Structural Organization of Nuclear Gene for Subunit Vb of Mouse Mitochondrial Cytochrome c Oxidase*

We have reported recently the isolation of a cDNA for nuclear encoded subunit Vb of mouse cytochrome c oxidase by screening mouse bone marrow and kidney cDNA libraries. In the present study, this cDNA was used as a probe to screen a mouse genomic library and isolate the complete gene encoding subunit Vb. South- ern blot hybridization of mouse genomic DNA with the cDNA probe suggested the occurrence of multiple genes including many retroinserts. Restriction analysis followed by Southern blot hybridization of genomic clones was used to identify the putative retroinserts from the intron containing genes. Of the 10 initial genomic clones isolated, one clone (MG3) showing the most complex hybridization pattern was found to contain the complete gene for subunit Vb. The DNA sequence analysis show that the subunit Vb gene contains four exons of 149, 73, 99, and 189 bases interrupted by three relatively small introns of 520, 165, and 648 nucleotides in a gene spanning about 2.5 kilobase pairs. As determined by a combination of primer extension and S 1 protection analyses, the major transcription start site appears to be located 49 nucleotides upstream of the translation initiation codon. The ability of the 5' upstream DNA to initiate transcription was studied using the chloramphenicol acetyltransferase (CAT) expression plasmids in NIH 3T3 cells. Using this system we observed that a segment of the gene spanning nucleotides -574 to +45 can drive the transcription of CAT gene in an orientation dependent manner. The upstream region of subunit Vb


Structural Organization of Nuclear Gene for Subunit Vb of Mouse
Mitochondrial Cytochrome c Oxidase* (Received for publication, October 16,1990) Aruna Basu and Narayan G. Avadhanit From the Laboratories of Biochemistrv. DeDartment of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, Pennsylv&aA19104 ' We have reported recently the isolation of a cDNA for nuclear encoded subunit Vb of mouse cytochrome c oxidase by screening mouse bone marrow and kidney cDNA libraries. In the present study, this cDNA was used as a probe to screen a mouse genomic library and isolate the complete gene encoding subunit Vb. Southern blot hybridization of mouse genomic DNA with the cDNA probe suggested the occurrence of multiple genes including many retroinserts. Restriction analysis followed by Southern blot hybridization of genomic clones was used to identify the putative retroinserts from the intron containing genes. Of the 10 initial genomic clones isolated, one clone (MG3) showing the most complex hybridization pattern was found to contain the complete gene for subunit Vb. The DNA sequence analysis show that the subunit Vb gene contains four exons of 149, 7 3 , 99, and 189 bases interrupted by three relatively small introns of 520, 165, and 648 nucleotides in a gene spanning about 2.5 kilobase pairs. As determined by a combination of primer extension and S 1 protection analyses, the major transcription start site appears to be located 49 nucleotides upstream of the translation initiation codon. The ability of the 5' upstream DNA to initiate transcription was studied using the chloramphenicol acetyltransferase (CAT) expression plasmids in NIH 3T3 cells. Using this system we observed that a segment of the gene spanning nucleotides -574 to +45 can drive the transcription of CAT gene in an orientation dependent manner. The upstream region of subunit Vb gene lacks the TATA and CAAT elements, although it contains several GC rich elements and a pyrimidine rich stretch around the transcription start site.
Cytochrome c oxidase (EC 1.9.3.1), the terminal oxidase of the mitochondrial electron transport chain, is an oligomeric complex containing seven to nine subunits in the lower eukaryotes like slime mold and yeast and 13 subunits in all of the animal cells examined thus far (1,2). The biogenesis of cytox' is a complex process requiring the coordinate expression of both nuclear and mitochondrial genes. The three largest subunits of the complex involved in the catalytic function are encoded by the mitochondrial genomes and the remaining smaller subunits are coded by the nuclear genes and targeted to mitochondria by a post-translational mechanism (2,3). Although the precise functions of the nuclear encoded subunits remain unknown, a number of these subunits in the animal cells are expressed as tissue-specific isologs, and some of them respond to hormonal and developmental signals (2,4). Furthermore, mutational studies using homologous recombination in the yeast have demonstrated that subunits like IV, Va, plus Vb and VIIa are necessary for the biogenesis of functional cytox ( 5 ) . In support of this, recent clinical studies on human patients with neuromuscular and renal dysfunctions imply that the aberrant expression of some of the nuclear encoded subunit(s) might be the reason for cytox deficiencies and resulting mitochondrial dysfunction in these diseases (6)(7)(8). Thus, cytox provides an attractive model system to study the molecular mechanisms of nuclear and mitochondrial gene interaction, and also the biogenesis of a complex oligomeric enzyme whose function is affected in a number of genetic and metabolic diseases.
