Dual Mutations Reveal Interactions Between Components of Oxidative Phosphorylation in Kluyveromyces lactis

Corresponding author: G. D. Clark-Walker, Molecular Genetics and Evolution Group, Research School of Biological Sciences, The Australian National University, PO Box 475, Canberra, ACT, 2601, Australia., dcw{at}rsbs.anu.edu.au (E-mail)

Communicating editor: B. J. ANDREWS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Loss of mtDNA or mitochondrial protein synthesis cannot be tolerated by wild-type Kluyveromyces lactis. The mitochondrial function responsible for ⁰-lethality has been identified by disruption of nuclear genes encoding electron transport and F₀-ATP synthase components of oxidative phosphorylation. Sporulation of diploid strains heterozygous for disruptions in genes for the two components of oxidative phosphorylation results in the formation of nonviable spores inferred to contain both disruptions. Lethality of spores is thought to result from absence of a transmembrane potential, , across the mitochondrial inner membrane due to lack of proton pumping by the electron transport chain or reversal of F₁F₀-ATP synthase. Synergistic lethality, caused by disruption of nuclear genes, or ⁰-lethality can be suppressed by the atp2.1 mutation in the ß-subunit of F₁-ATPase. Suppression is viewed as occurring by an increased hydrolysis of ATP by mutant F₁, allowing sufficient electrogenic exchange by the translocase of ADP in the matrix for ATP in the cytosol to maintain . In addition, lethality of haploid strains with a disruption of AAC encoding the ADP/ATP translocase can be suppressed by atp2.1. In this case suppression is considered to occur by mutant F₁ acting in the forward direction to partially uncouple ATP production, thereby stimulating respiration and relieving detrimental hyperpolarization of the inner membrane. Participation of the ADP/ATP translocase in suppression of ⁰-lethality is supported by the observation that disruption of AAC abolishes suppressor activity of atp2.1.

FOLLOWING the discovery that mutations in Kluyveromyces lactis can convert this petite-negative yeast into a petite-positive form resembling Saccharomyces cerevisiae (CHEN and CLARK-WALKER 1993 ), progress has been made in understanding differences between the wild types of these yeasts (CHEN and CLARK-WALKER 1999 ). S. cerevisiae, as a representative of petite-positive yeasts, readily loses its mitochondrial DNA (mtDNA) to form respiratory-deficient petite mutants when treated with DNA targeting drugs (FERGUSON and VON BORSTEL 1992 ; CONTAMINE and PICARD 2000 ), whereas loss of mtDNA and mitochondrial protein synthesis is lethal for K. lactis (⁰-lethality; CLARK-WALKER and CHEN 1996 ; MURRAY et al. 2000 ). Mutations in K. lactis that suppress ⁰-lethality, initially termed mgi, for mitochondrial genome integrity, occur in the ATP1, -2, and -3 genes encoding the -, ß-, and -subunits of F₁-ATPase (CHEN and CLARK-WALKER 1995 , CHEN and CLARK-WALKER 1996 ; CLARK-WALKER et al. 2000 ). Hence K. lactis strains containing a suppressor mutation resemble S. cerevisiae by readily losing mtDNA and forming petite mutants when treated with ethidium bromide.

F₁-ATPase, the site for suppressor mutations, is associated with a mitochondrial inner membrane complex, F₀. Under conditions permitting oxidative phosphorylation, the F₁F₀-ATP synthase catalyzes the formation of ATP from ADP in response to proton translocation from the intermembrane space to the mitochondrial matrix (BOYER 1997 ; WEBER and SENIOR 1997 ; SARASTE 1999 ). The site for ATP synthesis is F₁, which is composed of 3-, 3ß-, 1-, 1-, and 1-subunits that in S. cerevisiae and K. lactis are all encoded by nuclear genes (ATTARDI and SCHATZ 1988 ; POYTON and MCEWEN 1996 ; CHEN et al. 1998 ).

The F₀ complex, on the other hand, contains three subunits encoded by mtDNA (ATP6, -8, and -9; ATTARDI and SCHATZ 1988 ; COX et al. 1992 ; POYTON and MCEWEN 1996 ; FOURY et al. 1998 ) as well as three major nuclear-encoded components (ATP4, -5, and -7; VELOURS et al. 1988 ; UH et al. 1990 ; NORAIS et al. 1991 ). Thus loss of mtDNA inactivates the F₀ complex. Furthermore, by disrupting the three nuclear genes for F₀ in K. lactis, it has been demonstrated that suppression of ⁰-lethality by mutant F₁ can occur in the absence of all six F₀ subunits (CHEN et al. 1998 ). In addition to encoding F₀ subunits, mtDNA contains genes for the electron transport complexes III, ubiquinol-cytochrome c reductase (CYB), and IV, cytochrome c oxidase (COX1, -2, and -3). Both complexes pump protons out of mitochondria during passage of electrons to oxygen. Therefore loss of mtDNA removes both the electron transport/proton pumping and ATP synthesis components of oxidative phosphorylation. Such a profound change can be tolerated by S. cerevisiae but not by wild-type K. lactis. Accordingly, we want to know what determines ⁰-lethality of K. lactis and how wild-type S. cerevisiae and F₁ mutants of K. lactis survive loss of oxidative phosphorylation.

