is characterized by bilateral symmetric lesions in various brain regions, in particular the basal ganglia and brainstem. Children are typically born following normal pregnancy, but may then display failure to thrive and develop motor coordination dysfunctions, episodic vomiting, blindness, hypotonia, seizures, and ataxia (17, 18). Sadly, LS is fatal and most patients succumb to the disease before they reach the age of 2 years (6, 16, 19). For diagnostic purposes, genetic and biochemical analysis of LS patients are commonly carried out using muscle biopsies and patient primary skin fibroblasts (3, 20). The latter cells can be cultured (albeit for a limited number of passages) to study the cellular pathophysiology of Ndufs4 mutations (21–24).

CI activity in fibroblasts from identified NDUFS4 patients analyzed so far ranges from 15% up to 75% enzyme activity of the lowest control value (15, 22). During recent years, our group has focused on determining the cellular consequences of mutations in nDNA-encoded CI subunits, including NDUFS4 (2, 6, 25). Native gel analysis of patient fibroblasts revealed that Ndufs4 mutations induce CI assembly defects resulting in formation of a ∼830 kDa CI subcomplex lacking the NADH dehydrogenase module (26, 27). In addition, Ndufs4 mutations also influence CIII assembly, compatible with the reduced CIII activity observed in patient fibroblasts (3).

At the single-cell level, fibroblasts of patients with CI deficiency displayed aberrations in mitochondrial membrane potential (Δψ), mitochondrial NAD(P)H levels, reactive oxygen species (ROS) levels, Ca2+ signaling, ATP homeostasis, and mitochondrial morphology (16, 21, 22, 24). Explorative data analysis (EDA), consisting of a principal component analysis and hierarchical clustering analysis, was carried out on 26 cellular parameters obtained from 24 individual patients with CI deficiency (five of which were NDUFS4 patients). The EDA strategy revealed a connection between residual CI activity, cellular ROS levels, and mitochondrial morphology/function (28). Interestingly, compared to cells from patients with a relatively late age-of-onset (AOO)/age-of-death (AOD), fibroblasts from patients displaying an early AOO/AOD exhibited lower CI residual activity, more dramatically increased ROS and NAD(P)H levels, and mitochondrial fragmentation (28). This suggests that the cellular aberrations in the fibroblast model correlate with the clinical phenotype. Conversely, analysis of a cohort of 130 patients with CI deficiency (14 of which displayed an Ndufs4 mutation) revealed no correlation between the AOO/AOD of the patients and residual CI activity in the patient fibroblasts (15). This suggests that sole analysis of residual CI activity is not predictive of disease prognosis in CI deficient patients.

In a recent study, stimulation of PKA-mediated NDFUS4 phosphorylation rescued CI activity during oxidative stress by stimulating import of NDUFS4 precursors and exchange of damaged subunits in CI with de novo generated subunits (29). This agrees with another study indicating that newly formed nDNA-encoded CI subunits rapidly incorporate into holo-CI in a subunit-specific manner (7). However, although cellular ROS levels were increased, we did not detect alterations in glutathione metabolism or increased lipid peroxidation in CI-deficient patient fibroblasts (22). Therefore, the increased ROS levels in these fibroblasts might fulfill a signaling role (2, 30, 31). Indeed, transcriptional analysis suggests that the absence of ROS-mediated damage might be due to upregulation of antioxidant defense systems and involves Keap1-Nrf2, glutathione, thioredoxin, and peroxiredoxin (32).

