Ko Samuel. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2001. / Includes bibliographical references (p. 230-252). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

Atherosclerosis is the leading cause of death in the Western world. Although mitochondrial DNA (mtDNA) damage has been implicated in atherosclerosis, it is unclear whether the damage is sufficient to impair mitochondrial respiration, and mitochondrial dysfunction has not been demonstrated. Treatment of vascular smooth muscle cells (VSMCs) with an atherogenic lipid, oxidised low-density lipoprotein (OxLDL), dose dependently decreased basal and maximal respiration and fat-feeding of apolipoprotein E deficient (ApoE-/-) mice reduced mitochondrial DNA copy number relative to nuclear DNA in aortas. Mitochondrial respiration of ApoE-/- mouse aortas, assessed through a 24-well Seahorse extracellular flux analyser, was not affected prior to the development of atherosclerotic plaques. Developed human carotid atherosclerotic plaques were dissected into defined regions including healthy media, shoulder region, fibrous cap and core and their respiration was investigated. The respiratory reserve capacity (RRC) of the shoulder region was similar to the media. However, the cap RRC was significantly reduced compared to healthy media. In contrast, the extracellular acidification rates (ECAR) of the media, shoulder, cap and core regions were similar. In addition, mtDNA copy number was significantly reduced in tissues derived from human plaques compared to healthy arteries and expression of complexes I and II of the electron transfer chain (ETC) were significantly reduced in plaque VSMCs. OxLDL induced mitophagy in human VSMCs and plaque VSMCs demonstrated increased levels of mitophagy without compensatory upregulation of proteins involved in mitochondrial biogenesis. Understanding the role of mitochondrial metabolism and signalling is important for our understanding of disease progression and may lead to future therapeutic targets.

Curcumin is a natural compound that improves blood glucose management. While some evidence from isolated mitochondria indicates that curcumin uncouples electron transport from oxidative phosphorylation, the effects of curcumin on mitochondrial respiration and hydrogen peroxide emission in intact skeletal muscle cells are not known. By assessing rates of oxygen consumption, we demonstrated for the first time that curcumin (40 µM) reduced the mitochondrial coupling efficiency (percentage of oxygen consumption that supports ATP synthesis) of intact skeletal muscle cells. A 30-minute incubation with curcumin decreased mitochondrial coupling efficiency by 17.0 ± 0.4% relative to vehicle (p < 0.008). Curcumin also decreased the rate of hydrogen peroxide emission by 43 ± 13% compared to vehicle (p < 0.05). Analysis of cell respiration in the presence of curcumin revealed a 40 ± 4% increase in the rate of oxygen consumption upon curcumin administration (p < 0.05 compared to vehicle). In additional experiments, no difference in mitochondrial coupling efficiency was observed between vehicle- and curcumin-pretreated cells after permeabilization of cell membranes (p > 0.7). The possibility of synergistic effects between curcumin and ursolic acid, another natural compound that improves blood glucose management, was also examined. Interestingly, ursolic acid (0.12 µM) increased mitochondrial coupling efficiency by 4.1 ± 1.1% relative to vehicle (p < 0.008) and attenuated the effect of curcumin when the two compounds were used in combination (decreased mitochondrial coupling efficiency by 8.0 ± 0.9% compared to vehicle, p < 0.008). These results provide evidence for lower mitochondrial coupling efficiency and hydrogen peroxide emission as possible contributors to the increased glucose uptake and insulin sensitivity of subjects after treatment with curcumin but not ursolic acid. Unless cells are assessed in the intact condition, changes to mitochondrial coupling efficiency after curcumin treatment may go unnoticed.

