Gluing the Respiratory Chain Together

Cytochrome bc 1 complex (complex III) and cytochrome c oxidase complex (complex IV) are multisubunit homodimers that are essential components of the mitochondrial respiratory chain. Complexes III and IV associate to form a supercomplex that can be displayed using blue native polyacrylamide gel electrophoresis. Both homodimeric complexes contain tightly associated cardiolipin (CL) required for function. We report here that in a crd1Δ strain of yeast (null in expression of CL synthase) ∼90% of complexes III and IV were observed as individual homodimers; only the supercomplex was observed with CRD1wild type cells. Introduction of a plasmid born copy of theCRD1 gene under exogenous regulation by doxycycline made possible controlled variation in the in vivo CL levels. At an intermediate level of CL, a mixture of individual homodimers (30%) and supercomplex (70%) was observed. These results strongly indicate that CL plays a central role in higher order organization of components of the respiratory chain of mitochondria. Cytochrome bc 1 complex (complex III) and cytochrome c oxidase complex (complex IV) are multisubunit homodimers that are essential components of the mitochondrial respiratory chain. Complexes III and IV associate to form a supercomplex that can be displayed using blue native polyacrylamide gel electrophoresis. Both homodimeric complexes contain tightly associated cardiolipin (CL) required for function. We report here that in a crd1Δ strain of yeast (null in expression of CL synthase) ∼90% of complexes III and IV were observed as individual homodimers; only the supercomplex was observed with CRD1wild type cells. Introduction of a plasmid born copy of theCRD1 gene under exogenous regulation by doxycycline made possible controlled variation in the in vivo CL levels. At an intermediate level of CL, a mixture of individual homodimers (30%) and supercomplex (70%) was observed. These results strongly indicate that CL plays a central role in higher order organization of components of the respiratory chain of mitochondria. Broader understanding of how lipid-protein interactions determine protein structure and function is increasing. Recent crystallographic results of multimeric membrane-associated complexes have revealed lipids inserted between the protein subunits (1Essen L. Siegert R. Lehmann W.D. Oesterhelt D. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 11673-11678Crossref PubMed Scopus (410) Google Scholar, 2McAuley K.E. Fyfe P.K. Ridge J.P. Isaacs N.W. Cogdell R.J. Jones M.R. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 14706-14711Crossref PubMed Scopus (207) Google Scholar). Even monomeric proteins such as OmpF of Escherichia coli show outer membrane-specific lipid A molecules in the crystal structure (3Ferguson A.D. Welte W. Hofmann E. Lindner B. Holst O. Coulton J.W. Diederichs K. Struct. Fold. Des. 2000; 8: 585-592Abstract Full Text Full Text PDF Scopus (182) Google Scholar). The highly ordered state of these lipids necessary for their resolution supports a role for specific and tight association of defined lipid species with membrane proteins. Considerable evidence also supports the organization of the respiratory chain in both prokaryotes and eukaryotes into “respirasomes” or supercomplexes made up of smaller multimeric complexes (4Schagger H. Crit. Rev. 2001; 52: 119-128Google Scholar). For oxidative phosphorylation, such supercomplexes may provide added efficiency by eliminating the need for diffusion of substrates and products between individual complexes and provide a means for channeling intermediates between individual complexes. What role does the lipid environment play in supporting supercomplex formation? Are there specific protein-lipid interactions involved or is the collective property of the membrane responsible for higher order complex formation? In 1974, Skulachev (5Mikel'saar K. Severina I.I. Skulachev V.P. Usp. Sovrem. Biol. (in Russian). 