Isolation of a Chinese Hamster Ovary (CHO) cDNA Encoding Phosphatidylglycerophosphate (PGP) Synthase, Expression of Which Corrects the Mitochondrial Abnormalities of a PGP Synthase-defective Mutant of CHO-K1 Cells
1999; Elsevier BV; Volume: 274; Issue: 3 Linguagem: Inglês
10.1074/jbc.274.3.1828
ISSN1083-351X
AutoresKiyoshi Kawasaki, Osamu Kuge, Shao-Chun Chang, Philip Heacock, Minseok Rho, Kenji Suzuki, Masahiro Nishijima, William Dowhan,
Tópico(s)Protein Kinase Regulation and GTPase Signaling
ResumoPhosphatidylglycerophosphate (PGP) synthase catalyzes the first step in the cardiolipin (CL) branch of phospholipid biosynthesis in mammalian cells. In this study, we isolated a Chinese hamster ovary (CHO) cDNA encoding a putative protein similar in sequence to the yeast PGS1 gene product, PGP synthase. The gene for the isolated CHO cDNA was namedPGS1. Expression of the CHO PGS1 cDNA in CHO-K1 cells and production of a recombinant CHO PGS1protein with a N-terminal extension in Escherichia coliresulted in 15-fold and 90-fold increases of PGP synthase specific activity, respectively, establishing that CHO PGS1 encodes PGP synthase. A PGP synthase-defective CHO mutant, PGS-S, isolated previously (Ohtsuka, T., Nishijima, M., and Akamatsu, Y. (1993)J. Biol. Chem. 268, 22908–22913) exhibits striking reductions in biosynthetic rate and cellular content of phosphatidylglycerol (PG) and CL and shows mitochondrial morphological and functional abnormalities. The CHO PGS-S mutant transfected with the CHO PGS1 cDNA exhibited 620-fold and 7-fold higher PGP synthase activity than mutant PGS-S and wild type CHO-K1 cells, respectively, and had a normal cellular content and rate of biosynthesis of PG and CL. In contrast to mutant PGS-S, the transfectant had morphologically normal mitochondria. When the transfectant and mutant PGS-S cells were cultivated in a glucose-depleted medium, in which cellular energy production mainly depends on mitochondrial function, the transformant but not mutant PGS-S was capable of growth. These results demonstrated that the morphological and functional defects displayed by the PGS-S mutant are due directly to the reduced ability to make normal levels of PG and/or CL. Phosphatidylglycerophosphate (PGP) synthase catalyzes the first step in the cardiolipin (CL) branch of phospholipid biosynthesis in mammalian cells. In this study, we isolated a Chinese hamster ovary (CHO) cDNA encoding a putative protein similar in sequence to the yeast PGS1 gene product, PGP synthase. The gene for the isolated CHO cDNA was namedPGS1. Expression of the CHO PGS1 cDNA in CHO-K1 cells and production of a recombinant CHO PGS1protein with a N-terminal extension in Escherichia coliresulted in 15-fold and 90-fold increases of PGP synthase specific activity, respectively, establishing that CHO PGS1 encodes PGP synthase. A PGP synthase-defective CHO mutant, PGS-S, isolated previously (Ohtsuka, T., Nishijima, M., and Akamatsu, Y. (1993)J. Biol. Chem. 268, 22908–22913) exhibits striking reductions in biosynthetic rate and cellular content of phosphatidylglycerol (PG) and CL and shows mitochondrial morphological and functional abnormalities. The CHO PGS-S mutant transfected with the CHO PGS1 cDNA exhibited 620-fold and 7-fold higher PGP synthase activity than mutant PGS-S and wild type CHO-K1 cells, respectively, and had a normal cellular content and rate of biosynthesis of PG and CL. In contrast to mutant PGS-S, the transfectant had morphologically normal mitochondria. When the transfectant and mutant PGS-S cells were cultivated in a glucose-depleted medium, in which cellular energy production mainly depends on mitochondrial function, the transformant but not mutant PGS-S was capable of growth. These results demonstrated that the