INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The vast majority of mitochondrial proteins are directed to their final subcellular location with great specificity. Most proteins targeted to the mitochondrial matrix contain a cleavable amino-terminal presequence with basic and hydroxylated amino acids interspersed throughout their length (14, 15, 28, 29, 40, 56). The ability of these presequences to form an amphiphilic alpha-helical structure is thought to be a prime determinant of mitochondrial targeting (6, 47, 48, 58). Mitochondrial targeting signals facilitate efficient protein translocation across the outer mitochondrial membrane by mediating interactions with cytosolic cofactors and the translocase of the outer membrane (TOM) complex (1, 2, 4, 7, 11, 42). In Saccharomyces cerevisiae, the multisubunit TOM complex is thought to consist of a core comprised of Tom40p, Tom5p, Tom6p, and Tom7p. In addition, at least four outer membrane proteins, Tom20p-Tom22p and Tom70p-Tom37p, make up two receptor systems with partially overlapping specificity for precursor binding (2, 4, 18, 23, 26, 33, 38, 39).

Upon their synthesis on cytosolic polysomes, many precursors must be maintained in an extended conformation to be imported efficiently. It was previously shown that the presequence of the F₁-ATPase  subunit precursor (pre-F₁) acts as an intramolecular chaperone by maintaining the preprotein in an import-competent state (21). The precursor may also be maintained in an import-competent conformation by molecular chaperones such as Hsp70 and Ydj1. A unique molecular chaperone, mitochondrial import stimulating factor (MSF), has been purified from reticulocyte lysate and shown to selectively bind mitochondrial precursors (3, 19). However, a yeast homologue has not been identified. Current models hypothesize that the MSF-precursor complex specifically docks with the Tom70p-Tom37p receptor complex in an ATP-dependent manner. The precursor is then sequentially delivered to Tom20p-Tom22p and the core TOM complex (7, 23, 27). After crossing the outer mitochondrial membrane, the targeting signal is inserted into the inner mitochondrial membrane in a -dependent manner (15, 37, 41). It then interacts with the translocase of the inner membrane complex, and its transport into the matrix is facilitated by an ATP-dependent motor formed by mitochondrial Hsp70p, Mge1p, and Tim44p (35, 41, 46, 53, 55).

Recent studies have sought to delineate the function and specificity of the receptor components in binding mitochondrial precursor proteins. Studies with purified cytosolic domains of Tom20p, Tom22p, and Tom70p indicate that these receptor subunits carry out a division of labor in binding mitochondrial proteins. In vitro experiments showed that the cytosolic domains of Tom20p and Tom22p preferentially bind precursors that contain amino-terminal targeting signals. In contrast, Tom70p interacts most strongly with precursors that have internal targeting signals. However, Tom70p has also been reported to stimulate the import of several precursors with amino-terminal presequences (24, 25). Strains containing knockouts of Tom20p, Tom37p, or Tom70p are viable but display mild growth defects on nonfermentable carbon sources. Furthermore, the overexpression of TOM70 can restore the ability of a tom20 strain to grow on a nonfermentable carbon source. While the simultaneous disruption of TOM70 and TOM20 is lethal, growth can be restored by the overexpression of TOM22 (27). These results suggest that the proteins encoded by these genes have overlapping and cooperative roles in mitochondrial binding and import.

The pleiotropic drug resistance (PDR) pathway controls the expression of a number of genes that mediate resistance to a broad range of structurally and functionally unrelated compounds. Recently, the PDR pathway was also reported to play a role in monitoring the functional status of mitochondria (22). Hallstrom and Moye-Rowley demonstrated that mitochondrial defects arising from the loss of a nuclear-encoded gene involved in electron transport activity or maintenance of the mitochondrial genome resulted in the increased expression of PDR3, a transcription factor that regulates the expression of many genes in the PDR pathway (22). In some cases, this stimulation of Pdr3p expression was mediated by the retrograde signaling pathway, which has been shown to respond to mitochondrial defects through the activation of several nuclear genes (31, 36). In addition, it was previously shown that a hyperactivating allele of PDR3 results in an inability to respire on nonfermentable carbon sources (34). These results suggest a role for Pdr3p in regulating certain aspects of mitochondrial function.