1974; 78: 348-370PubMed Google Scholar) suggested that cardiolipin (CL) 1The abbreviations used are: CL, cardiolipin; PG, phosphatidylglycerol; Cox2p, subunit 2 of cytochromec oxidase complex; Cobp, b component of cytochrome bc 1 complex; BN, blue native might participate in linking together components of the respiratory chain. CL is a minor phospholipid with distribution limited to bacterial cytoplasmic membranes and to the inner membrane of mitochondria where energy transducing oxidative processes of cells are centered. Structured CL is associated with theb subunit of the cytochrome bc 1complex (complex III) in yeast (6Lange C. Nett J.H. Trumpower B.L. Hunte C. EMBO J. 2001; 20: 6591-6600Crossref PubMed Scopus (365) Google Scholar). Complex III is a homodimer of 10 different subunits in yeast and 11 subunits in mammalian cells (7Hunte C. Koepke J. Lange C. Rossmanith T. Michel H. Struct. Fold. Des. 2000; 8: 669-684Abstract Full Text Full Text PDF Scopus (514) Google Scholar). When CL was removed from mammalian complex III, structure was perturbed and the complex lost all function (8Gomez Jr., B. Robinson N.C. Biochemistry. 1999; 38: 9031-9038Crossref PubMed Scopus (130) Google Scholar). Only CL and its immediate metabolic precursor phosphatidylglycerol (PG) restored activity to the resolved complex. Although PG restored full activity, the affinity of the multiple subunits for each other was diminished. Similarly, mammalian cytochrome c oxidase complex (complex IV) contains tightly bound CL that when removed results in loss of structure and function (9Sedlak E. Robinson N.C. Biochemistry. 1999; 38: 14966-14972Crossref PubMed Scopus (131) Google Scholar). In yeast and mammalian cells complexes III and IV form a supercomplex composed of a heterodimer of individual homodimers that remains intact after detergent solubilization (10Cruciat C.M. Brunner S. Baumann F. Neupert W. Stuart R.A. J. Biol. Chem. 2000; 275: 18093-18098Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar, 11Schagger H. Pfeiffer K. EMBO J. 2000; 19: 1777-1783Crossref PubMed Scopus (1039) Google Scholar). Here we show that both complex III and IV in CL-lacking yeast mutants do not associate to form a supercomplex under conditions of analysis where such a supercomplex is observed in extracts from wild type cells (10Cruciat C.M. Brunner S. Baumann F. Neupert W. Stuart R.A. J. Biol. Chem. 2000; 275: 18093-18098Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar, 11Schagger H. Pfeiffer K. EMBO J. 2000; 19: 1777-1783Crossref PubMed Scopus (1039) Google Scholar). The amount of such a supercomplex is dependent on the CL content of the mitochondria and is reflected in the growth properties of the cells on a non-fermentable carbon source. These results strongly suggest that CL plays a specific role in supporting a critical interaction between complexes III and IV in vivo that cannot be substituted by the elevated levels of PG in the mutant. YPH500 (ade2–101, his3Δ200, leu2Δ1, lys2–801, trp1Δ63, ura3–52, MATα) (12Sikorski R.S. Hieter P. Genetics. 1989; 122: 19-27Crossref PubMed Google Scholar) was used as wild type strain in this study. YZD2 (ade2–101, his3Δ200, leu2Δ1, lys2–801, trp1Δ63, ura3–52, crd1Δ::HIS3, MATα), with the entire CRD1(encodes CL synthase) open reading frame disrupted, was constructed by homologous recombination with a polymerase chain reaction (PCR) product of HIS3 gene flanked by 5′ and 3′ regions that lie outside of the CRD1 gene coding sequence. Amplification was accomplished from plasmid pRS303 (12Sikorski R.S. Hieter P. Genetics. 1989; 122: 19-27Crossref PubMed Google Scholar) with primers 5′-ATCGGGAGTATACAATATTTACAATTGAAAAAATATAGAAAGGAGGCAATTGAGAAACAAGCAGGCCTGGTAGCATAGTTTGGTCCCtgatgcggtattttctccttacg-3′ and 5′-GTTGATTTAGTGTATTTTCCACTACATAAAAACTAAAAAACATTCAAAATGAAAAGTCAGGACCCTTTTCAAAAAGGATCGCAATTAatgcggcatcagagcagattgta-3′ (homology regions flanking CRD1 are in capital letters, and HIS3 homology regions are in lowercase letters). Yeast transformation experiments were done using the Alkali-CationTM yeast transformation kit (Bio 101, Inc.). ACRD1 null derivative (crd1Δ) of YPH500 was selected on histidine dropout minimal medium glucose plates (13Sherman F. Methods Enzymol. 1991; 194: 3-21Crossref PubMed Scopus (2543) Google Scholar). Genomic PCR with three sets of primers (external CRD1 and internal HIS3, internal HIS3 and externalCRD1, two primers external to CRD1) confirmed the disruption. The doxycycline-regulated plasmid, pYMZ1 (CEN4, ARS1, URA3,CMVp/tetR-VP16,tetO-CYC1p/CRD1/ADH-1t), was constructed as follows. The CRD1 gene was isolated by PCR amplification of DNA prepared from plasmid pYCLS10-1 (14Chang S.C. Heacock P.N. Mileykovskaya E. Voelker D.R. Dowhan W. J. Biol. Chem. 1998; 273: 14933-14941Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar) with primers 5′-ataggatcc ATGATTCAAATGGTGCCC-3′ and 5′-atactgcag CTATTTTAAAAGTTTAAAAGCG-3′ (CRD1 homology regions are capitalized, start and stop codons are in boldface, and the BamHI and PstI sites are in italics). An 852-bp fragment was cut with BamHI and PstI and ligated into plasmid pCM189 (15Gari E. Piedrafita L. Aldea M. Herrero E. Yeast. 