Properties of a mitochondrial suppressor mutation restoring oxidative phosphorylation in a nuclear mutant of the yeast Schizosaccharomyces pombe.

The growth in glucose of the nuclear pleiotropic respiratory-deficient mutant pet1 of the “petite-negative” yeast Schizosaccharomyces pornbe is limited to a few cell generations after the addition of either 0.2 PM bongkrekic acid or 0.1 PM antimycin to the growth medium. The mutation sup2, which is of extra nuclear heredity, restores totally the resistance to both inhibitors for growth in glucose. Even though the cellular respiration and the content of cytochromes absorbing at 553.7, 560.5, and 605.8 nm at 77 K were almost totally restored in raffinose-grown pet1 sup-Z, the growth on glycerol is not restored in this strain. The oxidative phosphorylation and other mitochondrial activities such as oligomycin-sensitive ATPase, NADH-cytochrome c reductase, and cytochrome c oxidase activities, which are deficient in petl, are 30 to 70% restored in pet1 sup-2 grown in 1% glucose or 10% raffinose. A pronounced repression of respiration is observed when pet2 sup-2 is grown in 10% glucose. Under these conditions, pet2 sup-2 as well as pet1 exhibit a small but significant residual antimycin-sensitive respiration. Addition of antimycin to these cultures, while inhibiting the antimycinsensitive respiiation, allows continued expression of an antimycin-insensitive respiration, the rate of which is lower in pet1 (Q02 = 1.3 pl of O2 x min-’ x rng-I, dry weight) than in pet1 sup-2 (Q,, = 2.8). This respiration is unusual in that it is insensitive to cyanide and to hydroxamates but sensitive to azide. L-Malate as well as glucose may be utilized as respiratory substrate with both cytochromes b,,,,, and b,,:,, (77 K) being reduced in the presence of antimycin. Subsequent addition of azide oxidizes both cytochromes b even in the presence of cyanide plus antimycin, indicating oxidation-reduction equilibrium with a terminal antimycin and cyanide-insensitive oxidase. It is concluded that continuous cellular multiplication of S. pombe requires a critical level of intramitochondrial ATP which can be supplied either by oxidative phosphorylation or by the influx of the extramitochondrial ATP. In addition, a low but critical level of oxidations (Qo2 of about 2.0) is

The growth in glucose of the nuclear pleiotropic respiratory-deficient mutant pet1 of the "petite-negative" yeast Schizosaccharomyces pornbe is limited to a few cell generations after the addition of either 0.2 PM bongkrekic acid or 0.1 PM antimycin to the growth medium. The mutation sup-2, which is of extra nuclear heredity, restores totally the resistance to both inhibitors for growth in glucose. Even though the cellular respiration and the content of cytochromes absorbing at 553.7, 560.5, and 605.8 nm at 77 K were almost totally restored in raffinose-grown pet1 sup-Z, the growth on glycerol is not restored in this strain.
The oxidative phosphorylation and other mitochondrial activities such as oligomycin-sensitive ATPase, NADH-cytochrome c reductase, and cytochrome c oxidase activities, which are deficient in petl, are 30 to 70% restored in pet1 sup-2 grown in 1% glucose or 10% raffinose.
A pronounced repression of respiration is observed when pet2 sup-2 is grown in 10% glucose. Under these conditions, pet2 sup-2 as well as pet1 exhibit a small but significant residual antimycin-sensitive respiration. Addition of antimycin to these cultures, while inhibiting the antimycinsensitive respiiation, allows continued expression of an antimycin-insensitive respiration, the rate of which is lower in pet1 (Q02 = 1.3 pl of O2 x min-' x rng-I, dry weight) than in pet1 sup-2 (Q,, = 2.8). This respiration is unusual in that it is insensitive to cyanide and to hydroxamates but sensitive to azide. L-Malate as well as glucose may be utilized as respiratory substrate with both cytochromes b,,,,, and b,,:,, (77 K) being reduced in the presence of antimycin.
Subsequent addition of azide oxidizes both cytochromes b even in the presence of cyanide plus antimycin, indicating oxidation-reduction equilibrium with a terminal antimycin and cyanide-insensitive oxidase. It is concluded that continuous cellular multiplication of S. pombe requires a critical level of intramitochondrial ATP which can be supplied either by oxidative phosphorylation or by the influx of the extramitochondrial ATP. In addition, a low but critical level of oxidations (Qo2 of about 2.0) is * This work is Publication 1401 of the EURATOM Biology Division.
required which can be supplied by any combination of antimycin-sensitive and antimycin-insensitive respirations.
