当前位置: 首页 > 期刊 > 《核酸研究》 > 2004年第21期 > 正文
编号:11370083
Complete set of orthogonal 21st aminoacyl-tRNA synthetase-amber, ochre
http://www.100md.com 《核酸研究医学期刊》
     Department of Biology, Room 68-671, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

    * To whom correspondence should be addressed. Tel: +1 617 253 4702; Fax: +1 617 252 1556; Email: bhandary@mit.edu

    ABSTRACT

    We describe the generation of a complete set of orthogonal 21st synthetase-amber, ochre and opal suppressor tRNA pairs including the first report of a 21st synthetase-ochre suppressor tRNA pair. We show that amber, ochre and opal suppressor tRNAs, derived from Escherichia coli glutamine tRNA, suppress UAG, UAA and UGA termination codons, respectively, in a reporter mRNA in mammalian cells. Activity of each suppressor tRNA is dependent upon the expression of E.coli glutaminyl-tRNA synthetase, indicating that none of the suppressor tRNAs are aminoacylated by any of the twenty aminoacyl-tRNA synthetases in the mammalian cytoplasm. Amber, ochre and opal suppressor tRNAs with a wide range of activities in suppression (increases of up to 36, 156 and 200-fold, respectively) have been generated by introducing further mutations into the suppressor tRNA genes. The most active suppressor tRNAs have been used in combination to concomitantly suppress two or three termination codons in an mRNA. We discuss the potential use of these 21st synthetase-suppressor tRNA pairs for the site-specific incorporation of two or, possibly, even three different unnatural amino acids into proteins and for the regulated suppression of amber, ochre and opal termination codons in mammalian cells.

    INTRODUCTION

    Significant progress has been made recently in the synthesis of proteins in vivo carrying an unnatural amino acid at a specific site. The most common strategy involves the readthrough of an amber (UAG) stop codon by an amber suppressor tRNA aminoacylated with the desired unnatural amino acid (1). In one approach, the suppressor tRNA is aminoacylated by a mutant aminoacyl-tRNA synthetase (aaRS), which aminoacylates the suppressor tRNA with the desired unnatural amino acid instead of a normal amino acid. This approach has several key requirements: (i) an orthogonal suppressor tRNA that is not aminoacylated by any of the endogenous aaRSs in the cell, (ii) an orthogonal aaRS that aminoacylates only the suppressor tRNA but no other tRNA in the cell, and (iii) a mutant of the orthogonal aaRS that aminoacylates the suppressor tRNA with the unnatural amino acid but not with a normal amino acid. Another requirement is transport of the desired unnatural amino acid from the medium into the cell. Because most cells contain twenty aaRSs, the orthogonal synthetase-suppressor tRNA pairs are often called 21st synthetase-tRNA pairs. Using this approach, Schultz, Yokoyama and their co-workers have described the site-specific insertion of several unnatural amino acids into proteins in vivo (2–4).

    The above approach requires the isolation of aaRS mutants, one at a time, for each specific unnatural amino acid. Furthermore, the unnatural amino acid has to be closely related in size and structure to the natural amino acid. An alternative approach, that does not require a mutant aaRS, involves the import (by injection, transfection or electroporation) into cells of a suppressor tRNA already aminoacylated with the unnatural amino acid (5–8). This approach is quite flexible in that the same suppressor tRNA can be chemically aminoacylated with virtually any unnatural amino acid. The only requirement, also common to the approach mentioned above, is that the suppressor tRNA used should not be aminoacylated by any of the endogenous aaRSs. Previously, we described the import of purified suppressor tRNAs into mammalian cells by means of transfection and the identification of an amber suppressor tRNA (supF), derived from Escherichia coli tyrosine tRNA (tRNA1Tyr), that fulfilled the above requirements for site-specific insertion of unnatural amino acids (6).