1997; 13: 837-848Crossref PubMed Scopus (504) Google Scholar) behind the tetracycline-regulated promoter. DNA was amplified and verified by DNA sequencing of the junction points of the new plasmid. Yeast spheroplast transformations were accomplished as mentioned above. Transformants were selected on yeast nitrogen base uracil dropout medium containing glucose (13Sherman F. Methods Enzymol. 1991; 194: 3-21Crossref PubMed Scopus (2543) Google Scholar). YZD2/pYMZ1 was maintained on YPEG (1% yeast extract, 2% peptone, 1% ethanol, and 3% glycerol) medium (13Sherman F. Methods Enzymol. 1991; 194: 3-21Crossref PubMed Scopus (2543) Google Scholar) in the absence of doxycycline (derepression conditions). Phospholipid composition analyses were performed as described previously (16Ostrander D.B. Zhang M. Mileykovskaya E. Rho M. Dowhan W. J. Biol. Chem. 2001; 276: 25262-25272Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar). Briefly, cells were pregrown overnight at 30 °C in either complete synthetic defined media with 1% ethanol and 3% glycerol (13Sherman F. Methods Enzymol. 1991; 194: 3-21Crossref PubMed Scopus (2543) Google Scholar) or in YPEG. For doxycycline-regulated cells, the overnight culture was supplemented with 10 μg/ml of doxycycline (near complete repressed conditions). Cells were diluted into the same growth medium as the overnight cultures but supplemented with 32PPi (American Radiolabeled Chemicals, Inc.) at 10 μCi/ml and the indicated doxycycline concentration. Cultures were grown for four to five generations at 30 °C to constant radio specific activity and harvested during exponential phase of growth. Isolated phospholipids were separated by thin layer chromatography using HPTLC Silica Gel 60 plates (Whatman) impregnated with boric acid and chloroform/methanol/acetic acid (65/28/8) as solvent (17Fine J.B. Sprecher H. J. Lipid Res. 1982; 23: 660-663Abstract Full Text PDF PubMed Google Scholar). Co-migration with standard phospholipids (Sigma) was used to identify individual spots. Phospholipids were quantified by phosphor imaging using a Molecular Imager FX (Bio-Rad) and normalized to the total amount of organic soluble 32PPidisplayed after separation. Phospholipid composition is expressed as mole % of each lipid species with correction for 2 mol of phosphate/mol of CL. Crude mitochondria were prepared from cells in exponential phase of growth after spheroplast formation using zymolyase (16Ostrander D.B. Zhang M. Mileykovskaya E. Rho M. Dowhan W. J. Biol. Chem. 2001; 276: 25262-25272Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar). Sample preparation and BN-PAGE was carried out essentially as previously described (10Cruciat C.M. Brunner S. Baumann F. Neupert W. Stuart R.A. J. Biol. Chem. 2000; 275: 18093-18098Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar). Briefly, mitochondria (200 μg of protein) were solubilized in 40 μl digitonin buffer (1% (w/v) digitonin, 50 mm potassium acetate, 30 mm HEPES-KOH, pH 7.4, 10% glycerol, 0.1 mg/ml α2-macroglobulin, 0.1% (v/v) protease inhibitor mixture set III (Calbiochem)). After solubilization, samples were centrifuged at 125,000 × g for 30 min (TLA100.3 rotor, Beckman TL-100 ultracentrifuge). 4 μl of sample buffer (5% (w/v) Serva Blue G (Serva) in 500 mmaminocaproic acid) were added prior to electrophoresis using a 4–8% gradient of polyacrylamide. Proteins were transferred to nitrocellulose sheets for Western blot analysis and specifically located using monoclonal antibody (Molecular Probes) specific for Cox2p (cytochromec oxidase (complex IV) subunit 2) or polyclonal antibody (Gottfried Schatz, Biozentrum, Basel, Switzerland) specific for Cobp (cytochrome b subunit of complex III). Final detection was with secondary antibodies linked to horseradish peroxidase using chemiluminescence (Amersham Biosciences) and a Bio-Rad FX for quantification. High molecular weight markers from Amersham Biosciences were used as standards. Respiratory activity of isolated mitochondria was measured with a Clark oxygen electrode using NADH or ascorbate plus diaminoduren (cytochrome c oxidase activity) as substrate in the presence of the uncoupler carbonyl cyanide-p-(trifluoromethoxy)phenylhydrazone (14Chang S.C. Heacock P.N. Mileykovskaya E. Voelker D.R. Dowhan W. J. Biol. Chem. 1998; 273: 14933-14941Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar). Protein