THE JOURNAL OF CHEMISTRY Vol. 266, No. 12, …THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 266, No. 12, Issue

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 266, No. 12, Issue of April 25, pp. 7688-7693, 1991 Printed in U. S. A.

Subunit Function in Eukaryote Cytochrome c Oxidase A MUTATION IN THE NUCLEAR-CODED SUBUNIT IV ALLOWS ASSEMBLY BUT ALTERS THE FUNCTION AND STABILITY OF YEAST CYTOCHROME c OXIDASE*

(Received for publication, October 26, 1990)

Robert Lightowlers, Zosia Chrzanowska-Lightowlers, Michael Marusich, and Roderick A. Capaldi From the Institute of Molecular Biology, University of Oregon, Eugene, Oregon 97403

Strains of the yeast Saccharomyces cerevisiae disrupted in YCOX4, the nuclear gene encoding cytochrome c oxidase subunit IV, do not assemble a functional or spectrally visible oxidase. We report the char- acterization of a yeast strain, RM1, expressing a mutated YCOX4 gene which is temperature sensitive for respiration at 37 "C, but incorporates cytochrome aas over all growth temperatures. The mutant enzyme is less stable than the wild type, with subunit IV readily proteolyzed without gross denaturation of the complex but with a concomitant loss of oxidase activity. When grown fermentatively at 37 "C, cytochrome c oxidase from the mutant strain had a turnover number of less than 3% of the normal complex, while K,,, values and subunit levels were comparable to normal. Thus alt- erations in subunit IV can perturb the enzyme structure and alter its catalytic rate, implying a role for this subunit in cytochrome c oxidase function as distinct from assembly.

Cytochrome c oxidase (EC 1.9.3.1) is the terminal member of the eukaryote respiratory chain. This enzyme couples the four-electron transfer from ferrocytochrome c to molecular oxygen with the translocation of protons across the mitochondrial inner membrane, generating a proton electrochemical gradient which is then used to synthesize ATP (for recent reviews, see Refs. 1-3).

The subunit composition of cytochrome c oxidase from several organisms and tissues has been reported (4-10). Fully functional three-subunit bacterial aa3-type oxidases have been isolated which contain the two copper and two heme pros- thetic groups common to the eukaryote oxidase (9,lO). These show homology to the three oxidase polypeptides encoded by the mitochondrial genome of eukaryotes. In addition to the mitochondrial gene products, eukaryotic cytochrome c oxidases also contain nuclear-encoded proteins, varying in number from four in Dictyostelium discoideum (6) to 10 in mammals (4, 5, 7, 8). The role of these nuclear-coded subunits remains poorly understood. One approach taken to define their functions has been to generate null mutants in these nuclear genes, utilizing the genetics of the yeast Saccharo- myces cereuisiae. Genes encoding each of the six nuclear- coded subunits have now been disrupted in yeast, establishing that all except the smallest polypeptide, subunit VIII, are required for maintaining respiration (11-13). In strains disrupted in the genes for subunits IV, V, VI, VII, and VIIa, the

* This work was supported by National Institutes of Health Grant HL 22050. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

enzyme is not assembled, making it difficult to decide whether these polypeptides function in electron transfer driven proton translocation and/or regulation of this function, in addition to being required for assembly or homeostasis of the complex.

To approach the issue of the role of the nuclear-coded subunits of cytochrome c oxidase, we have begun a search for respiratory-defective strains with specific mutations in each of these polypeptides. In this paper we describe a mutant in subunit IV of cytochrome c oxidase of S. cerevisiae which assembles the enzyme, but in which activity is markedly reduced along with proteolytic digestion of the mutant subunit. To our knowledge, this is the first detailed account of a missense mutation in any nuclear oxidase gene.

EXPERIMENTAL PROCEDURES

Strains, Plasmids, and Growth Media-All strains and plasmids used in this paper are documented in Table I. Rich media used to grow S. cereuisiae were: 1% (w/v) yeast extract, 2% (w/v) peptone supplemented with 2% (w/v) glucose (YEPD), 1% (w/v) raffinose, 50 mM potassium phosphate (pH 5.3) (YEPR) or 2% (v/v) glycerol, 2% (v/v) ethanol, 50 mM potassium phosphate (pH 5.3) (YEPGE). Se- lective media were synthetic dextrose media (14) supplemented with required amino acids to 20 pg/ml and uracil to 50 pg/ml. Escherichia coli strains were grown in standard L broth. Ampicillin was used as a bacteriostatin in all large scale preparations of S. cereuisiae (40 pg/ ml) and in selective media for E. coli (50 pg/ml).

