2.6.1 AOX expression in cells
Ciona AOX was first introduced into human cells in 2006 (Hakkaart et al., 2006).
Human embryonic kidney (HEK293T) cells induced to express AOX showed resistance to antimycin, an inhibitor of Complex III, and cyanide, a well-known inhibitor of Complex IV of the RC. This resistant respiration was blocked by the AOX inhibitor n-propyl gallate (nPG), implying that the resistance was AOX dependent. Addition of nPG to AOX–expressing cells in the absence of cyanide had only a mild effect on oxygen consumption, showing that in non-stressed conditions the presence of AOX was not interfering with the function of the classical OXPHOS complexes. Overall, the enzyme did not affect the proliferation of the cells under standard culture conditions, but did decrease superoxide dismutase (SOD) activity, suggesting a role as a functional antioxidant (Hakkaart et al., 2006).
In cell models with cytochrome c oxidase (COX) deficiency, resulting from RNAi knockdown of COX10 in HEK293T-derived cells, or a mutation in COX15 in patient-derived fibroblasts, AOX was able to restore respiration by maintaining electron flow and improving growth in the absence of glucose (Dassa et al., 2009).
On the other hand, an increase in glucose was found to impair AOX function in this cell-line, which was deducted to be due to Complex I downregulation (Cannino et al., 2012).
2.6.2 AOX expression in Drosophila
AOX was first introduced into Drosophila as a highly expressed UAS-construct, to be expressed using the GAL4 or GeneSwitch system (Fernandez-Ayala et al., 2009). It has also been expressed FRQVWLWXWLYHO\ XQGHU WKH ơ-tubulin promoter as tub-AOX (Kemppainen et al., 2014b). The enzyme, expressed by either means, caused no
48
detrimental phenotype in the flies under standard growth conditions. The flies developed normally and were fertile. However, when AOX expression was driven by GAL4, flies showed a slight developmental delay of 0.5 day, as well as an exaggerated posteclosion weight loss.
Transgenic AOX in Drosophila was inferred to be active in vivo, since AOX-expressing flies showed resistance to both antimycin and cyanide (Fernandez-Ayala et al., 2009). When introduced into Drosophila models of mitochondrial dysfunction, AOX was able to rescue COX deficiencies caused by partial knockdown of either a structural subunit of Complex IV, Cox6c (cyclope) or the Complex IV-assembly factor Surf1. AOX was also able to rescue the locomotor phenotype of a dj-Ƣ hypomorph, considered as a Drosophila model for Parkinson’s disease (Fernandez-Ayala et al., 2009). Further studies showed that even with the lower expression levels of tub-AOX, the enzyme was able to partially rescue deficiencies caused by knockdown of different COX subunits many of which are lethal or cause severe neurological phenotypes (Kemppainen et al., 2014b).
In Drosophila, downregulation of JNK signaling in the dorsal thorax causes a thoracic closure defect that was corrected by expression of AOX. However, the defect could not be overcome by AOX when it was caused by downregulating late steps of the JNK pathway, nor if downstream targets of JNK were targeted (Andjelkovic et al., 2018). AOX has also been introduced into a fly model of Alzheimer’s disease (AD), that expresses human amyloid-Ƣ $Ƣ peptide that, by forming plaques, is widely considered to be the main pathological cause of AD. In the Drosophila AD model, AOX was able to modestly increase the shortened lifespan and EORFNHGR[LGDWLYHVWUHVVWKDWLVSUHVXPHGWREHLQYROYHGLQIRUPDWLRQRIWKH$Ƣ plaques (El-Khoury et al., 2016).
However, AOX has not been able to compensate for mitochondrial gene expression defects in Drosophila models. AOX expression had no effect on defects caused by mutated POLGWKHFDWDO\WLFVXEXQLWRIWKHPLWRFKRQGULDOUHSOLFDVHSROƣ nor on defects caused by mutated mitochondrial replicative DNA helicase Twinkle.
