FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2017309092> ?p ?o ?g. }

• W2017309092 endingPage "28721" @default.
• W2017309092 startingPage "28715" @default.
• W2017309092 abstract "The citrus phospholipid hydroperoxide glutathione peroxidase (cit-PHGPx) was the first plant peroxidase demonstrated to exhibit PHGPx-specific enzymatic activity, although it was 500-fold weaker than that of the pig heart analog. This relatively low activity is accounted for the catalytic residue of cit-PHGPx, which was found to be cysteine and not the rare selenocysteine (Sec) present in animal enzymes. Sec incorporation into proteins is encoded by a UGA codon, usually a STOP codon, which, in prokaryotes, is suppressed by an adjacent downstream mRNA stem-loop structure, the Sec insertion sequence (SECIS). By performing appropriate nucleotide substitutions into the gene encoding cit-PHGPx, we introduced bacterial-type SECIS elements that afforded the substitution of the catalytic Cys41 by Sec, as established by mass spectrometry, while preserving the functional integrity of the peroxidase. The recombinant enzyme, whose synthesis is selenium-dependent, displayed a 4-fold enhanced peroxidase activity as compared with the Cys-containing analog, thus confirming the higher catalytic power of Sec compared with Cys in cit-PHGPx active site. The study led also to refinement of the minimal sequence requirements of the bacterial-type SECIS, and, for the first time, to the heterologous expression inEscherichia coli of a eukaryotic selenoprotein containing a SECIS in its open reading frame. The citrus phospholipid hydroperoxide glutathione peroxidase (cit-PHGPx) was the first plant peroxidase demonstrated to exhibit PHGPx-specific enzymatic activity, although it was 500-fold weaker than that of the pig heart analog. This relatively low activity is accounted for the catalytic residue of cit-PHGPx, which was found to be cysteine and not the rare selenocysteine (Sec) present in animal enzymes. Sec incorporation into proteins is encoded by a UGA codon, usually a STOP codon, which, in prokaryotes, is suppressed by an adjacent downstream mRNA stem-loop structure, the Sec insertion sequence (SECIS). By performing appropriate nucleotide substitutions into the gene encoding cit-PHGPx, we introduced bacterial-type SECIS elements that afforded the substitution of the catalytic Cys41 by Sec, as established by mass spectrometry, while preserving the functional integrity of the peroxidase. The recombinant enzyme, whose synthesis is selenium-dependent, displayed a 4-fold enhanced peroxidase activity as compared with the Cys-containing analog, thus confirming the higher catalytic power of Sec compared with Cys in cit-PHGPx active site. The study led also to refinement of the minimal sequence requirements of the bacterial-type SECIS, and, for the first time, to the heterologous expression inEscherichia coli of a eukaryotic selenoprotein containing a SECIS in its open reading frame. selenocysteine glutathione peroxidase phospholipid hydroperoxide glutathione peroxidase citrus PHGPx selenocysteine insertion sequence open reading frame polymerase chain reaction polyacrylamide gel electrophoresis 4-(2-aminoethyl)-benzene-sulfonyl fluoride high pressure liquid chromatography Selenoproteins, which contain the rare amino acid selenocysteine (Sec)1 in their primary structure, have been so far identified in diverse organisms such as bacteria, archea, and mammals but not in yeast nor in plants (1Bock A. Forchhammer K. Heider J. Leinfelder W. Sawers G. Veprek B. Zinoni F. Mol. Microbiol. 1991; 5: 515-520Crossref PubMed Scopus (553) Google Scholar, 2Stadtman T.C. Annu. Rev. Biochem. 1996; 65: 83-100Crossref PubMed Scopus (814) Google Scholar). Because of the ionization and redox properties of selenium, Sec residues usually play a crucial role in the biological activity of selenoproteins; indeed, most have been characterized as oxido-reductase enzymes. Among these selenoenzymes, the mammalian family of glutathione peroxidases (GPx), which catalyze the reduction of hydrogen peroxide, lipid hydroperoxides, and other organic hydroperoxides by utilizing glutathione, was mostly investigated (3Flohe L. Yagi K. Lipid Peroxides in Biology and Medicine. Academic Press, New York1982: 149-159Crossref Google Scholar, 4Ursini F. Maiorino M. Brigelius F.R. Aumann K.D. Roveri A. Schomburg D. Flohe L. Methods Enzymol. 1995; 252: 38-53Crossref PubMed Scopus (667) Google Scholar). A member of this family, namely the phospholipid hydroperoxide glutathione peroxidase (PHGPx), was shown to play a key role in the cell protection against oxidative damage, presumably by its ability to control the peroxidation of unsaturated phospholipids comprised in biological membranes (5Maiorino M. Gregolin C. Ursini F. Methods Enzymol. 1990; 186: 448-457Crossref PubMed Scopus (227) Google Scholar,6Arai M. Imai H. Koumura T. Yoshida M. Emoto K. Umeda M. Chiba N. Nakagawa Y. J. Biol. Chem. 1999; 274: 4924-4933Abstract Full Text Full Text PDF PubMed Scopus (260) Google Scholar). The citrus PHGPx (cit-PHGPx), encoded by the csa gene, was isolated and characterized in our laboratory and found to exhibit structure and substrate specificity similar to that of the animal PHGPx (7Ben-Hayyim G. Faltin Z. Gepstein S. Camoin L. Strosberg A.D. Eshdat Y. Plant Sci. 1993; 88: 129-140Crossref Scopus (47) Google Scholar, 8Faltin Z. Camoin L. Ben-Hayyim G. Perl A. Beeor-Tzahar T. Strosberg A.D. Holland D. Eshdat Y. Physiol. Plant. 