The genes for all of the six nuclear coded subunits of cytox in yeast have been extensively characterized, and the nature of cis-acting DNA elements and trans-acting factors involved in the expression of some of these genes are being characterized (5, 9,10). Similarly, the structural features of the gene coding for subunit V (mammalian counterpart of subunit IV) of Neurospora and the respiratory defects resulting from deletions in this gene have been reported (11). Although cDNA for three different tissue-specific isologs, and also a number of generally expressed subunits of cytochrome c oxidase from various animal sources have been characterized (1,2,(12)(13)(14)(15)(16)(17), thus far the structural organization of genes for only two of the subunits, i.e. subunit IV of rat (18) and mouse (19) and VIc of rat (20), have been characterized. The occurrence of multiple genes, including pseudogenes, and retroinserts for the majority of nuclear encoded subunits in the animal cells has presented difficulty for the successful isolation of the functional genes. In a previous study from this laboratory, we reported the isolation and sequencing of cDNA for mouse cytox subunit Vb (21). In the present paper we report the complete nucleotide sequence of subunit Vb gene containing a functional 5' transcription promoter sequence. Isolation and Sequencing of Genomic DNA-A library of adult DBA 25 mouse liver genomic DNA in phage EMBL-3 vector was screened with 'I'P labeled DNA probe as described by Maniatis et al. (22).

Materials-Almost
Escherichia coli NM538 cells were infected with about 5 X lo4 plaqueforming units and plated on a 135-mm LB plate. About lo6 plaques from 20 plates were lifted on the Nytran membranes and hybridized with the labeled cDNA for subunit Vb described recently from this laboratory (21). Positive clones were isolated by repeated screening, and the DNA from resultant clones were isolated on a preparative scale by plate lysis followed by adsorption to Lambda Sorb, an affinity matrix from Promega Biotec Corp., and phenol extraction. The genomic DNA inserts (8-14 kb) from the resultant DNA preparations were released by partial digestion with Sal1 alone or Sal1 and BamHI and cloned in pGEM 32 or 72 plasmids. Restriction digests of these clones were subcloned to generate smaller and overlapping inserts of 0.8-1.5 kb for further characterization. Miniprep double-stranded DNA was sequenced by the dideoxy chain termination method of Sanger et al. (23), using the T7 polymerase based sequencing kit.
Southern Blot Analysis-DNA was isolated from Ehrlich ascites mouse tumor cells using the proteinase K digestion method (22). About 10 pg of DNA was digested overnight with EcoRI, HindIII, and BamHI and resolved by electrophoresis on a 0.8% agarose gel. About 0.5 pg of DNA from the genomic clones were digested with the appropriate enzymes for 2 h and resolved on a 1.0% agarose gel. The DNA was transblotted to Nytran membrane by the standard procedures (22,24) using 10 X SSC (1 SSC = 0.15 M NaCl, 0.015 M sodium citrate, pH 7.0) as the transfer medium. The blots were prehybridized in a buffer containing 50% formamide, 6 X SSC, 0.1% potassium pyrophosphate, 5 X Denhardt's (1 X Denhardt's = 0.02% Ficoll, 0.02% polyvinylpyrollidone, 0.02% bovine serum albumin), 150 kg/ ml denatured calf thymus DNA, and 0.5% sodium dodecyl sulfate for 6 h a t 42 "C. The blots were then hybridized with the '"P-labeled probes (2 X lo6 cpm/ml) in the prehybridization buffer for 16-24 h at 42 "C, washed under moderate stringency conditions specified in the Schleicher & Schuell manual, and exposed to Kodak SB5 x-ray film for 12-16 h.  (21), was synthesized and end labeled to a specific activity of 5 X lo5 cpm/ng using [y-:"P]ATP and polynucleotide kinase (22). For primer extension analysis, 10 ng, of labeled oligonucleotide was annealed to 50 pg of total RNA isolated from mouse liver in 0.12 M KC], 1 mM EDTA, 10 mM Tris-C1 (pH 8.0), and 2 unitslpl of RNasin in a total volume of 16 pl and incubated in a water bath a t 68 "C for 20 min. The water bath was steadily cooled down to 37 "C over a period of 3 h as described before (25). The reaction mixture was adjusted to contain 75 mM Tris-C1 (pH 8.0), 10 mM MgC12, 10 mM dithiothreitol, 70 mM KC1, 800 p~ each of dATP, dTTP, dCTP, and dGTP, and 1 unit/pl of avian myeloblastosis virus reverse transcriptase in a total volume of 32 pl. The reaction was carried out at 42 "C for 2 h and stopped by adding 5 mM EDTA. The reaction products were extracted with phenol-chloroform, precipitated with ethanol, denatured by heating a t 80 "C for 5 min in 80% formamide, and analyzed on a 6% polyacrylamide, 8 M urea sequencing gel. A sequencing ladder of the genomic DNA containing the 5' region of the gene was generated by using the same 23-mer oligonucleotide and used as the size marker.