Isolation of Yeast Mitochondria-Cells were grown at 30 or 37 "c to late log/early stationary phase in 4- or 10-liter batches of YEPGE or YEPR as indicated, with an aeration rate of 12 liters air/min. Cells were harvested by centrifugation at 3,500 X g in a Sorvall GS3 for 7 min, washed in distilled water, and pelleted by a similar centrifugation. Preparation of spheroplasts was essentially by the method of Daum et al. (15), except Zymolyase lOOT (ICN) was added to 10 mg/ 15 g of wet weight cells. Homogenization, differential centrifugation, and washing of mitochondria were as detailed (15). Depending on the source of sugar and size of preparation, a typical mitochondrial yield varied between 8 and 18 mg/lO g of wet weight cells. Mitochondria were resuspended in 0.6 M mannitol, 10 mM Tris-HC1 (pH 7.2), 1 mM phenylmethylsulfonyl fluoride to a final protein concentration of 10- 17 mg/ml and if not used immediately, were snap frozen in liquid nitrogen in aliquots of 0.1, 0.5, 1.0, or 2.0 mg and stored at -80 "C. No aliquot was reused after thawing.

Spectroscopic Analyses-All scans were performed on samples either at 20 "C, or in liquid nitrogen, using the SLM Aminco DW- 2000 dual wavelength spectrophotometer and low temperature acces- sory. Data were averaged from three repeated 2 nm/s scans through a slit width of 0.6 nm utilizing the slow data filter. For estimations of cytochrome concentrations (at 20 "C), samples (2 mg/ml) were solubilized in 100 mM potassium phosphate (pH 7.2), 25 mM NaCl, 1% lauryl maltoside. The following wavelength pairs and extinction coef- ficients were used for the Na dithionite-reduced minus K ferricyanide- oxidized spectra; cytochrome au3, A605 nm, 12.1 mM"/Cm (16); cytochrome c and cl, 550-540 nm, 20.1 mM"/cm (17); cytochrome b, 562-575 nm, 28.5 mM-l/Cm (18). All scans shown were taken of mitochondrial protein (0.6-1 mg/ml) at liquid nitrogen temperatures in 25 mM Tris-HC1 (pH 7.2), 25 mM NaCl, 50% (v/v) glycerol in the presence or absence of lauryl maltoside as indicated. Samples were

7688

Nuclear-encoded Subunit Mutant of Yeast Cytochrome c Oxidase 7689 TABLE I

Sources and genotypes of strains and plasmids used in this study

S. cerevisiae WD1

Y 15

JHRY99-

Y14 10a

Y 18

RM 1

Y5

MATa,p+leu2-112 COX1V::LEUQ his 3-11,15 ura3-328, 372 lac-

MATa as WDl:HIS+ containing plasmid pGALHO

MATa,p+leu2-3,112 ura3-

Diploid of RM1 X JHRY99-10a MATa,p"pepZl-l trp-1

A52,Ade-6

MATa,p+WDl containing plas-

MATa,p+WDl containing plasmid YpRLlO9

mid pMC4

Plasmids pMC4 COX IV, URA3 CEN4 ARSl

YpRLlO9 COX IV, URA3 CEN4 ARSl

pGALHO HISS, CEN4, ARSl

E. coli MC1061 F- hsdR hsdM+ araD139 (Aara-

ABOIC,leu) 7697 AlacX74 galU/K

endAl recAl gyrA96 thi-, hsdR17 (rk-,mk+) relAl supE44 X Alac-proAB F'ltraD36 proAB lacIq ZAM151

JM109

A kind gift from G. Schatz (28) ~0x4- strain used to select for missense mutants

Opposite mating type of WD1, produced by switching mating type of WD1 by method detailed in Kolodkin et al. (29)

Aggeler and Capaldi (13)

This paper This paper. Rho" tester strain constructed

from BJ2187 (a kind gift from F. Stahl, Eugene, OR)

This study

This study

A kind gift from G. Shatz (30). COX4 is expressed constitutively via the ADHl promoter

This study. Carries hydroxylamine-generated missense mutations in cox4

Plasmid used to facilitate mating switch of WD1 (29)

Casabadan and Cohen (31)

Yanisch-Perron et al. (32)

rapidly frozen and allowed to devitrify a t room temperature before low temperature analysis.