Typically, OXPHOS dysfunction mostly affects cell types that heavily rely on mitochondrial ATP generation. This means that mitochondrial disorders are often associated with malfunction of the brain, eyes, heart, liver, kidney, and muscle (25). Calcium (Ca2+) homeostasis and signaling constitute important ATP-dependent processes in most cells. CI-deficient patient fibroblasts displayed aberrations in Ca2+ and ATP handling upon stimulation with the hormone bradykinin (Bk) (33). In the case of Ndufs4 mutations, no change in resting free cytosolic Ca2+ concentration ([Ca2+]c) was observed, but the Ca2+ content of the ER (ERCa) was slightly (but significantly) reduced (24, 33). This reduction was associated with a lower Bk-induced increase in [Ca2+]c, a reduced increase in mitochondrial free Ca2+ concentration ([Ca2+]m) and a smaller elevation in mitochondrial free ATP concentration ([ATP]m). As a consequence, active Ca2+ removal from the cytosol was 10–70% slower in NDUFS4 cells during Bk stimulation (24, 33). We hypothesized that Ndufs4 mutations result in a small but significant Δψ depolarization, causing a reduction in mitochondrial ATP production capacity. In resting cells, this diminished capacity reduces the ATP supply to the Ca2+ pump on the endoplasmic reticulum (the sarco(endo)plasmic reticulum Ca2+-ATPase or SERCA). Upon cell stimulation with Bk, the ensuing reduction in ERCa results in a lower peak of [Ca2+]c and less Ca2+ is taken up by the mitochondria (i.e., a lower peak [Ca2+]m). As a consequence, Ca2+-stimulated mitochondrial ATP production is decreased and mitochondrial ATP supply to the SERCAs is hampered, leading to a slower removal of Ca2+ from the cytosol (24, 33). In agreement with these results, repetitive cytosolic Ca2+ transients induced by a train of glutamate stimulations decayed slower in differentiated neurons generated from mouse embryonic stem cells harboring mutations in mtDNA-encoded CI subunits (i.e., m.13887Cins in Mtnd6 and m.12273G>A in Mtnd5) compared to control neurons (34).

Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models

  1. Top of page
  2. Abstract
  3. Introduction
  4. Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts
  5. Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models
  6. The Validity of Cellular and Animal Models
  7. Acknowledgements
  8. References

The NDUFS4 Whole-Body Knockout Mouse

The first animal model for NDUFS4 dysfunction, the Ndufs4 whole-body knockout (WBKO) mouse, was recently presented (35). In this model, exon 2 of the Ndufs4 gene is deleted, mimicking a frame shift mutation. The phenotype of WBKO mice includes loss of hair and vision, ataxia, motor dysfunctions, and, ultimately, death before adulthood is reached due to severe decreases in lung function and encephalomyopathy (36). Examination of the brain in WBKO mice revealed that deletion of the Ndufs4 gene is associated with neurodegeneration in the motor centers, including the cerebellum, basal ganglia, and brainstem (35). Studying the effects of therapeutics in WBKO mice by means of behavioral testing is therefore challenging, not only given the fact that these animals usually do not survive for more than 50 days after birth but also because these animals show severe motor impairments at 5–6 weeks of age. Analysis of muscle fibers revealed that Ndufs4 KO muscles were essentially normal. However, NADH oxidase activity was significantly decreased in glycolytic and oxidative muscle fibers from KO animals (35). This decrease probably relates to the assembly of the CI holo-enzyme, which was hampered in various tissues of the WBKO mouse (37). Similar to NDFUS4 patient fibroblasts, WBKO mouse tissues (i.e., pancreas, kidney, liver, lung, brain, heart, and muscle) contained an inactive ∼830 kD CI subassembly (native gel analysis) and displayed a maximal CI activity up to 44% of wild-type mice (37). However, we also detected a ∼200 kDa CI subcomplexes that did display enzymatic activity. Taken together, our analysis of the WBKO tissues led us to propose that the ∼830 kDa CI subassembly lacks the tip (i.e., the ∼200 kDa subassembly) of CI, and that this tip represents the NADH dehydrogenase module (37).

