Skeletal muscle gene expression profiling in mitochondrial disorders

Abstract

Extremely variable clinic and genetic features characterize mitochondrial encephalomyopathy (MEM). Pathogenic mitochondrial DNA (mtDNA) defects can be divided into large-scale rearrangements and single point mutations. Clinical manifestations become evident when a threshold percentage of the total mtDNA is mutated. In some MEM, the "mutant load" in an affected tissue is directly related to the severity of the phenotype. However, the clinical phenotype is not simply a direct consequence of the relative abundance of mutated mtDNA. Other factors, such as nuclear background, can contribute to the disease process, resulting in a wide range of phenotypes caused by the same mutation. Using Affymetrix oligonucleotide cDNA microarrays (HG-U133A), we studied the gene expression profile of muscle tissue biopsies obtained from 12 MEM patients [4 common 4977 bp deleted mtDNA and 8 A3243G: 4 progressive external ophthalmoplegia (PEO) and 4 mitochondrial myopathy, encephalopathy, lactic cidosis, and stroke-like episodes syndrome (MELAS) phenotypes] compared with age-matched healthy individuals. We found several differentially expressed genes: 35 were markedly up-regulated in the mtDNA macro-deletion group (vs. the control group) and 4 decreased; 56 genes were dysregulated in A3243G-related disorders (53 down-regulated in PEO and 3 up-regulated in MELAS). Finally, 12 genes were similarly regulated in the majority of the MEM patients under study. Amongst these, we identified an increased expression of genes related to the metabolism of the amino groups, as well as of several genes involved in genetic information processing. Moreover, few genes were similarly decreased in MEM patients vs. the control group. Real-time PCR demonstrated excellent reproducibility of the microarray-based findings. The observed expression changes are likely to represent a molecular signature for mitochondrial disorders. Furthermore, the differential expression profile of MELAS(A3243G) vs. PEO(A3243G) may support a role of nuclear background in contributing to these different clinical phenotypes. MEM microarray data are available from GEO database (http://www.ncbi.nlm.nih.gov/geo/) with the accession number: GSE1462.