1998; 104: 741-746Crossref Scopus (32) Google Scholar). However, the cit-PHGPx catalytic residue, as well as that of other plant analogs so far identified, is Cys and not Sec, which is the one present in mammalian enzymes (8Faltin Z. Camoin L. Ben-Hayyim G. Perl A. Beeor-Tzahar T. Strosberg A.D. Holland D. Eshdat Y. Physiol. Plant. 1998; 104: 741-746Crossref Scopus (32) Google Scholar). As predicted, considering the lower reductive power under physiological conditions of Cys as compared with Sec, this difference in their catalytic site affects the enzymatic properties of the plant peroxidase. We found the enzymatic activity of the cit-PHGPx to be 500-fold weaker than that of the pig heart PHGPx, and similar to that of the engineered pig heart PHGPx, in which the catalytic Sec was replaced by a Cys residue by point mutagenesis (9Maiorino M. Aumann K.D. Brigelius-Flohe R. Doria D. van den Heuvel J. McCarthy J. Roveri A. Ursini F. Flohe L. Biol. Chem. Hoppe-Seyler. 1995; 376: 651-660Crossref PubMed Scopus (218) Google Scholar,10Eshdat Y. Holland D. Faltin Z. Ben-Hayyim G. Physiol. Plant. 1997; 100: 234-240Crossref Google Scholar). In all organisms investigated so far, Sec is encoded by a UGA opal codon, usually a STOP codon (11Zinoni F. Birkmann A. Leinfelder W. Bock A. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 3156-3160Crossref PubMed Scopus (177) Google Scholar, 12Tate W.P. Mansell J.B. Mannering S.A. Irvine J.H. Major L.L. Wilson D.N. Biochemistry. 1999; 64: 1342-1353PubMed Google Scholar). However, the presence of a downstream mRNA stem-loop structure, designated as the Sec insertion sequence (SECIS), precludes termination of the polypeptide biosynthesis and promotes Sec incorporation into the nascent protein (2Stadtman T.C. Annu. Rev. Biochem. 1996; 65: 83-100Crossref PubMed Scopus (814) Google Scholar, 13Baron C. Bock A. Soll D. RhajBhandary U. tRNA: Structure, Biosynthesis and Function. American Society for Microbiology, Washington, D.C.1995: 529-544Google Scholar). Trans elements are also required for this decoding process, including a peculiar tRNASec, encoded by theselC gene, which carries a UCA anticodon (14Leinfelder W. Zehelein E. Mandrand-Berthelot M.A. Bock A. Nature. 1988; 331: 723-725Crossref PubMed Scopus (314) Google Scholar). This tRNASec is recognized by a specific elongation factor, SelB, which interacts specifically with both tRNASec and the mRNA stem-loop structure (15Kromayer M. Wilting R. Tormay P. Bock A. J. Mol. Biol. 1996; 262: 413-420Crossref PubMed Scopus (98) Google Scholar). The molecular mechanism that affords Sec incorporation into the polypeptide chain was identified in the bacterial and animal kingdoms but differs substantially between the two. In eukaryotes, the SECIS is located in the 3′-untranslated region of the mRNA. It has been shown that fusing a mammalian SECIS to the mRNA of any protein can direct Sec incorporation at each UGA codon introduced in the open reading frame (ORF) (16Berry M.J. Banu L. Harney J.W. Larsen P.R. EMBO J. 1993; 12: 3315-3322Crossref PubMed Scopus (348) Google Scholar, 17Shen Q. Chu F.F. Newburger P.E. J. Biol. Chem. 1993; 268: 11463-11469Abstract Full Text PDF PubMed Google Scholar). On the other hand, in the Escherichia coli fdhF gene, encoding the Sec-containing formate dehydrogenase H, the stem-loop structure is immediately adjacent to the 3′ side of the UGA codon and is therefore part of the ORF (18Zinoni F. Heider J. Bock A. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 4660-4664Crossref PubMed Scopus (208) Google Scholar). Consequently, expression of a eukaryotic selenoprotein in transformed bacteria would require the presence of a bacterial-type SECIS inside the ORF of the recombinant protein. Because of these species-specific requirements, attempts to obtain heterologous expression of genes coding for eukaryotic selenoproteins have been unsuccessful so far (19Tormay P. Bock A. J. Bacteriol. 1997; 179: 576-582Crossref PubMed Google Scholar), except for the thioredoxin reductase (20Gladyshev V.N. Krause M. Xu X.M. Korotkov K.V. Kryukov G.V. Sun Q.A. Lee B.J. Wootton J.C. Hatfield D.L. Biochem. Biophys. Res. Commun. 1999; 259: 244-249Crossref PubMed Scopus (76) Google Scholar, 21Arner E.S. Sarioglu H. Lottspeich F. Holmgren A. Bock A. J. Mol. Biol. 1999; 292: 1003-1016Crossref PubMed Scopus (202) Google Scholar). In this protein, Sec is the C-terminal penultimate residue; thus, the bacterial-type SECIS could be fused to the 3′ end of the gene without introducing any further change in the ORF. Other strategies were developed to overcome the constraint of engineering SECIS in the ORF. Sec insertion into the polypeptide structure has been achieved by chemical modification of Ser residue (22Ding L. Liu Z. Zhu Z. Luo G. Zhao D. Ni J. Biochem. J. 1998; 332: 251-255Crossref PubMed Scopus (77) Google Scholar, 23Liu J.-Q. Jiang M.-S. Luo G.-M. Yan G.-L. Shen J.-C. Biotechnol. Lett. 1998; 20: 693-696Crossref Scopus (25) Google Scholar) or mischarging of tRNACys with Sec when Cys was omitted from the growth medium (24Muller S. Senn H. Gsell B. Vetter W. Baron C. Bock A. Biochemistry. 