morphological and functional defects displayed by the PGS-S mutant are due directly to the reduced ability to make normal levels of PG and/or CL. In animal tissue, the anionic phospholipids phosphatidylglycerol (PG) 1The abbreviations used are: PG, phosphatidylglycerol; CL, cardiolipin; PGP, phosphatidylglycerophosphate; CHO, Chinese hamster ovary; EST, expressed sequence tag; kb, kilobase pairs(s); ORF, open reading frame; cPGS1, CHO PGS1 gene product; yPGS1, yeast PGS1gene product. and cardiolipin (CL) are thought to be necessary for many cellular functions. PG represents approximately 1% of total lipid phosphorous in mammalian tissues except for lung and is found in many intracellular locations, such as mitochondrial, nuclear, and microsomal membranes. In lung, PG represents approximately 5% of total phospholipid; it is found there predominantly in lamella body membranes (reviewed in Ref. 1Hostetler K.Y. Hawthorne J.N. Ansell G.B. Phospholipids. Elsevier/North Holland Biomedical Press, Amsterdam1982: 215-261Google Scholar) and is also one of the main components of lung surfactant (1Hostetler K.Y. Hawthorne J.N. Ansell G.B. Phospholipids. Elsevier/North Holland Biomedical Press, Amsterdam1982: 215-261Google Scholar, 2Hallman M. Gluc L. Biochim. Biophys. Acta. 1975; 409: 172-191Crossref PubMed Scopus (92) Google Scholar). Recent biochemical analysis suggests that PG is a potential activator of the protein kinase C family, including protein kinase C-θ (3Pietromonaco S.F. Simons P.C. Altman A. Elias L. J. Biol. Chem. 1998; 273: 7594-7603Abstract Full Text Full Text PDF PubMed Scopus (179) Google Scholar) and nuclear protein kinase C-βII (4Murray N.R. Fields A.P. J. Biol. Chem. 1998; 273: 11514-11520Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar). CL represents from 0.2 to 15% of total lipid phosphorous in various animal tissues and is located primarily in the inner mitochondrial membrane (1Hostetler K.Y. Hawthorne J.N. Ansell G.B. Phospholipids. Elsevier/North Holland Biomedical Press, Amsterdam1982: 215-261Google Scholar). Biochemical analysis suggests that CL is required for many enzymatic activities, such as cytochrome c oxidase (5Awasthi Y.C. Chuang T.F. Keenan T.W. Crane F.L. Biochim. Biophys. Acta. 1971; 226: 42-52Crossref PubMed Scopus (153) Google Scholar) and carnitine acylcarnitine translocase (6Noel H. Pande S. Eur. J. Biochem. 1986; 155: 99-102Crossref PubMed Scopus (86) Google Scholar), and is involved in cellular functions, such as mitochondrial protein import (7Rietveld A. Sijens P. Verkleij A.J. de Kruijff B. EMBO J. 1983; 2: 907-913Crossref PubMed Google Scholar, 8Ou W.J. Ito A. Umeda M. Inoue K. Omura T. J. Biochem. 1988; 103: 589-595Crossref PubMed Scopus (50) Google Scholar, 9Schleyer M. Neupert W. Cell. 1985; 43: 339-350Abstract Full Text PDF PubMed Scopus (268) Google Scholar, 10Ardail D. Privat J.P. Egret-Charlier M. Levrat C. Lerme F. Louisot P. J. Biol. Chem. 1990; 265: 18797-18802Abstract Full Text PDF PubMed Google Scholar) and binding of matrix Ca2+ (11Krebs J.J.R. Hauser H. Carafoli E. J. Biol. Chem. 1979; 254: 5308-5316Abstract Full Text PDF PubMed Google Scholar). In yeast cells, most of the requirement for CL for mitochondrial functions appears to be substituted by an increase in PG content (12Chang S.-C. Heacock P.N. Mileykovskaya E. Voelker D.R. Dowhan W. J. Biol. Chem. 1998; 273: 14933-14941Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar). Disruption of the CRD1 gene, which encodes CL synthase, resulted in no detectable CL and 5-fold elevation of PG content in yeast cells. The yeast crd1-null strain can grow on both fermentable and nonfermentable carbon sources (12Chang S.