| Human Molecular Genetics | Pages |
©1999 Oxford University Press |
Maturation of frataxin within mammalian and yeast mitochondria: one-step processing by matrix processing peptidase
Introduction
Results
FRDA import into mammalian and yeast mitochondria
Import of Yfh1p and FRDA into wild-type and ssq1-1 mutant mitochondria
Effect of I->F mutation on import of FRDA and Yfh1p
Processing of FRDA by bacterially expressed MPP
Maturation of FRDA occurs in one step
Discussion
Materials And Methods
Constructs
In vitro transcription, translation and import
In vitro processing by MPP
Acknowledgements
References
Maturation of frataxin within mammalian and yeast mitochondria: one-step processing by matrix processing peptidase
Received July 1, 1999; Revised and Accepted August 17, 1999
Friedreich's ataxia is a neurodegenerative disease caused by mutations in the nuclear gene encoding frataxin (FRDA). FRDA is synthesized with an N-terminal signal sequence, which is removed after import into mitochondria. We have shown that FRDA was imported efficiently into isolated mammalian or yeast mitochondria. In both cases, the processing cleavage that removed the N-terminal signal sequence occurred in a single step on import, generating mature products of identical mobility. The processing cleavage could be reconstituted by incubating the FRDA preprotein with rat or yeast matrix processing peptidase (MPP) expressed in Escherichia coli. We used these assays to evaluate the import and processing of an altered form of FRDA containing the disease-causing I154F mutation. No effects on import or maturation of this mutated FRDA were observed. Likewise, no effects were observed on import and maturation of the yeast frataxin homolog (Yfh1p) carrying a homologous I130F mutation. These results argue against the possibility that the I154F mutation interferes with FRDA function via effects on maturation. Other mutations can be screened for effects on FRDA biogenesis as described here, by evaluating import into isolated mitochondria and by testing maturation with purified MPP.
INTRODUCTION
Friedreich's ataxia (FA) is a neurodegenerative disease with an autosomal recessive pattern of inheritance. It is the most common inherited ataxia with a prevalence of 1 in 50 000 individuals (reviewed in ref. 1). The progressive neuro- degeneration in FA primarily affects dorsal root ganglia in the central nervous system. The neurological symptoms include gait and limb ataxia, lower limb areflexia, loss of proprioception and dysarthria. Patients also develop skeletal abnormalities, hypertrophic cardiomyopathy and, often, impaired glucose tolerance (2). These symptoms progress with age, causing most patients to die before they reach the age of 30 years. The disease results from inherited defects in the gene that encodes a protein designated frataxin (FRDA). The most common molecular cause of FA is the hyperexpansion of a polymorphic GAA trinucleotide repeat (<39 repeat units in normal individual versus 66-1700 in FA patients) located in the first intron of the FRDA gene, resulting in a reduced level of FRDA mRNA (3,4). Several FA patients carry one allele with a hyperexpansion of the GAA repeat element and one allele with a missense mutation (4-6). The most common disease-causing missense mutation in FRDA is I154F. Another missense mutation in FRDA, G130V, in combination with a hyperexpanded allele, is associated with a milder and more slowly progressing disease course.
The FA gene was identified by positional cloning (4). The encoded protein, FRDA, was localized to mitochondria by immunofluorescence and immunoelectron microscopy (3). Major clues about the cellular function of the protein came from work in the model organism Saccharomyces cerevisiae (7-12). Yeast possesses a homologous protein, Yfh1p (yeast frataxin homolog), which like its human counterpart is localized to yeast mitochondria (7-9). The deletion of the corresponding gene from the haploid organism results in respiratory deficiency because of the loss of mitochondrial DNA (8). Furthermore, [Delta]yfh1 yeast mutants accumulate iron within mitochondria (7,11). The yeast mutant phenotype is mirrored by the cells of affected tissues from patients with FA. Endomyocardial biopsies of FA patients show diminished activity of mitochondrial respiratory complexes (12) and high iron accumulation is found in heart tissue from patients with FA (13). Finally, FRDA and Yfh1p have highly homologous sequences at their C-termini and the human protein expressed in the yeast mutant is able to complement some of the mutant phenotypes if the first 39 amino acids of the FRDA precursor are replaced by the first 34 amino acids of Yfh1p. The I154F disease-associated mutation in the mature FRDA protein reduces its ability to complement the yeast mutant (8).