1994; 33: 3404-3412Crossref PubMed Scopus (148) Google Scholar, 25Boschi-Muller S. Muller S. Van Dorsselaer A. Bock A. Branlant G. FEBS Lett. 1998; 439: 241-245Crossref PubMed Scopus (57) Google Scholar). However, these strategies do not afford directed substitution, because all Ser or Cys residues along the recombinant protein are subjected to Sec conversion. Consequently, the structure-function characterization of selenoproteins and the potential use of novel engineered selenoenzymes, carrying Sec as a catalytic residue and possessing enhanced biological activity, have been hampered. Here we report for the first time the successful heterologous expression in E. coli of an engineered eukaryotic protein, namely cit-PHGPx, in which the native catalytic Cys residue was replaced by Sec. This substitution conferred enhanced peroxidase activity to the recombinant enzyme. This study clearly demonstrates the ability to engineer and express eukaryotic selenoproteins in bacteria, which is of interest for the biochemical study of this class of proteins and for the design of novel selenoproteins. E. coli strains DH10B and DH5α (Life Technologies, Inc.) were used as host for the cloning and expression experiments throughout this study. Plasmid pBluescript SK− (Stratagene) was used as the cloning vector for mutagenesis and expression experiments. E. coli cells were grown in LB medium containing 50 μg/ml of ampicillin and 5 μm sodium selenite (Sigma). Higher concentration of sodium selenite (20 μm) was suspected to be toxic because reddish bacterial pellets were obtained after centrifugation. All recombinant DNA manipulations were carried out by standard procedures. Restriction enzymes were obtained from New England Biolabs and from Roche Molecular Biochemicals. The Vent DNA polymerase (New England Biolabs) was used for PCR amplification. We generated a series of mutated csagenes carrying different mRNA SECIS motifs. The mutagenesis was done by means of PCR technique. The pARO1 plasmid (26Holland D. Faltin Z. Perl A. Ben-Hayyim G. Eshdat Y. FEBS Lett. 1994; 337: 52-55Crossref PubMed Scopus (28) Google Scholar), which carries the csa gene encoding the cit-PHGPx protein, served as a template. The strategy for the csa mutagenesis is presented in Fig. 1. The required mutations were introduced into the csa gene using the appropriate primers during PCR amplification (Fig. 1 A). SECIS elements were encoded by two groups of primers designated as prTGT or prTGA, and prST1, prST1b, or prST2. The prTGT and prTGA primers possessed the anticodon encoding the catalytic residue of cit-PHGPx variants (amino acid at location 41 in the native enzyme), either Cys or Sec, respectively. The prST1, prST1b, and prST2 primers encoded the mRNA stem-loop of the corresponding SECIS motif ST1,ST1b, and ST2 (Fig. 1 B). The prST+ and prST− primers were hybridized to the ColE1 origin and contained the AlwNI restriction site. PCR fragments, resulting from amplification of the sequence lying between prST+ and prTGT or prTGA on one hand and between prST− and prST1, prST1b, or prST2 on the other hand were purified on agarose gel. The first fragment was then restricted withAlwNI and HpaI enzymes, whereas the second fragment was restricted with the AlwNI enzyme. After ligation, the resulting plasmid was transformed into competent cells. The mutated csa sequences were verified by DNA sequence analysis. The synthetic selC gene was constructed as described previously (27Normanly J. Masson J.M. Kleina L.G. Abelson J. Miller J.H. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 6548-6552Crossref PubMed Scopus (126) Google Scholar, 28Chen G.T. Axley M.J. Hacia J. Inouye M. Mol. Microbiol. 1992; 6: 781-785Crossref PubMed Scopus (19) Google Scholar). The entire tRNASec coding region (14Leinfelder W. Zehelein E. Mandrand-Berthelot M.A. Bock A. Nature. 1988; 331: 723-725Crossref PubMed Scopus (314) Google Scholar) was synthesized by annealing five complementary oligonucleotides, as shown in Fig. 2. This synthetic gene, containing an EcoRI restriction site at the 5′ end and aPstI restriction site at the 3′ end, was inserted into the unique EcoRI and PstI sites of pSU18 plasmid (29Martinez E. Bartolome B. de la Cruz F. Gene (Amst.). 1988; 68: 159-162Crossref PubMed Scopus (253) Google Scholar). In the resulting construct, the synthetic selC gene is located downstream the lacZ promoter. This plasmid was then restricted with FspI and SgrAI enzymes, and the resulting fragment was inserted into the unique NgoMIV andBsaAI restriction sites of the pARO1 plasmid. Purification and enzymatic characterization of the modified cit-PHGPx were performed using bacteria overexpressing selC gene (see below). Bacteria (5 ml,A 600 = 3) were centrifuged, the pellet was resuspended in 100 μl of 50 mm Tris-HCl, 5 mmEDTA, and 1 mm AEBSF, pH 7.5, frozen in liquid nitrogen and thawed three times. Cellular debris was removed by centrifugation (12,000 × g, 20 min). Total protein concentration was determined according to Bradford using the Bio-Rad protein assay. Soluble protein samples (10 μg/well) were analyzed by SDS-PAGE followed by Coomassie Blue staining or by Western blotting using anti-cit-PHGPx antibodies, peroxidase-conjugated second antibody, and ECL as substrate (7Ben-Hayyim G. Faltin Z. Gepstein S. Camoin L. Strosberg A.D. Eshdat Y. Plant Sci. 1993; 88: 129-140Crossref Scopus (47) Google Scholar). Estimation of the relative amounts of protein