In the present study, we utilized a well-characterized mutant form of pre-F₁ with a minimal targeting signal to further examine the function of mitochondrial protein import receptors (5, 6, 21). Using an in vivo kinetic analysis, we found that the minimal targeting signal greatly increased the dependence of the pre-F₁ precursor on Tom70p for mitochondrial import. Experiments using an in vitro mitochondrial protein import system suggested that this requirement for Tom70p is related to an inability to maintain this mutant precursor in an import-competent conformation. The resulting import defect was so severe that a tom70 strain expressing this mutant precursor exhibited a cold-sensitive phenotype on a nonfermentable carbon source. A multicopy suppressor screen revealed that the overexpression of PDR3 stimulated the import of this defective precursor and restored growth under these conditions. Our results indicate that the import of this precursor is increased by the PDR3-mediated stimulation of TOM72 expression, a little studied TOM70 homologue.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Strains and growth conditions. The S. cerevisiae strains used in this study were SEY6215 (MATa his3-200 leu2-3,112 ura3-52 trp1-901 atp2::LEU2), YDB210 (MAT his3-200 leu2-3,112 ura3-52 trp1-901 tom70::HIS3 atp2::LEU2), YDB403 (MATa his3-200 leu2-3,112 ura3-52 trp1-901 atp2::LEU2 tom72::hisG), and YDB404 (MAT his3-200 leu2-3,112 ura3-52 trp1-901 tom70::HIS3 atp2::LEU2 tom72::hisG). The TOM72 mutant strains were derived from SEY6215 and YDB210 using standard yeast genetic techniques (54). Yeast transformations were performed by the alkali cation technique (30, 50) and selected on SD medium (54) containing required supplements.

Gene disruptions. A gene disruption of TOM70 was generated from pVH19 by deleting a 948-bp BglII fragment that included the AUG initiation codon of TOM70. The resulting 3.6-kb linear plasmid was ligated with a 1.8-kb BamHI fragment containing HIS3. The resulting plasmid was digested with BamHI and EcoRI to generate a 4.2-kb fragment that was transformed into SEY6215 by one-step gene disruption. Transformants were selected on SD plates for the histidine auxotrophic marker, and the correct integration event was verified by Southern blotting. The resulting strain was designated YDB210 for this study.

Construction of plasmids. The yeast shuttle plasmid pDB425 containing WT (WT) pre-F₁ was constructed by cloning the 2.6-kb EcoRI-HindIII fragment containing the ATP2 structural gene and its promoter region into Ycplac22 (16). pDB427 was constructed by replacing the 1.575-kb HindIII fragment of pDB425 with the 1.539-kb HindIII fragment of pDB51 encoding 1,2 pre-F₁. The 2µm plasmid containing TOM72, pDB561, was constructed by cloning the 3.020-kb EcoRI-BamHI fragment from the pGAL-TOM72 plasmid into the EcoRI-BamHI sites of pSEY8.

Expression and purification of recombinant proteins. Plasmids expressing WT pre-F₁-dihydrofolate reductase (DHFR) (pDB421), 1,2 pre-F₁-DHFR (pDB423), WT pre-F₁-DHFR* (pDB507), and 1,2 pre-F₁-DHFR* (pDB509) have been described (21). These precursors were expressed and purified from the Escherichia coli strain BL21(DE3). Purified precursors were stored in urea buffer (8 M urea, 20 mM Tris [pH 7.5], 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride) until use (20).

Mitochondrial protein import assays. Mitochondria were prepared from the WT yeast strain D273-10B and the tom70 yeast strain HR1. Strains were grown on YMM (minimal salts supplemented with 0.1% glucose, 2% DL-lactate, and 0.3% yeast extract). Import assays were carried out as described previously (20, 21). Import reactions were stopped by a 10-fold dilution into ice-cold SEM buffer (0.25 mM sucrose, 1 mM EDTA, 10 mM morpholinepropanesulfonic acid [MOPS] [pH 7.2]) supplemented with 0.5 µg of valinomycin, 20 µg of oligomycin, and 8 µg of antimycin per ml. The samples were treated with 0.1 mg of proteinase K per ml for 30 min on ice followed by the addition of 1 mM phenylmethylsulfonyl fluoride. The samples were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and quantitation was performed using a PhosphorImager system (Molecular Dynamics). In control experiments, protease protection was not observed when 0.5 µg of valinomycin per ml and 20 µg of oligomycin per ml were present in the import reaction or when Triton X-100 was added during protease treatment. This confirmed that the protease-protected precursor proteins had undergone mitochondrial import.