Regulation of the coordinated synthesis and the assembly of the components of the inner mitochondrial membrane is poorly understood. A possible approach to study this phenomenon is to investigate situations in which this regulation is upset. This might be the case in pleiotropic yeast mutants where one single-gene nuclear mutation produces several protein deficiencies in the inner mitochondrial membrane. Since their first observation by Sherman and Slonimski in 1964 (11, several nuclear pleiotropic respiratory-deficient mutants deficient in a set of proteins (such as cytochrome uuR, cytochrome b, and oligomycin-sensitive ATPase) containing mitochondrially synthesized peptides have been described in Saccharomyces cerevisiae (2-9) and Schizosaccharomyces pombe (lo-15). In S. cerevisiae, these nuclear pleiotropic respiratorydeficient mutants do often exhibit a high frequency of spontaneous induction of mitochondrial p-mutations which complicates their genetic and biochemical study (1,4,6,9). In contrast, no viable p-can be induced in the "petite-negative" yeast S. pombe (10,16,17). This property has greatly facilitated the study of the nuclear pleiotropic mutant S. pombe pet1 which was previously reported to be deficient in cytochrome ua3, cytochrome b, and the oligomycin-sensitive ATPase (13). The present paper describes two additional characteristics of petl. Its growth on glucose is inhibited by addition of either antimycin, a respiratory chain inhibitor (18) or of bongkrekic acid, an inhibitor of mitochondrial translocation of adenine nucleotides (19). Both traits are suppressed by the introduction into pet1 of the sup-2 mutation. The latter mutation was recently reported to be of mitochondrial heredity (20). We describe here that the suppressor mutation largely restores the deficient oxidative phosphorylation ofpetl With the possible exception of a rather similar situation in Paramecium recently reported (21), such mitochondrial-nuclear interaction has not been recognized so far. Furthermore, investigation of the effects of the suppressor mutation on the sensitivity of respiration to antimycin leads to the description of a new type of "alternative" respiration which is characterized by sensitivity to azide and by insensitivity not only to cyanide and antimycin but also to hydroxamates. The homogenate was centrifuged at 1000 x g for 5 min. In some cases, the pellet was ground again in the same conditions, The two supernatants were combined and centrifuged at 25,000 x g for 20 min; the pellet was suspended in one volume of 0.6 M sorbitol, 50 mM Tris/acetate, pH 7.5, and centrifuged at 1000 x g for 5 min. The latter supernatant was centrifuged at 25,000 x g for 20 min and the final pellet was suspended in 1 volume of 0.6 M sorbitol, 50 mM Tris/acetate, pH 7.5. When the ratios of ATP/oxygen were determined, mitochondria were isolated from prospheroplasts. In this case, the cells were grown on 1 g of glucose, 1 g of yeast extract/l00 ml, and HCl to bring to pH 4.5. The following procedure modified from that of Foury and Goffeau (221 was used. Two liters or 1.5 liters of culture medium were inoculated with 10" cells/ml and incubated at 30" in 6-liter Erlenmeyer flasks on a rotary shaker at 120 rounds/min. After 15 h of growth, 0.5 g of glucose/100 ml was added to the medium and the culture was further incubated for 30 min. A final concentration of 1.5 mM 2-deoxy-nglucose was then added to the culture and the cells were collected 10 min later by centrifugation at 3500 x g for 5 min and washed twice with 25 mM potassium phosphate buffer, pH 6.5, and 1 mM ethylenediaminetetraacetic acid. Thirty-five to fifty grams wet weight of cells were suspended in 350 to 500 ml of 1. addition of 0.18 pM antimycin to the culture medium, the cellular growth yield is decreased by further addition of bongkrekic acid, even though antimycin alone has no effect on cell multiplication. Similar situations have been described for Saccharomyces cerevisiae and have been interpreted as indicating the existence of a requirement for cell division of an unknown metabolite or process which is diluted out when the intramitochondrial ATP supply is limited (25). The sensitivity to bongkrekic acid of the growth on glucose of pet1 provides thus a convenient screening test to select suppressor mutations such as pet1 sup-2 able to grow on glucose in the presence of bongkrekic acid (Table I).
