JPH08509381A - Posttranscriptional regulation of genes by trace minerals - Google Patents

Abstract

  (57) [Summary] A first nucleic acid encoding a heterologous polypeptide, wherein at least one codon of the mRNA transcribed from said first nucleic acid has been replaced with codon UGA, and a second nucleic acid operably linked to said first sequence , Wherein the second nucleic acid is capable of directing translation of the UGA codon as selenocysteine, comprising the step of regulating the expression of a heterologous protein in vivo and in vitro. ..

Description

Detailed Description of the Invention Posttranscriptional regulation of genes by trace minerals BACKGROUND OF THE INVENTION   The present invention relates to post-transcriptional regulation of expression of heterologous genes.   One of the key goals of biotechnology is medically and commercially valuable The gene encoding the protein is expressed in the transfected gene. Stable transfection into cell and animal systems under current controllable conditions It is to operate. To date, the most widely used method is inductive By placing the gene of interest under the control of the Romano, it is possible to Controlling the expression of the gene. However, currently available induction Most of the sex promoters have significant background levels under non-inducing conditions. It allows the generation of genes, thus making the utility of these methods low-level transformers. Limited to applications where the transfected gene product has no significant effect on cells To do. Furthermore, most inducible promoters limit the expression of genes under inducing conditions. Only the increased gain (usually about 3 times) can be produced.   In addition, current methods transfer the gene of interest into cells that lack the gene product. It usually requires transfection. Because of the natural gene product Presence often results in signal-to-noise ratios that are difficult to assess and This interferes with the use of functional assays to test the generated gene product. is there In some cases, when cells lacking the gene product are not available, the transfer The protein derived from the transfected gene is a “epitope tag”. Or a small amino acid called "peptide flag" The addition of the sequence allows it to be differentially identified from natural cells. However , This approach is often limited to applications that do not require functional proteins . The sequence of the added peptide is essential for the functional activity of the protein. Includes protein folding and post-transcriptional processing It can interfere with a certain number of processes. SUMMARY OF THE INVENTION   We are novel amino acids of one or more amino acids in a polypeptide , Of the polynucleotide sequence resulting in the substitution by selenocysteine (SeCys). Discovered that alterations provide a way to control expression of genes at the level of translation did. Accordingly, in one aspect, the invention provides (a) heterologous polypeptides. Encoding the first nucleic acid, wherein at least one of the mRNAs transcribed from the first nucleic acid One codon is replaced by the codon UGA, and is operably linked to the first sequence The second nucleic acid, which has been generated, directs the translation of the UGA codon as selenocysteine. Is prepared, and (b) the heterologous polypeptide. Under conditions where the production of selenium is controlled by the level of selenium available to the cells Production of a heterologous polypeptide in a eukaryotic cell, including the step of growing the cell On how to control.   The method of the invention can be applied to any type of eukaryotic cell that can be maintained in cell culture. It can be performed in vitro on cells. Preferably the cell is a eukaryotic cell, eg For example, mammalian tissue culture cells (eg, COS-1, HL-60, CV-1, C-6, LLC / PK-1, 3T3L1 or CHO cells) or yeast cells, eg For example, Saccharomyces cerevisiae (Saccharomyces  cere visiae ). In one preferred embodiment, the cells used are recombinant cells. It does not contain naturally occurring proteins that are substantially homologous to the peptides. Only However, such cells are either not available or are homologous natural proteins. In the presence of substantial deficiencies as compared to cells containing cytoplasm. The peptide is a recombinant polypeptide for the nucleophile due to the presence of a selenocysteine residue. Due to increased reactivity of the peptide or radioisotope⁷⁵Radiation using Se Sex labeling can distinguish it from the native protein.   The first nucleic acid and the second nucleic acid are recombinant capable of autonomously replicating in a cell. Can be introduced and maintained in cells in a vector, or It can be stably integrated into the genome of cells according to standard techniques. Heterogeneous Polypep Tide production is controlled by the amount of selenium, a trace amount of minerals in the medium in which cells are cultured. To be done. Available selenium is available when it is desired to inhibit expression of the polypeptide. Is substantially absent, that is, the concentration of selenium in the medium is lower than 1 ng / ml. Cells are preferably maintained in medium below 0.1 ng / ml. Different type For inducing expression of the polypeptide, the cell culture medium is typically 1-50 ng / m 1, preferably 2-40 ng / ml, most preferably 5-25 ng / ml To do.   Alternatively, the method of the invention comprises deriving the first nucleic acid and the second nucleic acid from a non-human mammal. Stable integration into the derived germ cell genome and transfection of non-human mammals It can be carried out in vivo by obtaining transgenic offspring. "Transgenic", as used herein, artificially inserts into a cell. A DNA sequence that has been incorporated and is part of the genome of the animal that develops from the cell. It means a mammal including. Such a transgene is transgene It can be partially or completely heterologous to the animal. Trance Non-human mammals that can be produced by genetic techniques are included in the present invention; preferred In addition to mice, other mammals include rats, cows, pigs, sheep, goats, rabbits, Includes guinea pigs, hamsters, and horses. "Embryonic cell" is here When used, it includes embryonic stem (ES) cells and fertilized oocytes. Fertilized In the case of isolated oocytes, the preferred method of transgenic induction is micro For ES cells, the preferred method is electroporation. It is an option. However, virus delivery systems such as retroviruses Other methods, including transfection of liposomes, or fusion of liposomes can be used . After introduction of the transgene into the germ cells, the cells are introduced into pseudopregnant females and To obtain progeny that are heterozygous for the transgene. Then heterozygous Stable strains of animals of this species can be maintained by appropriate backcrossing to the original animal strain. Or heterozygous offspring can be bred to obtain homozygous animals. it can. Transgenic when trying to inhibit expression of a heterologous polypeptide Animals are maintained on a diet containing less than 0.02 mg / kg food and Current induction supplements the diet with selenium at concentrations of 0.1 mg / kg or higher Be started by.   