Examination of the amino acid sequences of FRDA and its homologs does not immediately shed light on how the protein controls the build up of iron in organelles. We have previously proposed a link between Yfh1p maturation and its function in mitochondrial iron homeostasis (9). We demonstrated that on import into mitochondria the maturation of Yfh1p takes place in two steps: an initial proteolytic cleavage removing ~2 kDa and a subsequent cleavage removing an additional ~4 kDa from the N-terminus of the preprotein. The second step of processing is delayed in a yeast mutant lacking the mitochondrial chaperone Ssq1p. Furthermore, the phenotypes of ssq1-1 and [Delta]yfh1 yeast mutants are very similar (7-9,11,14-16). The most striking similarity is that both mutants exhibit an increase in mitochondrial iron content. Based on these results, the two-step processing of Yfh1p seems important in mitochondrial iron homeostasis in yeast (9).
The possibility that FRDA exerts its effects via interactions with other mitochondrial proteins has been examined. Using a yeast two-hybrid assay, Koutnikova et al. (17) have identified the mitochondrial matrix processing peptidase (MPP) as an FRDA partner protein. These studies apparently suggest that the maturation of FRDA takes place in two steps: MPP cleavage of FRDA first results in an intermediate form that is processed further to the mature form. Furthermore, the disease-causing I154F mutation appears to slow the processing of FRDA by MPP. The slower processing rate of the mutated protein is thought to contribute to a functional FRDA deficiency in FA patients (17). However, these processing experiments were done in vitro with bacterially expressed rat MPP and in vivo by overexpression of frataxin in COS cells followed by immunoblotting. A more definitive conclusion requires that import and maturation experiments be carried out using purified mitochondria. Such studies for FRDA are lacking.
Here we describe the import and maturation of wild-type and I154F mutant FRDA using purified mammalian and yeast (wild-type and ssq1-1 mutant) mitochondria. Surprisingly, in all cases imported FRDA was processed to the mature form in one step by MPP. No intermediate form of FRDA was detected during its maturation within the organelle. In addition, we found that the I154F mutation had no detectable effect on import and/or maturation of FRDA. These results demonstrate that, although the C-terminal domains of FRDA and Yfh1p are functionally homologous, the processing of their N-terminal signal sequences within mitochondria is quite different. These studies provide the basis for understanding the role of preprotein processing in the functions of FRDA and Yfh1p.
RESULTS
FRDA import into mammalian and yeast mitochondria
FRDA in humans and Yfh1p in yeast exhibit sequence similarity and are functional homologs (7-12,16). The machinery for preprotein import into the mitochondrial matrix and subsequent maturation of imported molecules is also conserved between yeast and humans (18-21). We therefore compared the import of reticulocyte lysate-synthesized FRDA precursor into isolated mammalian (rat) and yeast mitochondria. Import into mitochondria was assayed by three criteria: (i) cleavage of the signal sequence; (ii) protection of the signal-less mature polypeptide from digestion by an exogenous protease; and (iii) sedimentation of the imported and protected molecules with mitochondria on centrifugation (22).
The mobility of FRDA preprotein on an SDS gel (Fig. 1, ~30 kDa) deviated considerably from the size of 23.1 kDa estimated from its amino acid sequence. FRDA was imported with comparable efficiency in rat or yeast mitochondria (Fig. 1, compare m in lanes 2 and 6). In both cases, FRDA preprotein was processed to a mature form (m) of ~18 kDa which remained protected from externally added trypsin (Fig. 1, lanes 4 and 8). The size of this mature form was consistent with the reported size of mature FRDA in various human and mouse tissues as observed on immunoblots (3). Import was completely inhibited by valinomycin (Fig. 1, lanes 3 and 5 for rat, lanes 7 and 9 for yeast) which dissipates the membrane potential across the inner membrane. These observations suggest that FRDA follows the same import pathway into yeast mitochondria as it does into mammalian mitochondria.
