Portal de Periódicos da CAPES

Cloning and Expression of Mitochondrial Translational Elongation Factor Ts from Bovine and Human Liver

1995; Elsevier BV; Volume: 270; Issue: 29 Linguagem: Inglês

10.1074/jbc.270.29.17243

1083-351X

Hong Xin, Velinda L. Woriax, William Burkhart, Linda Spremulli,

Viral Infectious Diseases and Gene Expression in Insects

The sequences of the cDNAs for the mitochondrial translational elongation factor Ts (EF-Tsmt) from bovine and human liver have been obtained. The deduced amino acid sequence of bovine liver EF-Tsmt is 338 residues in length and includes a 55-amino acid signal peptide and a mature protein of 283 residues. The sequence of the mature form of bovine EF-Tsmt is 91% identical to that of human EF-Tsmt and 29% identical to Escherichia coli EF-Ts. Southern analysis indicates that there are two genes for EF-Tsmt in bovine liver chromosomal DNA. A 224-base pair intron is located near the 5′-end of at least one of these genes. Northern analysis using a human multiple tissue blot indicates that EF-Tsmt is expressed in all tissues, with the highest levels of expression in skeletal muscle, liver, and kidney. Both the mature and precursor forms of bovine liver EF-Tsmt have been expressed in E. coli as histidine-tagged proteins. The mature form of EF-Tsmt forms a complex with E. coli elongation factor Tu. This complex is active in poly(U)-directed polymerization of phenylalanine. The precursor form is expressed as a 42-kDa protein, which is rapidly degraded in the cell. The sequences of the cDNAs for the mitochondrial translational elongation factor Ts (EF-Tsmt) from bovine and human liver have been obtained. The deduced amino acid sequence of bovine liver EF-Tsmt is 338 residues in length and includes a 55-amino acid signal peptide and a mature protein of 283 residues. The sequence of the mature form of bovine EF-Tsmt is 91% identical to that of human EF-Tsmt and 29% identical to Escherichia coli EF-Ts. Southern analysis indicates that there are two genes for EF-Tsmt in bovine liver chromosomal DNA. A 224-base pair intron is located near the 5′-end of at least one of these genes. Northern analysis using a human multiple tissue blot indicates that EF-Tsmt is expressed in all tissues, with the highest levels of expression in skeletal muscle, liver, and kidney. Both the mature and precursor forms of bovine liver EF-Tsmt have been expressed in E. coli as histidine-tagged proteins. The mature form of EF-Tsmt forms a complex with E. coli elongation factor Tu. This complex is active in poly(U)-directed polymerization of phenylalanine. The precursor form is expressed as a 42-kDa protein, which is rapidly degraded in the cell. In Escherichia coli, elongation factor Tu (EF-Tu)1 1The abbreviations used are: EF-Tuelongation factor TuEF-Tselongation factor TsEF-1elongation factor 1EF-Tumtmitochondrial EF-TuEF-Tsmtmitochondrial EF-Tspre-EF-Tsmtprecursor of bovine liver EF-TsmtPCRpolymerase chain reaction5′-RACE5′-rapid amplification of cDNA endsbpbase pair(s)IPTGisopropyl-1-thio-β-D-galactopyranosidePAGEpolyacrylamide gel electrophoresis. facilitates the binding of aminoacyl-tRNA to the ribosome during the elongation cycle of protein biosynthesis(1Reinbolt J. Metz M.-H. Ehresmann P.R.B. Ehresmann C. Nierhaus K.H. Franceschi F. Subramanian A.R. Erdmann V.A. Wittmann-Liebold B. The Translational Apparatus. Plenum Press, New York1993: 285-294Crossref Google Scholar). Following A-site binding of the correct aminoacyl-tRNA, EF-Tu catalyzes the hydrolysis of GTP, and EF-Tu•GDP is released from the ribosome. Elongation factor Ts (EF-Ts) catalyzes the nucleotide exchange reaction promoting the formation of EF-Tu•GTP from EF-Tu•GDP(2Sprinzl M. Trends Biochem. Sci. 1994; 19: 245-250Abstract Full Text PDF PubMed Scopus (84) Google Scholar). The guanine nucleotide exchange reaction occurs through the formation of an intermediate EF-Tu•Ts complex(3Miller D.L. Weissbach H. Biochem. Biophys. Res. Commun. 