The polypeptide encoded by the first nucleic acid has a known nucleotide sequence. Can be any desired polypeptide. Modify the polypeptide Methods for incorporating amino acid residues in selenocysteine are well known, and here Described in. Preferably, the selenocysteine residue is a protein found in nature. Amino acids at positions in the polypeptide that do not destroy normal biological activity of quality. , For example, is replaced with a nonessential amino acid. Such amino acids are It can be identified by means well known to those skilled in the Positions not related to the activity of the medium or binding (eg determined by mutation analysis ), Or a position considered important for the structural integrity of the polypeptide Usually (for example, as predicted by computer analysis or crystallography) Will exist. Most often, selenocysteine usually has cysteine residues. Will be inserted at a position in the polypeptide.   In a preferred embodiment, the second nucleic acid has a stem in the mRNA transcribed therefrom. -Comprising flanking sequences of nucleotides capable of forming loop secondary structure, Here, the stem-loop formed by mRNA is the above-mentioned selenocysteine. UGA translations can be commanded. In one preferred embodiment, the second nucleic acid Is 3'of the gene encoding naturally-occurring mammalian selenocysteine -Derived from approximately 90 contiguous nucleotides from the untranslated region. For example, 2 nucleic acids are human selenocysteine and glutathione peroxidase shown in FIG. Containing nucleotide sequences substantially homologous to nucleotides 654-740 of Mu. In another preferred embodiment, the second nucleic acid is synthetically derived and the stem- The sequence 5'-NAAAUNNUAAAN-3 'in the loop at the tip of the loop Can form a stem-loop, and the stem forms a bubble Contains at least 12, preferably at least 14 non-complementary nucleotides Where the bubbles are symmetrically opposed to each half of the array 5'-NUAGU. N-3 '; preferably the stem-loop contains approximately 80 nucleotides. Contains, the bubble is located approximately 17 nucleotides from the base of the stem, Then, the loop is almost 11 nucleotiated from the bubble at the tip of the stem-loop structure. It is located in the city.   Therefore, the present invention also provides a nucleoside capable of forming a stem-loop secondary structure. For single-stranded nucleic acids comprising flanking stretches of tide, the stem-loop loop is Contains the sequence 5'-NAAAUNNUAAAN-3 ', and the stem forms a bubble. Containing at least 12, preferably at least 14 non-complementary nucleotides , Where the bubbles are symmetrically opposed to each half of the array 5'-NUAGU. It contains N-3 '. Nucleic acid is operably linked to mRNA molecules containing UGA codons. When chained, it can direct the translation of the codon, UGA, as selenocysteine. You can Preferably, the stem-loop nucleic acid comprises approximately 90 nucleotides. Has a bubble located approximately 17 nucleotides from the base of the stem, and a loop Is located approximately 11 nucleotides from the bubble at the tip of the stem-loop structure It Also preferably, each half of the bubble is approximately 7 nucleotides, and The loop contains approximately 12 nucleotides.   The present invention also provides a double-stranded DNA containing DNA encoding the single-stranded nucleic acid of the present invention. Regarding nucleic acids.   A "heterologous" nucleic acid is a part or whole of the cell or animal into which it is introduced. Nucleic acid that is foreign to the Other than containing selenocysteine substituted in the mino acid, By a nucleic acid that is homologous to an endogenous gene.   The term "operably linked," as used herein, includes a stem-loop Adjacent stretches of nucleotides that form secondary structures encode proteins A protein that is sufficiently close to the nucleic acid to be translated as selenocysteine It is meant to allow translation of UGA codons in quality. Preferably the stem -The loop is in the 3'-untranslated region of the mRNA molecule that encodes the polypeptide. Preferably within 2000 nucleotides of the UGA codon, more preferably 40 Within 0 to 1500 nucleotides, most preferably 500 to 1200 nucleotides Inserted inside.   "Functionally active" is any in-character characteristic of a protein found in nature.   Refers to the processing of activity in vivo or in vitro.   "Homology", as used herein, refers to two polypeptide molecules or It refers to sequence similarity between two nucleic acid molecules. In both of the two compared sequences Are occupied by identical nucleotide bases or amino acid subunits , The molecules are homologous in their position. Thus, "substantially homologous "Is a nucleotide or amino acid that is, if not completely, homologous to Means an array of.   A "heterologous" nucleic acid is partial or partial to the animal into which it is transfected. Or a completely foreign nucleic acid, or within a transgenic animal. Homologous to the causative gene but at a position different from that of the natural gene By a nucleic acid that is inserted into the genome of an animal.   Unless otherwise noted, all technical and scientific terms used herein are It has the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Book methods and materials similar or equivalent to those described herein. Although preferred in practice and testing of the invention, preferred methods and materials are described herein. Enter. All publications mentioned below are incorporated herein by reference. Furthermore, the substance The methods and examples are illustrative only and not intended to be limiting.   The method of the present invention goes beyond currently used methods of gene expression. It offers several advantages. First, the SeCys The substitution absolutely depends on the supply of selenium and thus at the level of translation trans It allows virtually absolute control of the amount of gene product that has been transfected. Secondly , SeCys has all of the biological properties of amino acid residue Cys, thus C The replacement of SeCys with ys is usually done in the transfected gene product. Does not significantly alter the biological activity of. Third, a transfer containing SeCys The processed gene product has increased reactivity towards its nucleophile. Or⁷⁵Easily distinguished from native cellular proteins by Se incorporation Can be   Other features and advantages of the invention will be apparent from the following detailed description and claims. Will. Detailed description   The drawings are first briefly described.Drawing   FIG. 1A is a schematic of human cell glutathione peroxidase cDNA constructs. It is a figure. Open reading frame (ORF) and 3'UTR are wide Indicated by bars; plasmid and 5'UTR by flanking line It is shown. The nucleotide number is at the beginning of the open reading frame. The ATG start codon is present in nt1-3 and the TGA selenocystis Is present at nt 142-144, and the TAG stop codon is nt 607. -609. The arrow indicates the position of the restriction endonuclease site. Line diagram The lower line represents the position of the indicated deletion. The hatched bar below the diagram shows the epi Shows the position where the taupe labeling sequence is inserted and the region of the replaced cDNA .   FIG. 1B shows UGA in the coding region of human Gpx mRNA.₁₄₂SE Schematic and potential deletion of potential secondary structure immediately downstream of the lenocysteine codon Illustrates the locations of ORF-D1, ORF-D2, ORF-D3, and ORF-D4 To do.   FIG. 1C shows another potential mutation in the coding region of human Gpx mRNA. It is a schematic diagram of the next structure, where UGA₁₄₂The selenocysteine codon has a hairpin structure Exist in the building. The deletion ORF-D5 is also indicated.   Figure 2A shows pCMV4 (lane 1), native GPx (lane 2), or lacking. Using the mutants ORF-D1 to ORF-D4 (lanes 3 to 6 respectively) Immunoprecipitated after transfection⁷⁵Se-labeled COS-1 cell extract 2 is an autoradiograph of the SDS-polyacrylamide gel.   FIG. 2B shows pCMV4 vector (lane 1), native GPx (lane 2), Or transfection using the deletion mutant ORF-D5 (lane 3) Immunoprecipitation after⁷⁵SDS-polyacryle of Se-labeled COS-1 cell extract It is an autoradiograph of a luamide gel.   