Figure 1. FRDA import into isolated rat or yeast mitochondria. FRDA precursor was synthesized in reticulocyte lysate and import reactions were performed at 30°C for 30 min in the absence (-) or presence (+) of valinomycin. Samples were treated with trypsin where indicated and analyzed by SDS-PAGE followed by autoradiography. RM, rat mitochondria; YM, yeast mitochondria; Val, valinomycin. p and m denote the precursor and the mature forms of FRDA, respectively.
Import of Yfh1p and FRDA into wild-type and ssq1-1 mutant mitochondria
Yfh1p preprotein is imported into mitochondria and then processed in two steps, removing first ~2 kDa and then an additional ~4 kDa from the N-terminus (9). FRDA imported into yeast or mammalian mitochondria appeared to be processed in only one step (Fig. 1). It remained a possibility that conversion of the intermediate to the mature form of FRDA occurred so rapidly at 30°C that it was complete during the 30 min incubation. We therefore tested FRDA import into wild-type yeast mitochondria at a lower temperature (20°C) and for a shorter time period (5 min). As an additional control, we evaluated import of Rip1p; this precursor is processed in two steps and the processing has been extensively characterized (23-26). As can be seen in Figure 2A, FRDA processing to the mature form occurred rapidly in one step (lanes 5 and 6), even under conditions in which processing of Yfh1p (lanes 1 and 2) and of Rip1p (lanes 3 and 4) was incomplete and the corresponding intermediate forms of the latter two proteins were easily detected.
Figure 2. Import of FRDA and Yfh1p into wild-type and ssq1-1 mutant mitochondria. (A) Preproteins (FRDA, Yfh1p and Rip1p) were synthesized in reticulocyte lysate. Import was carried out using wild-type mitochondria at 20°C for 5 min. Untreated and trypsin-treated samples were analyzed by SDS-PAGE followed by autoradiography. p, i and m denote the precursor, the intermediate and the mature forms, respectively. The bands marked with an asterisk (*) indicate presumed N-terminally truncated forms of preproteins generated during in vitro translation. (B) Import of reticulocyte lysate-synthesized Yfh1p was carried out using wild-type (WT) or ssq1-1 (M) mitochondria at 20°C for 20 min. Untreated and trypsin-treated samples were analyzed by SDS-PAGE followed by autoradiography. p, i and m denote the precursor, the intermediate and the mature forms of Yfh1p, respectively. An asterisk (*) indicates a presumed N-terminally truncated form of Yfh1p generated during in vitro translation. (C) Import of reticulocyte lysate-synthesized FRDA into ssq1-1 mitochondria (M) was assayed at 20°C for either 5 or 20 min. Samples were analyzed either directly or after treatment with trypsin. p and m denote the precursor and the mature forms of FRDA, respectively.
Mutants of yeast mt-Hsp70 (ssq1-1) exhibit altered processing of Yfh1p (9). The characteristic alteration is a delay in the second cleavage of the two-step maturation process. We wondered whether examining FRDA import into ssq1-1 yeast mutant mitochondria would allow detection of processing intermediates that could not be detected in the wild-type mitochondria. Of note is that the human homolog of Ssq1p has not been identified, although evidence exists for evolutionary conservation of the Hsp70s and members of the various branches of this protein family (27). The data presented in Figure 2B show the import characteristics of Yfh1p into wild-type and ssq1-1 mitochondria. The initial processing cleavage removed ~2 kDa from the N-terminus of the preprotein and generated an intermediate size polypeptide (i) migrating at ~27 kDa. A subsequent cleavage removed ~4 kDa from the N-terminus of the intermediate form, generating a mature product (m) of ~23 kDa. Both in the wild-type and the ssq1-1 mutant, the import of Yfh1p was efficient, as judged by the appearance of the protease-resistant imported molecules (i + m). However, in contrast to the wild-type mitochondria, conversion of the intermediate (i) to the mature form (m) was significantly delayed in the mutant (Fig. 2B, compare lanes 2 and 3 with lanes 4 and 5, respectively). An analogous import experiment was carried out with FRDA for two different time periods (Fig. 2C). Import was time dependent and efficient, but no intermediate form of FRDA was detected during its maturation.