In relation to the death of K. lactis on loss of mtDNA, it appears a priori that the mitochondrial genome contains one or more vital genes. Identification of the genes required for survival involved solving a conundrum. For instance, it is known that K. lactis can sustain deletions in mtDNA removing ATP9 (CLARK-WALKER et al. 1997 ) and, in a separate strain, COX1 and -2 (HARDY et al. 1989 ). Likewise, K. lactis can survive disruptions in nuclear genes that inactivate F₁ (ATP1, -2, and -3, ATP, and ATP) and F₀ (ATP4, -5, and -7), as well as genes of the electron transport chain [CYC1, QCR7, QCR8, PETIII, and COX18 (for review see CHEN and CLARK-WALKER 1999 )]. From these results it can be concluded that neither electron transport nor ATP synthesis/hydrolysis is essential for K. lactis. Consequently, it would seem that either K. lactis mtDNA contains a novel but essential gene not found in petite-positive yeasts or lethality could be due to simultaneous loss of genes for both electron transport and ATP synthesis/hydrolysis components of oxidative phosphorylation. The second possibility, termed the two-component model, has been investigated by a genetic test where disruptions in nuclear genes for electron transport and the F₀ complex are brought together in haploid spores by meiosis.

The second question addressed in this study concerns the mechanism of ⁰-lethality suppression in K. lactis. Suppression may occur by a process invoked to explain survival of S. cerevisiae petite mutants. In wild-type cells, the electrochemical potential across the mitochondrial inner membrane, , is generated by proton export during electron transport or, in some circumstances, by back pumping of protons through F₀ following hydrolysis of ATP by F₁-ATPase (MITCHELL 1979 ; NICHOLLS and FERGUSON 1992 ; BOYER 1997 ). Neither of these mechanisms for generation of is available to petite mutants. It has been proposed that the combined activities of F₁-ATPase and ADP/ATP translocase are responsible for maintenance of an adequate (DUPONT et al. 1985 ). ADP/ATP translocase, located in the mitochondrial inner membrane, is the third component of oxidative phosphorylation that, under aerobic conditions, exchanges ATP in the mitochondrial matrix for ADP in the cytosol (KLINGENBERG 1984 ). In petite mutants, it is thought that ADP/ATP translocase reverses to import ATP with four negative charges and export ADP with three negative charges (KLINGENBERG and ROTTENBERG 1977 ; LARIS 1977 ; DUPONT et al. 1985 ). The action of F₁-ATPase is to hydrolyze ATP to ADP in the absence of F₀. It is presumed that the combined action of the two enzymes provides sufficient electrogenic exchange of ATP for ADP to maintain at a level sufficient for protein import into mitochondria. In view of the postulated role for ADP/ATP translocase for survival of S. cerevisiae petite mutants, we investigated whether this enzyme participates in suppression of ⁰-lethality in K. lactis.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and media:
Yeast strains used in this study are listed in Table 1. Diploid strain CW75, homozygous for atp2.1, was obtained as a zygotic colony from CK429-5D a atp2.1 lysA1 ura3.1 crossed with CK429-7C atp2.1 ade1 ura3.1. Complete medium (GYP) contains 0.5% Bacto yeast extract, 1% Bacto Peptone, and 2% glucose. Glycerol medium (GlyYP) contains 2% glycerol in place of glucose. Minimal medium (GMM) contains 0.67% Difco (Detroit) yeast nitrogen base without amino acids and 2% glucose. Nutrients essential for auxotrophic strains were added at 25 µg/ml for bases and 50 µg/ml for amino acids. Ethidium bromide (EB) medium is GYP plus EB at 16 µg/ml. For sporulation of K. lactis, ME medium contains 5% malt extract and 2% Bacto agar. S. cerevisiae was sporulated on 1.5% K-acetate agar. Antimycin (Sigma, Castle Hill, New South Wales) was added to GYP medium after autoclaving to give a concentration of 1 µM. G418 (Geneticin; GIBCO BRL, Grand Island, NY) was added to GYP medium after autoclaving to give a concentration of 200 µg/ml.

Gene disruption:
Isolation of K. lactis genes and their disruption have been described in detail in previous publications (Table 1). Genes coding for the OSCP (ATP5) and d (ATP7) subunits of F₀ were disrupted by insertion of kan and URA3 (CHEN et al. 1998 ). The gene for cytochrome c (CYC1) was disrupted by insertion of URA3 (CHEN and CLARK-WALKER 1993 ) and PET111, encoding a translational activation factor for COX2, was disrupted by insertion of kan (COSTANZO et al. 2000 ). The unique gene for ADP/ATP translocase (AAC) in plasmid pUC19-KlAAC (VIOLA et al. 1995 ) and a copy of AAC disrupted by URA3 in plasmid pUC-KlAAC-URA were gifts from Cesira Galeotti. The disruption, in addition to removing the majority of the AAC gene, deletes 453 bp of downstream sequence containing many stop codons in all three reading frames.