Ca2+-mediated regulation of gene expression plays an important role in neuronal plasticity. NFAT (Nuclear Factor of Activated T-cells) is a Ca2+/calcineurin (CaN)-dependent transcription factor that has been implicated in a number of neuronal functions including axon outgrowth, presynaptic remodeling and neural survival. NFAT is activated by Ca2+/CaN-dependent dephosphorylation, whereas re-phosphorylation by glycogen synthase kinase-3β (GSK3β) and several other protein kinases deactivates NFAT and triggers its export from the nucleus. In addition to electrically-mediated Ca2+ signals, neurotrophins can potently regulate NFAT function in neurons as well. However the mechanisms of NFAT activation by electrical activity and neurotrophins are not completely understood. In aim 1, I found that electrical stimulation produced a mitochondrial Ca2+ cycling-mediated prolonged [Ca2+]i elevation (plateau), which profoundly affected NFAT activity. The elimination of the [Ca2+]i plateau by blocking mitochondrial Ca2+ uptake or release strongly reduced nuclear import of NFAT. Furthermore, preventing Ca2+ mobilization from mitochondria diminished NFAT-mediated transcription. In aim 2, I found that NGF, a family of neurotrophins, potentiated NFAT-dependent transcription triggered by electrical activity through the TrkA-PI3K-Akt-GSK3β pathway and this effect was mediated primarily by NFATc3. Monitoring NFATc3 movement in DRG neurons in real time showed that NGF slowed the rate of NFATc3 nuclear export, which was mimicked by inhibiting GSK3β, whereas blockade of PI3K prevented this effect. Taken together, I proposed that mitochondrial Ca2+ cycling functions as a novel regulatory mechanism for NFAT activation and NFATc3 serves as an integrator of electrical activity and neurotrophin signaling in the regulation of gene expression in DRG neurons.

Mitochondrial Ca2+ buffering is an important physiological modulator of neuronal signaling and bioenergetics, but this propensity toward Ca2+ regulation proves pathological during excitotoxic insult. Specifically, excessive mitochondrial Ca2+ uptake is a key component of glutamate toxicity within the penumbra surrounding the ischemic core following stroke. This mitochondrial toxicity and Ca2+ dyshomeostasis may be visualized in real time as delayed calcium deregulation (DCD). DCD is a predictor of neuronal, excitoxic death, and is composed of three phases: 1) an initial response; 2) a latent period of elevated, but stable cytosolic Ca2+; and 3) failure of mitochondrial Ca2+ retention, termed deregulation. The duration of the latent period is an index of neuronal resistance. Mitochondria are dynamic organelles that rapidly and reversibly undergo fission and fusion (MFF). MFF is tightly regulated by the phosphoregulation of fission inducing Drp1 at serine 656. Drp1-S656 phosphorelation is mediated by PKA/AKAP1, and it is dephosphorylated by PP2A/Bβ2. Phosphorylation of Drp1-S656 inactivates this contractile GTPase resulting in inhibition of mitochondrial fission and a shift toward elongated mitochondria. This PKA/AKAP1 dependent Drp1-S656 phosphorylation has proven to be neuroprotective. Likewise, attenuation of PP2A/Bβ2 signaling enhances neuronal survival during ischemia and excitotoxic insult. Based on the mitochondrial buffering role in excitotoxicity and MFF modulation of neuronal survival, we began investigating the role of Ca2+ buffering as a function of MFF during glutamate toxicity. Noted above, resistance to excitoticity is visualized by the duration of the DCD latent period. Overexpression of AKAP1 in cultured hippocampal neurons greatly prolonged DCD latency in a PKA dependent manner, while Bβ2 ablation prolonged DCD latency by hours. Pharmacological modulation of PKA required PDE4 inhibition to reproduce the AKAP1 observations. Preliminary experiments studying the effect of Bβ2 overexpression on matrix Ca2+ load suggests possible mechanism of MFF regulated of matrix Ca2+ accumulation. Using mtPericam DRG neurons as a model system for individual mitochondrial Ca2+ recording, we discovered impaired extrusion kinetics in mitochondria fragmented by both Drp1 and Bβ2 overexpression. Ca2+ uptake was comparable to that of control. Extreme elongation of mitochondria via dominant negative Drp1-K38A enhanced recovery. Understanding these observations, however, requires knowledge of the mitochondrial Ca2+ buffering mechanism. Mitochondrial uptake candidates include MCU and ccdc109b. Our neuronal characterization of MCU confirms a role in mitochondrial Ca2+ buffering, but not a requirement; other components must be involved. Ccdc109b remains an inconclusive candidate, but may be an important regulator of MCU. Mitochondrial efflux transporters include Letm1 and NCLX. Though Letm1 observations are hindered by control artifact, preliminary evidence supports a role in extrusion. The role of NCLX is complicated by possible tissue specificity. Functional expression experiments utilizing Na+ free Li+ external solution suggests absence of NCLX in hippocampal neurons; DRG neurons were capable of Li+ exchange. The above observations confirm the significance of mitochondrial Ca2+ extrusion in neuronal survival. Understanding the mechanisms and regulation of mitochondrial Ca2+ transport has the potential to provide novel therapeutic targets in pathologies of excitotoxic etiology.

