MOLECULAR AND CELLULAR BIOLOGY, July 1998, p. 4043–4052 0270-7306/98/$04.0010 Copyright © 1998, American Society for Microbiology. All Rights Reserved.
Vol. 18, No. 7
Prohibitin Family Members Interact Genetically with Mitochondrial Inheritance Components in Saccharomyces cerevisiae KAREN H. BERGER
AND
MICHAEL P. YAFFE*
Department of Biology, University of California, San Diego, La Jolla, California Received 4 February 1998/Returned for modification 17 March 1998/Accepted 21 April 1998
Phb2p, a homolog of the tumor suppressor protein prohibitin, was identified in a genetic screen for suppressors of the loss of Mdm12p, a mitochondrial outer membrane protein required for normal mitochondrial morphology and inheritance in Saccharomyces cerevisiae. Phb2p and its homolog, prohibitin (Phb1p), were localized to the mitochondrial inner membrane and characterized as integral membrane proteins which depend on each other for their stability. In otherwise wild-type genetic backgrounds, null mutations in PHB1 and PHB2 did not confer any obvious phenotypes. However, loss of function of either PHB1 or PHB2 in cells with mitochondrial DNA deleted led to altered mitochondrial morphology, and phb1 or phb2 mutations were synthetically lethal when combined with a mutation in any of three mitochondrial inheritance components of the mitochondrial outer membrane, Mdm12p, Mdm10p, and Mmm1p. These results provide the first evidence of a role for prohibitin in mitochondrial inheritance and in the regulation of mitochondrial morphology. Mitochondrial inheritance is an essential and active process by which daughter cells receive mitochondria prior to the completion of cytokinesis. In budding yeast, factors specifically required for mitochondrial inheritance have been identified and characterized through the analysis of conditional mutants (7, 25). Three distinct proteins of the mitochondrial outer membrane, Mdm10p, Mmm1p, and Mdm12p, have been found to constitute one class of mitochondrial inheritance factors. Each protein is required for normal mitochondrial morphology and inheritance, and mdm10, mmm1, and mdm12 loss-of-function mutants exhibit similar phenotypes of temperature-sensitive growth and enlarged, round mitochondria (6, 9, 39). At least one of these proteins, Mdm12p, has been evolutionarily conserved and possesses a homolog in the fission yeast Schizosaccharomyces pombe (6). While the location of these proteins in the mitochondrial outer membrane suggests that they may interact with cytoskeletal elements to mediate normal mitochondrial distribution, their molecular activity remains to be defined. To explore Mdm12p function, high-copy-number plasmidborne suppressors able to bypass the cellular requirement for Mdm12p were identified. This paper describes the characterization of a plasmid-borne suppressor encoding a prohibitinrelated protein localized to the mitochondrial inner membrane and exhibiting genetic interactions with mitochondrial outer membrane inheritance components. Prohibitins are a family of conserved proteins whose first member was identified as a negative regulator of cell division in cultured animal cells (29). Prohibitin homologs have been identified in diverse organisms and cell types and have been localized to mitochondria in animal and plant cells (20, 38). The function of prohibitin at the molecular level is unknown.
for yeast were prepared as described previously (33). Yeast were transformed with lithium acetate (21). Respiration-deficient isolates were obtained by plating cells on YPDG medium (1% yeast extract, 2% Bacto Peptone, 0.1% glucose, 3% glycerol) and screening for small colonies. Candidate respiration-deficient strains were shown to be unable to grow on YPG medium (1% yeast extract, 2% Bacto Peptone, 3% glycerol). The mitochondrial DNA of these respiration-deficient cells is likely to be partially or fully deleted, as DAPI (4,6-diamidino-2-phenylindole) staining of cells failed to reveal any mitochondrial fluorescence. We refer to such respiration-deficient cells as [rho2] cells. For inhibition of mitochondrial protein synthesis, cells were grown for several generations in 1% yeast extract–2% Bacto Peptone–2% galactose at 30°C, with either no drug, erythromycin (2 mg/ml), or chloramphenicol (4 mg/ml). Genetic and molecular biological techniques. The plasmids used are described below. Escherichia coli DH5a was used to propagate plasmid DNA. General molecular biological methods were as described previously (34). PCR amplification was performed with Taq DNA polymerase (Fisher Scientific, Pittsburgh, Pa.) in Taq reaction buffer supplemented with 2.5 mM MgCl2 and 0.2 mM concentrations of deoxynucleoside triphosphates (Boehringer Mannheim Corp., Indianapolis, Ind.) by using an ERICOMP (San Diego, Calif.) thermal cycler. Specific oligonucleotides were synthesized by Operon Technologies (Alameda, Calif.). Site-directed mutagenesis was carried out by oligonucleotide-mediated mutagenesis with the Transformer kit (Clontech Laboratories, Palo Alto, Calif.). Isolation of total yeast RNA and Northern analysis were performed essentially as described previously (37). DNA probes for the PHB1, PHB2, and MDM1 genes were prepared from PCR products corresponding to the respective coding sequences and labeled with [32P]dCTP by random priming with a DNA labeling kit (Boehringer Mannheim). NIH Image (version 1.61) software was used for quantitative comparison of RNA levels following Northern analysis and of protein levels following immunoblotting. For yeast genomic library construction, yeast genomic DNA from strain MYY629 was partially digested with BamHI and BglII and ligated into the 2mm plasmid YEp13 (8), which had been digested with BamHI and treated with calf alkaline phosphatase. Isolation of high-copy-number plasmid suppressors of mdm12. Cells of mdm12-null strain MYY623 were transformed with a yeast genomic DNA library constructed in the LEU2 2mm vector YEp13 and plated onto selective media lacking leucine. Plates were incubated at 23°C (permissive temperature) until colonies had formed and were replica plated to yeast extract-peptone-dextrose (YPD) at 37°C (nonpermissive temperature). Colonies able to grow at 37°C after replica plating were further analyzed to identify those for which growth at high temperature was plasmid dependent. Of ;3,400 Leu1 transformants screened, one clone that provided partial suppression of the mdm12 mutant phenotype was identified. Construction of phb1-null and phb2-null strains. Replacement of wild-type chromosomal DNA corresponding to PHB1 and PHB2 coding sequences with prototrophic markers was performed by one-step gene replacement with PCRgenerated cassettes essentially as described previously (4). Gene replacement oligonucleotide sequences are available on request. All gene replacements were generated in diploid cells and subsequently verified by PCR analysis of haploid progeny following sporulation. Independently, PHB1 was replaced with LEU2
MATERIALS AND METHODS Strains and media. The Saccharomyces cerevisiae strains used in this work are listed in Table 1. All strains were derived from MYY290 or MYY291 (37). Media
* Corresponding author. Mailing address: University of California, San Diego, Department of Biology, 0347, La Jolla, CA 92093-0347. Phone: (619) 534-4769. Fax: (619) 534-4403. E-mail:
[email protected]. 4043
4044
MOL. CELL. BIOL.
