Suppressor Analysis of Mutations in the 5'-Untranslated Region of COB mRNA Identifies Components of General Pathways for Mitochondrial mRNA Processing and Decay in Saccharomyces cerevisiae

Corresponding author: Carol L. Dieckmann, Department of Biochemistry, University of Arizona, P.O. Box 210106, Tucson, AZ 85721-0106., dieckman{at}u.arizona.edu (E-mail)

Communicating editor: M. CARLSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The cytochrome b gene in Saccharomyces cerevisiae, COB, is encoded by the mitochondrial genome. Nuclear-encoded Cbp1 protein is required specifically for COB mRNA stabilization. Cbp1 interacts with a CCG element in a 64-nucleotide sequence in the 5'-untranslated region of COB mRNA. Mutation of any nucleotide in the CCG causes the same phenotype as cbp1 mutations, i.e., destabilization of both COB precursor and mature message. In this study, eleven nuclear suppressors of single-nucleotide mutations in CCG were isolated and characterized. One dominant suppressor is in CBP1, while the other 10 semidominant suppressors define five distinct linkage groups. One group of four mutations is in PET127, which is required for 5' end processing of several mitochondrial mRNAs. Another mutation is linked to DSS1, which is a subunit of mitochondrial 3' 5' exoribonuclease. A mutation linked to the SOC1 gene, previously defined by recessive mutations that suppress cbp1 ts alleles and stabilize many mitochondrial mRNAs, was also isolated. We hypothesize that the products of the two uncharacterized genes also affect mitochondrial RNA turnover.

GENE expression in yeast mitochondria is a coordinated process that requires the functions of both nuclearly and mitochondrially encoded proteins. Mitochondrial COB mRNA, which encodes cytochrome b, is a good model with which to study this type of fine regulation in Saccharomyces cerevisiae. Nuclear-encoded protein factors specific to the COB transcript have been shown to be required for processing of introns, production of the 5' end of the mRNA, mRNA stability, and mRNA translation. In this study, we have used genetic suppressor analysis to uncover factors involved in general pathways of mRNA decay in yeast mitochondria.

The nuclear-encoded protein Cbp1 stabilizes COB mRNA but none of the other six mitochondrial mRNAs (DIECKMANN et al. 1984 ). COB mRNA is undetectable in cbp1 mutants and thus the mutants are respiratory-deficient; they can grow on a fermentable carbon source, such as glucose, but not on a nonfermentable carbon source, such as glycerol. Our recent study supports the idea that Cbp1 protein physically interacts with the RNA to protect it from nucleolytic degradation (CHEN and DIECKMANN 1997 ). We showed that a CCG element at positions -944 to -942 of the COB 5'-untranslated region (5'-UTR; with start codon AUG of COB defined as +1) plays a critical role in COB mRNA stabilization. Mutation of any single nucleotide of this CCG eliminates COB mRNA accumulation and reduces the level of pre-COB RNA fivefold, a phenotype equivalent to that of cbp1 mutants. Because the single-nucleotide mutations affect COB mRNA levels drastically, the respiratory growth of the mutants on glycerol was affected. Mutant CAG (CCG CAG) is like mutant AAU (CCG AAU), and does not grow at all on glycerol medium at any temperature. However, mutants ACG (CCG ACG) and CCU (CCG CCU) grow slowly at 25 and 30°, and faster-growing pseudorevertants arise spontaneously (CHEN and DIECKMANN 1997 ).

In this article, we have characterized 11 of these pseudorevertants. The suppressor mutations are nuclearly encoded. One is a dominant, single nucleotide mutation in CBP1 (CHEN and DIECKMANN 1997 ). The other 10 suppressors are semidominant and fall into five different linkage groups. One of the linkage groups is defined by pet127 mutations. PET127 functions in 5' processing and turnover of mitochondrial RNAs (WIESENBERGER and FOX 1997 ). Another suppressor is linked to DSS1. DSS1 encodes one of the three protein components comprising the 3' 5' exoribonuclease activity in yeast mitochondria (referred to as mtEXO; DMOCHOWSKA et al. 1995 ). A mutation linked to the SOC1 locus was also recovered. SOC1 was originally defined by recessive mutations that suppress cbp1 ts alleles (STAPLES and DIECKMANN 1994 ). soc1 mutations increase the stability of many of the mitochondrial mRNAs. The other two linkage groups define loci of unknown function.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and media:
The S. cerevisiae strains used in this study are listed in Table 1. The media in which yeast or Escherichia coli strains were grown were as described previously (CHEN and DIECKMANN 1997 ).

Isolation of spontaneous pseudorevertants:
ACG and CCU mutant strains were patched on YEPD plates and incubated at 30° overnight. The patches were replicated onto YEPG plates and incubated at 30° for 6–10 days until faster-growing colonies arose in the patches. Five independent colonies were recovered from each of the ACG and CCU strains and purified. The pseudorevertants isolated from ACG were named "sup-a" followed by numbers 1 to 5; those from CCU were named "sup-u" followed by numbers 6 to 10. The pseudorevertants were grown in liquid YEPD medium overnight, diluted, and plated for single colonies on YEPD plates, which after 2 days incubation at 30° were replica-plated to YEPG medium. Over 95% of the colonies were able to grow on glycerol. This test guarantees that suppression is not due to heteroplasmic elements with mitochondrial DNA rearrangements (DIECKMANN et al. 1984 ; MULLER et al. 1984 ). Suppression is otherwise not maintained on nonselectable media because of segregation of the suppressing, rearranged molecule from the grande mitochondrial DNA.

Yeast mating type switches and changes of genetic markers:
A switch of mating type from MAT to MATa was implemented by transformation with the HO gene. Plasmid pRS413/HO was constructed, with the sequence of HO from -1360 to +2510 (RUSSELL et al. 1986 ) ligated into pRS413 (SIKORSKI and HIETER 1989 ). Transformants were selected on histidine dropout plates. The mating types were tested by mating with testers D237-10B/A21 (MAT) and CB11 (MATa). Only those that could mate with D237-10B/A21, but not with CB11, were MATa. To get rid of the HO plasmid after the mating type was switched, the strains were plated for single colonies on YEPD after overnight growth in liquid YEPD. His^- colonies were identified by replica-plating on histidine dropout plates. The new strains were named "a" followed by their original names.