in bands of interest was obtained by densitometry, using TINA 2.07d software. The recombinant proteins were purified by reverse-phase HPLC using an Aquapore BU-300 column (100 × 2.1 mm). Lyophilized crude extract from E. coli was diluted up to 1 ml with 0.1% trifluoroacetic acid in water (solution A) and was centrifuged at 1400 × g for 10 min. The supernatant was immediately applied on the Aquapore column prewashed with solution A. Elution was performed for 92 min at a flow rate of 0.8 ml/min by applying a linear gradient starting from solution A up to 0.09% trifluoroacetic acid in 70% acetonitrile, using a Waters 625 LC System. Absorbance between 200 and 300 nm was monitored during the chromatography using Waters 991 Photodiode array detector. Fractions were collected manually and frozen immediately after their elution from the column and kept at −20 °C. Protein samples were dissolved in water:methanol:formic acid (50:50:5) and were introduced to API 365 triple-quadrupole mass spectrometer (Perkin-Elmer-Sciex, Thornill, Canada) with a syringe pump (5 μl/min; Harvard Apparatus, South Natick, MA). The device was equipped with an atmospheric pressure ion source, used for sampling positive ions produced from a pneumatically assisted electrospray interface. The ion spray probe tip was held at 4.5 kV, and the orifice voltage was set at 130 V. The mass spectrometer was scanned continuously from m/z 800 to 1700, with a scan step of 0.1 and a dwell time/step of 2.0 ms, resulting in scan duration of 22 s. Ten scans were averaged for each analysis. Mass calibration of the instrument was accomplished by matching ions of polypropylene glycol to known reference masses stored in the mass calibration table of the mass spectrometer. Data was collected on a Power Macintosh 8600/200 and processed through the Biotoolbox 2.3 software obtained from Sciex. Pellets, corresponding to 50-ml cultures of transformed bacteria, were resuspended in 500 μl of buffer (100 mm Tris-HCl, 5 mm EDTA, 1 mm AEBSF, and 5 mm β-mercaptoethanol, pH 8), frozen in liquid nitrogen, and thawed three times. Cellular debris was removed by centrifugation at 12,000 × g for 30 min. PHGPx activity was measured at room temperature (30Beeor-Tzahar T. Ben-Hayyim G. Holland D. Faltin Z. Eshdat Y. FEBS Lett. 1995; 366: 151-155Crossref PubMed Scopus (84) Google Scholar) based on the assay described by Maiorino et al. (5Maiorino M. Gregolin C. Ursini F. Methods Enzymol. 1990; 186: 448-457Crossref PubMed Scopus (227) Google Scholar). The assay mixture (total volume, 1 ml) contained 100 mm Tris-HCl, pH 8.3, 5 mm EDTA, 15 mm glutathione, 0.15 mmNADPH, 3 units of glutathione reductase, and 250 μg of total protein sample. After 5 min of preincubation, phosphatidylcholine hydroperoxide was added to give a final concentration of 20 μm, and the rate of the change in the absorbance at 340 nm was monitored, as compared with the absorbance change of the same assay mixture in the absence of phosphatidylcholine hydroperoxide. Specific activity of Cys-containing cit-PHGPx proteins was calculated as the ratio between the total PHGPx activity measured in crude extracts and the amount of the enzyme actually synthesized, as estimated by the densitometry analysis. Specific activity of cit-PHGPx variants was expressed as a percentage of native cit-PHGPx specific activity. The relative specific activity of the ST2Secprotein versus ST2Cys was calculated as the ratio between the total PHGPx activity measured in crude extracts, and the amount of enzyme that was determined by Western blot analysis. Nontransformed E. coli cells do not exhibit any endogenous PHGPx-like activity. We aimed at introducing in thecsa ORF variants of the fdhF SECIS (Fig.3 A, panel a) (18Zinoni F. Heider J. Bock A. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 4660-4664Crossref PubMed Scopus (208) Google Scholar), which would afford Sec incorporation into the recombinant cit-PHGPx while preserving the enzymatic activity. The constructs were generated taking into consideration three essential requirements (31Heider J. Baron C. Bock A. EMBO J. 1992; 11: 3759-3766Crossref PubMed Scopus (154) Google Scholar, 32Liu Z. Reches M. Groisman I. Engelberg-Kulka H. Nucleic Acids Res. 1998; 26: 896-902Crossref PubMed Scopus (82) Google Scholar): (i) the eight nucleotides defining the fdhF loop domain should be kept unchanged; (ii) base pairing on the fdhF bulge domain should be preserved and the bulged U17 remain invariant; and (iii), the length of the intermediary domain should comprise 11–14 nucleotides (Fig. 3 A). Accordingly, two SECIS variants called ST1 andST2 were designed (Fig. 3 A, panels band d). As compared with the fdhF SECIS, we particularly altered the bulge domain of ST1 stem-loop by attempting to do conservative substitutions like Tyr48 → Phe or Thr49 → Ser to preserve the cit-PHGPx activity (Fig. 3 B). In contrast, the bulge domain of theST2 motif was more preserved so as to ensure its stability and its recognition by the translation machinery. Moreover, to shorten the length of the intermediary domain to 14 nucleotides, we deleted in both SECIS variants the three nucleotides encoding the residue Ser46 in cit-PHGPx. This deletion and the mutation of the flanking Asn45 and Asn47 to Gly and Gln, respectively, are based on the protein sequence of the human pGPx (33Chu F.F. Esworthy R.S. Doroshow J.H. Doan K. Liu X.F. Blood. 