Screening of WT yeast genomic multicopy library. The WT yeast genomic library was constructed by insertion of genomic Sau3a DNA fragments into the Yep24 2µm yeast shuttle vector and was a generous gift from Vytas Bankaitis. The library was transformed into YDB210 expressing pDB427 and transformants were screened on SMD plates selective for the auxotrophic markers. Each transformant was then streaked out on YP lactate plates for single colonies and incubated at 15°C for 10 days. More than 3,000 transformants were screened, 15 of which grew under the repressive conditions. Plasmids were then rescued, propagated in E. coli, and transformed back into YDB210 expressing pDB427. Transformants were grown under repressive conditions in order to confirm that the transforming plasmid conferred suppression of the growth defect on a nonfermentable carbon source, and 10 were confirmed to suppress the growth defect. The plasmids containing the complementing insert were subjected to restriction mapping. The E. coli transformants were screened for the presence of either the pDB427 or library plasmid by HindIII enzyme digests. Those transformants that contained the library plasmid were then subjected to BamHI and HindIII enzyme digestion. Of the 10 library plasmids screened by restriction mapping, 8 contained inserts with similar or identical restriction digests. The inserts were then sequenced by automated flourescent sequencing with primers that anneal to flanking regions of the plasmid insertion site. The sequences were then used to search the Saccharomyces Genome Database (Stanford University) by BLAST search to identify the genes contained within the insert. All eight inserts mapped to chromosome II between nucleotides 209,089 and 227,504. The library containing the smallest complementing insert was subjected to further restriction mapping in order to identify the gene conferring increased growth under the repressive conditions (see Fig. 2). Each subclone was then transformed into YDB210 expressing pDB427 to identify which gene suppressed the growth defect. In order to ultimately define the suppressive region of the insert, an 8.266-kb BamHI-EcoRV fragment was subcloned into pSEY8. A 7.7-kb BglII fragment was then subcloned into pSEY8 resulting in the plasmid termed pDB562 and transformed into YDB210 expressing pDB427.

Cell labeling and immunoprecipitation. The S. cerevisiae strains SEY6215 and YDB210, transformed with pDB425 or pDB427 and pDB562, were grown in synthetic medium (49) containing 2% glucose to a cell density of 0.5 to 0.7 A₆₀₀ units/ml at 30°C. The cells were resuspended in fresh medium to 4 A₆₀₀ units/ml and incubated for 15 min at 30°C with shaking. The labeling reaction was initiated by the addition of 0.2 mCi of [³⁵S]EXPRESS protein labeling mix (DuPont NEN). The labeling reaction was terminated by the addition of either 0.1 mg of cycloheximide per ml and 0.02 mg of methionine per ml or 0.3% yeast extract per ml, 0.02 mg of methionine per ml, and 0.02 mg of cysteine per ml. Incubation with the label termination mix was continued throughout the chase period. To terminate the chase period, a 0.5-ml aliquot of cells was added directly to trichloroacetic acid (5% final concentration) and the mixture was incubated on ice for 30 min. Immunoprecipitations were carried out as described previously (5). The efficiency of import (mature F₁ [%]) represents mature protein/(total precursor plus mature protein) × 100.

Immunoassay blots and enzyme assays. Mitochondria were prepared from the yeast strains as described previously (17). Published procedures were used to assay ATPase activity (44) and protein concentration (8). For immunoblot analysis of purified mitochondria, mitochondrial extracts were resuspended in SDS-PAGE sample buffer and boiled for 5 min. Ten micrograms of mitochondrial protein from each sample was analyzed by SDS-PAGE. Proteins were transferred to Immobilon-P membranes (Millipore) with a Trans-Blot apparatus (Bio-Rad). Following a 1-h incubation with 5% nonfat milk in phosphate-buffered saline-0.5% Tween 20, the blot was incubated with monoclonal antisera directed against pre-F₁ and polyclonal antisera directed against porin for 2 h at room temperature with rocking. The blot was then treated for 1 h with rabbit antisera directed against mouse immunoglobulin G. The signal was then detected by incubation with ¹²⁵I-protein A (Amersham) followed by autoradiography.