BERGER AND YAFFE TABLE 1. S. cerevisiae strains used
Strain
MYY290 MYY291 MYY298 MYY503 MYY504 MYY623 MYY624 MYY629 MYY630 MYY631 MYY632 MYY633 MYY634 MYY635 MYY636 MYY637 MYY638 MYY639 MYY640 MYY641 MYY642 MYY643 MYY644 MYY646 MYY647 MYY648
Relevant genotype
MATa his3 leu2 ura3 MATa his3 leu2 ura3 MATa/a his3/his3 leu2/leu2 ura3/ura3 MATa his3 leu2 ura3 mdm10::URA3 MATa his3 leu2 ura3 mdm10::URA3 MATa his3 leu2 ura3 mdm12::URA3 MATa his3 leu2 ura3 mdm12::URA3 MATa his3 leu2 ura3 mdm12::URA3 SOT1 MATa his3 leu2 ura3 phb1::LEU2 MATa his3 leu2 ura3 phb1::LEU2 MATa his3 leu2 ura3 phb1::HIS3 MATa his3 leu2 ura3 phb1::HIS3 MATa his3 leu2 ura3 phb2::HIS3 MATa his3 leu2 ura3 phb2::HIS3 MATa his3 leu2 ura3 phb1::LEU2 phb2::HIS3 MATa his3 leu2 ura3 phb1::LEU2 phb2::HIS3 [rho2] derivative of MYY290 [rho2] derivative of MYY291 [rho2] derivative of MYY630 [rho2] derivative of MYY631 [rho2] derivative of MYY634 [rho2] derivative of MYY635 [rho2] derivative of MYY636 MATa/a his3/his3 leu2/leu2 ura3/ura3 mdm12::URA3/MDM12 PHB1/phb1::LEU2 MATa/a his3/his3 leu2/leu2 ura3/ura3 mdm12::URA3/MDM12 PHB2/phb2::HIS3 MATa his3 leu2 ura3 mdm12::URA3 phb2::HIS3 SOT1
and with HIS3 to generate phb1::LEU2 and phb1::HIS3, respectively, and PHB2 was replaced with HIS3 to generate phb2::HIS3. Determination of replicative life span. Replicative life span was determined essentially as described previously (27). Briefly, virgin mother cells were isolated as new daughters and were separated by micromanipulation from each daughter. The original mother was retained until it lysed or stopped dividing. The number of generations for a given mother was equal to the number of buds put forth by that cell. Replicative life spans were determined for cells grown on YPD at 30°C, with plates refrigerated overnight for the course of the experiment. Life spans were determined for at least 54 mother cells for each strain analyzed, and these data were used to determine mean and maximum replicative life spans. Statistical analysis was performed by using Student’s two-tailed t test. Gene constructions. For site-directed oligonucleotide mutagenesis of the YEp13-based PHB2-tetA suppressing clone (plasmid pKB38), the selection primer in each case was oligonucleotide XhoI-KILL (59GAAGTTCTCCTGGA GGATTTAG-39) (mutated nucleotides are underlined throughout), which destroys the unique XhoI site in this plasmid. An in-frame stop codon was created between the 39-truncated PHB2 and the 59-truncated tetA in plasmid pKB38 with the mutagenic primer PT-STOP (59-CGATGCGTCCGGTCTAGAGGATCTT TGC-39), yielding plasmid pKB39 (PHB2-STOP). To destroy TetA transporter activity, Asp-469 of Phb2p-TetA (corresponding to Asp-287 of native TetA from pBR322) was mutated to Ala with the mutagenic oligonucleotide TetA-DA (59-CCCAGCGCGGCGGCCGCCAT-39). Mutations were confirmed by nucleotide sequence analysis. For construction of a hemagglutinin (HA)-tagged version of PHB2-tetA, a unique BamHI site at the 39 end of tetA in plasmid pKB38 was created by oligonucleotide-directed mutagenesis using the mutagenic primer TetA-BamHI (59-CGAGGTGGCCCGGATCCATGCACCGCG-39) to yield plasmid pKB40. Sequences encoding three tandem repeats of the influenza virus HA epitope were liberated from plasmid pGTEP1 (36) by digestion with NotI, fragment ends were filled with Klenow fragment, and the fragment was ligated into plasmid pKB40, which had been digested with BamHI and treated with Klenow fragment and calf alkaline phosphatase, yielding plasmid pKB41. Full-length wild-type versions of PHB1 and PHB2 were cloned following PCR amplification from genomic DNA of strain MYY290. For PHB1, primers PHB1-U2 (59-GGGGATTCTTCAGGGAAAGGGAGTTTGACGAT-39) and PHB1-L2 (59-GGGGATTCAGCAGAAGGAGGGCAAGAAGACAA-39) were used to amplify the coding region along with approximately 150 bp of 59 noncoding sequence and 400 bp of 39 noncoding sequence, flanked by BamHI sites. For PHB2, the coding region and approximately 570 bp of 59 noncoding sequence and 250 bp of 39 noncoding sequence were amplified with flanking BamHI sites by using primers PHB2-U2 (59-GGGGATCCAACAAGGAAGG TTTGGAGTGTAGC-39) and PHB2-L2 (59-GGGGATCCGAATATACCACG CAAGCCGAATGT-39). A centromere-based plasmid containing the HA-tagged PHB2-tetA suppressor
Reference or source
37 37 26 39 39 6 6 6 This This This This This This This This This This This This This This This This This This
study study study study study study study study study study study study study study study study study study
was constructed by digestion of pKB41 with NheI and PvuII, which recognize unique sites 59 and 39 of PHB2-tetA-HA in YEp13 sequences, and ligation with plasmid pRS315 (36a), which was digested with XbaI and SmaI. Expression of the centromere vector-based Phb2p-TetA-HA at reduced levels relative to the 2mm-based suppressing clone was verified by Western analysis using anti-HA antibodies. A plasmid for high-level expression of both PHB1 and PHB2 was generated by subcloning the BamHI fragment containing PHB2 into YEp13::PHB1. High-level expression of Phb1p and Phb2p from cells harboring the 2mm-PHB1-PHB2 plasmid was verified by Western analysis. Preparation of antibodies. To generate antibodies specific for Phb1p and Phb2p, vectors were constructed to express fusion proteins LacZ-Phb1p and LacZ-Phb2p, respectively. For plasmid pKB48, the 39 portion of PHB1 (from the unique ClaI site to the unique PstI site 39 of the gene) was introduced into plasmid pTRB1 (10), which was digested with BamHI, treated with Klenow fragment, and then digested with PstI. Plasmid pKB49 consisted of the 39 portion of PHB2 (from the unique ApaLI site) introduced into the BamHI site of plasmid pTRB1 following modification with Klenow fragment. The fusion proteins were expressed in bacterial cells, purified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electroelution, and injected into rabbits (18). The LacZ-Phb1p fusion protein contained the C-terminal 279 amino acids of Phb1p, and the LacZ-Phb2p protein contained the C-terminal 265 amino acids of Phb2p. At the antibody concentrations used for immunoblot detection of Phb1p and Phb2p, antiserum cross-reactivity was not observed. Western analysis, indirect immunofluorescence, and fluorescence microscopy. Whole-cell protein extracts were prepared by glass bead lysis in radioimmunoprecipitation assay buffer (150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, 50 mM Tris-HCl [pH 8.0]) (18) with the addition of 1 mM phenylmethylsulfonyl fluoride. SDS-PAGE and immunoblotting were performed as described previously (39, 42). Protein concentrations were determined by bicinchoninic acid assays (Pierce, Rockford, Ill.). Methods for indirect immunofluorescence, staining with 2-(4-dimethylaminostryl)-1-methylpyridinium iodide (DASPMI) and DAPI, and fluorescence microscopy were as described previously (25). Additional antisera were specific for Tom70p (31), OM45 (44), F1b (22), Mdm10p (39), or Mas2p (22), and have been described previously. The monoclonal antibody 12CA5 (Berkeley Antibody Co., Richmond, Calif.) was used for immunodetection of the HA epitope (15). Subcellular fractionation. Yeast cells were grown in semisynthetic lactate medium (12) at 30°C, converted to spheroplasts, homogenized, and subjected to differential centrifugation to isolate subcellular fractions as previously described (35, 43). Purified inner and outer mitochondrial membranes were obtained following osmotic shock, sonication, and sucrose density gradient centrifugation (12). Mitoplasts, possessing largely intact mitochondrial inner membranes and osmotically disrupted outer membranes, were prepared as described previously (12).