Enzymatic Assays-Succinate-cytochrome c reductase activity of 100 pg of mitochondrial protein was measured at 20 "C in 1 ml of 100 mM sodium phosphate buffer (pH 7.5), 15 mM Na succinate, 0.3 mM KCN and was monitored at 500 nm by the rate of reduction of oxidized yeast cytochrome c (40 pM) using the Ar550 = 21.1 mM"/cm (19). Oligomycin-sensitive ATPase activity was measured with an ATP-regenerating system as described in the presence or absence of 5 p~ oligomycin (20). Cytochrome c oxidase activities were measured at 20 "C in 1 ml of 25 mM Tris-HCI (pH 7.2), 25 mM NaCl, 0.1% lauryl maltoside. Unless otherwise stated, reduced cytochrome c was added to 6 p ~ . Mitochondrial protein varied from 5 to 100 pg depending on cytochrome a n 3 levels. The initial rate of oxidation at 550 nm was determined and expressed as mole of cytochrome c oxidized per mol of cytochrome aas/s, using the A,SSO noted above.

GdnHCl Denaturation-Mitochondrial protein was pelleted and solubilized in 2% lauryl maltoside, 25 mM NaCl, and 25 mM Tris- HC1 (pH 7.2), and GdnHC1' was added from a 6 M stock to the indicated concentration, maintaining a protein concentration of 2 mg/ml. The samples were left at 20 "C for 5 min and the levels of intact oxidase were calculated from the a-band absorbance at 605 nm uersus a base-line drawn from 590 to 620 nm of the reduced minus oxidized 20 "C spectrum for each sample. Levels of intact oxidase at each GdnHCl concentration are presented as a percentage of the calculated level for untreated preparations.

Immunochemical Methods-Rabbit antisera were raised against 100 pg of purified yeast oxidase holoenzyme, boosted three times with equal amounts of antigen over a 60-day interval. Monoclonal antibodies were produced in the Monoclonal Antibody Facility of the University of Oregon. Cell fusions and growth conditions are detailed in Marusich (21). Mice were injected with 50 pg of purified yeast oxidase (either in the detergent-solubilized form or after acetone denaturation) after being emulsified in Freund's complete adjuvant and were boosted twice with 50 pg of the same, emulsified in incom- plete adjuvant. Culture media from IgG-secreting cell fusions were initially screened against total mitochondrial protein, followed by a

' The abbreviation used is: GdnHCl, guanidine hydrochloride.

second screen against the purified enzyme. The specificity of monoclonal antibody to subunit IV was confirmed by binding to the protein from an E. coli strain constructed to overexpress this subunit. The monoclonal antibody to subunit I1 was obtained by screening for antibodies which reacted with purified yeast oxidase in Western analysis by staining a band at position I1 but did not bind mitochondrial protein from the pa tester strain Y18. Monoclonal antibodies were concentrated from large scale culture media (22).

For Western analysis, 100 pg of mitochondrial protein was dissociated in an equal volume of 10% sodium dodecyl sulfate, 8 M urea, 0.1% bromphenol blue, 100 mM dithiothreitol, and separated through a high resolution polyacrylamide gel as described by Kadenbach et al. (23). For polyclonal antibody binding, protein was transferred onto nitrocellulose (24), and for monoclonals onto polyvinylidine difluoride membrane (25) (Millipore). Antibody binding and visualization used the goat anti-rabbit or goat anti-mouse conjugated alkaline phosphatase assay system (Bio-Rad), with the primary polyclonal antibody at a titer of 1:40 and with the monoclonals at 1:2000.

Miscellaneous-All chemicals used were of the highest purity avail- able. Reduction and oxidation of yeast cytochrome c (Sigma Type VIII-b) for use in enzyme assays were produced by the addition of sufficient Na ascorbate or K ferricyanide, respectively, to 10-pmol aliquots, followed by separation from reductant/oxidant by passage through a Sephadex G-25 column pre-equilibrated in 20 mM K phosphate (pH 7.0). Aliquots were stored at -80 "C. DNA sequence analysis was by the dideoxy chain termination method of Sanger et al. (26), after subcloning of relevant endonuclease-generated fragments into M13mp19. The U. S. Biochemical DNA sequencing kit utilizing the T4 Sequenase polymerase was employed. For hydroxylamine mutagenesis of plasmid pMC4, 50 pg of DNA was heated at 65 "C in 50 mM K phosphate (pH 6.0), 0.4 M NH*OH, 0.5 mM EDTA. Aliquots were taken at 30-min intervals over a 2-h period and the treated DNA was ethanol precipitated in 0.3 M Na acetate (pH 4.8). Traces of NH,OH were removed by washes with 70% ethanol and two further precipitations. Yeast transformation was by the lithium acetate method of Ito et al. (27). Plasmid purification, production, and ligation of endonuclease-generated DNA fragments and DNA- mediated transformation of E. coli were by standard procedures.

