Cytochrome Oxidase Assembly does not Require Catalytically Active Cytochrome c *

Cytochrome c oxidase (COX), the terminal enzyme of the mitochondrial respiratory chain, catalyzes the transfer of electrons from reduced cytochrome c to molecular oxygen. COX assembly requires the coming together of nuclear- and mitochondrial-encoded subunits and the assistance of a large number of nuclear gene products acting at different stages of maturation of the enzyme. In Saccharomyces cerevisiae, expression of cytochrome c, encoded by CYC1 and CYC7, is required not only for electron transfer but also for COX assembly through a still unknown mechanism. We have attempted to distinguish between a functional and structural requirement of cytochrome c in COX assembly. A cyc1/cyc7 double null mutant strain was transformed with the cyc1-166 mutant gene (Schweingruber, M. E., Stewart, J. W., and Sherman, F. (1979) J. Biol. Chem. 254, 4132-4143) that expresses stable but catalytically inactive iso-1-cytochrome c. The COX content of the cyc1/cyc7 double mutant strain harboring non-functional iso-1-cytochrome c has been characterized spectrally, functionally, and immunochemically. The results of these studies demonstrate that cytochrome c plays a structural rather than functional role in assembly of cytochrome c oxidase. In addition to its requirement for COX assembly, cytochrome c also affects turnover of the enzyme. Mutants containing wild type apocytochrome c in mitochondria lack COX, suggesting that only the folded and mature protein is able to promote COX assembly.


5
The respiratory competent haploid yeast W303-1B was transformed with the linear 1.5 kb EcoR1-HindIII fragment containing the cyc1 null allele. A uracil prototrophic transformant was verified by PCR 1 to be deleted for CYC1. This mutant, designated W303∆CYC1, was transformed with the linear fragment containing the cyc7 null allele.
Several uracil-and tryptophan-independent clones obtained from the transformation were confirmed by PCR to have the CYC7 deletion. One of the double mutants W303∆CYC1,7 was used for further studies.
Construction of the cyc1 mutant gene-The cyc1-166 mutation (8)  The fragment with the mutation was digested with a combination of Xma1 and KpnI and was substituted for the corresponding fragment in pCYC1/ST1 yielding pCYC1/W65S. The mutant gene was recovered as an XmaI-HindIII fragment and was transferred to YIp351 and YEp351 (14) yielding pCYC1/ST11 and pCYC1/ST12, respectively. The mutation was confirmed by sequencing of the insert in pCYC1/ST11. The mutant gene was integrated at the leu2 locus of W303-1B after linearization of pCYC1/ST11 at the ClaI site of the LEU2 marker in the plasmid.
Cytochrome oxidase assays-COX was assayed either spectrophotometrically by following the oxidation of ferro-cytochrome c at 550 nm (15)  Miscellaneous procedures-Standard procedures were used for the preparation and ligation of DNA fragments, and for transformation and recovery of plasmid DNA from E.
Mitochondrial protein synthesis was assayed in vivo in the presence of cycloheximide as described previously (11). Proteins were separated by PAGE in the buffer system of Laemmli (18). Cytochrome c was detected on Western blots using a rabbit polyclonal antibody raised against SDS-denatured yeast cytochrome c purchased from Sigma Chemical Co. (St Louis, MO). Antibody-antigen complexes were visualized by a secondary reaction with the Super Signal detection kit (Pierce, Rockford, IL). Alternatively Western blots were first treated with antibody against cytochrome c, followed by incubation with 125 Iprotein and quantitation of the signals with a Storm PhosphorImager (Molecular Dynamics, Inc, Sunnyvale, CA). Protein concentrations were determined by the method of Lowry et al. (19).