To enable the selection of diploids when the isogenic pseudorevertants were crossed to each other, the auxotrophies of the MAT leu2 his3 strains were changed to MAT ura3 his3 by the following method. Plasmid U::L-F was constructed with LEU2 inserted into the ScaI site of URA3, which is carried in the pUC18 (NORRANDER et al. 1983 ) backbone. "F" refers to LEU2 in the same transcriptional orientation as URA3. A total of 1 µg of U::L-F was digested with HindIII and the digestion mixture was directly transformed into the pseudorevertants. Transformants were selected on leucine dropout plates. To confirm the disruption of the URA3 locus, chromosomal DNAs were prepared as described previously (ELION and WARNER 1986 ) except that the zymolyase buffer contained 1.2 M sorbitol, 75 mM KPO₄ (pH 7.5), 2.5% ß-mercaptoethanol, 0.2 mM EDTA (pH 7.5), and 1 mg of zymolyase (Seikagaku Corp., Ijamsville, MD). A total of 2 µg of each sample was digested with HindIII and separated on a 1% agarose gel. The Nytran membrane, onto which the DNA was transferred, was probed with a random-priming-labeled URA3-HindIII fragment. All strains showed a shifted URA3::LEU2 fragment (3.3 kb) compared with the wild-type URA3-HindIII fragment (1.2 kb). Pseudorevertants with changes of the auxotrophies have "u" added to their original names, respectively. aLL20/rho⁰ (MATa leu2 his3), u20/rho⁰ (MAT ura3 his3), or au20/rho⁰ (MATa ura3 his3) are derivatives of LL20/rho⁰ (MAT leu2 his3).

Backcrosses and linkage analysis:
The u pseudorevertant strains were crossed with aLL20/rho⁰. The diploids were selected on glucose medium without any amino acid (WO) + histidine medium and then grown on YEPG plates to test the dominance of the suppressor mutations by examining the respiratory capability of the sup/+ diploids at 30 and 25°. To examine whether each pseudorevertant strain contains a single nuclear mutation, the resulting diploids were sporulated in 1% potassium acetate solution at room temperature. After zymolyase digestion, the tetrads were dissected by micromanipulation. After 2 days, spores were tested for auxotrophies and respiration, at 30° for [rho^+ACG] and 33° for [rho^+CCU].

To cross the 10 pseudorevertants with each other, the MATa leu2 his3 strains were mated with the MAT ura3 his3 strains. Diploids were selected on WO + histidine medium and then sporulated. From 20 to 40 tetrads were examined for each of the crosses. To determine the linkage between "sup-a" suppressors and "sup-u" suppressors, u sup-u strains were made [rho⁰] (FOX et al. 1991 ) and then crossed with a sup-a strains containing [rho^+ACG]. The resulting diploids were sporulated as above.

Allele specificity tests of the suppressors:
To examine whether the suppressors cause a respiratory-deficiency phenotype of their own, to examine whether sup-a1, -a2, -a3, -a4, and -a5 could suppress [rho^+CCU] or [rho^+CAG], and to examine whether sup-u6, -u7, -u8, -u9, and -u10 could suppress [rho^+ACG] or [rho^+CAG], all 10 pseudorevertants were made [rho⁰] by ethidium bromide-mutagenesis (FOX et al. 1991 ). Lack of mitochondrial DNAs was confirmed by ultraviolet fluorescence microscopy of cells suspended in 1 µg/µl of 4', 6'-diamidino-2-phenylindole (DAPI). Using karyogamy-deficient strains carrying the wild-type mitochondrial genome or genomes with different mutations in their COB genes (JC3/rho⁺, JC3/rho^+ACG, JC3/rho^+CAG, and JC3/rho^+CCU strains), the nuclear genomes containing the 10 different suppressors were combined with various mitochondrial genomes by cytoduction (BERLIN et al. 1991 ).

Suppression of single-nucleotide COB mutants by overexpression of CBP1:
Plasmid pG60/T31 (DIECKMANN et al. 1984 ), which carries the entire CBP1 gene in the yeast 2µ vector YEp13, was transformed into the ACG, CAG, and CCU mutant strains. YEp13 with no insert was transformed as a control. Transformants were selected on leucine dropout plates, replicated onto YEPG plates, and grown at 30°.

Suppression of single-nucleotide COB mutants by soc1 and pet127:
To obtain strains with the mitochondrial genomes of ACG, CAG, and CCU in the mutant soc1 and pet127 nuclear backgrounds, the mitochondrial genomes [rho^+ACG], [rho^+CAG], and [rho^+CCU] were transferred into aRSY29/rho⁰ (STAPLES and DIECKMANN 1994 ) and pet127/rho⁰ via cytoduction using strains JC3/rho⁰ and JC7/rho⁰.

Linkage analysis of the suppressors with SOC1, PET127, SUV3, and DSS1 loci:
Because SOC1 is not yet cloned, the relationship of the suppressors to SOC1 was carried out by examining respiration of the diploids generated by crossing the suppressors (MAT leu2 his3 sup [rho⁰]) with a soc1[rho⁰] strain [aRSY29/rho⁰] (STAPLES and DIECKMANN 1994 ). Diploids were selected on WO + leucine medium, replicated onto YEPG plates, and grown at 30°. Diploids were sporulated and tetrads dissected.

To examine whether a wild-type copy of PET127, SUV3, or DSS1 could complement the suppressor functions, plasmids pGW694 (WIESENBERGER and FOX 1997 ), YEp24(T1), and pAD15 (DMOCHOWSKA et al. 1995 ) were transformed into the pseudorevertants, respectively. Transformants were selected on uracil or leucine dropout plates and then grown on glycerol plates (without uracil or leucine supplied) at 30°. To examine whether the suppressors are linked to the PET127, SUV3, or DSS1 loci, pseudorevertants were crossed with pet127/rho⁰, suv3/rho⁰, and dss1/rho⁰ separately. Diploids were selected and sporulated. The isogenic pet127/rho⁰ strain was obtained by disruption of the PET127 gene in the au20/rho⁰ strain. Plasmid p(pet127) was constructed by ligating the URA3 gene into SalI- and XbaI-digested pGW694, which truncates PET127 on the 5' end. The pet127::URA3 fragment was amplified by PCR with primers PET127-U2 (5'-cagggcacttgagagagcac-3') and PET127-L5 (5'-aagcgaatggtgtgatgaaatc-3') and transformed into the au20/rho⁰ strain. Plasmid Yep351-SUV3U (D. G. ROBERTS and C. L. DIECKMANN, unpublished results) was cut with BanI, and the 2.2-kb fragment containing a suv3::URA3 disruption allele was used to transform au20/rho⁰ to obtain the suv3/rho⁰ strain. Plasmid pKS(dss1), which carries a dss1::URA3 disruption allele in the pKS vector, was linearized, and transformed into au20/rho⁰ to make the isogenic dss1/rho⁰ strain. Ura⁺ transformants were selected in each of the three transformations on medium lacking uracil. Disruption of PET127, SUV3, and DSS1 was confirmed by Southern blot (data not shown).