1970; 38: 1016-1022Crossref PubMed Scopus (65) Google Scholar). In contrast to E. coli, during the elongation cycle of protein synthesis in Thermus thermophilus, a dimeric form of EF-Ts binds two molecules of EF-Tu, forming an (EF-Tu•EF-Ts)2 structure, which is extremely stable and cannot be dissociated in the absence of protein-denaturing reagents(4Arai K.-I. Arai N. Nakamura S. Oshima T. Kaziro Y. Eur. J. Biochem. 1978; 92: 521-531Crossref PubMed Scopus (53) Google Scholar). In this organism, GDP present in the EF-Tu•GDP complex is thought to exchange directly with GTP present in the (EF-Tu•EF-Ts•GTP)2 dimer. Mammalian mitochondrial EF-Tu and EF-Ts (EF-Tumt and EF-Tsmt) have been purified as a tightly associated complex (EF-Tu•Tsmt) from bovine liver(5Schwartzbach C.J. Spremulli L.L. J. Biol. Chem. 1989; 264: 19125-19131Abstract Full Text PDF PubMed Google Scholar, 6Schwartzbach C.J. Spremulli L.L. J. Biol. Chem. 1991; 266: 16324-16330Abstract Full Text PDF PubMed Google Scholar). The EF-Tu•Tsmt complex is very stable and cannot be dissociated even in the presence of high concentrations of guanine nucleotides. In this respect, the mitochondrial factors differ significantly from the corresponding E. coli factors and show some resemblance to thermophilic EF-Tu and EF-Ts. elongation factor Tu elongation factor Ts elongation factor 1 mitochondrial EF-Tu mitochondrial EF-Ts precursor of bovine liver EF-Tsmt polymerase chain reaction 5′-rapid amplification of cDNA ends base pair(s) isopropyl-1-thio-β-D-galactopyranoside polyacrylamide gel electrophoresis. The stability of the EF-Tu•EF-Ts complex is thought to be determined largely by the nature of the EF-Ts component. For example, EF-Ts from thermophilic bacteria forms strong complexes with E. coli EF-Tu, whereas E. coli EF-Ts produces only weak complexes with thermophilic EF-Tu(7Wittinghofer A. Guariguata R. Leberman R. J. Bacteriol. 1983; 153: 1266-1271Crossref PubMed Google Scholar). In addition, chloroplast EF-Ts from Euglena gracilis forms a tighter complex with E. coli and chloroplast EF-Tu than does the E. coli factor (8Spremulli G.H. Spremulli L.L. Biochem. Biophys. Res. Commun. 1987; 148: 1490-1495Crossref PubMed Scopus (5) Google Scholar). It is not clear what features of EF-Ts modulate the strength of its interaction with EF-Tu. The genes for EF-Ts from several prokaryotes have been cloned and sequenced, as has the gene for chloroplast EF-Ts from the thermophilic red algae Galdieria sulphuraria. In general, the overall sequence for EF-Ts is far less conserved than the sequence for EF-Tu, but there is some conservation located in the NH2-terminal one-third of the protein. In this work, cDNAs encoding EF-Tsmt have been cloned and sequenced from both bovine and human liver. In addition, bovine liver EF-Tsmt has been expressed in E. coli. EF-Tu•Tsmt was purified as described(5Schwartzbach C.J. Spremulli L.L. J. Biol. Chem. 1989; 264: 19125-19131Abstract Full Text PDF PubMed Google Scholar). The EF-Tumt and EF-Tsmt components were separated by reverse-phase high performance liquid chromatography using a Brownlee RP300 column (2.1 × 100 mm) with a linear gradient of acetonitrile, 0.1% trifluoroacetic acid (20-64%) over 60 min. The fraction identified as EF-Tsmt was dried, dissolved in 8 M urea, and incubated at 50°C for 30 min. The solution was diluted to 4 M urea with 0.2 M Tris-HCl, pH 8.5. Sequence-grade endoproteinase Lys-C was added (5 μg), and the sample was incubated at 37°C for 20 h. The resultant peptides were separated on a Brownlee RP300 column (1.0 × 250 mm) with a linear gradient of acetonitrile, 0.1% trifluoroacetic acid (8-64%) over 90 min. The prominent peaks were sequenced on an Applied Biosystems 477A liquid-pulse sequencer connected to an Applied Biosystems 120A phenylthiohydantoin analyzer. Total RNA was extracted from bovine liver by the guanidinium thiocyanate procedure(9Chirgwin J.M. Przybyla A.E. MacDonald R.J. Rutter W.J. Biochemistry. 1979; 18: 5294-5299Crossref PubMed Scopus (16654) Google Scholar). Poly(A)+ RNA was purified by oligo(dT)-cellulose chromatography(10Jacobson A. Methods Enzymol. 