RESULTS AND DISCUSSION
Phenotype of the cyc1, and the cyc1/7 null mutants -Deletion of CYC1 reduces the rate but does not abolish growth of yeast on rich glycerol/ethanol medium (ref. 4, Fig. 1A). In contrast, deletion of both CYC1 and CYC7 completely blocks growth on the non-fermentable carbon sources (Fig. 1A). Wild type yeast has been shown to contain 95% iso-1-cytochrome c encoded by CYC1 gene and 5% iso-2-cytochrome c encoded by CYC7 (2). Western blot analysis of mitochondrial iso-2-cytochrome c expressed in the cyc1 null mutant is 12% of the total cytochrome c detected in the parental wild type strain ( Fig. 1B). This value is 2 -3 times higher than the 5% iso-2-cytochrome c reported previously. This could be due to a difference in the strains or carbon sources used in two studies. It is also possible that the absence of CYC1 results in an increased expression of iso-2-cytochrome c.  Fig.4). Although there is some reduction in Cox3p and some of the imported subunits (Cox4p and Cox5p), these constituents appear to be more stable (Fig. 4).
Phenotype of a cyc1/cyc7 double mutant expressing iso-1-cytochrome c with a W65S mutation -Several mutant alleles of cyc1 were previously shown to express catalytically inactive forms of iso-1-cytochrome c (4). One such allele ( cyc1-166), coding for a W65S amino acid substitution, produces a protein that is stable at 24 o C but is unable to mediate electron transfer from the bc 1 complex to COX (8). These properties of iso-1-cytochrome c with the W65S mutation made it possible to examine if COX assembly depends on a redox active protein.
A gene with the cyc1-166 mutation was made by PCR amplification of the appropriate coding region with the mutation in one of the primers. The mutant gene was integrated into the chromosomal DNA of the cyc1/cyc7 double mutant by targeted insertion at the leu2 locus (W303∆CYC1,7/ST11). The mutant was also transformed with the gene on a high copy plasmid (W303∆CYC1,7/ST12). Even though mitochondria from both transformants had immunochemically detectable cytochrome c ( Fig. 1B), neither the multicopy nor the integrated mutant gene was able to rescue the growth defect of the cyc1/cyc7 mutant strain on glycerol at either 24 o C or 30 o C (Fig. 1A).
Western blot analysis of mitochondria from the transformant with the integrated mutant gene indicated that the level of iso-1-cytochrome c is approximately 85% of the iso-1cytochrome c in wild type (Fig. 1). The mitochondrial concentration of iso-1-cytochrome c was somewhat lower in the transformant with the gene on a high-copy plasmid (68% of wild type). The failure of the W 65S mutant protein to support growth on non-fermentable carbon sources is in agreement with the previously noted deleterious effect of the mutation on the catalytic activity of iso-1-cytochrome c (8).
The cyc1-166 mutant was reported to contain spectrally detectable cytochrome c but at lower concentrations than wild type yeast (8). This was also true of the cyc1/cyc7 null strain transformed with the cyc1-166 gene in the integrative or episomal plasmid Since the amount of iso-2-cytochrome c in the mutant detected immunologically is 75% of wild type (Fig. 1B), only 30% of the W65S protein contains heme.
The status of COX in the cyc1/cyc7 null strain harboring the cyc1-166 allele was examined in several ways. Spectra of mitochondrial cytochromes indicated the presence of cytochromes aa 3 when the mutant protein was expressed either from the chromosomally integrated or plasmid-borne gene (Fig. 2). The presence of the integrated or episomal copy of the cyc1-166 allele restored cytochrome aa 3 to more than 50% of the level seen in wild type mitochondria ( Fig. 2A). The ability of iso-1-cytochrome c with the W65S mutation to rescue the COX deficiency of the cyc1/cyc7 strain was confirmed by enzyme assays and by Western analysis of COX subunits proteins, which indicated that the steady-state levels of the mitochondrially encoded Cox1p, Cox2p, and Cox3p were restored to nearly wild type levels (compare Figs. 4 and Fig. 6C).
Two different assays were used to measure COX activity. The first relied on the reduction of endogenous cytochrome c in mitochondria by ascorbate in the presence of TMPD. Using this assay the cyc1 null mutant had 40% of wild type COX, while no activity was detected in the cyc1/cyc7 double mutant (Table II). Predictably, the double mutant transformed with the cyc1-166 gene was also completely inactive in catalyzing ascorbic acid oxidation by oxygen (Table II). When the assays were repeated in the presence of exogenous cytochrome c, the specific activity of COX in the cyc1 mutant was raised to 70% of wild type, while the cyc1/cyc7 strain with the integrated cyc-166 gene was comparable to that of wild type (Table II). Similar results were obtained when the COX activity was a ssayed spectrophotometrically by measuring oxidation of substrate amounts of reduced cytochrome c (Table II). Surprisingly, the COX activity measured by both assays was lower in the transformant expressing the W65S protein from a multicopy plasmid. This was not true of the NADH-cytochrome c reductase activities, which were nearly the same in all the strains (Table II). The spectrum of mitochondrial cytochromes also showed a lower concentration of cytochromes aa 3 in the high copy transformant than in the strain with the integrated gene ( Fig. 2).