    Essentially, all of the work on unnatural amino acid mutagenesis has involved the use of an amber suppressor tRNA along with an amber stop codon at the site of interest in the protein gene. The availability of other suppressor tRNA/stop codon pairs would greatly add to the versatility of unnatural amino acid mutagenesis and allow site-specific insertion of two or more different unnatural amino acids into proteins. Recently, we showed that an ochre (UAA) suppressor tRNA, derived from E.coli tRNA1Tyr, was also not aminoacylated by any of the mammalian cytoplasmic aaRSs and that import of aminoacylated ochre suppressor tRNA led to specific suppression of an ochre codon in a firefly luciferase reporter gene (9). We further showed that import of a mixture of aminoacylated ochre and amber suppressor tRNAs led to the concomitant suppression of an ochre and an amber codon in a mRNA, suggesting that this approach could be used for the synthesis of proteins carrying two different unnatural amino acids in mammalian cells. Along these lines, Schultz and co-workers have very recently described a frameshift suppressor tRNA for use in E.coli (10) and an opal suppressor tRNA for use in mammalian cells (11).

    With the objective of expanding the nature and number of unnatural amino acids that can be introduced into proteins, we have been interested in generating new orthogonal 21st synthetase-suppressor tRNA pairs that also include ochre and opal suppressor tRNAs. Furthermore, while highly active suppressor tRNAs are useful for increased yield in synthesis of proteins carrying unnatural amino acids, such suppressor tRNAs are often toxic to cells and are actively selected against (12,13). It is, therefore, important to also develop orthogonal suppressor tRNAs, that have a range of activities in suppression of the corresponding codons. Here, we report on a complete set of orthogonal 21st synthetase-amber, ochre and opal suppressor tRNA pairs derived from E.coli glutamine tRNA (tRNAGln; Figure 1), with each suppressor tRNA having a wide range of activity in suppression (36-fold for amber, 156-fold for ochre, and 200-fold for opal suppressor tRNA). Activity of each suppressor tRNA is dependent upon expression of E.coli glutaminyl-tRNA synthetase (GlnRS). We have used the suppressor tRNAs in combination to suppress two different stop codons (UAG and UAA; UAG and UGA) and even all three stop codons in the same mRNA. Therefore, following the isolation of a mutant of E.coli GlnRS, which aminoacylates the amber, ochre and opal suppressor tRNAs with a glutamine analog, these new 21st synthetase-tRNA pairs can be used along with the 21st synthetase-amber suppressor tRNA pair and/or the 21st synthetase-opal suppressor tRNA pair developed by others (3,11) for the synthesis of proteins carrying two or, possibly, even three different unnatural amino acids in mammalian cells. Additionally, because each suppressor tRNA is specific for its cognate codon, the suppressor tRNAs described here can be aminoacylated in vitro with different unnatural amino acids and imported into mammalian cells by means of transfection.

    Figure 1. Cloverleaf structures of suppressor tRNAs derived from E.coli tRNAGln. The mutated anticodon sequences and the C9 to A9 mutation are circled.

    MATERIALS AND METHODS

    Plasmids

    The dual-luciferase reporter system coding for the Renilla luciferase (Renilla reniformis; RLuc) and firefly luciferase (Photinus pyralis; FLuc) fusion protein has been described previously (9,14). The DNA sequences encoding Renilla and firefly luciferase were fused to express a single protein with two bioluminescent activities (Figure 2). Plasmid pRF.wt was used to express a fusion protein that provides RLuc activity through its N-terminal domain and FLuc activity through its C-terminal domain. Site-specific mutagenesis (Quikchange; Stratagene) was performed to introduce amber, ochre and opal codons into the FLuc coding region to generate plasmids pRF.Y70am (9), pRF.Y70oc (9), pRF.Y70op, pRF.Q162am, pRF.Q162oc, pRF.Q162op, pRF.Y165am, pRF.Q283op, pRF.Y70oc/Y165am (9), pRF.Y70op/Y165am and pRF.Y70oc/Y165am/Q283op. In addition, tyrosine 70, glutamine 162 and tyrosine 165 of the wild-type FLuc gene were replaced with glutamine and serine codons, respectively, to yield plasmids pRF.Y70Q, pRF.Y70S, pRF.Q162S, pRF.Y165Q and pRF.Y165S.

    Figure 2. Schematic representation of the luciferase reporter mRNA encoding a Renilla luciferase–firefly luciferase (RLucFLuc) fusion protein. Internal stop codon mutations in the firefly luciferase gene are indicated. The luciferase reporter mRNA has two termination signals at the 3'-terminus separated by a UUC codon (...UAAUUCUAG...polyA...; termination codons are underlined).