To further enhance our understanding of the cellular consequences of Ndufs4 gene deletion, we generated immortalized mouse embryonic fibroblasts (MEFs) from wt and WBKO mice (38). In contrast to NDUFS4 patient fibroblasts and WBKO tissue samples, Ndufs4 KO MEFs displayed virtually no CI activity. Western blot analysis revealed that mitochondria-enriched fractions from WBKO MEFs did not contain detectable amounts of NDUFS4 protein. Similar to WBKO tissue samples, Ndufs4 KO MEFs also contained a ∼830 kDa CI subassembly (enzymatically inactive) and a ∼200 kDa CI subcomplexes (active). The levels of other fully assembled OXPHOS complexes (i.e., complexes II–V) were not affected by Ndufs4 gene deletion, although the activity of CII, CIII, and CIV was reduced in Ndufs4 KO MEFs. Intact KO cells exhibited a moderate reduction in routine and maximal oxygen (O2) consumption, which was fully inhibited by acute application of the CI inhibitor rotenone. The latter demonstrates that catalytically active CI is present in Ndufs4 KO MEFs (38). Additionally, a reduction in NAD+ levels and increase in NADH levels were observed, paralleled by a decreased and increased extracellular glucose concentration and lactic acid, respectively. This suggests that Ndufs4 KO MEFs are slightly more glycolytic than wt MEFs, although no differences in key glycolytic proteins (i.e., hexokinases and phosphorylated pyruvate dehydrogenase) were observed between wt and Ndufs4 KO MEFs (38). Integrating the above results obtained from tissues and MEFs from the WBKO animal, we hypothesized that CIII stabilizes CI in the absence of the NDUFS4 subunit (37), and that due to the metabolic properties of the immortalized KO MEFs, Ndufs4 gene deletion has only modest effects in these cells (38).

NDUFS4-Point Mutation Mice, NDUFS4 Purkinje Cell-Knockout Mice, and Neuron- and Glia-Specific Knockout Mice

Ndufs4-point mutation (PM) mice were generated by introducing a PM in the Ndufs4 gene, leading to a loss of the last 10–15 amino acids of the last exon of the NDUFS4 protein (39). Animals heterozygous for the Ndufs4-PM displayed lower CI activity, reduced CI-mediated O2 consumption, and increased organ lactate levels; the homozygous genotype, however, was embryonic lethal. To date, the clinical phenotype of the heterozygous Ndufs4-PM does not mimic human symptoms of LS. The Ndufs4-Purkinje Cell (PC) and neuron and glia-specific knockout (NesKO) mouse strains were created to study the brain-specific effects of Ndufs4 KO. The clinical phenotype of these mice is very similar to that of WBKO (i.e., not brain-specific) animals (40). NesKO mice display neuronal damage in brain areas associated with movement, including the cerebellum, striatum, and basal ganglia (41), while Ndufs4-PC mice show enhanced neuronal damage in the PCs of the cerebellum. However, Ndufs4-PC mice only show mild behavioral and neuropathological abnormalities, compared to the WBKO mouse (42).

The Validity of Cellular and Animal Models

  1. Top of page
  2. Abstract
  3. Introduction
  4. Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts
  5. Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models
  6. The Validity of Cellular and Animal Models
  7. Acknowledgements
  8. References

To test the validity of the animal and cellular models described above, we used the following criteria: (i) Do these models display mitochondrial dysfunction as a result of Ndufs4 mutations (“face validity”), (ii) are these models predictive with respect to disease progression of patients with Ndufs4 mutations (“predictive validity”), and (iii) to what extent can these models be used to develop therapeutic interventions for CI deficiency (“construct validity”)?

The Validity of Patient Fibroblasts and NDUFS4 KO Mouse-Derived Cells as a Model

The best line of evidence for diagnosis and molecular investigation into disease pathology comes from isolated primary human skin fibroblasts. Patient fibroblasts demonstrate good face validity in residual CI activity, CI complex assembly, and OXPHOS dysfunction similar to that observed in muscle biopsies. Extensive live cell analysis of patient fibroblasts further demonstrates that the defects caused by mutations in CI subunits have effects on mitochondrial function that reasonably correlate with the clinical phenotype (i.e., AOO and AOD) of the patients.

Patient fibroblasts display heterogeneity between mutations as well as heterogeneity in genetic background unlike the genetically engineered models. However, in situ fibroblasts are commonly not affected in CI deficiency and might become increasingly glycolytic under standard culture conditions. Voets et al. (32) demonstrated specific changes when patient cells were cultured under OXPHOS-stimulated (i.e., “galactose medium”) conditions when compared with “normal” (i.e., “glucose medium”) conditions. However, it remains to be established how these changes translate to brain and muscle tissue in patients. In terms of validity, the adaptation of patient fibroblasts to culture conditions and the CI mutation affect the construct validity. However, stimulation of adaptive programs by exogenous means might be important for future drug development. It should be kept in mind that therapeutic interventions in CI deficient patient fibroblasts might not de facto result in improvement of the phenotype in resting cells due to (a downregulated?) contribution of OXPHOS during routine cell culture, but could have a significant influence in active cells or under more OXPHOS-dependent conditions.