7690 Nuclear-encoded Subunit Mutant of Yeast Cytochrome c Oxidase

RESULTS

Mutant Production-Plasmid pMC4 (ycox4) was treated in uitro with the chemical mutagen hydroxylamine, as outlined under "Experimental Procedures." This preparation was then used to transform strain WDl (ycox4- to uracil prototrophy (see Table I for a complete list of strains and plasmids). All prototrophs were then screened for respiratory competence by plating onto YEPGE. Of 800 uracil prototrophs generated, 52 showed no apparent growth on glycerol/ethanol solid media after 3 days incubation a t 30 "C. 36 of these proved to be p- when mated with the tester strains Y15 (ycox4- MATa) and Y18 (p" MATa) to analyze for production of petites, indicating a spontaneous rate of petite formation of approximately 5%.

Small scale cultures of the 16 remaining strains were grown on the nonrepressive sugar raffinose (33), and their isolated mitochondria analyzed for the presence of subunit IV by Western analysis (results not shown). Mitochondria from one respiratory-deficient strain only, RM1, contained a steady- state level of subunit IV equal to that of the normal control strain Y5 (ycox4-, MATa:YCOX4). Plasmid DNA was isolated from strain RM1, used to transfer E. coli to ampicillin resistance, reisolated, and then used to transform WD1 to uracil prototrophy. This plasmid, YpRLlO9, reproducibly con- ferred the respiratory-deficient phenotype. The complete YCOX4 gene harbored on YpRL109 was sequenced by the chain termination procedure of Sanger et al. (26). Four A-G transitions were found, resulting in three predicted amino acid replacements as follows:

A42T. E45K, G48N - CCT -+ ACT GAA -+ AAA B T + &T

Growth Characteristics and Western Analysis of Mutant Strain RMI-Doubling times for RM1 and several control strains when grown on various carbon sources at either 30 or 37 "C are shown in Fig. 1. A more rigorous study of strain RM1 revealed that on solid YEPGE, growth could be seen a t 30 "C after 5 days. However, no growth of RM1 was observed on respiratory substrates a t 37 "C, even after 2 weeks. The respiratory deficiency was shown to be recessive by mating R M l with the wild type strain JHRY99-10A, with the result- an t diploid strain maintaining a similar doubling time to the haploid wild type on respiratory substrates (Fig. 1). Several attempts to produce suppressors of strain RMl that would grow on glycerol/ethanol at 37 "C either spontaneously or by ethylmethane sulfate mutagenesis were unsuccessful.

The subunit composition of RM1 and other strains was

Y5 RYI OD1 Y14 JHRYlOA mnn- -cOu llavm rmDm mwrws

cvnmm

FIG. 1. Growth properties of various yeast strains. Doubling times for strains are compared after growth at either 30 or 37 "C on nonfermentable carbon sources (YEPCE) or fermentation media (YEPR). A, growth at 30 "C on YEPGE; R, growth at 37 "C on YEPGE; C, growth at 30 "C on YEPR; D, growth at 37 'C on YEPR. The negative control strain, WDI, is unable to respire at any temperature. (Doubling time for this strain at 37 "C on YEPR was similar to 30 "C.)

examined by Western blotting using polyclonal antibodies against the yeast holoenzyme. Fig. 2 shows an analysis of 100 pg of mitochondrial protein from various strains probed with a polyclonal antibody reacting with subunits 11, IV, V, VI, and VII/VIIa. As shown, mitochondria from the control ycox4- strain WD1 contain no detectahle subunit IV and are low in subunit I1 (see also Dowham et al. (28)). The mutant strain RM1, however, maintains good steady-state levels of each of the observable oxidase suhunits and appears to correctly process the mutant subunit IV. The levels of oxidase polypeptides in the mutant strain grown either at 30 or 37 "C on YEPR were similar.

Spectral Properties and Actiuities of Mitochondrial En- zymes-Table I1 lists the specific activities of succinate-cytochrome c reductase and oligomycin-sensitive ATPase of mitochondria isolated from mutant strain RM1 and the control strains WDl and Y5 along with estimated cytochrome levels for these strains under the indicated growth conditions. Levels of cytochrome bc, and related activities were very similar in mitochondria from all strains tested. ATPase activities were also in good agreement, implying that the mutation in strain RM1 did not have pleiotropic effect.. upon other complexes involved in oxidative phosphorylation.