Southern blot analysis and sequencing:
Mitochondrial DNAs were prepared from the 10 pseudorevertants (BONITZ et al. 1980 ). Approximately 1 µg of each sample was digested with MboI. The fragments were separated on a 1% agarose gel and transferred to Nytran membrane, which was probed with a random-priming-labeled COB -1350 +319 fragment (CHEN and DIECKMANN 1994 ). The COB sequence from -1360 to -648 from each of the 10 pseudorevertants was amplified by PCR and cloned into pBluescript KS (Stratagene, La Jolla, CA) and sequenced as previously described (CHEN and DIECKMANN 1997 ).

Primer extension analysis:
Total cellular RNAs were isolated from SUF63-F and all of the pseudorevertants (CAPONIGRO et al. 1993 ). Primer extension reactions and signal strength analyses were performed as described previously (CHEN and DIECKMANN 1997 ).

PCR, T/A cloning, and sequencing:
To examine the PET127 genes in the pseudorevertants bearing group II suppressors, chromosomal DNAs were prepared from these strains. The PET127 genes were amplified by PCR using primers PET127-U2 (5'-cagggcacttgagagagcac-3') and PET127-L2 (5'-cccaacgctgactactgtct-3'). The PCR products were directly ligated to the pGEM-T Easy (Promega, Madison, WI) vector. The cloned PET127 genes from the pseudorevertants were sequenced (UA facility), with primers PET127-U2, PET127-L2, PET127-U3 (5'-cgctacaaaaattgagagata-3'), and PET127-L3 (5'-cctcgtactcaggtgaca-3').

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Single-nucleotide mutations in COB mRNA allow recovery of spontaneous pseudorevertants:
In a previous study of the COB 5'-UTR, a 64-nucleotide sequence was defined as sufficient for COB mRNA stabilization (MITTELMEIER and DIECKMANN 1993 ). The SUF63-F strain has only the -962 to -898 sequence of the leader retained in a deletion from -1098 to -707 and has wild-type levels of COB mRNA. This strain has been used as the wild type in the present study. Within the 64-nucleotide cis-element, the sequence CCG located at positions -944 to -942 was shown to be particularly important, because mutation of any of the 3 nucleotides leads to degradation of COB messages and thus respiratory deficiency (CHEN and DIECKMANN 1997 ). The AAU mutant, in which the CCG is mutated to AAU, and the CAG mutant, with CCG mutated to CAG, disallow growth on glycerol medium at all temperatures. However, the ACG mutant, with CCG mutated to ACG, and the CCU mutant, with CCG mutated to CCU, grow slowly on medium containing a nonfermentable carbon source and give rise to faster-growing colonies (CHEN and DIECKMANN 1997 ). Ten independent strains carrying suppressors were isolated, 5 from the ACG mutant, sup-a1, sup-a2, sup-a3, sup-a4, and sup-a5, and 5 from the CCU mutant, sup-u6, sup-u7, sup-u8, sup-u9, and sup-u10 (Figure 1). All 10 suppressors were shown to be mitotically stable (see MATERIALS AND METHODS); suppression was not lost after overnight growth in rich glucose medium, which is nonselective for respiration. This implies that the suppressors are stable and heritable genomic mutations.

View larger version (35K): In this window In a new window Download PPT slide   Figure 1. Respiratory growth of the spontaneous pseudorevertants of the ACG and CCU mutant strains. All strains were streaked on YEPD plates and after overnight incubation replicated onto YEPG plates. Pictures were taken after 4 days of incubation on YEPG at 30°. SUF63-F is the wild-type control strain.

To determine whether the suppressor mutations were nuclear or in the mitochondrial genome, the mitochondrial genomes of the 10 pseudorevertants were transferred into the wild-type LL20 nuclear background via cytoduction using a karyogamy-deficient strain. The resulting strains were as respiratory-deficient as the original ACG and CCU mutants (data not shown). Thus, the suppressor mutations were in the nuclear genome.

To test dominance, the pseudorevertants (MAT ura3 his3) were outcrossed to the aLL20/rho⁰ strain and respiration of the diploids was examined. As shown in Figure 2, only the sup-a2/+[rho^+ACG] diploids showed respiratory competence similar to that of the haploid, which implies that sup-a2 contains a dominant suppressor. One might have expected that the +/+[rho^+ACG] and +/+[rho^+CCU] diploids would grow similarly to the respective +[rho^+ACG] and +[rho^+CCU] haploid strains. However, the diploids grew more slowly than the haploids on glycerol (i.e., +/+[rho^+ACG] < +[rho^+ACG] and +/+[rho^+CCU] < +[rho^+CCU]). Upon combination of any one of the other nine suppressors with wild-type (sup/+), a level of growth between that of the respiratory-incompetent +/+ diploids and diploid suppressor strains sup/sup was observed. Thus, except for sup-a2, all of the other nine suppressors are semidominant (+/+<sup/+ <sup/sup).

View larger version (63K): In this window In a new window Download PPT slide   Figure 2. Dominance test of the suppressors. The 11 suppressor strains, wild-type control strain SUF63-F, and the ACG and CCU mutant strains derived from SUF63-F were crossed to aLL20/rho0. All haploid and diploid strains were streaked on YEPD plates, and then replicated onto YEPG plates after overnight incubation. Pictures were taken after 6 days of incubation at 30° on YEPG. All strains are named by nuclear genome, followed by mitochondrial genome. (A) The mitochondrial genomes are [rho+ACG] if not specified. The plate was incubated at 28° instead of 30° to better show the growth difference between +/+[rho+ACG] diploids and sup/+[rho+ACG] diploids. (B) The mitochondrial genomes are [rho+CCU] if not specified.

To examine whether each of the pseudorevertants contained single nuclear mutations, the diploids were sporulated and 20–40 tetrads were dissected for each diploid. The tetrads were grown on YEPD and replica-plated onto YEPG medium. Eight of the diploid strains yielded tetrads that were 2:2 for respiration, demonstrating that these eight pseudorevertants contain single nuclear mutations. The majority of the sup-u9 diploids that were dissected showed 2:2 segregation; however, one diploid yielded two unlinked suppressors. The second suppressor was isolated from a nonparental ditype (NPD) tetrad of this sup-u9 diploid, and was named sup-u11. sup-u11 showed very similar features to the other five CCU pseudorevertants, containing a single nuclear semidominant mutation (Figure 2). Because the diploids do not respire well, we believe that the second suppressor arose during the selective growth in the acetate-containing sporulation medium, which requires the ability to respire.