1987; 152: 254-261Crossref PubMed Scopus (164) Google Scholar). Single-stranded cDNAs were synthesized by reverse transcription of 5 μg of mRNA using primer 3 or 4 (see Table 1) or primer NS (GGAATTCCCTGCCTGTTTGAGATCCCCGC; the underlined sequence represents an EcoRI adaptor). Bovine liver chromosomal DNA was prepared as described(11Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Smith J.A. Seidman J.G. Struhl K. Current Protocols in Molecular Biology. John Wiley & Sons, Inc., New York1987Google Scholar).TABLE I Open table in a new tab PCR amplification reaction mixtures (100 μl) contained 0.2 mM dNTPs, 2.5 units of Taq DNA polymerase, 50 pmol of the appropriate primers, the buffer system purchased with Taq DNA polymerase (Promega), and either 1 μg of bovine liver chromosomal DNA or an aliquot of the specifically primed cDNA. When the cDNA was used as template, the first five cycles were done at 94°C for 1 min (denaturation), 56°C for 1.5 min (annealing), and 72°C for 2 min (polymerization). For the remaining 35 cycles, annealing was carried out at 61°C. When chromosomal DNA was used as template, primer annealing was done at 50°C during the first five cycles and at 55°C in the 35 remaining cycles. In the last cycle, the reaction time at 72°C was extended to 5 min to allow completion of chains. Nested PCRs involving a second or third round of amplification were carried out as described above, except that they contained 1 μl of the reaction mixture obtained from the previous round of PCR as the template. The reaction mixtures were analyzed on 1.5% agarose or 3% NuSieve GTG-agarose gels. Specific bands were identified by ethidium bromide staining and eluted from the gel using a Geneclean or Mermaid kit (BIO 101, Inc.). Approximately 5 × 106 plaques from two bovine liver libraries (Stratagene) and 2 × 106 plaques from a human liver cDNA library (CLONTECH) were screened by hybridization with a bovine liver EF-Tsmt cDNA probe obtained by PCR amplification and labeled using random priming (12Gottesman M.M. Methods Enzymol. 1987; 151: 363-364Google Scholar). Hybridizations with bovine liver libraries were carried out at 65°C, while hybridization with the human library was done at 55°C. Positive plaques were replated until purified, and the pBluescript SK(-) phagemid clones were excised according to the manufacturer's instructions (Stratagene). Single-stranded cDNA was synthesized as described above using the reverse primer NS. A poly(dA) tail was added to this cDNA by terminal transferase. This cDNA was used as template for 5′-RACE-PCR (13Frohman M.A. Dush M.K. Martin G.R. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8998-9002Crossref PubMed Scopus (4336) Google Scholar). A specific PCR product, ∼260 bp, was obtained, and two BamHI fragments from it were cloned into pTZ18R. EF-Tsmt clones were sequenced by the dideoxynucleotide chain termination method (14Biggin M.D. Gibson T.J. Hong G.F. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 3963-3965Crossref PubMed Scopus (1411) Google Scholar) and subjected to autosequencing in the University of North Carolina DNA Sequencing Facility. All clones were sequenced completely in both directions. Analysis of the sequence was done with Genetics Computer Group sequence analysis programs running on a VAX computer. Single-stranded 5′-RACE products were sequenced after two rounds of nested PCR. The single-stranded DNA was generated in the second round of PCR by using one primer in a 50-fold molar excess over the other primer. Northern analysis of poly(A)+ RNA was performed using 1% agarose gels run in the presence of 1 M glyoxal and 50% dimethyl sulfoxide as described(15Carmichael G.G. McMaster G.K. Methods Enzymol. 