In terms of validation, MEFs obtained from the Ndufs4 KO mouse demonstrate face validity similar to that observed in patient fibroblasts, although Ndufs4 KO MEFS do not demonstrate increased ROS production as is seen in patient fibroblasts (Table 1). Neuronal cultures obtained from Ndufs4 KO mice also showed no increased ROS levels (43, 44), but further support for the role of ROS and CI activity is shown by Valsecchi et al. (45), who found that isolated CI deficiency in primary muscle and skin fibroblasts from Ndufs4 KO mice is indeed associated with increased ROS levels, but that factors such as cell type and cellular metabolic state can have an effect on whether increases in ROS are detectable.

Table 1. Mammalian and cellular models for complex-I subunit NDUFS4-related disorders
inline image

A feature shared by all cellular models is the lack of a predictive validity. In particular, Ndufs4 KO MEFs fail to show predictive validity, as these cells become metabolically altered due to immortalization. Tissue extracts and cultured cells, primary or immortalized, should be considered as single time points and therefore the moment of “sampling” greatly determines how a cell will “perform,” which pertains not only to primary cells obtained from patients, but also to genetically engineered animal models. In contrast to the lack of predictive capability, cellular models do have a high degree of construct validity in the search for therapeutic interventions. This validity works two ways: first, identification of potential targets for therapeutic interventions and second, application of a potential therapeutic compound.

The Validity of NDUFS4 Knockout Mice as a Model

Not all mice are created equal. In their article examining the generation of appropriate mouse models for human disease, Rivera and Tessarollo (46) point out that while mice share several similarities with humans, genetically speaking, it is extremely difficult to generate a model that perfectly mimics human disease. As mutations affecting OXPHOS in man are extremely heterogeneous, it is difficult to generate a single mouse model that can accurately measure all aspects of the disease. The Ndufs4 KO mouse demonstrably shows face validity, as the mouse phenotype definitely mimics human symptomology: ataxia, failure to thrive, and ultimately, death at a young age, resulting from encephalomyopathy and respiratory failure (36). However, the mouse models available to date are far from perfect: the models mentioned above mainly show damage in the cerebellum, whereas most human patients show extensive damage in the basal ganglia (47).

Predictive validity presents more of a hurdle: how comparable is the disease progression of the mouse to that of man? Given the differences in brain morphology, it is more difficult to argue that the Ndufs4 knockout mouse is a perfect model for the disease. However, with our increasing knowledge of the human genome, it is now also becoming more common to create models that mimic only a specific symptom of the disease spectrum, which may be more beneficial, considering the extremely heterogeneous nature of Ndufs4 dysfunction-related diseases. The failure of scientists to produce animal models with the expected phenotypes has generated doubt as to the importance of in vivo testing. Part of this inability to produce robustly valid animal models may lie in the assumptions that a mouse will show similar symptoms as a man. The lack thereof may be due to several factors, perhaps in part due to differences in metabolism and genetic differences between mouse strains. Inbreeding has led to extreme genetic homogeneity, further distancing mice from man.

The Ndufs4 knockout models may prove themselves quite valuable in terms of construct validity. In several studies examining treatment with antioxidants and mitochondria-targeted treatments, many have found that exposure to these compounds had positive effects in the Ndufs4 mouse model (for a complete overview, please refer to (40)). An additional complication lies in the translation of preclinical pharmacokinetic data to clinical trials: finding the correct dose for human use, a dose that is not so low as to have no effect, but not so high as to be toxic, is a challenge that still plagues researchers today (48).

Further, knockout animal models cannot be applied to other human diseases unconditionally. A recent study of Parkinson's disease (PD) progression and therapeutic strategies using Ndufs4 KO mice (43, 44, 49, 50) demonstrated that although Ndufs4 KO mice display motor dysfunctions similar to PD patients, the Ndufs4 gene deletion does not affect dopaminergic neurons in the substantia nigra pars compactica in vivo. There were also no morphological aberrations seen in dopaminergic neurons in isolated primary mesencephalic cultures or increased basal levels of apoptosis or increased ROS levels. Moreover, whereas PD generally is associated with aging, Ndufs4 KO animals die before reaching adulthood. Therefore, in terms of validity, the Ndufs4 mouse demonstrates neither face predictive nor construct validity for the study of PD. Chemical inhibition of CI, however, did reveal off-target effects in the Ndufs4 KO. Therefore, this method may be more valid for testing of putative new PD therapies.