Effect of I->F mutation on import of FRDA and Yfh1p
The disease-causing I154F mutation in FRDA has been suggested to slow the processing of FRDA by MPP (17). Therefore, the kinetics of import and maturation of wild-type FRDA were compared with those of mutant FRDA (Fig. 3A). Import reactions were carried out with ssq1-1 mitochondria in an effort to uncover subtle processing defects. Import and maturation were time dependent and efficient for both wild-type and I154F mutant FRDA and no significant difference between the wild-type and mutant FRDA was detected. These studies were extended by examining the effect of an analogous I130F mutation on Yfh1p import (Fig. 3B). Import kinetics and the maturation pattern remained unaffected by the I130F mutation in Yfh1p. These results suggest that the functional defects in FRDA and Yfh1p resulting from the respective I->F mutations do not result from effects on import into mitochondria or subsequent maturation by processing peptidase(s) within the organelle.
Figure 3. Import kinetics of wild-type and mutant FRDA and Yfh1p. Wild-type and mutant preproteins were synthesized in reticulocyte lysate. Import was carried out using ssq1-1 mitochondria at 20°C for different time periods. Untreated and trypsin-treated samples were analyzed by SDS-PAGE followed by autoradiography. (A) (Left) FRDA; (right) FRDA(I154F). (B) (Left) Yfh1p; (right) Yfh1p(I130F). The bands marked with an asterisk (*) indicate presumed internal initiation products generated during in vitro translation.
Processing of FRDA by bacterially expressed MPP
To evaluate whether maturation of FRDA to the corresponding mature form was mediated by MPP alone, we examined the cleavage pattern generated by bacterially expressed MPP in the absence of mitochondria (Fig. 4A). The enzyme consists of two subunits ([alpha] and [beta]) and both are required for activity. The plasmid pVG18 allows the synthesis of a single mRNA encoding both subunits of yeast MPP (28,29). Both subunits were expressed in bacteria and the [alpha]/[beta]-His6 MPP complex was purified to homogeneity by Ni-NTA chromatography (not shown). Likewise, the plasmid pGEMabMlu allows bacterial expression of both the [alpha] and [beta] subunits of rat MPP (30). When FRDA synthesized in reticulocyte lysate was incubated with rat (Fig. 4A, lane 3) or yeast (lane 4) MPP, an additional band was generated in each case at the expense of the preprotein. The cleavage products of yeast MPP and rat MPP were identical in size. The cleavage was specific because: (i) no product was detected after incubation with control bacterial extract (lane 2); and (ii) activity was abolished by addition of EDTA (lane 5) or o-phenanthroline (data not shown), inhibitors of the MPP metalloprotease.
Figure 4. Maturation of wild-type and mutant FRDA by purified MPP. (A) FRDA precursor was synthesized in reticulocyte lysate and treated with bacterially expressed rat or yeast MPP in the absence or presence of 5 mM EDTA. An equivalent amount of bacterial extract containing no MPP served as a negative control. In parallel, import reactions were performed at 20°C for 30 min, followed by trypsin treatment where indicated. Samples were analyzed by SDS-PAGE and autoradiography. rMPP, bacterial extract containing rat MPP; yMPP, purified yeast MPP. p and m denote the precursor and the mature forms of FRDA, respectively. (B) Preproteins were synthesized in reticulocyte lysate and incubated with purified yeast MPP at 20°C for different time periods. Samples were analyzed by SDS-PAGE followed by autoradiography. p and m denote the precursor and the mature forms, respectively.
The MPP-cleaved products (Fig. 4A, lanes 3 and 4) were then compared with the signal-cleaved product of FRDA obtained after in vitro import into yeast mitochondria (Fig. 4A, lanes 6 and 7). The products generated by yeast or rat MPP were identical in size to the mature form of FRDA generated during its maturation within mitochondria. These results strongly suggest that the maturation of FRDA is mediated by MPP alone. We also examined the kinetics of maturation of FRDA and FRDA(I154F) by purified yeast MPP (Fig. 4B). No significant difference in the maturation process was observed. In both cases the mature form could be detected within 1 min of incubation and ~60% of precursor molecules were cleaved within the next 7 min. These results further substantiate that the I154F mutation does not interfere with maturation of FRDA.