In addition to the inhibition by bongkrekic acid, the cellular division in glucose of the respiratory-deficient mutant pet1 is sensitive to the presence of antimycin in the culture medium. Fig. lZ3 shows that the addition of 0.1 PM antimycin to the culture medium limits the growth of pet1 in glucose to six generations compared to nine for the wild type. In the same conditions, the strain pet1 sup-2 behaves like the wild type. Table I also demonstrates marked synergistic effects of the combination of bongkrekic acid and antimycin in all strains. We therefore conclude that both inhibitors reach their intracellular target and that the suppressions of sensitivity to bongkrekic acid or antimycin in pet1 sup-2 is not due to a modification of the cellular permeability or to a modification of the (intracellular) inhibitors binding sites. The antimycin resistance of the growth in glucose of pet1 sup-2 was used to demonstrate that the sup-2 gene is of extranuclear (and probably mitochondrial) heredity (20). The resistance to antimycin given by sup-2 must not be confused with the ANTR trait recently described in S. pombe (26). The latter gene is also of mitochondrial heredity but is expressed by resistance to antimycin of the growth in glycerol while the suppressor mutations here described confer resistance of the growth in glucose but not that in glycerol (20).

Oxidative Phosphorylation in Wild
Type -Suppression of bongkrekic sensitivity in pet1 sup-2 suggests that in this strain, the mitochondria are independent of the cytosol for the supply of their intramitochondrial ATP and therefore that they are able to carry out oxidative phosphorylation. Until now it was not possible to measure mitochondrial ATP/oxygen in S. pombe because of the lack of availability of a method for isolation of intact mitochondria.
We have derived, therefore, a new method to prepare S. pombe prospheroplasts, rendering the cell wall fragile by addition of 2 deoxy-Dglucose to glucosegrowing cells. These prospheroplasts were further treated with a mixture of snail gut and Arthrobacter enzymes as described under "Materials and Methods." After osmotic lysis, fairly intact mitochondria with a respiratory control of 2.6 for NADH (see Fig. 2) and 1.3 for a-glycerophosphate were isolated from the wild type. Respiratory controls of 2.6, 2.4, 2.3. and 1.8 were obtained with a-ketoglutarate, citrate, malate, and succinate, respectively, provided that 1.6 mM pyruvate (which alone is poorly oxidized) was added to each of the above substrates (data shown for succinate in Fig. 2). where the phosphorylation Site I is absent (27). The high antimycin-sensitive oxidation rate of externally added NADH and its ATP/oxygen value of 1.75 suggest that, like in S. cerevisiae (28), an external NADH dehydrogenase is directly hooked to the second phosphorylation site in the inner mitochondrial membrane of S. pombe. On the other hand, the rather low ATP/oxygen ratios obtained with a-ketoglutarate are unexplained. Substrate level phosphorylation, however, is likely to be present because the oligomycin-insensitive ATP formed during the oxidation of cY-ketoglutarate is significantly higher than that with external NADH (Table II).

Oxidative
Phosphorylation in Mutants - Table II shows that no appreciable oxidations or phosphorylations are detected in pet1 mitochondria and that both oxidations and phosphorylations are very significantly restored in pet1 sup-2. The phosphorylation associated with the oxidation of a-ketoglutarate is largely oligomycin-insensitive in pet1 sup-2 suggesting that the substrate level phosphorylation is also functioning. It must be noted that no endogenous mitnchondrial ATP is detected in any of the strains when no substrates are added.

Presence
of Antimycin-insensitive and Antimycin-sensitive Respirations - Table III shows that a small antimycin-sensitive respiration (Qo, = 0.8) is observed inpetl grown in highly repressive conditions, such as exponential growth in 10% glucose. Under the same conditions, the rate of antimycin-sensitive respiration of pet1 sup-2 is significantly higher (QoZ = 1.6).
In addition, Table III shows that S. pombe develops an antimycin-insensitive respiration under conditions where the antimycin-sensitive respiration is reduced. The rates of the antimycin-insensitive respiration of pet1 sup-2 (Q,, = 2.8) is significantly higher than that of pet1 (Qoz = 1.3). Table IV shows that the antimycin-insensitive respiration observed in the mutants is not inhibited by 3 mM salicylhydroxamic acid, a typical inhibitor of the so-called "alternative-respiration" observed in many organisms and plant mitochondria (29)(30)(31)(32)(33)(34).  No endogenous ATP was detected in the absence of substrates.
In each assay, 1 to 3.0 mg of protein were used.