Eleven suppressors define six linkage groups:
Because the mutations behaved in a dominant or semidominant fashion, it was not possible to use complementation analysis to group the suppressors. Thus, linkage analysis was required for determining groupings. To be able to cross the pseudorevertants to each other, the original MAT strains were converted to isogenic MATa strains by mating-type switching. To facilitate the selection of diploids on minimal medium, the URA3 locus of each of the MAT strains was disrupted by insertion of LEU2 (see MATERIALS AND METHODS). Diploids of MAT ura3 his3 sup1 x MATa leu2 his3 sup2 were selected on WO + histidine medium. The diploids from all crosses were respiratory competent, which is consistent with our observation that the suppressors are semidominant. The diploids were sporulated and tetrads were dissected. After scoring the tetrads for respiration from all crosses, the 11 suppressors were categorized into six linkage groups (Table 2). The dominant sup-a2 suppressor was shown to be a cbp1 mutation and has been described in detail previously (CHEN and DIECKMANN 1997 ). None of the other 10 suppressors was linked to CBP1, and they fell into five different linkage groups.

Suppressors show some allele specificity and have no phenotype of their own:
One possibility for gain-of-function suppressors is that they may have a phenotype of their own when separated from the original mutation (ADAMS and BOTSTEIN 1989 ). The semidominant suppressors do not have a respiratory-deficient phenotype of their own when combined with a wild-type mitochondrial genome (Table 3). This result is consistent with the hypothesis that these suppressors are loss-of-function mutations.

Some dominant reciprocal suppressors are also allele specific. To examine whether the suppressors of [rho^+ACG] can suppress [rho^+CAG] and [rho^+CCU] and whether the suppressors of [rho^+CCU] can suppress [rho^+ACG] and [rho^+CAG], different "sup" nuclear backgrounds were combined with different mitochondrial mutant genomes by cytoduction. The respiratory growth at 30° of all new strain combinations is summarized in Table 3. All of the [rho^+ACG] suppressors can suppress [rho^+CCU]; all of the [rho^+CCU] suppressors can suppress [rho^+ACG], though most of the suppressors are unable to suppress [rho^+CAG]. Thus, the suppression of COB 5'-UTR mutations by the suppressors isolated here is partially but not completely allele-specific. This result is also consistent with these mutations being loss-of-function alleles.

Overexpression of CBP1 suppresses the ACG and CCU mutations but not the CAG mutation:
Previous genetic evidence showed that Cbp1 interacts with COB mRNA (CHEN and DIECKMANN 1997 ). We wondered whether the interaction with mutated COB mRNA could also be suppressed by increasing the level of wild-type Cbp1 protein in vivo. The pG60/T31 plasmid, which contains the CBP1 gene in the 2µ YEp13 backbone, was transformed into the [rho^+ACG] (ACG), [rho^+CAG] (CAG), and [rho^+CCU] (CCU) strains. Transformation of YEp13 alone into these strains was used as a control. Figure 3 shows growth of the transformants on glycerol medium. Approximately 10-fold overexpression of CBP1 suppressed the respiratory deficiency of the [rho^+ACG] and [rho^+CCU] mutants, but not the [rho^+CAG] mutant. Consistent with our previous findings, this is another piece of evidence that the Cbp1 protein interacts with COB mRNA. Presumably the binding affinity between Cbp1 and COB 5'-UTR is lowered because of the mutations in CCG, but the affinity is strong enough that binding can be restored by increasing the level of Cbp1 severalfold in the cell.

View larger version (100K): In this window In a new window Download PPT slide   Figure 3. Suppression of the ACG, CAG, and CCU mutations by overexpression of CBP1. Plasmid p60/T31, which contains CBP1 in the 2µ vector YEp13 [lane "p(CBP1)"] and an empty YEp13 (lane "vector"), was transformed into control strain SUF63-F, and the mutants ACG, CAG, and CCU. The picture was taken after 6 days of incubation at 30° on YEPG.

RNA analysis reveals processing defects in some pseudorevertant strains:
To begin to analyze the function of the suppressors we examined COB mRNA in these strains. Mitochondrial function was induced by growing the pseudorevertants in YEPG media, and total cellular RNAs were prepared. To compare COB RNA processing in the pseudorevertants (all lanes labeled sup) with that in the wild-type strain (lanes labeled SUF63-F), primer extension reactions were used to examine the ratio of COB pre-RNA to mature message (Figure 4A). Strains containing group II suppressors (sup-a1, sup-u6, sup-u7, sup-u8) showed a striking RNA phenotype; COB precursor accumulated to a level 4- to 5-fold higher than that in the wild-type strain. However, no mature COB message was detected. This implies that COB pre-RNA can be translated because the strains respire (discussed below). Strains with suppressors in the other linkage groups accumulated both COB precursor and mature message, however, at levels 2- to 5-fold lower (precursor) and 7- to 33-fold lower (mature) than those in wild type (Table 4). Figure 4B shows the primer extension assay for the four spores from an NPD tetrad of sup-u9 (group III) and sup-u11 (group VI).

View larger version (50K): In this window In a new window Download PPT slide   Figure 4. Accumulation of COB pre-RNAs and mature messages in the suppressor strains analyzed by primer extension assay. Total cellular RNAs were prepared from each strain. A total of 10 pmol of COB6B and COX4242 primers was end-labeled and hybridized in a saturation ratio to 8 µg of RNA. cDNAs were synthesized by reverse transcriptase and separated on a sequencing gel. Signals from COX2 mRNA were used for normalization. (A) Strains sup-a1, -a2, -a3, -a4, -a5, and sup-u6, -u7, -u8, -u10. (B) Strains sup-u9 and sup-u11 shown are segregants from the NPD tetrad.

Survey of RNA turnover genes:
Because the phenotype of the group II suppressor strains was indicative of a loss-of-function in 5' trimming of COB pre-mRNA, we decided to survey genes encoding known components of mitochondrial mRNA processing and turnover machinery as likely candidates mutated in our pseudorevertants. PET127 was a good candidate because it was shown to have a role in RNA 5' trimming and turnover of a broad range of mitochondrial transcripts including COB (WIESENBERGER and FOX 1997 ). pet127 mutants accumulate COB pre-RNA and have no mature message, similar to the group II suppressor strains in this study. We also tested to see whether our suppressors are in SUV3 or DSS1. Along with a third, 75-kD protein, Suv3 and Dss1 constitute a mitochondrial 3' 5' exonuclease (mtEXO) activity (DMOCHOWSKA et al. 1995 ; MARGOSSIAN et al. 1996 ). Another good candidate is encoded by the nuclear SOC1 gene. soc1 mutations suppress cbp1^ts mutations and cause accumulation of precursor and mature COB mRNA in levels similar to those of groups III–VI suppressors. soc1 mutations increase RNA accumulation in mitochondria generally (STAPLES and DIECKMANN 1994 ). Therefore, all the candidates described above play a role in general RNA turnover pathways in yeast mitochondria.