VOL. 18, 1998
PROHIBITIN AND MITOCHONDRIAL INHERITANCE
4045
FIG. 1. Plasmid-mediated suppression of growth and mitochondrial morphology defects. (A) Suppression of the mutant growth defect. Equal numbers of mdm12-null cells harboring the high-copy-number plasmids indicated were plated on selective medium lacking leucine and incubated at 30°C (semipermissive temperature). Lane x shows cells plated at a 10-fold dilution relative to cells shown in lane 10x. Plasmids were vector, YEp13; PHB2-tetA, the smallest suppressing subclone; PHB2-STOP, the subclone with an in-frame stop codon between PHB2 and tetA; and PHB21, a vector with the wild-type PHB2 gene. (B) Partial suppression of mitochondrial morphology defects. Mitochondria were visualized by fluorescence microscopy following DASPMI uptake. Cells harbored either vector (YE13) or the PHB2-tetA plasmid. Representative wild-type MDM10 MDM12 (strain MYY290), mdm12-null (MYY623), and mdm10-null (MYY503) cells are shown. Bar 5 2 mm.
RESULTS Identification of a high-copy-number suppressor of mdm12. Cells lacking Mdm12p are viable at 23°C but fail to grow at 37°C (6). To investigate further the cellular role of Mdm12p, a screen for high-copy-number plasmid-borne suppressors of the temperature-sensitive growth defect of mdm12-null cells was initiated (see Materials and Methods). A single clone which conferred improved growth of mdm12-null cells at elevated temperature was identified (Fig. 1A). Mutant cells harboring the suppressing clone also showed a partial restoration of normal mitochondrial morphology (Fig. 1B). In unsuppressed mdm12-null cells, the wild-type mitochondrial reticulum appeared to have collapsed into one or two giant spherical organelles which were largely defective for entry into buds even
at the permissive temperature, 23°C (6) (Fig. 1B). In mdm12null cells harboring the suppressor plasmid, mitochondria appeared smaller and more fragmented than in unsuppressed mutant cells and also sometimes exhibited elongated morphologies (Fig. 1B). In addition, while mitochondria were typically restricted to the mother portions of unsuppressed mdm12-null cells (6), mitochondria were often distributed in both mother and bud portions of suppressed cells. Cells harboring the mdm10-null mutation, which exhibit phenotypes extremely similar to those of mdm12-null cells, were also transformed with the suppressing plasmid. The mdm10-null transformants also exhibited partial suppression of aberrant mitochondrial morphology and distribution (Fig. 1B) as well as of temperature-sensitive growth (data not shown). Cells with an mmm1-
4046
BERGER AND YAFFE
FIG. 2. Schematic diagram of the suppressing PHB2-tetA clone. PHB2 sequences were cloned as an in-frame gene fusion with the 39 portion of tetA in the 2mm-based vector YEp13, as described in Materials and Methods. Illustrated is the smallest suppressing subclone (plasmid pKB38), generated by deletion of the insert DNA up to the unique ApaI site 59 of PHB2. Numbering refers to nucleotides.
null mutation, which are phenotypically similar to mdm10-null and mdm12-null cells (9), also showed improved growth and mitochondrial morphology when harboring the suppressing plasmid (data not shown). Mitochondrial morphology in otherwise wild-type (MDM10 MDM12) cells did not appear altered by the suppressing plasmid (Fig. 1B). The temperature sensitivity of other mutants functionally unrelated to the mdm10 and mdm12 mutants, namely the mas1 (45), mas3 (37), and mas5 (2) mutants, and of the pep12/vps6- (5) and vps18null (32) mutants was not suppressed by transformation of these strains with the plasmid, indicating that suppressing activity was restricted to a limited class of temperature-sensitive mutants. Suppression is conferred by a prohibitin homolog. An analysis of sequences in the suppressing plasmid revealed that suppression was conferred by a single S. cerevisiae open reading frame. The cloned S. cerevisiae sequences corresponded to a previously uncharacterized open reading frame, recently designated PHB2 (Saccharomyces Genome Database, Stanford, Calif.) (11). PHB2 encodes a predicted protein product of 315 amino acids and 34.1 kDa which is homologous to a second S. cerevisiae protein, prohibitin, the product of PHB1. Prohibitin (Phb1p) and the prohibitin-related protein encoded by PHB2 (Phb2p) in yeast display 52% amino acid identity and 63% similarity, with conserved residues distributed throughout the lengths of the proteins (data not shown). The proteins show the greatest divergence at their N and C termini. Phb1p and Phb2p are homologous to prohibitins and prohibitin-related proteins previously identified in a wide range of organisms (11, 14, 28, 29, 38, 41). In mammalian cells, prohibitin has been implicated as a regulator of cell growth, and this protein may function in diverse cellular processes including development and tumor suppression (24, 29). DNA sequence analysis of the suppressing plasmid revealed that genomic library construction had fortuitously generated an in-frame gene fusion between PHB2 and the bacterial tetracycline resistance gene, tetA, carried by the pBR322-derived vector YEp13 (Fig. 2). The 59 PHB2 portion of the PHB2-tetA gene fusion encoded all but the C-terminal 34 amino acids of the 315-amino-acid Phb2p; the 39 tetA sequences encoded the last 297 amino acids of the 396-amino-acid product of the bacterial tetracycline resistance gene (30). The predicted PHB2-tetA product is a 578-amino-acid protein of 62.3 kDa. To determine whether tetA sequences were required for PHB2tetA-mediated suppression of mdm12, an in-frame stop codon between PHB2 and tetA sequences in the suppressing clone was generated by site-directed mutagenesis. The resulting clone no longer suppressed mdm12 defects (PHB2-STOP; Fig. 1A).
MOL. CELL. BIOL.