    Plasmid pSVB.hsup2am contains the gene for the hsup2am amber suppressor tRNA derived from the E.coli tRNAGln . This tRNA was previously called hsup2A9am. Ochre (hsup2oc) and opal (hsup2op) suppressor tRNAs were generated by introducing C34 to U34 and C34U35 to U34C35 changes, respectively, in the anticodon of the tRNA. Plasmids pSVB.hsup2am, pSVB.hsup2oc and pSVB.hsup2op were altered to introduce additional U32 to C32, C38 to A38 or U32C38 to C32A38 mutations. Plasmids carrying amber, ochre and opal suppressor tRNAs derived from the human serine tRNA (pSVB.hseram, pSVB.hseroc, pSVB.hserop) have been described previously (16).

    Suppressor tRNA genes hsup2am, hsup2oc, hsup2.C32A38am and hsup2.C32A38oc were cloned into pBAD-araC (Invitrogen) for inducible expression of suppressor tRNAs in E.coli. The tRNA genes were amplified by PCR (forward primer: 5'-GGGGCCATGGACCAATTTGTTGGGGTATAGCCAAGCGGTAAGG-3'; reverse primer: 5'-GGGGTACGTATTGAATAAATTGGCTGGGGTACGAGG-3') using the respective pSVB plasmids as templates. The 110 bp PCR fragment was cut with NcoI and SnaBI and ligated into pBAD-araC cut with the same enzymes.

    The 1.7 kb DNA fragment encoding E.coli GlnRS was amplified by PCR (forward primer: 5'-CCCGAATTCGCCACCATGCATCACCATCACCATCACAGTGAGGCAGAAGCCC-3'; reverse primer: 5'CCCGCGGCCGCTTACTCGCCTACTTTCGCCC-3') from pESC-LEU.GlnRS (17) and inserted into the EcoRI/NotI sites of pCMVTNT (Promega). The resulting plasmid pTNT.EcGlnRS allows expression of E.coli GlnRS in mammalian cells with a His6-tag at the N-terminus of the protein.

    Transfection of mammalian cells

    HEK293T cells were maintained in DMEM (with 4500 mg/l of glucose; Cellgro) supplemented with 10% fetal bovine serum (Atlanta Biologicals Inc.), 2 mM glutamine, 100 U/ml of penicillin and 100 μg/ml of streptomycin (Invitrogen) at 37°C in a 5% CO2 atmosphere. Eighteen to twenty hours before transfection, cells were subcultured into 24-well plates. Transfection of HEK293T cells with plasmid DNA using Effectene (Qiagen) was as described previously (6) with minor modifications. Briefly, cells at 60–70% confluence were co-transfected with 0.5 μg of pRF plasmid carrying the luciferase reporter gene, 0.5 μg of pSVB plasmid carrying the tRNA gene and 5–10 ng of pCMVTNT plasmid carrying the E.coli GlnRS gene. The mixture of plasmid DNAs was diluted in 25 μl of EC buffer, supplied by the manufacturer, and then mixed with 2.5 μl of Enhancer and 5 μl of Effectene. The complexes were diluted with 0.25 ml of prewarmed (37°C) DMEM and added to the cells. 0.275 ml of medium supplemented with 10% serum and 10 mM sodium butyrate (Sigma) was added 3 h after transfection. Cells were harvested 48 h post-transfection.

    Assay for luciferase activity

    The Dual-Luciferase Reporter System (DLR; Promega) was used to measure luciferase activities in mammalian cell extracts as described (9). Measurement of luciferase activities was carried out on a Sirius tube luminometer (Berthold Detection Systems). For standard DLR assays, a 10-s pre-measurement delay and a 15-s measurement period were programmed. Luciferase activities are given as relative luminescence units (RLU) per μg of total cell protein. The protein concentration of cell lysates was determined with a BCA protein assay (Pierce) using BSA as standard. The values shown in the tables and figures represent the averages of at least three independent experiments; variations among experiments were <15%.