Group II suppressors are pet127 mutations:
PET127 was originally identified by a suppressor of a mutation in the mitochondrial translational activator, Pet122, which is required for translation initiation of COX3 (HAFFTER and FOX 1992 ). Recently it was found that Pet127 is required for proper 5' trimming and turnover of mitochondrial RNAs (WIESENBERGER and FOX 1997 ). We found that our group II suppressor strains had a phenotype similar to that of the pet127/cox3 double mutants in that study. To examine the possibility that group II suppressors were in the PET127 locus, plasmid pGW694, which carries a wild-type copy of PET127 on the Yplac33 backbone, was transformed into all 11 suppressor strains containing the URA3::LEU2 disruption. Transformants of linkage group II strains were as respiratory-deficient as the original [rho^+ACG] and [rho^+CCU] mutants, whereas transformants of the other groups maintained respiratory competence.

To test whether a mutant pet127 nuclear background would suppress mutations in CCG, [rho^+ACG], [rho^+CAG], and [rho^+CCU] mitochondrial genomes were combined with a pet127 nuclear genome via cytoduction. Respiratory growth of recombined strains pet127/rho^+ACG, pet127/rho^+CAG, and pet127/rho^+CCU are shown in Figure 5A. The pet127 mutation did suppress the respiratory deficiency of the ACG and CCU mutants and allowed very slow growth of the CAG mutant at 30° (observed after 7 days of incubation).

View larger version (28K): In this window In a new window Download PPT slide   Figure 5. Suppression of the ACG, CAG, and CCU mutations in COB by pet127 (A) and soc1 (B) nuclear mutations. Pictures were taken after 4 days of incubation at 30° on YEPG. (A) Respiratory growth of the strains pet127/rho+ACG, pet127/rho+CAG, and pet127/rho+CCU. The pet127/rho+CAG strain started to show light growth after 7 days (data not shown). (B) Respiratory growth of the strains soc1/rho+ACG, soc1/rho+CAG, and soc1/rho+CCU. Growth of soc1/rho+CAG could not be observed after 10 days of incubation (data not shown).

To examine whether group II suppressors are indeed linked to PET127, they were crossed to an isogenic pet127/rho⁰ strain. The diploids were sporulated, and 20 tetrads were scored from each cross. All tetrads segregated 4⁺:0^- for respiration, implying that group II suppressors are tightly linked to PET127.

To examine the molecular nature of the pet127 genes in group II suppressor strains, the genes were amplified by PCR, cloned, and sequenced. Compared with the wild-type PET127 gene in the LL20 strain, all four group II suppressor strains were found to contain mutations in PET127. sup-a1 contained a single nucleotide deletion at position +585 (with A of start codon AUG defined +1), which results in a premature termination at codon 310 of the wild-type protein. sup-u6 contained a single nucleotide mutation of G T at position +1597, which changes codon 533 from Glu (GAA) to a stop codon (UAA). sup-u7 contained a single nucleotide deletion at position +997, which results in a peptide truncated at codon 345. Finally, sup-u8 contained three mutations; the most upstream one was a G T mutation at +391, which changes codon 131 from Glu to a stop codon. These data show that all group II suppressor strains contain either nonsense mutations or frameshift mutations in PET127, which are likely knockouts of gene function.

The suppressors are not suv3 mutations:
SUV3 was originally identified by a nuclear suppressor that affects a variety of post-transcriptional processes in yeast mitochondria (ZHU et al. 1989 ) and was recently identified as encoding a helicase component of the mitochondrial 3' 5' exonuclease (mtEXO; MARGOSSIAN et al. 1996 ). suv3 mutations lead to pleiotropic mitochondrial RNA defects including accumulation of excised group I introns, unspliced aI5ß- or bI3-containing precursors, and reduced COB and COX1 mRNA levels (CONRAD-WEBB et al. 1990 ). We did observe intron bI4- and bI5-containing processing intermediates in some of the group II and group III cob suppressor strains (Northern blot not shown), consistent with the observations of CONRAD-WEBB et al. 1990 . To study whether any of the suppressors might be in the SUV3 locus, plasmid YEp24(T1), which contains a wild-type SUV3 fragment in the YEp24 backbone, was transformed into all 11 pseudorevertants. Transformation did not cause any difference in respiratory growth of the strains, which implies that none of the suppressors is in the SUV3 locus. To confirm the transformation data, a suv3rho⁰ strain was mated to all of the suppressor strains. The diploids respired only as well as the ACG and CCU mutant strains when mated to wild-type rho⁰. Even though the diploids respired very weakly, 27–32 tetrads were dissected in crosses to representatives of groups III, IV, and V. No linkage was observed to the SUV3 locus.

The sup-u11 suppressor is tightly linked to DSS1:
DSS1 was identified as a multicopy suppressor of a suv3 disruption strain that cannot stably maintain mitochondrial genomes (DMOCHOWSKA et al. 1995 ). The Dss1 protein encodes another of the three protein components of the mtEXO complex and has sequence homology to E. coli 3' 5' exonuclease RNaseII. Plasmid pAD15, which carries a wild-type copy of the DSS1 gene on the YEp434 backbone, was transformed into all 11 suppressor strains. Introducing a wild-type copy of the DSS1 gene partially affected respiratory growth of group III and group VI suppressors, but did not affect suppressors in the other groups. Crosses of group III suppressors to the isogenic dss1/rho⁰ strain did not yield respiratory-competent diploids, which implies that group III suppressors are not dss1 mutations. However, the cross of sup-u11 to dss1/rho⁰ yielded respiratory-competent diploids. All 40 tetrads from this cross showed 2:2 segregation for respiration. Because knockout mutations in DSS1 lead to loss of mitochondrial DNA (DMOCHOWSKA et al. 1995 ), the results of the tetrad analysis indicate that sup-u11 is tightly linked to DSS1.

Group IV suppressor is a soc1 mutation:
soc1 mutations were identified previously as nuclear suppressors of cbp1^ts mutations (STAPLES and DIECKMANN 1994 ). soc1 behaves as a true recessive suppressor of the cbp1^ts mutations. cbp1^ts soc1 mutants showed accumulation of both COB precursor and mature message, though both in reduced levels compared with those in the wild-type strain. To examine whether any of the suppressors in the present study were soc1, the 10 semidominant suppressor strains were crossed with aRSY29/rho⁰, which is a soc1 mutant strain containing no mitochondrial DNA. All diploids showed the codominance phenotype associated with crosses of the suppressor strains collected in this study to each other. The diploids were sporulated and tetrads dissected from representatives of groups III, IV, and V. A total of 28 tetrads segregated 4:0 respiratory-competent to respiratory-deficient progeny, indicating that sup-a5 is tightly linked to SOC1.

