A novel variant in COX16 causes cytochrome c oxidase deficiency, severe fatal neonatal lactic acidosis, encephalopathy, cardiomyopathy, and liver dysfunction

Abstract

COX16 is involved in the biogenesis of cytochrome-c-oxidase (complex IV), the terminal complex of the mitochondrial respiratory chain. We present the first report of two unrelated patients with the homozygous nonsense variant c.244C>T(p. Arg82*) in COX16 with hypertrophic cardiomyopathy, encephalopathy and severe fatal lactic acidosis, and isolated complex IV deficiency. The absence of COX16 protein expression leads to a complete loss of the holo-complex IV, as detected by Western blot in patient fibroblasts. Lentiviral transduction of patient fibroblasts with wild-type COX16 complementary DNA rescued complex IV biosynthesis. We hypothesize that COX16 could play a role in the copper delivery route of the COX2 module as part of the complex IV assembly. Our data provide clear evidence for the pathogenicity of the COX16 variant as a cause for the observed clinical features and the isolated complex IV deficiency in these two patients and that COX16 deficiency is a cause for mitochondrial disease.

Keywords: COX16; OXPHOS; assembly factor; cardio-encephalopathy; mitochondrial complex IV deficiency.

Figure 1

Lentiviral complementation of patient (Pat) fibroblasts with wild‐type COX16 complementary DNA (cDNA) with (COX16/V5) or without V5 tag (COX16) rescues complex IV deficiency: (a) Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS‐PAGE)/Western blot of mitochondrial fractions of the transduced Patient (Pat) and control (Ctrl) cell lines shows an expression of the transgenic proteins GFP/V5 and COX16/V5 detected with the anti‐V5 antibody. (b) Graph of the relative activities of the oxidative phosphorylation (OXPHOS) enzymes in COX16‐transduced Pat and Ctrl cell lines expressed as a percentage of the activities of the GFP‐transduced in the same cell lines (%GFP). This shows a significant rescue of complex IV activity in the COX16 cDNA‐transduced Pat cell lines (Pat COX16 and Pat COX16/V5) (***p < .001, two‐way analysis of variance [2way ANOVA]). Except for the minor change in complex I transduced with COX16/V5 cDNA (**p < .01, two 2way ANOVA), there are no significant changes in the relative activities of the other OXPHOS enzymes (ns). The error bars indicate standard deviation. C) Blue native‐PAGE (BN‐PAGE)/western blot of mitochondrial fractions of the transduced Pat and Ctrl cell lines with wild‐type COX16 cDNA shows a clear rescue of the assembly of complex IV in the Pat cell lines, whereas the GFP cDNA complemented Pat cell line has undetectable complex IV. Antibodies were used against NDUFA9 for complex I, UQCRC2 for complex III, ATP5A for complex V, COX4 for complex IV, SDHA for complex II. (d) SDS‐PAGE/Western blot of mitochondrial fractions of the transduced Pat and Ctrl cell lines shows a clear expression of both the COX16 and the transgenic COX16/V5 proteins. In addition, the expression levels of all the tested subunits, COX1, COX2, and to a lesser extend COX4 and COX5A, are increased in Pat cells expressing wild‐type COX16 cDNA. Antibodies were used against COX16, COX1, COX2, COX4, COX5A; loading Ctrl CII SDHA subunit was detected with anti‐SDHA. (e) 2D‐BN‐PAGE/SDS‐PAGE of mitochondrial fractions (60 µg of protein) of wild‐type COX16 cDNA and GFP cDNA‐transduced Pat cell lines are compared with a Ctrl cell line to analyze subassembly products. The Pat + GFP shows more COX1 in the mitochondrial translation regulation assembly intermediate of cytochrome c oxidase assembly intermediate than Ctrl cells; this reduces and shifts towards the holo‐complex after complementation with COX16/V5 and COX16. COX2 assembly intermediates are not visible and COX2 is barely visible in the holo‐complex in the Pat + GFP. After complementation with both COX16 cDNAs, the COX2 could be detected in the holo‐complex. Antibodies were used against COX1 and COX2 subunits. The Western blot data in this figure was obtained in N = 1 independent experiment for (a) and (e), and N = 2 for (c) and (d). The data of the enzyme activities of (b) are obtained in N = 1 experiment with two duplicates. Duplicates are measured in two independent experiments and were only accepted when each of the duplicate values was within a 10% range of their average