Mitochondrial complex IV deficiency, caused by mutated COX6B1, is associated with encephalomyopathy, hydrocephalus and cardiomyopathy

Abstract

Isolated cytochrome c oxidase (COX) deficiency is a prevalent cause of mitochondrial disease and is mostly caused by nuclear-encoded mutations in assembly factors while rarely by mutations in structural subunits. We hereby report a case of isolated COX deficiency manifesting with encephalomyopathy, hydrocephalus and hypertropic cardiomyopathy due to a missense p.R20C mutation in the COX6B1 gene, which encodes an integral, nuclear-encoded COX subunit. This novel mutation was predicted to be severe in silico. In accord, enzymatic activity was undetectable in muscle and fibroblasts, was severely decreased in lymphocytes and the COX6B1 protein was barely detectable in patient's muscle mitochondria. Complementation with the wild-type cDNA by a lentiviral construct restored COX activity, and mitochondrial function was improved by 5-aminoimidazole-4-carboxamide ribonucleotide, resveratrol and ascorbate in the patient's fibroblasts. We suggest that genetic analysis of COX6B1should be included in the investigation of isolated COX deficiency, including patients with cardiac defects. Initial measurement of COX activity in lymphocytes may be useful as it might circumvent the need for invasive muscle biopsy. The evaluation of ascorbate supplementation to patients with mutated COX6B1 is warranted.

Figure 1.

Head CT. Head CT performed at age 2 years shows prominent dilatation of the ventricles, VP shunt in the right occipital region, and of note is the very thin cortex.

Figure 2

Structure-based assessment of effect the R20C and R20H mutations in the COX6B1 protein. (a) Structure of the wild-type protein (PDB id 2eij).^{, ,} Note the major contribution of R20 to a network of hydrogen bonds involving residues D18-R20-D36-R39 that orients the n-terminal tail (including residues D18 and D20) relative to the adjacent helix (including D36 and R39). The local area shown in the following panels is highlighted by a square. (b) In order to assess the effects of the mutations R20C (from this study) and R20H, respectively, these mutations were introduced in silico: Wild type and mutations are shown in the same one figure (upper panel), and for better clarity, also in separate panels for R20C (middle panel) and R20H (lower panel). Notably, neither histidine nor cysteine are able to form adequate hydrogen bonds, owing to both size and orientation. However, the effect of cysteine is predicted to be more dramatic due to loss of positive charge.

Figure 3.

Western blotting analysis. Equal amount of protein (8 μg) containing equal amount of CS activity (3.5 mU), from control (C) and patient (P) muscle mitochondria was separated by Urea-SDS-PAGE and subjected to western blotting with primary antibodies: anti COX6B1, anti-MTCO1 (COX I), anti-Complex IV subunit II (COX II), anti-complex II 70 KDa subunit (SDH). These were subsequently detected by peroxidase conjugated secondary antibodies and visualized by chemiluminescence (a). Band intensities quantified depicted in the graph relatively to SDH showed severely decreased COX6B1 and partially decreased COX II and COX I (b).

Figure 4

The effect of small molecules on patient's fibroblasts. Control (C) fibroblasts (n=5), patient (P) fibroblasts or patient fibroblast transfected with wt gene (P+COX6B1) were seeded and grown in GLU or GAL medium. Patient's fibroblasts were also grown on GAL in the presence of either of the following compounds: AICAR; ascorbate; bezafibrate; N-acetylcysteine (NAC) or resveratrol or vehicle (no additive). Cell growth (a) was measured by methylene blue at 620 nm (A620). ATP content (b) was measured by luciferin-luciferase and normalized to growth. Values are presented as normalized mean±SEM. $ Statistically significant P<0.05, when compared with normal control. Asterisk (*) statistically significant when compared with untreated patient cells.