    Western blot analysis

    Cell lysates were prepared as described above and concentrated by acetone precipitation. Proteins were resolved by SDS–PAGE, transferred onto Immobilon PVDF membrane (Millipore) and probed with primary antibodies against FLuc (AB3256; polyclonal; Chemicon), RLuc (MAB4410; monoclonal; Chemicon) and actin (sc-9104; monoclonal; Santa Cruz Biotechnologies). The horseradish peroxidase-conjugated secondary antibodies were anti-goat IgG (Promega), anti-mouse IgG and anti-rabbit IgG (both Amersham). Signals were visualized using enhanced oxidase/luminol reagents (ECL; Perkin Elmer Life Sciences).

    Analysis of in vivo state of tRNAs

    Total RNAs were isolated from mammalian cells under acidic conditions using TRI-Reagent (Sigma) or TRIzol (Invitrogen). The tRNAs were separated by acid urea polyacrylamide gel electrophoresis (18), electroblotted onto Hybond-N+ membrane (Amersham) and detected by RNA blot hybridization. Membranes were pre-hybridized at 42°C in 10x Denhardt's solution/6x SSC/0.5% SDS. Hybridization was performed at 30°C in 6x SSC/0.1% SDS in the presence of a 5'-32P-labeled oligonucleotide, complementary to nucleotides 57–72 of the hsup2am tRNA. A 5'-32P-labeled oligonucleotide complementary to nucleotides 7–22 of the human serine tRNA was also used as an internal standard. Membranes were washed at room temperature, once with 6x SSC/0.1% SDS followed by two washes with 6x SSC, and then subjected to autoradiography. Northern blots were quantified by PhosphorImager analysis using ImageQuant software (Molecular Dynamics).

    Expression of mutant suppressor tRNAs in E.coli

    Transformants of E.coli CA274 (HfrH lacZ125am trpEam) carrying pBAD.hsup2am, pBAD.hsup2oc, pBAD.hsup2/C32A38am and pBAD.hsup2/C32A38oc, respectively, were grown in LBAmp medium at 37°C to mid-log phase (A600 of 0.5–0.6). Arabinose was added to a final concentration of 0.002% to induce transcription from the PBAD promoter. Cells were then grown for 80 min at 37°C and two more hours at room temperature (20°C), harvested by centrifugation, and analyzed for ?-galactosidase activity using the Beta-Glo assay system (Promega). Relative ?-galactosidase activities were normalized to the specific activities of ?-lactamase in the same extract (19) and to cell density at the time of harvest.

    RESULTS

    A complete set of orthogonal amber, ochre and opal suppressor tRNAs derived from E.coli tRNAGln (hsup2am, hsup2oc, hsup2op)

    We described previously (15) the expression of an amber suppressor tRNA derived from E.coli tRNAGln in mammalian cells (Figure 1). The E.coli suppressor tRNA gene, flanked by the original 5' and 3' sequences of the human initiator tRNAMet, was cloned into the mammalian expression vector pSVBpUC. An additional C9 to A9 mutation was introduced to improve transcription efficiency by mammalian RNA polymerase III. The resulting plasmid was transfected into mammalian cells, and the tRNA hsup2A9am (which we rename here as hsup2am for the sake of simplicity) was expressed along with or without E.coli GlnRS. The data indicated clearly that the suppressor tRNA was active in COS1 cells . Furthermore, its activity as an amber suppressor was strictly dependent upon co-expression of E.coli GlnRS . This work provided the first example of an orthogonal suppressor tRNA in mammalian cells.

    Figure 3. Acid urea PAGE/northern blot analysis of hsup2am, hsup2oc and hsup2op tRNAs. Total tRNA was isolated under acidic conditions and separated by acid urea PAGE. Suppressor tRNAs were visualized by RNA blot hybridization using a 5'-32P-labeled oligonucleotide complementary to nucleotides 57–72 of E.coli tRNAGln. A 5'-32P-labeled oligonucleotide complementary to nucleotides 7–22 of the human tRNASer was used as internal standard for quantitation of RNA and aminoacylation levels by PhosphorImager analysis.