1980; 65: 380-391Crossref PubMed Scopus (177) Google Scholar). Bovine chromosomal DNA (30 μg) was digested with an optimal amount of the indicated restriction enzyme. Digests were run on 1.5% agarose gels at 60 V for 19 h. Nucleic acids were transferred to Zeta-Probe blotting membranes as recommended by the manufacturer (Bio-Rad), and blots were probed as indicated. PCR was used to add an NdeI cutting site to the 5′-end and an XhoI cutting site to the 3′-end of the bovine liver cDNA encoding the precursor form of EF-Tsmt and to the portion of the cDNA encoding the mature form of EF-Tsmt. These cDNAs were then cloned into pET24c(+). E. coli BL21(DE3) was used as the host for expression. Purification of the mature and precursor forms under denaturing conditions was performed using nickel-nitrilo-triacetic acid (Ni-NTA) affinity chromatography as described by QIAGEN Inc. For the large-scale purification of the mature and precursor forms under native conditions, expression was induced by exposure of cells (1.0-1.2 A600 units/ml) to 0.025 mM IPTG. Cells were collected after 1 h of induction and lysed by grinding with two times the cell weight of alumina for 20 min on ice. The cell paste was resuspended in 4 volumes of buffer containing 50 mM Tris-HCl, pH 7.6, 60 mM KCl, 7 mM MgCl2, 7 mM β-mercaptoethanol, 0.1 mM phenylmethylsulfonyl fluoride, and 10% glycerol. Alumina was removed by centrifugation at 11,000 × g at 4°C for 15 min. The supernatant was collected and incubated with DNase I (5 μg/ml) for 15 min on ice. The extract was then subjected to centrifugation at 15,000 × g for 30 min. Ni-NTA resin (0.4 ml of a 50% slurry) equilibrated in the same buffer was added to the extract for each 1 liter of original culture. This slurry was shaken at 4°C for 1 h. The Ni-NTA resin was collected by centrifugation at 15,000 × g for 10 min, rinsed with 35 ml of wash buffer (50 mM Tris-HCl, pH 7.6, 1 M NH4Cl, 5 mM β-mercaptoethanol, 10 mM imidazole, and 10% glycerol), poured into a small column, and washed with an additional 60 ml of wash buffer. Protein was eluted from the resin using three aliquots (1 ml each) of elution buffer (50 mM Tris-HCl, pH 7.6, 40 mM KCl, 5 mM β-mercaptoethanol, 0.15 M imidazole, and 10% glycerol). The eluted protein was dialyzed immediately against a 100-fold excess of buffer containing 20 mM Hepes/KOH, pH 7.0, 40 mM KCl, 1 mM MgCl2, 0.1 mM EDTA, and 10% glycerol. The protein concentrations were determined by the Micro-Bradford method (Bio-Rad). The activity of the complex containing E. coli EF-Tu and EF-Tsmt (E. coli EF-Tu•EF-Tsmt) was measured by its ability to catalyze the poly(U)-directed polymerization of phenylalanine on E. coli ribosomes(5Schwartzbach C.J. Spremulli L.L. J. Biol. Chem. 1989; 264: 19125-19131Abstract Full Text PDF PubMed Google Scholar, 6Schwartzbach C.J. Spremulli L.L. J. Biol. Chem. 1991; 266: 16324-16330Abstract Full Text PDF PubMed Google Scholar). One unit is defined as the incorporation of 1 pmol of [14C]Phe into polypeptide at 37°C using a 30-min incubation. Polyclonal antibodies against EF-Tu•EF-Tsmt were produced by Pel-Freez Biologicals (Rogers, AK).2 2V. Woriax and L. Spremulli, manuscript in preparation. Western blotting was done by using the enhanced chemiluminescent detection system of Amersham Corp. EF-Tsmt is the product of a nuclear gene in mammals. To obtain cDNA clones of this factor, it was necessary to obtain partial peptide sequence information. The EF-Tu•Tsmt complex was dissociated, and the two factors were separated by reverse-phase high performance liquid chromatography. EF-Tsmt was subjected to NH2-terminal Edman degradation and to internal peptide sequence analysis following digestion with endoproteinase Lys-C. The sequences of six peptides ranging in size from 9 to 26 residues including the NH2-terminal peptide were obtained (Table 1). Several of the peptides have only a low sequence identity to the sequences for the corresponding prokaryotic factors, and except for the NH2-terminal peptide, their relative positions could not be predicted. Degenerate oligonucleotide primers were designed from these sequences. Forward primers Np-1 and 1 were derived from the NH2-terminal peptide, and their relative positions with respect to each other were known exactly (Fig. 1). Reverse primers 3 and 4 were predicted to be located either in the middle or the COOH-terminal region of EF-Tsmt. The positions of primers 3 and 4 