So the questions then remain: how valid are data obtained from animal and cellular studies, can we reliably translate results from these data to humans, and what is the future of these kinds of experimentation? Given the complexity of the brain, for example, there may never be an in silico model that can perfectly predict the disease progression and/or outcome of treatment with a putative new therapy. Therefore, cellular and animal studies will remain necessary, but the methods in which they are used may change considerably. In many cases, including Ndufs4 abnormalities, advances in genetics allow us to model specific aspects of human disease. These genetically modified animals are in many respects ideal, as they can be much more valid than currently available models, but in these times of rampant genetic modification, we are losing sight of the big picture: an animal model for diseases in their entirety. However, until we know the specific mechanisms behind genetic mutations, and their connections to the development of diseases, the generation and use of specific genetically altered animal and cellular models will most likely become the new norm for animal and cellular testing. Research into mitochondrial, neurodegenerative, and complex metabolic diseases has come a long way since the rise of genetics in the field of medicine. To date, significant quantities of genes and proteins involved in mitochondrial function have been identified. However, the biological consequences of these mutations, and the pathophysiological mechanism of these diseases, are less well defined. As isolated CI deficiencies often affect the brain, we are forced to use model systems to diagnose the disease, investigate disease progression and disease pathology, and to examine therapeutic interventions which ultimately should be used to treat patients.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts
  5. Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models
  6. The Validity of Cellular and Animal Models
  7. Acknowledgements
  8. References

