U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Madame Curie Bioscience Database [Internet]. Austin (TX): Landes Bioscience; 2000-2013.

Cover of Madame Curie Bioscience Database

Madame Curie Bioscience Database [Internet].

Show details

Animal Models of OXPHOS Disorders

and .

Dysfunction of the mitochondrial respiratory chain has been associated with a wide range of human diseases ranging from diabetes to cardiomyopathy. Mutations in a number of nuclear as well as mitochondrial genes have been implicated in causing these diseases. Several animal models have now been created which reproduce some of the clinical pathology observed in human patients suffering from OXPHOS disorders. In this chapter we review some of these animal models of OXPHOS disorders and how they have led to a further understanding of both mitochondrial respiratory chain function and dysfunction.


The link between mitochondrial dysfunction, due to an impaired respiratory chain, and human disease has been well documented within the last decade. There are between 80-100 protein subunits comprising the respiratory chain and only 13 of these are encoded by mitochondrial DNA (mtDNA) with the remainder being nuclear DNA encoded. It is not surprising, therefore, that both nuclear and mitochondrial DNA mutations can directly affect the function of mitochondria. A large number of mtDNA point mutations and rearrangements have been identified as the primary cause of OXPHOS disease in affected patients. In addition, an increasing number of nuclear DNA mutations are being discovered in mitochondrial disorders.13 Despite this increasing knowledge in the underlying gene defects causing OXPHOS disorders, the molecular pathogenesis events that link the mutated gene to the observed clinical phenotypes are largely still undetermined.

In addition to the well-defined genetic disorders linked to mitochondrial dysfunction, there is circumstantial evidence that mitochondrial dysfunction perhaps plays a role in common disorders such as neurodegeneration (Parkinson and Alzheimer diseases), diabetes mellitus and heart failure. There have also been reports that an age-dependent accumulation of somatic mtDNA deletions may contribute to ageing.4 To be able to ameliorate or cure these disorders, it is necessary for us to understand in more detail the mechanisms leading to the phenotypes observed. An ATP deficiency in affected cells is often assumed to be the main cause of pathology, however, no conclusive evidence has yet been presented to confirm this. Along with an ATP deficiency, an impaired respiratory chain also leads to alterations of cellular reduction-oxidation (redox) status, induction of the mitochondrial pathway for apoptosis or increased production of reactive oxygen species (ROS), all of which can lead to the pathogenesis observed in patients.

Whilst there is limited availability of human tissues, the ability to produce model organisms with specific nuclear or mtDNA mutations gives us the necessary tissue to study in-depth the resulting biochemical defects. Basic biochemical defects due to nuclear and mtDNA mutations have been studied in lower model organisms such as budding yeast, worms and fruit flies (one model of which is discussed here), however, it is clear that many physiological differences between these organisms and humans exist. Hence, there are a number of advantages in creating mouse models of OXPHOS disorders since both mice and humans display similar gene content, along with comparable types of internal organs and physiology. In this chapter we discuss a Drosophila model and a number of mouse models in which nuclear genes have been modified to reproduce mitochondrial respiratory chain disorders and also describe some of the most recent advances in directly manipulating mouse mtDNA to recreate symptoms observed in humans carrying mtDNA point mutations or mtDNA rearrangements. Table 1 provides a summary of the animal models discussed here.

Table 1. Animal models of OXPHOS disorders.

Table 1

Animal models of OXPHOS disorders.

A Drosophila Model of Mitochondrial Deafness

In the fruit fly Drosophila melanogaster, a mutant phenotype of bang sensitivity (technical knockout, tko) i.e., temporary paralysis resulting from mechanical vibration, was isolated and has been extensively studied.57 This temporary paralysis is associated with a failure of signalling from mechanoreceptor neurons and a comparison was made between this phenotype and that of sensorineural deafness observed in humans resulting from a signalling failure in the mechanosensory receptor cells of the inner ear. It was found that by transgenically complementing a 3.2-kb fragment of nuclear DNA it was possible to rescue the tko phenotype.8 This segment of genomic DNA encodes a single transcript for a homologue of bacterial ribosomal protein S12. The N-terminal sequence of this protein contained features commonly observed in mitochondrial targeting presequences indicating a probable mitochondrial localization for this protein.9 Sequencing of the gene revealed a single amino acid change in the tko mutants which affects a conserved leucine residue.

A more detailed study of these mutant tko flies was undertaken to more clearly establish the phenotype.10 A severe hearing deficiency was discovered along with developmental delays and behavioural and sensory abnormalities. Decreased activity of the mitochondrial complexes I, III, and IV were observed in mutant larvae, along with a reduction in the level of small subunit (12S) to large subunit (16S) mitochondrial rRNA, when compared with controls. Conclusive evidence that it is the point mutation L85H that causes the mutant phenotype was obtained by transgenically complementing homozygous mutants with the wild type gene for tko and consequently rescuing the mutant phenotype. Homozygous mutants complemented in the same way with the tko gene carrying the L85H mutation maintained the mutant phenotype. These results indicate a role for the tko protein in mitochondrial protein translation and a potentially useful model for human mitochondrial disease. Many mitochondrial mutations resulting in a dysfunction of mitochondrial protein translation are known to have a phenotype of sensorineural deafness, however, the reasons why a mitochondrial translation defect causes such a tissue-specific disorder is not understood. The availability of this model organism is of huge benefit to study the evidently essential role of mitochondria in the mechanosensory cells, and will aid the understanding of diseases leading to human sensorineural deafness.

Mouse Models of Nuclear DNA Mutations

Mammalian mitochondrial DNA encodes 13 polypeptides, all of which are involved in oxidative phosphorylation, along with 22 tRNAs and 2 rRNA subunits. Hence, the majority of proteins that make up the respiratory chain and all the proteins responsible for mtDNA maintenance and transcription are nuclear encoded, translated in the cytoplasm and imported into mitochondria. In more recent years, mutations in these nuclear encoded mitochondrial proteins have been found to be connected with mitochondrial disorders e.g., Leigh syndrome caused by mutations in the SURF1 protein.1112 The ability to reproduce phenotypic symptoms of these diseases by mutating the associated nuclear genes would be a major breakthrough for not only understanding the mechanisms of these disorders but also for developing effective treatments. To date no such mouse models have been successfully created. However, a number of nuclear genes encoding mitochondrial proteins have been knocked out to create either germ line or tissue specific knockout mice. These genes include Ant1 connected with mitochondrial bioenergetics, Tfam a mtDNA maintenance factor, the antioxidant genes GPx1 and Sod2, and a gene associated with the Frataxin syndrome, Frda.

Tissue Specific Adenine Nucleotide Translocator Knockout

The main function of the adenine nucleotide translocator (ANT) is to exchange mitochondrial matrix ATP for cytosolic ADP across the inner mitochondrial membrane, utilising the electrochemical gradient. In humans there are three isoforms of this protein, ANT1, ANT2 and ANT3, whereas there are only two isoforms in the mouse (Ant1 and Ant2). The gene Ant1 is predominantly expressed in heart and skeletal muscle and has been designated as a heart/muscle specific isoform. By knocking out Ant1 in mouse, an ATP deficiency would result in the heart and skeletal muscle tissues, effectively creating a tissue specific knockout. It was hoped pathological features arising in this mouse model would be similar to those observed in patients suffering from mtDNA mutation syndromes. Perhaps there would be a correlation between ATP deficiency and clinical manifestation giving further insight into the mechanism behind the diseases. More recently, two heterozygous missense mutations in Ant1 were identified in patients suffering from adPEO (autosomal dominant progressive external ophthalmoplegia).13 Large-scale mtDNA deletions are characteristic of this disorder, suggesting ANT1 may play a role in mtDNA maintenance.