relative to each other could not be predicted. Hence, two groups of nested reverse transcriptase-PCRs were carried out (Fig. 1). Two specific cDNAs were synthesized, one using primer 3 (cDNA3) and the other using primer 4 (cDNA4). In the first round of PCR, the primer used for cDNA synthesis was used in combination with primer Np-1 derived from the NH2-terminal sequence. No specific bands were visible after the first round of PCR in either group. A second round of PCR was performed using the product of cDNA3 and primers Np-1 and 4. No specific bands could be observed after this second round of PCR, suggesting that primer 3 lies to the 5′-side of primer 4. In contrast, a very strong band of ∼650 bp could be observed after the second round of PCR using the product of the first round of PCR derived from cDNA4 with primers 3 and Np-1 (Fig. 1). This observation confirms the idea that primer 3 lies to the 5′-side of primer 4, allowing the nested PCR to succeed. To further confirm the identity of the 650-bp product, this fragment was amplified using primers 3 and primer 1. This reaction gave a product of ∼600 bp, ∼60 bp shorter than the starting DNA. This size difference corresponds to the distance between primers Np-1 and 1. The 650-bp product was cloned into vector pTZ18R and sequenced. The deduced amino acid sequence of this fragment contained five of the peptides obtained by sequence analysis, confirming that this fragment is indeed a partial cDNA coding for bovine liver EF-Tsmt. Bovine liver λZap II and λMAX-1 cDNA libraries were screened for additional portions of the EF-Tsmt cDNA. Three positive plaques were isolated from the λZap II cDNA library, and the pBluescript plasmids carrying the cDNA inserts of interest were excised in vivo. DNA sequence analysis indicated that these clones encompassed the entire 3′-region of the EF-Tsmt cDNA, including the poly(A) tail. Five clones were isolated in a similar manner from the λMAX-1 cDNA library. They contained inserts of ∼135 bp from the middle of the EF-Tsmt coding sequence. None of eight clones isolated contained the 5′-region of the bovine liver EF-Tsmt cDNA, and all of the inserts had the same 5′-end. Subsequent analysis of the sequence of the EF-Tsmt cDNA indicated that there is a G:C-rich region just upstream of this position. Presumably, secondary structure in the mRNA resulted in a strong stop for reverse transcriptase during cDNA synthesis. To obtain the sequences from the 5′-end of the EF-Tsmt cDNA, 5′-RACE-PCR was carried out(13Frohman M.A. Dush M.K. Martin G.R. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8998-9002Crossref PubMed Scopus (4336) Google Scholar). This approach allowed the cloning of an additional 176 bp from the 5′-end of the EF-Tsmt cDNA. Analysis of the cDNA clones described above provided 1297 bp of sequence including the poly(A) tail and encompassing the entire coding region (Fig. 2). The size of the cDNA obtained corresponds well to the size estimated for the mRNA by Northern analysis (data not shown). The long open reading frame codes for the entire EF-Tsmt polypeptide, including a 55-amino acid mitochondrial localization signal and a mature protein of 283 amino acids. The cDNA sequence indicates that bovine liver EF-Tsmt has a very short 5′-untranslated region (18 bp long). Although few eukaryotic cytoplasmic mRNAs have leader regions as short as 18 nucleotides, work by Kozak (16Kozak M. Gene Expr. 1991; 1: 111-115PubMed Google Scholar, 17Kozak M. Gene Expr. 1991; 1: 117-125PubMed Google Scholar, 18Kozak M. J. Cell Biol. 1989; 108: 229-241Crossref PubMed Scopus (2809) Google Scholar) indicates that a leader of this length is generally sufficient to allow initiation. It is possible that all of the clones obtained by 5′-RACE-PCR are shorter than the actual mRNA, reflecting a strong barrier to reverse transcriptase at this position. However, 27 clones obtained by 5′-RACE-PCR all terminated within this region. The initiation codon (designated position +1) is preceded by an A residue at position −3 and followed by a T residue