FIG. 3 shows pCMV4 vector (lane 1), epitope-tagged GPx (ray). 2) or deletion mutants UTR-D1 to UTR-D3 (lanes, respectively) Immunoprecipitated after transfection using 3-5)⁷⁵Se labeled CO Fig. 2 is an autoradiograph of SDS-polyacrylamide gel of S-1 cell extract. It   FIG. 4 shows pCMV4 vector (lane 1), epitope-tagged GPx (ray). 2), deletion mutant UTR-D4 (lane 3), or deletion mutant UT Immunoprecipitated after transfection using R-D5 (lane 4)⁷⁵S e-labeled COS-1 cell extract It is an autoradiograph of the SDS-polyacrylamide gel of a product.   FIG. 5 is a schematic diagram of the potential secondary structure of the 3'UTR of human Gpx mRNA. It   FIG. 6 shows the product of an RNase protection assay using labeled riboprobes. It is an autoradiograph of a polyacrylamide gel. Lane 1, undigested pro Lane; lane 2, RNA from untransfected COS-1 cells and high Hybridized probe; lane 3, epitope-tagged GPx COS -1 probe hybridized with transfectant; lane 4 , UTR-D4 COS-1 transfectants and hybridization Probe; lane 5, UTR-D5 COS-1 transfectants Hybridized probe.   FIG. 7. Transfection with rab5b opal mutant and fusion construct. Immunoprecipitated³⁵S-labeling (lanes 1-4) and⁷⁵Se labeling (ray 5-8) Autoradiography of SDS-polyacrylamide gel of COS-1 cells It's rough. Lanes 1 and 5, pCMV4 vector; lanes 2 and 6, ra. b5b (opal) GPx3'UTR; lanes 3 and 7, rab5b (opal) ); Lanes 4 and 8, rab5b (wt) GPx5'UTR.   FIG. 8 shows nucleotidies of human glutathione peroxidase containing 3'UTR. Draw the array.   FIG. 9 shows the sequence of the “optimized” selenocysteine insertion sequence (SECIS) and Depict secondary structure.Selenocysteine   Bacterial formate dehydrogenase, mammalian glutathione peroxidase (GPx) family (Mullenbach et al., N ucleic Acids Res.15: 5484, 1987; Chambu. s et al., EMBO J .;5: 1221, 1986; Esworthy et al., Arch. ． Biochem. Biophys.286: 330, 1991; Takaha si et al. Blood68: 640, 1986), type I iodothyronine 5'-de-yoyo Enzyme (Berry et al. (1991) Nature349: 438-440) , And selenoprotein P (Read et al. (1990) J. Biol. Chem. ．26517899-17905), including small numbers of eukaryotic and prokaryotic Sex proteins are a unique group of polypeptides containing unusual selenocysteine Belong to Selenoprotein production is tightly regulated by the level of exogenous selenium. Was reported. For example, Knight et al. (J. Nutr.117: 73 2, 1987) show that the activity of glutathione peroxidase is a diet lacking selenium. Was not detected in the rats fed with the subject (≦ 0.02 ppm, 0.016 mg / kg) It was reported to be reduced to a possible level. Chanoine et al. (Endocri noology131: 1787, 1992) is also a diet lacking selenium. Rats fed 6 weeks of treatment received levels of both type I and type II 5'-deiodinase. Reported to have a significant reduction (≦ 20% normal). Speier et al. J. Biol. Chem.260: 8951, 1985) is in vitro. , Glutathione peroxidase activity was greater than 1 ng / ml in medium cells. The optimum activity depends on the concentration of ren and the optimal activity is 5 ng / ml sodium selenate (2.6 × 10^-8M), but cells grown in medium not supplemented with Se It becomes deficient in glutathione peroxidase, and the activity of Se supplemented cells It proved to have 1 to 3%. Chada et al. (Blood74: 25 35, 1989) and Chu et al. (Nu cleic Acids Res.18: 1531, 1990) Of glutathione peroxidase activity between deficient cells and selenium-filled cells Reported ~ 50-fold difference.   Regulation of selenoprotein production by exogenous selenium is regulated at the UGA codon. Post-transcriptional by antitranslational uptake of selenocysteine ds Biochem. Sci.16, 463-467), and this UGA codon Unique selenocysteine-supplying tRNAs that usually contain the appropriate UCA anticodon Acts as a transcription termination codon (Hawkes et al. (1982) Biochim. Biophys. Acta699, 183-191; Lee Et al. (1989) J. Am. Biol. Chem.264, 9724-9727). like this Thus, regulation of selenoproteins is driven by the translational process at the mRNA UGA codon. It is most likely to proceed by controlling the space. In selenocysteyl-tRNA Incorporated selenium will allow the translation to be read through, but the absence of selenium Underneath, the selenocysteine tRNA remains unacylated and then U The GA codon will function to stop translation.   Since the initial identification of UGA codon usage for selenocysteine incorporation, A key issue in the interpretation of this "extended genetic code" is that ribosome translation assembly The presence of selenocysteine mRNA from the terminating UGA codon in other mRNA species. How to identify special UGA codons in the open reading frame It is.   Necessary and sufficient for signaling UGA translation as selenocysteine In order to identify all of the elements, we use human selenocysteine, glutathione The open reading frame of the gene encoding peroxidase and And 3'-untranslated region (3 ' The functional importance of sequences from both the UTR) and glutathione peroxidase Uptake of selenocysteine in both and unrelated non-selenoproteins I analyzed it.Construction of GPx and rab5b subclones   Vector pBluescript KS (Stratagene [Stratage ne]), the GPx subclone GPxR is replaced with all GPx deletion subclones. It was used as a common template to construct the lane. G in the same vector PX1 cDNA (Chu et al. (1990) Nucleic Acids Res ．18, 1531-1539). Of this clone DNA sequencing ("Sequenase" kit [US B standard dideoxy sequencing techniques using iochemicals) Previously reported codon 11 (GCC) (Mullenbach (1987) Nu cleic Acids Res.15, 5484; Chada et al. (1990) G. enomics6, 268-271), and the codon 92 we reported.   CTG (Chada et al. (1990) Genomics6, 268-271) Immediately upstream of, the CAG (Mullenbac) observed by Mullenbach et al. h (1987) Nucleic Acids Res.15, 5484) (inheritance Instead of the child bank acceptance numbers Y00369 and M21304), add one Additional GCG codons are indicated. The latter insert allows us to use other normal GPX1 sequences. The polymorphism observed in.   Unless otherwise stated, GPx deletion subclones were prepared using standard methods (Ho et al. (1989). Gene7751-59), overlap extension polymerase chain reaction (PCR), Perkin-Elmer Setsu (Perkin-Elmer) Cetus) Thermal Service It was constructed using a thermal cycler and reagents. This PCR method uses two flanking primers and two flanking primers that determine the size of the final product. And two mutually complementary primers that direct the desired mutations in the target sequence. Needed. Flanking and complementary mutagenic primers One sequence for each pair is listed in Table 1. The final PCR product was pBluescript It was inserted back into pt KS and the sequence was verified by standard methods. Then , Each mutated GPx sequence with the eukaryotic expression vector pCMV4 (Anderson et al. (1989) J. Biol. Chem.264, 8222-8229), Subsequent transfections into COS-1 cells as described below. I've roasted.   “Epitope labeling” of GPx is the open reading frame of GPx. 30n encoding the first 12 nucleotides (nt) of the ATG start codon t-sequence and subsequent 9 amino acids of human influenza hemagglutinin protein 27 bases encoding an epitope (Chada et al. (1989) Blood 74, 2535-2541) (illustrated in FIG. 1). Like). The two oligonucleotides listed in Table 1 were annealed and then Generated in GPx wild type or pBluescript KS And / or ClaI and N in the mutant subclone in pCMV4 Inserted through the heI restriction site. In this method, amino acids 2-4 of GPx Deleted and having a net increase of 6 amino acid residues over wild type GPx Labeled "GPxEPI" was generated. Rabbit antiserum against this epitope is It was possible, but its binding to the labeled GPx molecule was related to the GPx peptide sequence. Since it was much lower than that of the antisera, the latter was used for the detection of labeled GPx molecules. I used it.   