Cloning, sequencing and expression of the two genes encoding the mitochondrial single-stranded DNA-binding protein in Xenopus laevis

Gene 184 (1997) 65–71

Cloning, sequencing and expression of the two genes encoding the mitochondrial single-stranded DNA-binding protein in Xenopus laevis Anne-Marie Champagne, Catherine Dufresne, Laure Viney, Monique Gue´ride * Centre de ge´ne´tique mole´culaire, CNRS, 91198 Gif-sur-Yvette, France Received 16 May 1996; accepted 24 June 1996; Received by D. Schlessinger

Abstract In Xenopus laevis the single-stranded DNA binding protein imported into the mitochondria consists of two highly related polypeptides. The establishment of the genomic nucleotide sequences reveals that they are encoded by two different genes, XLSSB1 and XLSSB2. The deduced amino acid sequence is identical to the direct amino acid sequence determined by Edman degradation of the mitochondrial polypeptides [Ghrir, R., Lecaer, J.P., Dufresne, C. and Gueride, M. (1991) Primary structure of the two variants of Xenopus laevis mtSSB, a mitochondrial DNA binding protein. Arch. Biochem. Biophys. 291, 395–400]. Both genes are organized in seven exons and six introns, the sequence of the peptide leader is interrupted by an intervening sequence (intron 2). The exon/intron junctions are in exactly conserved positions, splitting the same codon. A high level of identity is observed between corresponding introns of the two genes over part or most of their lengths. Structural features of intronic sequences reveal multiple rearrangements and exchanges during the evolution of X. laevis species. A CCAAT box and the potential regulatory elements NRF-2 and Sp 1 are observed in the 5∞-flanking region of both genes. During oogenesis, XLSSB gene expression is correlated with the replicative activity of the mitochondrial DNA. Keywords: Sequence comparison; Introns structure; Genome duplication; mRNA

1. Introduction In vertebrates most of the genetic information required for the biogenesis and the functions of mitochondria is nuclear, in particular all the components of the replication and transcription machinery are encoded by nuclear genes. In a previous study we characterized a single-stranded DNA binding protein, mtSSB, isolated from highly purified Xenopus laevis oocyte mitochondria. In vivo, the mtSSB is associated with the mitochondrial nucleoid: by electron microscopy it was shown to be bound to the single strand of DNA displaced during replication (Barat and Mignotte, 1981). In vitro, by its interaction with the template, it modulates the level of replication (Mignotte et al., 1988) and * Corresponding author. Tel.: +33 1 69823207; Fax +33 1 69824372; e-mail: [email protected] Abbreviations: aa, amino acid(s); bp, base pair(s); cDNA, DNA complementary to RNA; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; kb, kilobase or 1000 bp; MRP, mitochondrial RNA processing; mt, mitochondrial; NRF, nuclear respiratory factor; nt, nucleotide(s); SSB, single-stranded DNA-binding protein; TFA, transcription factor A. 0378-1119/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved PII S 03 7 8 -1 1 1 9 ( 9 6 ) 0 0 5 74 - 4

transcription (Barat-Gueride et al., 1989) mediated by homologous mitochondrial polymerases. The protein, tetrameric in its native form, consists of two slightly different polypeptides called mtSSBs and mtSSBr depending on their sensitivity to the limited proteolysis with a-chymotrypsin. Their primary structure was determined (Ghrir et al., 1991); the amino acid sequence of mtSSBs which agrees with the amino acid sequence deduced from the cDNA ( Tiranti et al., 1991) consists of 129 amino acids. Sequence comparison with other SSBs indicates a high degree of identity between regions of X. laevis mtSSB and domains of E. coli SSB known to be essential for replication, repair or recombination (Meyer and Laine, 1990). When compared with the mitochondrial SSB of Saccharomyces cerevisiae encoded by the nuclear gene RIM1 which is an essential component of the yeast mt DNA replication ( Van Dyck et al., 1992), there is 49% similarity (including conserved substitutions). A high level of conservation is observed among mature mitochondrial SSBs as more than 50% of similarity is observed between X. laevis, Drosophila melanogaster (Stroumbakis et al., 1994), rat (Hoke et al., 1990, Tiranti et al., 1993) and human ( Tiranti et al., 1993) proteins. Although precise functions of the

66

A.-M. Champagne et al./Gene 184 (1997) 65–71

mitochondrial SSB remain to be elucidated it seemed important to establish the nuclear sequence of the gene, to identify the putative regulatory elements and to measure the gene expression in relation with mtDNA replication. We report here the cloning and sequencing data of X. laevis genes encoding the two mtSSB polypeptides and their expression during oogenesis.