Maturation of FRDA occurs in one step
If maturation of FRDA by MPP occurs in one step, it may be possible to detect the cleaved signal peptide as a single fragment. MPP cleavage of many mitochondrial preproteins involves an arginine at position -2 and, frequently, another at position -3 with respect to the cleavage site (18). Based on this motif, MPP cleavage of the FRDA precursor is likely to occur between residues 41 and 42, releasing a fragment of ~4.5 kDa. In order to look for the cleaved signal peptide, we took advantage of a modified Tricine-SDS gel system that allows separation and identification of small peptides (31). When FRDA precursor was incubated with purified MPP, some of the precursor was cleaved to generate the mature form and the signal peptide (Fig. 5, lanes 1 and 2). The band indicated by an asterisk likely represents an N-terminally truncated form of FRDA lacking the signal sequence, since it was not imported into mitochondria (data not shown). Therefore, as expected, this band remained unaffected by MPP.
Figure 5. Identification of the cleaved FRDA signal peptide. FRDA was synthesized in reticulocyte lysate. The preprotein was treated with bacterially expressed and purified yeast MPP. Samples were analyzed by 15% Tricine-SDS-PAGE followed by autoradiography. The band marked with an asterisk (*) indicates a presumed N-terminally truncated form of FRDA generated during in vitro translation. p, FRDA precursor; m, mature FRDA; sp, FRDA signal peptide. The arrow indicates the migration position of the Nua1p signal peptide with a molecular mass of 3.8 kDa. Note that FRDA migrates with an apparent molecular mass of ~23 kDa on Tricine-SDS-PAGE as opposed to ~30 kDa on usual Tris/glycine-SDS-PAGE (Figs 1-4). This migration discrepancy is likely due to different gel buffers and/or different SDS concentrations in the loading buffers, in the gels and also in the running buffers.
The molecular mass of the cleaved FRDA signal peptide was calculated to be ~4.2 kDa, based on the migration pattern of protein standards. As an additional control for the size determination, another mitochondrial preprotein (Nua1p; Saccharomyces Genome Database) with a predicted cleavable signal sequence of 34 amino acids was subjected to cleavage by MPP under identical conditions (data not shown). As expected, the resulting cleaved signal peptide of Nua1p (Fig. 5, indicated by an arrow; 3.8 kDa) migrated slightly faster than the FRDA signal peptide. Taken together, these results strongly suggest that FRDA preprotein is processed in one step by MPP, generating the mature form.
DISCUSSION
FA is an inherited neurodegenerative disease (1). The molecular basis for this disease is a defect in the nuclear gene that encodes the mitochondrial protein FRDA. FRDA is synthesized on cytoplasmic ribosomes as a preprotein with an N-terminal mitochondrial signal sequence. Here we have studied the biogenesis of FRDA. We have demonstrated that the precursor form of FRDA could be imported efficiently into mammalian and yeast mitochondria. This result is consistent with the fact that elements of the yeast mitochondrial protein import machinery have homologs in humans (19,20). Taken together, our results substantiate that the mechanism of mitochondrial protein import is very similar for yeast and humans. Observations regarding FRDA import into yeast mitochondria may therefore be relevant to the processes that occur in humans.
Maturation of FRDA in yeast and mammalian mitochondria was also indistinguishable. On import into mammalian or yeast mitochondria, the N-terminal signal sequence of FRDA with a molecular mass of ~4.2 kDa was efficiently cleaved in one step. The protease responsible for this N-terminal processing was MPP. This heterodimeric metallopeptidase of the mitochondrial matrix is highly conserved between yeast and mammals (18). Furthermore, we have shown that correct processing of the FRDA signal sequence was achieved in the absence of mitochondria, using bacterially expressed and purified MPP from yeast or rat. Consistent with the MPP cleavage site motif, the FRDA precursor was likely cleaved between amino acids 41 and 42. Careful kinetic studies of the import and processing of FRDA revealed only the precursor and mature forms. Our results are in apparent conflict with a published report in which maturation of FRDA was suggested to take place in two steps (17). However, these studies were performed using FRDA overexpressed in COS cells and evaluated by immunoblotting. Some of the overexpressed FRDA was non-specifically degraded and this might have complicated identification of the mature FRDA.