Northern blot analysis. Total cellular RNA was prepared by SDS-phenol extraction from strains grown on YMM at 30°C and was harvested at 1.2 to 1.5 A₆₀₀ units/ml (52). Equal amounts of RNA from each culture (20 µg) were subjected to agarose gel electrophoresis. Following transfer to nitrocellulose, ATP2 mRNA was detected by using a 617-bp probe that spanned nucleotides 136 to 753 of the coding sequence. The DNA probe was amplified from the ATP2 gene with primers DB697 (5'-GTTACCGCTGTCATTGGTGCCATT-3') and DB69 (5'-CACTAAGGCGACCTTGGATTCACCTT-3') by PCR and was labeled with [-³²P]dATP by random hexamer priming. TOM72 mRNA was detected by a probe corresponding to nucleotides 71 to 760 of the coding sequence that was amplified from genomic DNA with primers DB766 (5'-GGACTGCTGCTGTGGGTGCTTATT-3') and DB767 (5'-TGCGAGCCTCTGCCTTCATCCTT-3'). TOM37 mRNA was detected by a probe coresponding to nucleotides 366 to 797 of the coding sequence that was amplified from genomic DNA with primers DB764 (5'-TTCCAGACTAACGGACTACCAGC-3') and DB765 (5'-TGAACGGATGCGGTTACCATCAG-3'). TOM22 mRNA was detected by a probe corresponding to nucleotides 198 to 421 of the coding sequence that was amplified from genomic DNA with primers DB791 (5'-AGACATTGTCCCCCCAGGTAAGAG-3') and DB792 (5'-TGCAGCATCTTTTTCACCTTGGGC-3'). TOM40 mRNA was detected by a probe corresponding to nucleotides 53 to 647 of the coding sequence that was amplified from genomic DNA with primers DB789 (5'-CCCTTTCTCCTTTGACTGCGAAGC-3') and DB790 (5'-CGCTACCGTCAGTTCTGGAGTATA-3'). TOM20 mRNA was detected by a probe corresponding to nucleotides 5 to 516 of the coding sequence that was amplified from genomic DNA with primers DB787 (5'-CCCAGTCGAACCCTATCTTACTG-3') and DB788 (5'-TCGTTAGCTTCAGCAACCGCATCA-3'). Following hybridization and washing, the radioactive bands were visualized by autoradiography.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

A tom70 mutant inefficiently imports a derivative of pre-F₁ with a minimal import signal. We previously reported that the mitochondrial import signal of pre-F₁ contains redundant targeting information within its amino-terminal 34 amino acids (5, 6). 1,2 pre-F₁ is a mutant precursor that contains two small nonoverlapping deletions within its targeting signal, leaving a "minimal targeting signal" capable of facilitating import into mitochondria in vivo (Fig. 1A). Previous in vivo studies showed that this reduction to a minimal import signal has little effect on the steady-state levels of F₁ protein (or ATPase activity) in mitochondria, and 1,2 pre-F₁ is normally processed upon entry into the matrix (5).

View larger version (25K): [in this window] [in a new window]   FIG. 1.   Mitochondrial import mediated by WT and 1,2 pre-F1 targeting signals. (A) Amino acid sequences of the amino termini of WT pre-F1 and 1,2 pre-F1 as predicted from nucleotide sequences (5). Deletions are indicated by hatched boxes and basic residues are indicated by plus signs. (B) WT and tom70 cells expressing either WT or 1,2 pre-F1 were pulse-labeled for 5 min at 30°C, followed by chase times extending to 5 h. Cell lysates were subjected to immunoprecipitation with pre-F1-specific antiserum as described in Materials and Methods. (C) In vitro mitochondrial protein import of precursors containing WT or 1,2 pre-F1 targeting signals fused to the DHFR or DHFR* passenger proteins. Import reactions were carried out using WT or tom70 mitochondria as indicated.

Genetic screen for multicopy suppressors of the cold-sensitive growth defect. F₁ is an essential subunit of the F₁-ATPase complex, which carries out the generation of ATP by oxidative phosphorylation. As a result, atp2 strains are unable to grow on nonfermentable carbon sources such as lactate. We next asked whether the low level of pre-F₁ import that occurred in a tom70 strain led to a growth defect on nonfermentable carbon sources. A TOM70 strain expressing 1,2 pre-F₁ exhibited only a slight growth defect when incubated on YP lactate plates at either 30 or 15°C (Fig. 2A). However, tom70 cells expressing 1,2 pre-F₁ exhibited a much more severe cold-sensitive phenotype under these conditions (Fig. 2B). This is presumably due to the inability of the tom70 strain to assemble enough of the ATP synthase complex to sustain growth at 15°C.

