Introduction: Leigh Syndrome French Canadian (LSFC) is a fatal neurodegenerative disorder with onset in infancy. It is a founder-effect disease in the Saguenay-Lac St Jean region of Quebec, and the majority of identified cases are homozygous for a missense mutation in the LRPPRC gene on chromosome 2p21. LRPPRC gene encodes a leucine-rich protein of the pentatricopeptide repeat family (PPR), and it primarily localizes to the mitochondrial compartment. Land plants have hundreds of PPR proteins, but only seven have been identified in humans. All PPR family members that have been investigated mediate some aspect of RNA metabolism. LSFC is characterized biochemically by a severe decrease in cytochrome c oxidase (COX) activity in brain and liver, and a moderate decrease in heart, kidney, skeletal muscle, and skin fibroblasts. A complex I assembly defect has also been reported in the skeletal muscles of LSFC patients. The molecular basis for this tissue specificity remains unknown. The loss of function mutation destabilizes LRPPRC, resulting in a large decrease in the steady-state levels of most mitochondrial mRNAs; however, the exact mechanism by which LRPPRC stabilizes mitochondrial mRNAs remains unknown. Our goal is to investigate the function of LRPPRC using animal models of LSFC and human cell lines and tissues.

Methods and results: BN-PAGE analysis of the conditional LRPPRC knockout livers showed a severe decrease in the steady-state level of LRPPRC protein and a marked reduction in the assembly of complexes IV and V after 5- and 10-weeks of age. The assembly defect correlated with a marked reduction in the steady-state levels of the corresponding mRNAs. To test whether this might be due to changes in poly-A tail length, we measured this in the corresponding mRNAs in LRPPRC knock out liver and demonstrated decreased abundance of polyadenylated mRNAs and an increase in the abundance of oligo- or non-adenylated mRNAs. This finding prompted us to analyze the polyadenylation status of COX1 and COX2 mRNAs in tissues from an LSFC patient in addition to ND3 mRNA, an mRNA whose abundance is not altered by LRPPRC. The population of oligo-adenylated COX1 and COX2 mRNAs was increased in all LSFC tissues examined (liver, heart, and skeletal muscle); however, the level of polyadenylated ND3 mRNA was not altered. Unexpectedly, we observed marked differences in poly-A tail length amongst control tissues, suggesting that this may be a factor in the tissue- specific expression of the biochemical defects in LSFC. The translation of only a subset of mtDNA-encoded mRNAs is altered by the loss of LRPPRC. To investigate whether this might involve some compensation in the translation machinery, we investigated mitochondrial ribosome assembly by sucrose density gradient analysis. This showed increased level of large and small ribosomal subunits in the LRPPRC knock out mice.

Conclusion: The loss of LRPPRC protein in the liver-specific knock out mice results in complex IV and V assembly defects and increased levels of large and small ribosomal subunits. LRPPRC appears to control the stability of some but not all mitochondrial mRNAs perhaps by protecting it from degradation by 3’ nucleases. The mechanism by which LRPPRC protects the 3’ termini of a subset of mitochondrial mRNAs remains unknown; however, there is correlation between the loss of LRPPRC and reduction in the levels of those mRNAs. In addition, the significance of the tissue-specific differences in poly-A tail length will require further study.