SI Nuclease Protection-The same primer used for the primer extension was used to generate a 187 nucleotides long S1 probe. 5' end-labeled primer (100 ng) was annealed to 15 pg of alkali denatured DNA from a genomic subclone containing the first 1-kb region of the gene and extended using the Klenow fragment of DNA polymerase I (22). The double-stranded DNA thus generated was digested with AccI, and the 187-nucleotide single-stranded DNA probe was gelpurified. About 2 X lo5 cpm of the probe was hybridized with 100 pg of mouse liver RNA in 80% formamide, 40 mM PIPES (pH 6.4), 1 mM EDTA, and 0.4 M NaCl a t 42 "C overnight and treated with 100-200 Vogt units of S1 nuclease at 37 "C for 30 min essentially as described before (26). The S1-resistant fragments were denatured and resolved on a 6% sequencing gel as described for the primer extension analysis. Assays for Transcription Promoter Actiuity-The SalI-AuaI segment of the gene spanning nucleotide number -574 to +45 relative to the mRNA cap site was tested for transcription promoter activity using expression plasmids containing the bacterial CAT reporter gene. The two plasmids used in this study have a common pUC19 backbone and intact CAT coding sequence. One of them named pCAT-basic lacks both promoter and enhancer elements, whereas the second plasmid named pCAT-enhancer contains an SV40 enhancer element, but no promoter (Promega Corporation). For cloning the SalI-AuaI fragment in the sense orientation, the AuaI end was bluntended by gap filling with dNTPs and T4 polymerase (22), and the 638-bp DNA was released by digesting with SalI. This fragment contains the region +45 to -574 of the gene, and a 19 nucleotide long polylinker sequence at the upstream end 5' to the cap site. The pCAT-basic and pCAT-enhancer plasmids were first cut with XbaI, and the protruding ends were blunt-ended by gap filling. The plasmid was subsequently cut with Sal1 and ligated to the 638-bp DNA modified as described above. For cloning the SalI-AuaI insert in the reverse orientation, the pCAT-basic plasmid was first cut with HindIII, blunt-ended by gap filling, and the modified 638-bp DNA was subsequently cloned in the blunt ended HindIII and Sal1 sites.
Using this method three different plasmid constructs were generated, each containing the cytox gene insert immediately upstream of the CAT gene as follows: pCAT-Bs (basic, sense orientation), pCAT-Br (basic, reverse orientation), and pCAT-Es (enhancer containing, sense orientation) The promoter activities of the plasmid constructs were tested by transfecting the DNA to NIH 3T3 cells. The cells were maintained in 75-cm2 flasks using Dulbecco's modified Eagle's high glucose medium supplemented with 2 mM glutamine, 50 pg/ml gentamycin, and 10% heat-inactivated fetal bovine serum. For transfection experiments, the cells were seeded a t a density of 8 X lo5 cells/lO cm dish and transfected 24 h later with 15 pg of DNA using the calcium phosphate coprecipitation method (27). After a 40-h incubation, the cells were harvested, and the cell extracts were assayed for protein content, and CAT activity. CAT assays were performed using ["C] chloramphenicol as the substrate and 25-50 pg of cellular protein as the enzyme source as described by Gorman et al. (28). Acetylated forms of chloraphenicol were separated by thin layer chromatography on Silica gel plates and viewed by autoradiography. The extent of conversion was quantitated by scanning the x-ray films through an LKB Ultroscan-XL (Pharmacia).

