The mitochondrial genomes of Chlamydomonad algae lack the cox2 gene that encodes the essential subunit COX II of cytochrome c oxidase. COX II is normally a single polypeptide encoded by a single mitochondrial gene. In this work we cloned two nuclear genes encoding COX II from both Chlamydomonas reinhardtii and Polytomella sp. The cox2agene encodes a protein, COX IIA, corresponding to the N-terminal portion of subunit II of cytochrome c oxidase, and thecox2b gene encodes COX IIB, corresponding to the C-terminal region. The cox2a and cox2b genes are located in the nucleus and are independently transcribed into mRNAs that are translated into separate polypeptides. These two proteins assemble with other cytochrome c oxidase subunits in the inner mitochondrial membrane to form the mature multi-subunit complex. We propose that during the evolution of the Chlorophyte algae, thecox2 gene was divided into two mitochondrial genes that were subsequently transferred to the nucleus. This event was evolutionarily distinct from the transfer of an intact cox2gene to the nucleus in some members the Leguminosae plant familyAF305078AF305079AF305080AF305540AF305541AF305542AF305543 . The mitochondrial genomes of Chlamydomonad algae lack the cox2 gene that encodes the essential subunit COX II of cytochrome c oxidase. COX II is normally a single polypeptide encoded by a single mitochondrial gene. In this work we cloned two nuclear genes encoding COX II from both Chlamydomonas reinhardtii and Polytomella sp. The cox2agene encodes a protein, COX IIA, corresponding to the N-terminal portion of subunit II of cytochrome c oxidase, and thecox2b gene encodes COX IIB, corresponding to the C-terminal region. The cox2a and cox2b genes are located in the nucleus and are independently transcribed into mRNAs that are translated into separate polypeptides. These two proteins assemble with other cytochrome c oxidase subunits in the inner mitochondrial membrane to form the mature multi-subunit complex. We propose that during the evolution of the Chlorophyte algae, thecox2 gene was divided into two mitochondrial genes that were subsequently transferred to the nucleus. This event was evolutionarily distinct from the transfer of an intact cox2gene to the nucleus in some members the Leguminosae plant familyAF305078AF305079AF305080AF305540AF305541AF305542AF305543 . high precision liquid chromatography mtDNA mitochondrial DNA nucleotide(s) polymerase chain reaction maximum local hydrophobicity of a sequence segment hydrophobic moment rapid amplification of cDNA ends bacterial artificial chromosome Mitochondria are thought to descend from free-living α-proteobacteria (1Gray M.W. Lang B.F. Cedergren R. Golding G.B. Lemieux C. Sankoff D. Turmel M. Brossard N. Delage E. Littlejohn T.G. Plante I. Rioux P. Saint-Louis D. Zhu Y. Burger G. Nucleic Acids Res. 1998; 26: 865-878Crossref PubMed Scopus (302) Google Scholar), whose closest extants are bacteria of the genusRickettsia (2Andersson S.G.E. Zomorodipour A. Andersson J.O. Sicheritz-Pontén T. Alsmark U.C.M. Podowski R.M. Nslund A.K. Eriksson A.-S. Winkler H.H. Kurland C.G. Nature. 1998; 396: 133-140Crossref PubMed Scopus (1316) Google Scholar). After the endosymbiotic event, there was a transfer of genes from the protomitochondrion to the nucleus such that few genes now remain in mitochondrial genomes (3Gray M.W. Int. Rev. Cytol. 1992; 14: 233-357Crossref Scopus (440) Google Scholar). Those genes that remain encode only a sub-set of the mitochondrial proteins needed for oxidative phosphorylation and a portion of the factors necessary for their expression (4Attardi G. Schatz G. Annu. Rev. Cell Biol. 1988; 4: 289-333Crossref PubMed Scopus (1062) Google Scholar). The mtDNA-encoded respiratory chain subunits are highly hydrophobic polytopic proteins that contain two or more transmembrane stretches (5von Heijne G. FEBS Lett. 1986; 198: 1-4Crossref PubMed Scopus (121) Google Scholar, 6Popot J.L. de Vitry C. Annu. Rev. Biophys. Biophys. Chem. 1990; 19: 369-403Crossref PubMed Scopus (105) Google