To examine whether a soc1 allele originally isolated as a suppressor of a ts cbp1 mutation can suppress CCG cob mutations, soc1/rho^+ACG, soc1/rho^+CAG, and soc1/rho^+CCU strains were made via cytoduction. Respiratory growth of these strains is shown in Figure 5B. soc1 can indeed suppress the ACG and CCU mutations in COB mRNA, but not the CAG mutation. This finding strengthens the idea that SOC1 plays a role in mitochondrial RNA decay.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast mitochondrial RNAs are different from yeast cytoplasmic RNAs. They do not have the m⁷Gppp cap structure at the 5' ends, and polyadenylation of 3' ends has not been reported. They do have long AU-rich 5' and 3' UTRs and a common motif at the 3' end, a AAUAA(U/C)AUUCUU dodecamer sequence (OSINGA et al. 1984 ). Little is known about the turnover pathways for mitochondrial RNAs in yeast. A general 3' 5' exonucleolytic decay pathway is suggested by the presence of both a degradative enzyme and a common protection mechanism against its action. Three polypeptide units comprise the 3' 5' exonuclease activity, two of which are encoded by SUV3 and DSS1 (for a review, see MARGOSSIAN and BUTOW 1996 ). A three-polypeptide protein complex binds to the 3' dodecamer sequence of each of the mitochondrial mRNAs and protects them from 3' 5' degradation (MIN and ZASSENHAUS 1993 ). In addition to the 3' 5' pathway, there are emerging hints that a 5' end-dependent pathway may also be important for mRNA maturation and decay. In a recent study by WIESENBERGER and FOX 1997 , and in our current study, pet127 nulls were recovered as suppressors of both cox3 and cob 5'-UTR mutations that render the individual mRNAs unstable. A careful analysis showed that a pet127 knockout affects the stability and 5' end processing of many mitochondrial mRNAs and 15S rRNA (WIESENBERGER and FOX 1997 ). Whether 3' 5' or 5' end-dependent degradation is the major pathway of mitochondrial mRNA turnover has not been determined. Such an analysis will require pulse-chase experiments and analysis of decay intermediates.

COB mRNA clearly requires a specific stabilization system at the 5' end in addition to the general 3' end one. The nuclear-encoded protein Cbp1 is required uniquely for COB mRNA accumulation by protecting the 5' end. By deletion assay and site-directed mutagenesis (MITTELMEIER and DIECKMANN 1993 ; CHEN and DIECKMANN 1997 ), we located a cis-element in the 5'-UTR of COB that is important for Cbp1-dependent mRNA accumulation. All of our data support the hypothesis that Cbp1 recognizes and interacts with a CCG-containing element. Mutation of any of these three nucleotides leads to degradation of COB mRNA and respiratory-deficient cell growth. Nevertheless, two different single-nucleotide mutations in CCG allow suppressors to arise spontaneously.

To identify compensatory mutations in Cbp1 or other proteins that specifically interact with mutated COB 5'-UTR, and also to begin to identify components of the general 5' end-dependent degradation pathway, we used a genetic analysis of the spontaneous suppressors to identify factors that affect mRNA stability as an alternative to biochemical approaches, e.g., affinity chromatography and gel retardation assays, which have been used widely and successfully to identify RNA-binding proteins and enzymes (LEE et al. 1983 ; SUN and ANTONY 1996 ). Because none of the ACG and CCU suppressors are rearrangements of mitochondrial DNA, and all are nuclear mutations, genetic analysis of these suppressors had the potential to reveal protein-RNA interactions, protein-protein interactions, and novel proteins.

As it has been difficult to obtain purified Cbp1 protein, we have not been able to show direct binding of this protein with COB mRNA yet. Discovery of a dominant suppressor of a cob mutation in CBP1 in the present study provided the first evidence that Cbp1 interacts directly with COB mRNA (CHEN and DIECKMANN 1997 ). That a 5- to 10-fold overexpression of CBP1 suppresses two of the CCG mutations strengthens the idea that Cbp1 physically interacts with COB mRNA. Our model is that the conditional ACG and CCU mutations cause a decrease in the binding efficiency between Cbp1 and COB mRNA, and that overexpression or mutation of Cbp1 overcomes the reduction in affinity. This suppressor system is similar to suppression of mutations in actin, ACT1, by mutations in the actin-binding protein Sac6 (ADAMS and BOTSTEIN 1989 ).

Interestingly, this suppressor analysis revealed not only an allele of CBP1, a specific protector of COB mRNA, but also loci that affect mitochondrial mRNAs in general. Suppressor sup-u11 was found to be tightly linked to DSS1, which encodes a subunit of the 3' 5' exonuclease. Studies are being continued to identify the dss1 mutation in the sup-u11 strain. Because knockout mutations in DSS1 cause loss of the mitochondrial genome, the sup-u11 mutation is likely a missense mutation rather than a nonsense or frameshifting mutation. The linkage between sup-u11 and DSS1 strongly implies that the 3' 5' exonuclease is important in mitochondrial mRNA decay. It is curious that impairment of the 3' 5' exonuclease is implicated in suppression of a 5'-UTR mutation. Is there direct communication between the 5'- and 3'-UTRs, or is dysfunction of any major turnover enzyme able to raise the concentration of the mutant mRNA?

It was interesting to have isolated pet127 mutations as suppressors of COB mRNA mutations in this study. Only recently was it discovered that a pet127 knockout mutation has a pleiotropic effect on the processing and turnover of several mitochondrial RNAs (WIESENBERGER and FOX 1997 ). It was proposed that Pet127 plays a surveillance role in RNA metabolism (i.e., to remove unnecessary or incorrectly processed RNAs) or acts as a general RNA degradation factor. Our findings in this study favor the model that Pet127 is involved in a general 5' end-dependent RNA decay pathway in yeast mitochondria. The RNA phenotype of the pet127 suppressor strains revealed another unexpected phenomenon; they contain unprocessed pre-COB RNA only, but they respire as well as wild type. Thus, it is likely that the COB precursor is being translated in these strains. Provided with COB pre-RNA at a level as high as that of the mature COB message in a wild-type strain, the group II suppressor strains generate enough cytochrome b protein to be able to respire. In cbp1 mutants, which also do not contain any mature COB RNA, the COB precursor content is 20-fold lower than in the pet127 strains and is not enough to support respiration. Such levels of mature mRNA support respiratory growth. Either the precursor is translated less efficiently than the mature message, or Cbp1 is also required for translation.

A third previously characterized component important for mitochondrial turnover, Soc1, was also identified in the current study. soc1 mutations allow accumulation of higher than normal levels of many mitochondrial mRNAs and stabilize COB mRNA manyfold in a strain with a ts cbp1 allele (STAPLES and DIECKMANN 1994 ). The COB transcript pattern of groups I and III–VI suppressor strains is the same as the transcript pattern for soc1 suppression of cbp1^ts at the restrictive temperature (STAPLES and DIECKMANN 1994 ). Both COB pre-RNA and mature message accumulate to levels lower than those in the wild-type strain. The suppressors do not greatly increase the steady-state level of COB precursor, but permit the return of a low level of mature COB mRNA. Identification of the two unknown genes defined by the groups III and V suppressors will likely reveal key components of the mitochondrial RNA turnover machinery.