Furthermore, the high-level expression of a wild-type copy of PHB2 also did not confer suppression of mutant growth (Fig. 1A) or mitochondrial morphology, nor did expression of these constructs appear to affect mitochondrial morphology in wildtype (MDM12) cells (data not shown). Finally, suppression could not be mediated by lower-level expression of PHB2-tetA from a centromere-based plasmid, as shown by the failure of such a construct to improve growth or mitochondrial morphology of mdm12-null transformants (data not shown). The native TetA protein is located in the bacterial inner membrane where it spans the membrane multiple times and confers tetracycline resistance by an active efflux mechanism. To test whether transporter activity of the TetA moiety of the Phb2p-TetA fusion protein was required for mdm12 suppression, an aspartate residue required for transporter activity of native TetA (1) was mutated to alanine (D469A; corresponds to D287A in native TetA) by oligonucleotide-directed mutagenesis. The mutated protein remained fully competent for suppression (data not shown), indicating that the transporter activity was not required for this function. Characterization of phb1 and phb2 mutants. To investigate prohibitin function, null alleles of both PHB1 and PHB2 were generated by integrative transformation and gene replacement. Either singly or together, phb1-null and phb2-null mutations did not confer any apparent phenotype on otherwise wild-type cells. Growth of phb1-null and/or phb2-null mutant cells on various carbon sources (ethanol, galactose, glycerol, glucose, and raffinose) and at different temperatures (14, 23, 30, and 37°C) appeared indistinguishable from that of the parental (PHB1 PHB2) strain. Mitochondrial morphology and distribution, visualized by DASPMI staining and fluorescence microscopy, appeared unaltered by the loss of Phb1p and/or Phb2p (data not shown). Homozygous phb1/phb1 phb2/phb2 diploids remained competent for sporulation and did not exhibit reduced spore viability. Finally, phb1-null and phb2-null mutations did not increase the frequency of loss of respiratory capacity associated with deletion of the mitochondrial genome (generation of [rho2] cells) compared to that for the parental strain (data not shown). To assess further a possible mitochondrial role for PHB1 and PHB2, mitochondria in phb1-null phb2-null cells which had lost respiratory function due to the deletion of mitochondrial DNA ([rho2]) were examined. These cells did not exhibit any apparent difference from otherwise isogenic PHB1 PHB2 [rho2] cells in growth on fermentable carbon sources at 14, 23, 30, or 37°C. However, indirect immunofluorescence microscopy revealed that the phb1 phb2 [rho2] cells displayed substantial alterations of mitochondrial morphology and distribution (Fig. 3). Wild-type (PHB1 PHB2) [rho2] cells contained mitochondria with typical tubular morphology and peripheral distribution, and most cells (77%; n 5 207) contained at least one snake-like mitochondrion that spanned at least one-fourth of the length of the cell. In contrast, most mitochondria in phb1-null phb2-null [rho2] cells lacked the normal reticular morphology and organized mitochondrial distribution, and only a minority of cells (20%; n 5 303) exhibited even one tubular mitochondrion that was at least one-fourth the length of the cell. In single-mutant phb1-null or phb2-null [rho2] cells, mitochondria also appeared fragmented and disorganized compared to those in PHB1 PHB2 [rho2] cells, similar to what was found for the double-null mutant (data not shown). These results indicate that Phb1p and Phb2p are important for normal mitochondrial morphology and distribution specifically in cells which lack a functional mitochondrial genome. To address whether the mitochondrial morphology defect in phb1 and/or phb2 [rho2] cells was related to the absence of the
VOL. 18, 1998
PROHIBITIN AND MITOCHONDRIAL INHERITANCE
4047
FIG. 3. Altered mitochondrial morphology in phb1 phb2 [rho2] cells. Cells with deficiencies of mitochondrial DNA which possessed either wild-type (left panels) or null (right panels) alleles of PHB1 and PHB2 were grown in YPD liquid medium at 30°C, fixed, and processed for indirect immunofluorescence microscopy. Mitochondria were visualized with antibodies to mitochondrial outer membrane protein OM14, and nuclear DNA was stained with DAPI. Bar 5 2 mm.
mitochondrial genome or to defective mitochondrial function, phb1 phb2 mutant cells which were respiration deficient due to a mutation in the cox4 gene were examined by indirect immunofluorescence microscopy. These cells exhibited mitochondrial morphology indistinguishable from that of PHB1 PHB2 cox4 cells (data not shown). Mitochondrial morphology was also examined in PHB1 PHB2 and phb1 phb2 [rho1] cells following inhibition of mitochondrial protein synthesis with either chloramphenicol or erythromycin. These treatments did not affect mitochondrial morphology in either PHB1 PHB2 or phb1 phb2 cells (data not shown). These results suggest that mitochondrial DNA (or associated proteins) may play a structural role which is revealed in the absence of Phb1p and Phb2p. At the time of this work, the function of PHB1 and PHB2 in budding yeast had not been previously described. In unpublished data, loss of function of the PHB1-encoded prohibitin in yeast was reported to increase replicative life span, defined as the number of times a single yeast cell produces daughter buds before ceasing to divide (3). The effect of phb1-null and phb2null mutations on replicative life span for both cells containing a wild-type mitochondrial genome ([rho1]) and cells with the mitochondrial DNA deleted ([rho2]) was examined. In these cells, loss of Phb1p and/or Phb2p function did not increase the replicative life span. In fact, loss of Phb1p and/or Phb2p function reduced replicative life span, as measured by either mean or maximum life span (Table 2). For [rho1] cells, either the phb1-null or the phb2-null mutation significantly reduced mean replicative life span (P , 0.01). The double phb1-null phb2-null mutant also showed a similarly reduced life span relative to wild-type cells (P , 0.001). The single phb1-null mutation had a more severe effect on replicative life span than the phb2-null or the double phb1-null phb2-null mutation (P , 0.001). In [rho2] cells, both the phb2-null and the phb1-null phb2-null mutants showed reduced life spans compared to wild-type [rho2] cells (P , 0.01), but the [rho2] phb1-null mutant did not. These data are substantially in agreement with another
recent analysis of Phb1p and Phb2p function in budding yeast (11), although we did not observe an enhanced effect on replicative life span in the double mutant. Phb1p and Phb2p are proteins of the mitochondrial inner membrane. Phb1p and Phb2p homologs have been localized to mitochondria in higher animal and plant cells (20, 38) but have not been previously localized in yeast. Further, the proteins’ submitochondrial distribution has not been firmly established in any cell type. To address the site of activity of Phb1p and Phb2p, polyclonal antisera recognizing each of the proteins were generated and used for immunoblot analysis of different
TABLE 2. Replicative life spana Genotype
Replicative life span (generations) Mean 6 SD
Median
Maximum
[rho ] PHB1 PHB2 phb1 PHB2 PHB1 phb2 phb1 phb2
25.6 6 7.3 17.3 6 4.8 20.9 6 8.2 20.5 6 6.9
27.4 17.7 21.3 20.1
36.0 23.9 33.3 30.6
[rho2] PHB1 PHB2 phb1 PHB2 PHB1 phb2 phb1 phb2
14.8 6 6.6 14.7 6 6.4 10.4 6 4.4 11.8 6 6.2
14.1 14.4 10.0 12.2
24.8 24.1 16.3 21.2
1
a Life spans were determined for n individual cells with the relevant partial genotypes indicated. The maximum life span is the number of generations after which 90% of the cells in a population have ceased to divide (10% still dividing). Strains analyzed were as follows: [rho1] PHB1 PHB2, MYY290 and MYY291 (n 5 60); [rho1] phb1 PHB2, MYY630 and MYY631 (n 5 103); [rho1] PHB1 phb2, MYY634 and MYY635 (n 5 64); [rho1] phb1 phb2, MYY636 and MYY637 (n 5 117); [rho2] PHB1 PHB2, MYY638 and MYY639 (n 5 103); [rho2] phb1 PHB2, MYY640 and MYY641 (n 5 106); [rho2] PHB1 phb2, MYY642 and MYY643 (n 5 100); [rho2] phb1 phb2, MYY644 (n 5 54).