Mice lacking Ant1 were created by removing exons 1-3 producing a mutant allele that is null for Ant1 activity.14 Homozygous mutants were viable, fertile and showed comparable growth to wild-type animals, however, there was a complete absence of the Ant1 protein and no Ant1 gene expression. Skeletal muscle from mutant mice showed ragged-red fibres (RRF) as seen in mitochondrial disorder patients, along with increased succinate dehydrogenase (SDH) and cytochrome c oxidase (COX) activities. Hearts from mutant animals 4-6 months old showed cardiac hypertrophy and were enlarged when compared with wild-type hearts, however, no difference in heart size between mutants and wild-types was observed in the younger 6-8 week old mice. Again there was a marked mitochondrial proliferation in older mutant hearts. Mitochondrial respiration rates in skeletal muscle were compared in mutant and wild-type age-matched individuals and found to be significantly reduced in the Ant1 knockouts, with a lack of ADP stimulated respiration; this was expected due to the mutants inability to transport ADP into or ATP out of it's skeletal muscle mitochondria. Consistent with the drop in respiration levels, homozygous mutant animals had an exercise intolerance, only being capable of completing 54% of an exercise program that all wild-type mice completed. A metabolic profile from blood was compiled and the Ant1 mutant mice showed a profile associated with OXPHOS deficiency. In conclusion, by knocking out the Ant1 gene in mice it was possible to create an animal model presenting features commonly associated with mitochondrial myopathy and cardiomyopathy. In doing so it was demonstrated that mitochondrial ATP deficiency can lead to at least some of the clinical phenotypes manifested in humans with mitochondrial disorders.

As one of the morphological phenotypes of the Ant1 mutant mice was mitochondrial proliferation in both skeletal muscle and heart it was hoped that by looking at gene expression profiles of these tissues it would be possible to identify factors involved in mitochondrial biogenesis as well as genes up-regulated due to OXPHOS dysfunction. By using differential display reverse transcription-polymerase chain reaction techniques 17 genes were found to be up regulated in the muscle of homozygous mutant mice. This included both mtDNA and nuclear DNA encoded respiratory chain components, some mitochondrial tRNA and rRNA genes and intermediary metabolism genes. Clearly, this is an attempt by the cells to increase energy metabolism. Additionally, genes connected with apoptosis (Mcl-1) and some genes potentially involved with mitochondrial biogenesis (SKD3) were increased in expression.

Another proposed mechanism for the onset of symptoms due to mitochondrial disease is the overproduction and toxicity of mitochondrial reactive oxygen species (ROS). Mitochondrial ROS are by-products of energy production by the respiratory chain and include the superoxide anion (O2-), hydrogen peroxide (H2O2), and hydroxyl radical (OH). Manganese superoxide dismutase (Sod2) and glutathione peroxidase-isoform 1 (Gpx1) are antioxidant enzymes located in the mitochondria to convert ROS to harmless products. Due to the close proximity of mtDNA to the site of production of ROS it is essential that they are dissipated fast and efficiently to eliminate DNA damage. However, over the course of a lifetime a gradual build-up of ROS damage to mtDNA has been observed.1516 Skeletal muscle and heart from Ant1 mutant mice were analysed for their levels of production of the ROS H2O2.17 Both tissues were found to have maximal production when compared to age-matched wild-type individuals. Along with increased H2O2 was a concomitant increase in mRNA levels of Sod2 as well as increased steady state protein levels of both Sod2 and Gpx1. A greater number of mtDNA rearrangements were to be found in heart and to a lesser extent in the skeletal muscle of Ant1 mutant animals as would be expected from the increased ROS production. Why do Ant1 mutant mice produce more mitochondrial ROS? By inhibiting ADP/ATP transfer, proton transport through the ATP synthase complex is blocked and consequently inhibits the electron transport chain (ETC). Electrons would build up in the ETC complexes and be transferred to oxygen, generating O2- and hence H2O2 levels will rise.

What conclusions can be drawn from the creation of the Ant1 knockout mouse model? By inhibiting ADP/ATP transfer it is possible to recreate some of the symptoms observed in humans with mitochondrial disorders with a striking similarity to patients suffering from adPEO. Consistent with the phenotype of these animals is the dysfunctional OXPHOS system and an increased production of ROS with a concomitant rise in antioxidant levels. The increased mitochondrial ROS may well have precipitated the accumulation of rearranged mtDNA molecules leading to a negative feed back system; increasing proportions of damaged mtDNA due to ROS, giving rise to a more dysfunctional ETC and increasing ROS production. The mitochondrial manganese superoxide dismutase (MnSOD) knockout mouse model, discussed later, shows the role that mitochondrial ROS play in the pathophysiology of mitochondrial diseases.

Mice Deficient in Cellular Glutathione Peroxidase

Reactive oxygen species (ROS) are produced during cellular processes involving oxygen and include highly reactive molecules such as singlet oxygen (O2-), hydrogen peroxide (H2O2) and hydroxyl radical (OH). Aerobic organisms have a defence mechanism to prevent oxidative damage to tissues and cells and includes an array of antioxidant enzymes, such as superoxide dismutases, glutathione peroxidases and catalase. In mammals there are five known glutathione peroxidase (Gpxs) isoenzymes that vary in expression levels between tissues. The function of Gpxs is to catalyse the reduction of H2O2 to water as well as a number of hydroperoxides e.g., DNA and lipid peroxides, to alcohols (Fig. 1). However, it is uncertain whether the primary role of these enzymes is to protect against oxidative damage under normal physiological conditions or if it has a more protective role during oxidative stress.

Figure 1. Reactions catalysed by glutathione peroxidase (a) reduction of hydrogen peroxide to form water; (b) reduction of various hydroperoxides to form corresponding alcohols.

Figure 1

Reactions catalysed by glutathione peroxidase (a) reduction of hydrogen peroxide to form water; (b) reduction of various hydroperoxides to form corresponding alcohols. GSH = reduced glutathione, GSSG = glutathione disulfide, ROOH = organic hydroperoxide. (more...)

In order to map the expression pattern of glutathione peroxidase isoform 1 (Gpx1) a ‘knock-in’ mouse model was created by homologous recombination to insert the β-galactosidase gene into exon 1 of Gpx1.18 In situ detection of β-galactosidase activity was performed by tissue staining with X-gal to reveal expression of Gpx1. In heterozygous mutant adult mice, high levels of Gpx1 expression were detected in liver and kidney cortex, with a small amount detected in heart, skeletal muscle, lung and spleen.

The gene encoding Gpx1 was first knocked out in mice in 1997 by Ho et al19 by disruption of exon 2 with an inserted neo cassette. Although there were no detectable mRNA levels of the gene in brain, heart, kidney, liver or lung and virtually no Gpx1 activity, there appeared to be no physiological differences between homozygous knockouts (Gpx1-/-) and wild-type mice. Gpx1-/- mice grew normally and were apparently healthy up to 20 months of age. In addition, no protein or lipid oxidation was found and there was no increased sensitivity to hypoxia. Whilst glutathione peroxidase activity has been diminished in certain tissues of the Gpx1-/- mouse, indicating the large role this isoform plays, there is still sufficient activity to prevent accumulation of H2O2 under physiological conditions. Replacing exons 1 and 2 with a PGKneo cassette created a second Gpx1 knockout mouse where again homozygous mutants Gpx1-/- were viable and fertile, although by eight months they exhibited growth retardation.18 Mitochondria were isolated from liver, kidney, heart, skeletal muscle, and brain of Gpx1-/- animals and found to contain no Gpx1 activity. In addition the ability of Gpx1-/- liver mitochondria to reduce H2O2 was down to 1.5% compared to wild-type liver mitochondria, suggesting Gpx1 is the only liver mitochondrial Gpx. Interestingly, there was no detectable Gpx activity at all in control heart mitochondria, implicating that Gpx1 plays a role in protecting liver mitochondria from oxidative stress but does not function in heart mitochondria. Quantification of ATP production and respiration rates by Gpx1-/- liver mitochondria showed a reduction by about one-third compared to wild-type controls. Meanwhile heart mitochondria from both knockout and wild-type animals showed comparable respiration rates. Livers from Gpx1-/- mice showed an increase in lipid oxidation and isolated liver mitochondria appeared to produce a greater amount of H2O2. Results from this animal model clearly show the importance of Gpx1 activity in mouse liver mitochondria, which appears to be a location that lacks any of the other glutathione peroxidase isoforms. The role of Gpx1 in heart mitochondria does not seem to be as important, at least under physiological conditions. This may be due to the presence of catalase, another H2O2 oxidoreductase, in heart mitochondria.

To determine the role Gpx1 plays during oxidative stress, de Haan et al20 treated Gpx1-/- mice to the oxidant paraquat. Paraquat is known to generate free radicals within cells. Within five hours of injecting paraquat intraperitoneally, the Gpx1-/- mice died, contrasting greatly with the wild-type controls that showed no lethality even up to 10 days later. Activity of Gpx1 in the lungs of wild-type mice was found to be elevated 2-fold following paraquat treatment that was subsequently found to be due to increased transcription of the Gpx1 gene. Gpx1-/- mice showed no Gpx1 activity in the lungs either before or after paraquat injection. Cultured neuronal cells derived from Gpx1-/- mice also showed a higher sensitivity to low doses of H2O2 added to the culture medium when compared to wild-type derived neuronal cultures that were completely viable under such conditions. These results suggest that Gpx1 plays a role in protection against oxidative stress induced by certain factors, however, its' role under physiological conditions may be limited. The role Gpx1 takes in protection of cells and tissues under conditions of oxidative stress induced by disease has yet to be elucidated.