GPx subclones in which the 3'UTR sequence was partially or completely deleted were It was constructed by DNA recombination technology. Briefly, pBluescript 250nt from epitope-tagged GPx subclone GPxEPI in KS Removal of the AvrII-SpeI fragment and subsequent religation of the remaining large fragment. Construct a subclone UTR-D3 in which the entire GPx3'UTR has been deleted did. Removal of the AvrII-XhoI fragment and subsequent plasmid XhoI part GPx3 from pBluescript KS containing position-excluded GPxEPI The subclone UTR-D2 was recombined by recombination of the remaining large fragments in the'UTR. Configured. removed from the construct GPxEPI in pBluescript KS Adhesive Cl The GPxEPI-containing fragment with the ends of aI and end-filled XhoI was cloned into ClaI and And inserted into the expression vector pCMV4 via SmaI polylinker restriction sites By doing so, a subclone UTR-D1 was obtained.   The overlap extension PCR method was also used to use Ras-related GTPases Mutations in the rab5b gene and fusion subclones encoding one member of (Wilson et al. (1992) J. Clin. Invest.8 9 , 996-1005). Has a 1.6 Kb rab5b cDNA clone Plasmid pMT2 was transformed into D. B. Wilson (Harvard School of Medicine (Herva rd Medical School, Boston, MA) did. Construct rab5b (opal) GPx3'UTR is a GPx3'UTR Rab5 with opal (UGA) mutation at codon 63 fused to a row It contained the fusion product of the b coding region. Flanking and mutagenesis The oligonucleotide sequences of the primers are listed in Table 1. 3'PCR Frankin Guprimer removes the native rab5b TGA stop codon, and its TAG At least 3 codons in the GPx open reading frame, including the stop codon Resulted in the substitution of Natural GPxR clone in pBluescript KS Contains the entire GPx3'UTR derived from the ClaI-AvrII double digestion of The resulting rab5b (opal) in the pBluescript KS construct ) The mutant was inserted. The gene fusion product was then transformed into pCMV as described above. 4 was subcloned into Also, using the same strategy, rab5b (WT ) GPx3'UTR configured, but in this case, flanking Normal PCR was applied using only the immers and the fusion products (WT, Chi, opal The wild type without mutation) was inserted into pCMV4. GPx3'UTR Rab5b (opal) GPx3 deleted as an AvrII-EcoRI fragment 'UTR subclone and Rab5b3' UTR sequence of approximately 900 nt NheI By removing the native rab5b3'UTR by fusing the EcoRI fragment Construct containing the opal mutation in the coding region rab5b (opal ) Was configured. The resulting rab5b (opal) sequence is then p It was inserted into CMV4.Transfection, labeling and lysis of COS-1 cells   Modified calcium phosphate mediated or electroporation method (Mani atis et al. (1990) Molecular Cloning: A Labor. atory Manual, ColdSpring Harbor Labor by Atory, Cold Spring Harbor), GPx or rab5 transfect COS-1 cells for transient expression of the b construct, then 10% fetal bovine serum, 5 ng / ml sodium selenate, 25 mM HE PES pH 7.4, and 1X penicillin-streptomycin-fundizo The cells were cultured in DMEM medium supplemented with Gibco-BRL. All The experiment was performed 2 to 4 times.   As a control for transfection efficiency, COS-1 cells were treated with Nichols -Supplied by the Nichols Institute 2 μg of plasmid pX included in the human growth hormone transient expression assay system Co-transfected with GH5. Crystal multi-detector RIA system (Yuna Ited Technologies Packard [United Technology] by using a radio-immunoassay using Then, human growth hormone secreted into the medium was detected.   ⁷⁵Has an original specific activity of 750 to 1000 Ci / g for the labeling of Se, 10 μCi as selenious acid diluted in nitric acid⁷⁵Se (Misori Research Reac University of University [University of Missouri Research rch ReactorFacility]) in each plate Added to the extruded cells, and incubate the cells at 37 ° C for an additional 3 hours. Was added.   ³⁵For the labeling of S, the transfected cells in each plate were First, 10% dialyzed calf serum, 1 x glutamine (Gibco), Methionine-free and Glutamine-free, supplemented with and 25 mM HEPES Incubated for 30 minutes in DMEM medium (Gibco). Next And has a specific activity of 1140 Ci / mmole for methionine, 250 μm Ci Express³⁵S-protein labeling mixture (NEN Dupont)) was added to the plate and the cells were incubated at 37 ° C. And incubated for 2 hours.   ⁷⁵Se or³⁵5 or 1 μl (respectively) of diisopropyl after S labeling Add fluorophosphate to the ice-cold labeling mixture in a plate of COS-1 cells. Added After 5 minutes, aspirate this mixture and 1.5 ml of COS cell lysis buffer. (50 mM HEPES pH 7.8, 1% Triton X-100, 10 mM E DTA, 1 mM phenylmethylsulfonyl chloride) was added to each plate. 4 After shaking for 20 minutes at 0 ° C, the lysed cell suspension was added to a microfuge (mi cell fuge tube and centrifuge at 14,000 xg for 10 minutes to disrupt cells. The piece was removed. Sodium dodecyl sulfate (SDS) was added to the supernatant to a final concentration of 0.5%. Add to concentration and boil for 5 minutes Heat in boiling water, then cool on ice.Immunoprecipitation and protein electrophoresis   A synthetic peptide sequence from the GPx polypeptide chain, one from residues 26-46, And two rabbit antibodies raised against the other from residues 174-192. Serum (Berkeley Antibody Company [Berkeley Antib ody Co. ] Richmond, CA) was used in immunoprecipitation . 15 μl of each antiserum, +20 μl of protein A-Sepharose (Sepha rose) CL-4B beads (Sigma [Sigma]) were added to each plate, The mixture was then incubated overnight at 4 ° C with constant tumbling. I got it. The beads are subsequently allowed to settle and washed buffer (50 mM HEPES p H7.8, 150 mM NaCl, 1% Triton X-100, 0.5% deo Xycolate, 0.1% SDS), washed twice, and 50 mM HEPES   Washed once with pH 7.8, 30 μl SDS-gel loading buffer (50 mM T ris-HCl, pH 6.8, 100 mM dithiothreitol, 2% SDS , 0.1% bromophenol blue, 10% glycerol) and boil Heat in water for 3 minutes, then settle in a microfuge. Then the supernatant Sample solution for SDS-polyacrylamide gel electrophoresis (SDS-PAGE) Collected in.   For COS-1 cells transfected with the rab5b construct, The order was the same as above, except that 0.2% SDS was used for cell lysis. And was raised against a synthetic peptide from the hypervariable domain of rab5b , 8 μl of affinity purified rabbit antibody (obtained from DB Wilson ) Was added.   Electrophoresis of proteins is performed using standard techniques (Maniatis et al. (1990) Mol. Equal Cloning: A Laboratory Manual, C old Spring Harbor Laboratory, Cold Spring Harvard) on a 12% SDS-polyacrylamide gel. .RNase protection assay   RNA of whole cells was analyzed by the guanidine-HCl method (Ginsburg et al. (1985) ) Science228, 1401-1406). T7 Promo RNA transcription kit (Stratagene) Starting at the ClaI site of the 5'-untranslated region of the GPxEPI transcript. 224 nt complementary to the 179 nt segment³²P-labeled RNA transcription Riboprobes were generated by synthesizing the body. The template is GPxEPI Of the construct formed by recycling of the large fragment of end-filled SpeI-RsrII It was a ClaI fragment. 