View larger version (46K): [in this window] [in a new window]   FIG. 2.   Growth defects of yeast strains on a nonfermentable carbon source at low temperature. (A) WT cells and tom70 cells expressing either WT or 1,2 pre-F1 were grown on a YP lactate plate for 3 days at 30°C. (B) WT and tom70 cells expressing either WT or 1,2 pre-F1 were grown on a YP lactate plate for 8 days at 15°C.

View larger version (39K): [in this window] [in a new window]   FIG. 3.   Identification of a multicopy suppressor of the growth defect observed in a tom70 strain expressing 1,2 pre-F1. (A) Sequencing primers complementary to vector sequences on either side of the genomic insert were used to generate partial sequences, and the genes included in the complementing insert were identified by a BLAST search of the Saccharomyces Genome Database. Different fragments (indicated by solid black lines under the insert maps) were then tested for the ability to restore growth of the tom70 strain expressing 1,2 pre-F1, as determined on YP lactate plates at 15°C. Those clones that contained complementing inserts are indicated by a positive sign, while clones that did not complement the cold-sensitive phenotype are indicated by negative signs. (B) The tom70 strain expressing 1,2 pre-F1 was streaked on a YP lactate plate beside the same strain harboring a 2µm plasmid containing PDR3. The plate was incubated at 15°C for 8 days.

Multicopy PDR3 or PDR1 stimulates import of 1,2 pre-F₁ in a tom70 strain. We next examined how the expression of the PDR3 gene from a multicopy plasmid affected the import kinetics of 1,2 pre-F₁ in a tom70 strain. Since Pdr1p is a regulator of Pdr3p and both regulate many PDR genes in common, we examined the effects of expressing each of these genes from a multicopy plasmid. tom70 strains expressing 1,2 pre-F₁ that carried either the PDR3 or PDR1 genes on a multicopy plasmid were pulse-labeled for 5 min with [³⁵S]methionine-cysteine and then chased for various lengths of time in the presence of cycloheximide to prevent further protein synthesis. We found that PDR3 or PDR1 overexpression resulted in a significant increase in the amount of 1,2 pre-F₁ that was imported into mitochondria (Fig. 4A). While only 10 to 20% of 1,2 pre-F₁ was imported into mitochondria in the tom70 strain, 40 to 50% of the precursor protein was imported when multicopy plasmids expressing either PDR3 or PDR1 were present. Shorter chase times confirmed that this increase in overall yield was accompanied by an increase in the initial rate of import (data not shown). These results were not due to the stimulation of ATP2 expression (which encodes pre-F₁), since Northern blots of mRNA isolated from strains grown on media containing lactate as carbon source indicated that the steady state ATP2 mRNA abundance was not increased in the strain overexpressing PDR3 (Fig. 4B). Instead, our results indicate that the overexpression of PDR3 or PDR1 can partially restore 1,2 pre-F₁ import in a tom70 strain.

View larger version (19K): [in this window] [in a new window]   FIG. 4.   Effect of PDR3 and PDR1 overexpression on 1,2 pre-F1 import. (A) tom70 cells expressing 1,2 pre-F1 and harboring 2µm plasmids containing either PDR3 or PDR1 were pulse-labeled for 5 min at 30°C and then chased for times extending to 20 h. Cell lysates were subjected to immunoprecipitation with a pre-F1-specific antiserum as described in Materials and Methods. (B) Northern blots were performed on mRNA isolated from WT and mutant cells and mutant cells overexpressing PDR3. The probe was directed against ATP2 as described in Materials and Methods. Strains were grown at 30°C in YP lactate medium. The amount of mRNA was normalized to actin mRNA (n = 3).

View larger version (18K): [in this window] [in a new window]   FIG. 5.   Steady-state levels of F1 and ATPase activity following PDR3 and PDR1 overexpression in tom70 cells harboring 1,2 pre-F1. (A) Mitochondria isolated from WT or mutant strains harboring control plasmids or plasmids containing PDR3 or PDR1 were subjected to immunoblot analysis with F1-specific monoclonal antibodies. The steady-state level of mature F1 in mitochondria from each strain was expressed relative to the level observed in WT mitochondria. In each case, samples were normalized to the level of porin measured on parallel blots (n = 3). (B) Mitochondria freshly isolated from the strains described above were assayed for mitochondrial ATPase activity. The ATPase specific activity (Spec. Act.) (micromoles per minute per milligram of protein) in mitochondria from each strain was expressed relative to the specific activity measured in WT mitochondria (n = 3).