XLSSB2 gene coding for mtSSBr. It has been suggested that a whole genome duplication took place in X. laevis species 10–30 million years ago (Bisbee et al., 1977); this event could have produced the duplication of the SSB gene.

2. Results and discussion

Sequences of 10 068 and 6193 nucleotides have been established for XLSSB1 and XLSSB2 genes, respectively ( Fig. 2). Both genes are organized in seven exons, 52–177 nt long, and six introns, 356–5210 nt long. All intron/exon junctions ( Table 1) contain conserved GT and AG sequences at the 5∞- and 3∞-ends of the introns (Breathnach and Chambon, 1981). Since the cDNA sequence of XLSSB2 gene is not known, the 5∞-end of exon 1 corresponding to the transcription start point cannot be precisely localized. However, assuming that the 5∞-extremity of exon 1 is adjacent to the codon CAG (nt 877), exon 1 of XLSSB2 would be 109 nt in length and would share 88% identity with the 127 nt of exon 1 of XLSSB1. Exon 7 encodes 12 amino acids and includes the 3∞-untranslated regions. One polyadenylation signal sequence -AATAAA- is found 111 nt and 74 nt downstream from the stop codon TAG in XLSSB1 and XLSSB2, respectively. In the mature XLSSB1 mRNA the poly A stretch is added at the end of a 3∞-untranslated region of 136 nt. Comparison of the sequence encompassing the polyadenylation site, GTTTATTGaaatctcca, with the sequence of XLSSB2 gene, GTTTATTTaaatcttca, suggests a 126-nt 3∞-untranslated region sharing 77% identity with XLSSB1. In both genes the nucleotide sequence of the peptide leader (17 amino acids) is split into two parts by intron 2 between amino acids −10 and −9 (Gln-Val ). Introns 3, 4, 5 and 6 occur at identical positions splitting the same codons in both genes as indicated in Table 2.

2.1. Isolation of genomic clones for Xenopus laevis mtSSB genes Genomic clones containing SSB sequences were isolated by screening a X. laevis genomic library using xlmtssb cDNA ( Tiranti et al., 1991), which encodes the polypeptide mtSSBs (Ghrir et al., 1991), as hybridization probe. Seventeen positive clones were isolated; the size of the insert ranges from 11.6 to 19.9 kb. They could be organized in two main classes based on their restriction map and hybridization with the cDNA probe ( Fig. 1); this classification was confirmed later by sequencing data. Regions of interest of two clones from class I (7 and 11) and of four clones from class II (1, 3, 5 and 8) were subcloned in pBluescript II KS (−) and their nucleotide sequences were established using the exonuclease III/mung bean deletion method ( Fig. 2). Exons have been located with respect to the xl-mtssb cDNA sequence and to the amino acid sequences of the two polypeptides. From these data we infer that the clones in class I correspond to the XLSSB1 gene coding for mtSSBs while the clones in class II correspond to the

2.2. Genomic organization of XLSSB1 and XLSSB2 genes

2.3. Identification of putative regulation elements in the 5∞-flanking regions

Fig. 1. Physical map of genomic clones encoding X. laevis SSB genes. A genomic DNA library (Stratagene) of about 9.2×105 Lambda FIX II phages containing a partial Sau3AI digest of muscle DNA from an individual X. laevis was screened using the plaque hybridization method (Sambrook et al., 1989) and 32P-labeled xl-mtssb cDNA as probe. Seventeen positive clones were obtained. Phage DNAs were isolated from purified plaques, digested with restriction endonucleases, separated by electrophoresis on a 0.5% agarose gel and transferred onto Hybond N+ membrane (Amersham). Restriction fragments that hybridized with xl-mtssb cDNA probe are hatched underlined. NotI was used to excise the inserts: BamHI, #; EcoRI, n).