Disruption of genes in K. lactis was performed by the one-step replacement method (ROTHSTEIN 1983 ). Verification that diploids were heterozygous for single-copy disruption was undertaken by digestion of genomic DNA followed by Southern blotting and hybridization according to the protocols described in previous publications. Construction of strains heterozygous for two disruptions was by transformation of a strain with a single disruption. Strain CK11/5 atp7::URA3/+ pet111 ::kan/+ was formed by disruption of ATP7 in CK11/4, and strain CK11/6 atp5::kan/+ cyc1::URA3/+ was obtained by disruption of ATP5 in CK11/3. Likewise, CW75/3 atp7:: URA3/+ pet111::kan/+ was obtained by disruption of PET111 in CW75/1. Strains CW75/5 aac::URA3/+ atp5::/+ and CW75/6 aac::URA3/+ pet111::kan/+ were isolated from CW75/4 aac::URA3/+ following transformation with the respective gene disruption cassettes.

Complementation of aac-disruptant lethality by wild-type AAC:
To confirm disruption of AAC, the wild-type gene was reintroduced into CK11/7. The 1661-bp MscI-HpaI fragment from pUC19-KlAAC containing KlAAC (Fig 3) was ligated with pCXJ30 cut with SmaI. pCXJ30 is a multicopy vector of 5579 bp containing the S11 K. lactis origin of DNA replication and a kan gene conferring resistance to G418 (X. J. CHEN, unpublished observations). The resulting plasmid, pCXJ30-AAC containing a single copy of AAC, was introduced into CK11/7 by selecting for resistance to G418 following transformation. A selected transformant, CK11/7-pCXJ30-AAC, was sporulated on ME agar containing 200 µg/ml G418 to maintain the plasmid.

View larger version (45K): In this window In a new window Download PPT slide   Figure 1. Response of K. lactis wild type (PM6-7A) and strains lacking components of F0, encoded by mtDNA, Gly-3.9 [ATP9] or nuclear DNA, CK350 (atp5::kan), and PH1 (atp7::kan) to antimycin (1 µM) in the presence of 2% glucose. Aliquots (10 µl) of cultures, serially diluted to 10-2–10-5, were added to plates that were incubated for 3 days at 28° before photography.

View larger version (71K): In this window In a new window Download PPT slide   Figure 2. Tetrad dissection of asci and phenotype determination of surviving colonies obtained from CK11/5 (atp7:: URA3/+ pet111::kan/+). After 4 days at 28°, colonies were replica plated to GYP medium containing G418 (200 µg/ml), GlyYP, and GMM minimal medium supplemented with adenine and lysine.

View larger version (21K): In this window In a new window Download PPT slide   Figure 3. Disruption of AAC in K. lactis. (A) Restriction site map of AAC and replacement of a BclI fragment containing the majority of the gene by URA3 from S. cerevisiae. The PCR-amplified DNA of 514 bp used as a probe contains a small portion of the AAC gene from the 5' region. (B) Autoradiograph of DNA from CK11 (wild type) and CK11/7, heterozygous for disruption of AAC. Genomic DNA was digested with MscI-HpaI or MscI before separation of fragments in 0.8% agarose. Hybridization was performed with the AAC-PCR product labeled with [32P]dATP. (C) Ascus dissection of CK11/7 (aac::URA3/+). The photo was taken after 4 days at 28°.

Ascus dissection and phenotype determination:
Asci were dissected with a Singer MSM Series 200 System (Singer Instruments, Somerset, UK) following brief treatment with Zymolyase (Seikaguku, Tokyo). Tetrad colonies were photographed after 4 days at 28°. Phenotypes were determined by replica plating to G418 medium for kan, GMM Ade Lys lacking uracil for URA3, and to GlyYP for glycerol growth. Genotypes have been inferred from phenotypes according to alleles listed in Table 1.

Preparation of AAC probe:
A 514-bp segment of the K. lactis AAC gene was amplified from pUC19-KlAAC by the polymerase chain reaction using primers 5' CGATAACTAACAACGTACC 3' in the forward direction and 5' GTAGACATCGACTAGACC 3' in the reverse direction. Location of the amplified segment is shown in Fig 3.

DNA manipulation:
Standard conditions, described in detail in previous publications (CHEN and CLARK-WALKER 1993 , CHEN and CLARK-WALKER 1995 , CHEN and CLARK-WALKER 1996 ), were used for DNA isolation, restriction enzyme digestion, electrophoresis, transfer of DNA to Nylon, hybridization, and autoradiography.