The lower COX activity in the multicopy transformant could be explained by a kinetic block due to limited accessibility or exchange of the wild type substrate cytochrome c with the W65S mutant protein. This was tested by depletion of the iso-1cytochrome c from mitochondria prior to the assay. Following sonic irradiation of wild type mitochondria in the presence of 1 M KCl, COX activity was reduced to 15% of the starting values (Table III). The specific activity returned to normal levels when the depleted mitochondria were assayed polarographically or spectrophotometrically in the presence of added cytochrome c. Under these conditions, however, the specific activity of mitochondria from the high copy transformant measured in the presence of cytochrome c was even lower after depletion (Table III). At present, therefore, the reason for the observed difference COX between the single and multicopy transformants is not clear.
Only a fraction of the W65S mutant iso-1-cytochrome c is in a protease protected compartment of mitochondria-As indicated, only 30% of the W65S mutant iso-1cytochorme c in mitochondria of cells with the integrated cyc1-166 allele contains heme.
The intra-mitochondrial location of the wild type and of the W65S mutant proteins was compared by testing their sensitivity to proteinase K in mitochondria and mitoplasts.
Most of the iso-1-cytochrome c in the strain with the chromosomally integrated cyc1-166 by guest on March 24, 2020 http://www.jbc.org/ Downloaded from allele was found to be susceptible to digestion by the protease in intact mitochondria (Fig.   5A). A small fraction corresponding to 10%, however, was in a proteinase K protected compartment. Even though all the protein sedimented with mitoplasts, it was completely sensitive to proteinase K.
In contrast, cytochrome c in wild type mitochondria is digested by the protease only when they are converted to mitoplasts (Fig. 5B). Sco1p, an inner membrane protein previously shown to face the intermembrane space (22), cytochrome b 2 , a soluble intermembrane marker, and α-ketoglutarate dehydrogenase, a soluble matrix protein showed the expected properties in the mitochondria and mitoplasts (Fig. 3B). These results indicate that some 90% of the mutant protein is associated with mitochondria in a manner that makes it accessible to proteinase K. Whether it is bound to the outer membrane or is only partially inserted into the intermembrane space has not been determined.
The small fraction of W65S protein resistant to proteinase K in mitochondria probably corresponds to mature protein located in the intermembrane space while the more abundant fraction digested by the protease, is probably mostly mutant apoprotein. Since transport of cytochrome c to the intermembrane space of mitochondria has been shown to be coupled to heme addition (23), the protease sensitivity of most of the W65S protein suggests notwithstanding the fact that the tryptophan at residue 65 is not covalently linked to heme, its replacement by a serine must reduce the efficiency of heme attachment to the apoprotein. Mutations in the cysteine ligands of heme have also been shown to impede apocytochrome c import into mitochondria in vitro (23). In these studies most of the mutant apocytochrome c was also found to cosediment with mitochondria even though it was not protected against the protease (23). incubation. Wild type cytochrome c was also reduced at the higher temperature, although the extent, even after overnight incubation, was much less (Fig. 6A).

Role of cytochrome c in stability of COX-
Incubation of the mutant cells at 37 o C caused greater than 90% loss of COX activity after 14 hours of incubation (Fig. 6A). The decrease in the specific activity was accompanied by a partial reduction in the cytochromes aa 3 absorption bands (Fig. 6B).