Upstream from the transcription start point, 1197 and 879 nt have been determined for XLSSB1 and XLSSB2 genes, respectively. Both sequences contain the same putative regulatory elements (Fig. 2), they lack TATA boxes (Gannon et al., 1979) but reveal CCAAT elements (Benoist et al., 1980). Though several GC boxes are not found, the sequence GGGGCGGAA could represent a binding site for the transcription factor Sp1 (Dynan and Tjian, 1983). Tandem binding sites for the nuclear respiratory factor 2 (NRF-2) ( Virbasius and Scarpulla, 1991), with the GGAA motif characteristic of the ETSdomains, are clustered around the transcription start point. Both genes lack sequences homologous to NRF-1

A.-M. Champagne et al./Gene 184 (1997) 65–71

a

67

b

Fig. 2. Nucleotide sequences of XLSSB1 and XLSSB2 genes and deduced aa sequences. Sequencing analysis was done using the exonuclease III/mungbean deletion kit according to the manufacturer’s protocol (Stratagene) and the nucleotide sequences were determined by the dideoxy chain termination method. Exons are indicated in capital letters, non-transcribed regions in lower case letters. The aa are given in the single letter code; they are numbered negatively in the peptide leader, positively in the mature polypeptide; the stop codon is marked with an asterisk. The putative transcription factor binding sites CCAAT, NRF-2, Mt1, Mt3 and Mt4 are underlined, Sp1 is overlined. The arrow above the nt sequence indicates the transcription start point. The polyadenylation signal is underlined; the nt sequences have been assigned the EMBL/GenBank/ DDBJ databases accession No. X83673 for XLSSB1 and No. X83674 for XLSSB2.

( Evans and Scarpulla, 1990), OXBOX or REBOX (Chung et al., 1992) elements present in a number of nuclear genes specifying mitochondrial respiratory functions. Sequences homologous to the Mt1, Mt3 and Mt4 elements (Suzuki et al., 1990), whose precise function has not yet been determined, are also present in the 5∞-flanking region of the XLSSB genes. Apart from these potential regulatory sequences no similarity was found between the two 5∞-flanking regions. 2.4. Structural features of intronic sequences Several observations emerge from a comparative survey of the intronic sequences.

Introns 1 and 5 of XLSSB1 as well as introns 1, 3 and 5 of XLSSB2 contain repeated sequences. Depending on the intron, the unit is 14–33 nt long and repeated 6–58 times. These repeated sequences represent 30–60% of the length of the corresponding intron and are flanked at their boundaries by short repeats, 6–13 nt. Similar sequences were found in the database but not as repeated elements. Introns having the same position in the two genes share sequences (≥75% identity) either on most of their length as introns 2 and 6 or only over a few tens of nucleotides (Fig. 3). Sequences of various lengths exhibiting at least 80% identity are observed in several introns. They have

68

A.-M. Champagne et al./Gene 184 (1997) 65–71

Fig. 3. Intronic sequence comparison of XLSSB1 and XLSSB2 genes. Optimal alignments between intronic sequences were obtained using BestFit program. The length in nucleotides are given in normal type or italicized numbers for similar or different sequences, respectively. The percentage of identity is indicated.

Table 1 Exon-intron junctions in XLSSB1 and XLSSB2 genes

structural features which could be reminiscent of some kind of mobile elements (class II ). As an example, one sequence 450 nt long contains a 23-nt palindromic sequence and a 100-nt inverted repeat at each extremity. It is duplicated in XLSSB1, intron 3 (nt 7017–7947), it is found with a few modifications four times in intron 1 (nt 4636–6475) and once in reverse in XLSSB2, intron 2 (nt 2384–2786). This sequence has similarity with a probable transposase in the 21S rRNA intron of S. cerevisiae (28% identity, 50% similarity over 28 aa residues corresponding to nt 7742–7825 in XLSSB1). In various introns of both genes such sequences of 100–500 nt exhibit 74–96% identity with sequences of Table 2 Intron spliced codons in XLSSB1 and XLSSB2 genes

a XLSSB1 gene. b XLSSB2 gene. c See Shapiro and Senapathy (1987). d Nucleotide combinations: M=A/C; R=A/G; Y=C/T; N= A/C/G/T.

A.-M. Champagne et al./Gene 184 (1997) 65–71

introns, 5∞- or 3∞-regions of other genes, always from X. laevis species. In intron 4 of XLSSB1, successive insertions seem to have taken place. A sequence of 326 nt (nt 8399–8724) shared by six X. laevis genes is found inserted inside a 136-nt sequence (nt 8257–8355 and nt 8901–8937) which has been observed in nine other X. laevis genes (in intronic or 5∞-flanking regions). 2.5. Sequence comparison Comparison of the nucleotide sequence of the exons of both genes reveals a high level of identity, 91–96% for exons 2–6 which correspond to the translated sequence. Their average GC content is 54% compared to 42% for the introns (the usual deficit in the dinucleotide CG is observed in all the introns). The restriction analysis and partial sequencing of several clones allow the identification of two allelic sequences for XLSSB1 and three for XLSSB2 gene. The same stretch of DNA of about 1650 nt, including exon 5 (91 nt), was sequenced in clones 1, 3 and 5 of XLSSB2 gene (Fig. 4). In 1091 nt (excluding the repeated region), one conservative substitution was found in exon 5, with only one addition and two substitutions at the same positions in the introns. Sequences of 739, 830 and 549 nt were determined in the repeated regions of intron 5 of clones 1, 3 and 5, respectively: the number of repeated