A spectrally visible level of cytochrome a a 3 was absent from the negative control strain WD1 (Fig. 30) and was diminished in mitochondria from strain RM1 when compared to the normal control Y5 (Table 11, Fig. 3R, cf. A ) . Although unable to respire a t 37 "C when grown on raffinose RM1 contained levels of cytochrome a u 3 that were similar to, hut consistently slightly higher than when grown on either YEPGE (Fig. 3C, cf. R ) or YEPR at 30 "C (data not shown). The Eadie-Hofstee plots of cytochrome c oxidase activity over a range of cytochrome c concentrations from 0.1 to 15 jtM gave straight line plots from which K,,, values of 3.2 j t ~ for the RM1 oxidase and 3.5 p~ for the wild type enzyme were calculated. Under the assay conditions detailed under "Experimental Proce- dures," no evidence was obtained for a second high affinity cytochrome c-binding site which has been noted previously (see Ref. 34; but also see Ref. 35). The Vma, of the mutant oxidase was lower than that of the normal enzyme hy as much as 3.5-fold. However, when the enzyme was assayed from the mutant strain grown a t 37 "C (YEPR), the K,,, for cytochrome c increased to 4.8 p ~ , while the estimated Vma. was dramatically altered, being less than 3% of that of the normal enzyme. K,,, and V,,,,,= values of the wild type strain were the same

A B C D t " , -

c o r I V - :ox V l - cor '1 -

ror Y W , Y I I ~ -

FIG. 2. Western blot analysis of mitochondrial protein from various yeast strains probed with anti-holooxidaseantihodies. Mitochondria were isolated from strains grown at 30 "C on Y K P H unless stated otherwise. 100 pg of protein was dissociated and separated through a 198 polyacrylamide gel and transferred to a nitrocellulose membrane as detailed under "Experimental Procedures." Binding was attempted in 20 mM Tris-HCI (pH 7 5 ) , 0.5 M NaCI. I T skimmed milk powder supplemented with 0.0.5'X (v/v) Tween 20. A, wild type (Y5); H, negative control (WDI); C', mutant strain KMI; I). mutant strain R M I grown at 37 "C.

Nuclear-encoded Subunit Mutant of Yeast Cytochrome c Oxidase 7691 TABLE I1

Selected enzymatic activities and cytochrome contents of mitochondrial protein from various yeast strains Cytochromes Specific activity

Strain Phenotype Succinatecytochrome (temp:media) b C,Cl aa3 c" reductase Oligomycin-sensitive Cytochrome c

ATPaseb oxidase'

nmol/mg mitochondrial protein

Y5 Wild type COX4 0.33 0.29 0.140 1.03 3.9 331.0 (30 "C:YEPGE) control

(30 "C:YEPR) control

(30 "C:YEPGE)

(37 "C:YEPR)

WD1 Negative ~ 0 x 4 - 0.27 0.28 0.96 4.2

RM1 Missense cox4 0.29 0.31 0.089 0.97 5.1 91.3

RM1 Missense cox4 0.28 0.26 0.110 0.94 4.0 7.4

' Micromole of cytochrome c reduced per min/mg of protein at 20 "C. * Micromole of ATP hydrolyzed per min/mg of protein at 37 "C. e Mole of cytochrome c oxidized per s/mole of a a 3 at 20 "C.

500 550 600 650

FIG. 3. Spectral analysis of mitochondrial protein. Reduced oxidized difference spectra of mitochondrial protein, 0.6-1 mg/ml, at liquid nitrogen temperatures are shown after solubilization in 1% lauryl maltoside, 25 mM Tris-HC1 (pH 7.2), 25 mM NaCl, and addition of glycerol to 50% (v/v). Mitochondria were isolated from strains grown as follows: A, wild type (Y5), YEPGE at 30 "C; B, mutant strain RM1, YEPGE at 30 "C; C, mutant strain RM1, YEPR at 37 "C; D, negative control strain (WDl), YEPR at 30 "C.

whether grown in YEPR or YEPGE at 30 or 37 "C. Lauryl Maltoside Solubilization of Cytochrome c Oxidase-