The C-terminal domains of FRDA and Yfh1p are highly conserved and functionally homologous (8). We have demonstrated that the disease-causing I->F mutation in this domain did not interfere with import and maturation of FRDA or Yfh1p. Thus, it is unlikely that the I->F mutation hinders FRDA (or Yfh1p) function via effects on the protein's biogenesis. Using the assays described herein, however, other FA disease-causing mutations can be evaluated for subtle effects on mitochondrial import and MPP processing. Mutations in the N-terminus might block targeting functions of the signal peptide, and mutations in other domains might interfere with folding required to become an efficient substrate for processing.
How, then, does the I154F mutation affect function? In the simplest scenario, the mutant form of the protein might fold incorrectly, leading to rapid aggregation or degradation. It is interesting that Yfh1p has been shown to mediate iron efflux from mitochondria (16). It would be worthwhile, therefore, to determine whether the I154F mutation (I130F in yeast) interferes with this activity. Furthermore, the identification of proteins that interact with FRDA (or Yfh1p) may provide clues to their function. For example, Yfh1p does not contain any predicted transmembrane domain, hence it is unlikely to act as an iron pump. It is possible that Yfh1p mediates iron efflux through interactions with Atm1p, an ABC transporter located in the mitochondrial inner membrane (32). Likewise, Yfh1p may participate in a cascade of reactions involving Ssq1p (9). The phenotypes of [Delta]yfh1 (7,11), [Delta]atm1 (33) and ssq1-1 (9) are very similar, in particular with respect to their high levels of accumulated mitochondrial iron. It would therefore be interesting to see whether Yfh1p physically interacts with Atm1p and/or Ssq1p and, if so, whether the I->F mutation abolishes this interaction.
Most proteins imported into the matrix contain a signal sequence that is removed by MPP. However, a subset of preproteins undergoes a second processing step. This second processing step is mediated by mitochondrial intermediate peptidase (MIP; encoded by OCT1) in the matrix or by inner membrane proteases (Imp1p/Imp2p) in the intermembrane space (18). These processing proteases are highly conserved between yeast and mammals, as are many of their substrates. However, there are exceptions. RIP1 encodes the iron-sulfur protein component of the multisubunit respiratory complex III. Rip1p in yeast is synthesized in the cytosol, imported into mitochondria and processed by two successive cleavages, mediated by MPP followed by MIP (23-25). In contrast, processing of the homologous Rieske iron-sulfur protein of bovine heart mitochondria occurs in one step by MPP; MIP is not involved (26). The data demonstrating different processing of FRDA and Yfh1p are reminiscent of observations of Rip1p. Processing of the FRDA preprotein (humans) occurs in one step and is mediated by MPP. Processing of Yfh1p preprotein (yeast) occurs in two steps (9) and both cleavages appear to be mediated entirely by MPP (our unpublished observation). Why this difference in processing of the human frataxin and the yeast Yfh1p should have evolved is unclear.
The differences in processing are likely to be relevant to function. Thus, FRDA was able to complement the yeast [Delta]yfh1 mutant only if the N-terminal 39 amino acids were replaced with 34 amino acids from the yeast precursor (8). The requirement for the yeast N-terminal domain for function in yeast may result from the species-specific two-step processing of Yfh1p. Further studies of in vivo complementation of various yeast mutants together with in vitro import and maturation will help in understanding structure-function relationships of wild-type and mutant forms of FRDA.
MATERIALS AND METHODS
Constructs
(i) pSP64T/YFH1. The YFH1 open reading frame (ORF) was amplified from the plasmid pET21b/YFH1 (9) using primers 5[prime]-AT-ATAAGAATTCCATATGATTAAGCGGTCTCTCGCAAGT-3[prime] (sense) and 5[prime]-TTCTTAAGATCTTTACTCGAGTTGGCTTTTAGAAATGGCCTT-3[prime] (antisense). The resulting product (EcoRI-NdeI-YFH1 ORF-XhoI-stop codon-BglII) was cloned into the EcoRI and BglII sites of pSP64T (34), creating plasmid pSP64T/YFH1. We then used this plasmid for the cloning of all other NdeI-XhoI fragments.