Scholar). The genes for nad1,nad2, nad3, nad4, nad4L,nad5, and nad6 (encoding subunits 1, 2, 3, 4, 4L, 5, and 6 of NADH:ubiquinone oxidoreductase), cob (encoding cytochrome b of the bc1 complex),cox1, cox2, and cox3 (encoding subunits COX I, COX II, and COX III of cytochrome coxidase), and atp6 and atp8 (encoding subunitsa and A6L of the F0 portion of ATP synthetase) are found in the mitochondrial genomes of most organisms. The transfer of mitochondrial genes to the nucleus is an ongoing process, as shown by the presence of particular genes encoded in both the mitochondrial and the nuclear genomes in certain species. These are exemplified by COX II in some members of the family leguminosae (7Nugent J.M. Palmer J.D. Cell. 1991; 66: 473-481Abstract Full Text PDF PubMed Scopus (283) Google Scholar, 8Covello P.S. Gray M.W. EMBO J. 1992; 11: 3815-3820Crossref PubMed Scopus (134) Google Scholar, 9Adams K.L. Song K. Roessler P.G. Nugent J.M. Doyle J.L. Palmer J.D. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 13863-13868Crossref PubMed Scopus (116) Google Scholar) and by subunit 9 of ATP synthetase in Neurospora crassa(10van den Boogaart P. Samallo J. Agsteribbe E. Nature. 1982; 298: 187-189Crossref PubMed Scopus (127) Google Scholar). The algae of the family Chlamydomonadaceae, includingChlamydomonas reinhardtii and Polytomella sp., lack some of the genes that are typically found in mitochondrial genomes, including nad3, nad4L, cox2,cox3, atp6, and atp8(11Michaelis G. Vahrenholz C. Pratje E. Mol. Gen. Genet. 1990; 223: 211-216Crossref PubMed Scopus (101) Google Scholar, 12Denovan-Wright E.M. Nedelcu A.M. Lee R.W. Plant Mol. Biol. 1998; 36: 285-295Crossref PubMed Scopus (65) Google Scholar, 13Kroymann J. Zetsche K. J. Mol. Biol. 1998; 47: 431-440Google Scholar) 1S. Funes, A. Antaramian, and D. González-Halphen, unpublished results. 1S. Funes, A. Antaramian, and D. González-Halphen, unpublished results.. We have shown that, in at least two members of this family, C. reinhardtiiand Polytomella sp., the cox3 gene was transferred to the nucleus, and the corresponding mitochondrial copy has been lost (14Pérez-Martı́nez X. Vázquez-Acevedo M. Tolkunova E. Funes S. Claros M.G. Davidson E. King M.P. González-Halphen D. J. Biol. Chem. 2000; 275: 30144-30152Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). Several mitochondrial respiratory chain complexes have been isolated and characterized from the colorless alga Polytomella sp. (15Gutiérrez-Cirlos E.B. Antaramian A. Vázquez-Acevedo M. Coria R. González-Halphen D. J. Biol. Chem. 1994; 269: 9147-9154Abstract Full Text PDF PubMed Google Scholar, 16Atteia A. Dreyfus G. González-Halphen D. Biochim. Biophys. Acta. 1997; 1320: 275-284Crossref PubMed Scopus (27) Google Scholar, 17Brumme S. Kruft V. Schmitz U.-K. Braun H.-P. J. Biol. Chem. 1998; 273: 13143-13149Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar), a close relative of Chlamydomonas, including an active, cyanide-sensitive cytochrome c oxidase preparation (14Pérez-Martı́nez X. Vázquez-Acevedo M. Tolkunova E. Funes S. Claros M.G. Davidson E. King M.P. González-Halphen D. J. Biol. Chem. 2000; 275: 30144-30152Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). In this work, we show that COX II is present as a heterodimer in this complex. All COX II sequences that have been described to date are single polypeptides encoded by one gene, normally in the mitochondrial genome. In both Polytomella sp. and C. reinhardtii, COX II is encoded by two nuclear genes that were named cox2a and cox2b. The cox2a gene encodes a protein, COX IIA, corresponding to the N-terminal half of a typical one-polypeptide COX II, and the cox2b gene encodes a protein, COX IIB, equivalent to the C-terminal half of the same subunit. We propose that these separate genes give rise to a heterodimeric COX II that results from the noncovalent assembly of the COX IIA and COX IIB polypeptides in cytochrome c oxidase of the inner mitochondrial membrane. Polytomellasp. (198.80, E. G. Pringsheim), from the Sammlung von