-C. Heacock P.N. Mileykovskaya E. Voelker D.R. Dowhan W. J. Biol. Chem. 1998; 273: 14933-14941Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar, 13Tuller G. Hrastnik C. Achleitner G. Schiefthaler U. Klein F. Daum G. FEBS Lett. 1998; 421: 15-18Crossref PubMed Scopus (123) Google Scholar, 14Jiang F. Rizavi H.S. Greenberg M.L. Mol. Microbiol. 1997; 26: 481-491Crossref PubMed Scopus (154) Google Scholar), although with reduced efficiency on the latter. On the other hand, disruption of the yeast PGS1 gene (also known as PEL1), which encodes phosphatidylglycerophosphate (PGP) synthase resulted in no detectable CL or PG (15Chang S.-C. Heacock P.N. Clancey C.J. Dowhan W. J. Biol. Chem. 1998; 273: 9829-9836Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar) and the inability to grow on nonfermentable carbon sources (15Chang S.-C. Heacock P.N. Clancey C.J. Dowhan W. J. Biol. Chem. 1998; 273: 9829-9836Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar, 16Janitor M. Subik J. Curr. Genet. 1993; 24: 307-312Crossref PubMed Scopus (48) Google Scholar, 17Janitor M. Obernauerova M. Kohlwein S.D. Subic J. FEMS Microbiol. Lett. 1996; 140: 43-47PubMed Google Scholar). Early enzymological studies revealed the biosynthetic pathway for PG and CL in animal cells as shown in Fig.1. PGP synthase catalyzes the committed step in CL biosynthesis with the displacement of the CMP moiety of CDP-diacylglycerol by sn-glycerol 3-phosphate to produce PGP (18Kiyasu J.Y. Pieringer R.A. Paulus H. Kennedy E.P. J. Biol. Chem. 1963; 238: 2293-2298Abstract Full Text PDF PubMed Google Scholar). PGP is rapidly dephosphorlylated to generate PG that is utilized as a substrate along with CDP-diacylglycerol for CL synthesis. PGP synthase activity is predominantly localized in mitochondria, but measurable activity is also detected in endoplasmic reticulum (19Jelsema C.L. Morré D.J. J. Biol. Chem. 1978; 253: 7960-7971Abstract Full Text PDF PubMed Google Scholar). Solubilization and partial purification of PGP synthase from pig liver mitochondria was reported (20Nicholson D.W. McMurry W.C. Biochim. Biophys. Acta. 1986; 856: 515-525Crossref PubMed Scopus (7) Google Scholar, 21McMurry W.C. Jarvis E.C. Can. J. Biochem. 1978; 56: 414-419Crossref PubMed Scopus (21) Google Scholar), but neither complete purification nor isolation of cDNA for PGP synthase from somatic cells has been reported. Chinese hamster ovary (CHO) cell mutants are a good model for studying mammalian phospholipid biosynthesis and functions (22Nishijima M. Kuge O. Hanada K. Trends Cell Biol. 1997; 7: 324-329Abstract Full Text PDF PubMed Scopus (13) Google Scholar). Previously, we isolated a temperature-sensitive CHO mutant (PGS-S) that is defective in PGP synthase activity. The PGP synthase activity in the mutant was 1% of that in wild type CHO-K1 cells, and biosynthetic rate and cellular content of PG and CL were also markedly reduced in the mutant (23Ohtsuka T. Nishijima M. Akamatsu Y. J. Biol. Chem. 1993; 268: 22908-22913Abstract Full Text PDF PubMed Google Scholar). Moreover, this mutant displays mitochondrial morphological and functional abnormality (24Ohtsuka T. Nishijima M. Suzuki K. Akamatsu Y. J. Biol. Chem. 1993; 268: 22914-22919Abstract Full Text PDF PubMed Google Scholar). These previous results implied that PG and/or CL is essential for mitochondrial morphology and function in CHO-K1 cells. In this study, we report the first isolation of a somatic cell cDNA from CHO cells that encodes a PGP synthase (cPGS1). Using this cDNA, we also demonstrate that the mitochondrial morphological defects and inability to carry out oxidative phosphorylation to support growth on glucose-depleted medium displayed by mutant PGS-S are directly due to the reduced ability to make normal levels of the anionic phospholipids PG and/or CL. All chemicals were reagent grade or