was measured using the Bradford method (Bio-Rad) with bovine serum albumin as standard. A “biological reagent” was developed in which the CL content of yeast mitochondria could be regulated in vivo. A low copy “tet-off” plasmid system, pYMZ1, was employed to afford exogenous regulation of a plasmid copy of the CRD1 gene in a crd1Δ background (YZD2). In this system exogenous addition of the repressor doxycycline to the growth medium showed a dose-dependent reduction of CL levels with a parallel increase in PG levels (Fig.1). Although fully repressed or derepressed levels of CL did not reach that observed in acrd1Δ or CRD1 strain, respectively, the use of the regulated system plus these latter two strains afforded a complete range of CL levels. As expected, the only lipid that increased in a dose dependent manner was PG, the immediate precursor to CL. The steady state levels of other phospholipids were not significantly changed by either mutation or doxycycline levels supporting a specific effect on mitochondrial phospholipid composition. Since doxycycline is not metabolized by or has no known regulatory effects in yeast, this system is superior to uncover effects directly related to specific gene expression. Use of natural regulated promoters in yeast results in large unrelated changes in cell metabolism when switching between metabolizable regulatory molecules. CRD1 cells grew considerably faster and to a higher final A 600 than crd1Δ cells (Fig. 2) as has been observed previously (18Jiang F. Rizavi H.S. Greenberg M.L. Mol. Microbiol. 1997; 26: 481-491Crossref PubMed Scopus (154) Google Scholar, 19Tuller G. Hrastnik C. G. U. F. G. 1998; PubMed Scopus Google Scholar). no effect on the growth properties cells, and YZD2/pYMZ1 cells in the absence of doxycycline grew as as wild type cells with a wild type CL YZD2/pYMZ1 in the presence of doxycycline μg displayed growth properties intermediate between wild type and crd1Δ cells with its intermediate CL content mol YZD2/pYMZ1 cells grown in levels of doxycycline showed growth and 600 not cell reflected the A 600 in all isolated from both CRD1 and crd1Δ cells of NADH as as cytochromec oxidase activity with results (14Chang S.C. Heacock P.N. Mileykovskaya E. Voelker D.R. Dowhan W. J. Biol. Chem. 1998; 273: 14933-14941Abstract Full Text Full Text PDF PubMed Scopus (170) Google F. M. M. Greenberg M.L. J. Biol. Chem. 2000; 275: Full Text Full Text PDF PubMed Scopus Google Scholar). As digitonin extracts of mitochondria isolated cells to BN-PAGE showed supercomplex formation between complex IV and complex III (Fig. as by antibody specific for Cox2p and Cobp, (10Cruciat C.M. Brunner S. Baumann F. Neupert W. Stuart R.A. J. Biol. Chem. 2000; 275: 18093-18098Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar, 11Schagger H. Pfeiffer K. EMBO J. 2000; 19: 1777-1783Crossref PubMed Scopus (1039) Google Scholar) that cells grown on non-fermentable carbon displayed a supercomplex composed of a heterodimer containing homodimer each of complex III and complex the other extracts of mitochondria from crd1Δ cells displayed molecular species of smaller that with antibody specific for a component of either complex IV (Fig. or complex III (Fig. The of these smaller species are with of a homodimer of complex IV or a homodimer of complex III (10Cruciat C.M. Brunner S. Baumann F. Neupert W. Stuart R.A. J. Biol. Chem. 2000; 275: 18093-18098Abstract Full Text Full Text PDF PubMed Scopus (215) Google was the molecular standard the absence of of complexes III and of Western from experiments showed of complexes III and IV as individual homodimers and not into a the affinity of complex III and complex IV for each other is when these complexes are in cells containing PG in of CL. of YZD2/pYMZ1 cells in the presence of 10 μg/ml doxycycline mol % in a mixture of individual homodimers of complex III (Fig. individual homodimers of complex IV (Fig. and or of the or of YZD2/pYMZ1 cells in the absence of doxycycline mol % in only supercomplex formation the of a supercomplex made up of homodimers of complexes III and IV is highly dependent on CL in a dose-dependent We report here a to uncover of the phospholipid CL in cell function. The to CL content of the mitochondria an to in a dose-dependent manner with the level of CL to strongly an of CL in a A parallel increase in the level of PG cannot be but in experiments can be used to a role for other phospholipids were significantly by either regulation or of the of the be in the mitochondria. as the of of CL in the presence of PG is by the of effect of the crd1Δ on growth of yeast at 30 °C on glucose as the carbon (14Chang S.C. Heacock P.N. Mileykovskaya E. Voelker D.R. Dowhan W. J. Biol. Chem. 1998; 273: 14933-14941Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar, F. Rizavi H.S. Greenberg M.L. Mol. Microbiol. 