69

units of 19 nt is variable and a few identical substitutions (about one per unit) were observed at particular positions. Actually, in all cases, the repeated sequences account for the main differences between alleles. It was verified that the length of the repeated sequences had not been modified during subcloning in E. coli by comparison of the restriction profiles of plasmid subclones and original recombined l DNA. The deduced amino acid sequence of XLSSB1 gene is identical to the amino acid sequence previously determined by Edman degradation of the mature polypeptide mtSSBs (Ghrir et al., 1991). Two conservative substitutions were found between this genomic sequence and the cDNA sequence ( Tiranti et al., 1991): G (nt 8972)A and T (nt 10619)C; these differences reflect the polymorphism among the X. laevis population. Only 82 over 131 amino acid residues were identified for the mtSSBr polypeptide encoded by XLSSB2 gene, it had been suggested that the COOH extremities of both polypeptides could be identical. As predicted the COOH termini of both polypeptides are identical with the exception of the last residue LeuPro. One difference is observed between the direct and the deduced amino acid sequence from XLSSB2: at position 32 the nucleotide sequence predicts an Asp residue (codon GAC ), whereas the direct amino acid sequence showed Glu (codon GAG or GAA); as both amino acids are dicarboxylic this change has probably no consequence

Fig. 4. Comparison of XLSSB2 clones 1, 3 and 5. For the three clones, sequencing was done from the 5∞-end up to the close BamHI/EcoRI sites: BamHI, #; EcoRI, n; compare with Fig. 1). The base substitutions and the additional T are indicated. Repeated sequences are indicated by hatched line, numbers refer to their length in nucleotides.

70

A.-M. Champagne et al./Gene 184 (1997) 65–71

on the secondary and tertiary structure of the protein and may correspond to two allelic forms. As usual for most of the polypeptides imported into mitochondria, the nt sequence reveals the existence of a peptide leader. For both genes it consists of 17 amino acids with only one difference at position −12: Ala (codon GCT ) in mtSSBs is replaced by Val (codon GTC ) in mtSSBr, both amino acids being hydrophobic. 2.6. Expression of XLSSB genes During oogenesis of X. laevis which lasts about 2 years, the growth of the oocytes is achieved in successive steps at various rates. The level of mitochondriogenesis, in particular the increase in mtDNA mass, is correlated with cell growth (Callen et al., 1980); about 12 rounds of mtDNA replication have been performed when the oocyte reaches 0.3 mm in diameter at the end of the previtellogenic period, five more rounds take place up to the full size of the oocyte (1.3 mm). As the mtSSB protein is supposed to be essential for mtDNA replication, the gene expression could be modulated during oogenesis. The steady-state amount of mRNA was determined in oocytes at different stages. Oocytes from a young female were sorted out and divided into four classes according to their size. A ‘Northern’ blot analysis, using xl-mtssb cDNA as probe, clearly shows a much higher amount of mRNA in previtellogenic oocytes (diameter<0.5 mm) indicating some correlation between XLSSB gene expression and the mtDNA replication ( Fig. 5). The lengths and sequences of transcripts of the two genes are too close to allow their separation

Fig. 5. Expression of XLSSB genes during X. laevis oogenesis and in X. laevis cell cultures. Total RNA was extracted from X. laevis oocytes and XTC cell cultures (Pudney et al., 1973) with guanidium-thiocyanate followed by centrifugation through a CsCl cushion as described by Sambrook et al. (1989). Total RNA was denatured, separated by electrophoresis on a 1.25% agarose gel containing 0.66 M formaldehyde, transferred onto a Hybond-N membrane (Amersham) and hybridized successively with 32P-labeled xl-mtssb and GAPDH (Fort et al., 1985) probes. Lanes A and B: 25 mg of RNA from confluent (A) and exponential (B) XLTC X. laevis cell cultures. Lanes 1–4: 25 mg of RNA from oocytes of diameter<0.5 mm (1); 0.5–0.7 mm (2); 0.7–0.9 mm (3); 0.9–1.3 mm (4). mRNA from GAPDH is 1272 bp long, mRNA from XLSSB1 is 753 bp long.

on an agarose gel so we have no evidence of possible differential expression. As expected, no significant difference was observed between the mRNA steady state of exponential and early confluent X. laevis cell cultures.