at position +4. Analysis of numerous translational start sites indicates that the consensus sequence has a purine at position −3 and a G residue at position +4(19Kozak M. Cell. 1986; 44: 283-292Abstract Full Text PDF PubMed Scopus (3586) Google Scholar). The 3′-untranslated region is 190 bp in length and contains a polyadenylation signal (AAUAAA) 16 nucleotides before the poly(A) tail. An analysis of the encoded amino acid sequence is provided below. The bovine liver cDNA was used as a probe to screen a human liver λZap II library. Two of the 10 positive clones obtained had inserts of ∼1 kilobases, including a poly(A) tail. These two clones included the entire coding region for the mature form of bovine liver EF-Tsmt and a portion of the mitochondrial import signal (Fig. 2). NH2-terminal sequence analysis indicates that the mature form of bovine liver EF-Tsmt begins with Ser-56 in the long open reading frame (Fig. 2). The mitochondrial import signal for EF-Tsmt is thus 55 residues long. Mitochondrial import sequences are not highly conserved in primary sequence. However, they generally lack acidic residues, are enriched in basic and hydroxylated amino acids, and can form an amphiphilic α-helix or β-sheet. The transit peptide for bovine EF-Tsmt lacks acidic residues, but is not particularly enriched in either basic or hydroxylated residues. It does not appear to be able to form an amphiphilic α-helix or β-sheet. At least two different pathways are believed to be involved in the processing of proteins imported into mitochondria(20Hendrick J.P. Hodges P.E. Rosenberg L.E. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 4056-4060Crossref PubMed Scopus (264) Google Scholar). One pathway uses a single mitochondrial processing peptidase that recognizes Arg at position −2 relative to the processing site(20Hendrick J.P. Hodges P.E. Rosenberg L.E. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 4056-4060Crossref PubMed Scopus (264) Google Scholar). The precursor for bovine liver EF-Tsmt does not appear to fit into this group. A second pathway involves sequential cleavage by two proteases. Proteins processed by this pathway generally have Arg at position −10, a hydrophobic residue at position −8, and Gly, Ser, or Thr at position −5 relative to the cut site. The precursor for bovine liver EF-Tsmt has His at position −10, Phe at position −8, and Gly at position −5 and may possibly be processed by this two-step pathway. The mature form of bovine EF-Tsmt is 283 amino acids in length (Fig. 2) and has a molecular mass of 30,739 Da. The mature form of human liver EF-Tsmt also appears to have 283 residues. The two mammalian EF-Tsmt sequences are 91% identical (Table 2). The sequence of EF-Ts from three prokaryotes (E. coli, Spiroplasma citri, and Spirulina platensis), one chloroplast EF-Ts sequence (G. sulphuraria), and several eukaryotic EF-1βγδ sequences have been reported(21An G. Bendiak D.S. Mamelak L.A. Friesen J.D. Nucleic Acids Res. 1981; 9: 4163-4172Crossref PubMed Scopus (78) Google Scholar, 22Chevalier C. Saillard C. Bove J.M. J. Bacteriol. 1990; 172: 2693-2703Crossref PubMed Google Scholar, 23Sanangelantoni A.M. Calogero R.C. Buttarelli F.R. Gualerzi C.O. Tiboni O. FEMS Microbiol. Lett. 1990; 66: 141-146Crossref Google Scholar, 24Cormier P. Osborne H.B. Morales J. Bassez T. Minella O. Poulhe R. Belle R. Mulner-Lorillon O. Nucleic Acids Res. 1993; 21: 743Crossref PubMed Scopus (23) Google Scholar, 25Morales J. Cormier P. Mulner-Lorillon O. Poulhe R. Belle R. Nucleic Acids Res. 1992; 20: 4091Crossref PubMed Scopus (39) Google Scholar). The β- and δ-subunits of the cytoplasmic factor have both been reported to function as nucleotide exchange factors(24Cormier P. Osborne H.B. Morales J. Bassez T. Minella O. Poulhe R. Belle R. Mulner-Lorillon O. Nucleic Acids Res. 1993; 21: 743Crossref PubMed Scopus (23) Google Scholar, 25Morales J. Cormier P. Mulner-Lorillon O. Poulhe R. Belle R. Nucleic Acids Res. 1992; 20: 4091Crossref PubMed Scopus (39) Google Scholar). The bacterial factors are ∼250-300 residues in