It has been shown that reactive oxygen species (ROS) are released in response to articular cartilage injury. The excessive release of ROS has been shown to be mitochondrial in nature and leads to chondrocyte death which in turn can lead to post-traumatic osteoarthritis (PTOA). Evidence suggests that mitochondria are attached to chondrocytes' cytoskeleton. Upon tissue level deformation, it is believed that mitochondria also experience deformation in response to cytoskeletal strain, releasing ROS. Therefore, it was hypothesized that inhibition of chondrocytes' cytoskeleton would prevent mitochondrial distortion rendering them unable to release ROS in response to the applied strain, saving chondrocytes. Osteochondral explants treated with cytoskeletal inhibitors were found to reduce mitochondrial ROS production directly after impact and increase chondrocyte viability 24 hours after impact. The release of mitochondrial ROS is an important mechanotranduction pathway in the initiation of PTOA.

No description available.

Myostatin is a negative regulator of muscle growth. When it is not present or non-functional double-muscling occurs, the primary characteristic of this phenotype being an increase in muscle mass. Another characteristic of double-muscling is an increased proportion of type IIB muscle fibres, which rely on glycolysis as their primary energy source, as opposed to type IIA and type I fibres which rely on oxidative phosphorylation. This switch in muscle metabolism directly impacts on the mitochondria, as mitochondria from glycolytic muscle fibres have been shown to have differences in metabolic activity. The increased proportion of glycolytic muscle fibres present in myostatin knockout animals provides a unique model to investigate alterations in muscle fibre type metabolism. The mouse model of myostatin knockout utilised during this study was generated by genetic deletion of exon three of the myostatin gene. Verification of this knockout was attempted by western blot analysis, but only the latency associated protein (LAP) was detected. Interestingly, the LAP was barely detectable in the knockout muscle suggesting deletion of exon three affects binding of anti-myostatin antibodies to the LAP, as that part of the gene is not deleted. A comparison of the basal mitochondrial stress levels was made, also by western blot analysis. The knockout mitochondria showed no change in levels of heat shock protein 60 or superoxide dismutase 2, indicating that they are not being subjected to any increased stress due to the myostatin knockout phenotype. A comparative proteomics approach was used to detect changes in the mitochondrial proteome of myostatin knockout gastrocnemius muscle to gain clues to how mitochondria from glycolytic muscle fibres differ from those present in oxidative fibres. This was undertaken using two-dimensional electrophoresis (2-DE), in-gel tryptic digests and peptide mass fingerprinting by mass spectrometry. A 2-DE gel protein loading of 220 g was shown to give the best protein spot resolution and the most crucial step in the loading process was found to be the laying of the immobilized pH gradient, which had to be performed very carefully to obtain a consistent loading pattern. This study resolved only around 160 protein spots out of the estimated 1,000 to 2,000 proteins present in the mitochondria. Modulation of six proteins was seen at a plt0.1 level, but were unable to be identified using the current methodology. More abundant mitochondrial proteins were able to be identified, but showed no significant modulation. Malate dehydrogenase and 3-hydroxyacyl-CoA dehydrogenase, which were identified during this study, have been reported to have decreased activity in mitochondria from glycolytic muscle fibres. This study suggests that the change in activity observed by other researchers is due to inhibition of these enzymes in the glycolytic fibres or activation in the oxidative fibres.