    Table 1. Amber, ochre and opal suppression in HEK293T cells

    We now describe the generation of ochre and opal suppressor tRNAs derived from E.coli tRNAGln by changing the anticodon of the hsup2am tRNA to UUA (ochre; hsup2oc) and UCA (opal; hsup2op), respectively. To test and compare the activities of hsup2am, hsup2oc and hsup2op tRNAs in suppression, plasmids carrying amber, ochre or opal stop codon mutations in codon 162 of the firefly luciferase (FLuc) gene (Figure 2) were transfected into HEK293T cells along with plasmids carrying the genes for the suppressor tRNAs and E.coli GlnRS. Cells were harvested 48 h post-transfection and extracts were assayed for luciferase activity. Table 1 summarizes the results. No FLuc activity is detected over background in HEK293T cells that express the suppressor tRNAs but do not contain E.coli GlnRS (Table 1; lines 2, 5 and 8). Thus, along with the hsup2am, the hsup2oc and hsup2op tRNAs are also not recognized by any of the endogenous mammalian aaRSs. Suppression of the amber, ochre and opal codon in the FLuc gene was only observed in the presence of E.coli GlnRS (Table 1; lines 3, 6 and 9) yielding FLuc activities of 0.79 x 106 RLU/μg, 0.024 x 106 RLU/μg and 0.044 x 106 RLU/μg, respectively. These data represent the first example of a complete isogenic set of orthogonal amber, ochre and opal suppressor tRNAs and provide the first report of a 21st synthetase-ochre suppressor tRNA pair suitable for expression in mammalian cells.

    Interestingly, hsup2am tRNA yielded significantly higher levels of FLuc activity, 20–30 fold over the hsup2oc and hsup2op tRNAs, with the ochre suppressor having the lowest activity. These striking differences in suppression efficiencies can be explained, at least partly, by more efficient in vivo aminoacylation of the amber suppressor tRNA by E.coli GlnRS, as shown by acid urea PAGE followed by RNA blot hybridization of total tRNA isolated from HEK293T cells, using a probe directed against nucleotides 57–72 of the tRNA (Figure 3). PhosphorImager analysis indicates that hsup2am tRNA is aminoacylated almost quantitatively (87%), whereas hsup2oc and hsup2op tRNAs are aminoacylated to lower levels, 32 and 45%, respectively. These findings are not completely surprising since nucleotides in the anticodon of E.coli tRNAGln, changed to generate these suppressor tRNAs, are critical recognition elements for E.coli GlnRS based on the crystal structure of the tRNAGln–GlnRS complex (20) and on biochemical studies (21). An additional faster migrating band was detected for hsup2oc and hsup2op tRNAs. These bands probably represent tRNA species with additional modifications in the anticodon-loop of the suppressor tRNA (e.g. U34) or conformational variants of the tRNA. The effect of base modifications on the mobility of tRNAs in acid urea PAGE has been described previously (22).

    Mutants of the orthogonal amber, ochre and opal suppressor tRNAs with enhanced suppressor activity in mammalian cells

    The activity and aminoacylation specificity of tRNAs is affected by sequences in and around the anticodon loop and stem and by base modifications, especially those in the anticodon loop (23–26). For example, the modification of A37 located next to the anticodon is known to be important for the suppressor activity of tRNAs by strengthening the interaction between codon and anticodon (27,28). The enzyme responsible for modifying the A37 residue, the dimethylallyl diphosphate:tRNA dimethylallyl transferase (DMAPP-transferase) encoded by the miaA gene, has been identified in E.coli, yeast and mammalian cells and its substrate requirements have been characterized. The minimum recognition motif on the tRNA consists of a stretch of three A's, A36-A37-A38 .

    To improve the activity of hsup2 derived tRNAs, we introduced the following mutations in the anticodon loop of the hsup2am, hsup2oc and hsup2op tRNA genes (Figure 1): mutation of U38 to A38 to generate a potential recognition motif for the DMAPP-transferase; mutation of U32 to C32; and a double mutation of U32 and U38 to C32 and A38, respectively. The C32A38 double mutation generates an anticodon loop sequence, which mimics the sequence found in most strong suppressor tRNAs from prokaryotic and eukaryotic sources (16,23,30). The C32 mutation also removes a potential transcription termination signal (a string of 4 U residues U32–U35) for RNA polymerase III in the hsup2oc tRNA (31,32). Yarus and co-workers have previously generated an E.coli tRNAGln mutant similar to the C32A38 mutant by transplanting part of the anticodon stem and the entire anticodon loop of E.coli tRNATrp onto E.coli tRNAGln (24).