This work was supported by the Centres for Systems Biology Research initiative (CSBR09/013V) of the “Nederlandse organisatie voor Wetenschappelijk Onderzoek” (NWO, Netherlands Organisation for Scientific Research).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Human CI Deficiency Due to NDUFS4 Mutations: Patient Fibroblasts
  5. Human CI Deficiency Due to NDUFS4 Mutations: Knockout Animal Models
  6. The Validity of Cellular and Animal Models
  7. Acknowledgements
  8. References
  • 1
    Smeitink, J. A. M., van den Heuvel, L. P., and DiMauro, S. ( 2001) The genetics and pathology of oxidative phosphorylation. Nat. Rev. Genet. 2, 342352.
  • 2
    Koopman, W. J., Nijtmans, L. G., Dieteren, C. E., Roestenberg, P., Valsecchi, F. et al. ( 2010) Mammalian mitochondrial complex I: biogenesis, regulation, and reactive oxygen species generation. Antioxid. Redox Signal. 12, 14311470.
  • 3
    Budde, S. M., van den Heuvel, L. P., Janssen, A. J., Smeets, R. J., Buskens, C. A. et al. ( 2000) Combined enzymatic complex I and III deficiency associated with mutations in the nuclear encoded NDUFS4 gene. Biochem. Biophys. Res. Commun. 275, 6368.
  • 4
    Petruzzella, V., Vergari, R., Puzziferri, I., Boffoli, D., Lamantea, E. et al. ( 2001) A nonsense mutation in the NDUFS4 gene encoding the 18 kDa (AQDQ) subunit of complex I abolishes assembly and activity of the complex in a patient with Leigh-like syndrome. Hum. Mol. Genet. 10, 529535.
  • 5
    Papa, S., Rasmo, D. D., Technikova-Dobrova, Z., Panelli, D., Signorile, A. et al. ( 2012) Respiratory chain complex I, a main regulatory target of the cAMP/PKA pathway is defective in different human diseases. FEBS Lett. 586, 568577.
  • 6
    Koopman, W. J. H., Distelmaier, F., and Smeitink, J. A. M. ( 2012) OXPHOS mutations and neurodegeneration. EMBO J. 13, PMID 23149385.
  • 7
    Dieteren, C. E., Koopman, W. J., Swarts, H. G., Peters, J. G., Maczuga, P. et al. ( 2012) Subunit-specific incorporation efficiency and kinetics in mitochondrial complex I homeostasis. J. Biol. Chem. 287, 4185141860.
  • 8
    Lazarou, M., McKenzie, M., Ohtake, A., Thorburn, D. R., and Ryan, M. T. ( 2007) Analysis of the assembly profiles for mitochondrial- and nuclear-DNA-encoded subunits into complex I. Mol. Cell. Biol. 27, 42284237.
  • 9
    Vogel, R. O., Smeitink, J. A. M., and Nijtmans, L. G. ( 2007) Human mitochondrial complex I assembly: a dynamic and versatile process. Biochim. Biophys. Acta 1767, 12151227.
  • 10
    Papa, S., Sardanelli, A. M., Cocco, T., Speranza, F., Scacco, S. C. et al. ( 1996) The nuclear-encoded 18 kDa (IP) AQDQ subunit of bovine heart complex I is phosphorylated by the mitochondrial cAMP-dependent protein kinase. FEBS Lett. 379, 299301.
  • 11
    De Rasmo, D., Palmisano, G., Scacco, S., Technikova-Dobrova, Z., Panelli, D. et al. ( 2010) Phosphorylation pattern of the NDUFS4 subunit of complex I of the mammalian respiratory chain. Mitochondrion 10, 464471.
  • 12
    Papa, S., Scacco, S., De Rasmo, D., Signorile, A., Papa, F. et al. ( 2010) cAMP-dependent protein kinase regulates post-translational processing and expression of complex I subunits in mammalian cells. Biochim. Biophys. Acta 1797, 649658.
  • 13
    van den Heuvel, L. P., Ruitenbeek, W., Smeets, R., Gelman-Kohan, Z., Elpeleg, O. et al. ( 1998) Demonstration of a new pathogenic mutation in human complex I deficiency: a 5-bp duplication in the nuclear gene encoding the 18-kD (AQDQ) subunit. Am. J. Hum. Genet. 62, 262268.
  • 14
    Iuso, A., Scacco, S., Piccoli, C., Bellomo, F., Petruzzella, V. et al. ( 2006) Dysfunctions of cellular oxidative metabolism in patients with mutations in the NDUFS1 and NDUFS4 genes of complex I. J. Biol. Chem. 281, 1037410380.
  • 15
    Koene, S., Rodenburg, R. J., van der Knaap, M. S., Willemsen, M. A., Sperl, W. et al. ( 2012) Natural disease course and genotype-phenotype correlations in Complex I deficiency caused by nuclear gene defects: what we learned from 130 cases. J. Inherit. Metab. Dis. 35, 737747.
  • 16