<p>As a consequence of the presence of both mitochondria and chloroplasts in plant cells there is a higher sorting requirement in a plant cell than that in a non-plant cell. Reflecting this, protein import to mitochondria and chloroplasts has been shown to be highly specific. However, there is a group of proteins which are encoded by a single gene in the nucleus, translated in the cytosol and targeted to both mitochondria and chloroplasts. These proteins are referred to as dual targeted proteins. The first protein shown to be dual targeted was pea glutathione reductase (GR). The focus of this thesis is the targeting properties of the dual targeted protein glutathione reductase.</p><p>In order to overcome the limitations with traditional in vitro import systems we have developed an import system for simultaneous import of precursor proteins into mitochondria and chloroplasts (dual import system). The chloroplastic precursor of the small subunit of ribulose bisphosphate carboxylase/oxygenase (SSU) was mis-targeted to pea mitochondria in a single import system, but was imported only into chloroplasts in the dual system. The dual GR reductase precursor was targeted to both mitochondria and chloroplasts in both the single and dual import system.</p><p>We have investigated the targeting and processing properties of the GR targeting signal. Using N-terminal truncations we have demonstrated that the GR targeting signal has a domain organisation. Our results show that GR has evolved a dual targeting signal with the C-terminal part being sufficient for chloroplast import, the internal part required for the mitochondrial import and the N-terminal part housing a “fine-tuning” function. Furthermore, we have constructed a range of point mutations on the GR signal sequence changing positive amino acid residues and stretches of hydrophobic amino acid residues. Overall single mutations had a greater effect on mitochondrial import compared to import into chloroplasts. We have also shown that the recognition of the GR processing site differs between MPP and SPP. Single amino acid substitutions in the vicinity of the processing site clearly affected processing by MPP while processing by SPP showed low sensitivity to single mutations.</p>

Electron paramagnetic resonance (EPR) and Mössbauer spectroscopy were used to examine the distribution of iron in mitochondria from Saccharomyces cerevisiae. These organelles were packed into EPR and Mössbauer cuvettes, affording spectra with unprecedented signal/noise ratios. EPR spectra of as-isolated intact mitochondria exhibited fourteen distinct signals, some of which were assigned according to previously reported g-values obtained using isolated proteins. Signals from adventitious manganese (II) and iron (III) were largely removed when mitochondria were isolated in buffers supplemented with the metal chelators EDTA or EGTA. Signals were simulated and intensities were quantified to afford spin concentrations and estimates of the concentration of EPR-active species in mitochondria. The effects of treating samples with chemical modifiers were examined. Packed samples were analyzed for protein and metal content, affording averaged values of 50 mg/mL [protein], 590 µM [Fe], 340 µM [Cu], and 17 µM [Mn]. 57Fe-enriched intact mitochondria isolated in the presence of metal chelators exhibited Mössbauer spectra dominated by three components. Approximately 60% of the 57Fe in the sample gave rise to a quadrupole doublet, most of which was diamagnetic. The parameters of this doublet are typical of S = 0 [4Fe-4S]2+ clusters and S = 0 ferrous heme groups. Spectra of samples reduced with dithionite, pH 8.5, suggested that at least half of this doublet arose from [4Fe-4S]2+ clusters. The second major component exhibited in the Mössbauer spectra arose from high-spin ferrous ions (10%-30%). The third major component (15%) came from iron exhibiting magnetic hyperfine interactions and is likely reflected in the Fe-containing species observed by EPR. The results presented here suggest that mitochondria contain ~ 600 µM of Fe overall, ~ 200 – 400 µM organized as [4Fe-4S]2+ clusters, with about 25 µM due to the [4Fe-4S]2+ cluster of aconitase. Approximately 60 µM – 200 µM of the Fe in mitochondria is high-spin ferrous ions, ~ 40 µM as the Rieske S = 1/2 [2Fe-2S]+ cluster of cytochrome bc1, and ~20 µM as the S = 1/2 [2Fe-2S]+ cluster of succinate dehydrogenase. The high-spin ferric hemes of the a3:CuB site of cytochrome oxidase and cytochrome c peroxidase each account for ~ 4 µM of Fe.