The loss of enzymatic activity, however, preceded the reduction in the cytochrome aa 3 bands. For example, even though less than 25% of the starting activity remained after 4 hours at 37 o C, the cytochromes aa 3 peaks were only marginally reduced ( Fig. 6A and   6B). Western analyses of COX subunits also indicated very partial losses of Cox1p, Cox2p, Cox3p and Cox5p during the first 4 hours at 37 o C (Fig. 6C). The most significant change was a large decrease of Cox1p in the mutant after the overnight incubation. In the wild type strain COX was highly stable at 37 o C, with more than 90% of enzyme and activity after 14 hr of incubation at the high temperature (Fig. 6A). Western analysis also failed to reveal any significant reductions in the COX subunits after different times at  (Fig. 6D).
The marked decrease of COX activity in mutant cells exposed to 37 o C for 4 hours even though cytochromes aa 3 and steady-state concentrations of COX subunits are unaffected during this period, suggests some more subtle changes in the quaternary structure or some other aspect of the enzyme. This is also seen in the wild type strain but to a lesser degree. These results suggest that the main role of cytochrome c is in assembly but that it also contributes towards the stability of the enzyme. The latter role is especially evident after prolonged incubation at 37 o C, which leads to turnover of Cox1p, more extensive loss of cytochromes aa 3 , and a virtually complete absence of enzyme activity (Fig. 6). The degradation of COX in mutant cells undergoing cytochrome c degradation may be similar to the loss of COX induced by mitochondrial cytochrome c release in wild type yeast committed to programmed cell death (24).  Fig 7A).

Apocytochrome c does not promote COX assembly-
To test if apocytochrome c is able to promote COX assembly, a cyc3 null mutant was transformed with CYC1 and CYC7 on high copy plasmids (ST13 and ST5 respectively. Both yeast transformants accumulate some apocytochrome c (Fig. 7B). The concentration of apocytochrome c i n mitochondria of the cyc3 mutant transformed with the respective genes was lower than in wild type and was further reduced after treatment with proteinase K. The mitochondrial concentrations of iso-1 or iso-2-apocytochrome c in the proteinase K protected compartment is comparable to the amount of iso-2cytochrome c present in the cyc1 mutant, which is able to express at least 70% of the normal amount of COX (Fig. 8). Neither of the two cyc3 transformants, however, contained COX either by spectral (Fig. 7 A) or enzymatic criteria (not shown).
Additionally, Western analysis of COX subunits indicated that despite the presence of apocytochrome c in mitochondria of the two transformants there was no increase in the steady-state concentrations of Cox2p and Cox3p. These results indicate that apocytochrome c is unable to promote COX assembly.
The present study shows that assembly of COX depends on the presence of cytochrome c in mitochondria even when the latter is unable to function in electron transport. The requirement for cytochrome c, therefore, is not related to either reduction or oxidation of some group in a subunit or assembly intermediate of COX. We estimate that the molar concentration of cytochrome c in mitochondria of wild type yeast is approximately t he same as that of COX. Assembly of COX, therefore, does not depend on stoichiometric concentration of cytochrome c. Iso-2-cytochrome c whose concentration in a cyc1 mutant is only 12% of the total amount of cytochrome c in wild type yeast is able to support the expression of 70% of normal amounts of COX. This lack of requirement for stoichiometry is also supported by the results obtained with the W65S mutant. Since the apoprotein cannot substitute for the mature cytochrome, the function of cytochrome c probably depends on a properly folded protein. This is consistent with a structural role in COX assembly. In addition to its requirement for assembly, the results of obtained with the W65S mutant exposed to 37 o C indicate that cytochrome c also affects turnover of COX.      Incorporation of 35 S-methionine into the mitochondrial translation products was allowed to proceed for 20 min at 30 o C as described previously (11). Excess 80 mM cold methionine and 4 µg/ml puromycin were added (0 t ime) and samples taken after 30 and 90 min of chase. Equivalent amounts of total cellular proteins were separated by SDS-PAGE on a 17.5% polyacrylamide gel, transferred to a nitrocellulose membrane, and exposed to X -ray film. The mitochondrially translated ribosomal protein Var1p, subunits 1 (Cox1p), subunit 2 (Cox2p) and 3 (Cox3p) of COX, cytochrome b (Cyt. b), and subunit 6 (Atp6p) and subunit 8 and 9 (Atp8/9) of the oligomycin-sensitive ATPase are identified in the margin.