Algenkulturen (Gottingen, Germany), was grown as previously described (15Gutiérrez-Cirlos E.B. Antaramian A. Vázquez-Acevedo M. Coria R. González-Halphen D. J. Biol. Chem. 1994; 269: 9147-9154Abstract Full Text PDF PubMed Google Scholar). Cytochrome c oxidase fromPolytomella sp. was obtained as previously described (14Pérez-Martı́nez X. Vázquez-Acevedo M. Tolkunova E. Funes S. Claros M.G. Davidson E. King M.P. González-Halphen D. J. Biol. Chem. 2000; 275: 30144-30152Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). Polyacrylamide gel electrophoresis was performed as in Schäggeret al. (18Schägger H. Link T.A. Engel W.D. von Jagow G. Methods Enzymol. 1986; 126: 224-237Crossref PubMed Scopus (219) Google Scholar) using 16% acrylamide gels. For tryptic digestion analysis, gels were stained with Amido Black, and the polypeptide of interest was excised from the gel. Polypeptides were isolated as previously described (15Gutiérrez-Cirlos E.B. Antaramian A. Vázquez-Acevedo M. Coria R. González-Halphen D. J. Biol. Chem. 1994; 269: 9147-9154Abstract Full Text PDF PubMed Google Scholar) for N-terminal sequencing. Amino-terminal Edman degradation was carried out on an Applied Biosystems Sequencer at the Laboratoire de Microséquençage des Protéines, Institut Pasteur, Paris, France. An 18.6-kDa polypeptide was isolated from polyacrylamide gels and subjected to tryptic and endolysin-C digestion and separation on DEAE-C14 and DEAE-C18 HPLC2 columns. Peaks eluted from the columns were subjected to N-terminal sequence analysis. Total DNA and total RNA fromPolytomella sp. and C. reinhardtii were isolated as previously described (14Pérez-Martı́nez X. Vázquez-Acevedo M. Tolkunova E. Funes S. Claros M.G. Davidson E. King M.P. González-Halphen D. J. Biol. Chem. 2000; 275: 30144-30152Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). PCR fragments were cloned into pMOS blue-T (Amersham Pharmacia Biotech) or pGEM-T easy (Promega). cDNA was prepared from 1–2 μg of total RNA with Moloney murine leukemia virus reverse transcriptase (Promega) or Superscript II reverse transcriptase (Life Technologies). All standard molecular biology techniques were as described (19Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York1989Google Scholar). Sequencing was carried out by the Kimmel Cancer Center DNA Sequencing Facility, Thomas Jefferson University, and at the Unidad de Biologı́a Molecular, Instituto de Fisiologı́a Celular, Universidad Nacional Autónoma de México. A genomicPolytomella sp. cox2b fragment was amplified by PCR using two degenerate primers: F1 (5′- CA(A/G) GA(C/T) AG(C/T) GC(C/T) AC(A/T) AG(C/T) CA(G/A) GC(C/T) CA(A/G) G-3′) based on internal sequence (QDSATSQAQA) of COX II, and B1 (5′-TG (G/A)TT (C/T)AA (A/G)CG (A/T)CC (A/T)GG (A/G)AT (A/G)GC (A/G)TC CAT-3′) based on the internal sequence MDAIPGRLNQ. The resulting 300-nt product was used to isolate genomic cox2b clones from a library ofPolytomella DNA PstI fragments of ∼2 kilobases in length in pBluescript. The cox2b cDNA sequence from Polytomella sp. was obtained by PCR and 5′-RACE (20Frohman M.A. Methods Enzymol. 1993; 218: 340-356Crossref PubMed Scopus (465) Google Scholar) using primers based on the genomic sequence obtained above. A poly(dT) tail was added to the 5′ end of the cDNA with a terminal transferase (Roche Molecular Biochemicals). The forward primer was oligo(dT)/adapter primer: 5′-GACTCGAGTCGACATCGATTTTTTTTTTTTTTTTT-3′, and the reverse primer (B2) was 5′-AGCTGTTTAAGACCATGACTTC-3′. Acox2a cDNA fragment was amplified using the degenerate primers F2 (5′-GA(G/A) GC(T/C) CC(T/C) GT(T/C) GC(T/C) TGG CAG CT(T/G) GG-3′), based on the N-terminal protein sequence EAPVAWQLG, and B3 (5′-CCA (A/G)TA CCA CTG (A/G)TG (A/G/T)CC (A/G)AT (A/G)GC C-3′), based on the internal conserved sequence KAIGHQWYW. Nested PCR was done with degenerate primers F3 (5′-CAG GA(T/C) TC(T/C) GC(T/C) AC(T/C) TC(T/C) CAG GC(T/C) CAG G-3′), based on the N-terminal sequence of the protein QDSATSQAQA and B4 (5′-GA(A/G) TA(A/G) AG(A/G) AG(A/G) GC(A/G) AA(A/G)GA(A/G) GG-3′), based