In order to characterize the cytochrome c oxidase complex of RM1 more completely, preliminary experiments to define conditions for isolating the complex were undertaken. Thus, several detergent conditions for solubilizing the enzyme from mitochondrial membranes were tested. In all cases, disruption of the membrane led to inactivation of the enzyme. Fig. 44 shows a time course of the loss of activity of the oxidase after addition of lauryl maltoside to mitochondria. Solubilization of the membranes containing wild type enzyme had little effect on cytochrome c oxidase activity, however, activity of the mutant oxidase was mostly lost within 20 min. Western analysis of cytochrome c oxidase subunits I1 and IV at time intervals after solubilization with lauryl maltoside show a disappearance of subunit IV in a time course that parallels the loss of activity (Fig. 4B). This disappearance of subunit IV is due to endogenous protease digestion and could be slowed dramatically, but not prevented completely, by adding a mixture of protease inhibitors including EDTA (5 mM), phenylmethylsulfonyl fluoride (5 mM), and 1,lO-phenanthro- line (1 mM). No cleavage products of subunit IV were detected either with the monoclonal antibody or a polyclonal antibody, suggesting that small peptides are generated. Loss of subunit IV occurs without significant cleavage of subunit I1 (Fig. 4) or subunits V, VI, VII, and VIIa based on Western analysis (result not shown) and there is no effect on the cytochrome spectrum (Fig. 4C). Thus the conformational effect of the

mutated subunit IV is a localized one and does not result in denaturation of the complex.

The conformation of cytochrome c oxidase incorporating the mutant subunit IV was also compared with wild type enzyme by its susceptibility to denaturation with GdnHCl. As shown in Fig. 5, the concentration of GdnHCl at which the cytochrome c m 3 spectrum (absorbance at 605 nm) was 50% of the original was 0.7 M, cf. 2.0 M for the wild type enzyme, indicating that the complex containing the mutant subunit IV was more readily denatured than the wild type enzyme.

DISCUSSION

Isolation of bacterial aa3-type oxidases, which are fully functional but contain only three polypeptides (homologous to the mitochondrially encoded subunits) has led to uncer- tainty about whether nuclear-coded subunits are bona fide components of eukaryote oxidases, and if so, what functions they play. Disruption of the YcOX4 gene in S. cereukiae has been shown to prevent assembly of cytochrome c oxidase (30), establishing that this is a component of the complex but its role remains uncertain. Subunit IV, the homologue of subunit Vb in mammals, is thought to provide ligands for the Zn atom absent from prokaryotes but a component of the eukaryote enzyme. Extended x-ray absorption fine structure studies indicate that three of the ligands to this metal ion are Cys residues (36). There are only 5 Cys in all six nuclear-coded subunits of yeast oxidase combined (37); the three in subunit IV are conserved among eukaryote oxidases. We have produced site-specific substitutions of Cys-109 and Cys-112 of subunit IV, but mutation of either of these residues prevents assembly of the cytochrome c oxidase complex.2

Strain RM1 (this paper) expresses a missense copy of subunit IV which contains 3 amino acid substitutions. This strain is affected in its ability to respire at all temperatures, but is unable to grow at all at 37 "C on nonfermentable carbon sources. However, unlike the ~0x4- strain WD1 or strains expressing substitutions for Cys-109 or Cys-112, cytochrome c m 3 is assembled into a partially active complex. RM1 is unable to grow on respiratory substrates at 37 "C, but can be grown at this temperature on raffinose (a nonrepressive sugar) as sole carbon source when it exhibits levels of cytochrome a a 3

only marginally lower than in the wild type control strain grown under similar conditions. Kinetic constants calculated for the oxidase in mitochondrial preparations from the wild type control strain Y5 and the mutant strain RM1 grown

R. Lightowlers, unpublished results.

7692 Nuclear-encoded Subunit Mutant of Yeast Cytochrome c Oxidase

1 A

FIG. 4. Western blot analyses and cytochrome c oxidase activities of mitochondria from wild type Y 5 ( A ) , and mutant RM1 ( B ) strains (YEPGE at 30 "C) after solubilization in lauryl maltoside. Two mg of the indicated mitochondrial protein were solubilized in 2% (w/v) lauryl maltoside, 25 mM Tris-HC1 (pH 7.2), 25 mM NaCl and incubated a t 30 "C. Oxidase activities were determined at the time points indicated after solubilization and are expressed as a percentage of the initial activity. 100-pg aliquots were removed at each time point then dissociated, separated through a 19% polyacrylamide gel, and transferred to polyvinylidine difluoride membranes as detailed under "Ex- perimental Procedures." Blots were subjected to anti-I1 and anti-IV monoclonal antibodies. (For mutant RM1 mitochondria (R), the alkaline phosphatase reac- tion with anti-IV antibody was allowed to proceed for a comparatively longer time to emphasize its time-dependent disappearance.) C, spectral analysis of mitochondrial protein (0.6 mg/ml after lauryl maltoside solubilization. Reduced oxidized difference spectra were re- corded at liquid nitrogen temperatures, after lauryl maltoside solubilization and incubation for 5 h a t 30 "C. A, wild type (Y5) mitochondria; R, mutant (RMI) mitochondria.

