Mitochondrial disorders in the South African context : a clinical and biochemical approach
In the past, patients with suspected mitochondrial disorders (MDs) were identified only phenotypically in South Africa. However, the specific population-related characteristics were unknown and not documented. Up to as late as 1998, no facility was available in South Africa to confirm the diagnosis of MDs. The diagnosis of MDs is challenging under the best of circumstances, and thus it posed an opportunity to develop imaginative diagnostic strategies in a developing country such as South Africa with other major health-related issues. In order to develop a comprehensive service in a country burdened by tuberculosis and human immunodeficiency virus, it was important firstly to define the patient population. It was found that the MD phenotype in the South African population was unique compared with described populations in other countries. African patients predominantly presented with a myopathy and combined enzyme deficiencies in contrast to Caucasian patients, who predominantly presented with an encephalopathy or encephalomyopathy and tended to have more single enzyme deficiencies. Interesting case presentations were identified, including a young adult male patient who presented with Kearns–Sayre Syndrome (KSS). A novel deletion of 3,431 base pairs (bp) between positions 7,115 and 10,545, flanked by a five bp direct repeat sequence, was found in 80% of this patient‘s muscle mitochondrial DNA (mtDNA). It was also demonstrated in this case that the absence of mtDNA-encoded ATPase6 and ATPase8 genes resulted in the aberrant synthesis of adenosine triphosphate (ATP) synthase. Obtaining muscle biopsies in children was extremely difficult and was also complicated by patients living in remote areas with limited access to health care facilities. Consequently, an alternative, less invasive option of analyzing urine was investigated. A metabolomics approach was evaluated by firstly investigating the organic acid-containing section of the metabolome, obtaining urine of patients with respiratory chain disorders (RCDs). It was possible to compile the first comprehensive list of 24 metabolites associated with RCDs, which were, both statistically and practically, significantly elevated. Secondly, a global metabolic profile involving carbohydrate, amino acid and fatty acid catabolism was also constructed. It clearly illustrated the diversity and complexity of the complex biochemical consequences in RCDs and that there was no single characteristic organic acid biomarker profile to distinguish between the complex I (CI), CIII and multiple complex deficiencies. Thirdly, amino acid and carnitine analyses were added to the metabolic profile to assist further in the development of an explorative biosignature. Finally, a biosignature comprising of six organic acids, six amino acids and one other marker was constructed. It included succinic acid, lactic acid, 3-OH-isobutyric acid, 3-OH-valeric acid, 3-OH-3-Me-glutaric acid, 2-OH-glutaric acid, α-aminoadipic acid, glutamic acid, alanine, glycine, serine, tyrosine, and creatine. Differences between population groups, as observed in the clinical study, were not observed in the metabolomics studies, but the statistical processes in variable and case selection aimed to have complete separation between controls and patients. This resulted in a more homogenous patient group, a prerequisite to identify markers for RCDs. A limited number of Caucacian patients was finally included in the two different metabolomics studies, 11/39 (28.2%) and 5/20 (25.0%) respectively. Except for one Caucasian patient with a pure encephalopathy, all the others had muscle involvement as well. The clinical differences observed between African and Caucasian patients therefore remain to be investigated on a biochemical level in a follow up study. This study was a multi-disciplinary project, with a clinical-biochemical approach, since 2009, which successfully described the South African RCD patient profile and which can lead to the development of a refined diagnostic service in South Africa. In addition, a significant outcome of the study was the development of a potential biosignature in urine to assist in and simplify the diagnostic process in future. The scientific contributions of this study resulted in five publications: two articles were published and one was submitted for publication in the Journal of Inherited Metabolic Disease, one article was published in Metabolomics and one was published in the South African Paediatric Review.