Kinetic parameters of F₁-ATPase:
Mitochondria were prepared from 750-ml cultures grown to stationary phase in GYP medium at 28°. Spheroplasts were produced by digestion with 10,000 units Zymolyase-20T (Seikagaku) for 60 min at 30°, resuspended in ice-cold 0.6 M sorbitol/10 mM TES [N-Tris (hydroxymethyl) methyl-2-aminoethanesulfonic acid] pH 7.4/0.1 mM 4-(2-aminoethyl)-benzene-sulfonyfluoride (AEBSF), an HCl protease inhibitor (Calbiochem, La Jolla, CA), and broken in a French press at 1000 p.s.i. Mitochondria were recovered by centrifugation at 9000 rpm for 20 min in a Sorvall SS34 rotor after prior removal of cellular debris by two rounds of centrifugation at 3500 and 5000 rpm for 5 min. Mitochondrial pellets were resuspended in 500–700 µl of 5 mM TES (pH 7.5), 15% w/v glycerol, 0.4 mM EDTA, 0.4 mM dithiothreitol containing 0.2 mM AEBSF protease inhibitor. ATPase activity was estimated according to LAW et al. 1995 . Freshly prepared mitochondria (50 µl) were added to 1.2 ml of 10 mM Tris-HCl (pH 8.2), 2 mM MgCl₂, 200 mM KCl, and to final ATP concentrations of 150, 200, 300, 500, 1000, and 3000 µM. Incubation was at 30° and 0.5-ml aliquots were removed at 1 and 2 min and added to 50 µl 3 M trichloroacetic acid. Inorganic phosphate concentration was measured with Sumner buffer against a standard curve and protein was estimated by Dc-Protein Assay (Bio-Rad Laboratories, Richmond, CA). ATPase activity is expressed as micromoles of ATP hydrolyzed per minute per milligram protein. V_max and K_m estimations, according to the Michaelis-Menton equation, were obtained from a Linweaver-Burk plot. Values are averages from three separate mitochondrial preparations in each case.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Sensitivity of F₀ mutants to antimycin:
K. lactis strains lacking F₀ subunits, coded by either the mitochondrial or nuclear genomes, were examined for their response to antimycin on glucose medium. Antimycin specifically inhibits electron transport but does not affect glycolysis (SLATER 1973 ). As shown in Fig 1, strains lacking F₀ subunit 9, (Gly-3.9), the OSCP protein coded by ATP5 (CK350), or subunit d, coded by ATP7 (PH1), are sensitive to 1 µM antimycin, whereas the wild type, PM6-7A, is less affected. Hence inhibition of respiration by antimycin is more severe for strains lacking a functional F₀ complex, which suggests that K. lactis may not tolerate simultaneous inhibition of both electron transport and F₁F₀-ATP synthase.

Colethality of disruptions for two components of oxidative phosphorylation:
In the absence of mutations in mtDNA that inactivate genes for both electron transport and the F₀ complex, an indirect genetic test involving disrupted nuclear genes was applied. Diploid strains of K. lactis were constructed by serial inactivation of genes required for the two components of oxidative phosphorylation. As a preliminary step in the construction of double-disruptant strains, the survival of spores from strains containing single disruptions was examined. Strains with disruptions in genes specifying electron transport components cytochrome c, cyc1::URA3 (CK11/3, 32/35 give 4:0 viable colonies) or the translational activation factor for Cox2p, pet111::kan (CK11/4, 36/39) show a slight decrease in spore survival compared to ones inactivated in F₀ subunits OSCP, atp5::kan (CK11/1, 38/39) or subunit d, atp7::URA3 (CK11/2, 39/40). The small decrease in spore survival revealed by these numbers provides a background for studies using double disruptants.

Following meiosis and ascus dissection of strains containing two disruptions, CK11/5 (atp7::URA3 pet111 ::kan) and CK11/6 (atp5::kan cyc1::URA3), phenotypes were examined by replica plating to medium containing G418, minimal medium lacking uracil, and GlyYP containing glycerol (Fig 2; Table 2). For the 95 tetrads dissected from CK11/5 the results are straightforward. There is no surviving colony containing both G418 resistance, as a marker of PET111 disruption, and growth in the absence of uracil, as a marker for ATP7 inactivation. Microscopic examination shows that spores inferred to contain two disruptions have germinated to form microcolonies containing between 100 and 1000 cells that do not proliferate. From the illustration it can be seen that large colonies growing on glycerol do not contain inactivated genes (as they require uracil and are sensitive to G418). Consequently the nonviable colonies in these tetrads can be inferred to contain two inactivated genes. Likewise, in tetrads showing three surviving colonies, the nonviable colony can be inferred to contain two inactivated genes on the basis of the phenotypes of the viable colonies. In tetrads lacking a Gly⁺ colony, G418⁺ and Ura⁺ phenotypes segregate from one another in the four viable spores. The number of asci showing 4:0, 3:1, or 2:2 survival of 16:59:20 is close to the ratio of 1:4:1 for parental ditype, tetratype, and nonparental ditype, respectively. As a 1:4:1 ratio indicates independent assortment, it can be concluded that the ATP7 and PET111 genes of K. lactis are unlinked. Another aspect of results from sporulation of CK11/5 is that all spores containing single disruptions survive.

Some nonviable colonies inferred to contain a single gene disruption have been observed on dissection of CK11/6. Examination of phenotypes from CK11/6, containing disruptions in ATP5 and CYC1, shows that five spores inferred to contain cyc1::URA3 do not survive (Ura⁺/G418^s; Table 2). The relevant asci show 3:1 and 2:2 ratios of viable to nonviable spores. However, the results from CK11/6 support the previous observation from CK11/5 that spores inferred to contain double disruptions do not survive. Thus not a single surviving colony, lacking both components of oxidative phosphorylation, has been found in 170 asci.

Suppression of double-disruptant lethality:
To ascertain whether a ⁰-lethality suppressor mutation can compensate for loss of both oxidative phosphorylation components due to disruption of nuclear genes, a diploid strain, CW75, was constructed that is homozygous for the suppressor allele atp2.1. Subsequently, a strain, CW75/3, that is heterozygous for disruptions in both ATP7 and PET111 was obtained. Sporulation and ascus dissection of CW75/3 revealed that 54 out of 60 tetrads produced four viable colonies, while 6 tetrads contained one nonviable spore (Table 2). Phenotypic analysis showed that two nonviable spores were inferred to contained atp7::URA3, one had pet111::kan, two contained both disruptions, and one spore lacked any disruption. From these results it is apparent that the atp2.1 allele can suppress colethality of atp7::URA3 and pet111::kan.