' 7

0 5 10 15 20 30 60

TIME (min)

M+ll

cox IV b

0 5 10 1 5 2 0 30 60 TIME (rnin)

C

0)

2 P, v) A P

B

I 500

" I . . - , - . T

0 .0 0.5 1 .o 1.5 2 .0 2.5 3.0

Concenlraliun of (hanidinium-IICL (M) FIG. 5. Stability of cytochrome c oxidase in isolated mito-

chondria from wild type strain Y 5 and mutant strain RM1. Mitochondria (2 mg/ml) from each strain (YEPGE at 30 'C) were solubilized in 2% lauryl maltoside and subjected to increasing concentrations of guanidinium hydrochloride as detailed under "Experi- mental Procedures." The intact oxidase in various concentrations of GdnHCl was calculated as a percentage of visible cytochrome aa3 in comparison to the untreated mitochondrial preparation. B, mitochondria from mutant strain RM1; e, mitochondria from wild type strain Y5.

550 nm 600 650

either at 30 (YEPGE) or 37 "C (YEPR), showed similar K,,, values for reduced cytochrome c (3.2-4.8 p ~ ) , but very differ- ent turnover numbers, 339 s" for Y5 at 30 "C, 92 s" for RM1 at 30 "C, and 7.7 s" for RM1 grown at 37 "C (Table 11).

Denaturation of the oxidase with the chaotrope GdnHCl showed that the lauryl maltoside-solubilized complex from mutant strain RM1 was less stable than the wild type complex, as judged by disappearance of the characteristic cytochrome a a 3 a-band at 605 nm. When enzyme solubilized from RM1 was incubated at 30 "C, subunit IV levels were quickly depleted by endogenous proteases at a rate similar to the measured loss of cytochrome c oxidase activity. However, no concomitant loss of visible cytochrome a a 3 was noted. These results suggest that the mutant subunit IV is not bound or folded correctly. It may be in equilibrium between the folded and unfolded states, with the unfolded state being favored at the nonpermissive temperature, where it may be only loosely bound or free of the complex. At the nonpermissive temperature or after degradation of subunit IV by proteases, the remaining oxidase is still able to bind substrate but electron transfer to molecular oxygen is greatly impaired. There are several possible explanations of this observation. Removal of subunit IV may open up a core subunit for protease digestions. Western analysis of mitochondrial samples from strain RM1, either after growth at 37 "C or after solubilization and incubation at 30 "C, show no degradation of subunit 11, although this subunit is particularly sensitive to added proteases, which leave subunit IV intact.3 Cleavage of subunit I11 is unlikely to affect electron transfer activity as this subunit has been removed completely without effect (38, 39). This leaves the

Y.-Z. Zhang, unpublished results.

Nuclear-encoded Subunit Mutant of Yeast Cytochrome c Oxidase 7693

possibility that subunit I, for which we have no antibody, is 17. Van Gelder, B. F., and Slater, E. C. (1962) Biochem. Biophys. being cleaved to decrease enzyme turnover, but such protease Acta 118,47-57 digestion must occur without loss or alteration of heme a or 18. Berden, J., and Slater, E. C. (1970) Biochem. Biophys. Acta 2 1 6 ,

CuA site is affected by removal of subunit IV such that it is J. Biol. Chem. 2 5 1 , 1104-1115 unable to accept or transfer electrons. Chan and co-workers 20. Lotscher, H. R., deJong, C., and Capaldi, R. A. (1984) Biochem- (40) have shown that chemically perturbing the CuA site istry 23,4128-4134 causes a considerable loss of oxidase activity. Malmstrom (3, 21. Marusichy M. F. (lgW) J. Zmmurwz. Methods ll4* 155-159 41) has suggested that a conformational change induced on 22. R e k L. M., Maims, S. L., Ryan, D. E., Levin, W., Bandiera, S.,

reduction Of the CuA/heme a site may be coup1ed to the 23. Kadenbach, B., Jarausch, J., Hartmann, R., and Merle, P. (1982) and Thomas, P. E. (1987) J. Zmmunol. Methods 100 , 123-130

proton translocating function of cytochrome oxidase. Aber- Anal. Biochem. 129,517-521 rant binding or loss of the mutant subunit IV could prevent 24. Zhang, Y-Z., Lindorfer, M., and Capaldi, R. A. (1988) Biochem- this transition, resulting in extremely low enzyme turnover istry 27,1389-1394 rates. 25. Matsudaira, P. (1987) J. Biol. Chem. 2 6 2 , 10035-10038

a3 based on spectral analysis. Another Possibility is that the 19. Ferguson-Miller, S., Brautigan, D. L., and Margoliash, E. (1976) 237-249