    Distelmaier, F., Visch, H. J., Smeitink, J. A. M., Mayatepek, E., Koopman, W. J. et al. ( 2009) The antioxidant Trolox restores mitochondrial membrane potential and Ca2+ -stimulated ATP production in human complex I deficiency. J. Mol. Med. (Berl.) 87, 515522.
  • 17
    Distelmaier, F., Koopman, W. J., van den Heuvel, L. P., Rodenburg, R. J., Mayatepek, E. et al. ( 2009) Mitochondrial complex I deficiency: from organelle dysfunction to clinical disease. Brain 132, 833842.
  • 18
    Koene, S. and Smeitink, J. A. M. ( 2011) Mitochondrial Medicine, A Clinical Guideline, 1st edn. P. 135, Khondrion B.V., Nijmegen, The Netherlands.
  • 19
    Leigh, D. ( 1951) Subacute necrotizing encephalomyelopathy in an infant. J. Neurol. Neurosurg. Psychiatry 14, 216221.
  • 20
    Janssen, A. J., Trijbels, F. J., Sengers, R. C., Wintjes, L. T., Ruitenbeek, W. et al. ( 2006) Measurement of the energy-generating capacity of human muscle mitochondria: diagnostic procedure and application to human pathology. Clin. Chem. 52, 860871.
  • 21
    Koopman, W. J., Visch, H. J., Verkaart, S., van den Heuvel, L. P., Smeitink, J. A. M. et al. ( 2005) Mitochondrial network complexity and pathological decrease in complex I activity are tightly correlated in isolated human complex I deficiency. Am. J. Physiol. Cell Physiol. 289, C881C890.
  • 22
    Verkaart, S., Koopman, W. J., Cheek, J., van Emst-de Vries, S. E., van den Heuvel, L. P. et al. ( 2007) Mitochondrial and cytosolic thiol redox state are not detectably altered in isolated human NADH:ubiquinone oxidoreductase deficiency. Biochim. Biophys. Acta 1772, 10411051.
  • 23
    Verkaart, S., Koopman, W. J., van Emst-de Vries, S. E., Nijtmans, L. G., van den Heuvel, L. P. et al. ( 2007) Superoxide production is inversely related to complex I activity in inherited complex I deficiency. Biochim. Biophys. Acta 1772, 373381.
  • 24
    Visch, H. J., Koopman, W. J., Leusink, A., van Emst-de Vries, S. E., van den Heuvel, L. P. et al. ( 2006) Decreased agonist-stimulated mitochondrial ATP production caused by a pathological reduction in endoplasmic reticulum calcium content in human complex I deficiency. Biochim. Biophys. Acta 1762, 115123.
  • 25
    Koopman, W. J., Willems, P. H. G. M., and Smeitink, J. A. M. ( 2012) Monogenic mitochondrial disorders. N. Engl. J. Med. 366, 11321141.
  • 26
    Scacco, S., Petruzzella, V., Budde, S., Vergari, R., Tamborra, R. et al. ( 2003) Pathological mutations of the human NDUFS4 gene of the 18-kDa (AQDQ) subunit of complex I affect the expression of the protein and the assembly and function of the complex. J. Biol. Chem. 278, 4416144167.
  • 27
    Ugalde, C., Janssen, R. J., van den Heuvel, L. P., Smeitink, J. A. M., and Nijtmans, L. G. ( 2004) Differences in assembly or stability of complex I and other mitochondrial OXPHOS complexes in inherited complex I deficiency. Hum. Mol. Genet. 13, 659667.
  • 28
    Blanchet, L., Buydens, M. C., Smeitink, J. A. M., Willems, P. H. G. M., and Koopman, W. J. ( 2011) Isolated mitochondrial complex I deficiency: explorative data analysis of patient cell parameters. Curr. Pharm. Des. 17, 40234033.
  • 29
    De Rasmo, D., Signorile, A., Larizza, M., Pacelli, C., Cocco, T. et al. ( 2012) Activation of the cAMP cascade in human fibroblast cultures rescues the activity of oxidatively damaged complex I. Free Radic. Biol. Med. 52, 757764.
  • 30
    Distelmaier, F., Valsecchi, F., Forkink, M., van Emst-de Vries, S., Swarts, H. G. et al. ( 2012) Trolox-sensitive reactive oxygen species regulate mitochondrial morphology, oxidative phosphorylation and cytosolic calcium handling in healthy cells. Antioxid. Redox Signal. 17, 16571669.
  • 31
    Koopman, W. J., Verkaart, S., Visch, H. J., van Emst-de Vries, S., Nijtmans, L. G. et al. ( 2007) Human NADH:ubiquinone oxidoreductase deficiency: radical changes in mitochondrial morphology? Am. J. Physiol. Cell Physiol. 293, C22C29.
  • 32
    Voets, A. M., Huigsloot, M., Lindsey, P. J., Leenders, A. M., Koopman, W. J. et al. ( 2012) Transcriptional changes in OXPHOS complex I deficiency are related to anti-oxidant pathways and could explain the disturbed calcium homeostasis. Biochim. Biophys. Acta 1822, 11611168.