Although S. cerevisiae can survive lack of both components of oxidative phosphorylation due to loss of mtDNA, it has not been determined if it can tolerate loss of nuclear genes encoding these complexes. Consequently, to determine whether S. cerevisiae is directly analogous to ⁰-lethality suppressor mutants of K. lactis, segregants from a diploid strain, CWS7, heterozygous for disruptions in genes for cytochrome c₁, cyt1::kan, and F₀ subunit b, atp4::kan, were examined for survival. As anticipated, no lethality of colonies containing double disruptions was observed in 20 tetrads (not illustrated), thereby formally demonstrating that baker's yeast behaves as though it contains a suppressor analogous to atp2.1 of K. lactis.

Disruption of AAC is lethal in K. lactis:
Examination of whether ADP/ATP translocase is required for suppression of ⁰-lethality was performed in the first instance by disrupting one copy of the gene, AAC, in the wild-type diploid CK11. Correct disruption of AAC in strain CK11/7 was ascertained by hybridization with a PCR probe specific to the 5' region of the sequence that is retained in the disrupted gene (Fig 3). The wild type, CK11, has a single hybridizing fragment of 1661 bp, produced by MscI and HpaI digestion, whereas the heterozygous disruptant, CK11/7, contains an additional band of 2950 bp from MscI cleavage, corresponding to the expected size for disruption of AAC that has lost the HpaI site due to replacement by URA3 (Fig 3). Sporulation and ascus dissection of CK11/7 gave only two viable spores per tetrad (Fig 3). In some tetrads, minicolonies could be seen that did not grow on subculture. Replica plating showed that all viable spores in 51 tetrads examined were Ura^- and consequently lacked a disrupted AAC gene (not illustrated).

To confirm that lethality is a consequence of AAC disruption, the wild-type gene on a multicopy plasmid, pCXJ30, was transformed into CK11/7. Sporulation on ME agar containing G418 to maintain the plasmid, followed by ascus dissection, showed that some tetrads produced four or three viable colonies (Fig 4). Replica plating to minimal medium indicated that two or one of the colonies in these tetrads had a Ura⁺ phenotype representing aac::URA3. Colonies having a Ura⁺ phenotype also grew on G418, marking the retention of the plasmid containing the wild-type AAC gene. In addition, some Ura^- colonies also retained the plasmid as they grew on G418 (Fig 4). From these data it appears that lethality of haploids following meiosis of CK11/7 is a consequence of AAC disruption and is not due to inactivation of a gene at an extraneous site. This conclusion is supported by observations that all three AAC genes of S. cerevisiae can restore viability of K. lactis disrupted in AAC (G. D. CLARK-WALKER, unpublished observations).

View larger version (99K): In this window In a new window Download PPT slide   Figure 4. Tetrad dissection of asci and phenotype determination of colonies obtained from CK11/7 (aac::URA3/+) transformed with the multicopy plasmid pCXJ30-AAC containing the wild-type gene from K. lactis. After 4 days at 28°, colonies were replica plated to GYP medium containing G418 (200 µg/ml) and GMM minimal medium supplemented with adenine and lysine.

Suppression of aac-disruptant lethality by atp2.1:
To examine participation of the ADP/ATP translocase in suppression of ⁰-lethality, strain CW75, homozygous for atp2.1, was disrupted in one copy of AAC to produce CW75/4. Sporulation and ascus dissection of CW75/4 showed that 67 of the 78 tetrads gave four viable colonies while the other 11 had one nonviable colony. In each case the nonviable spore corresponded to the inferred presence of aac::URA3.

The survival of haploid strains containing a disrupted AAC gene in the presence of the suppressor allele atp2.1 permitted an assessment of whether such strains were dependent on mtDNA. Examination of whether four haploid strains from a single ascus could survive loss of mtDNA was determined by growth on a GYP plate containing 16 µg/ml ethidium bromide (not illustrated). Two strains, with wild-type AAC as determined by Southern blotting (not illustrated), and the suppressor allele atp2.1, grew on EB where they lose mtDNA, whereas the two segregants, with a disrupted AAC gene, did not grow. Furthermore, prolonged incubation of AAC-disruptants on EB did not lead to the appearance of resistant papillae. In other words, strains containing atp2.1 are not able to lose mtDNA when they lack ADP/ATP translocase.

Suppression of aac-disruptant lethality by atp2.1 requires both F₀ and electron transport:
The demonstration that mtDNA cannot be lost from haploid strains containing aac::URA3 and atp2.1 raised a question as to whether electron transport/proton pumping or F₁F₀-ATP synthase activities are required. To examine this question, two strains, CW75/5 and CW75/6, were made by disruption of one copy of ATP5 and one copy of PET111, respectively, in CW75/4 containing aac::URA3. Spore viability and phenotypes of surviving colonies following dissection of asci are shown in Table 2. In both cases there are no viable colonies containing two disruptions in 38 asci from CW75/5 and 74 asci from CW75/6. With CW75/5, 2 asci have a 3:1 ratio and 2 have a 2:2 ratio of viable to nonviable colonies with one nonviable spore inferred to contain only aac::URA3 in each case. Likewise, in tetrads derived from CW75/6 there are nonviable spores inferred to contain single disruptions in AAC: 2 with a 3:1 ratio and 8 with a 2:2 ratio. The 2 remaining tetrads have a nonviable spore inferred to contain a single disruption in PET111. Despite the decreased viability of spores containing single disruptions, these results are in accord with the number of nonviable spores derived from the parental strain, CW74/4, containing just an AAC disruption, where 11 nonviable spores were found in 78 tetrads (Table 2). However, the conclusion obtained from the above analysis is that both electron transport and F₁F₀-ATP synthase activities are required to support viability of strains lacking ADP/ATP translocase, even though the suppressor allele, atp2.1, is present.