Oxidation rates Addition 972h-petf pet1 sup-2 nmol0, ATPIO nmol0, ATPIO run0102 ATPIO X min-' x min-' X minx mg-' x mg-' x mg-' 1.6 IIIM NADH io" Furthermore, while the "alternative respirations" described so far are insensitive to NaN:, (331, it appears that the antimycininsensitive oxygen uptake of S. pombe is inhibited by NaN,,. As illustrated in Fig. 3, the levels of sensitivity to azide of the antimycin-sensitive and antimycin-insensitive respirations are markedly different. Half-inhibition of the first one requires 7.5 pM azide compared to 175 @M for the latter. We conclude that cytochrome oxidase, which is sensitive to low concentrations of azide, is not involved in the antimycininsensitive respiration. This is supported by the observation that 1 mM KCN which totally inhibits the respiration of the wild type is not more inhibitory than antimycin in the mutants. The development of an antimycin-insensitive respiration is not specifically dependent on the pet1 mutations since Table  III shows that an antimycin-insensitive respiration also develops in the wild type grown in 10% glucose when the antimycinsensitive respiration has been blocked by the presence of antimycin during the growth as well as during the harvest and the respiration measurements. Under these conditions also, the however, is much more pronounced in nonrepressive conditions. Table III shows that, when the wild type is grown in 10% r&nose for 24 h, the antimycin-sensitive cellular Qol reaches 37.0, compared to 19.1 for exponential cells in 10% glucose. The mutant pet1 does derepress sig&cantly when grown on r&nose, but the rate of its antimycinsensitive respiration remains low (QOB = 3.9). When grown in 1% glucose, the antimycin-sensitive Qoz of pet1 is even lower (Qo, = 1.2) which explains why no appreciable oxidative phosphorylation was observed in mitochondria isolated from pet1 derepresses markedly when grown in rafinose since its antimycin-sensitive QoZ reaches 23.0 compared to 1.6 in 10% glucose. It also becomes clear that when the antimycin-sensitive respiration reaches a sufficient level, as in wild type in glucose or raffnose or pet1 sup-2 in rafflnose, the antimycininsensitive respiration disappears. Restoration of respiration in pet1 sup-2 under nonrepressive conditions is confirmed by mitochondrial enzyme-activity measurements.  is further illustrated by Fig. 4, showing absorption spectra of mitochondria isolated from raffinose-grown stationary phase cells. At liquid nitrogen temperature, the wild type exhibits the cytochrome c absorption bands (cu, = 548.7 nm and LYE = 543.8 nm), two cytochrome b bands absorbing at 553.3 and 560.5 nm and a cytochrome aa:, band at 605.8 nm. The cytochrome cl absorbing at 550.5 nm is masked by cytochrome c (see Ref. 35 for discussion of mitochondrial pigments in S. pornbe).
The cytochromes b and a are very low in pet1 On the other hand, the pet1 sup-2 absorption spectrum is similar to that of the wild type; the 553.3, 560.5, and 605.8 nm peaks being even more prominent in the suppressed strain than in the wild type.

Involvement of Cytochromes b in Antimycin-insensitive
Respiration -Antimycin-insensitive respiration rates of about 2.0plof0, x h ' x rng-', dry weight, were elicited not only by glucose but also by L-malate as respiratory substrates for wild type cells grown in glucose plus antimycin. Under the same conditions, oxalacetate, succinate. fumarate, citrate, pyruvate, L-and wlactate, n-ketoglutarate, glycerol, cu-glycerophosphate, ethanol, formate, acetate, glutamate, aspartate, phydroxybutyrate either were oxidized poorly or not at all. Fig.  5 (b and d) shows that the onset of antimycin-insensitive respiration by L-malate or glucose in wild type cells grown in glucose plus antimycin produces a marked increase of absorbance at about 554.5 and 560 nm (77 K). These absorption peaks A=005 J FIG. 4. Differential absorption spectra of mitochondrial fractions isolated from Schizosaccharomyces pombe wild type and mutants grown in nonrepressive conditions. Dithionite-reduced minus,oxidized spectra of mitochondrial fractions prepared by mechanical grinding were carried out at liquid nitrogen temperature in 2-mm cuvettes with the Aminco DW2 spectrophotometer. The strains 972h~ and pet1 sup-2 were oxidized with molecular oxygen.
To oxidizeprtl , 2 rnM potassium ferricyanide was used. The slit width was 1.5 nm for 972h-andpetl and 0.5 nm for pet1 sup-2. The scan speed was 1.0 nm x sm' and the chart speed was 25 nm x inch '. The protein concentrations were 6.3 mg x ml-' for 972hm, 5.8 mg x ml-' forpetl sup-2, and 8.2 mg x ml-' forpetl. d, 320 rnM L-malate, Q,,,, = 2.1; e, 320 mM L-ma&e for 5 min, followed by 0.5 mM KCN for-3 min; f, 320 mM Lmalate for 5 min, followed by 0.5 mM KCN for 3 min, followed by 0.5 mM NaN, during 3 min. Q,,.. < 0. oxidative phosphorylation rate of the wild type is obtained. This is however sufficient to restore resistance of growth to bongkrekic acid. The latter trait is therefore a powerful test for screening strains with very low or no oxidative phosphorylation.