1992; 79: 3233-3238Crossref PubMed Google Scholar), where a one-residue deletion appears at position 46 as compared with other GPxSec (Fig. 3 B). The genes encoding ST1Cys and ST2Cys proteins contained theST1 and ST2 SECIS motifs, respectively, and still possessed a UGU codon for Cys41 as the catalytic residue. The genes encoding ST1Sec and ST2Sec carried a UGA opal codon that aimed to direct Sec incorporation at position 41 and not termination of the polypeptide synthesis (Fig.3 B). At first, ST1Cys and ST2Cys mutants were characterized to determine whether the SECIS mutagenesis affected cit-PHGPx biochemical properties. As shown in Fig. 4, ST1Cys and ST2Cys proteins (lanes 3 and 4) were visualized by SDS-PAGE followed by Coomassie Blue staining and were found to migrate slightly faster than cit-PHGPx (lane 2). We evaluated by densitometry analysis that the bands assigned to ST1Cys and ST2Cys presented intensity corresponding approximately to 75 and 113% of that of cit-PHGPx, respectively (Table I).Table IPeroxidase activity of Cys-containing cit-PHGPx variants towards phosphatidylcholine hydroperoxidecit-PHGPx variantsPHGPx activity1-aEnzymatic activity was measured in crude extract prepared from E. coli grown in LB medium. A unit is 1 nmol of NADPH oxidized/min, with respect to protein extract under the conditions described under “Experimental Procedures.”PHGPx quantitation1-bAmount of cit-PHGPx variants in crude extract was evaluated by Coomassie Blue staining and densitometry analysis. cit-PHGPx concentration was arbitrarily set at 100.Specific activity1-cPercentage of cit-PHGPx specific activity.units/mg of extract protein% cit-PHGPxcit-PHGPx13.8 ± 2.3100100ST1Cys6.5 ± 2.775 ± 2162.5 ± 9.6ST2Cys11.4 ± 4.1113 ± 1772 ± 9.5Values are given as the means ± S.E. of at least four independent measurements.1-a Enzymatic activity was measured in crude extract prepared from E. coli grown in LB medium. A unit is 1 nmol of NADPH oxidized/min, with respect to protein extract under the conditions described under “Experimental Procedures.”1-b Amount of cit-PHGPx variants in crude extract was evaluated by Coomassie Blue staining and densitometry analysis. cit-PHGPx concentration was arbitrarily set at 100.1-c Percentage of cit-PHGPx specific activity. Open table in a new tab Values are given as the means ± S.E. of at least four independent measurements. Determination of the enzymatic activities of the different protein preparations (Table I) revealed that the ST1Cys mutant displayed a specific activity attaining 62.5% of that of the cit-PHGPx, whereas the ST2Cys mutant exhibited a specific activity corresponding to 72% of that of cit-PHGPx. Therefore, despite the introduction of SECIS motifs that led to seven amino acids replacements in the primary sequence of cit-PHGPx (Fig. 3 B), most of the peroxidase activity was preserved in the Cys-containing variants. To examine Sec incorporation into the ST1Sec and ST2Secpolypeptides, bacteria expressing the engineered cit-PHGPx were grown in LB medium complemented or not with 5 μm sodium selenite. Protein extracts were analyzed by SDS-PAGE and Western blotting. As shown in Fig. 5, the addition of selenium to the growth medium enabled the synthesis of ST1Sec and ST2Sec (lanes 3 and5), whose apparent molecular weight was identical to that of ST1Cys and ST2Cys. The yield of ST2Sec was higher than that of ST1Sec. Therefore, suppression of the termination codon was more efficient for the ST2Sec mutant than for ST1Sec, correlating with the fact that ST2 stem-loop resembled more closely thefdhF SECIS than the ST1 motif (Fig.3 A). Nevertheless, the amount of Sec-containing cit-PHGPx appeared to be much lower than that of the native cit-PHGPx under the same conditions (Fig. 5, lanes 1, 3, and5). In an attempt to increase ST1Sec expression, we designed the ST1b SECIS, in which A17 and U35are base paired (Fig. 3 A, panel c). The compensatory mutation A35 → T35 was expected to improve the stability of the ST1 stem-loop. However, as shown in Fig. 5 (lane 4), the SDS-PAGE analysis did not reveal any production of ST1bSec protein, demonstrating that the substitution A35 → T35 led to the complete abolition of Sec incorporation. To examine the effect of tRNASec concentration on the level of cit-PHGPxSec expression, a synthetic selC gene was inserted into the plasmids coding for the cit-PHGPx variants (see “Experimental Procedures”). As shown in Fig. 5 (lane 6), the overexpression of tRNASec led to a significant increase of the ST2Sec production; a similar result was obtained for ST1Sec as well (data not shown). After quantitative determination of immunoblot bands, the estimated amount of ST2Sec protein in crude extract was about 8-fold lower than that of the ST2Cys protein (Fig.6). Overexpression of tRNASecalso enhanced the synthesis of a cit-PHGPx protein in the absence of selenium complementation (Fig. 5, lane 6). We assumed that such level of synthesis was due to selenium traces, which are certainly present in LB medium (see “Discussion”). Recombinant cit-PHGPx and its ST2Cys and ST2Sec mutants were partially purified from soluble fraction of bacterial extract using one-step HPLC. All three proteins were eluted after 56 min and gave very distinct peaks (Fig. 7 A). ST2Cys was purified almost to homogeneity (Fig.7 B, lane 2), whereas