Kinetic parameters of F₁-ATPase:
Previous studies have shown that F₁-ATPase activity does not correlate with ability of atp alleles to suppress ⁰-lethality (CLARK-WALKER et al. 2000 ). However, ⁰-suppression was shown to require some ATPase activity, as a double mutant, lacking ability to hydrolyze ATP, yet containing atp3.1, could not lose mtDNA. In view of these results it was decided to compare kinetic parameters of F₁-ATPase from wild-type and atp 2.1 mutant strains. For this comparison we used the previously constructed haploid isogenic strains, CW36-ATP2 containing wild-type ATP2 and CW36-H (atp2.1), which differ only by the introduced mutation (CLARK-WALKER et al. 2000 ). As shown in Table 3, F₁-ATPase of mitochondria from the atp2.1 strain has a fivefold lower K_m for ATP than the wild type, although the latter strain has a greater V_max. The increased affinity of mutant F₁ for ATP in the hydrolysis reaction suggests an explanation for suppressor activity as discussed below.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The assembly of two mutated genes into a haploid nucleus by meiosis has allowed us to examine their combined effect on dissected spores. By this means, it has been found that loss of the electron transport and F₀ components of oxidative phosphorylation are synergistically lethal in K. lactis. Inactivation of electron transport has been achieved either by disruption of the PET111 gene, encoding a translational activation factor for Cox2p synthesis (COSTANZO et al. 2000 ), or by disruption of the unique gene for cytochrome c, CYC1 (CHEN and CLARK-WALKER 1993 ). Elimination of a functional F₀ complex has been accomplished by disruption of either the ATP5 or ATP7 genes encoding the OSCP and d subunits, respectively (CHEN et al. 1998 ). By contrast, in S. cerevisiae, combined loss of electron transport and F₀ complexes, by either elimination of mtDNA or inactivation of nuclear genes, is not lethal. Hence the combined evidence supports the conclusion that loss of mtDNA from wild-type K. lactis is lethal because both electron transport and F₀ components of oxidative phosphorylation are missing. Lethality is thought to be due to lack of a mechanism for generating sufficient for import of proteins through the inner membrane.

In accord with expectations from previous studies, introduction of the atp2.1 mutation results in survival of spores containing double disruptions. A possible explanation for suppression of ⁰-lethality and double disruptant-lethality by this allele comes from the observation that mutant F₁-ATPase has a decreased K_m for ATP compared to wild type. In other words, the atp2.1 allele has greatly increased the affinity of F₁ for ATP. As a consequence it is likely that mutant F₁ can hydrolyze sufficient ATP to allow the ADP/ATP translocase to electrogenically exchange ADP in the matrix for ATP in the cytosol. Conversely, because the wild-type enzyme has a lower affinity for ATP, it may not be able to supply an adequate level of ADP to support exchange.

In the second part of this study we used coincidence of mutations to investigate the role of ADP/ATP translocase in suppression of ⁰-lethality. It has been proposed that if ADP/ATP translocase participates in suppression of ⁰-lethality then disruption of the AAC gene in K. lactis would not allow mtDNA to be lost from cells containing a F₁ suppressor allele because there would be no way to maintain . In other words, cells lacking ADP/ATP translocase would need to maintain by proton pumping from electron transport and/or reversal of F₁F₀-ATP synthase.

However, in an initial experiment we found that disruption of AAC is lethal and that introduction of the wild-type gene restores viability. This result was unanticipated as it has been reported that inactivation of AAC is nonlethal in haploid K. lactis (VIOLA et al. 1999 ), although attempts by another group to disrupt the gene in haploid K. lactis were not successful (TREZEGUET et al. 1999 ). A possible explanation for the conflicting results could rest on genetic differences between strains. Because the atp2.1 allele can suppress AAC-disruption lethality in our strain, it is possible that some genes required for energy transduction in mitochondria could be the most likely candidates for allelic differences.

Indeed, differences may exist between genes required for the efficiency of the F₁F₀-ATP synthase that could be connected to the mechanism of atp2.1 suppression of AAC-disruption lethality. For instance, according to observations from other organisms, it is believed that disruption of ADP/ATP translocase leads to hyperpolarization of the mitochondrial inner membrane (ESPOSITO et al. 1999 ; VANDER HEIDEN et al. 1999 ). Relief of hyperpolarization by atp2.1 is proposed to occur by partial uncoupling in the forward direction of ATP synthesis. Support for this view comes from the knowledge that atp suppressor mutants are all less efficient at making ATP than wild type as evidenced by their poorer growth on a nonfermentable substrate (CLARK-WALKER et al. 2000 ). Accumulation of ADP and lowering of the ATP/ADP ratio would stimulate respiration (NICHOLLS and FERGUSON 1992 ; ARNOLD and KADENBACH 1999 ; KADENBACH and ARNOLD 1999 ), thereby preventing the accumulation of electrons in the respiratory chain and the formation of detrimental reactive oxygen species (SKULACHEV 1996 ; FINKEL and HOLBROOK 2000 ). Both an intact respiratory chain and a functional F₁F₀-ATP synthase complex would be required to relieve hyperpolarization by mutant F₁ in agreement with experimental results. Thus the suppression of both ⁰ and AAC-disruption lethalities is based on the two functions of F₁. In the forward direction, in association with F₀, mutant F₁ is partially uncoupled, whereas in the reverse reaction it has a greater affinity for ATP. Both activities would result in a higher level of ADP in mitochondria compared to wild type, stimulating respiration in one case and promoting electrogenic exchange in the other instance.