1997; 26: 481-491Crossref PubMed Scopus (154) Google G. Hrastnik C. G. U. F. G. 1998; PubMed Scopus Google Scholar). At 30 °C on a non-fermentable carbon crd1Δ cells are in both growth and final cell or cell in a CL dose-dependent the of this effect on carbon and the association of CL with energy transducing complexes of the inner mitochondrial a molecular for these be in organization function of these complexes. have that complex III and IV form a supercomplex made up of a heterodimer of homodimers of complex III and complex Since this supercomplex was at digitonin of protein) to be displayed under native on is that detergent supercomplex formation or that this complex is in vivo H. Pfeiffer K. EMBO J. 2000; 19: 1777-1783Crossref PubMed Scopus (1039) Google Scholar). The and here is the complete of this supercomplex in extracts of cells CL to wild type cells and the detection of an intermediate level of supercomplex at intermediate CL levels. These results strongly indicate that CL plays a central vivo in the association between complexes III and As the homodimers may have limited affinity for each other to the presence of PG is required for their individual The effect of of both PG and CL on components of the chain cannot be null mutants are in of the subunits of these complexes (16Ostrander D.B. Zhang M. Mileykovskaya E. Rho M. Dowhan W. J. Biol. Chem. 2001; 276: 25262-25272Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar). complex III and IV from contain tightly bound phospholipid molecules CL. CL is associated with and to be required for the structure and function of mammalian complex IV (9Sedlak E. Robinson N.C. Biochemistry. 1999; 38: 14966-14972Crossref PubMed Scopus (131) Google Scholar) as by its resolution in the crystal structure T. M. H. E. H. T. R. K. R. S. A. 1999; Scholar). lipid molecules have been resolved in the crystal structure of complex III from (6Lange C. Nett J.H. Trumpower B.L. Hunte C. EMBO J. 2001; 20: 6591-6600Crossref PubMed Scopus (365) Google Scholar). is the of a CL bound between b 1 of the with in the to which CL is bound grew but these did not in a parallel of activity of complex CL levels also the growth properties of yeast a effect on respiratory chain activity as measured in isolated mitochondria. suggest the loss in growth efficiency is related to a role of CL that cannot be by The for CL revealed by results is in the association of complexes III and IV into a supercomplex that in vivo might also to efficiency of energy transducing and cell growth might CL a specific and interaction between complex III and crystal of multimeric membrane protein complexes strongly a role for lipids as a than between subunits of these complexes P.K. K.E. Isaacs N.W. Cogdell R.J. Jones M.R. Sci. 2001; 26: Full Text Full Text PDF PubMed Scopus Google Scholar). in this role is the to and between the membrane specific at the In membrane proteins with tightly bound lipids on their but these lipids also into this and may also provide for interaction between tightly associated proteins. A by of cytochrome c 1 and cytochrome b has been suggested as a possible of interaction between complex III and complex IV (4Schagger H. Crit. Rev. 2001; 52: 119-128Google Scholar). We suggest CL may this and provide a between the complexes. The between complexes III and IV to be the supercomplex at low H. Pfeiffer K. EMBO J. 2000; 19: 1777-1783Crossref PubMed Scopus (1039) Google Scholar). is with that supercomplex can be in cells CL but with elevated PG that may low affinity A molecular of PG may not the and the of two PG molecules may in also that there is no for CL in homodimers of complex III or complex IV and is with the of PG to for CL in of activity for the highly mammalian complex III (8Gomez Jr., B. Robinson N.C. Biochemistry. 1999; 38: 9031-9038Crossref PubMed Scopus (130) Google Scholar). results are with the that complexes in mitochondria and might be in PubMed Scopus Google Scholar, E. Dowhan W. D. L. 2001; PubMed Scopus Google Scholar, E. Dowhan W. J. 2000; PubMed Scopus Google Scholar). We Stuart for for BN-PAGE We are also to for with antibody used in this