Coevolution predicts direct interactions between mtDNA-encoded and nDNA-encoded subunits of oxidative phosphorylation...

Area of Science:

• Molecular Biology
• Biochemistry
• Genetics

Background:

• The structure and subunit interactions of NADH-ubiquinone oxidoreductase (complex I), the largest mammalian oxidative phosphorylation (OXPHOS) complex, remain incompletely understood.
• Complex I comprises subunits encoded by both mitochondrial DNA (mtDNA) and nuclear DNA (nDNA), necessitating cytonuclear coevolution.
• Previous work identified nDNA-encoded subunits with accelerated amino acid replacement, suggesting adaptation to mtDNA's faster evolution and potential interaction sites.

Purpose of the Study:

• To investigate and map physical cytonuclear interactions within human complex I.
• To validate a bioinformatic approach for predicting subunit interactions based on coevolutionary signals.
• To provide evidence for direct interactions between nDNA- and mtDNA-encoded subunits in complex I.

Main Methods:

• Bioinformatic analysis of coevolutionary signals in 10 protein complexes with solved structures to establish prediction accuracy.
• Application of the coevolutionary analysis to predict interactions between specific nDNA- and mtDNA-encoded subunits of complex I.
• Experimental validation of predicted interactions using the yeast two-hybrid system in human cells.

Main Results:

• The coevolutionary analysis accurately predicted physically interacting subunits in known complexes with nearly 90% accuracy.
• Predicted interactions include NDUFC2 with ND5/ND4 and NDUFA1 with ND5/ND4/ND1, and among mtDNA-encoded subunits.
• Experimental validation confirmed interactions between NDUFC2-ND4 and NDUFA1-ND1/ND4, supporting the predictive model.

Conclusions:

• This study provides the first evidence of direct physical interactions between nDNA- and mtDNA-encoded subunits in human OXPHOS complex I.
• The findings elucidate the intricate cytonuclear interaction network within complex I, crucial for cellular energy production.
• The developed method, ComplexCorr, offers a valuable tool for predicting subunit interactions in multi-subunit complexes, especially those lacking structural data.