	ACKNOWLEDGMENTS

We thank Cesira Galeotti for plasmids containing the AAC wild-type and disrupted genes of K. lactis and Lijun Ouyang for skilled technical assistance.

Manuscript received April 10, 2001; Accepted for publication August 3, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ARNOLD, S. and B. KADENBACH, 1999  The intramitochondrial ATP/ADP-ratio controls cytochrome c oxidase activity allosterically. FEBS Lett. 443:105-108[Medline].

ATTARDI, G. and G. SCHATZ, 1988  Biogenesis of mitochondria. Annu. Rev. Cell Biol. 4:289-333.

BOYER, P. D., 1997  The ATP synthase—a splendid molecular machine. Annu. Rev. Biochem. 66:717-749[Medline].

CHEN, X. J. and G. D. CLARK-WALKER, 1993  Mutations in MGI genes convert Kluyveromyces lactis into a petite-positive yeast. Genetics 133:517-525[Abstract].

CHEN, X. J. and G. D. CLARK-WALKER, 1995  Specific mutations in - and -subunits of F₁-ATPase affect mitochondrial genome integrity in the petite-negative yeast Kluyveromyces lactis. EMBO J. 14:3277-3286[Medline].

CHEN, X. J. and G. D. CLARK-WALKER, 1996  The mitochondrial genome integrity gene, MGI1, of Kluyveromyces lactis encodes the beta-subunit of F₁-ATPase. Genetics 144:1445-1454[Abstract].

CHEN, X. J. and G. D. CLARK-WALKER, 1999  The petite mutation in yeasts: 50 years on. Int. Rev. Cytol. 194:197-238.

CHEN, X. J., P. M. HANSBRO, and G. D. CLARK-WALKER, 1998  Suppression of ⁰ lethality by mitochondrial ATP synthase F₁ mutations in Kluyveromyces lactis occurs in the absence of F₀. Mol. Gen. Genet. 259:457-467[Medline].

CLARK-WALKER, G. D. and X. J. CHEN, 1996  A vital function for mitochondrial DNA in the petite-negative yeast Kluyveromyces lactis. Mol. Gen. Genet. 252:746-750[Medline].

CLARK-WALKER, G. D., F. FRANCOIS, X. J. CHEN, M. R. VIEIRA DA SILVA, and M. L. CLAISSE, 1997  Mitochondrial ATP synthase subunit 9 is not required for viability of the petite-negative yeast Kluyveromyces lactis. Curr. Genet. 31:488-493[Medline].

CLARK-WALKER, G. D., P. M. HANSBRO, F. GIBSON, and X. J. CHEN, 2000  Mutant residues suppressing ⁰-lethality in Kluyveromyces lactis occur at contact sites between subunits of F₁-ATPase. Biochim. Biophys. Acta 1478:125-137[Medline].

CONTAMINE, V. and M. PICARD, 2000  Maintenance and integrity of the mitochondrial genome: a plethora of nuclear genes in the budding yeast. Microbiol. Mol. Biol. Rev. 64:281-315[Abstract/Free Full Text].

COSTANZO, M. C., N. BONNEFOY, E. H. WILLIAMS, G. D. CLARK-WALKER, and T. D. FOX, 2000  Highly diverged homologs of Saccharomyces cerevisiae mitochondrial mRNA-specific translational activators have orthologous functions in other budding yeasts. Genetics 154:999-1012[Abstract/Free Full Text].

COX, G. B., R. J. DEVENISH, F. GIBSON, S. M. HOWITT and P. NAGLEY, 1992 The structure and assembly of ATP synthase, pp. 283–315 in Molecular Mechanisms in Bioenergetics, edited by L. ERNSTER. Elsevier Science, Amsterdam.

DUPONT, C. H., J. P. MAZAT, and B. GUERIN, 1985  The role of adenine nucleotide translocation in the energization of the inner membrane of mitochondria isolated from ⁺ and ⁰ strains of Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 132:1116-1123[Medline].

ESPOSITO, L., S. MELOV, A. PANOV, B. A. COTTREL, and D. C. WALLACE, 1999  Mitochondrial disease in mouse results in increased oxidative stress. Proc. Natl. Acad. Sci. USA 96:4820-4825[Abstract/Free Full Text].

FERGUSON, L. R. and R. C. VON BORSTEL, 1992  Induction of the cytoplasmic ‘petite’ mutation by chemical and physical agents in Saccharomyces cerevisiae. Mutat. Res. 265:103-148[Medline].

FINKEL, T. and N. J. HOLBROOK, 2000  Oxidants, oxidative stress and the biology of ageing. Nature 408:239-247[Medline].