ST2Sec, whose amount in crude extract is substantially lower than that of ST2Cys, was estimated to be 50% pure (Fig. 7 B,lane 1). Using an integration procedure that measures the peak absorbance in comparison with a quantified cit-PHGPx standard, it was estimated that ST2Cys and ST2Secpurification yielded about 2.30 and 0.23 mg of protein, respectively, per liter of culture. These amounts are in agreement with the relative concentration of ST2Sec versusST2Cys previously calculated (Fig. 6). The fractions containing either ST2Cys or ST2Sec, obtained by the HPLC procedure (Fig. 7), did not exhibit any catalytic activity, probably because of denaturation of the protein during the purification procedure. They were used, however, to establish the chemical composition of the amino acid 41 of ST2Cys and ST2Sec by mass spectrometry. The HPLC fractions corresponding to ST2Sec and ST2Cys were analyzed by mass spectrometry (Fig. 8). A major peak representing a mass of 18258 was obtained for the purified ST2Cys, corresponding to the predicted mass of cit-PHGPx variant containing Cys41. As for ST2Sec, a major peak corresponding to a mass of 18,305 was obtained. The mass difference of 47 between ST2Cys and ST2Secpeaks, corresponding to the atomic weight difference between selenium and sulfur, clearly demonstrated Sec incorporation into the ST2Sec enzyme. Moreover, ST2Sec profile did not reveal any other peak corresponding to the incorporation at position 41 of an amino acid other than Sec. Therefore, the ST2Secprotein synthesized by bacteria grown in selenium-supplemented medium was exclusively a Sec-containing enzyme. The PHGPx activity exhibited by the soluble fractions of E. coli expressing cit-PHGPx variants was determined. It appeared that the Sec-containing cit-PHGPx was actually capable of catalyzing the reduction of phosphatidylcholine hydroperoxide (Fig.9). Under conditions of selenium deficiency in LB medium, a total PHGPx activity of 0.6 and 0.7 unit was measured in the crude extracts obtained from bacteria expressing ST1Sec and ST2Sec, respectively (Fig. 9). When growth medium was supplemented with 5 μm sodium selenite, PHGPx activity from bacteria expressing ST1Sec and ST2Secincreased significantly, attaining 3.2 and 6 units, respectively, confirming that selenium complementation induced cit-PHGPxSec synthesis as shown in Fig. 5. Protein extracts from bacteria expressing ST2Sec and ST2Cys, used for the measurements of PHGPx activity (Fig. 9), were subjected to Western blot analysis as shown in Fig. 6. Considering that bacteria synthesizing ST2Sec displayed a total activity 2 ± 0.7-fold lower than that of bacteria expressing ST2Cys, whereas the corresponding amount of ST2Sec was 8.2 ± 0.6-fold lower than that of ST2Cys (Fig. 6), the specific activity of the ST2Sec peroxidase was estimated to be 4 ± 1.3-fold higher than that of the ST2Cysenzyme. In contrast to the simplicity of substituting Cys for Sec in selenoproteins via simple point mutagenesis (TGA to TGT), the reverse mutation, i.e. from Cys to Sec, is not straightforward because a point mutation from TGT to TGA will introduce a STOP codon, which normally terminates protein translation. In E. coli, TGA is decoded as a sense codon for Sec only if a bacterial-type SECIS element is located immediately downstream the TGA codon (2Stadtman T.C. Annu. Rev. Biochem. 1996; 65: 83-100Crossref PubMed Scopus (814) Google Scholar, 13Baron C. Bock A. Soll D. RhajBhandary U. tRNA: Structure, Biosynthesis and Function. American Society for Microbiology, Washington, D.C.1995: 529-544Google Scholar). In the present study, we aimed at substituting Sec for the catalytic Cys41 residue of cit-PHGPx, and to produce, in prokaryotic cells, a recombinant Sec-containing eukaryotic glutathione peroxidase. The major obstacle to such engineering lies in the fact that introduction of SECIS within the csa ORF should result in alteration of the cit-PHGPx amino acid sequence and may affect its function. Accordingly, on the basis of the bacterial fdhFSECIS, we attempted to engineer two SECIS-like motifs, designated asST1 and ST2, within the csa ORF, in a way that minimized the number of amino acid substitutions in cit-PHGPx (Fig. 3). We initially characterized the ST1Cys and ST2Cys proteins whose mRNA possessed an engineered SECIS motif but still contained a UGU codon encoding the catalytic Cys41. ST1Cys and ST2Cys displayed specific activity attaining 62.5 and 72% of that of cit-PHGPx, respectively (Table I), demonstrating that cit-PHGPx biosynthesis and most of its catalytic activity were preserved despite the structural modifications induced by the SECIS mutagenesis. Such result was probably due to the fact that the amino acid substitutions were designed in a rather poorly conserved domain of the PHGPx family (Fig.3 B). We then examined whether the engineered SECIS were able to direct Sec incorporation. The UGA codon inserted in the cit-PHGPx reading frame, in addition to the stem-loop, was recognized as a sense codon, because full-sized ST1Sec and ST2Sec proteins were synthesized (Fig. 5). Moreover, this synthesis was dependent upon the addition of selenium to the growth medium. This evidence for Sec incorporation into ST1Sec and ST2Sec was conclusively confirmed by mass spectrometry analysis of the purified ST2Sec protein (Fig. 8). Thus, despite drastic