on the internal conserved sequence PSFALLYS. For 3′-end cDNA cloning, oligo(dT)/adapter primer and primer F4 (5′-TCCTCTACCACATCGCCACCC-3′) were used. For nested PCR, adapter (5′-GACTCGAGTCGACATCGA-3′) and primer F5 (5′-ACTACACTAAGCAAGCTCTCCCTG-3′) were used. For 5′-RACE, primers B5 (5′-TCAGGGAGAGCTTGCTTAGTGTAG-3′), with oligo(dT)/adapter primer, and B6 (5′-TTGGTGGCGATGTGGTAGAGG-3′), with adapter for nested PCR, were used. Primers F6 (5′-AATGCTCGCCCAGCGTATC-3′) and B7 (5′-AAACCTTCACACACCCATAGGC-3′), derived from the 5′- and 3′-RACE products, were used to amplify the full-length cDNA of thecox2a gene. The genomic sequence of thePolytomella sp. cox2a gene was obtained by PCR amplification of total Polytomella sp. DNA with the same primers. The cDNAs for cox2a and cox2b were obtained by screening a C. reinhardtii cDNA library in λgt10 (21Franzén L.-G. Falk G. Plant Mol. Biol. 1992; 19: 771-780Crossref PubMed Scopus (26) Google Scholar) using the Polytomella sp. cox2a cDNA orcox2b genomic DNA as probes. A total of 17 positive clones were obtained from 5 × 104 plaque-forming units screened. PCR using two primers based on λgt10 sequences was used to identify the positive clones with the largest inserts. Phage DNA from these clones was isolated with the Qiagen Lambda mini kit. The 5′ end of the cox2a cDNA was completed by RACE using primers based on the cDNA sequence obtained. A bacterial artificial chromosome clone containingcox2b was obtained from a BAC genomic library from C. reinhardtii (22Lefebvre P.A. Silflow C.D. Genetics. 1999; 151: 9-14Crossref PubMed Google Scholar) by Genome Systems using the C. reinhardtii cox2b cDNA obtained above. BAC DNA was sequenced directly using internal primers. Mitochondrial targeting sequences were analyzed using MitoProt II (23Claros M.G. Comput. Appl. Biosci. 1995; 11: 441-447PubMed Google Scholar, 24Claros M.G. Vincens P. Eur. J. Biochem. 1996; 241: 779-786Crossref PubMed Scopus (1366) Google Scholar). The same program was used to calculate the segments with high local hydrophobicity () in a distance comprising 13 to 17 residues. The mesoH was determined by scanning each sequence for a maximum average hydrophobicity measured in windows from 60 to 80 residues and averaging the values (24Claros M.G. Vincens P. Eur. J. Biochem. 1996; 241: 779-786Crossref PubMed Scopus (1366) Google Scholar). More hydrophobicity scales were included to reduce the possibility of bias. Protein transmembrane stretches were predicted using the program TodPred II (25Claros M.G. von Heijne G. Comput. Appl. Biosci. 1994; 10: 685-686PubMed Google Scholar). Three-dimensional structure modeling was carried out using SWISS-MODEL (26Guex N. Peitsch M.C. Electrophoresis. 1997; 18: 2714-2723Crossref PubMed Scopus (9503) Google Scholar). The nucleotide sequences are in the DDBJ/EMBL/GenBankTM nucleotide sequence data base under the accession numbers AF305078 (Polytomella sp.cox2a cDNA), AF305079 (Polytomella sp.cox2b cDNA), AF305080 (C. reinhardtii cox2acDNA), AF305540 (C. reinhardtii cox2bcDNA), AF305541 (Polytomella sp. genomiccox2a), AF305542 (Polytomella sp. genomiccox2b), and AF305543 (C. reinhardtii genomiccox2b). An active cytochrome c oxidase fromPolytomella sp. was isolated as previously described (14Pérez-Martı́nez X. Vázquez-Acevedo M. Tolkunova E. Funes S. Claros M.G. Davidson E. King M.P. González-Halphen D. J. Biol. Chem. 2000; 275: 30144-30152Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). The four largest polypeptides with apparent molecular masses of 54.6, 29.6, 18.6, and 14.5 kDa were subjected to Edman degradation. The 54.6-kDa polypeptide, not susceptible to Edman degradation, was identified as subunit I of cytochrome c oxidase (COX I) since its mass was similar to that predicted for the mitochondrialcox1 gene sequence (54,781 Da (27Antaramian A. Coria R. Ramı́rez J. González-Halphen D. Biochim. Biophys. Acta. 1996; 1273: 198-202Crossref PubMed Scopus (16) Google Scholar)). The 29.6-kDa polypeptide (N-terminal sequence: SSDAGHHLSPRERYLV) was previously identified