26. Sanser. F., Nicklen. S.. and Coulson. A. R. (1977) Proc. Natl. REFERENCES

1. Capaldi, R. A. (1990) Annu. Reu. Biochem. 59,569-596 2. Capaldi, R. A. (1990) Arch. Biochem. Biophys. 2 8 0 , 252-262 3. Malmstrom, B. G. (1990) Arch. Biochem. Biophys. 2 8 0 , 233-241 4. Kadenbach, B., and Merle, P. (1981) FEBS Lett. 135 , l -11 5. Power, S. D., Lochrie, M. A., Sevarino, K. A., Patterson, T. E.,

and Poyton, R. 0. (1984) J. Biol. Chem. 2 6 9 , 6564-6570 6. Bisson, R., Schiavo, G., and Papini, E. (1985) Biochemistry 24,

7. Kadenbach, B., Stroh, A., Ungibauer, M., Kuhn-Nentwig, L., Buge, U., and Jarausch, J. (1986) Methods Enzymol. 126 , 32- 45

8. Yanamura, W., Zhang, Y-Z., Takamiya, S., and Capaldi, R. A. (1988) Biochemistry 27, 4909-4914

9. Raitio, M., Jalli, T., and Saraste, M. (1987) EMBO J. 6 , 2823-

7845-7852

10.

11.

12.

13.

14.

15.

16.

2825

Saraste, M. (1988) J. Biochem. (Tokyo) 103 , 606-610 Sone, N., Yokoi, F., Fu, T., Ohta, S., Metso, T., Raitio, M., and

Poyton, R. O., Trueblood, C. E., Wright, R. M., and Farrell, L.

Patterson, T. E., and Poyton, R. 0. (1986) J. Biol. Chem. 2 6 1 ,

Aggeler, R. J., and Capaldi, R. A. (1990) J. Biol. Chem. 2 6 5 ,

Sherman, F., Fink, G. R., and Hicks, J. B. (1986) in Methods in Yeast Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Daum, G., Bohni, P. C., and Schatz, G. (1982) J. Biol. Chem.

Weiss, H., and Sebald, W. (1978) Methods Enzymol. 5 3 , 66-73

E. (1988) Ann. N. Y. Acad. Sci. 5 5 0 , 289-307

17192-17197

16389-16393

257,13028-13033

A i d Sci. U. S. A: 74,5463-5467 '

teriol. 1 5 3 , 163-168 27. Ito, H., Fukuda, Y., Murata, K., and Kimura, A. (1983) J. Bac-

28. Dowham, W., Bibus, C. R., and Schatz, G. (1985) EMBO J. 4 ,

29. Kolodkin, A. L., Klar, A. J. S., and Stahl, F. W. (1986) Cell 46 ,

30. Bibus, C. R., Lemire, B. D., Suda, K., and Schatz, G. (1988) J.

31. Casabadan, M. F., and Cohen, S. N. (1980) J. Mol. Biol. 138 ,

32. Yanisch-Perron, C., Vieira, J., and Messing, J . (1985) Gene

33. Lustig, A., Padmanabadan, G., and Rabinowitz, M. (1982) Bio-

34. Trivedi. A.. Fantin. D. J.. and Tustanoff. E. R. (1986) Biochem.

179-184

733-740

Biol. Chem. 263,13097-13102

179-207

(Amst.) 33, 103-119

chemistry 21,309-316

35.

36.

37.

38.

39.

40.

41.

Cell i iol . '64, 1195-1210

J. Biol. Chem. 254, 11973-11981 Dethmers, J. K., Ferguson-Miller, S., and Margoliash, E. (1979)

Naqui, A., Powers, L., Lundeen, M., Constantinescu, A., and

McEwen, J . E., KO, C., Kloeckner-Gruissem, B., and Poyton, R.

Prochaska, L. J., and Fink, P. S. (1987) J. Bioenerg. Biomembr.

Saraste, M., Pentilla, T., and Wikstrom, M. (1981) Eur. J.

Nilsson, T., Gelles, J., Li, P. M., and Chan, S. I. (1988) Biochem-

Thornstrom, P-E., Nilsson, T., and Malmstrom, B. G. (1988)

Chance, B. (1988) J. Biol. Chem. 263 , 12342-12345

0. (1986) J. Biol. Chem. 261 , 11872-11879

19,143-167

Biochem. 115,261-268

istry 27,296-301

Biochem. Biophys. Acta 9 3 5 , 103-108

THE JOURNAL OF CHEMISTRY Vol. 266, No. 12, …THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 266, No. 12, Issue

Documents