Two Mitochondrial Superoxide Dismutase Knockout Mouse Models

Manganese superoxide dismutase (MnSOD) is one of several oxygen radical scavengers and has been located within mitochondria, a major cellular source of free radicals. MnSOD converts the mitochondrial superoxide radical (O2-) to hydrogen peroxide (H2O2), which is subsequently broken down to water by glutathione peroxidase-isoform 1 (Gpx1) (Fig. 2). To further establish the role mitochondrial ROS play in the pathophysiology of diseases especially age-related disorders, Li et al21 created a knockout mouse of the Sod2 gene, which encodes MnSOD. Sod2 was inactivated by deletion of exon 3, resulting in a loss of enzyme activity. Heterozygous mutant animals showed no abnormal phenotype whilst the deletion proved to be lethal in the homozygous knockouts with death occurring 4-10 days postnatally. These results are in sharp contrast to similar mouse models where inactivation of the cytosolic Sod1 or extracellular Sod3 genes had little effect on viability or fertility of the animals.2223 This suggests that mitochondria are not only a major source of O2- but also the toxicity of O2- is far greater in mitochondria than in the cytosol or extracellular matrix.

Figure 2. MnSOD (manganese superoxide dismutase) catalyses the reduction of the mitochondrial superoxide radical (O2-) to hydrogen peroxide (H2O2).

Figure 2

MnSOD (manganese superoxide dismutase) catalyses the reduction of the mitochondrial superoxide radical (O2-) to hydrogen peroxide (H2O2).

A detailed characterisation of the Sod2 mutant mice showed dilated cardiomyopathy, significant lipid accumulation in liver and skeletal muscle, a marked deficiency in succinate dehydrogenase (SDH or complex II) in the heart and skeletal muscle, and a partial deficiency of complex I and citrate synthase in heart mitochondria.24 In addition, activity of the mitochondrial enzyme aconitase (a component of the citric acid cycle) was shown to be greatly reduced in isolated heart and brain mitochondria. This iron-sulphur containing enzyme has previously been shown to be particularly sensitive to O2- damage.25 Evidently the increased levels of mitochondrial O2- has resulted in disruption of the iron-sulphur containing enzymes leading to a dysfunctional electron transport chain (ETC) and inhibition of the tricarboxylic acid cycle. To confirm this conclusion, a study of the rate of respiration of mitochondria isolated from the liver of homozygous mutant mice showed a 40% reduction of state III respiration, indicative of an ETC dysfunction. This defective respiratory chain system may lead to an increased production of mitochondrial ROS further weakening a system unable to process its harmful free radicals. As would be expected from an accumulation of mitochondrial ROS, there was a concomitant rise in oxidative damage to total DNA isolated from heart, liver and brain of Sod2 mutant mice.

The above Sod2 knockout mouse was created in the CD1 mouse background and at the same time a similar Sod2 knockout mouse was produced in the C57BL/6 mouse background, giving rise to a different set of phenotypes.26 Inactivation of the Sod2 gene was carried out by excision of exons 1 and 2, resulting in the removal of the transcription and translational start sites. Mutant and wild type offspring from heterozygous Sod2 mutant crosses were indistinguishable at birth with a typical Mendelian ratio of 1:2:1 being achieved. However, growth retardation of the homozygous mutants was obvious from postnatal day 2 up until their death on day 18. Electron microscopy carried out on the brain and spinal cord of 10 day old mutant animals showed signs of degenerative injury including extensive mitochondrial damage, especially to the neurons of the basal ganglia and brainstem. This result contrasted with electron microscopy data obtained from brain, heart, liver and skeletal muscle of 4-5 day old Sod2 mutant mice in the CD1 background where no mitochondrial damage was observed. Phenotypic similarities between the two Sod2 knockout mouse models include a dilated cardiomyopathy, lipid accumulation in the liver and progressive motor abnormalities, including early onset of fatigue. Clearly MnSOD is an important mitochondrial antioxidant enzyme which plays an essential role in the removal of mitochondrial ROS in tissues with a high energy demand. These are the first animal models where increased mitochondrial ROS can be associated with onset of mitochondrial disease. Increased mitochondrial O2- results in disruption of certain Fe-S containing enzymes leading to inactivation of the ETC and TCA with a subsequent loss in mitochondrial fatty acid oxidation and accumulation of fat in the liver. Energy starvation in the heart due to the dysfunctional ETC then leads to cardiac myopathy and heart failure.15

Due to mounting evidence that accumulation of mitochondrial ROS leads to mitochondrial defects and tissue pathologies, it is of importance to test antioxidant treatments, which will hopefully scavenge the harmful mitochondrial free radicals. Using the Sod2 mutant mice in the CD1 background, Melov et al2728 tested a number of such superoxide dismutase mimetics. Treatment of Sod2 mutant animals with MnTBAP (manganese 5,10,15,20-tetrakis (4-benzoic acid) porphyrin), a SOD mimetic, was capable of increasing the life-span of these animals from 10 days to 3 weeks along with ameliorating their cardiac myopathy and diminishing lipid accumulation in the liver. However, MnTBAP treatment revealed a neuropathology, previously undetected in these mice, which was characterised by a spongiform degeneration of the cortex and brain stem nuclei. Whilst MnTBAP is unable to cross the blood brain barrier, it is clearly able to slow the rapid accumulation of mitochondrial ROS in peripheral tissues. In addition, these results suggest that accumulation of mitochondrial ROS is extremely toxic to the brain.27 Three more catalytic antioxidant drugs were successfully used to rescue the pathologies seen in Sod2 mutant mice.28 All three salen manganese complexes (EUK-8, EUK-134 and EUK-189) contain SOD2 and catalase activities and were administered daily from postnatal day 3, at various doses to both mutant and wild type animals. Treatment of Sod2 mutants greatly increased their life-span, not only to 3 weeks as was the case with MnTBAP but up to 44 days with the cause of death at this age still undetermined. None of the neurobehavioral phenotypes revealed in the MnTBAP treated mice were observed, although mice over the age of 25 days developed a progressive movement disorder. This class of SOD-catalase mimetics can evidently cross the blood brain barrier and rescue the spongiform encephalopathy along with protecting peripheral tissues from mitochondrial ROS overproduction. Perhaps these drugs will provide a therapeutic approach to combating neurodegenerative diseases that have been linked to oxidative stress such as Alzheimer's and Parkinson's diseases as well as spongiform encephalopathies.

Characterisation of the Sod2 Heterozygous Knockouts

The homozygous Sod2 knockout mouse proved to be a useful animal model for acute mitochondrial O2- damage, however, the heterozygous Sod2+/- mice proved to be just as useful in the study of chronic O2- toxicity. A lifetime of oxidative phosphorylation exposes mammalian cells to an ever increasing amount of reactive oxygen species (ROS) which eventually leads to oxidation of lipids, proteins and DNA. This is the basis for the free radical theory of ageing which claims that age-related physiological decline is a result of oxidative damage to tissues and cells by ROS. Since the majority of ROS are produced in mitochondria it are the mitochondria that are the most affected by ROS toxicity, leading to an age-related accumulation of mitochondrial damage. To further investigate the link between accumulated mitochondrial ROS damage and ageing, Kokoszka et al29 characterised in more depth the heterozygous Sod2 knockout mice, which are subjected to a deficiency of the antioxidant enzyme MnSOD and potentially are exposed to greater levels of O2- toxicity. Protein levels of MnSOD were determined to be reduced ˜50% in Sod2+/- compared to wild-type mice. Mitochondrial membrane potential (Δψ) was found to be reduced in Sod2+/- mice but interestingly there was also an age-related decline of Δψ in both wild-type and Sod2+/- animals suggesting chronic oxidative damage could cause an increase in mitochondrial inner membrane proton permeability. Protons are pumped from the mitochondrial matrix across the mitochondrial inner membrane; creating an electrochemical gradient or Δψ which can be dissipated when protons are allowed to leak back across the inner mitochondrial membrane. Since the Δψ is used by ATP synthase (complex V of the ETC) to make ATP, a direct link can be made between oxidative phosphorylation and Δψ. Hence, a reduction in Δψ will result in reduced oxygen consumption, slowing down of mitochondrial oxidative phosphorylation with a concomitant increase in ROS production from electrons stranded in the inhibited ETC. A suggested reason for the increase in proton leakage was that oxidative damage to lipids in the mitochondrial inner membrane was rendering it leaky. Indeed it was shown that middle aged Sod2+/- animals had twice the level of lipid hydroxides compared to controls.