better. Restriction endonucleases and DNA modifying enzymes were from Promega Corp. and New England Biolabs. Thin layer chromatography Silica Gel 60 plates were from EM Science. Oligonucleotides were prepared commercially by Genosis Biotechnologies, Inc. QIAEXTM gel extraction kit was from Qiagen.sn-[U-14C]Glycerol 3-phosphate was from ICN Radiochemicals. [32P]Orthophosphate and [α-32P]dCTP were from Amersham Pharmacia Biotech. CDP-diacylglycerol was from Serdary Research Laboratory.sn-Glycerol 3-phosphate and trypsin were from Sigma. Ham's F-12 medium, newborn calf serum, Geneticin (G418), penicillin G, and streptomycin sulfate were from Life Technologies, Inc. All media for bacterial growth were obtained from BIO 101 or Difco. Oligonucleotides corresponding to parts of human expressed sequence tags (ESTs) (The Institute for Genome Research, gene identification numbers THC122139 and T12593) were used to amplify a CHO PGS1 cDNA fragment from a CHO cDNA library (25Kuge O. Saito K. Nishijima M. J. Biol. Chem. 1997; 272: 19133-19139Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar) by means of a two-stage polymerase chain reaction. The primers used for the first round of amplification were 5′-ATGGCATCCCTTTACCTGGG-3′ (sense) and 5′-ACGTAGCGGATCTGTCGGTT-3′ (antisense). Primers, 5′-TCTGAATTCGAGCAGGAACTGGTGGATTG-3′ (sense) and 5′-TCTCTCGAGATCTGTCGGTTGGTGAAGTA-3′ (antisense) containing anEcoRI site and XhoI site, respectively, were used for the second round of amplification. The amplification reactions were performed for 40 cycles of denaturation at 94 °C for 1 min, annealing at 50 °C for 2 min, and elongation at 72 °C for 2 min, with Taq polymerase (Perkin-Elmer) according to the manufacturer's instruction. For the second round of amplification, the first round reaction mixture was diluted 20-fold and used as template. The 0.35-kb product of the second round reaction was subcloned into theEcoRI/XhoI sites of pBluescriptII SK+(Stratagene) and sequenced. Based on the sequence of the cloned 0.35-kb DNA, an oligonucleotide (5′-CTCTGGAGAAGTCACTACAGTCG-3′ (sense)) was synthesized, biotinylated, and used for the enrichment of corresponding cDNA clones from a CHO cDNA library (25Kuge O. Saito K. Nishijima M. J. Biol. Chem. 1997; 272: 19133-19139Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar) by employing the GeneTrapperTM cDNA positive selection system (Life Technologies, Inc.) according to the manufacturer's instructions. The single stranded cDNA clones captured with the biotinylated oligonucleotide were electroporated into ElectroMAX DH10B (Life Technologies, Inc.), and ampicillin-resistant transformants were selected. Colonies of the resultant transformant were subjected to a polymerase chain reaction screening for colonies containing CHOPGS1 cDNA using the same unbiotinylated sense oligonucleotide and 5′-AGTCACTCAGGTTTGCACC-3′ (antisense, which corresponds to the sequence of human EST cDNA, The Institute for Genome Research number THC122139) as polymerase chain reaction primers. Methods for plasmid preparation, restriction enzyme digestion, DNA ligation, andEscherichia coli transformation were preformed as described previously (26Sambrook J. Fritsh E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). Plasmid DNA to be sequenced and to be transfected into CHO cells was prepared with Wizard DNA purification kits (Promega Corp.). DNA sequencing reactions were performed by the TaqDye-deoxy Terminator (Applied Biosystems) method with walking primers, and the reaction products were run on an Applied Biosystems sequencer. Both strands of two CHO PGS1 cDNA clones carrying a putative full-length open reading frame (ORF) were determined. Strain CHO-K1 was obtained from the American Type Culture Collection. Mutant PGS-P (parent strain CHO-K1) and mutant PGS-S (parent strain PGS-P) were previously described (23Ohtsuka T. Nishijima M. Akamatsu Y. J. Biol. Chem. 1993; 268: 22908-22913Abstract Full Text PDF PubMed Google Scholar). Unless otherwise indicated all strains were grown in Ham's F-12 medium (27Ham R.G. Prod. Natl. Acad. Sci. U. S. A. 1965; 53: 288-295Crossref PubMed Scopus (645) Google Scholar) supplemented with 10% (v/v) newborn calf serum, penicillin G (100 units/ml), and streptomycin sulfate (100 μg/ml), under 5% CO2 atmosphere at 100% humidity at either 40, 37, or 33 °C. Ham's F-12 medium containing galactose (1.8 mg/ml) instead of glucose as the carbon source was prepared with reagent grade chemicals and supplemented with 10% (v/v) dialyzed newborn calf serum, penicillin G (100 units/ml), and streptomycin sulfate (100 μg/ml). A plasmid (pSPORT1-cPGS1), carrying a putative CHOPGS1 cDNA (see Fig. 2), was cleaved at theSalI and NotI sites flanking the cDNA insert. The resulting fragment was inserted into the same restriction enzyme sites of mammalian expression vector, pSV-SPORT1 (Life Technologies, Inc.), under control of the SV40 promoter. The resulting construct, pSV-cPGS1, and pSV-SPORT1 were introduced into CHO-K1 cells using LipofectAMINE reagent (Life Technologies, Inc.) according to the manufacturer's instructions. A plasmid (pSPORT-cPGS1′) potentially expressing a chimeric derivative of the CHO PGS1 cDNA ORF underlacOP control was introduced into E. coli strain XL1-Blue. The cDNA insert in pSPORT-PGS1′ is identical to that shown in Fig. 2 except for an additional GG at the 5′-end. Because the cDNA library used to make this plasmid employed a SalI linker (5′-TCGACCCACGCGTCCG-3′) at the 5′-end, this plasmid should express a chimeric protein with following sequence fused to the N-terminal MET of the putative ORF encoded by the PGS1 cDNA: MTMITPS-SNTTHYRESWYACRYRSGIPGS-THAS-GRVS (the hyphens divide the sequence derived from β-galactosidase, the vector, the linker, and the region 5′ of the ORF, respectively). E. coli strain XL1-Blue transformed with pSPORT-cPGS1′ was cultivated in LB-medium containing 100 μg/ml of ampicillin at 30 °C. Protein production was induced by the addition of isopropyl β-d-thiogalactoside to the medium at a final concentration of 1 mm. After cultivation for 4 h in the presence of isopropyl β-d-thiogalactoside, cells were collected, suspended in a buffer (50 mm Tris-HCl (pH 7.8), 1 mm EDTA, 100 mm NaCl, and 0.1 mmphenylmethylsulfonyl fluoride), and sonically disrupted for enzyme assay. A mammalian expression vector (pSV-OKneo) containing the G418 resistant determinant (28Hanada K. Hara T. Nishijima J. Kuge O. Dickson R.C. Nagiec M.M. J. Biol. Chem. 1997; 272: 32108-32114Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar) was cleaved at the SalI and NotI sites and ligated with the CHO PGS1 cDNA generated by the digestion of pSPORT1-cPGS1 with same restriction enzymes. The resulting construct (pSVneo-cPGS1) was introduced into mutant PGS-S cells (23Ohtsuka T. Nishijima M. Akamatsu Y. J. Biol. Chem. 1993; 268: 22908-22913Abstract Full Text PDF PubMed Google Scholar) using LipofectAMINE reagent, and then the G418-resistant transformants were selected in medium containing 400 μg/ml of G418 at 37 °C. Stable transformant colonies were isolated with cloning cups, and finally purified by limited dilution to yield only one colony per well. PGP synthase activity was measured as described previously (23Ohtsuka T. Nishijima M. Akamatsu Y. J. Biol. Chem. 1993; 268: 22908-22913Abstract Full Text PDF PubMed Google Scholar) with some modification. The assay was performed at 37 °C for 30 min in the presence of 50 mmTris-HCl (pH 7.4), 0.25 mm CDP-diacylglycerol, 0.1 mm sn-[U-14C]glycerol 3-phosphate (20 mCi/mmol), 0.25% Triton X-100, and 0.4 mg/ml of CHO cell extract protein in a total volume of 100 μl or 0.5 mg/ml of E. coli cell extract protein in a total volume of 200 μl. After incubation, the lipids were extracted by the sequential addition of 600 μl of chloroform/methanol (1:2, v/v), 200 μl of chloroform, and 200 μl of phosphate-buffered saline. The lipid-containing chloroform phase was washed twice with 400 μl of chloroform/methanol/0.1m KCl (3:47:48, v/v), and the radioactivity in the chloroform phase was quantified as described previously (15Chang S.-C. Heacock P.N. Clancey C.J. Dowhan W. J. Biol. Chem. 1998; 273: 9829-9836Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar). Lipid extraction from 32Pi-labeled cells was performed according to Bligh and Dyer (29Bligh E.G. Dyer W.J. Can. J. Biochem. Physiol. 1959; 37: 911-917Crossref PubMed Scopus (42878) Google Scholar). Analysis of lipid composition and separation of lipids were performed by two-dimensional thin layer chromatography (30Nishijima M. Kuge O. Akamatsu Y. J. Biol. Chem. 1986; 261: 5784-5789Abstract Full Text PDF PubMed Google Scholar). Alternatively, one-dimensional thin layer chromatography was performed for the separation of PG or CL from other lipids. For the separation of PG, chloroform/methanol/acetic acid (65:25:10, v/v) was used as solvent. For the separation of CL, chloroform/methanol/water/ammonium hydroxide (130:75:6:2, v/v) was used with boric acid-impregnated Silica gel 60 thin layer plates (31Fine B.J. Sprecher H. J. Lipid Res. 1982; 23: 660-663Abstract Full Text PDF PubMed Google Scholar). Protein concentrations were determined by BCA protein assay reagent (Pierce) using bovine serum albumin as a standard. mRNA was isolated using the FastTrackTM 2.0 Kit (Invitrogen) according to the manufacturer's instructions from CHO-K1 and mutant PGS-S cells grown at 40 °C for 4 days. mRNA (4 μg) was separated on 1% agarose gels containing formaldehyde and transferred to a positively charged HybondTM N+ Nylon membrane (Amersham Pharmacia Biotech) by capillary action using 20× SSC (3 m NaCl, 0.3m sodium citrate, pH 7.0) at room temperature. RNA was cross-linked to the membrane by baking at 80 °C for 2 h. TheNotI/SalI CHO PGS1 cDNA fragment from pSV-cPGS1 and a 2.0-kb fragment of human β-actin cDNA were32P-labeled with the Rediprime DNA labeling system (Amersham Pharmacia Biotech). Hybridization reaction was performed in Rapid-Hyb buffer (Amersham Pharmacia Biotech). A mouse Multiple Tissue Northern (MTNTM) Blot (CLONTECH) was hybridized with 32P-labeled CHO PGS1 cDNA and human β-actin cDNA according to the manufacturer's instructions. The results were visualized by exposure to x-ray film, and quantitative analysis was preformed with a Packard Instant Imager (Packard Instruments Inc.). In order to identify a mammalian cDNA clone encoding PGP synthase activity, we searched within the The Institute for Genome Research human gene index data base (http://www.tigr.org/) (32Adams M.D. Kerlavage A.R. Fleischmann R.D. Fuldner R.A. Bult C.J. Lee N.H. Kirkness E.F. Weinstock K.G. Gocayne J.D. White O. et al.Nature. 1995; 377 (suppl.): 3-174Google Scholar) for human ESTs that encode a similar protein to the yeast PGS1gene product (yPGS1) (15Chang S.