(ii) pSP64T/YFH1 (I130F). The conserved isoleucine at amino acid position 130 of Yfh1p was changed to a phenylalanine using a QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer's protocol. The plasmid pSP64T/YFH1 was used as the template and the method required two oligonucleotides: 5[prime]-CAGCCTCCAAATAAGCAATTTTGGTTGGCATCACCATTG-3[prime] (sense) and 5[prime]-CAATGGTGATGCCAACCAAAATTGCTTATTTGGAGGCTG-3[prime] (antisense).
(iii) pSP64T/FRDA. The plasmid pTL1 containing the entire cDNA coding for FRDA was used as a template (3). The FRDA ORF was amplified by PCR using primers 5[prime]-CGCGGATCCCATATGTGGACTCTCGGGCGCCGCGCA-3[prime] (sense) and 5[prime]-CCG-GAATTCCTCGAGGGTACCAGCATCTTTTCCGGAATAG-GCCAA-3[prime] (antisense). The PCR product was inserted into the NdeI and XhoI sites of the vector pSP64T/YFH1, creating plasmid pSP64T/FRDA.
(iv) pSP64T/FRDA (I154F). The conserved isoleucine at residue 154 of FRDA was changed to phenylalanine as described above using oligonucleotides 5[prime]-CAGACGCCAAACAAGCAATTCTGGCTATCTTCTCCATCC-3[prime] (sense) and 5[prime]-GGATGGAGAAGATAGCCAGAATTGCTTGTTTGGCGTCTG-3[prime] (antisense).
(v) pSP64T/RIP1. The RIP1 ORF was amplified from a yeast genomic library (Promega, Madison, WI) using primers 5[prime]-GC-TCTAGACATATGTTAGGAATAAGATCATCTGTCAAG-3[prime] (sense) and 5[prime]-CGGGATCCCTCGAGGGTACCAACAATGA-CCTTATCACCATCGAA-3[prime] (antisense). The PCR product was inserted into the NdeI and XhoI sites of the vector pSP64T/YFH1, creating plasmid pSP64T/RIP1.
All constructs were confirmed by sequencing.
In vitro transcription, translation and import
The plasmids pSP64T/FRDA, pSP64T/YFH1 and pSP64T/RIP1 were linearized with BamHI and their transcription was carried out using a Ribomax-SP6 kit (Promega). Preproteins were synthesized in reticulocyte lysate (Promega) using the manufacturer's protocol. In vitro import reactions were performed using mitochondria isolated from S.cerevisiae (9,22,35,36) and from rat liver (37). Following import, reaction mixtures were treated with trypsin (0.2-0.4 mg/ml) for 30 min at 0°C. To inactivate trypsin, samples were diluted with 20 mM HEPES-KOH, pH 7.5, 0.6 M sorbitol, 0.1 mg/ml bovine serum albumin, 5 mg/ml soybean trypsin inhibitor, 100 U/ml Trasylol and 1 mM phenylmethylsulfonyl fluoride. Mitochondria were sedimented (15 000 g for 10 min at 4°C) and washed with 10% trichloroacetic acid. Samples were analyzed by SDS-PAGE and autoradiography (9,22,35,36).
In vitro processing by MPP
Escherichia coli BL21(DE3) cells carrying the plasmid pVG18 were used for co-expression of both the [alpha] and [beta] subunits of yeast MPP and the functional enzyme complex was purified as described (28,29). Likewise, BL21(DE3) cells carrying the plasmid pGEMabMlu were used for expression of rat MPP (30). The processing assay was performed essentially as described previously (28-30). Briefly, 35S-labeled preproteins were synthesized in reticulocyte lysate. MPP cleavage was performed in 50 mM Tris-HCl, pH 7.5, containing 0.15 M NaCl and 10% glycerol. A small aliquot of post-ribosomal supernatant (1-4 µl) containing the radiolabeled preprotein was incubated with purified yeast MPP (100 ng) or bacterial extract containing an equivalent amount of rat MPP in a total volume of 40 µl for different time periods at 20°C. Saturated ammonium sulfate solution (80 µl) was added to the reaction mixture and incubated on ice for 30 min. Precipitated proteins were sedimented and analyzed. This procedure was necessary to remove globin from the reaction mixture. To visualize cleaved signal peptides, samples were analyzed by 15% Tricine-SDS-PAGE (31).