as subunit III (14Pérez-Martı́nez X. Vázquez-Acevedo M. Tolkunova E. Funes S. Claros M.G. Davidson E. King M.P. González-Halphen D. J. Biol. Chem. 2000; 275: 30144-30152Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). The N-terminal sequence of the 14.5-kDa polypeptide, DANSSELVLEPTRKYKAGLATRELW, did not show similarity with any sequences in GenBankTM. The N-terminal sequence of the 18.6-kDa polypeptide, EAPVAWQLGFQDSATSQAQA, was similar to COX II from other species, allowing identification of this polypeptide as COX II. A second sequence, MDAIPGR(R/L)NQIWLTINREG, was obtained during the Edman degradation of the 18.6-kDa polypeptide region at a yield of less than 50% that of the yield of the first sequence. This sequence also was similar to COX II from other species and was identified as an internal fragment of COX II that was obtained after partial cleavage of the protein during Edman degradation. On the basis of the primary amino acid sequences obtained for COX II, two degenerate oligodeoxynucleotide primers were designed. Using these primers, a PCR amplification product of 300 nt was obtained using total DNA from Polytomella sp. as a template. This PCR product encoded the C-terminal portion of thecox2 gene but lacked the region that encodes the N-terminal sequence of COX II. The absence of an N-terminal sequence was attributed to nonspecific annealing of the primer based on the N-terminal amino acid sequence. The 300-nt cox2 gene fragment hybridized to a 2-kilobasePstI fragment of Polytomella sp. total DNA in Southern analyses. To obtain a full-length gene, a mini-library was constructed from PstI fragments of ∼2 kilobases, and a positive clone was isolated and sequenced. This genomic sequence contained a 462-nt open reading frame that encoded a 153-amino acid protein homologous to the C terminus of COX II. The gene was namedcox2b. There was no open reading frame corresponding to the N-terminal portion of COX II in the 960 nt preceding the 462-nt open reading frame. Using primers based on the genomic sequence, a portion of a cDNA corresponding to the cox2b open reading frame was amplified from Polytomella sp. total RNA using reverse transcription-PCR. The remainder of the 5′-cDNA sequence was obtained by 5′-RACE. The full-length cDNA obtained contained a 65-nt 5′-untranslated region and a 462-nt open reading frame identical to that of the genomic clone. This showed that this gene encoded only the C-terminal portion of COX II and did not contain introns. The overall organization of the cox2b gene is shown in Fig.1. The predicted protein contained the MDAIPGRLNQIWLTINREG internal sequence that was obtained from direct protein sequencing as well as the sequence GQCSEICG known to be the COX II binding site for binuclear copper (28Tsukihara T. Aoyama H. Yamashita E. Tomizaki T. Yamaguchi H. Shinzawa-Itoh K. Nakashima R. Yaono R. Yoshikawa S. Science. 1996; 272: 1136-1144Crossref PubMed Scopus (1915) Google Scholar). The major N-terminal sequence of COX II, determined by Edman degradation, was not present in the deduced protein sequence. The N-terminal 43 amino acids lacked homology to any COX II proteins. The remaining sequence was homologous with the C-terminal half of many COX II proteins. The highest similarity was with COX II from the alga Prototheca wickerhamii (29Wolff G. Plante I. Lang B.F. Kück U. Burger G. J. Mol. Biol. 1994; 237: 75-86Crossref PubMed Scopus (143) Google Scholar). Altogether, these data suggested that thecox2 gene had been split into two genes inPolytomella sp. The PCR amplification product of the cox2b gene ofPolytomella sp. was also used to isolate a cox2bcDNA from a λgt10 cDNA library of C. reinhardtii. A comparison of this cox2b cDNA with the genomic sequence (see below) showed that this cox2b cDNA sequence lacked the first four codons of the cox2b gene and a 5′-untranslated region. It exhibited 85% identity with thePolytomella sp. cox2b (Fig.2). The cox2b cDNA fromC. reinhardtii was used to isolate cox2b from aC. reinhardtii BAC genomic library. The