An interesting point in this study is that all of the suppressors described are semidominant (Figure 2). A prominent consequence of semidominance is that the suppressor phenotypes are additive in diploids; i.e., sup1+/+sup2 respires better than either sup1/+ or +/sup2. We propose two models for the semidominance feature. In model A, the Sup protein has multiple functions. In a sup strain, one of the functions is lost, but the other is not; e.g., the Sup protein encoded by sup in a sup/+ diploid has lost endo-/exonuclease activity, but it still retains mRNA-binding activity, which allows competition with or interruption of the binding and activity of wild-type Sup protein. In model B, the dosage of the Sup protein in the cell is a limiting factor. Because there is twice as much wild-type Sup protein in +/+ cells as in sup/+ cells, +/+ mRNAs degrade faster than sup/+ mRNAs, and sup/+ mRNAs degrade faster than sup/sup mRNAS. Model B explains why sup/+ diploids grow more like + haploids, rather than +/+ diploids, and explains the additive feature shown by the double suppressor mutants. Both models may be represented in the collection of the 10 semidominant suppressors. Because pet127 is a null and acts as a suppressor, and the other pet127 suppressors are likely to be nulls, the mechanism of suppression for the group II mutations is most likely described by model B.

	ACKNOWLEDGMENTS

We thank Dr. Tom Fox for sending the pet127 yeast strains and wild-type PET127 and disruption plasmids, Dr. Piotr Stepien for the DSS1 wild-type and disruption plasmids, and Dr. Alex Tzagoloff and Dr. Ronald Butow for the wild-type SUV3 plasmids. We thank Dr. Alison Adams, Dr. John Little, Dr. Richard Hallick, Dr. Bill Montfort, Dr. Karen Kindle, and Dr. Tim Ellis for critical reading of the manuscript. This research is supported by National Institutes of Health grant GM34893 to C.L.D. M.I-O. received partial support from Centro de Investigacion en Alimentacion y Desarrollo, A.C. Hermosillo, Sonora, Mexico.

Manuscript received July 2, 1998; Accepted for publication December 21, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADAMS, A. E. M. and D. BOTSTEIN, 1989  Dominant suppressors of yeast actin mutations that are reciprocally suppressed. Genetics 121:675-683[Abstract/Free Full Text].

ALEXANDER, N. J., P. S. PERIMAN, D. K. HANSON, and H. R. MAHLER, 1980  Mosaic organization of a mitochondrial gene: evidence from double mutants in the cytochrome b region of Saccharomyces cerevisiae.. Cell 20:199-206[Medline].

BERLIN, V., J. A. BRILL, J. TRUEHEART, J. D. BOEKE and G. R. FINK, 1991 Genetic screens and selections for cell and nuclear fusion mutants, pp. 774–792 in Guide to Yeast Genetics and Molecular Biology, edited by C. GUTHRIE and G. R. FINK. Academic Press, San Diego.

BONITZ, S. G., G. CORUZZI, B. E. THALENFELD, A. TZAGOLOFF, and G. MACINO, 1980  Assembly of the mitochondrial membrane system. Physical map of the Oxi3 locus of yeast mitochondrial DNA. J. Biol. Chem. 255:11922-11926[Abstract/Free Full Text].

CAPONIGRO, G., D. MUHLRAD, and R. PARKER, 1993  A small segment of the MATa1 transcript promotes mRNA decay in Saccharomyces cerevisiae: a stimulatory role for rare codons. Mol. Cell. Biol. 13:5141-5148[Abstract/Free Full Text].

CHEN, W. and C. L. DIECKMANN, 1994  Cbp1p is required for message stability following 5'-processing of COB mRNA. J. Biol. Chem. 269:16574-16578[Abstract/Free Full Text].

CHEN, W. and C. L. DIECKMANN, 1997  Genetic evidence for interaction between Cbp1 and specific nucleotides in the 5' untranslated region of mitochondrial cytochrome b mRNA in Saccharomyces cerevisiae.. Mol. Cell. Biol. 17:6203-6211[Abstract].

CONRAD-WEBB, H., P. S. PERLMAN, H. ZHU, and R. A. BUTOW, 1990  The nuclear SUV3-1 mutation affects a variety of post-transcriptional processes in yeast mitochondria. Nucleic Acids Res. 18:1369-1376[Abstract/Free Full Text].

DIECKMANN, C. L., G. HOMISON, and A. TZAGOLOFF, 1984  Assembly of the mitochondrial membrane system. Nucleotide sequence of a yeast nuclear gene (CBP1) involved in 5' end processing of cytochrome b pre-mRNA. J. Biol. Chem. 259:4732-4738[Abstract/Free Full Text].

DIECKMANN, C. L., T. J. KOERNER, and A. TZAGOLOFF, 1984  Assembly of the mitochondrial membrane system. CBP1, a yeast nuclear gene involved in 5' end processing of cytochrome b pre-mRNA. J. Biol. Chem. 259:4722-4731[Abstract/Free Full Text].

DMOCHOWSKA, A., P. GOLIK, and P. P. STEPIEN, 1995  The novel nuclear gene DSS-1 of Saccharomyces cerevisiae is necessary for mitochondrial biogenesis. Curr. Genet. 28:108-112[Medline].

ELION, E. A. and J. R. WARNER, 1986  An RNA polymerase I enhancer in Saccharomyces cerevisiae.. Mol. Cell. Biol. 6:2089-2097[Abstract/Free Full Text].

FOX, T. D., L. S. FOLLEY, J. J. MULERO, T. W. MCMULLIN, and P. E. THORSNESS et al., 1991  Analysis and manipulation of yeast mitochondrial genes. Methods. Enzymol. 194:149-165[Medline].

HAFFTER, P. and T. D. FOX, 1992  Suppression of carboxy-terminal truncations of the yeast mitochondrial mRNA-specific translational activator PET122 by mutations in two new genes, MRP17 and PET127.. Mol. Gen. Genet. 235:64-73[Medline].

LEE, J. C., B. HENRY, and Y. C. YEH, 1983  Binding of proteins from the large ribosomal subunits to 5.8S rRNA of Saccharomyces cerevisiae.. J. Biol. Chem. 258:854-858[Abstract/Free Full Text].

MARGOSSIAN, S. P. and R. A. BUTOW, 1996  RNA turnover and the control of mitochondrial gene expression. Trends Biol. Sci. 21:392-396.

MARGOSSIAN, S. P., H. LI, H. P. ZASSENHAUS, and R. A. BUTOW, 1996  The DExH box protein Suv3p is a component of a yeast mitochondrial 3'-to-5' exoribonuclease that suppresses group I intron toxicity. Cell 84:199-209[Medline].