3 μg total cellular RNA, 10 μg yeast tRNA, And 6 μl of riboprobe (400,000 TCA-precipitating cpm / μl) The RNase protection assay of the hybridization mixture was performed using standard techniques (Ma niatis et al. (1990) Molecular Cloning: A Lab. Oratory Manual, Cold Spring Harbor La laboratory, Cold Spring Harbor).Role of the nucleotide sequence around the UGA codon   We first determined the number in the open reading frame of GPx mRNA. The possibility that the cleotide sequence may serve as a signal for the insertion of selenocysteine Explored. Sequence analysis of GPx MRNA has been performed on both prokaryotic and eukaryotic cells. Together with nocysteine mRNA Did not reveal conserved sequences that are common, but UGA₁₄₂2 around codon One possible loop structure was predicted. (UGA₁₄₂Is the GPx cDNA sequence Call the starting codon in Leotide 142; all nucleotides for GPx As shown in Figure 1A, the number of Ochid is the first number of the open reading frame. Start at the base). UGA₁₄₂One putative stem just downstream of The structure of E. coli formate dehydrogenase mRNA and E. coli And stems similar to those found in related prokaryotic selenoenzyme genes- Create a loop structure (shown in Figure 1, panel B) (Zinoni et al. (1990) Proc. Natl. Acad. Sci. USA87, 4660-4664) . UGA at the tip of the "hairpin"₁₄₂Others that incorporate codons (shown in Figure 1C) ) Of several mammalian GPx mRNAs, as well as E. coli. ) Conserved between formate dehydrogenase mRNA sequences (Chada et al. 1988) Oxy-Radicalsin Molecular Biological y and Pathology (Cerutti, P., Fridovich , I. , And McCord, J. et al. , Ed.) Pp. 273-288, AlanR. , Liss Inc. ,New York). In the direction of uptake of selenocysteine To test the role of each of these potential secondary structures in construct a series of 5 sequential deletions in the open reading frame of x, ORF-D1 to ORF-D5 shown in FIG. 1 were designed.   The first four deletion subclones are UGA₁₄₂Putative immediately downstream of the codon Located within the stem-loop region. ORF-D1 is from codon 49 to codon 53 ORF-D2 lacks codons 65-69; ORF-D3 Lacks codons 54-63 And ORF-D4 lacks codons 71-74. ORF-D1, OR The sequences deleted from F-D2 and ORF-D3 are 5'of the stem, respectively. Part, the 3'portion of the stem, and the open reading frame of Gpx mRNA. Putative stem-loop structure in the genome region (Zinoni et al. (1990) P roc. Natl. Acad. Sci. USA87, 4660-4664) It corresponds to most of the loop (29 of 31 nt). ORF-D4 is E. coli For translation of selenocysteine in (E. coli) formate dehydrogenase Of the putative stem-loop structure, which corresponds to sequences reported to be important for Represents the 12 nt sequence immediately downstream (Zinoni et al. (1990) Proc. Na tl. Acad. Sci. USA87, 4660-4664). GPx Sub Lawn ORF-D5 forms part of the stem of another putative hairpin loop structure. UGA of GPx mRNA₁₄₂Codon 4 located immediately upstream of the codon 7 deletions (Chada et al. (1988) Oxy-Radicals i. n Molecular Biology and Pathology (Cer utti, p. , Fridovich, I .; , And McCord, J. et al. , Ed) pp. 273-288, Alan R. et al. , Liss Inc. ,New York) . These deletion subclones are supported by the eukaryotic expression vector pCMV4 , Individually transfected into COS-1 cells and expression of GPX To⁷⁵Se labeling, immunoprecipitation, SDS-polyacrylamide gel electrophoresis, and And autoradiography.   As shown in FIG. 2, panel A, the vector alone (lane 1) was used for transfer. Transfected COS-1 cells have a large size of 23 kD similar to that of human GPx. Have a texture⁷⁵Se-containing polypeptide (heaven Demonstrating a low background level (most similar to naive monkey cells GPx). Transient development of native human GPx cDNA and deleted ORF-D1 to ORF-D4 All (currently lanes 2 and 3-6, respectively) are of the GPx protein. High level of inside⁷⁵Shows the incorporation of Se. These deletions are few but real. It appears to show a reduction in GPx expression, which is not qualitative. Repeated experiments (epi Including the occurrence of deletions in the identification of the tope-labeled constructs) was also slightly reduced. Showed the present. Similarly, as shown in Figure 2B, the deletion ORF-D5 was introduced into GPx. Produces or does not produce a reduction in the insertion of selenocysteine. This Thus, a putative loop in the open reading frame of Gpx mRNA. Structure slightly modulates the expression of GPx, both of which are selenotypes in human GPx. UGA as cysteine₁₄₂It is not absolutely necessary for codon translation.Role of 3'UTR   To test the role of the 3'UTR in the insertion of selenocysteine, we have It contains GPx subclones containing deletions of various lengths in that region as described above. Configured Transiently expressed human G from COS-1 background In order to improve the resolution of Px products, epitope labeling (Chada et al. (19 89) Blood74, 2535-2541) among these subclones Incorporated. As illustrated in FIG. 1A, we have human influenza hemagglutination. Elementary protein (Chada et al. (1989) Blood74, 2535-254 30n encoding the ATG start codon of 1) and an epitope of 9 amino acids The sequence of t was replaced with the first 4 codons. Transiently expressed, epitope-tagged G Clear identification of Px was possible. Because the labeled GPx is Sufficient to separate at a position detectably higher than the band of labeled GPx. Because it moved completely. This difference in mobility is due to the coding region deletion structure described above. Transfection without the substantial overexpression required for adult assessment It was possible to evaluate the transient expression of the cloned constructs. In addition, The row was inserted into the wild-type GPx subclone GPxR to create a new GPx subclone. Of the transient expression of the GPx3'UTR deletion construct. Served as a positive control for. These deletions are also shown in Figure 1A. It   Into GPx in transfected COS-1 cells [⁷⁵Se ] -Effect of three large deletions of the 3'UTR on selenocysteine uptake Is shown in FIG. Lane 1 transfected with vector alone Demonstrate the background GPx signal in the treated cells. Slightly The larger epitope-tagged GPx has a GPxEPI whose 3'UTR is intact. Background expressed by the construct (lane 2) and of endogenous COS-1 Easily distinguished from de. Distal 100 nt 3'UTR deletion (UTR-D1, Lane 3) did not reduce the expression of transfected GPx. Shi However, the proximal 129 nt (construct UTR-D2, lane 4) or the entire 3'UTR (construct UTR-D3, lane 5) is detectable into GPx⁷⁵S The incorporation of e was completely eliminated. Distal and global 3'UTR deletions are 3'UT It did not yield Gpx mRNA without R. Because of human growth hormone The 3'UTR sequence of the gene was cloned into pCMV4 vector for fusion to the inserted sequence. Because it is built into the target (as in other GPx constructs, transcription termination If not separated by site).   University of Wisconsin Computer Group Software (University rity of Wisconsin Computer, Group so whole Gpx mRNA using the FOLD program of Computer analysis of its 3'UTR sequence (Devereux et al. (1984) N ucleic Acids Res.12387-395) is a rat and Found in human 5'deiodinase and 3'UTR of rat GPx gene Things (Berry et al. (1991) Nature353273-276) Reveals a potential secondary structure consisting of a long stem with two small loops that are similar I made it Moreover, two 4-nt sequences in the loop (UAAA in the first And UGAU in No. 2; shown in FIG. 4) is a report of the 5'deiodinase gene. The motif of the reported “selenocysteine insertion sequence” (Berry et al. (1991) Nature353273-276) at the same position. It was.   These two short sequences are necessary for the insertion of selenocysteine into GPx. We specifically excluded each of these sequences in order to test if they were important Two small deletion mutants, UTR-D4 and UTR-D5, were constructed. The results shown in FIG. 5 show that both of these short deletions (lanes 3 and 4) were transgenic. Detection into epitope-tagged GPx in transfected COS-1 cells Demonstrate complete destruction of available selenocysteine uptake.   