mutations in the bulge domain, especially for the ST1 SECIS in which six of the eight native nucleotides were changed (Fig. 3), the altered SECIS still afforded Sec incorporation, showing that a particular array of nucleotides in the bulge domain is not essential for function and may be modified. In addition to these nucleotide substitutions, the intermediary domain between the Sec codon and the stem-loop was three nucleotides longer than that of the bacterial fdhF SECIS. This additional codon probably led to a substantial decrease in the efficiency of Sec incorporation, as suggested in previous works done on the native fdhF SECIS (31Heider J. Baron C. Bock A. EMBO J. 1992; 11: 3759-3766Crossref PubMed Scopus (154) Google Scholar, 32Liu Z. Reches M. Groisman I. Engelberg-Kulka H. Nucleic Acids Res. 1998; 26: 896-902Crossref PubMed Scopus (82) Google Scholar). We attempted to obtain a higher amount of ST1Sec protein by substituting A35 by U to establish the base pairing A17-U35. Because the base A35 was not described as belonging to the minimal SECIS motif allowing Sec incorporation (32Liu Z. Reches M. Groisman I. Engelberg-Kulka H. Nucleic Acids Res. 1998; 26: 896-902Crossref PubMed Scopus (82) Google Scholar), we expected that this complementary mutation could improve the stability of the mRNA stem-loop and consequently, enhance the UGA-directed Sec incorporation. However, Sec incorporation was abolished by this single mutation (Fig. 5), suggesting that the base A35, which is, in fact, conserved among all the SECIS identified in fdh genes from E. coli,Enterobacter aerogenes, and Hemeophilus influenzae (34Wilting R. Vamvakidou K. Bock A. Arch. Microbiol. 1998; 169: 71-75Crossref PubMed Scopus (17) Google Scholar), plays a critical role and therefore should be added to the minimal SECIS requirements. By overexpressing tRNASec (28Chen G.T. Axley M.J. Hacia J. Inouye M. Mol. Microbiol. 1992; 6: 781-785Crossref PubMed Scopus (19) Google Scholar, 35Tormay P. Sawers A. Bock A. Mol. Microbiol. 1996; 21: 1253-1259Crossref PubMed Scopus (47) Google Scholar), the selenoprotein production was significantly increased, thus providing additional evidence that ST1Sec and ST2Sec expression was actually governed by the Sec incorporation mechanism. Probably because of the presence of selenium traces in LB medium, tRNASecoverexpression also enhanced the expression of a cit-PHGPX protein when sodium selenite was not added to the growth medium. Nevertheless, we cannot exclude the possibility that this synthesis was also due to an additional read-through mechanism resulting in Trp incorporation. Indeed, it has been recently shown that the presence of a G nucleotide adjacent to the UGA on its 3′ side in the fdhF SECIS, as found in the ST1 and ST2 motifs as well (Fig.3 A), promotes Trp-inserting opal suppression when selenium is omitted (36Liu Z. Reches M. Engelberg-Kulka H. J. Mol. Biol. 1999; 294: 1073-1086Crossref PubMed Scopus (17) Google Scholar, 37Sandman K.E. Noren C.J. Nucleic Acids Res. 2000; 28: 755-761Crossref PubMed Scopus (26) Google Scholar). However, the fact that mass spectrometry analysis of ST2Sec showed no indication for incorporation at position 41 of an amino acid other than Sec (Fig. 8), suggests that when bacteria were grown in sodium selenite-supplemented medium, Sec incorporation pathway was greatly favored. Substitution of native Sec by Cys in mouse cellular GPxSec(38Rocher C. Lalanne J.L. Chaudiere J. Eur. J. Biochem. 1992; 205: 955-960Crossref PubMed Scopus (136) Google Scholar) and in pig heart PHGPxSec (9Maiorino M. Aumann K.D. Brigelius-Flohe R. Doria D. van den Heuvel J. McCarthy J. Roveri A. Ursini F. Flohe L. Biol. Chem. Hoppe-Seyler. 1995; 376: 651-660Crossref PubMed Scopus (218) Google Scholar) led to a dramatic reduction of the enzymatic activity, about 1000- and 250-fold, respectively. Consistently, the naturally Cys-containing cit-PHGPx displays catalytic activity of only 0.2% of that of the pig heart PHGPxSec (10Eshdat Y. Holland D. Faltin Z. Ben-Hayyim G. Physiol. Plant. 1997; 100: 234-240Crossref Google Scholar). If the relatively low activity of cit-PHGPx is only due to the presence of a catalytic Cys instead of Sec, a superactive cit-PHGPx could possibly be generated just by replacing the Cys41 residue by Sec. Our results show that the recombinant ST2Sec PHGPx actually exhibited a 4-fold higher specific activity than that of the Cys-containing control analog, ST2Cys. Thus, it appeared that the cit-PHGPxSecvariants possibly do not possess the appropriate physico-chemical and conformational environment that confer to Sec its particularly high contribution to the catalytic efficiency of animal GPxSec. Further investigation on structure-function relationships of GPxSec is therefore required to produce, in bacteria, an improved GPxSec displaying an activity comparable with that of mammalian enzymes. Within this scope, the system developed here is of high potential for performing mutational analysis that is unconceivable in animal cell system and for producing and characterizing other novel selenoproteins as well. We thank Abdelkader Namane from the Pasteur Institute for carrying out the mass spectrometry analysis. Thanks are also due to Dominique Granger and Limor Mordechai for secretarial assistance. We are very grateful to Hermine Klepal for constant support." @default.