4048
BERGER AND YAFFE
MOL. CELL. BIOL.
FIG. 4. Localization of Phb1p, Phb2p, and Phb2p-TetA. (A) Immunoblot analysis of subcellular fractions. Subcellular fractions were obtained from mdm12-null cells harboring the PHB2-tetA suppressor. F1b and OM45 were used as markers for the mitochondrial inner and outer membranes, respectively. Samples in the Phb2p-TetA-HA section were immunoblotted with antibodies that recognize the HA epitope of the ;68-kDa fusion protein. These antibodies also recognize an unrelated, endogenous yeast protein of lower molecular mass that fractionates with the cytosol. Ho, total cell homogenate; L, low-speed pellet; M, mitochondrial pellet; I, intermediate-speed pellet; Hs, high-speed pellet; C, cytosolic fraction. (B) Phb1p, Phb2p, and Phb2p-TetA are localized to the mitochondrial inner membrane. Inner and outer mitochondrial membranes were resolved by sucrose density gradient centrifugation. Fractions were numbered in the order collected, from 1 (bottom) to 17 (top). Shown are fractions from the middle of the gradient which contained the peak concentrations of mitochondrial inner and outer membrane proteins. Control proteins were F1b, a mitochondrial inner membrane protein, and Tom70p, a mitochondrial outer membrane protein. Equal amounts of protein were loaded for each fraction. (C) Trypsin treatment of mitoplasts defines the topology of Phb1p, Phb2p, and Phb2p-TetA-HA in the mitochondrial inner membrane. Mitoplasts were digested with various amounts of trypsin either with or without the addition of 1% Triton X-100. Proteins were resolved by SDS-PAGE and analyzed by immunoblotting. Mas2p is a mitochondrial matrix protein protected from trypsin in the absence of detergent.
subcellular fractions. Both Phb1p and Phb2p were concentrated in fractions enriched for mitochondrial proteins (Fig. 4A). In subfractionation of purified mitochondrial membranes, both Phb1p and Phb2p behaved as components of the mitochondrial inner membrane (Fig. 4B). Additionally, neither Phb1p nor Phb2p was extracted from the membranes with alkali carbonate treatment (see below), indicating that both polypeptides behaved as integral membrane proteins. Identical subcellular and submitochondrial distributions of Phb1p and Phb2p in Phb2p-TetA-suppressed mdm12 cells (Fig. 4) and wild-type cells were observed (data not shown). The subcellular localization of the suppressing Phb2p-TetA fusion protein was also analyzed. Antibodies generated against native Phb2p did not recognize the predicted product of the PHB2-tetA gene fusion in immunoblot analysis. Accordingly, a version of this protein tagged at the extreme C terminus of TetA with three copies of the influenza virus HA epitope was generated (Phb2p-TetA-HA). The HA-tagged version of the Phb2p-TetA clone still conferred suppression of mdm12-null growth and mitochondrial morphology defects (data not shown). Immunoblot analysis of mdm12-null cells harboring the Phb2p-TetA-HA construct indicated that the fusion protein was localized to the mitochondria, like native Phb2p (Fig. 4A). Furthermore, submitochondrial fractionation also indicated that Phb2p-TetA-HA was a component of the mitochondrial inner membrane, like native Phb2p (Fig. 4B). To address the topology of Phb1p, Phb2p, and Phb2p-TetA in the mitochondrial inner membrane, mitoplasts, which possess broken mitochondrial outer membranes and intact inner membranes, were prepared from mdm12-null cells harboring the Phb2p-TetA-HA suppressing clone and examined for the accessibility of these proteins to trypsin in the presence and absence of detergent permeabilization (Fig. 4C). In these samples, Phb1p and Phb2p both behaved in a manner similar to Mas2p, a protein localized to the mitochondrial matrix and therefore protected from external protease in the absence of detergent (22). In intact mitoplasts, Phb1p and Phb2p were relatively resistant to trypsin treatment but became accessible
to protease degradation with the addition of detergent. In contrast, the HA epitope of Phb2p-TetA-HA appeared very sensitive to even low levels of trypsin either in the presence or absence of detergent, suggesting that at least the C-terminal portion of the fusion protein containing the HA epitope was exposed to the intermembrane space. Phb1p and Phb2p are interdependent. The characterization of Phb1p and Phb2p localization revealed that the presence of these two proteins in the cell was interdependent. In phb1-null cells, Phb2p was no longer detectable by Western analysis, and Phb1p was similarly reduced to undetectable levels in phb2null cells (Fig. 5A). Phb1p levels in phb2-null cells were restored by expression of PHB2 from a centromere-based plasmid, and Phb2p was likewise restored in phb1-null cells by PHB1 expression (Fig. 5A). To determine whether the interdependence of Phb1p and Phb2p occurred at the level of gene expression, levels of PHB1 and PHB2 mRNA were examined by Northern analysis (Fig. 5B). In phb2-null cells, PHB1 RNA levels were slightly reduced, and PHB2 RNA levels were slightly decreased in phb1-null cells (Fig. 5B). These modest reductions in RNA levels seemed unlikely to account for the corresponding absence of detectable polypeptide subunits, and the interdependence of Phb1p and Phb2p therefore appeared to be largely posttranscriptional. A potential model for the interdependence of Phb1p and Phb2p is that the proteins form a complex in the mitochondrial inner membrane, such that one protein is destabilized in the absence of the other. Attempts to detect such a complex by coimmunoprecipitation of Phb1p and Phb2p, either in the presence or absence of chemical cross-linking, were unsuccessful. Also, while loss of either Phb1p or Phb2p resulted in a decrease of the other protein to below detectable levels, increased expression of one of the proteins did not result in a comparable increase in levels of the other species (Fig. 5C). In cells expressing high levels of Phb1p, Phb2p levels were increased about twofold, while high-level expression of Phb2p had no effect on levels of Phb1p (Fig. 5C). Because Phb1p was not detected in immunoblot analysis of
VOL. 18, 1998
PROHIBITIN AND MITOCHONDRIAL INHERITANCE
4049
FIG. 5. Phb1p and Phb2p are interdependent. (A) Immunoblot analysis of total cell protein extracts from phb1-null (lanes 1 and 2) and phb2-null (lanes 3 and 4) cells harboring a centromere-based vector with no insert (lanes 1 and 3), with wild-type PHB1 (lane 2), or with wild-type PHB2 (lane 4). Protein extracts were blotted with antibodies against Phb1p, Phb2p, or F1b, as indicated. (B) Quantitation of Northern analysis of PHB1 and PHB2. Levels of PHB1 RNA in phb2-null cells harboring either centromere-based vector alone (2) or CEN-PHB2 (1) were compared (left side), as were PHB2 RNA levels in phb1-null cells harboring either vector (2) or CEN-PHB1 (1) (right side). PHB1 and PHB2 RNA levels were standardized by using MDM1 RNA as a control for recovery; levels of this message did not vary with mutations in PHB1 and/or PHB2. (C) Overexpression of Phb1p and Phb2p. Levels of Phb1p, Phb2p, F1b, and OM45 in protein extracts were analyzed by immunoblotting. Protein extracts in the left section (PHB1) were from phb1-null cells harboring either vector alone (2), centromere-based vector containing wild-type PHB1 (1), or 2mm-based vector containing PHB1 (11). The right section (PHB2) shows protein extracts from phb2-null cells harboring either vector alone (2), centromere-based vector containing wild-type PHB2 (1), or 2mm-based vector containing PHB2 (11).
phb2-null cells (and vice versa), it could not be simply determined whether the localization of one protein was altered in the absence of the other. However, Phb2p could be detected in phb1-null cells that harbored a high-copy-number PHB2 plasmid (Fig. 6A). In these cells, the subcellular distribution of Phb2p appeared unaltered compared to that of control fractions from wild-type cells expressing high levels of Phb2p (Fig. 6A). Additionally, Phb2p was not extracted from mitochondrial membranes by carbonate treatment whether or not Phb1p was present (Fig. 6B). These insults indicate that Phb1p does not play an essential role in the import or membrane insertion of Phb2p.