FOURY, F., T. ROGANTI, N. LECRENIER, and B. PURNELLE, 1998  The complete sequence of the mitochondrial genome of Saccharomyces cerevisiae.. FEBS Lett. 440:325-331[Medline].

HARDY, C. M., C. L. GALEOTTI, and G. D. CLARK-WALKER, 1989  Deletions and rearrangements in Kluyveromyces lactis mitochondrial DNA. Curr. Genet. 16:419-427[Medline].

KADENBACH, B. and S. ARNOLD, 1999  A second mechanism of respiratory control. FEBS Lett. 447:131-134[Medline].

KLINGENBERG, M., 1984 The ADP/ATP carrier in mitochondrial membranes, pp. 511–553 in The Enzymes of Biological Membranes, Vol. 4, edited by A. N. MARTOUOSI. Plenum Press, New York.

KLINGENBERG, M. and H. ROTTENBERG, 1977  Relation between the gradient of the ATP/ADP ratio and the membrane potential across the mitochondrial membrane. Eur. J. Biochem. 73:125-130[Medline].

LARIS, P. C., 1977  Evidence for the electrogenic nature of the ATP-ADP exchange system in rat liver mitochondria. Biochim. Biophys. Acta 459:110-118[Medline].

LAW, R. H. P., S. MANON, R. J. DEVENISH, and P. NAGLEY, 1995  ATP synthase from Saccharomyces cerevisiae. Methods Enzymol. 260:133-163[Medline].

MITCHELL, P., 1979  Keilin's respiratory chain concept and its chemiosmotic consequences. Science 206:1148-1159[Free Full Text].

MURRAY, G. L., W. G. BAO, H. FUKUHARA, X. M. ZUO, and G. D. CLARK-WALKER et al., 2000  Disruption of the MRP-L23 gene encoding the mitochondrial ribosomal protein L23 is lethal for Kluyveromyces lactis but not for Saccharomyces cerevisiae. Curr. Genet. 37:87-93[Medline].

NICHOLLS, D. G., and S. J. FERGUSON, 1992 Bioenergetics 2. Academic Press, London.

NORAIS, N., D. PROME, and J. VELOURS, 1991  ATP synthase of yeast mitochondria. Characterization of subunit d and sequence analysis of the structural gene ATP7. J. Biol. Chem. 266:16541-16549[Abstract/Free Full Text].

POYTON, R. O. and J. E. MCEWEN, 1996  Crosstalk between nuclear and mitochondrial genomes. Annu. Rev. Biochem. 65:563-607[Medline].

ROTHSTEIN, R. J., 1983  One-step gene disruption in yeast. Methods Enzymol. 101:202-211[Medline].

SARASTE, M., 1999  Oxidative phosphorylation at the fin de siècle. Science 283:1488-1493[Abstract/Free Full Text].

SKULACHEV, V. P., 1996  Role of uncoupled and non-coupled oxidations in maintenance of safely low levels of oxygen and its one-electron reductants. Q. Rev. Biophys. 29:169-202[Medline].

SLATER, E. C., 1973  The mechanism of action of the respiratory inhibitor, antimycin. Biochim. Biophys. Acta 301:129-154[Medline].

TREZEGUET, V., I. ZEMAN, C. DAVID, G. J. LAUQUIN, and J. KOLAROV, 1999  Expression of the ADP/ATP carrier encoding genes in aerobic yeasts; phenotype of an ADP/ATP carrier deletion mutant of Schizosaccharomyces pombe. Biochim. Biophys. Acta 1410:229-236[Medline].

UH, M., D. JONES, and D. M. MUELLER, 1990  The gene coding for the yeast oligomycin sensitivity-conferring protein. J. Biol. Chem. 265:19047-19052[Abstract/Free Full Text].

VANDER HEIDEN, M. G., N. S. CHANDEL, P. T. SCHUMACKER, and C. B. THOMPSON, 1999  Bcl-x_L prevents cell death following growth factor withdrawal by facilitating mitochondrial ATP/ADP exchange. Mol. Cell 3:159-167[Medline].

VELOURS, J., P. DURRENS, M. AIGLE, and B. GUERIN, 1988  ATP4, the structural gene for yeast F₀F₁ ATPase subunit 4. Eur. J. Biochem. 170:637-642[Medline].

VIOLA, A. M., C. L. GALEOTTI, P. GOFFRINI, A. FICARELLI, and I. FERRERO, 1995  A Kluyveromyces lactis gene homologue to AAC2 complements the Saccharomyces cerevisiae op1 mutation. Curr. Genet. 27:229-233[Medline].

VIOLA, A. M., T. LODI, and I. FERRERO, 1999  A Klaac null mutant of Kluyveromyces lactis is complemented by a single copy of the Saccharomyces cerevisiae AAC1 gene. Curr. Genet. 36:29-36[Medline].

WEBER, J. and A. E. SENIOR, 1997  Catalytic mechanism of F₁-ATPase. Biochim. Biophys. Acta 1319:19-58[Medline].

This article has been cited by other articles:

				J. Satrustegui, B. Pardo, and A. del Arco Mitochondrial Transporters as Novel Targets for Intracellular Calcium Signaling Physiol Rev, January 1, 2007; 87(1): 29 - 67. [Abstract] [Full Text] [PDF]