Suppression of Antimycin Sensitivity of Growth on Glucose -Glucose-repressed pet1 and pet1 sup-2 exhibit total (antimycin-sensitive plus antimycin-insensitive) respiration rates of 2.1 and 4.4 ~1 of 0, x h-l x rng-' dry weight, respectively. They, respectively, carry out six and nine generations in 10% glucose supplemented with 0.18 pM antimycin. Such differences are clearly detected not only by the cell density of liquid cultures but also by the size of colonies on agar plates containing 3% glucose and 0.5 pg/ml of antimycin (20). The test of antimycin sensitivity of the growth on glucose is thus remarkably precise since it allows discrimination of strains with such small rates of respiration.
The presence in glucose-grown pet1 of a low residual antimycin-sensitive cellular respiration (Q,, = 0.8) strongly suggests that this process is the inhibitory target of antimycin when added in the culture medium. If so, this low antimycinsensitive respiration must be essential for the cellular multiplication in glucose of the mutant pet1 since addition of antimycin restricts its growth. A similar suggestion has previously been made for the pleiotropic S. cereuisiae pet 936 respiratory-deficient strain (4). This implies that in the wild type as well as in the suppressed strains pet1 sup-2 which grows well in the presence of antimycin, the inhibition of the antimycinsensitive oxygen uptake is compensated by a distinct process which cannot fully develop in pet1 This process might well be the antimycin-insensitive respiration which is functioning in glucose-grown pet1 at a lower rate (Q,, = 1.3) than that of glucose-grown pet sup-2 (Q,,. = 2.8) or of the wild type grown in the presence of antimycin (Q,, = 2.3).
The antimycin-insensitive respiration rates observed in respiratory-deficient conditions are rather low in S. pombe. The highest antimycin-insensitive Qo, obtained in this work, is 2.8 pl of 0, x lu-' x rng-', dry weight, which is 10 to 50 times lower than the so-called alternative respiration observed in other species such as Neurospora crassa (291, Candida lipolytica (30), Moniliella tomentosa (31), or Paramecium tetraurelia (32). Furthermore, in contrast to the above species, the S. pombe antimycin and cyanide-insensitive respiration is not sensitive to salicylhydroxamic acid and is sensitive to sodium azide. To our knowledge, the presence of such antimycincyanide-and hydroxamate-insensitive, azide-sensitive oxidation has not been reported so far. It must be mentioned that this new type of alternative respiration is not restricted to the strains described in this paper; we have also observed its presence in 14 distinct mitochondrial mit-strains of S. pombe kindly provided by Dr. K. Wolf and G. Seitz from Munich University.
The inhibition by antimycin of the growth of respiratorydeficient strains could be explained by a requirement for continuous growth in glucose of a minimum level of either a "normal" antimycin-sensitive respiration or of an antimycininsensitive, azide-sensitive oxygen uptake or of a combination of both. This is in agreement with inhibition of growth in glucose of S. pombe by anaerobioses (17) or by azide (data not shown). In the growth conditions used in this work, the minimum level of the total cellular respiration yielding continuous cellular multiplication in glucose is estimated to be about 2.0 ~1 of 0, x h-' x rng-', dry weight, e.g. slightly lower or equal to that of glucose-grownpetl (Q,, = 2.1) but higher than that observed in the same strain put in presence of antimycin (Q, = 1.4). If this hypothesis is correct, the limited growth of pet1 in glucose in the presence of antimycin is due to its insufficient level of the antimycin-insensitive, azide-sensitive respiration. Conversely, the suppression by sup-2 of the antimycin sensitivity of the growth on glucose ofpetl must then result from the potentiality of the suppressed strain to induce a sufficient antimycin-insensitive modified respiration when grown on glucose in the presence of antimycin. It is not excluded that this might result from a higher content in cytochrome(s) b.
Cytochromes b are generally considered not to participate in the cyanide-insensitive pathway in plants (see Ref. 33 for discussion). However, it has recently been suggested that, in Paramecium, the branching of the antimycin-insensitive pathway is posterior to both cytochrome b,, and cymchrome b,,, (77 K), the latter possibly being an autooxidizable pigment (49). The situation might be similar in S. pombe. Indeed, the fact that azide oxidizes cytochromes b5j,.5 and b,,, (77 K) which were both previously reduced by L-malate in the presence of antimycin plus cyanide, indicates that azide blocks the flow of reducing equivalents prior to the two cytochromes b and also that the latter are in equilibrium with a cyanideinsensitive oxidase. Therefore, we cannot exclude the possibility that at least one of the two cytochromes b is directly involved in the antimycin-sensitive respiration of S. pombe and is directly or indirectly controlled by the mitochondrial sup-2 mutation.