FIG. 6. Localization and membrane association of overexpressed Phb2p. (A) Subcellular distribution of overexpressed Phb2p in PHB1 and phb1-null cells. Subcellular fractions are the same as those described in the legend for Fig. 4. (B) Membrane association of Phb1p and Phb2p. Mitochondria were extracted with 0.1 M Na2CO3 and supernatant (S) and pellet (P) fractions were isolated by centrifugation. 1, PHB1 PHB2 cells harboring vector only (MYY291-YEp13); 2, PHB1 PHB2 cells harboring high-copy-number PHB2 (MYY291-YEp13::PHB2); 3, phb1-null cells harboring high-copy-number PHB2 (MYY632-YEp13::PHB2). As previously documented (39), antiserum against Mdm10p recognizes two species: authentic Mdm10p, which is associated with the pellet, and an unrelated polypeptide of slightly higher molecular weight which fractionates with the supernatant.
Interestingly, the Phb2p-TetA suppressor protein did not appear to depend on either wild-type Phb1p or Phb2p for its stability, as it could be detected in protein extracts from phb1null or phb2-null cells at levels similar to those in wild-type (PHB1 PHB2) cells (data not shown). Additionally, the Phb2pTetA protein could not stabilize Phb1p in the absence of Phb2p and could not stabilize Phb2p in the absence of Phb1p (data not shown). High-level expression of both Phb1p and Phb2p in the same cell was not able to mimic the suppression of mdm12 mutant phenotypes, nor did such expression confer any apparent phenotype of growth or mitochondrial morphology on wild-type (MDM10 MDM12) cells (data not shown). Genetic interaction of prohibitin mutations with mdm mutations. Because PHB2 was identified in a genetic screen for high-copy-number suppressors of mdm12 loss of function, we examined whether the phb2 or phb1 mutation might confer a synthetic phenotype in combination with the mdm12 mutation. A genetic analysis of progeny of a cross of single-mutant phb2null and mdm12-null strains indicated that the double-mutant phb2 mdm12 strain was inviable (Table 3). A cross of phb1-null and mdm12-null mutant strains yielded similar results, indicating that the phb1 mdm12 double mutant was either inviable or marginally viable, forming only extremely small colonies. Mutations in either PHB1 or PHB2 also appeared deleterious in combination with the loss of function of either Mdm10p or Mmm1p, two other mitochondrial outer membrane proteins which, like Mdm12p, are required for normal mitochondrial inheritance and morphology (9, 39) (Table 3 and data not shown). Finally, phb1-null phb2-null mdm10-null and phb1-null phb2-null mdm12-null triple mutants were generated and also were inviable or extremely sick (data not shown). Viable phb2 mdm12 cells could be obtained if the heterozygous diploid parent strain harbored a plasmid encoding wild-type PHB2; such cells were unable to lose the plasmid even on rich medium. The phb1- and phb2-null mutations did not produce synthetic phenotypes when combined with other mdm mutations in genes which encode factors not localized to mitochondria, specifically, different mutant alleles of mdm1 (16) or an mdm2 mutation (40). Additionally, a phb1-null or phb2-null mutation did not produce synthetic inviability when combined with the loss of OM45, a mitochondrial outer membrane pro-
4050
BERGER AND YAFFE
MOL. CELL. BIOL.
TABLE 3. Synthetic inviability of prohibitin and mdm mutations Parental genotype
Spore phenotype 2
2
No. of spores Viable
Total
MDM12/mdm12::URA3 phb1::LEU2/PHB1
Ura Ura1 Ura2 Ura1
Leu Leu2 Leu1 Leu1
57 64 68 9a
59 70 70 59
MDM12/mdm12::URA3 phb2::HIS3/PHB2
Ura2 Ura1 Ura2 Ura1
His2 His2 His1 His1
87 75 69 0
88 80 80 88
MDM10/mdm10::URA3 phb1::LEU2/PHB1
Ura2 Ura1 Ura2 Ura1
Leu2 Leu2 Leu1 Leu1
15 15 17 4a
15 17 17 15
MDM10/mdm10::URA3 phb2::HIS3/PHB2
Ura2 Ura1 Ura2 Ura1
His2 His2 His1 His1
22 26 26 10a
22 26 26 22
a When viable, these double-mutant haploid progeny formed very small colonies. Total numbers of spores of each phenotypic class were inferred by tetrad analysis.
tein which does not play a role in mitochondrial inheritance (44), or when combined with a mutation in TOM70, encoding an outer membrane component of the mitochondrial protein import apparatus (19) (data not shown). To examine further the specificity of the synthetic interaction, mdm10 and mdm12 mutants were crossed to strains harboring mutations in other genes encoding mitochondrial proteins. The mdm10 and mdm12 mutations did not produce synthetic lethality when combined with mutations in genes not known to have roles in mitochondrial inheritance, namely, cox4 (encoding subunit IV of cytochrome c oxidase [13]), sdh2 (encoding the iron-sulfur subunit of succinate dehydrogenase [23]), and OM45 (an integral protein of the mitochondrial outer membrane [44]). Double mutants, which could be obtained in each case, did not exhibit apparent synthetic-growth phenotypes. The mdm10 and mdm12 mutations did appear to be synthetically lethal in combination with a yme2 lesion (17)
(data not shown), which is known to decrease mitochondrial structural integrity. Synthetic lethality of the mdm12 phb2 double mutant is suppressed by the SOT1 mutation. Previously, a novel mutation, the SOT1 mutation, was shown to suppress the growth and mitochondrial morphology defects of mdm12, mmm1, and mdm10 mutant cells (6). To examine the effect of the SOT1 mutation on the synthetic lethality of mdm12 and phb2 mutations, the haploid progeny of a cross between mdm12 SOT1 and phb2 strains were examined by tetrad analysis. A majority (13 of 21 cells) of the triple-mutant mdm12 phb2 SOT1 cells were viable, indicating that the SOT1 mutation could at least partially suppress the synthetic lethality. In these cells, mitochondrial morphology and inheritance appeared indistinguishable from those in otherwise isogenic cells with wild-type PHB2 (Fig. 7). The plasmid-encoded PHB2-tetA suppressor was also tested for its ability to abrogate the synthetic lethality of phb2 mdm12 cells. Double-mutant phb2 mdm12 cells could be obtained at a low frequency (2 of 7 cells) following sporulation if they harbored the PHB2-tetA suppressor, but these haploid progeny appeared extremely sick indicating, at most, very limited suppression. A similar experiment indicated that the PHB2-tetA suppressor likewise failed to rescue the phb1 mdm12 double mutant (data not shown). Finally, high-copynumber expression of PHB1 could not substitute for that of PHB2 in the phb2 mdm12 double mutant, nor could high-copynumber PHB2 rescue the phb1 mdm12 double mutant. DISCUSSION While investigating the role of Mdm12p in mitochondrial inheritance, we identified a genetic interaction between a class of inheritance components of the mitochondrial outer membrane and proteins belonging to the prohibitin family. This interaction was uncovered through the characterization of a fusion protein that partially suppresses an mdm12-null mutation when expressed from a high-copy-number plasmid. The fusion protein consists of the Phb2p protein fused to the bacterial tetracycline resistance protein, TetA, this fusion having arisen fortuitously through the construction of a yeast genomic library in a yeast-E. coli shuttle vector. The Phb2p-TetA fusion protein altered the mitochondrial morphology of mdm12 and mdm10 mutant cells but did not appear to affect mitochondria in otherwise wild-type cells. Although the mechanism of sup-
FIG. 7. Double mutant phb2 mdm12 cells harboring the SOT1 suppressor mutation are viable and exhibit the suppressed mitochondrial phenotype. SOT1-mutationsuppressed mdm12-null cells which carried either PHB2 (strain MYY629) or null mutation phb2 (strain MYY648) were grown in YPD at 23°C, stained with DASPMI, and examined by fluorescence microscopy. Bar 5 2 mm.