  • 33
    Willems, P. H. G. M., Valsecchi, F., Distelmaier, F., Verkaart, S., Visch, H. J. et al. ( 2008) Mitochondrial Ca2+ homeostasis in human NADH:ubiquinone oxidoreductase deficiency. Cell Calcium 44, 123133.
  • 34
    Trevelyan, A. J., Kirby, D. M., Smulders-Srinivasan, T. K., Nooteboom, M., Acin-Perez, R. et al. ( 2010) Mitochondrial DNA mutations affect calcium handling in differentiated neurons. Brain 133, 787796.
  • 35
    Kruse, S. E., Watt, W. C., Marcinek, D. J., Kapur, R. P., Schenkman, K. A. et al. ( 2008) Mice with mitochondrial complex I deficiency develop a fatal encephalomyopathy. Cell Metab. 7, 312320.
  • 36
    Quintana, A., Zanella, S., Koch, H., Kruse, S. E., Lee, D. et al. ( 2012) Fatal breathing dysfunction in a mouse model of Leigh syndrome. J. Clin. Invest. 122, 23592368.
  • 37
    Calvaruso, M. A., Willems, P. H. G. M., van den Brand, M., Valsecchi, F., Kruse, S. et al. ( 2012) Mitochondrial complex III stabilizes complex I in the absence of NDUFS4 to provide partial activity. Hum. Mol. Genet. 21, 115120.
  • 38
    Valsecchi, F., Monge, C., Forkink, M., de Groof, A. J., Benard, G. et al. ( 2012) Metabolic consequences of NDUFS4 gene deletion in immortalized mouse embryonic fibroblasts. Biochim. Biophys. Acta 1817, 19251936.
  • 39
    Ingraham, C. A., Burwell, L. S., Skalska, J., Brookes, P. S., Howell, R. L. et al. ( 2009) NDUFS4: creation of a mouse model mimicking a Complex I disorder. Mitochondrion, 9, 204210.
  • 40
    Roestenberg, P., Manjeri, G. R., Valsecchi, F., Smeitink, J. A. M., Willems, P. H. G. M. et al. ( 2012) Pharmacological targeting of mitochondrial complex I deficiency: the cellular level and beyond. Mitochondrion, 12, 5765.
  • 41
    Quintana, A., Kruse, S. E., Kapur, R. P., Sanz, E., and Palmiter, R. D. ( 2010) Complex I deficiency due to loss of Ndufs4 in the brain results in progressive encephalopathy resembling Leigh syndrome. Proc. Natl. Acad. Sci. U.S.A. 107, 1099611001.
  • 42
    Breuer, M. E., Willems, P. H. G. M., Russel, F. G., Koopman, W. J., and Smeitink, J. A. M. ( 2012) Modeling mitochondrial dysfunctions in the brain: from mice to men. J. Inherit. Metab. Dis. 35, 193210.
  • 43
    Choi, W. S., Kruse, S. E., Palmiter, R. D., and Xia, Z. ( 2008) Mitochondrial complex I inhibition is not required for dopaminergic neuron death induced by rotenone, MPP+, or paraquat. Proc. Natl. Acad. Sci. U.S.A. 105, 1513615141.
  • 44
    Choi, W. S., Palmiter, R. D., and Xia, Z. ( 2011) Loss of mitochondrial complex I activity potentiates dopamine neuron death induced by microtubule dysfunction in a Parkinson's disease model. J. Cell. Biol. 192, 873882.
  • 45
    Valsecchi, F., Grefte, S., Roestenberg, P., Joosten-Wagenaars, J., Smeitink, J. A. M. et al. ( 2012) Primary fibroblasts of NDUFS4-/- mice display increased ROS levels and aberrant mitochondrial morphology. Mitochondrion, 9, PMID 23234723.
  • 46
    Rivera, J. and Tessarollo, L. ( 2008) Genetic background and the dilemma of translating mouse studies to humans. Immunity 28, 14.
  • 47
    Leshinsky-Silver, E., Lebre, A. S., Minai, L., Saada, A., Steffann, J. et al. ( 2009) NDUFS4 mutations cause Leigh syndrome with predominant brainstem involvement. Mol. Genet. Metab. 97, 185189.
  • 48
    Teitelbaum, Z., Lave, T., Freijer, J., and Cohen, A. F. ( 2010) Risk assessment in extrapolation of pharmacokinetics from preclinical data to humans. Clin. Pharmacokinet. 49, 619632.
  • 49
    Ghosh, A., Chandran, K., Kalivendi, S. V., Joseph, J., Antholine, W. E. et al. ( 2010) Neuroprotection by a mitochondria-targeted drug in a Parkinson's disease model. Free Radic. Biol. Med. 49, 16741684.
  • 50
    Sterky, F. H., Hoffman, A. F., Milenkovic, D., Bao, B., Paganelli, A. et al. ( 2011) Altered dopamine metabolism and increased vulnerability to MPTP in mice with partial deficiency of mitochondrial complex I in dopamine neurons. Hum. Mol. Genet. 21, 10781089.