RESULTS AND DISCUSSION
Strategy for the Isolation of Cytochrome Oxidase Vb Gene-The complexity of cytox Vb gene in the mouse genome was determined by Southern hybridization of restriction enzyme digested genomic DNA with "'P-labeled cDNA probe. As shown in Fig. lA, the cDNA hybridized with multiple bands of 2-10 kb from mouse genomic DNA digested with EcoRI, HindIII, and BamHI, suggesting the existence of multiple genes. A previous study on the Northern blot analysis of RNA, and sequencing of cDNA, however, failed to detect any significant tissue specific heterogeneity in mouse subunit Vb (21). The results of Southern hybridization thus suggested the possible occurrence of pseudogenes as predicted for some of the cytochrome oxidase genes and also ribosomal protein genes (1, 2, 18, 19, 29-31).
Initially, we screened about IO6 plaque-forming units of mouse liver genomic DNA library in EMBL-3 vector using 3'P-labeled subunit Vb cDNA probe and isolated 10 positive clones. Inserts of 8-14 kb were extensively analyzed by restriction mapping, and the fragments hybridizing with the subunit Vb cDNA probe were subcloned in pGEM 32 and 72 plasmids for sequencing. Although not presented here, the first two clones we sequenced turned out to be the fully processed retroposon type of pseudogenes. We therefore decided to search for the intron-containing gene using a restriction mapping strategy. As shown in Fig. 2, the genomic DNA inserts were digested with NcoI and DdeI restriction enzymes and probed with "P-labeled NcoI-DdeI fragment of cDNA by Southern blot hybridization. The choice of these two restriction enzymes was based on our previous observations that the mouse cytox Vb cDNA contains a unique NcoI site close to the 5' end and a DdeI site 300 nucleotides downstream toward the 3' end. The Southern blot in Fig. 2 shows the restriction patterns of four different genomic DNA clones and the Vb cDNA digested with NcoI and DdeI. It is seen that two of the genomic clones contain a 300-bp fragment similar to the fragment from the cDNA digest suggesting the possible absence of introns in these genomic clones. Although not shown the remaining five genomic clones showed a similar 300-bp NcoI-DdeI fragment. The genomic clone designated as MG9 yielded a single band of 500 nucleotide, suggesting that it may be a partially processed gene, whereas clone MG3 exhibited the most complex pattern with fragments of 500 and 750 bp (see Fig. 2). The latter clone was further characterized and found to harbor the complete subunit Vb gene. Hybridization of mouse genomic DNA blots with a probe specific for the first intron of the gene yielded single bands (see Fig. lB), suggesting that there may be a single copy of the complete intron containing gene in the mouse genome. Our results, however, do not rule out the possible existence of an evolutionarily divergent Vb gene in the mouse genome.
Structural Features of Cytox Vb Gene-As shown in Fig. 3, subunit Vb gene contains one site each for SacI, BglI, and AccI, two sites for NcoI and AuaI, and four sites for HincII in a stretch of about 2.5-kb DNA. Restriction fragments of 0.5-1.5-kb DNA were subcloned in pGEM 32 or 72 plasmids for sequencing. Over 80% of the gene was sequenced in both directions using the strategy shown in Fig. 3.