VOL. 18, 1998
PROHIBITIN AND MITOCHONDRIAL INHERITANCE
pression is unknown, the Phb2p-TetA fusion protein could not bypass the cellular requirement for Phb1p and Phb2p in mdm10 and mdm12 mutant cells. Also, the fusion protein localized to the same submitochondrial site as the authentic Phb1p and Phb2p. One model of suppression consistent with these observations is that the suppressing protein activates or augments some normal activity of Phb1p and Phb2p. The effect of the fusion protein might be to cause changes in the physical properties of the mitochondrial membranes or to alter the interaction of the membranes with specific peripheral proteins, leading to more tubular mitochondria and a more dispersed mitochondrial distribution. Prohibitin was originally identified in mammalian cells as the product of a tumor suppressor gene (24, 29). In the yeast S. cerevisiae, prohibitin was proposed to play a role in cellular senescence (3). Contrary to earlier reports that a mutation in Phb1p increased replicative life span (3), Coates et al. (11) have recently presented data indicating that loss of either Phb1p or Phb2p decreased replicative life span. Our results demonstrating a modest reduction in replicative life span in both phb1 and phb2 mutants are consistent with the latter study. Prohibitin has now been identified in a wide range of species and localized to mitochondria in animal and plant cells (11, 20, 38). Our results have extended this localization to demonstrate that both Phb1p and Phb2p are integral proteins of the mitochondrial inner membrane in S. cerevisiae. Although we were not able to demonstrate the direct binding of Phb1p and Phb2p (as was shown by coimmunoprecipitation for prohibitin and prohibitin-related protein in animal cells [11]), our finding that the stability or maintenance of Phb1p depends on Phb2p (and vice versa) supports the idea that the two polypeptides function as a complex. In addition, each of the proteins appears to play distinct molecular roles, since overexpression of either species led to significant steady-state levels of the respective protein but could not complement the loss of the homolog. The specific cellular role or activity of prohibitin or prohibitin-related protein is unknown, but our identification of genetic interactions between PHB genes and a subset of mitochondrial inheritance components, along with the altered mitochondrial morphology of [rho2] phb1 and phb2 mutants, suggests that Phb1p and Phb2p function in the regulation of mitochondrial morphology and distribution. Loss of Mdm12p, Mdm10p, or Mmm1p results in giant mitochondria and severe defects in mitochondrial distribution, yet daughter buds occasionally receive small mitochondria, and cultures grow, although slowly, at 23°C (6, 9, 39). The “backup” mechanism that facilitates this inefficient mitochondrial inheritance may depend on the function of Phb1p and Phb2p, accounting for the inviability of the double-mutant strains. The molecular activity of these proteins may be further clarified by the identification of additional interacting components. ACKNOWLEDGMENTS We thank Jennifer Whistler (University of California, Berkeley) for advice on PCR-mediated gene disruption and Randy Hampton (University of California, San Diego [UCSD]) for the gift of plasmid pGTEP1. We thank Immo Scheffler (UCSD) for the sdh2-null mutant, Peter Thorsness (University of Wyoming, Laramie) for the yme2 mutant, Scott Emr (UCSD) for pep12 and vps18 mutants, and members of the Emr laboratory for their helpful advice. We are grateful to Randy Hampton, Anne-Laure Danquigny-Genestier, Harold Fisk, and Kelly Shepard for valuable comments on the manuscript. This work was supported by grant GM44614 from the National Institutes of Health. K.H.B. was supported by NIH grant GM16173 and a fellowship from the American Heart Association.
4051
REFERENCES 1. Allard, J. D., and K. P. Bertrand. 1992. Membrane topology of the pBR322 tetracycline resistance protein. J. Biol. Chem. 267:17809–17819. 2. Atencio, D. P., and M. P. Yaffe. 1992. MAS5, a yeast homolog of DnaJ involved in mitochondrial protein import. Mol. Cell. Biol. 12:283–291. 3. Austriaco, N. R., Jr. 1996. Review: to bud until death: the genetics of aging in the yeast, Saccharomyces. Yeast 12:623–630. 4. Baudin, A., O. Ozier-Kalogeropoulos, A. Denouel, F. Lacroute, and C. Cullin. 1993. A simple and efficient method for direct gene deletion in Saccharomyces cerevisiae. Nucleic Acids Res. 21:3329–3330. 5. Becherer, K. A., S. D. Rieder, S. D. Emr, and E. W. Jones. 1996. Novel syntaxin homologue, Pep12p, required for the sorting of lumenal hydrolases to the lysosome-like vacuole in yeast. Mol. Biol. Cell 12:283–291. 6. Berger, K. H., L. F. Sogo, and M. P. Yaffe. 1997. Mdm12p, a component required for mitochondrial inheritance that is conserved between budding and fission yeast. J. Cell Biol. 136:545–553. 7. Berger, K. H., and M. P. Yaffe. 1996. Mitochondrial distribution and inheritance. Experientia 52:1111–1116. 8. Broach, J. R., J. N. Strathern, and J. B. Hicks. 1979. Transformation in yeast: development of a hybrid vector and isolation of the CAN1 gene. Gene 8:121–133. 9. Burgess, S. M., M. Delannoy, and R. E. Jensen. 1994. MMM1 encodes a mitochondrial outer membrane protein essential for establishing and maintaining the structure of yeast mitochondria. J. Cell Biol. 126:1375–1391. 10. Burglin, T. R., and E. M. DeRobertis. 1987. The nuclear migration signal of Xenopus laevis nucleoplasmin. EMBO J. 6:2617–2625. 11. Coates, P. J., D. J. Jamieson, K. Smart, A. R. Prescott, and P. A. Hall. 1997. The prohibitin family of mitochondrial proteins regulate replicative lifespan. Curr. Biol. 7:607–610. 12. Daum, G., P. C. Bohni, and G. Schatz. 1982. Import of proteins into mitochondria. Cytochrome b2 and cytochrome c peroxidase are located in the intermembrane space of yeast mitochondria. J. Biol. Chem. 257:13028–13033. 13. Dowhan, W., C. R. Bibus, and G. Schatz. 