length, similar to the size of the mammalian mitochondrial factors. The one known gene for chloroplast EF-Ts encodes a protein of 199 residues, although it is unclear whether this gene actually encodes a functional product(23Sanangelantoni A.M. Calogero R.C. Buttarelli F.R. Gualerzi C.O. Tiboni O. FEMS Microbiol. Lett. 1990; 66: 141-146Crossref Google Scholar). The only chloroplast EF-Ts studied at the protein level to date is from E. gracilis(26Fox L. Erion J. Tarnowski J. Spremulli L. Brot N. Weissbach H. J. Biol. Chem. 1980; 255: 6018-6019Abstract Full Text PDF PubMed Google Scholar) and appears to be a monomer of 62 kDa, considerably larger that any of the corresponding factors known. The β- and δ-subunits of the cytoplasmic factor EF-1 are 227 and 265 residues in length, respectively(24Cormier P. Osborne H.B. Morales J. Bassez T. Minella O. Poulhe R. Belle R. Mulner-Lorillon O. Nucleic Acids Res. 1993; 21: 743Crossref PubMed Scopus (23) Google Scholar, 25Morales J. Cormier P. Mulner-Lorillon O. Poulhe R. Belle R. Nucleic Acids Res. 1992; 20: 4091Crossref PubMed Scopus (39) Google Scholar). A comparison of the sequences of these nucleotide exchange factors (Table 2) indicates that mammalian EF-Tsmt is 27-35% identical to the corresponding prokaryotic and chloroplast factors, but <21% identical to either the β- or δ-subunit of EF-1.TABLE II Open table in a new tab The alignment of the sequences of EF-Ts from prokaryotes, chloroplasts, and mammalian mitochondria (Fig. 3) indicates that conserved regions are clustered in the NH2-terminal one-third of the protein. The complete conservation of 22 residues in these factors is observed, with 20 of these amino acids being located within the first 90 residues. The longest stretch of completely conserved residues is 5 amino acids long. There is also a conserved stretch of 10 residues present if conservative substitutions are taken into account. Little information is available on the regions of EF-Ts that are important for the nucleotide exchange activity of this factor. The sequence alignment presented in Fig. 3 and the observation that EF-Tsmt will promote GDP exchange with E. coli EF-Tu suggest that the NH2-terminal one-third of the protein may play a particularly important role in this process. For determination of the number of copies of the gene encoding bovine liver EF-Tsmt, a Southern blot of total DNA digested with EcoRI, BamHI, HindIII, and BglII, respectively, was probed using nucleotides 177-1182 as a probe (Fig. 4). This region was selected as a probe because PCR analysis of genomic DNA and the cDNA indicated that no introns were present in this region of the gene (data not shown). EcoRI cuts the probe once, while the other three enzymes do not have a cutting site in the probe. As indicated in Fig. 4, EcoRI digestion resulted in the appearance of four bands, while BamHI, HindIII, and BglII digestion gave two bands on the Southern blot. These results suggest that two genes encoding EF-Tsmt exist in bovine chromosomal DNA. PCR analysis failed to amplify genomic sequences corresponding to the cDNA, suggesting that neither of the genes detected is a pseudogene (data not shown). It has recently been observed that there are also two copies of the EF-Tumt gene in bovine chromosomal DNA.3 3V. Woriax, W. Burkhart, and L. Spremulli, manuscript in preparation. An examination of the EF-Tsmt gene for the presence of intervening sequences was carried out by PCR amplification of chromosomal DNA using different combinations of oligonucleotide primers designed from the cDNA sequence. This analysis indicated the presence of an intron near the 5′-end of the bovine EF-Tsmt genes. The PCR analysis used here suggests that both genes contain a similarly sized intron. However, this procedure can only detect relatively small introns, and this interpretation must be viewed with caution. This intron (224 bp) was cloned and sequenced (Fig. 2B). There are no reports on the expression of EF-Tsmt in any mammalian species to date. To investigate the relative amounts of the EF-Tsmt