Chronic oxidative stress is known to activate the mitochondrial permeability transition pore (mtPTP) which when open allows the passage of molecules less than 1,500 Da between the mitochondrial matrix and cytosol.3031 Opening of the mtPTP channel results in a drop of Δψ, a loss of mitochondrial matrix solutes and the release of some apoptosis inducing factors such as cytochrome c. Middle to old age Sod2+/- mice were found to have a highly sensitised mtPTP compared with age matched controls and this was linked to the three fold increase in apoptotic hepatocytes observed in aged Sod2+/- livers. The decline in mitochondrial function observed in the Sod2+/- mice was accompanied by a rise in respiratory chain enzyme activity, a common phenomenon seen in mitochondrial disorders in a futile attempt to compensate for the energy deficiency. Throughout this study it was noted that the mitochondrial oxidative damage seen in tissues of Sod2+/- individuals was also accumulated in aged controls, although at a much later stage. In brief, this study has shown that increased mitochondrial ROS can result in a decrease in Δψ, leakage of protons across the mitochondrial inner membrane, oxidation of mitochondrial lipids, sensitisation of the mtPTP and subsequent initiation of apoptosis. In addition, it further supports the hypothesis that increased mitochondrial ROS leads to a decline in mitochondrial function and the initiation of apoptosis, a common feature in the ageing process.

Manipulation of Mitochondrial Transcription Factor A Expression in Mice

Regulation of mitochondrial gene expression in a mouse model was achieved by disrupting the nuclear encoded mitochondrial transcription factor A (Tfam).32 As illustrated in Figure 3, two loxP-sites were introduced either side of exons 6 and 7 of the Tfam gene by homologous recombination in embryonic stem cells creating TfamloxP mice. TfamloxP mice were then crossed with mice ubiquitously expressing cre-recombinase under the β-actin promoter, which excised exons 6 and 7 of the Tfam gene by cre-mediated recombination. Heterozygous knockout animals were viable even though levels of Tfam transcripts and protein were reduced by approximately 50%. A reduction of mtDNA copy number of up to 34% was observed in all tissues analysed, however, there were no significant reductions in mtRNA transcripts or respiratory chain complex activities with the exception of the heart. These results were interesting considering the reduced mtDNA copy numbers and suggest a compensatory increased stability of mitochondrial mRNA transcripts and polypeptides. Homozygous knockouts of Tfam were embryonic lethal with death occurring between embryonic day (E) 8.5 and 9.5 at which stage no mtDNA was detectable. These results confirmed that Tfam is essential for mtDNA maintenance in vivo and is an essential protein during embryogenesis in the mouse.

Figure 3. Schematic drawing of the construction of conditional Tfam knockout mice.

Figure 3

Schematic drawing of the construction of conditional Tfam knockout mice. Initially a targeting vector containing loxP-flanked Tfam was transfected into ES cells, which were then injected into blastocysts. Chimeric founder animals were mated to obtain (more...)

Germ-line disruption of Tfam confirmed the efficiency of the cre-loxP recombination system in generating knockout models and this system was then used to create a series of tissue-specific Tfam knockout mice. TfamloxP mice were mated with transgenic mice carrying cell type specific expression of cre recombinase.

Heart Specific Disruption of Tfam

Two mouse models of a heart specific knockout of the Tfam protein have been generated using either the muscle creatine kinase (Mck) promoter or the α-myosin heavy chain (Myhca) promoter to control cre recombinase expression.3334 Tfam expression was disrupted in both heart and skeletal muscle of homozygous TfamloxP mice expressing cre recombinase by the Mck promoter from E13. These mutant animals (TfamloxP/TfamloxP, +/Mck-cre) had a normal respiratory chain function in the heart at birth, however, they had a mean survival of just 20 days and at death exhibited a dilated cardiomyopathy. The Mck promoter is active only in cardiomyocytes of the heart where a reduction of Tfam protein levels was observed, as well as in skeletal muscle. There was a concomitant drop in steady-state levels of mitochondrial transcripts and mtDNA levels in both heart and skeletal muscle tissues but not in any other tissues analysed. Several cardiomyocytes lacked COX activity confirming the breakdown in respiratory chain function. Morphological analysis of the hearts by electron microscopy revealed abnormally enlarged mitochondria containing tubular cristae. This animal model has reproduced many pathophysiological features of human diseases resulting from either deletions or point mutations in mitochondrial tRNA genes: (i) impaired mtDNA expression leading to a dysfunctional respiratory chain in a tissue specific manner; (ii) a mosaic pattern of respiratory chain-deficient cells in affected tissues; (iii) appearance of morphologically abnormal mitochondria; and (iv) a time dependent deterioration of respiratory chain function.

A second heart specific knockout of Tfam was created with cre recombinase expression being controlled by the Myhca promoter that results in cre expression from E8. The resulting knockouts (TfamloxP/TfamloxP, +/Myhca) had onset of cardiomyopathy during embryogenesis with the majority (75%) of knockouts dying in the neonatal period. Interestingly, the 25% of mutant animals that survived lived for several months before dying from dilated cardiomyopathy. This suggested modifying gene(s) played a role in determining the lifespan of the knockouts. Both the Mck and Myhca animal models demonstrate that by regulating cre expression in the heart it is possible to control the age of onset of cardiac respiratory chain dysfunction and will therefore be a very useful tool for the study of other tissue specific knockouts of Tfam.

A Mouse Model for Mitochondrial Diabetes

A tissue-specific disruption of Tfam in mouse pancreatic β-cells was achieved by crossing mice with a loxP-flanked Tfam with transgenic mice expressing cre recombinase by the rat insulin-2 promoter.35 Seven week old mutant mice showed a severe mtDNA depletion in islets along with a COX deficiency. Pancreatic sections of mutant mice were analysed by electron microscopy and revealed abnormally appearing mitochondria with tubular cristae in the insulin-producing cells, a morphology commonly found in mitochondria with a dysfunctional respiratory chain. Along with the drop in mtDNA expression and concomitant respiratory chain deficiency in pancreatic β-cells, mutant mice from the age of just 5 weeks developed diabetes with decreased blood insulin concentrations.

At 7 weeks of age mutant mice showed a normal distribution of β-cells within the pancreas, yet were diabetic, suggesting an impaired β-cell stimulus-secretion coupling. Metabolism of glucose in the β-cell is directly coupled to insulin secretion. During mitochondrial oxidation of glucose metabolites, there is an increase in the concentration of ATP that results in closure of the ATP-dependent potassium channels (KATP channel). Depolarisation of the plasma membrane occurs when the KATP channels are closed and a subsequent opening of the voltage-sensitive L-type Ca channels occurs allowing Ca to enter the cytosol. The incoming Ca activates fusion of the insulin-containing vesicles, causing release of insulin. Hence, it is clear that mitochondrial production of ATP plays a key role in the release of insulin by regulating the KATP channel and that to block the ETC will subsequently lead to inhibiting plasma membrane depolarisation and insulin secretion. Islets isolated from both mutant and control mice were subjected to glucose stimulation and the levels of intracellular Ca were measured. Mutant animals showed a decreased rise in Ca levels compared with controls. These results confirm that mitochondrial dysfunction in the β-cells of the mutant mice are the principle cause of the diabetic phenotype observed. Not only do these mice reproduce pathophysiological features of mitochondrial diabetes but also give evidence that the respiratory chain plays a central role in normal glucose-induced insulin secretion.