ACKNOWLEDGEMENTS
A cDNA encoding frataxin was the generous gift of M. Koenig. We thank V. Géli for the plasmid pVG18, J. Adamec for the plasmid pGEMabMlu, and H.K. Anandatheerthavarada and N.G. Avadhani for rat liver mitochondria. We also thank Debleena Pain for technical help and David Schwartz for his comments on the manuscript. This work was supported by grants from the National Institutes of Health (GM57067) and the American Heart Association (9951300U) to D.P. A.D. is supported by the NIH grant DK53953. D.M.G. is supported by NIH training grant HL07027-24.
REFERENCES
+To whom correspondence should be addressed. Tel: +1 215 573 7305; Fax: +1 215 573 5851; Email: pain{at}mail.med.upenn.edu
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I. Condo, N. Ventura, F. Malisan, A. Rufini, B. Tomassini, and R. Testi In vivo maturation of human frataxin Hum. Mol. Genet., July 1, 2007; 16(13): 1534 - 1540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Gu, D. M. Gordon, B. Amutha, and D. Pain A GTP:AMP Phosphotransferase, Adk2p, in Saccharomyces cerevisiae: ROLE OF THE C TERMINUS IN PROTEIN FOLDING/STABILIZATION, THERMAL TOLERANCE, AND ENZYMATIC ACTIVITY J. Biol. Chem., May 13, 2005; 280(19): 18604 - 18609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Gordon, M. Kogan, S. A.B. Knight, A. Dancis, and D. Pain Distinct roles for two N-terminal cleaved domains in mitochondrial import of the yeast frataxin homolog, Yfh1p Hum. Mol. Genet., February 1, 2001; 10(3): 259 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Cavadini, C. Gellera, P. I. Patel, and G. Isaya Human frataxin maintains mitochondrial iron homeostasis in Saccharomyces cerevisiae Hum. Mol. Genet., October 1, 2000; 9(17): 2523 - 2530. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-J. Cho, M. G. Lee, J. K. Yang, J. Y. Lee, H. K. Song, and S. W. Suh Crystal structure of Escherichia coli CyaY protein reveals a previously unidentified fold for the evolutionarily conserved frataxin family PNAS, July 19, 2000; (2000) 160270897. [Abstract] [Full Text]
|
|
|
|
|
|
H. Puccio Recent advances in the molecular pathogenesis of Friedreich ataxia Hum. Mol. Genet., April 1, 2000; 9(6): 887 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Dhe-Paganon, R. Shigeta, Y.-I. Chi, M. Ristow, and S. E. Shoelson Crystal Structure of Human Frataxin J. Biol. Chem., September 29, 2000; 275(40): 30753 - 30756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Cavadini, J. Adamec, F. Taroni, O. Gakh, and G. Isaya Two-step Processing of Human Frataxin by Mitochondrial Processing Peptidase. PRECURSOR AND INTERMEDIATE FORMS ARE CLEAVED AT DIFFERENT RATES J. Biol. Chem., December 22, 2000; 275(52): 41469 - 41475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Kim, S. Saxena, D. M. Gordon, D. Pain, and A. Dancis J-domain Protein, Jac1p, of Yeast Mitochondria Required for Iron Homeostasis and Activity of Fe-S Cluster Proteins J. Biol. Chem., May 11, 2001; 276(20): 17524 - 17532. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-J. Cho, M. G. Lee, J. K. Yang, J. Y. Lee, H. K. Song, and S. W. Suh Crystal structure of Escherichia coli CyaY protein reveals a previously unidentified fold for the evolutionarily conserved frataxin family PNAS, August 1, 2000; 97(16): 8932 - 8937. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||