Mitochondrial bioinformatics analysis
Bioinformatics analysis of the mitochondrial proteome and the mutational spectrum of the Opa1 protein
The mitochondrial proteome is composed of a wide variety of proteins inherited in part from the prokaryotic ancestors of mitochondria and mainly encoded by the nuclear genome, as well as a smaller number of proteins encoded by the mitochondrial genome. So far, about 700 mitochondrial proteins have been characterized in humans by means of proteomics, genomics and bioinformatics. However, the actual number of mitochondrial proteins may be much higher. The characterization of these proteins is essential to the understanding of many common genetic diseases associated with mitochondrial dysfunction.
We compared in silico sequences of human mitochondrial proteins with those of prokaryotes and showed that most of the proteins involved in such diseases are homologous to prokaryotic proteins. We then developed a bioinformatics research strategy to identify new mitochondrial proteins on the basis of their prokaryotic origin and the presence of a characteristic N-terminal extension. All known prokaryotic proteomes were compared to the human genome and various filtering tools were developed to identify new proteins.
In parallel to this overall strategy of screening, we focused on the study of the Opa1 protein, one of the proteins associated with autosomal dominant optic atrophy, which is involved in mitochondrial fusion. Opa1, a dynamin GTPase, is involved in the remodeling of the mitochondrial inner membrane, apoptosis, maintenance of mitochondrial DNA, and energy metabolism. We developed an international database listing the variations of Opa1 so as to characterize its mutational spectrum(http://lbbma.univ-angers.fr/eOPA1). This tool was used as a complement to a multicentric clinical study involving nearly a thousand patients with optic neuropathy. Our work has led to the development of novel bioinformatics tools that should contribute to a better understanding of mitochondrial pathophysiology.
- Tourmen Y., Ferre M., Malthiery Y., Dessen P., and Reynier P., Mitochondrial diseases preferentially involve proteins with prokaryote homologues.C R Biol, 2004.327(12): p. 1095-101.
- Ferre M., Amati-Bonneau P., Tourmen Y., Malthiery Y., and Reynier P., eOPA1: an online database for OPA1 mutations. Hum Mutat, 2005. 25(5): p. 423-8.
- Ferre M., Bonneau D., Milea D., Chevrollier A., Verny C., Dollfus H., Ayuso C., Defoort S., Vignal C., Zanlonghi X., Charlin J.F., Kaplan J., Odent S., Hamel C.P., Procaccio V., Reynier P., and Amati-Bonneau P., Molecular screening of 980 cases of suspected hereditary optic neuropathy with a report on 77 novel OPA1 mutations. Hum Mutat, 2009. 30(7): p. E692-705.
Fig PH (a) Seventy-three nuclear-encoded human mitochondrial proteins of the oxidative phosphorylation according to their eukaryotic (PH+) or prokaryotic (PH-) origin. The 13 proteins of prokaryote origin encoded by mitochondrial DNA (mtDNA) are also shown. (b) Origin of respiratory chain and Krebs cyclenuclear genes associated with human mitochondrial diseases.
Fig eOPA1. Data obtained from the eOPA1 locus-specific database (Ferre ́et al., Hum Mutat 2005), updated July 2010. Distribution of the 213 pathogenic mutations in the OPA1 gene according to: (a) the exons involved, including their intronic neighbourhood; and (b) their consequence at the protein level. OPA1 mutations involved in particular clinical presentations (ADOAD and “ADOA plus”) are shown in red, including 2 compound heterozyguous mutations shown in purple. GED: GTPase effector domain. The frequency of the OPA1 mutations in the patients is symbolized by stars, the number of stars being correlated to the mutation frequency in each exon.