-C. Heacock P.N. Clancey C.J. Dowhan W. J. Biol. Chem. 1998; 273: 9829-9836Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar). Two human ESTs (The Institute for Genome Research gene identification number, THC122139 and T12593) were found to encode peptides that exhibited 41.6% identity with yPGS1 in a stretch of 178 amino acids and 43.4% identity in a stretch of 79 amino acids, respectively. A cDNA fragment that has similarity to yPGS1 in sequence was amplified from a CHO cDNA library using polymerase chain reaction with primers corresponding to the sequence of human ESTs. In order to isolate a full-length cDNA clone, an oligonucleotide derived from the sequence of the amplified CHO cDNA fragment was used for screening against a CHO cDNA library as described under "Experimental Procedures." Eight positive plasmid clones were identified that have cDNA inserts of approximately 2.2 kb. Two of the clones, named pSPORT1-cPGS1 and pSPORT1-cPGS1′, were used for further analysis, and the gene corresponding to the isolated CHO cDNA was named PGS1. The CHO PGS1cDNA contained a large ORF encoding a protein of 553 amino acid residues with a calculated molecular mass of 62,329 Da (Fig. 2). Fifteen bases upstream of the poly(A) tail, there is a sequence (TATAAA) similar to the consensus poly(A) attachment signal (AATAAA) that is also functional as a poly(A) attachment signal (33Simonsen C.C. Levinson A.D. Mol. Cell. Biol. 1983; 3: 2250-2258Crossref PubMed Scopus (96) Google Scholar). Because the size of PGS1 mRNA was estimated at about 2.4 kb by Northern blot analysis (see below) and poly(A) length is usually from 150 to 200 bases (34Darnell J.E. Sci. Am. 1983; 249: 72-82Crossref Scopus (21) Google Scholar), the 2.2-kb CHO PGS1 cDNA lacking the poly(A) tail appears to be the complete cDNA. The amino acid sequence of cPGS1 deduced from the cDNA sequence exhibited 30% amino acid sequence identity with yPGS1, and the identical amino acids were scattered throughout the sequence (Fig.3 A). In Fig. 3 A,the underlined sequence indicates a region of high homology (79%) between cPGS1 and yPGS1. Within this region is the HKD motif (HXKXXXXDXXXXXXG). This motif (Fig.3 B) is found in several hydrolases and phosphotransferases (35Koonin E.V. Trends Biochem. Sci. 1996; 21: 242-243Abstract Full Text PDF PubMed Scopus (140) Google Scholar) involved in phospholipid metabolism such as E. coliphosphatidylserine synthase, bacterial CL synthases, eukaryotic phospholipase D, and yPGS1. The presence of the HKD motif is consistent with the CHO PGS1 gene product encoding a PGP synthase (see below). To examine whether the CHO PGS1 gene encodes PGP synthase activity, the CHO PGS1 cDNA was cloned into mammalian expression vector, pSV-SPORT1, and the resultant plasmid, designated as pSV-cPGS1, and pSV-SPORT1 were introduced into CHO-K1 cells. The cell extract derived from the transient transfectant with pSV-cPGS1 exhibited a 15-fold higher specific activity (3550 pmol/min/mg of protein) of PGP synthase than that from CHO-K1 cells transfected with pSV-SPORT1 (230 pmol/min/mg of protein); the average of duplicate assays varied by less than 10%. This result suggested that the CHOPGS1 gene encodes PGP synthase activity. To obtain further evidence that CHO PGS1 encodes PGP synthase, E. coli strain XL1-Blue was transformed with pSPORT1-cPGS1′. The resultant transformant was cultured, and production of recombinant cPGS1 was induced by the addition of isopropyl β-d-thiogalactoside. PGP synthase activity in cell extracts in which recombinant protein production was induced (781 pmol/min/mg of protein) was 90-fold higher than that of cell extracts in which recombinant protein production was not induced (8.68 pmol/min/mg of protein); the average of duplicate assays varied by less than 10%. This result provides definitive evidence that CHOPGS1 encodes cPGS1. PGS-S is a temperature-sensitive CHO mutant strain in which PGP synthase activity is 1% of that in wild type CHO-K1 at the restrictive temperature of 40 °C (Ref. 23Ohtsuka T. Nishijima M. Akamatsu Y. J. Biol. Chem. 1993; 268: 22908-22913Abstract Full Text PDF PubMed Google Scholar and Table I). The mutant was isolated by introducing a second mutation into mutant PGS-P, in which PGP synthase activity