Disrupting Oxidative Phosphorylation to Produce Mitochondrial Late-Onset Neurodegeneration (MILON) Mice

By crossing TfamloxP/TfamloxP mice with mice heterozygous for a transgene expressing cre recombinase from the calcium-dependent calmodulin kinase II promoter it was possible to create MILON mice.36 The Tfam gene was knocked-out in neocortex with maximum Tfam recombination having occurred by 1 month of age. At 2 months mutant mice had much reduced levels of Tfam protein, however, there was no overt phenotype at this age. From 5-6 months of age, mutant (MILON) mice showed abnormal behaviour with a physical deterioration that degenerated rapidly towards death approximately one week after the first symptoms appeared. By 2 months of age there was a ˜40% drop in mtDNA levels and by 4 months the levels of mtRNA transcripts were also reduced by ˜40%. As would be expected with the reduction in mtDNA transcripts, the activities of complex I and IV in neocortical samples were shown to be reduced in 4- and 5-months old MILON mice. Cell death, as detected by TUNEL-positive nuclei, was only observed in the brains of symptomatic animals, no TUNEL-positive cells were seen in 2-4 months old presymptomatic MILON mice. Massive neurodegeneration was displayed in end-stage symptomatic MILON mice. To determine if MILON mice were more susceptible to stress-induced neuronal death both controls and MILON mice were injected with kainic acid which induces seizures. Both sets of mice exhibited high level seizures but only the MILON mice developed an increased number of TUNEL-positive cells, as assessed by performing TUNEL staining on brains harvested 24 hours after kainic acid injection, indicating an increased sensitivity to neuronal stress in these animals. Human neurodegenerative disorders generally develop later in life, as is the case in this animal model where MILON mice are healthy well into adulthood. Interestingly, before onset of neurodegenerative symptoms MILON mice are more vulnerable to excitotoxic challenges indicating that perhaps patients with mitochondrial encephalomyopathies may exhibit a clinical deterioration in response to moderate stress, e.g., viral infections. In addition, it was observed that neurons could survive for at least one month after oxidative phosphorylation was inhibited. Such a time lag has been observed in human patients with mitochondrial neurodegeneration and possibly reflects the ability of neurons to withstand long periods of OXPHOS deficiency. As there are many similarities between this mouse model of mitochondrial dysfunction in neurons and human sufferers of mitochondrial neurodegeneration, these mice will prove useful in future studies of the molecular pathways leading to neuronal cell death and in the testing of pharmacological treatments for these disorders.

Embryonic Lethality Due to Loss of Frataxin

Frataxin is a 210 amino acid nuclear-encoded mitochondrial protein that has been highly conserved through evolution and is thought to play a role in mitochondrial iron homeostasis. Analysis of frataxin in mouse has shown expression of the gene from E10.5. Tissue specific expression has been observed in humans and mice with frataxin being expressed in heart, dorsal root ganglia, pancreas, liver and skeletal muscle.3738 A large GAA triplet repeat expansion in the first intron of the gene encoding frataxin has been associated with the most common hereditary ataxia, Friedreich ataxia (FRDA). FRDA is a neurodegenerative disorder characterised by progressive ataxia of the limbs, dysarthria and other coordination disorders. These symptoms are believed to be a result of neuronal degeneration, however, other manifestations such as cardiomyopathy and diabetes are also common. Tissue biopsies from patients reveal activities of mitochondrial iron-sulphur (Fe-S) containing enzymes to be greatly impaired as well as both mitochondrial and cytosolic aconitase.39 It was believed that a loss of functional frataxin protein led to an accumulation of mitochondrial iron which is a potent oxidising agent, hence resulting in oxidation of the iron-sulphur containing enzymes with a subsequent loss of respiratory chain function.

Cossée et al40 created a frataxin knockout mouse by deleting exon 4, a highly conserved sequence in the frataxin gene, by homologous recombination. Deletion of exon 4 results in a frameshift and a truncation of the frataxin protein. Heterozygous knockouts were crossed but no homozygous mutant offspring were obtained and subsequent analysis of the embryos from such a cross revealed gross morphological abnormalities in approximately one third of all embryos. Abnormal embryos between E7.5 and E8.5 were genotyped and discovered to be either homozygous or heterozygous knockouts for frataxin. TUNEL staining (to detect apoptotic cells) was carried out on E6.75 embryo sections and abnormal embryos showed substantial positive staining indicating extensive apoptosis compared with virtually no staining in normal embryos. Reabsorption of abnormal embryos occurred by approximately E9.5. Contrary to the belief that loss of functional frataxin results in mitochondrial iron accumulation, no iron deposits were observed in any of the abnormal embryos analysed. This knockout mouse model shows that frataxin plays an important role in development with complete loss of the protein being lethal in mice. No FRDA patients have been described with a truncated point mutation in both alleles of the frataxin gene possibly due to this early lethality. The length of the intronic GAA triplet repeat expansion observed in FRDA patients can be directly correlated with severity and age of disease onset. Smaller repeats produce milder symptoms and a later onset of disease as well as enabling the synthesis of residual levels of frataxin when compared with large expansions.

To try and mimic the symptoms observed in FRDA patients two conditional knockout mice for the frataxin gene were created, one muscle and the other neuron specific.41 These conditional knockouts were produced by flanking exon 4 in the gene for frataxin (Frda) with loxP sites. Mice expressing the cre recombinase gene in specific tissues can then, by homologous recombination, excise out exon 4 of Frda creating a knockout. Homozygous mice for the conditional knockout allele of Frda were mated with mice that carried the Cre transgene either under the control of muscle creatine kinase (MCK) or the neuron-specific enolase (NSE) promoters. Upon expression of cre in these specific tissues there would subsequently be a tissue specific loss of frataxin. The skeletal muscle and heart specific (MCK) mutants begin to lose weight at 7 weeks and die at about 11 weeks of age. There is an absence of full-length transcript and protein in both skeletal muscle and heart of these knockouts and cardiomyopathy is observed in the latter stages, a similarity with the human FRDA disease. By contrast, the neuron-specific enolase (NSE) knockout mice have a low birth weight and develop ataxia and a progressive neurological phenotype with death occurring at about 24 days of age. Expression patterns show a lack of full-length frataxin transcript in not only neuronal tissues but also heart and liver. There is an earlier manifestation of cardiomyopathy in the NSE knockouts in addition to degeneration and necrosis of the nervous system, another common feature seen in FRDA patients. The hearts of both animal models revealed large-scale mitochondrial abnormalities suggestive of a mitochondrial dysfunction which was confirmed with biochemical analysis of heart tissue showing deficiencies in the Fe-S containing enzymes of the respiratory chain complexes I-III and aconitase. There were limited amounts of mitochondrial iron accumulation in MCK animals but no iron deposits were observed at all in the NSE mice suggesting that damage to Fe-S containing enzymes is not due to the accumulation of iron as previously thought. Frataxin is located in the mitochondrial inner membrane and its main role was believed to be transport of iron out of mitochondria. Since loss of function of frataxin does not primarily lead to mitochondrial iron accumulation but instead to a deficiency of Fe-S centre enzymes it is more likely that frataxin plays a larger role in synthesis of Fe-S clusters or protection from oxidative stress. It is hoped that these mouse models of FRDA will reveal the role that frataxin plays in iron metabolism and the biogenesis of Fe-S containing enzymes.

Transmitochondrial Mouse Models

With the exception of both sperm and mature oocytes, between 103-104 copies of mtDNA are present in each cell. Mitochondrial DNA sequences are known to differ between unrelated individuals by approximately 0.3% (˜50 nucleotides), however, it is normal for a given individual to contain just one variant of the mtDNA sequence. Heteroplasmy occurs when a cell contains more than one sequence variant of the mitochondrial genome and many known pathogenic mtDNA mutations are heteroplasmic in presentation. Therefore it would be of great value to have a mouse model with which to study transmission and tissue distribution of mutant mtDNA, along with a comparison between mutant load and observation of a pathogenic phenotype.

A number of techniques have been employed to develop a way of creating mice containing a heteroplasmic population of mtDNA. From these experiments two approaches succeeded in producing transmitochondrial mice, i.e., mice carrying two different types of mtDNA. The first method involves fusion of a cytoplast containing one form of mtDNA, with a one-cell embryo, which after a brief culture can be implanted into the fallopian tubes of a pseudopregnant female mouse as a two cell embryo.42 The second approach again involves a cytoplast containing one form of mtDNA but this time it is fused to an undifferentiated female mouse embryonic stem cell with subsequent injection into a mouse blastocyst and implantation into a foster mother.43

Maternal Transmission of a Neutral mtDNA Polymorphism

Under the first method, fusing cytoplasts from either NZB/BINJ or BALB/cByI zygotes to embryos of the opposite origin produced chimeric embryos carrying two variants of mtDNA, as illustrated in Figure 4. From the resulting offspring, nine were found to contain a mixed population of mtDNA and of these, five females were used for backcrossing. These females carried between 3-7% donor mtDNA independent of the source of this donor mtDNA (either NZB/BINJ or BALB/cByI). Proportions of the donor genotype in each female were equivalent to the mean proportion of donor mtDNA in all of her offspring, however, individual offspring varied from 0-30%. Clearly, rapid segregation of mtDNA genotypes was occurring between these two generations of mice through maybe one of two hypothesised mechanisms: 1) Expansion of mtDNA copy number from 103 to 105 occurs during maturation of primary oocytes.44 Replication of a subset of mtDNA molecules during this process could easily shift the proportions of a mtDNA genotype within one generation. 2) Before an embryo initiates mtDNA replication it undergoes a number of cell divisions with partitioning of mtDNA being completely random to each daughter cell.45 The early blastocyst consists of an inner cell mass which gives rise to the three germ layers of the embryo, whilst the remaining cells become extraembryonic tissues. Hence, an unequal distribution of mtDNA molecules within embryonic cells at this stage of development could also lead to a shift in mtDNA genotypes between generations.