PDR3 overexpression increases TOM72 mRNA abundance. Since 1,2 pre-F₁ has been shown to rapidly fold into an import-incompetent conformation (5, 6, 21), we next asked whether the overproduction of these transcription factors stimulate 1,2 pre-F₁ import by increasing the expression of one or more cytosolic molecular chaperones. Recently, a microarray analysis identified 26 genes whose expression was stimulated by hyperactivating alleles of PDR1 and PDR3 (13). Most, but not all, of these genes contain a PDR element (PDRE) in their promoter region that has been shown to mediate positive control by both of these transcription factors. We performed a pattern analysis search on the S. cerevisiae genome with the PDRE sequence (5'-CCGCGG-3') and identified two putative molecular chaperone genes, XDJ1 and PDR15, that contained upstream PDRE elements. PDR15 encodes a protein with limited homology to the Hsp70 family, while XDJ1 is a hypothetical open reading frame whose product has 40% homology with Ydj1p, a chaperone previously shown to aid precursor binding to the outer mitochondrial membrane (11). However, Northern blot analysis indicated that the steady-state mRNA level of these genes was unaffected by PDR3 overexpression (results not shown).

View larger version (17K): [in this window] [in a new window]   FIG. 6.   Effects of TOM72 overexpression in mutant cells. (A) Northern blots were performed on mRNA isolated from a WT or mutant strain harboring a control plasmid or a PDR3 plasmid with DNA probes directed against TOM72 as described in Materials and Methods. Strains were grown on lactate as carbon source at 30°C. The amount of mRNA was normalized to mRNA levels of actin (n = 3). (B) Mitochondria isolated from a WT or mutant strain harboring a control plasmid or TOM72 on a 2µm plasmid were subjected to immunoblot analysis with monoclonal antibodies directed against pre-F1. Relative steady-state levels show the levels of mature F1 relative to that of the WT. The mitochondrial protein concentration was standardized for levels of porin (n = 3). (C) Mitochondria freshly isolated from the strains described above were monitored for mitochondrial ATPase specific activity. Relative activity expresses the ATPase specific activity of each strain relative to that of the WT (n = 3).

View larger version (16K): [in this window] [in a new window]   FIG. 7.   Steady-state levels of F1 and ATPase activity following PDR3 overexpression in tom70 or tom70 tom72 cells harboring 1,2 pre-F1. (A) Mitochondria isolated from WT and mutant strains and mutant strains harboring PDR3 or PDR1 on 2µm plasmids were subjected to immunoblot analysis with F1-specific monoclonal antibodies. The steady-state level of mature F1 in mitochondria isolated from each strain was expressed relative to the level observed in WT mitochondria. In each case, samples were normalized to the level of porin measured on parallel blots (n = 3). (B) Mitochondria freshly isolated from the strains described above were assayed for mitochondrial ATPase activity. The ATPase specific activity (micromoles per minute per milligram of protein) in mitochondria from each strain was expressed relative to the specific activity measured in WT mitochondria (n = 3).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Tom70p stimulates mitochondrial import of a precursor containing a defective amino-terminal signal. Mitochondrial targeting signals are involved in several distinct steps of the mitochondrial protein import process. This includes interactions with cytosolic factors, import receptors, components of the TOM and translocase-of-the-inner-membrane complexes, and chaperones within the mitochondrial matrix. In our in vivo pulse-chase analysis, we found that 1,2 pre-F₁, a precursor containing a minimal targeting signal, exhibited a reduced rate of import compared to the WT pre-F₁ precursor. We found that the tom70 mutation greatly exacerbated this import defect, resulting in a severe decrease in the overall yield of imported precursor. One possible explanation for this result is that the increased dependence of the 1,2 targeting signal on the Tom70p receptor results from a reduced ability of the Tom20p receptor to efficiently recognize this minimal import signal. To test this possibility, we examined the ability of the 1,2 targeting signal to facilitate import into tom70 mitochondria. We previously reported that the presequence of pre-F₁ acts as an intramolecular chaperone that maintains the precursor in a loosely folded, import-competent conformation (21). While the 1,2 pre-F₁ import signal imported a DHFR fusion protein much less efficiently than the WT pre-F₁ signal did, the minimal targeting signal could import a structurally destabilized form of DHFR (DHFR*) into isolated mitochondria as well as the WT pre-F₁ import signal could. In the present study, we found that 1,2 pre-F₁-DHFR* was also capable of import into tom70 mitochondria as efficiently as WT pre-F₁-DHFR*. These results demonstrate that the 1,2 import signal retains the ability to mediate efficient import in the absence of the Tom70p receptor and suggest that the 1,2 pre-F₁ targeting signal retains the ability to interact efficiently with the Tom20p receptor. Since the only difference between DHFR and DHFR* is the inability of the latter protein to fold into a stable conformation, these results suggest that Tom70p acts to maintain this precursor in a loosely folded conformation prior to import.