The complete nucleotide sequence of the subunit Vb gene has been presented in Fig. 4. The gene contains four exons of 149, 73, 99, and 189 bp, which are interrupted by introns of 520, 165, and 648 nucleotides in the 5' to 3' direction. The first intron starts 10 nucleotides downstream from the N terminus of the mature protein. The remaining two introns also occur within the protein coding region of the gene. The intronlexon boundary sequences conform to published consensus sequence (32). The intron splice phase is type 1 (interrupting the codons) for the first and third introns, and the second intron is type 0, since it does not interrupt the codon. As seen from Fig. 4, each splice donor begins with a GT and each acceptor site ends with an AG, preceded by a polypyrimidine stretch. Furthermore, the exon sequences are identical to the corresponding cDNA sequence (21)

1626
"""""__________""""""""""""""...."..."""""."""""""""""""""" The hydropathy profile of mouse cytox Vb protein with respect to exon-intron organization has been presented in Fig.  5. As previously shown for this subunit from bovine heart (41, the protein shows clusters of hydrophilic domains as expected of a subunit extrinsic to the membrane. Although the intron positions coincide with the transition regions, the functional implications of these domains remain unclear at this time. It is also seen that the N-terminal signal sequence shows an unusual biphasic nature with the second half of the signal showing hydrophilicity. Furthermore the cytox Vb presequence contains Thr at position -4, Ala at position -7, and Arg at position -9, similar to the "conserved three-amino a c i d motif predicted for mitochondrial proteins undergoing a two-step proteolytic processing (33). The relative location of the hydroxylated hydrophobic (although weakly), and the basic amino acid residues in the cytox Vb protein, however, is shifted by one residue as compared with the predicted locations in other proteins (33). In view of models on cytox subunit topology (2, 4) suggesting that subunit Vb is an ectopic protein facing the cytosolic side of the membrane, it remains to be seen if the putative two step processing signal has any role in the transport of the protein to cytoplsmic side of the innermembrane. The transcription start site(s) of cytox Vb gene was determined using a combination of primer extension and S1 nuclease protection analysis. The same 23-mer synthetic oligonucleotide containing the anti-sense sequence of the translated region (see Fig. 4) was used as the primer for the RNAdependent extension and also for preparing the S1 probe as described under "Experimental Procedures." As shown in Fig.  6 (lane 2 ) , primer extension against mouse liver RNA template shows a major and longest product corresponding to a C residue located 49 nucleotides upstream from the translation initiation site. It is also seen from lanes 9 and 10 that the major and also the largest fragment resistant to S1 digestion is of identical size. These results show that the major transcription of cytox Vb gene begins at 49 nucleotides upstream of the translation start site. This position, the putative mRNA cap site has been designated as +1 nucleotide of the gene. The results of primer extension and the S1 analysis also suggest the possible occurrence of minor transcription initiations at about +9 and +20 nucleotides of the gene.

G C T G G~T G A C T G T C r U C C G T~T~C C T T T C A G G T G r U G A~A C M C l G T A C~~A l C~G T T T l G G C~A C~G G C G A G A G~A~A l~C C C A A
Characteristics of the 5'-Upstream Sequences-The sequence upstream of the transcription initiation site revealed several structural features which might be important in transcription regulation. As shown for a number of housekeeping genes and genes coding for some of the mitochondrial proteins t Exon 1 I Exon 2 I Exon 3

2 G A T C 3 4 5 6
v "~ " :; Localization of transcription start site b y p r i m e r extension and S1 nuclease protection methods. The 5' endlabeled 23-mer oligonucleotide complementary to the mRNA (see Fig.  4) was annealed to 50 pg of mouse liver RNA or E. coli tRNA and extended as described under "Experimental Procedures." A 5' endlabeled 187 nucleotide single-stranded S1 probe was prepared by using the same 23-mer primer. About 2 X lo5 cpm of the probe was hybridized with 100 pg of mouse liver RNA or E. coli tRNA and treated with S1 nuclease for 30 min at 37 "C also as described under "Experimental Procedures." A sequencing ladder was created using the same primer and 5 pg of alkali denatured DNA from genomic clone MG(3). Lane I , primer extension with tRNA; lane 2, primer extension with mouse liver RNA; lane 3, S1 probe hybridized to tRNA without nuclease treatment; lane 4, probe hybridized to tRNA was treated with 100 units of S1 nuclease; lane 5, probe hybridized to mouse liver RNA was treated with 100 units of S1 nuclease; and lane 6, sample as in lane 5 was treated with 200 units of S1 nuclease. The nucleotide sequence of this region has been given and the major transcription start site has been indicated with an arrow. (34)(35)(36)(37), the cytox Vb gene lacks the TATA and CAAT sequences which act in concert as the transcription promoter in a wide variety of cells from lower eukaryotes and animal cells. The sequences around the major transcription start site are rich in pyrimidine residues and show part homology to the "initiator" sequence (PyPyCAPyPyPyPyPy) found in a number of TATA containing genes (38). There are also a number of GC-rich sequences showing partial or complete homology to the transcription factor SP1 binding sites (39)(40)(41). The sequence GGCGGCGGG at positions -133 to -141 and partly inverted sequence CGCCCC a t positions -145 to -150 are some of the examples. In addition there is one enhancer core like sequence GTGGAAG (positions -288 to -294), which is nearly identical to the consensus sequence GTGGAAAG (42). In the vicinity of GC boxes we found sequence TGACTGA a t positions -156 to -162, which closely resembles the GCN4 or AP-1 transcription factor binding sequence TGAG(C)TCA (43,44). In addition, there is a sequence CAGCTAGCGG on the opposite strand at positions -169 to -178 that is homologous to the AP-4 binding site of the SV40 late promoter (45). A similar stretch of sequence has recently been shown to occur a t about 123 nucleotides upstream of the major transcription start site of rat cytox subunit IV gene (18). An octanucleotide sequence TTCGTGGT partially overlapping the putative enhancer core sequence is located at positions -293 to -300. This sequence is reminiscent of the UAS2 element (TNPyTTGGT) which is known to be present in the upstream regions of Yeast nuclear genes coding for mitochondrial proteins (46)(47)(48)(49) and implicated in binding transcription factor Hap 2/Hap 3. Similar sequences are detected in nuclear genes coding for cytox subunit VIc (20), subunit IV (18,19), and also other mitochondrial heme proteins in animal cells (50,51).