Figure 4. Diagram to illustrate the method used to create chimeric embryos containing a mixed population of mitochondria from either the BALB or NZB genetic background.

Figure 4

Diagram to illustrate the method used to create chimeric embryos containing a mixed population of mitochondria from either the BALB or NZB genetic background.

However, neither of these previously proposed hypotheses were found to be true for this animal model. Segregation of mtDNA was found to occur in oogonia, which are the precursors of primary oocytes, and contain a relatively small number of mtDNA templates, hence a possible explanation for the rapid segregation between generations. Both the NZB and BALB mtDNA genotypes should be completely neutral variants and any segregation of these genotypes could only occur by random drift as opposed to selection. Contrary to this belief a follow up study on this mouse model revealed a different rate of segregation in different tissues of the adult mice, with the BALB mtDNA genotype increasing in blood and spleen whilst the NZB genotype increased in liver and kidney.46 No conclusion has been reached as to why this should occur, however, it is clear there are strong tissue-specific selective pressures for different mtDNA genotypes in these animals. A more specific study of hepatocytes from these heteroplasmic mice revealed the selection for NZB mtDNA in the liver was not biased due to an increased oxidative phosphorylation capacity.47 In addition the sequences known to be important in the regulation of mtDNA replication are identical in both NZB and BALB mtDNA genotypes eliminating a difference in replication efficiency. This leaves nuclear encoded factors that function in mtDNA maintenance e.g., proteins involved in expression, replication and mtDNA structure, as likely candidates involved in segregation of different mtDNA genomes. Identification of these factors would certainly help in understanding the segregation patterns observed in different tissues of patients suffering from pathologies due to mtDNA mutations.

Transmitochondrial Mice Harbouring Mutated mtDNA

Embryonic stem (ES) cell cybrids have been successfully used to produce transmitochondrial mice carrying mtDNA from a partially respiratory-deficient chloramphenicol-resistant (CAPR) cell line in two separate experiments.4849 In both cases, CAPR cell lines were enucleated and fused with ES cell cybrids. The mtDNA population of these cybrids were made homoplasmic by Levy et al49 (i.e., carrying exclusively CAPR mtDNA) by treatment with the dye rhodamine-6-G to eliminate CAPS mtDNA. Injection of CAPR ES cell cybrids into blastocysts and subsequent implantation into foster mothers resulted in offspring with a low level of chimerism in various tissues along with detectable levels of CAPR mtDNAs.

In order to create a viable and fertile female transmitochondrial mouse that could generate transgenic mice with altered mtDNA a slightly different approach was taken , as illustrated in Figure 5.43 A female ES cell was chosen that would produce fertile oocytes. This ES cell line CC9.3.1 was first treated with R-6G to inactivate mitochondrial function, then fused with synaptosomes from the brain of an NZB mouse that would act as a neutral mtDNA variant. Selected ES cell cybrids were then injected into B6 cytoplasts and chimeric animals obtained. Using this technique it was possible to maternally transmit the NZB mtDNA variant through multiple generations with a heteroplasmic distribution for the NZB and common haplotype mtDNA throughout the animal.

Figure 5. Diagram showing the creation of ES cells containing mtDNA from the NZB origin.

Figure 5

Diagram showing the creation of ES cells containing mtDNA from the NZB origin. These ES cells were then injected into B6 blastocysts and implanted into pseudopregnant females to produce chimeric mice containing a heteroplasmic distribution of NZB and (more...)

Using the same procedure, CAPR mtDNA was introduced into the CC9.3.1 cell line and ES cytoplasts fused with blastocytes to produce chimeric mice. Characterisation of the CAPR chimeric mice showed ocular abnormalities, including congenital cataracts, decreased retinal function and hamaratomas of the optic nerve. Chimeric females were successful in producing offspring carrying CAPR mtDNA. With the exception of just one pup, all animals born with the CAPR genotype (either homoplasmic or heteroplasmic) died within 12 hours of birth, whilst many more died in utero. The failure to establish mouse lines with maternal transmission of the CAPR mutation makes it difficult to critically evaluate the phenotype.

Mice with Mitochondrial Dysfunction Carrying Rearranged mtDNA

A transmitochondrial mouse model was produced by Hayashi and coworkers50 to replicate the human pathology caused by deletions within the mtDNA genome. In humans large-scale mtDNA deletions (ΔmtDNA) directly affect respiratory chain function resulting in mitochondrial diseases as well as being associated with age-related disorders such as diabetes and neurodegenerative diseases.51 Single deletions of mtDNA generally occur spontaneously with ΔmtDNA not being maternally transmitted and are always as a heteroplasmic mixture with wild-type mtDNA.5253 Levels of ΔmtDNA vary greatly between different tissues and indeed between individual cells within an organ. As with most pathogenic mtDNA mutations a minimum threshold of mutant mtDNA is required before respiratory chain dysfunction is detected.

To generate mice carrying ΔmtDNA, respiratory deficient mouse cybrid cells harbouring deletion mutant mtDNAs were cultured and a single clone containing a 4,696 bp deletion was isolated and determined to contain 30% ΔmtDNA which later increased to 83% before remaining stable. This deletion spanned 6 tRNA genes and 7 structural genes. High percentage mutant cybrids were enucleated and electrofused to pronucleus-stage embryos before being transferred to the oviduct of pseudopregnant females. Out of the resulting F0 offspring 24 mice contained ΔmtDNA at a level of 6-42% and transmission of ΔmtDNA was demonstrated through the female germline from F0 progeny to their offspring. Interestingly, these mice all contained a partially duplicated form of mtDNA consisting of one full-length mtDNA and one ΔmtDNA, this was in addition to the wild type and deleted mtDNA molecules that are present in the original mouse cybrid clone. Maternally transmitted human mitochondrial diseases are commonly associated with partially duplicated mtDNA and not ΔmtDNA so it raised the possibility that maybe ΔmtDNA was not being directly transmitted from one generation of these mice to the next. Instead perhaps ΔmtDNA was interacting with wild-type mtDNA generating partially duplicated mtDNA, allowing transmission to progeny and then later being rearranged to yield ΔmtDNA again.

Analysis of the ΔmtDNA mice revealed COX negative fibres only where ΔmtDNA was predominant with a threshold of 85% mutant mtDNA molecules being necessary. However, the typical RRFs observed in muscle samples from human patients of mitochondrial diseases were not found. The majority of F1-F3 mice died within 200 days of birth due to renal failure with COX activity reduced to 28% that of a wild-type kidney. Mitochondrial dysfunction in the kidney causing renal failure has been reported in children with multisystem disorders due to wide-spread tissue distribution of ΔmtDNA.5456 A more in-depth analysis of individual skeletal muscle fibres to determine the COX activity of individual mitochondria revealed an interesting result.57 Skeletal muscle from a mouse with overall 85% ΔmtDNA had 41% COX negative fibres which were subsequently shown to contain 89% ΔmtDNA whilst the remaining 59% COX positive fibres carried just 80% ΔmtDNA. Electron micrographs showed a complete lack of coexistence of COX positive and COX negative mitochondria within the muscle fibres. Evidently, there must be some form of complementation between ΔmtDNA and wild-type mtDNA with a threshold of more than 85% ΔmtDNA being required before COX negative mitochondria are observed, correlating with the onset of disease phenotypes.

Clearly a very useful tool for the study of human mitochondrial disorders has been established, opening up the possibility to investigate transmission of rearranged mtDNA and the subsequent tissue distribution of ΔmtDNA. In addition, the mechanism of in vivo mitochondrial complementation can be further investigated. This method of introducing mutant mtDNA molecules into a mouse germ line will hopefully aid in the generation of mouse strains which reproduce the phenotypes observed in human sufferers of mitochondrial diseases.