genomiccox2b sequence contained the complete coding region and was identical to the cDNA except for the presence of one intron of 187 nt, located 21 nucleotides upstream of the stop codon. There was no open reading frame encoding a known protein in the 1.7 kilobases preceding the start codon for the C. reinhardtii cox2b gene. The predicted COX IIB protein was 153 amino acids long and was 85% identical and 92% similar to COX IIB of Polytomella sp. Both C. reinhardtii and Polytomella sp. are therefore likely to use two genes to code for COX II. To clone the gene that encoded the N-terminal region of COX II, nested PCR was performed with primers derived from the major N-terminal sequence obtained from the protein, and internal sequences (KAIGHQWYW and PSFALLYS) were conserved among COX II proteins. UsingPolytomella sp. cDNA as a template, a 250-nt PCR product was obtained that exhibited similarity with other cox2genes. The full-length cDNA, obtained using 5′- and 3′-RACE (Fig.1), contained an open reading frame of 816 nt predicted to encode a protein of 271 amino acids including the sequence EAPVAWQLGF determined for the N terminus of the 18.6-kDa polypeptide. This gene was namedcox2a, since it encoded the N-terminal portion of the COX II protein (COX IIA). The N terminus of the mature Polytomella sp. COX IIA protein, derived from direct sequencing, corresponded to Glu-131 of the predicted sequence, indicating that this protein contains a mitochondrial targeting sequence of 130 amino acids. The mature protein is predicted to be 141 amino acids long and to contain two putative transmembrane stretches, from Ile-28 to Thr-48 and from Val-69 to Leu-89, and the highly conserved sequence GRQWYWSY present in all sequences of COX II subunits known to date (Fig. 2). The sequence from Glu-131 to Glu-246 was homologous with the N-terminal portion of many COX II proteins. This sequence was most similar to COX II from the algaP. wickerhamii. The COX IIA protein contained a C-terminal 20-amino acid region, lacking similarity to conventional COX II proteins, that had a high density of charged amino acids. Primers corresponding to the 5′ and 3′ ends of the cox2acDNA sequence of Polytomella sp. were used to amplify a portion of the cox2a nuclear gene from total DNA. This 1773-nt PCR product was cloned and sequenced. The genomiccox2a gene contained 6 introns, ranging in size from 84 to 136 nt. The Polytomella sp. cox2a cDNA was used to isolate a partial cDNA clone of cox2a from a C. reinhardtii cDNA library. The complete sequence of C. reinhardtii cox2a was obtained by 5′-RACE (Fig. 1). The C. reinhardtii cox2a cDNA contained a 5′- untranslated region of 30 nt, an open reading frame of 855 nt, encoding a protein of 284 amino acids, and a 3′-untranslated region of 201 nt. The predicted C. reinhardtii COX IIA mature polypeptide exhibited 72% identity and 81% similarity with the sequence predicted for the COX IIA protein from Polytomella sp. (Fig. 2). The gene sequence predicts an extension of 21 residues at the C-terminal end that lacked homology to COX II proteins but that was highly similar to the extension predicted by the cox2a gene from Polytomella sp. The nuclear location of bothcox2a and cox2b was confirmed by Southern analysis. After electrophoresis of Polytomella sp. total DNA through agarose gels, the mtDNA was detected as a discrete band below the bulk of the nuclear DNA (Fig.3 A, left panel). This was confirmed with a cox1 gene probe (27Antaramian A. Coria R. Ramı́rez J. González-Halphen D. Biochim. Biophys. Acta. 1996; 1273: 198-202Crossref PubMed Scopus (16) Google Scholar) that hybridized with the mtDNA and a β-tubulin gene probe (30Conner T.W. Thompson M.D. Silflow C. Gene. 1989; 84: 345-358Crossref PubMed Scopus (22) Google Scholar) that hybridized with the nuclear DNA (Fig. 3 A, left panel). The cox2a and cox2b genes ofPolytomella sp. hybridized with the major DNA fraction and not with the mtDNA band, confirming their nuclear localization. A similar analysis was carried out with total DNA from C. reinhardtii using a cox1 gene probe that hybridized with the mtDNA (31Gray M.W. Boer P.H. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 1988; 319: 135-147Crossref PubMed Scopus (99) Google Scholar) and a cytochrome c gene probe (32Amati B.B. Goldschmidt-Clermont M. Wallace C.J. Rochaix J.D. J. Mol. Evol. 1988; 28: 151-160Crossref PubMed Scopus (21) Google Scholar) that hybridized with the nuclear DNA (Fig. 3 A, left panel). Those results confirmed that the cox2a andcox2b genes were also nuclear-encoded in C. reinhardtii. To determine whether cox2a and cox2b were present as single-copy genes in the genomes of Polytomella sp. andC. reinhardtii, additional Southern blot analyses were performed. Single hybridization bands were obtained for thecox2a and cox2b genes of Polytomellasp. and C. reinhardtii with several restriction enzymes (Fig. 3, panel B and data not shown). This suggested that both genes were present in only one copy in their respective genomes. The expression of cox2a and cox2b ofPolytomella sp. and C. reinhardtii was examined by Northern analyses. Probes derived from the cox2a andcox2b genes hybridized to independent transcripts of sizes consistent with the corresponding cDNA sequences for both algae (Fig. 3, panel C). The presence of a larger, mature transcript that could suggest a transpliced product was not observed in any case. The cox2b gene of Polytomella sp. exhibited a double band. Since the genomic sequence of thiscox2b gene contained three putative polyadenylation sites in the 3′-noncoding region, it is possible that these bands correspond to mRNAs that have different sites of polyadenylation. There is a significant bias in codon usage in these genera of algae (30Conner T.W. Thompson M.D. Silflow C. Gene. 1989; 84: 345-358Crossref PubMed Scopus (22) Google Scholar), and this bias differs between mitochondrial and nuclear genes (14Pérez-Martı́nez X. Vázquez-Acevedo M. Tolkunova E. Funes S. Claros M.G. Davidson E. King M.P. González-Halphen D. J. Biol. Chem. 2000; 275: 30144-30152Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). Analysis of the codon usage for the cox2a andcox2b genes of Polytomella sp. and C. reinhardtii indicated that the codon usage was consistent with their nuclear localization (data not shown). In addition, the conserved polyadenylation signals TGTAA (33Silflow C. Rochaix J.-D. Goldschmidt-Clermont M. Merchant S. The Molecular Biology of Chloroplasts and Mitochondria in Chlamydomonas. Kluwer Academic Publishers Group, Dordrecht, Netherlands1998: 25-40Google Scholar), present in the vast majority of nuclear genes in the Chlamydomonad family, were present at the 3′ ends of the cDNA sequences of cox2a and cox2b for both algae. Since the primary protein sequences derived from the 18.6-kDa region of a polyacrylamide gel corresponded to the predicted amino acid sequences of both COX IIA and COX IIB fromPolytomella sp., it is likely that this region of the gel contained both subunits. To confirm that Polytomella sp. contained two independent COX II polypeptides, the purified cytochromec oxidase complex of this alga was subjected to matrix-assisted laser desorption-time of flight mass spectrometry analysis. The complex contained two major polypeptides in the expected mass range for COX IIA and COX IIB, one of 15,984 Da (theoretical 16,222 Da for mature COX IIA) and one of 17,169 Da (theoretical 17,219 Da for full-length COX IIB). The differences between the predicted and observed masses are likely due to post-translational modifications of the proteins. The 17,169-Da polypeptide was isolated by reverse phase-HPLC, trypsinized, and analyzed by matrix-assisted laser desorption-time of flight mass spectrometry analysis. The trypsin digestion products exhibited molecular masses that were consistent with the theoretical molecular masses expected for COX IIB peptides (TableI).Table IMass spectrometry analysis of the tryptic fragments obtained from the COX IIB subunit of Polytomella spSequence of the tryptic peptideTheoretical molecular massExperimental