49
In fact, introduction of AOX slightly worsened the larval lethality in Twinkle mutants (Rodrigues et al., 2018). AOX was also unable to rescue the phenotype of technical knockout25t (tko25t) mutant flies, harbouring a point mutation in the gene coding for mitoribosomal protein S12. The tko25t phenotype consists of multiple defects, including developmental delay and sensitivity to seizures caused by mechanical stress known as bang sensitivity, neither of which AOX was able to improve (Kemppainen et al., 2014a). Although able to alleviate RC defects caused by mutations in single subunits of the enzyme complexes and decrease oxidative stress in disease models where ROS is involved, it seems that AOX cannot rescue mitochondrial dysfunction caused by mtDNA expression and translation defects that commonly lead to multiple RC deficiencies.
2.6.3 AOX expression in mice
An AOX-expressing mouse, MitAOX, was first reported in 2013 by El-Koury et al.
(2013). The enzyme caused no deleterious effects on the physiology of the animals, although they were slightly lighter in weight and had mildly increased body temperature. The mice showed resistance to cytochrome chain inhibition when exposed to lethal doses of gaseous cyanide. However, the insertion of the AOX being conducted by a non-targeted lentiviral transduction resulted in insertions at multiple genomic sites and low expression levels of the enzyme. In addition, AOX was not expressed in all tissues (El-Khoury et al., 2013). Another transgenic mouse expressing AOX was later reported by Szibor et al. (2017). The gene was expressed as a single copy introduced by targeted insertion at the Rosa26 locus and was confirmed to produce a functional, ubiquitously expressed enzyme that provided the animal with resistance to cyanide (Szibor et al., 2017).
Mutation in BCS1L, a mitochondrial inner membrane translocase required for assembly of Complex III, causes a severe pathological phenotype in humans known as GRACILE syndrome (fetal growth restriction, aminoaciduria, cholestasis, liver
50
iron overload, lactic acidosis, and early death during infancy). When crossed with mice with homozygous knock-in of the GRACILE mutation in BCS1L, AOX was able to alleviate multiple pathologies, e.g. increasing lifespan, preventing lethal cardiomyopathy and restoring RC function (Rajendran et al., 2019). AOX has also been shown to protect from cigarette smoke-induced tissue damage in mice (Giordano et al., 2018) and ROS-induced inflammation in a broader study on the role of Complex II in sepsis (Mills et al., 2016).
However, when AOX was introduced into COX15 knockout mice suffering from mitochondrial myopathy, instead of providing rescue, AOX expression worsened the phenotype by shortening lifespan and exacerbating the myopathy (Dogan et al., 2018). This was unexpected since AOX provided partial rescue in COX-knockdown models of Drosophila (Kemppainen et al., 2014b). Physiological characterization of AOX-expressing mouse model demonstrated no obvious physiological differences in body composition or cardiac performance on high-fat or ketogenic diet compared to non-transgenic mice (Dhandapani et al., 2019), while AOX-expressing mammalian cells have been found to respond to nutrient availability differently from controls (Cannino et al., 2012). On the other hand, it highlights the need for better understanding of the metabolic effects and regulation of AOX in transgenic models at both the organismal and tissue level. The therapeutic benefits of AOX seem to be limited to specific types or contexts of mitochondrial dysfunction and mapping potential limitations of the enzyme will give a better insight to the type of diseases in which it may be used in as a treatment in future.
51
3 AIMS OF THE STUDY
The overall aim of this study was to investigate the metabolic consequences and potential disadvantages of the presence of the alternative oxidase in higher eukaryotes/metazoans under environmental stress. The therapeutic potential of AOX demonstrated in several disease models has made it crucial to establish what effects AOX may have on the metabolic state of the organism, particularly in stress inducing conditions that are often encountered in the natural environment.
The strategy I adopted was to expose transgenic Drosophila melanogaster expressing AOX to environmental stress by two different approaches, namely by reproductive competition (I) and restricting the availability of nutritional resources (II and III).
1 The aim of the first part of the study was to test the potential selective disadvantage brought on by AOX expression. This was conducted by a natural selection assay, using the sperm competition paradigm and followed up by studying the physiological consequences of AOX expression on the male reproductive organs.
2 In the second part of the study, the aim was to study the consequences of AOX expression on the development of Drosophila reared under nutritional restriction. The objective was to identify potential detrimental effects that the enzyme might have when the organism is exposed to a nutritionally constrained environment.
3 The final aim of the study was to determine the mechanism by which AOX expression disturbs Drosophila development in a low-nutrient environment, with a focus on metabolic imbalances.
52