1985. The cytoplasmically-made subunit IV is necessary for assembly of cytochrome c oxidase in yeast. EMBO J. 4:2764–2771. 14. Eveleth, D. D., and J. L. Marsh. 1986. Sequence and expression of the Cc gene, a member of the dopa decarboxylase gene cluster in Drosophila: possible translational regulation. Nucleic Acids Res. 14:6169–6183. 15. Field, J., J. Nikawa, D. Broek, B. MacDonald, L. Rodgers, I. A. Wilson, R. A. Lerner, and M. Wigler. 1988. Purification of a RAS-responsive adenylyl cyclase complex from Saccharomyces cerevisiae by use of an epitope addition method. Mol. Cell. Biol. 8:2159–2165. 16. Fisk, H. A., and M. P. Yaffe. 1997. Mutational analysis of Mdm1p function in nuclear and mitochondrial inheritance. J. Cell Biol. 138:485–494. 17. Hanekamp, T., and P. E. Thorsness. 1996. Inactivation of YME2/RNA12, which encodes an integral inner mitochondrial membrane protein, causes increased escape of DNA from mitochondria to the nucleus in Saccharomyces cerevisiae. Mol. Cell. Biol. 16:2764–2771. 18. Harlow, E., and D. Lane. 1988. Antibodies. A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 19. Hines, V., and G. Schatz. 1993. Precursor binding to yeast mitochondria. A general role for the outer membrane protein Mas70p. J. Biol. Chem. 268: 449–454. 20. Ikonen, E., K. Fiedler, R. G. Parton, and K. Simons. 1995. Prohibitin, an antiproliferative protein, is localized to mitochondria. FEBS Lett. 358:273– 277. 21. Ito, H., Y. Fukuda, K. Murata, and A. Kimura. 1983. Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 153:163–168. 22. Jensen, R. E., and M. P. Yaffe. 1988. Import of proteins into yeast mitochondria: the nuclear MAS2 gene encodes a component of the processing protease that is homologous to the MAS1-encoded subunit. EMBO J. 7:3863– 3871. 23. Lombardo, A., and I. E. Scheffler. 1989. Isolation and characterization of a Saccharomyces cerevisiae mutant with a disrupted gene for the IP subunit of succinate dehydrogenase. J. Biol. Chem. 264:18874–18877. 24. McClung, J. K., E. R. Jupe, X.-T. Liu, and R. T. Dell’Orco. 1995. Prohibitin: potential role in senescence, development, and tumor suppression. Exp. Gerontol. 30:99–124. 25. McConnell, S. J., L. C. Stewart, A. Talin, and M. P. Yaffe. 1990. Temperature-sensitive yeast mutants defective in mitochondrial inheritance. J. Cell Biol. 111:967–976. 26. McConnell, S. J., and M. P. Yaffe. 1992. Nuclear and mitochondrial inheritance in yeast depends on novel cytoplasmic structures defined by the MDM1 protein. J. Cell Biol. 118:385–395. 27. Mortimer, R. K., and R. K. Johnson. 1959. Lifespan of individual yeast cells. Nature 183:1751–1752. 28. Narasimhan, S., M. Armstrong, J. K. McClung, F. F. Richards, and E. K. Spicer. 1997. Prohibitin, a putative negative control element present in Pneumocystis carinii. Infect. Immun. 65:5125–5130. 29. Nuell, J. J., L. Stewart, V. Walker, C. M. Friedman, G. A. Wood, J. R. Owens, E. L. Smith, R. Schneider, R. T. Dell’Orco, C. K. Lumpkin, D. B. Danner,
4052
BERGER AND YAFFE
and J. K. McClung. 1991. Prohibitin, an evolutionarily conserved intracellular protein that blocks DNA synthesis in normal fibroblasts and HeLa cells. Mol. Cell. Biol. 11:1372–1381. 30. Peden, K. W. 1983. Revised sequence of the tetracycline-resistance gene of pBR322. Gene 22:277–280. 31. Riezman, H., R. Hay, S. Gasser, G. Daum, G. Schneider, C. Witte, and G. Schatz. 1983. The outer membrane of yeast mitochondria: isolation of outside-out sealed vesicles. EMBO J. 2:1105–1111. 32. Robinson, J. S., T. R. Graham, and S. D. Emr. 1991. A putative zinc finger protein, Saccharomyces cerevisiae Vps18p, affects late Golgi functions required for vacuolar protein sorting and efficient a-factor prohormone maturation. Mol. Cell. Biol. 11:5813–5824. 33. Rose, M. D., F. Winston, and P. Hieter. 1990. Methods in yeast genetics: a laboratory course manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 34. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 35. Schauer, I., S. Emr, C. Gross, and R. W. Schekman. 1985. Invertase signal and mature sequence substitutions that delay intercompartmental transport of active enzyme. J. Cell Biol. 100:1664–1675. 36. Schneider, B. L., W. Seufert, B. Steiner, Q. H. Yang, and A. B. Futcher. 1995. Use of polymerase chain reaction epitope tagging for protein tagging in Saccharomyces cerevisiae. Yeast 11:1265–1274. 36a.Sikorski, R. S., and P. Hieter. 1989. A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122:19–27.
MOL. CELL. BIOL. 37. Smith, B. J., and M. P. Yaffe. 1991. A mutation in the yeast heat-shock factor gene causes temperature-sensitive defects in both mitochondrial protein import and the cell cycle. Mol. Cell. Biol. 11:2647–2655. 38. Snedden, W. A., and H. Fromm. 1997. Characterization of the plant homologue of prohibitin, a gene associated with antiproliferative activity in mammalian cells. Plant Mol. Biol. 33:753–756. 39. Sogo, L. F., and M. P. Yaffe. 1994. Regulation of mitochondrial morphology and inheritance by Mdm10p, a protein of the mitochondrial outer membrane. J. Cell Biol. 126:1361–1373. 40. Stewart, L. C., and M. P. Yaffe. 1991. A role for unsaturated fatty acids in mitochondrial movement and inheritance. J. Cell Biol. 115:1249–1257. 41. Terashima, M., K. M. Kim, T. Adachi, P. J. Nielsen, M. Reth, G. Kohler, and M. C. Lamers. 1994. The IgM antigen receptor of B lymphocytes is associated with prohibitin and a prohibitin-related protein. EMBO J. 13:3782– 3792. 42. Towbin, H., T. H. Staehlin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350–4354. 43. Yaffe, M. P. 1991. Analysis of mitochondrial function and assembly. Methods Enzymol. 194:627–643. 44. Yaffe, M. P., R. E. Jensen, and E. C. Guido. 1989. The major 45 kDa protein of the yeast mitochondrial outer membrane is not essential for cell growth or mitochondrial function. J. Biol. Chem. 264:21091–21096. 45. Yaffe, M. P., and G. Schatz. 1984. Two nuclear mutations that block mitochondrial protein import in yeast. Proc. Natl. Acad. Sci. USA 81:4819–4823.