3. Conclusions (1) The sequencing data reported here clearly show that the two polypeptides, mtSSBs and mtSSBr, correspond to two individual genes and not to allelic forms of a unique gene. This situation already described for many other X. laevis genes results probably from the duplication of the genome which is supposed to have arisen (10–30)×106 years ago (Bisbee et al., 1977). (2) Both genes consist of seven exons and six introns. Exons are more than 90% identical. A striking feature is that intronic sequences appear to have been subjected to multiple rearrangements and exchanges, through recombination or transposition, during the evolution of X. laevis: stretches of DNA of various lengths are shared by introns within one gene or between the two different genes. It is tempting to speculate that long sequences conserved in the two XLSSB genes with more than 80% identity as in intron 2, could reflect a possible function. (3) The expression of the XLSSB genes appears to be regulated during oogenesis. The level of mRNA is the highest in the class of oocytes which has the highest mt DNA replicative activity (Callen et al., 1980), indicating some correlation between the expression of the XLSSB genes and mtDNA replication. As the exact function of mtSSB remains to be elucidated in higher eukaryotes, experiments were carried out to determine if the XLSSB gene product could rescue the defect of the RIM1 gene in S. cerevisiae as the function of this gene was shown to be essential for the mtDNA replication apparatus ( Van Dyck et al., 1992). Several constructions were made with the yeast-inducible GAL10 promoter or the homologous RIM1 promoter upstream from the X. laevis cDNA sequence. In all cases the results were negative. (4) A CCAAT box and the potential regulatory elements NRF-2 and Sp1 are observed in the 5∞-flanking region of both XLSSB genes. Functional binding sites for nuclear respiratory factors, NRF-1 or NRF-2 with adjacent binding sites for the common transcriptional activator Sp1 have been observed in mammals, mainly in genes of the respiratory chain, but also in the promoter region of the mitochondrial transcription factor mtTFA ( Virbasius and Scarpulla, 1994) and of the MRP RNA gene ( Evans and Scarpulla, 1990) coding for the RNA associated with RNase MRP (Chang and

A.-M. Champagne et al./Gene 184 (1997) 65–71

Clayton, 1987), a possible component of the mitochondrial replication apparatus. Our results show that NRF elements are not restricted to mammals and strengthen the hypothesis that nuclear respiratory factors could contribute to the coordination of the expression of respiratory chain subunits with components required for mitochondrial transcription and replication.

Acknowledgement We acknowledge Dr. N. Touzet for providing the XTC X. laevis cell line. Financial support was provided by the Centre National de la Recherche Scientifique, the EU contract CHRX CT940491 Mitochondrial Biogenesis in Development and Disease and the Association Franc¸aise contre les Myopathies.

References Barat, M. and Mignotte, B. (1981) A DNA binding protein from Xenopus laevis oocyte mitochondria. Chromosoma 82, 583–593. Barat-Gueride, M., Dufresne, C. and Rickwood, D. (1989) Effect of DNA conformation on the transcription of mitochondrial DNA. Eur. J. Biochem. 183, 297–302. Benoist, C., O’Hare, K., Breathnach, R. and Chambon, P. (1980) The ovalbumin gene sequence of putative control regions. Nucleic Acids Res. 8, 127–142. Bisbee, C.A., Baker, A.C., Wilson, I., Haji-Azimi and Fishberg, M. (1977) Albumin phylogeny for clawed frogs (Xenopus). Science 195, 785–787. Breathnach, R. and Chambon, P. (1981) Organization and expression of eukaryotic split genes coding for proteins. Ann. Rev. Biochem. 50, 349–383. Callen, J.C., Dennebouy, N. and Mounolou, J.C. (1980) Development of the mitochondrial mass and accumulation of mtDNA in previtellogenic stages of Xenopus laevis oocytes. J. Cell. Sci. 41, 307–320. Chang, D.D. and Clayton, D.A. (1987) A novel endoribonuclease cleaves at a priming site of mouse mitochondrial DNA replication. EMBO J. 6, 409–417. Chung, A.B., Stepien, G., Haraguchi, Y., Li, K. and Wallace, D.C. (1992) Transcriptional, control of nuclear genes for the mitochondrial muscle ADP/ATP translocator and the ATP synthase b subunit. Multiple factors interact with the OXBOX/REBOX promoter sequences. J. Biol. Chem. 267, 21154–21161. Dynan, W.S. and Tjian, R. (1983) Isolation of transcription factors that discriminate between different promoters recognized by RNA polymerase II. Cell 32, 669–680. Evans, M.J. and Scarpulla, R.C. (1990) NRF-1: a trans-activator of