MAYER, S. A. and C. L. DIECKMANN, 1989  The yeast CBP1 gene produces two differentially regulated transcripts by alternative 3'-end formation. Mol. Cell. Biol. 9:4161-4169[Abstract/Free Full Text].

MIN, J. and H. P. ZASSENHAUS, 1993  Identification of a protein complex that binds to a dodecamer sequence found at the 3' ends of yeast mitochondrial mRNAs. Mol. Cell. Biol. 13:4167-4173[Abstract/Free Full Text].

MITTELMEIER, T. M. and C. L. DIECKMANN, 1993  In vivo analysis of sequences necessary for CBP1-dependent accumulation of cytochrome b transcripts in yeast mitochondria. Mol. Cell. Biol. 13:4203-4213[Abstract/Free Full Text].

MÜLLER, P. P., M. K. REIF, S. ZONGHOU, C. SENGSTAG, and T. L. MASON et al., 1984  A nuclear mutation that post-transcriptionally blocks accumulation of a yeast mitochondrial gene product can be suppressed by a mitochondrial gene rearrangement. J. Mol. Biol. 175:431-452[Medline].

NORRANDER, J., T. KEMPE, and J. MESSING, 1983  Construction of improved M13 vectors using oligodeoxynucleotide-directed mutagenesis. Gene 26:101-106[Medline].

OSINGA, K. A., E. DE VRIES, G. VAN DER HORST, and H. F. TABAK, 1984  Processing of yeast mitochondrial messenger RNAs at a conserved dodecamer sequence. EMBO J. 3:829-834[Medline].

RUSSELL, D. W., R. JENSEN, M. J. ZOLLER, J. BURKE, and B. ERREDE et al., 1986  Structure of the Saccharomyces cerevisiae HO gene and analysis of its upstream regulatory region. Mol. Cell. Biol. 6:4281-4294[Abstract/Free Full Text].

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[Abstract/Free Full Text].

STAPLES, R. R. and C. L. DIECKMANN, 1994  Suppressor analysis of temperature-sensitive cbp1 strains of Saccharomyces cerevisiae: the product of the nuclear gene SOC1 affects mitochondrial cytochrome b mRNA post-transcriptionally. Genetics 138:565-575[Abstract].

SUN, X. L. and A. C. ANTONY, 1996  Evidence that a specific interaction between an 18-base cis-element in the 5'-untranslated region of human folate receptor-alpha mRNA and a 46-kDa cytosolic trans-factor is critical for translation. J. Biol. Chem. 271:25539-25547[Abstract/Free Full Text].

TEN BERGE, A. M. A., G. ZOUTEWELLE, and R. B. NEEDLEMAN, 1974  Regulation of maltose fermentation in Saccharomyces carlsbergensis.. Mol. Gen. Genet. 131:113-121[Medline].

TZAGOLOFF, A., F. FOURY, and A. AKAI, 1976  Assembly of the mitochondrial membrane system XVIII: genetic loci on mitochondrial DNA involved in cytochrome b biosynthesis. Mol. Gen. Genet. 149:33-42[Medline].

WIESENBERGER, G. and T. D. FOX, 1997  Pet127p, a membrane-associated protein involved in stability and processing of Saccharomyces cerevisiae mitochondrial RNAs. Mol. Cell. Biol. 17:2816-2824[Abstract].

ZHU, H., H. CONRAD-WEBB, X. S. LIAO, P. S. PERLMAN, and R. A. BUTOW, 1989  Functional expression of a yeast mitochondrial intron-encoded protein requires RNA processing at a conserved dodecamer sequence at the 3' end of the gene. Mol. Cell. Biol. 9:1507-1512[Abstract/Free Full Text].

This article has been cited by other articles:

				Z. Fekete, T. P. Ellis, M. S. Schonauer, and C. L. Dieckmann Pet127 Governs a 5' -> 3'-Exonuclease Important in Maturation of Apocytochrome b mRNA in Saccharomyces cerevisiae J. Biol. Chem., February 15, 2008; 283(7): 3767 - 3772. [Abstract] [Full Text] [PDF]

				C. Nouet, M. Bourens, O. Hlavacek, S. Marsy, C. Lemaire, and G. Dujardin Rmd9p Controls the Processing/Stability of Mitochondrial mRNAs and Its Overexpression Compensates for a Partial Deficiency of Oxa1p in Saccharomyces cerevisiae Genetics, March 1, 2007; 175(3): 1105 - 1115. [Abstract] [Full Text] [PDF]

				A. T. Rogowska, O. Puchta, A. M. Czarnecka, A. Kaniak, P. P. Stepien, and P. Golik Balance between Transcription and RNA Degradation Is Vital for Saccharomyces cerevisiae Mitochondria: Reduced Transcription Rescues the Phenotype of Deficient RNA Degradation Mol. Biol. Cell, March 1, 2006; 17(3): 1184 - 1193. [Abstract] [Full Text] [PDF]

				T. P. Ellis, M. S. Schonauer, and C. L. Dieckmann CBT1 Interacts Genetically With CBP1 and the Mitochondrially Encoded Cytochrome b Gene and Is Required to Stabilize the Mature Cytochrome b mRNA of Saccharomyces cerevisiae Genetics, November 1, 2005; 171(3): 949 - 957. [Abstract] [Full Text] [PDF]

				B. SCHAFER, M. HANSEN, and B. F. LANG Transcription and RNA-processing in fission yeast mitochondria RNA, May 1, 2005; 11(5): 785 - 795. [Abstract] [Full Text] [PDF]

				K. Krause, R. Lopes de Souza, D. G.W. Roberts, and C. L. Dieckmann The Mitochondrial Message-specific mRNA Protectors Cbp1 and Pet309 Are Associated in a High-Molecular Weight Complex Mol. Biol. Cell, June 1, 2004; 15(6): 2674 - 2683. [Abstract] [Full Text] [PDF]

				A. Dziembowski, J. Piwowarski, R. Hoser, M. Minczuk, A. Dmochowska, M. Siep, H. van der Spek, L. Grivell, and P. P. Stepien The Yeast Mitochondrial Degradosome. ITS COMPOSITION, INTERPLAY BETWEEN RNA HELICASE AND RNase ACTIVITIES AND THE ROLE IN MITOCHONDRIAL RNA METABOLISM J. Biol. Chem., January 10, 2003; 278(3): 1603 - 1611. [Abstract] [Full Text] [PDF]

				M. A. Islas-Osuna, T. P. Ellis, L. L. Marnell, T. M. Mittelmeier, and C. L. Dieckmann Cbp1 Is Required for Translation of the Mitochondrial Cytochrome b mRNA of Saccharomyces cerevisiae J. Biol. Chem., October 4, 2002; 277(41): 37987 - 37990. [Abstract] [Full Text] [PDF]