Tom72p functions as a mitochondrial import receptor. Tom72p is a poorly characterized protein that is located on the outer mitochondrial membrane. It shares 53% amino acid identity with Tom70p and has been shown to be associated with other components of the outer membrane translocase complex (51). It has also been shown to cross-link with a 30- to 35-kDa protein that is distinct from the Tom70p binding partner, Tom37p. Despite its strong homology to Tom70p and similar subcellular localization, a knockout of the TOM72 gene was not found to significantly affect the import of any of the mitochondrial proteins examined. It was also found that the weak mitochondrial protein import defects observed in a strain carrying disruptions of both TOM70 and TOM72 were similar to the defects observed in a strain carrying a tom70 mutation alone, other than a slightly increased temperature-sensitive growth defect on a nonfermentable carbon source (51). Our finding that the overexpression of TOM72 suppresses the import defect of 1,2 pre-F₁ in tom70 cells leads us to conclude that the functions of Tom70p and Tom72p overlap. We previously demonstrated that the 1,2 pre-F₁ targeting signal was unable to maintain the precursor in an import-competent conformation (21), and data from our in vitro experiments described in this study further support that conclusion. This suggests that both Tom70p and Tom72p bind mitochondrial precursors and maintain them in an import-competent conformation prior to engaging the downstream components of the TOM complex, as previously suggested for Tom70p (25). The genome of S. cerevisiae has undergone a duplication in recent evolutionary time, and the TOM70 gene (chromosome I) and TOM72 gene (chromosome VIII) occur on one such duplicated region of the genome (32). It is presently unknown whether the duplicated copy of the gene encoding this mitochondrial protein import receptor is unique for this organism.

The PDR pathway can influence the mitochondrial import machinery. The ability of Pdr3p overexpression to suppress defects in mitochondrial protein import reveals a previously unappreciated role for the PDR pathway in mitochondrial biogenesis. Our results show that the overexpression of Pdr3p or Pdr1p can restore the ability of this strain to grow on a nonfermentable carbon source at 15°C. This suppression was mediated by an increase in both the rate and yield of import of this defective precursor. Together, these effects raised the steady-state level of ATP synthase activity above the threshold required for growth under the selective conditions. These results clearly indicate that the PDR pathway can play a previously unrecognized role in mitochondrial biogenesis. A recent study found that this pathway is also activated in response to mitochondrial dysfunction. Hallstrom and Moye-Rowley reported that PDR3 expression is up-regulated in strains containing mutations that compromise the electron transport chain or the maintenance of the mitochondrial genome (22). It was shown that the response to mitochondrial dysfunction by Pdr3p in some cases also requires activation by the retrograde signaling pathway, which modulates the expression of nuclear genes in response to changes in the mitochondrial status (31, 36). Our results demonstrate that Pdr3p overproduction can also stimulate the import of precursors that are incapable of efficient mitochondrial import when a mitochondrial stress is imposed. In this case, the inability to utilize a nonfermentable carbon source due to limiting ATP synthase activity did not induce TOM72 transcription, suggesting that this stimulus alone cannot activate the PDR pathway. However, we found that TOM72 transcription was induced under these conditions when the PDR3 gene dosage was increased. It is possible that other forms of mitochondrial stress that provide a stronger activation of the PDR pathway may stimulate the import of some mitochondrial proteins by increasing the level of Tom72p. We propose that this mechanism represents a physiological response that optimizes mitochondrial biogenesis during mitochondrial stress or dysfunction (Fig. 8).

View larger version (18K): [in this window] [in a new window]   FIG. 8.   Model showing that Pdr3p responds to mitochondrial dysfunction by stimulating mitochondrial protein import.