71

nuclear-encoded respiratory genes in animal cells. Genes Dev. 4, 1023–1034. Fort, Ph., Marty, L., Piechaczyk, M., El Sabrouty, S., Dani, C., Jeanteur, P., Blanchard, J.M. (1985) Various rat adult tissues express only one major mRNA species from the glyceraldehyde-3-phosphate multigenic family. Nucleic Acids Res. 13, 1431–1442. Gannon, F., O’Hare, K., Perrin, F., LePennec, J.P., Benoist, C., Cochet, M., Breathnach, R., Royal, A., Garapin, A., Cami, B. and Chambon, P. (1979) Nature 278, 428–434. Ghrir, R., Lecaer, J.P., Dufresne, C. and Gueride, M. (1991) Primary structure of the two variants of Xenopus laevis mtSSB, a mitochondrial DNA binding protein. Arch. Biochem. Biophys. 291, 395–400. Hoke, G.D., Pavco, P.A., Ledwith, B.J. and Van, T.G. (1990) Structural and functional studies of the rat mitochondrial single strand DNA binding protein P16. Arch. Biochem. Biophys. 282, 116–124. Meyer, R.R. and Laine, P.S. (1990) The single-stranded DNA-binding protein of Escherichia coli. Microbiol. Rev. 54, 342–380. Mignotte, B., Marsault, J. and Barat-Gueride, M. (1988) Effects of the Xenopus laevis mitochondrial single-stranded DNA-binding protein on the activity of DNA polymerase c. Eur. J. Biochem. 174, 479–484. Pudney, M., Varma, M.G.R. and Leake C.J. (1973) Establishment of a cell line ( XTC-2) from the South African clawed toad, Xenopus laevis. Experientia 29, 466–467. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Shapiro, M.B. and Senapathy, P. (1987) RNA splice junctions of different classes of eukaryotes: sequence statistics and functional implications in gene expression. Nucleic Acids Res. 15, 7155–7174. Stroumbakis, N.D., Li, Z. and Tolias, P.P. (1994) RNA- and singlestranded DNA-binding (SSB) proteins expressed during Drosophila melanogaster oogenesis: a homolog of bacterial and eukaryotic mitochondrial SSBs. Gene 143, 171–177. Suzuki, H., Hosokawa, Y., Toda, H., Nishikimi, M. and Ozawa, T. (1990) Common protein-binding sites in the 5∞-flanking regions of human genes for cytochrome c and ubiquinone-binding protein. 1 J. Biol. Chem. 265, 8159–8163. Tiranti, V., Barat-Gueride, M., Bijl, J., DiDonato, S. and Zeviani, M. (1991) A full-length cDNA encoding a mitochondrial DNA-specific single-stranded DNA binding protein from Xenopus laevis. Nucleic Acids Res. 19, 4291. Tiranti, V., Rocchi, M., DiDonato, S. and Zeviani, M. (1993) Cloning of human and rat cDNAs encoding the mitochondrial singlestranded DNA-binding protein (SSB). Gene 126, 219–225. Van Dyck, E., Foury, F., Stillman, B. and Brill, S.J. (1992) A singlestranded DNA binding protein required for mitochondrial DNA replication in S. cerevisiae is homologous to E. coli SSB. EMBO J. 11, 3421–3430. Virbasius, J.V. and Scarpulla, R.C. (1991) Transcriptional activation through ETS domain binding sites in the cytochrome c oxidase subunit IV gene. Mol. Cell. Biol. 11, 5631–5638. Virbasius, J.V. and Scarpulla, R.C. (1994) Activation of the human mitochondrial transcription factor A gene by nuclear respiratory factors: A potential regulatory link between nuclear and mitochondrial gene expression in organelle biogenesis. Proc. Natl. Acad. Sci. USA 91, 1309–1313.