To rule out the possibility that deletion mutants affect the level of GPx transcripts In addition, we found that Gpx m in transfected COS-1 cells RNA levels were measured by RNase protection assay. As shown in FIG. , Untransfected COS-1 cells (lane 2) are specific for epitope-tagged GPx transcripts Contains no detectable mRNA by simple riboprobe and epitope tag COS-1 cells transfected with modified wild-type GPx (lane 3) are U Those transfected with the TR-D4 and UTR-D5 deletions (rays 4 and 5). Also, open reading frames Deletions (ORF-D1, -D2, and -D3) in the genome It produced no detectable change in levels (data not shown). Trang Sfection efficiency depends on co-transfection with vectors encoding human growth hormone. Assayed by transfection and grouped in these experiments. Similar to (data not shown).GPx3'UTR is sufficient for insertion of selenocysteine   Sequences in the 3'UTR of GPx mRNA are responsible for translational insertion of selenocysteine. We have now proved that this 3'UTR is irrelevant Is sufficient to direct the same process to the UGA codon in the coding sequence of I studied whether or not. The selected target gene rab5b is Ras-related GTP Encodes a 25 kD GTP-binding protein that is a member of the ASE superfamily ( Wilson et al. (1992) J. Am. Clin. Invest.89, 996-10 05). This gene was used for three constructs:rab5b (opal) Changed to a UGA (opal) mutant with the native rab5b3'UTR Had the codon 63 UGU (cysteine) assigned;lab5b (opal) Gp x3'UTR Replaces its stop codon (UAG), and the entire GPx3'UTR GPx cDNA including the last 3 codons of the GPx coding region, including 5b (opal) outfit fused to the 3'part Ing sequence; andlab5b (wt) Gpx3'UTRHa It was also a rab5b-GPx fusion product, but not the opal mutation. It had a raw codon 63. The fusion construct can be UGU (cysteine) or UGA ( The potential selenocysteine) codons and GPx3 in the native GPx transcript. Placed the same number of nt upstream from the'UTR.   FIG. 7: Representative transient expression of these constructs in COS-1 cells. Represents the results of the test. The expression of rab5b is³⁵S (lanes 1-4) or⁷⁵Se (re After labeling with a radioisotope of sequence 5-8) It was detected by a nity-purified rabbit antibody. Transfection with vector alone COS-1 cells (lanes 1 and 5) were suitable for rab5b Showed no detectable immunoreactive protein at a unique 25 kD molecular mass . All three constructs are detectable, but at widely different levels 25kD³⁵Directs the synthesis of S-tagged polypeptides, but does Le) GPx3'UTR, ie fusion production of rab5b with opal mutation. Only things⁷⁵It was bound to a GPx3'UTR incorporating Se (lane 6). ra b5b (opal) transfectants had very low levels³⁵S-labeled tag Another opal nonsense suppression mechanism that expresses protein and possibly in COS-1 cells (Hatfield, D. (1985) Trends Bio). chem. Sci.10, 201-204).⁷⁵Se even after a very long exposure It is not detectable and is not detected at the UGA codon in the presence of the rab5b3'UTR. It indicates that no cysteine insertion occurred. The truncated polypeptide is³⁵S sign Undetectable on derivatized immunoprecipitates, the product of short polypeptides is unstable Or not immunoreactive with antiserum I suggested that. rab5b (wt) GPx3'UTR fusion construct Transfection is expected for the wild type rab5b coding region As detectable⁷⁵Expression of immunoreactive proteins without Se uptake occured. For these experiments, a vector encoding human growth hormone Transfection and measurement of secreted growth hormone Sfection efficiency was once again confirmed.   The importance of the distance between the UGA codon and the selenocysteine insert sequence remains unknown Is. Our target for induced selenocysteine uptake Rab5b has a similar number of amino acids to GPx and is mutated in UGA. Is UGA in GPx₁₄₂And the distance from the 3'UTR (550nt) is similar It However, in the 5'deiodinase, the span is approximately 1200 nt. The precise sequence between these elements required for selenocysteine uptake The distance is probably not critical.   These data demonstrate that the human GPx gene has a small 3'UTR segment. And conserved AAA and UG within the potential stem-loop structure. Although the AU sequence is essential for translation of selenocysteine in human GPx, , A potential stem-loop or hairpin structure in the coding region is essential There is no. Moreover, these data also prove, as evidenced, by the GPx3'UT R alone is an open reading of an unrelated non-selenocysteine, rab5b Opal mutation (UGA) in frame as selenocysteine, Sufficient to signal the translation.Synthesis of "optimized" selenocysteine inserts   Testing of the genes encoding various known mammalian selenocysteines has been shown to be 3 'UTR has little primary sequence similarity and Have a potential stem-loop structure (Hill et al., J. Biol. ． Chem.266: 10050, 1991; Zinoni et al., Proc. Na tl. Acad. Sci.87: 4660, 1990; Ho et al., Nucleic.   Acids Res.16: 5207, 1988; Berry et al., Nature. e353: 273, 1991). Of homology between the 3'UTRs of these genes The lack is due to the extension of the two 3-4 nucleotides of the putative stem-loop structure. Human glutathione proved essential for uptake of nocysteine Combined with our analysis of the peroxidase 3'UTR, Synthetic nucleotides containing and capable of forming orientation-independent stem-loop structures Allowed the design of the array. Is this "optimized" synthetic sequence a stem-loop base? 12 nucleotides at the top of the "bubble" and subsequent structure to 17 nucleotides Contains an additional 11 nucleotide stem with a loop of leotide or balloon Have. In order to make the “optimized” stem loops non-specific, 3-4 A mirror image of the targeting element of cleotide, a suitable bubble of artificial stem-loop And position over the balloon area. The structure of this optimization element is shown in FIG. . In this figure, MRS is the insertion of the element into the appropriate cloning vector. Means multiple restriction sites to facilitate and N represents any nucleotide. S; N_xMeans an extension of 2 or more nucleotides of any sequence; N : N means complementary base pairs.   Nucleotide sequences containing this optimized stem-loop have been used in molecular biology Can be constructed by standard techniques known to those skilled in the art. For example, we Applied Biosystems DN Use A synthesizer A 92-mer oligonucleotide containing a bubble-balloon was synthesized. Gel spirit After manufacture, this single-stranded oligonucleotide has 12 and / or 6 nucleotides. 22-mer sense and antisense PCR primers containing The Limer was used to serve as a template for PCR amplification:   1) Template oligonucleotide:   2) PCR primers: The PCR reaction was carried out according to standard methods with 0.1 μg / μl template oligonucleotide, 50 pmole / μl of each PCR primer for 1 min at 95 ° C 10 cycles at 50 ° C for 1 minute and 70 ° C for 1 minute did.   