    The FLuc activities in extracts of cells transfected with the various mutants derived from hsup2am are shown in Table 2. The hsup2/C32am tRNA yielded FLuc activities of 2.3 x 106 RLU/μg, representing a 3-fold increase of activity compared with the hsup2am tRNA. The A38 mutation resulted in a 15-fold increase of FLuc activity, whereas the combined C32 and A38 mutations resulted in a 36-fold increase of FLuc activity. Similarly, the FLuc activities for the hsup2oc mutants (Table 3) increased 3.9 and 6-fold for the C32 and A38 single mutants, respectively. The most striking effects were seen for the hsup2/C32A38oc and hsup2/C32A38op double mutants. The hsup2/C32A38oc mutant showed an activity of 3.76 x 106 RLU/μg corresponding to a 156-fold increase from the original hsup2oc tRNA. The FLuc activity in cells transfected with the mutant hsup2op tRNA also increased from 0.04 x 106 to 8.57 x 106 RLU/μg for the hsup2/C32A38op double mutant (Table 4, lines 3 and 5) corresponding to a 200-fold increase. Altogether, these mutants provide an isogenic set of amber, ochre and opal suppressor tRNAs, each with a range of suppression activities in mammalian cells.

    Table 2. Amber suppression in HEK293T cells

    Table 3. Ochre suppression in HEK293T cells

    Table 4. Opal suppression in HEK293T cells

    All of the suppressor tRNA mutants, including those with highest suppression activities still require E.coli GlnRS for their activity (Tables 2, 3 and 4) and are, therefore, completely orthogonal in HEK293T cells. The 21st synthetase-amber, ochre and opal suppressor tRNA pairs composed of the strongest C32A38 double mutants, and E.coli GlnRS had translational efficiencies of 35, 4.5 and 10.4%, respectively, as estimated by normalizing FLuc activities in cells transfected with the mutant RLucFLuc genes to those in cells transfected with the wild-type RLucFLuc gene (Tables 2, 3 and 4). These efficiencies compare favorably to those obtained with the homologous human serine amber, ochre and opal suppressor tRNAs (22.4, 6.1 and 27.8%, respectively), which are aminoacylated by the endogenous human seryl-tRNA synthetase (Table 5).

    Table 5. Activity of amber, ochre and opal suppressor tRNAs derived from the human serine tRNA (hseram, hseroc and hserop) in HEK293T cells

    The results of immunoblot analyses using anti-FLuc antibodies (Figure 4) also confirm the orthogonality of the enhanced amber, ochre and opal suppressor tRNAs. Thus, an 87 kDa protein corresponding to the full-length RLucFLuc fusion protein is detected only in cells co-transfected with plasmids carrying the genes encoding the reporter protein, the suppressor tRNA and E.coli GlnRS. Furthermore, the intensities of the full-length RLucFLuc fusion protein band parallel the luciferase activities in enzyme assays, providing additional evidence for the translational efficiencies of the E.coli tRNAGln derived suppressor tRNAs in the order amber > opal > ochre.

    Figure 4. Amber, ochre and opal suppression in HEK293T cells. Immunoblot analysis of proteins isolated from cells co-transfected with plasmids carrying the genes encoding the luciferase reporter, hsup2/C32A38am, hsup2/C32A38oc or the hsup2/C32A38op tRNAs and, when present, E.coli GlnRS. The RLucFLuc fusion protein was detected with an anti-FLuc antibody and E.coli GlnRS was detected with an anti-His4-antibody. An antibody against ?-actin was used as a loading control. RLucFLuc, full-length fusion protein; RLucFLuc*, truncated RLucFLuc fusion protein.