is 50% of that in the parental strain CHO-K1 (23Ohtsuka T. Nishijima M. Akamatsu Y. J. Biol. Chem. 1993; 268: 22908-22913Abstract Full Text PDF PubMed Google Scholar). The biosynthetic rate and content of PG in mutant PGS-S are both reduced to approximately 10% or less of that in CHO-K1. The biosynthetic rate and content of CL in mutant PGS-S are also reduced (23Ohtsuka T. Nishijima M. Akamatsu Y. J. Biol. Chem. 1993; 268: 22908-22913Abstract Full Text PDF PubMed Google Scholar). To examine whether the CHO PGS1 cDNA complements the defect in PG and CL biosynthesis of mutant PGS-S, the CHO PGS1 cDNA was cloned into pSV-OKneo, a mammalian expression vector with a G418-resistant determinant, and introduced into PGS-S cells. We analyzed five independent stable G418-resistant transformant cells for PGP synthase activity and found that PGP synthase activity of all transformants analyzed was higher than that of CHO-K1 cells. This result demonstrated that the CHO PGS1 cDNA complements the defect in PGP synthase activity in mutant PGS-S. A stable G418-resistant transformant, designated PGS-S/cPGS1, was purified and used for further characterization. The cell extract of transformant PGS-S/cPGS1 cultivated at 40 °C exhibited a 620-fold and 7.3-fold higher PGP synthase specific activity than that of mutant PGS-S and wild type CHO-K1, respectively (Table I).Table IPGP synthase activity of stable transformant PGS-S/cPGS1 cellsStrainPGP synthase activity(pmol/min)/mg of proteinCHO-K1193 ± 0.3PGS-S2.3 ± 0.8PGS-S/cPGS11420 ± 121Mutant PGS-S/cPGS1 cells with a temperature sensitive defect in PGP synthase were transfected with pSVneo-cPGS1, and a stable neo-resistant transformant (PGS-S/cPGS1) was isolated. Cells were grown at 40 °C for 4 days, and the cell extracts were assayed for PGP synthase activity. The data shown are the mean value ± S.D. of triplicate assays. Open table in a new tab Mutant PGS-S/cPGS1 cells with a temperature sensitive defect in PGP synthase were transfected with pSVneo-cPGS1, and a stable neo-resistant transformant (PGS-S/cPGS1) was isolated. Cells were grown at 40 °C for 4 days, and the cell extracts were assayed for PGP synthase activity. The data shown are the mean value ± S.D. of triplicate assays. To investigate the rate of PG biosynthesis in transformant PGS-S/cPGS1, cells were labeled with 32Pi for 2, 4, and 8 h at 40 °C, and the metabolically labeled lipids were analyzed. As shown in Fig. 4, PG biosynthetic rate in PGS-S/cPGS1 cells was approximately 13-fold and approximately 2-fold higher than that in PGS-S and CHO-K1 cells, respectively. The rate of CL biosynthesis in transformant PGS-S/cPGS1 was also higher than that in mutant PGS-S and similar to that in CHO-K1. Phospholipid composition of transformant PGS-S/cPGS1 was determined by long term labeling of intact cells with32Pi. When the cells were cultivated with32Pi for 4 days at 40 °C, the PG level in transformant PGS-S/cPGS1 was more than 30-fold higher than that in mutant PGS-S and about 2.5-fold higher than that in wild type CHO-K1. The CL level in transformant PGS-S/cPGS1 was 2.3-fold higher than that in mutant PGS-S cells and similar to that in CHO-K1 (TableII). Under our experimental conditions, the radioactivity incorporated into phospholipid per cell was virtually identical among the transformant, the mutant, and the wild type cells (data not shown). These results indicated that the expression of CHOPGS1 cDNA complements the defect of PG and CL biosynthesis in mutant PGS-S.Table IIPhospholipid composition of transformant PGS-S/cPGS1CHO-K1PGS-SPGS-S/cPGS1% of total phospholipidsPhosphatidylglycerol1.1 ± 0.1<0.12.8 ± 0.1Cardiolipin2.7 ± 0.61.2 ± 1.12.7 ± 0.6Phosphatidic acid1.5 ± 0.20.9 ± 0.13.0 ± 0.2Phosphatidylcholine
Referência(s)