• W2017309092 created "2016-06-24" @default.
• W2017309092 creator A5002143582 @default.
• W2017309092 creator A5021912565 @default.
• W2017309092 creator A5041239930 @default.
• W2017309092 creator A5046232163 @default.
• W2017309092 creator A5074184173 @default.
• W2017309092 date "2000-09-01" @default.
• W2017309092 modified "2023-10-16" @default.
• W2017309092 title "Substituting Selenocysteine for Catalytic Cysteine 41 Enhances Enzymatic Activity of Plant Phospholipid Hydroperoxide Glutathione Peroxidase Expressed in Escherichia coli" @default.
• W2017309092 cites W116937710 @default.
• W2017309092 cites W1208971950 @default.
• W2017309092 cites W1521051093 @default.
• W2017309092 cites W1571092378 @default.
• W2017309092 cites W1591054099 @default.
• W2017309092 cites W1831825906 @default.
• W2017309092 cites W1963506606 @default.
• W2017309092 cites W1978096883 @default.
• W2017309092 cites W1981458029 @default.
• W2017309092 cites W2010949672 @default.
• W2017309092 cites W2016051676 @default.
• W2017309092 cites W2021906281 @default.
• W2017309092 cites W2022058624 @default.
• W2017309092 cites W2026985909 @default.
• W2017309092 cites W2033876764 @default.
• W2017309092 cites W2036954969 @default.
• W2017309092 cites W2043038437 @default.
• W2017309092 cites W2061603863 @default.
• W2017309092 cites W2079288809 @default.
• W2017309092 cites W2079932180 @default.
• W2017309092 cites W2081645231 @default.
• W2017309092 cites W2083671881 @default.
• W2017309092 cites W2088114614 @default.
• W2017309092 cites W2090180026 @default.
• W2017309092 cites W2091836061 @default.
• W2017309092 cites W2096857780 @default.
• W2017309092 cites W2097235090 @default.
• W2017309092 cites W2122690964 @default.
• W2017309092 cites W2168912171 @default.
• W2017309092 cites W2169695201 @default.
• W2017309092 cites W2331629748 @default.
• W2017309092 cites W4243453188 @default.
• W2017309092 cites W4249076235 @default.
• W2017309092 cites W4251440173 @default.
• W2017309092 cites W4300932895 @default.
• W2017309092 doi "https://doi.org/10.1074/jbc.m004985200" @default.
• W2017309092 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/10874045" @default.
• W2017309092 hasPublicationYear "2000" @default.
• W2017309092 type Work @default.
• W2017309092 sameAs 2017309092 @default.
• W2017309092 citedByCount "81" @default.
• W2017309092 countsByYear W20173090922012 @default.
• W2017309092 countsByYear W20173090922013 @default.
• W2017309092 countsByYear W20173090922014 @default.
• W2017309092 countsByYear W20173090922015 @default.
• W2017309092 countsByYear W20173090922016 @default.
• W2017309092 countsByYear W20173090922017 @default.
• W2017309092 countsByYear W20173090922018 @default.
• W2017309092 countsByYear W20173090922019 @default.
• W2017309092 countsByYear W20173090922021 @default.
• W2017309092 countsByYear W20173090922022 @default.
• W2017309092 countsByYear W20173090922023 @default.
• W2017309092 crossrefType "journal-article" @default.
• W2017309092 hasAuthorship W2017309092A5002143582 @default.
• W2017309092 hasAuthorship W2017309092A5021912565 @default.
• W2017309092 hasAuthorship W2017309092A5041239930 @default.
• W2017309092 hasAuthorship W2017309092A5046232163 @default.
• W2017309092 hasAuthorship W2017309092A5074184173 @default.
• W2017309092 hasBestOaLocation W20173090921 @default.
• W2017309092 hasConcept C104317684 @default.
• W2017309092 hasConcept C155064970 @default.
• W2017309092 hasConcept C158086472 @default.
• W2017309092 hasConcept C162008176 @default.
• W2017309092 hasConcept C181199279 @default.
• W2017309092 hasConcept C185592680 @default.
• W2017309092 hasConcept C203580378 @default.
• W2017309092 hasConcept C203663253 @default.
• W2017309092 hasConcept C2776087148 @default.
• W2017309092 hasConcept C2778760513 @default.
• W2017309092 hasConcept C2778918659 @default.
• W2017309092 hasConcept C2779201268 @default.
• W2017309092 hasConcept C2781264002 @default.
• W2017309092 hasConcept C41625074 @default.
• W2017309092 hasConcept C538909803 @default.
• W2017309092 hasConcept C547475151 @default.
• W2017309092 hasConcept C55493867 @default.
• W2017309092 hasConceptScore W2017309092C104317684 @default.
• W2017309092 hasConceptScore W2017309092C155064970 @default.
• W2017309092 hasConceptScore W2017309092C158086472 @default.
• W2017309092 hasConceptScore W2017309092C162008176 @default.
• W2017309092 hasConceptScore W2017309092C181199279 @default.
• W2017309092 hasConceptScore W2017309092C185592680 @default.
• W2017309092 hasConceptScore W2017309092C203580378 @default.
• W2017309092 hasConceptScore W2017309092C203663253 @default.
• W2017309092 hasConceptScore W2017309092C2776087148 @default.
• W2017309092 hasConceptScore W2017309092C2778760513 @default.
• W2017309092 hasConceptScore W2017309092C2778918659 @default.
• W2017309092 hasConceptScore W2017309092C2779201268 @default.

• next