transcript in different human tissues, a multiple tissue Northern blot was analyzed using the full-length cDNA of EF-Tsmt as a probe. As shown in Fig. 5, EF-Tsmt transcripts of ∼1200 bases could be observed in all the human tissues analyzed. Human skeletal muscle had the most abundant level of transcripts among the tissues tested, followed by liver, kidney, and heart. Placenta, brain, pancreas, and lung had substantially lower levels of the mRNA for EF-Tsmt. Higher levels of expression were clearly observed in specialized tissues known to have high demands for energy production. Skeletal muscle contains not only the normal 1200-base mRNA, but also a larger transcript. Lower amounts of this other transcript are observed in several other tissues. This transcript may possibly arise from use of an alternative polyadenylation site (27Wahle E. Keller W. Annu. Rev. Biochem. 1992; 61: 419-440Crossref PubMed Google Scholar) or from transcription of the two separate genes or may represent a cross-hybridizing mRNA. Both the precursor of bovine liver EF-Tsmt (pre-EF-Tsmt) and the mature form of this factor (amino acids 56-338) were cloned into a pET expression vector. Both constructs carry a 6-residue histidine tag at the COOH terminus separated from the normal COOH terminus by a Leu-Glu linker. As indicated in Fig. 6(lane1), little, if any, material corresponding to EF-Tsmt was detectable in extracts of uninduced cells. However, following IPTG induction, a major band with a molecular mass of 34 kDa was observed (Fig. 6, lane2). Western analysis of samples before and after induction (lanes6 and 7) indicated that the new band reacted strongly with antibodies raised against bovine EF-Tu•Tsmt. This observation demonstrates that the cells are indeed expressing the bovine mitochondrial factor. The expressed EF-Tsmt migrated on SDS-PAGE to a position ∼2 kDa larger than that observed with EF-Tu•Tsmt prepared from mitochondria (Fig. 6, lanes2 and 3). This observation is not surprising since there are 9 extra amino acid residues including the His tag in the expressed form of the factor. An examination of the effects of induction of EF-Tsmt on the growth of the host E. coli cells indicated that the cells stopped growing within 1 h after induction, suggesting that the expression of the bovine mitochondrial factor is toxic to the cells. Maximal levels of induction were observed between 1 and 2 h following addition of IPTG, and concentrations of IPTG as low as 25 μM gave maximal levels of expression. The His-tagged form of pre-EF-Tsmt was expressed in E. coli as a 42-kDa protein. Pre-EF-Tsmt could be observed primarily in extracts made under denaturing conditions and when induction was carried out for only ∼30 min. Longer periods of induction resulted in the degradation of pre-EF-Tsmt (data not shown). When the mature form of EF-Tsmt was purified by Ni-NTA chromatography under nondenaturing conditions and then analyzed by SDS-PAGE (Fig. 6, lane4), a band with a molecular mass of 44 kDa copurified with EF-Tsmt. This band appears to have a molecular mass identical to that of E. coli EF-Tu (lane5), indicating that the mitochondrial factor will form a complex with bacterial EF-Tu. This observation is not surprising since EF-Tsmt can promote guanine nucleotide exchange with prokaryotic EF-Tu(5Schwartzbach C.J. Spremulli L.L. J. Biol. Chem. 1989; 264: 19125-19131Abstract Full Text PDF PubMed Google Scholar). The ability of the E. coli EF-Tu•EF-Tsmt complex to function in the poly(U)-directed polymerization of phenylalanine was examined. As shown in Fig. 7, this complex was active in the in vitro assay. The specific activity obtained for the heterologous complex (140,000 units/mg) is ∼30% of the activity that is obtained with the homologous EF-Tu•Tsmt complex from bovine liver. The E. coli EF-Tu•EF-Tsmt complex functions catalytically in the polymerization assay, with ∼10 rounds of phenylalanine incorporated into peptide for each complex present. A more detailed analysis of the properties of this heterologous complex is now underway.