Defective Nuclear-Mitochondrial DNA Interactions Resulting in Hearing Loss

Mitochondrial DNA mutations associated with age-related hearing loss (AHL) are generally homoplasmic in presentation, however, not all individuals with these mutations suffer with a hearing loss whilst others with identical mtDNA mutations will develop a severe hearing loss. These differences in phenotypic expression have lead to the conclusion that nuclear genes must determine whether or not an individual carrying a specific mtDNA mutation will develop symptoms of the disease. Furthermore, a number of inbred mouse strains exhibiting AHL have been shown to share a common gene (designated Ahl) on mouse chromosome 10.58

To test the involvement of mtDNA in AHL three of these inbred mouse strains (A/J, NOD/ LtJ and SKH2/J) were used in backcrosses with a wild-derived inbred mouse strain (CAST/Ei), that does not exhibit AHL.59 Progeny from the first cross (F1) between each of the three hearing impaired strains and CAST/Ei mice did not develop AHL. When the F1 mice were backcrossed only mice originating from the A/J hearing impaired strain developed AHL, mice from the other two backcrosses had normal hearing. It was observed that only those mice homozygous for the A/J allele at the Ahl locus developed AHL supporting the hypothesised link between the Ahl gene and AHL. In addition, mice carrying A/J mtDNA as opposed to CAST/Ei mtDNA had a more severe hearing impairment. By sequencing mtDNA from all three AHL mouse strains the only difference that was observed was a single adenine nucleotide insertion in the D-loop of the tRNA-Arg gene in A/J mtDNA. This insertion is in an adenine repeat where the published sequence contains 8 adenine repeats, NOD/LtJ and SKH2/J strains carry 9 adenines and A/J mtDNA shows 10 repeated adenines. There is no direct evidence showing that the extra adenine renders the mitochondrial tRNA-Arg dysfunctional and it is particularly curious that mice carrying A/J mtDNA but are not homozygous for the A/J Ahl gene have normal hearing. However, the results from these backcrosses do support a link between development of AHL and an interaction between nuclear and mitochondrial DNA. As yet no biochemical analysis on these mice has been carried out so there is no evidence of dysfunction in oxidative phosphorylation. Many human patients with AHL have shown a number of acquired mtDNA mutations in their auditory tissue, which suggests a link between respiratory chain dysfunction and the onset of hearing loss.60 It would be interesting to compare the respiratory chain function of mitochondria isolated from both normal and AHL mice created by the A/J x CAST/Ei backcrosses.