molecular massDaDaAFLTEYVK970.12970.448LVLPTNTLVR1125.371125.649LNQIWLTINR1270.481270.702LLVTASDV…AVPSLGIK2106.472123.172VPASQPIQ…DVQPGQLR2854.182854.383EGVFYGQC…VVEAISPR3141.563140.432The experimental values were obtained by matrix-assisted laser desorption-time of flight mass spectrometry analysis of tryptic digestion fragments of COX IIB purified by reverse transcription-HPLC from isolated cytochrome c oxidase. These masses are compared with the predicted molecular masses for the tryptic fragments from the cox2b gene product. Open table in a new tab The experimental values were obtained by matrix-assisted laser desorption-time of flight mass spectrometry analysis of tryptic digestion fragments of COX IIB purified by reverse transcription-HPLC from isolated cytochrome c oxidase. These masses are compared with the predicted molecular masses for the tryptic fragments from the cox2b gene product. We wished to determine whether COX IIB contained a cleavable mitochondrial-targeting sequence or if the novel N-terminal charged domain was present in the mature protein. The observed band of 18.6 kDa (containing COX IIA and COX IIB) was purified from polyacrylamide gels and subjected to digestion by trypsin or endolysin followed by HPLC separation and Edman degradation (Fig.4). Amino-ter
Summary It is accepted that Saccharomyces cerevisiae genome arose from complete duplication of eight ancestral chromosomes; functionally normal ploidy was recovered because of the massive loss of 90% of duplicated genes. There is evidence that indicates that part of this selective conservation of gene pairs is compelling to yeast facultative metabolism. As an example, the duplicated NADP‐glutamate dehydrogenase pathway has been maintained because of the differential expression of the paralogous GDH1 and GDH3 genes, and the biochemical specialization of the enzymes they encode. The present work has been aimed to the understanding of the regulatory mechanisms that modulate GDH3 transcriptional activation. Our results show that GDH3 expression is repressed in glucose‐grown cultures, as opposed to what has been observed for GDH1 , and induced under respiratory conditions, or under stationary phase. Although GDH3 pertains to the nitrogen metabolic network, and its expression is Gln3p‐regulated, complete derepression is ultimately determined by the carbon source through the action of the SAGA and SWI/SNF chromatin remodelling complexes. GDH3 carbon‐mediated regulation is over‐imposed to that exerted by the nitrogen source, highlighting the fact that operation of facultative metabolism requires strict control of enzymes, like Gdh3p, involved in biosynthetic pathways that use tricarboxylic acid cycle intermediates.
The atp6 gene, encoding the ATP6 subunit of F(1)F(0)-ATP synthase, has thus far been found only as an mtDNA-encoded gene. However, atp6 is absent from mtDNAs of some species, including that of Chlamydomonas reinhardtii. Analysis of C. reinhardtii expressed sequence tags revealed three overlapping sequences that encoded a protein with similarity to ATP6 proteins. PCR and 5'- and 3'-RACE were used to obtain the complete cDNA and genomic sequences of C. reinhardtii atp6. The atp6 gene exhibited characteristics of a nucleus-encoded gene: Southern hybridization signals consistent with nuclear localization, the presence of introns, and a codon usage and a polyadenylation signal typical of nuclear genes. The corresponding ATP6 protein was confirmed as a subunit of the mitochondrial F(1)F(0)-ATP synthase from C. reinhardtii by N-terminal sequencing. The predicted ATP6 polypeptide has a 107-amino acid cleavable mitochondrial targeting sequence. The mean hydrophobicity of the protein is decreased in those transmembrane regions that are predicted not to participate directly in proton translocation or in intersubunit contacts with the multimeric ring of c subunits. This is the first example of a mitochondrial protein with more than two transmembrane stretches, directly involved in proton translocation, that is nucleus-encoded.