Promoter Activity of the 5"Flanking Sequence-The ability of the 5"flanking region of cytox Vb gene to promote transcription initiation was examined using the CAT expression plasmids. The 638-bp DNA containing the genomic sequence +45 to -574 with respect to the cap site and 19 nucleotide region of the polylinker at the 5' end was cloned in the sense or anti-sense orientation in pCAT-basic and pCAT-enhancer plasmids (see "Experimental Procedures") and used for transfecting NIH 3T3 cells. As shown in Fig. 7, the 638-bp fragment introduced in the sense orientation in the promoter/enhancerless plasmid (pCAT-Bs) directed the expression of CAT to the extent of 35% conversion of the input substrate. When this fragment was placed in the sense orientation downstream of SV40 enhancer plasmid (pCAT-Es) the expression is increased significantly, yielding about 56% conversion (see  7). However, the same fragment cloned in the anti-sense orientation was unable to drive the transcription of CAT gene to any significant level. The failure to drive the CAT activity in the anti-sense orientation was not due to an aberrant translation initiation, since the insertion of the 638-bp DNA did not introduce any additional ATG codons. Furthermore, although not shown the DNA sequence from the polylinker region, including the 19 nucleotides attached to the cytox Vb upstream sequence, was unable to promote the transcription in either directions suggesting that the observed activity is indeed due to the upstream genomic DNA sequence. Thus, as shown for a number of transcription promoter elements the cytox Vb element used in these experiments is directional in nature.

Cytochrome Oxidas
The ability of the 619-bp upstream DNA to drive the transcription of CAT gene in transient expression studies (Fig. 7) in an orientation-dependent manner suggests that this region houses the putative transcription promoter elements. These results along with the results of Southern blot analysis presented in Fig. 1B provide direct proof that the subunit Vb gene described here is truly the expressed gene. As seen from Fig. 4, the gene lacks the classical TATA and CAAT consensus motifs. The absence of these structures and the occurrence of GC-rich boxes are the characteristics of transcription promoters of housekeeping genes which are transcribed in most tissues, although similar features are found in some genes expressed in a tissue-specific manner (52)(53)(54)(55). Genes with such features typically have several transcription start sites. In keeping with these observations, the results of primer extension and S1 protection experiments (Fig. 6) show multiple transcription start sites for cytox Vb gene as well. In many respects the promoter region of the subunit Vb gene resembles the TATA less subunit IV gene from both rat and mouse (18,19) and differs from the rat VIc gene which contains typical TATA and CAAT elements (20).
Although the promoter element(s) responsible to drive the transcription in the transient expression assay system has not yet been fine mapped, the 180-nucleotide region upstream of the cap site might be critical for the promoter function. The presence of the putative SP1, AP1, and AP4 binding sites within this region may be of special significance for transcription initiation. Some of the GC-rich elements can also function as enhancer elements in presence of transcription activator factors such as SP1 or ETF (56, 57) and as repressor elements in the presence of inhibitory factors such as GCF (58).
In summary this paper describes the isolation and characterization of mouse cytochrome oxidase subunit Vb gene with 5'-flanking sequences capable of promoting transcription initiation in NIH 3T3 fibroblast cells. Furthermore, the relatively simple restriction analysis strategy used in this study might be helpful in the identification and isolation of intron containing functional genes for other cytox subunits from higher animals.