Larsson NG, Clayton D. Molecular genetic aspects of human mitochondrial disorders. Annu Rev Genet. 1995;29:151–178. [PubMed: 8825472]
Smeitink J, van den Heuvel L, DiMauro S. The genetics and pathology of oxidative phosphorylation. Nat Rev Genet. 2001;2:342–352. [PubMed: 11331900]
Munnich A, Rustin P. Clinical spectrum and diagnosis of mitochondrial disorders. Am J Med Genet. 2001;106:4–17. [PubMed: 11579420]
Wallace DC. Mitochondrial diseases in man and mouse. Science. 1999;283:1482–1488. [PubMed: 10066162]
Judd BH, Shen MW, Kaufman TC. The anatomy and function of a segment of the X chromosome of Drosophila melanogaster. Genetics. 1972;71:139–156. [PMC free article: PMC1212768] [PubMed: 4624778]
Shannon MP, Kaufman TC, Shen MW. et al. Lethality patterns and morphology of selected lethal and semi-lethal mutations in the zest-white region of Drosophila melanogaster. Genetics. 1972;72:615–638. [PMC free article: PMC1212859] [PubMed: 4631596]
Ganetzky B, Wu CF. Indirect suppression involving behavioural mutants with altered nerve excitability in Drosophila melanogaster. Genetics. 1982;100:597–614. [PMC free article: PMC1201835] [PubMed: 17246073]
Royden CS, Pirrotta V, Jan LY. The tko locus, site of a behavioural mutation in Drosophila melanogaster, codes for a protein homologous to prokaryotic ribosomal protein S12. Cell. 1987;51:165–173. [PubMed: 3117373]
Shah ZH, O'Dell KMC, Miller SCM. et al. Metazoan nuclear genes for mitoribosomal protein S12. Gene. 1997;204:55–62. [PubMed: 9434165]
Toivonen JM, O'Dell KMC, Petit N. et al. technical knockout, a Drosophila model of mitochondrial deafness. Genetics. 2001;159:241–254. [PMC free article: PMC1461776] [PubMed: 11560901]
Tiranti V, Hoertnagel K, Carrozzo R. et al. Mutations of SURF-1 in Leigh disease associated with cytochrome c oxidase deficiency. Am J Hum Genet. 1998;63:1609–1621. [PMC free article: PMC1377632] [PubMed: 9837813]
Zhu Z, Yao J, Johns T. et al. SURF1, encoding a factor involved in the biogenesis of cytochrome c oxidase, is mutated in Leigh syndrome. Nat Genet. 1998;20:337–343. [PubMed: 9843204]
Kaukonen J, Juselius JK, Tiranti V. et al. Role of adenine nucleotide translocator 1 in mtDNA maintenance. Science. 2000;289:782–785. [PubMed: 10926541]
Graham BH, Waymire KG, Cottrell B. et al. A mouse model for mitochondrial myopathy and cardiomyopathy resulting from a deficiency in the heart/muscle isoform of the adenine nucleotide translocator. Nat Genet. 1997;16:226–234. [PubMed: 9207786]
Melov S, Hinerfeld D, Esposito L. et al. Multi-organ characterisation of mitochondrial genomic rearrangements in ad libitum and caloric restricted mice show striking somatic mitochondrial DNA rearrangements with age. Nuc Acid Res. 1997;25:974–82. [PMC free article: PMC146531] [PubMed: 9023106]
Tanhauser SM, Laipis PJ. Multiple deletions are detectable in mitochondrial DNA of aging mice. J Biol Chem. 1995;270:24769–75. [PubMed: 7559594]
Esposito LA, Melov S, Panov A. et al. Mitochondrial disease in mouse results in increased oxidative stress. Proc Natl Acad Sci USA. 1999;96:4820–4825. [PMC free article: PMC21775] [PubMed: 10220377]
Esposito LA, Kokoszka JE, Waymire KG. et al. Mitochondrial oxidative stress in mice lacking the glutathione peroxidase-1 gene. Free Radical Biology & Medicine. 2000;28:754–766. [PMC free article: PMC3049813] [PubMed: 10754271]
Ho Ye-S, Magnenat J-L, Bronson RT. et al. Mice deficient in cellular glutathione peroxidase develop normally and show no increased sensitivity to hyperoxia. J Biol Chem. 1997;272:16644–16651. [PubMed: 9195979]
de Haan JB, Bladier C, Griffiths P. et al. Mice with a homozygous null mutation for the most abundant glutathione peroxidase, Gpx1, show increased susceptibility to the oxidative stress-inducing agents paraquat and hydrogen peroxide. J Biol Chem. 1998;273:22528–22536. [PubMed: 9712879]
Li Y, Huang T-T, Carlson EJ. et al. Dilated cardiomyopathy and neonatal lethality in mutant mice lacking manganese superoxide dismutase. Nat Genet. 1995;11:376–381. [PubMed: 7493016]
Reaume AG, Elliott JL, Hoffman EK. et al. Motor neurons in Cu/Zn superoxide dismutase-deficient mice develop normally but exhibit enhanced cell death after axonal injury. Nat Genet. 1996;13:43–47. [PubMed: 8673102]
Carlsson LM, Jonsson J, Edlund T. et al. Mice lacking extracellular superoxide dismutase are more sensitive to hyperoxia. Proc Natl Acad Sci USA. 1995;92:6264–6268. [PMC free article: PMC41498] [PubMed: 7603981]
Melov S, Coskun P, Patel M. et al. Mitochondrial disease in superoxide dismutase 2 mutant mice. Proc Natl Acad Sci. 1999;91:12248–12252.
Gardner PR, Nguyen DD, White CW. Aconitase is a sensitive and critical target of oxygen poisoning in cultured mammalian cells and in rat lungs. Proc Natl Acad Sci USA. 1994;91:12248–12252. [PMC free article: PMC45414] [PubMed: 7991614]
Lebovitz RM, Zhang H, Vogel H. et al. Neurodegeneration, myocardial injury, and perinatal death in mitochondrial superoxide dismutase-deficient mice. Proc Natl Acad Sci USA. 1996;93:9782–9787. [PMC free article: PMC38506] [PubMed: 8790408]
Melov S, Schneider JA, Day BJ. et al. A novel neurological phenotype in mice lacking mitochondrial manganese superoxide dismutase. Nat Genet. 1998;18:159–163. [PubMed: 9462746]
Melov S, Doctrow SR, Schneider JA. et al. Lifespan extension and rescue of spongiform encephalopathy in superoxide dismutase 2 nullizygous mice treated with superoxide dismutase-catalase mimetics. The Journal of Neuroscience. 2001;21:8348–8353. [PMC free article: PMC6762800] [PubMed: 11606622]
Kokoszka JE, Coskun P, Esposito LA. et al. Increased mitochondrial oxidative stress in the Sod2 (+/-) mouse results in the age-related decline of mitochondrial function culminating in increased apoptosis. Proc Natl Acad Sci USA. 2001;98:2278–2283. [PMC free article: PMC30129] [PubMed: 11226230]
Petronilli V, Costantini P, Scorrano L. et al. The voltage sensor of the mitochondrial permeability transition pore is tuned by the oxidation-reduction state of vicinal thiols. Increase of the gating potential by oxidants and its reversal by reducing agents. J Biol Chem. 1994;269:16638–16642. [PubMed: 7515881]
Zoratti M, Szabo I. The mitochondrial permeability transition. Biochim Biophys Acta. 1995;1241:139–176. [PubMed: 7640294]
Larsson NG, Wang J, Wilhelmsson H. et al. Mitochondrial transcription factor A is necessary for mtDNA maintenance and embryogenesis in mice. Nat Genet. 1998;18:231–236. [PubMed: 9500544]
Wang J, Wilhelmsson H, Graff C. et al. Dilated cardiomyopathy and atrioventricular conduction blocks induced by heart-specific inactivation of mitochondrial DNA gene expression. Nat Genet. 1999;21:133–137. [PubMed: 9916807]
Li H, Wang J, Wilhelmsson H. et al. Genetic modification of survival in tissue-specific knockout mice with mitochondrial cardiomyopathy. Proc Natl Acad Sci. 2000;97:3467–3472. [PMC free article: PMC16263] [PubMed: 10737799]
Silva JP, Köhler M, Graff C. et al. Impaired insulin secretion and β-cell loss in tissue-specific knockout mice with mitochodnrial diabetes. Nat Genet. 2000;26:336–340. [PubMed: 11062475]
Sörensen L, Ekstrand M, Silva JP. et al. Late-onset corticohippocampal neurodepletion attributable to catastrophic failure of oxidative phosphorylation in MILON mice. J Neurosci. 2001;21:8082–8090. [PMC free article: PMC6763882] [PubMed: 11588181]
Koutnikova H, Campuzano V, Foury F. et al. Studies of human, mouse and yeast homologues indicate a mitochondrial function for frataxin. Nat Genet. 1997;16:345–351. [PubMed: 9241270]
Jiralerspong S, Liu Y, Montermini L. et al. Frataxin shows developmentally regulated tissue-specific expression in the mouse embryo. Neurobiol Dis. 1997;4:103–113. [PubMed: 9331900]
Rötig A, deLonlay P, Chretien D. et al. Frataxin gene expansion causes aconitase and mitochondrial iron-sulfur protein deficiency in Friedreich ataxia. Nat Genet. 1997;17:215–217. [PubMed: 9326946]
Cossée M, Puccio H, Gansmuller A. et al. Inactivation of the Friedreich ataxia mouse gene leads to early embryonic lethality without iron accumulation. Hum Mol Genet. 2000;9:1219–1226. [PubMed: 10767347]
Puccio H, Simon D, Cossée M. et al. Mouse models for Friedreich ataxia exhibit cardiomyopathy, sensory nerve defect and Fe-S enzyme deficiency followed by intramitochondrial iron deposits. Nat Genet. 2001;27:181–186. [PubMed: 11175786]
Jenuth JP, Peterson AC, Fu K. et al. Random genetic drift in the female germline explains the rapid segregation of mammalian mitochondrial DNA. Nat Genet. 1996;14:146–151. [PubMed: 8841183]
Sligh JE, Levy SE, Waymire KG. et al. Maternal germ-line transmission of mutant mtDNAs from embryonic stem cell-derived chimeric mice. Proc Natl Acad Sci. 2000;97:14461–14466. [PMC free article: PMC18941] [PubMed: 11106380]
Piko L, Taylor KD. Amounts of mitochondrial DNA and abundance of some mitochondrial gene transcripts in early mouse embryos. Dev Biol. 1987;123:364–374. [PubMed: 2443405]
Handyside AH, Hunter S. A rapid procedure for visualising the inner cell mass and trophectoderm nuclei of mouse blastocysts in situ using polynucleotidespecific flourochromes. J Exp Zool. 1984;231:429–434. [PubMed: 6209359]
Jenuth JP, Peterson AC, Shoubridge EA. Tissue-specific selection for different mtDNA genotypes in heteroplasmic mice. Nat Genet. 1997;16:93–95. [PubMed: 9140402]
Battersby BJ, Shoubridge EA. Selection of a mtDNA sequence variant in hepatocytes of heteroplasmic mice is not due to differences in respiratory chain function or efficiency of replication. Hum Mol Genet. 2001;10:2469–2479. [PubMed: 11709534]
Marchington DR, Barlow D, Poulton J. Transmitochondrial mice carrying resistance to chloramphenicol on mitochondrial DNA: Developing the first mouse model of mitochondrial DNA disease. Nat Med. 1999;5:957–60. [PubMed: 10426324]
Levy SE, Waymire KG, Kim YL. et al. Transfer of chloramphenicol-resistant mitochondrial DNA into the chimeric mouse. Transgenic Res. 1999;8:137–45. [PMC free article: PMC3049807] [PubMed: 10481313]
Inoue K, Nakada K, Ogura A. et al. Generation of mice with mitochondrial dysfunction by introducing mouse mtDNA carrying a deletion into zygotes. Nat Genet. 2000;26:176–181. [PubMed: 11017072]
Wallace DC. Mitochondrial diseases in man and mouse. Science. 1999;283:1482–1488. [PubMed: 10066162]
Larsson NG, Eiken HG, Boman H. et al. Lack of transmission of deleted mtDNA from a woman with Kerns-Sayre syndrome to her child. Hum Genet. 1992;50:360–363. [PMC free article: PMC1682461] [PubMed: 1734716]
Holt IJ, Harding AE, Morgan-Hughes JA. Deletions of muscle mitochondrial DNA in patients with mitochondrial myopathies. Nature. 1988;331:717–719. [PubMed: 2830540]
Rotig A, Bessis JL, Romero N. et al. Maternally inherited duplication of the mitochondrial genome in a syndrome of proximal tubulopathy, diabetes mellitus, and cerebellar ataxia. Am J Hum Genet. 1992;50:364–370. [PMC free article: PMC1682469] [PubMed: 1531167]
Jansen JJ, Maassen JA, van der Woude FJ. et al. Mutation in mitochondrial tRNA(Leu(UUR)) gene associated with progressive kidney disease. J Am Soc Nephrol. 1997;8:1118–1124. [PubMed: 9219161]
Szabolcs MJ, Seigle R, Shanske S. et al. Mitochondrial DNA deletion: A cause of chronic tubulointerstitial nephropathy. Kidney Int. 1994;45:1388–1396. [PubMed: 8072250]
Nakada K, Inoue K, Ono T. et al. Inter-mitochondrial complementation: Mitochondria-specific system preventing mice from expression of disease phenotypes by mutant mtDNA. Nat Med. 2001;7:934–940. [PubMed: 11479626]
Zheng QY, Johnson KR, Erway LC. Assessment of hearing in 80 inbred strains of mice by ABR threshold analyses. Hearing Res. 1999;130:94–107. [PMC free article: PMC2855304] [PubMed: 10320101]
Johnson KR, Zheng QY, Bykhovskaya Y. et al. A nuclear-mitochondrial DNA interaction affecting hearing impairment in mice. Nat Genet. 2001;27:191–194. [PMC free article: PMC2862214] [PubMed: 11175788]
Fischel-Ghodsian N, Bykhovskaya Y, Taylor K. et al. Temporal bone analysis of patients with presbycusis reveals high frequency of mitochondrial mutations. Hearing Res. 1997;110:147–154. [PubMed: 9282897]
Copyright © 2000-2013, Landes Bioscience.
Bookshelf ID: NBK6024


  • PubReader
  • Print View
  • Cite this Page

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...