The double-stranded PCR product was then added to the pCRII vector (Invitrogen [InV (ITROGEN], San Diego, CA). (InVitrogen) and combined with INV Transformed into αF 'cells. The sequence of the construct is Promega Using fmol PCR sequencing from (Madison, WI) Confirmed by sequencing the cleotide.Construction of recombinant selenocysteine containing polypeptide   Any desired polypeptide whose DNA sequence is known may be used in the method of the invention. And encode any amino acid that is not essential for the natural activity of the polypeptide. Can be used by codon substitution It Approaches to the production of these “TGA” mutants generally involve site By specific mutagenesis or oligonucleotide-based mutagenesis techniques , For example using a commercially available kit (Promega) Can be achieved. For example, human thyroid hormone receptor-β₁To The cDNA encoding the vector-p-alter (Promega [ Promega]) multiple cloning sites and Cysteine codons to TGA by oligonucleotide-based mutagenesis Mutated The cDNA is then rescued by standard laboratory procedures, And the mutation was confirmed by nucleotide sequencing.Expression of selenopolypeptide   Polypeptides according to the present invention can be used to select selenocysts using any suitable expression system. Linked to a recombinant nucleic acid containing a stem-loop structure required for translation of the in Expression from a recombinant nucleic acid having a sequence encoding the encoded polypeptide , Suitable expression vehicles, such as those described above. Transformation of a suitable eukaryotic host cell with the recombinant nucleic acid therein. Those skilled in molecular biology As will be appreciated, any of a wide variety of expression systems can be used to A selenocysteine containing protein can be provided. The exact inn to use The host cells are not critical to the present invention, and Saccharomyces cerevisiae (Saccharomyces   cerevisiae) Or mammalian cells ( For example, COS-1, HL-60, CV-1, LLC / PK-1, C-6, 3T 3L1 and CHO cells). Such cells have a wide range of sources (eg For example, American Type Culture Collection [American Ty pe Culture Collect Ion], Rockville, Maryland). Transformation or The method of transfection and the choice of expression vehicle depends on the polypeptide to be expressed. It will depend on the nature of the tide and the host system chosen. Transformation and trans For example, the method of the infection is described in Ausebel et al. (Current Prot. ocols in Molecular Biology, John Wiel y & Sons, New York, 1989); expression vehicles. Are well known in the art, such as Clonig Vector rs: A Laboratory Manual (PH Pouwels et al., 1985, Suppl. 1987).   For example, a cDNA encoding a desired polypeptide allows for expression Especially as a parental vector for selenocysteine expression systems in a designed orientation Preferred eukaryotic expression vectors pcDNA1 / neo and pRC / CMV (I In Vitrogen).   Alternatively, the selenocysteine containing polypeptide according to the invention is stable It can be produced by a transfected mammalian cell line. Feeding A number of vectors suitable for stable transfection of animal cells are commonly available Possible references, eg Pouwels et al., Supra; constructing such cell lines. Methods for doing so are also commonly available, eg, Ausebel et al., Supra.   Once the desired selenopolypeptide is stably transfected into the host system. Production, the production of the polypeptide should be regulated by the content of selenium in the medium. Can be. Cell culture systems lacking selenium have been described (Spairer. Et al., Supra; Chada et al., Supra). Normally, the production of selenocysteine protein is 0.1 ng / Inhibited at concentrations below ml medium and smells above 1 ng / ml Will be triggered. Optimal induction of selenopolypeptide production is approximately 5-25 occurs in ng / ml medium and concentrations above 50-100 ng / ml are It is cytotoxic, depending on the cell type.   Once the recombinant polypeptide is expressed, it is well known in the art. Can be isolated according to known methods, and the functional activity can be Assay suitable for tide, eg determined by enzymatic activity or affinity of binding can do. The desired selenopolypeptides with the same functional activity When expressed in cells containing the protein, the selenopolypeptide is described To have higher reactivity with nucleophiles for selenocysteine as described (Leonard et al., Biochim. Biophys. Acta.7 87 : 122, 1984), or as described herein,⁷⁵Use Se It is distinguished from the native protein by its radiolabelling.Expression of selenopolypeptides in vivo   Methods described herein to encode amino acid residues in selenocysteine A gene for any desired polypeptide modified according to Transgenic animals whose production is regulated by the selenium content of the animal's diet Can be used to produce. Method for producing transgenic animals Are well known (see, eg, Hogan et al., Manipulating).   the Mouse Embryo: A Laboratory manual 1, CSH Press, Cold Spring Harbor, New York, 1 986; Leder et al., U.S. Pat. No. 4,736,866). Typically, 0. Less than 016mg / kg When a diet containing dietary selenium is given to a transgenic animal, Expression of selenopolypeptide is blocked, but 0.1 mg / kg or more Selenium above (eg Na₂SeO₃, Sigma [Sigma]) containing animals When given, the protein will be produced at high levels.Other aspects   The method of the present invention also enhances any commercially available selenopolypeptide. Can be used to produce on a level. As discussed above, a valid selection The presence of selenium exceeds the levels produced under conditions of selenium selenopolypep It produces a 30-50 fold increase in the expression of tide. This level is, for example, selenium An expression vehicle that, when present, allows overexpression of tRNA under suitable conditions Cells cotransfected with a gene encoding selenocysteine tRNA It can be further increased by performing a function. For example, this The gene encoding the tRNA is placed under the control of an inducible promoter, then At the same time as or before supplying selenium to the medium, it is required for gene induction. Can be achieved by supplying a factor that

[Procedure amendment] [Submission date] December 14, 1995 [Amendment content] Claims 1. A method of controlling the production of a heterologous polypeptide in a eukaryotic cell, wherein a first nucleic acid encoding the heterologous polypeptide, wherein at least one codon of the mRNA transcribed from said first nucleic acid is replaced with codon UGA A second nucleic acid operably linked to the first sequence, the second nucleic acid is a translation of the UGA codon as selenocysteine only when selenium can be obtained from the medium in which the cells are grown. Is prepared, and the cells are grown under conditions in which the production of the polypeptide is controlled by the level of selenium available to the cell. How to be. 2. The method of claim 1, wherein the cells are grown in vitro. 3. The method of claim 2 wherein said cells are mammalian cells. 4. The method of claim 1, wherein the cells are yeast cells. 5. The method of claim 1 wherein said first and second nucleic acids are maintained in said cell in a recombinant vector capable of autonomously replicating in said cell. 6. The method of claim 1 wherein said first and second nucleic acids are stably integrated into the genome of said cell. 7. The cells are embryonic cells derived from a non-human mammal, and the method comprises the transgenic offspring of the non-human mammal containing the first nucleic acid and the second nucleic acid stably integrated into the genome. The method of claim 1 further comprising the step of obtaining. 8. 2. The method of claim 1 wherein said second nucleic acid is derived from approximately 90 contiguous nucleotides from the 3'-untranslated region of the naturally occurring mammalian selenoprotein-encoding gene. 9. 9. The method of claim 8 wherein the mammalian selenoprotein is human glutathione peroxidase and the second nucleic acid comprises a nucleotide sequence substantially homologous to the nucleotides: TGTCTCGG GGGGGTTTTC ATCTATGAGG GTG TTTCCTC TAAACCTACG AGGGAGGAAC ACCTGATCTT ACAGAAAATA CCACCTCGA. Ten. The method of claim 1 wherein said polypeptide is produced by said cells when the concentration of selenium is 1-25 ng / ml growth medium. 11. 8. The method of claim 7, wherein said polypeptide is produced in said transgenic progeny when the daily intake of said transgenic progeny contains more than 0.1 mg / kg selenium. 12. 2. The method of claim 1, wherein the cells are free of native proteins that are substantially similar to the recombinant polypeptide. 