Maturation of Wild-Type and Mutated Frataxin by the Mitochondrial Processing Peptidase

Hana Koutnikova 0 1 Victoria Campuzano 0 1 Michel Koenig 0 1 0 1 rue Laurent Fries, BP 163, 67404 Illkirch Cedex-Strasbourg, France 1 Institut de Gntique et de Biologie Molculaire et Cellulaire (IGBMC) , CNRS, INSERM, Universit Louis Pasteur Frataxin is a mitochondrial protein deficient in Friedreich ataxia (FRDA) and which is associated with abnormal intramitochondrial iron handling. We identified the mitochondrial processing peptidase b (MPPb ) as a frataxin protein partner using the yeast two-hybrid assay. In in vitro assays, MPPb binds frataxin which is cleaved by the reconstituted MPP heterodimer. MPP cleavage of frataxin results in an intermediate form (amino acids 41-210) that is processed further to the mature form. In vitro and in vivo experiments suggest that two C-terminal missense mutations found in FRDA patients modulate interaction with MPPb , resulting in a slower maturation process at the normal cleavage site. The slower processing rate of frataxin carrying such missense mutations may therefore contribute to frataxin deficiency, in addition to an impairment of its function. - The gene responsible for Friedreich ataxia (FRDA), an autosomal recessive neurodegenerative disease (1), codes for a novel 18 kDa protein, frataxin (2,3), located at mitochondrial membranes (35). Most patients are homozygous for a large trinucleotide expansion in the first intron of the gene, that causes a severe reduction of the transcript and protein steady-state levels. Four per cent of patients are compound heterozygotes for an expansion mutation and a point mutation. Two missense mutations have been reported, both being located at the C-terminal half of the protein, which might be relevant to the function of the protein (2,6). Frataxin is conserved from yeast to man and has a distant homologue in purple bacteria that share a common phylogeny with the prokaryotic mitochondrial ancestor. The study of yeast frataxin null mutants revealed a growth deficit on a non-fermentable carbon source, mitochondrial DNA instability, decreased activity in cytochrome c oxidase and high sensitivity to hydrogen peroxide, copper and iron (5,7,8). The most pronounced effect of yeast frataxin deficiency is an accumulation of iron in the mitochondria (5,8). In addition, iron deposits (9) and deficiency of ironsulfur enzymes (10) were found in the heart of FRDA patients, the former suggesting that frataxin plays a role in iron handling and the latter suggesting involvement of oxidative stress. We applied a yeast two-hybrid assay (11) to unravel frataxin function in mitochondria, and we identified mitochondrial processing peptidase b (MPPb ), a subunit of heterodimeric peptidase that is involved in proteolytic cleavage of N-terminal mitochondrial targeting sequences (12,13). Frataxin with Cterminal missense mutations found in FRDA patients showed decreased interaction in the yeast two-hybrid assay. Processing of wild-type and mutated frataxin was analysed in vitro with reconstituted MPP and in vivo by COS cell overexpression. The results suggest that the maturation efficiency of the missense mutants is reduced and may contribute to the pathogenicity of these mutations. We searched for protein partners of frataxin by yeast two-hybrid screening. An expression vector that encodes mouse frataxin fused to the DNA-binding domain of the transcription factor LexA was used as a bait in a two-hybrid screen of an embryonic (E9.5E12.5) cDNA library. Frataxin is expressed during mouse embryogenesis and, therefore, its putative partners are expected to be represented in such a library. From ~ 1.5 106 transformants, 11 independent positive clones were obtained as determined by activation of his and lacZ reporter genes. Seven clones code for known nonmitochondrial proteins and three code for proteins that are unlikely to be mitochondrial based on sequence similarity. A single mitochondrial protein, MPPb , was identified. The specificity of interaction between MPPb and frataxin was verified by retransformation into yeast cells. The MPPb two-hybrid clone encodes a protein homologous to rat MPPb from amino acids 40 to 489, and would therefore contain six additional N-terminal amino acids compared with the mature MPPb protein (14). In order to test whether the interaction between frataxin and MPPb is part of the recognition process that removes its mitochondrial targeting peptide, we constructed a series of frataxin deletion and point mutants and tested them using the yeast two-hybrid assay. The C-terminal domain of frataxin used as a bait did not activate the his and lacZ reporter genes, while a strong activation was detected when the N-terminal domain was used (Fig. 1). The C-terminal missense mutations found in FRDA patients, corresponding to the G127V and I151F changes on the mouse sequence, surprisingly reduced the activation of the reporter genes. We observed a 50% decrease in the activation of the lacZ reporter gene with the corresponding mouse I151F mutant and a 90% decrease with the G127V mutant, assuming an identical expression of the wild-type and point mutation constructs. Indeed, we found that the level of expression of the wild-type and point mutation frataxinLexA fusions are comparable by western blot analysis (data not shown). MPPb from mammals is structurally related to, but functionally distinct from, the core I protein of the respiratory chain complex III (15). We tested, therefore, whether the interaction of frataxin is specific for MPPb or also applies to the complex III core I protein. We found only a very weak interaction between frataxin and core I protein in vitro, by GST pull-down assay. In addition, no interaction was detected in vivo using the yeast two-hybrid assay and mature core I protein (amino acids 35472) as a prey (data not shown). Frataxin binds MPPb in vitro A GST pull-down assay was used to confirm mutual interaction of frataxin and MPPb in vitro. MPPb was cloned in the prokaryotic expression vector pGEX4T3 and expressed in Escherichia coli. Purified MPPb GST fusion protein was shown to bind ~ 10% of the in vitro translated frataxin while no binding was observed with a GST protein alone (Fig. 2). The binding of mouse I151F mutant to MPPb was reduced to ~ 5% of frataxin input, in agreement with the yeast two-hybrid results (Fig. 2). A similar study could not be performed with the G127V mutant since the 1G2 epitope encompasses the mutation (3). In vitro and in vivo processing of frataxin In order to test whether frataxin is indeed processed by MPP, we have reconstructed MPP in vitro by co-expression of the a and b subunits in E.coli. Total bacterial protein extract was assayed for processing activity using mouse wild-type and I151F and G127V mutant frataxins tagged at their C-terminus by five [35S]methionine residues. The b subunit of ATPase was used as a positive control (data not shown), and non-recombinant bacterial protein extract served as a negative control for non-spe (...truncated)