13. The cell contains a native protein that is substantially identical to the recombinant polypeptide, and the recombinant polypeptide is distinguished from the native protein by an increase in the reactivity of the recombinant polypeptide to a nucleophile. The present invention according to claim 1. 14. The ability of the recombinant polypeptide to contain the native protein substantially identical to the recombinant polypeptide, and wherein the recombinant polypeptide incorporates the radioactive isotope ⁷⁵ Se, rather than the native protein. The method of claim 1 wherein the method is distinguished from the native protein by. 15. A method of producing a radiolabeled recombinant polypeptide, comprising preparing a first nucleic acid encoding said heterologous polypeptide, wherein at least one codon of mRNA transcribed from said first nucleic acid is replaced with codon UGA. A second nucleic acid operably linked to the first sequence, the second nucleic acid comprising a contiguous sequence of nucleotides capable of forming a stem-loop secondary structure in the mRNA transcribed therefrom. The stem-loop can direct translation of the UGA codon as selenocysteine only when selenium can be obtained from the medium in which the cell is grown, and the first nucleic acid and the second nucleic acid are present in the cell. And is radioactive under conditions sufficient to allow incorporation of the isotope into the recombinant polypeptide translated from the first and second nucleic acids. Growing presence in said cells of Len isotope, a method which consists in. 16． An isolated single-stranded nucleic acid capable of forming a stem-loop secondary structure and consisting of a nucleotide sequence substantially identical to the nucleotides: TG TCTCGG GGGGGTTTTC ATCTATGAGG GTGTTTCCTC TAAACCTACG AGGGAGGAAC ACCTGATCTT ACAGAAAATA CCACCTCGA. 17. A double-stranded nucleic acid comprising DNA encoding the single-stranded nucleic acid according to claim 16. [Figure 8] [Figure 8]

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁶ Identification code Internal reference number FI C12N 9/08 9452-4B C12P 21/02 C C12P 21/02 9281-4B C12N 5/00 B // (C12N 9/08 C12R 1:91) (C12P 21/02 C12R 1: 865) (C12P 21/02 C12R 1:91)

Claims (1)

[Claims] 1. A first nucleic acid encoding a heterologous polypeptide, wherein at least one codon of the mRNA transcribed from said first nucleic acid has been replaced by the codon UGA, and a second nucleic acid operably linked to said first sequence , The second nucleic acid is capable of directing translation of the UGA codon as selenocysteine, and the production of the polypeptide is controlled by the level of selenium available to the cell. Proliferating said cells under certain conditions comprising controlling the production of the heterologous polypeptide in a eukaryotic cell. 2. The method of claim 1, wherein the cells are grown in vitro. 3. The method of claim 2 wherein said cells are mammalian cells. 4. The method of claim 1, wherein the cells are yeast cells. 5. The method of claim 1 wherein said first and second nucleic acids are maintained in said cell in a recombinant vector capable of autonomously replicating in said cell. 6. The method of claim 1 wherein said first and second nucleic acids are stably integrated into the genome of said cell. 7. The cells are embryonic cells derived from a non-human mammal, and the method comprises the transgenic offspring of the non-human mammal containing the first nucleic acid and the second nucleic acid stably integrated into the genome. The method of claim 1 further comprising the step of obtaining. 8. The method of claim 1 wherein said second nucleic acid comprises a flanking sequence of nucleotides capable of forming a stem-loop secondary structure in the mRNA transcribed therefrom. 9. 9. The method of claim 8 wherein said second nucleic acid is derived from approximately 90 contiguous nucleotides from the 3'-untranslated region region of the naturally-occurring mammalian selenoprotein-encoding gene. Ten. 10. The method of claim 9 wherein said mammalian selenoprotein is human glutathione peroxidase and said second nucleic acid comprises a nucleotide sequence substantially homologous to nucleotides 654-740 of FIG. 11. The second nucleic acid is synthetically derived, and the stem-loop loop comprises the sequence 5'-NAAAUNNUAAAN-3 ', and the stem-loop stem comprises at least 12 non-complementary nucleotides forming a bubble. The bubble comprises a symmetrically opposed sequence 5'-NUAGUN-3 'for each half thereof, wherein each N is independently selected from the group consisting of adenine, guanine, cytosine and uracil. Method of range 8. 12. The method of claim 2 wherein said polypeptide is produced by said cells when the concentration of selenium is 1-25 ng / ml growth medium. 13. 8. The method of claim 7, wherein said polypeptide is produced in said transgenic progeny when the daily intake of said transgenic progeny contains more than 0.1 mg / kg selenium. 14. 2. The method of claim 1, wherein the cells are free of native proteins that are substantially similar to the recombinant polypeptide. 15. The cell contains a native protein that is substantially similar to the recombinant polypeptide, and the recombinant polypeptide is distinguished from the native protein by an increase in the reactivity of the recombinant polypeptide to a nucleophile. The method according to claim 1. 16． The ability of said recombinant polypeptide, rather than said native protein, wherein said cell contains a native protein substantially similar to said recombinant polypeptide, and said recombinant polypeptide incorporates the radioactive isotope ⁷⁵ Se. The method of claim 1 wherein the method is distinguished from the native protein by. 17. A first nucleic acid encoding the heterologous polypeptide is provided, wherein at least one codon of the mRNA transcribed from the first nucleic acid is replaced with the codon UGA and is operably linked to the first sequence. A second nucleic acid is provided, said second nucleic acid comprising a contiguous sequence of nucleotides capable of forming a stem-loop secondary structure in the mRNA transcribed therefrom, said stem-loop comprising as a selenocysteine of said UGA codon. that can direct translation, the first nucleic acid and second nucleic acid is introduced into a cell, in order to enable ⁷⁵ Se incorporation into the polypeptide of the cells in the presence of ⁷⁵ Se Growing under sufficient conditions, which comprises producing a radiolabeled recombinant polypeptide. 18. A single-stranded nucleic acid comprising a contiguous stretch of nucleotides capable of forming a stem-loop secondary structure, the loop of the stem-loop comprising the sequence 5'-NAAA UNNUAAAN-3 ', and the stem-loop Stem contains approximately 14 non-complementary nucleotides forming a bubble, said bubble containing a symmetrically opposed sequence 5'-NUAGUN-3 'for each half thereof, each N being adenine, guanine, Independently selected from the group consisting of cytosine and uracil, and said nucleic acid is capable of directing the translation of the codon, UGA, as selenocysteine when it is operably linked to the mRNA containing said codon, Single-stranded nucleic acid. 19. 19. The nucleic acid of claim 18, wherein the stem-loop contains approximately 90 nucleotides, the bubble is located approximately 17 nucleotides from the base of the stem, and the loop is located approximately 11 nucleotides from the bubble. 20. 19. The nucleic acid of claim 18, wherein each half of said bubble is approximately 7 nucleotides. twenty one. 19. The nucleic acid of claim 18, wherein said loop contains approximately 12 nucleotides. twenty two. A double-stranded nucleic acid comprising DNA encoding the single-stranded nucleic acid according to claim 18.