    The increased FLuc activities in cells transfected with the various mutant suppressor tRNAs could be due to a combination of increased steady-state level of the tRNAs, increased extent of aminoacylation of the tRNAs and/or increased ribosomal activity of the tRNAs in suppression. To distinguish among these possibilities, the steady-state levels and extent of aminoacylation of all mutant tRNAs were analyzed by acid urea PAGE followed by RNA blot hybridization using probes directed against the mutant tRNAs and human tRNA3Ser as an internal control (Figure 5). The extent of aminoacylation remained essentially the same for all hsup2am mutants (87–95%), the appearance of an additional faster migrating band suggests heterogeneity in base modifications or the occurrence of conformational variants in some of the mutants (Figure 5A). Both the A38 mutation and the C32 mutation had similar effects. The extent of aminoacylation increased from 32% for hsup2oc and hsup2/A38oc tRNAs to 50% for hsup2/C32oc and hsup2/C32A38oc tRNAs (Figure 5B), whereas the extent of aminoacylation of the opal suppressor tRNA remained essentially unaltered at 50% (Figure 5C). In general, the relative intensity of the faster migrating band seemed to increase for all the C32A38 double mutants. Comparison of the total signals obtained for the suppressor tRNAs to that obtained for the human tRNA3Ser showed a maximal variation in steady-state levels of 2–2.3-fold for some of the mutant tRNAs, indicating a higher expression level or greater stability of these tRNAs. Taken together, these results suggest that the increased FLuc activities of 36, 156 and 200-fold seen in cells transfected with the C32A38 mutants of the hsup2am, hsup2oc and hsup2op tRNAs, respectively, are primarily due to increased activity of these tRNAs in suppression at the ribosomal level.

    Figure 5. Acid urea PAGE/northern blot analysis of additional mutants derived from hsup2am, hsup2oc and hsup2op tRNAs. (A) amber suppressor series; (B) ochre suppressor series; (C) opal suppressor series. Suppressor tRNAs were visualized by RNA blot hybridization using a 5'-32P-labeled oligonucleotide complementary to nucleotides 57–72 of tRNAGln. A 5'-32P-labeled oligonucleotide complementary to nucleotides 7–22 of the human tRNASer was used as internal standard (data not shown) for quantitation of RNA and aminoacylation levels by PhosphorImager analysis.

    Specificity of hsup2/C32A38am, hsup2/C32A38oc and hsup2/C32A38op tRNAs for their cognate codons

    The specificity of hsup2/C32A38am, hsup2/C32A38oc and hsup2/C32A38op tRNAs towards their cognate codons was investigated using the pRF.Q162am, Q162oc and Q162op reporter genes (Table 6). Despite their greatly enhanced activities towards their cognate codons, each suppressor tRNA translated only the corresponding cognate codon and had no significant activity towards a non-cognate stop codon. These results were confirmed with different luciferase stop codon mutations (at positions Y70, S163 and Y165) in different codon contexts (data not shown). The specificity of ochre suppressor tRNA mutants for the ochre codon in mammalian cells is in striking contrast to results obtained in E.coli. For example, expression of the same hsup2oc and hsup2/C32A38oc tRNAs in E.coli CA274 leads to significant suppression of an amber mutation in the chromosomal ?-galactosidase gene (Figure 6) by the ochre suppressor tRNAs.

    Table 6. Specificity of amber, ochre and opal suppression in HEK293T cells

    Figure 6. ?-galactosidase activity in cell extracts of E.coli with an amber mutation in the chromosomal ?-galactosidase gene transformed with plasmids carrying the hsup2am, hsup2/C32A38am, hsup2oc and hsup2/C32A38oc tRNA genes. Values represent the averages of at least three independent experiments.

    Concomitant suppression of two different termination codons (amber and ochre; amber and opal) in RLucFLuc mRNAs

    The hsup2/C32A38am and hsup2/C32A38oc tRNAs were used for concomitant suppression of amber and ochre codons using the RLucFLuc.Y70ocY165am reporter gene that had been used in earlier experiments (9). Table 7 summarizes the data. Co-expression of the hsup2/C32A38am and the hsup2/C32A38oc tRNAs and E.coli GlnRS in HEK293T cells resulted in a significant level of FLuc activity of 2.6 x 106 RLU/μg (Table 7; line 3). This level of FLuc activity is similar to that found in cells co-expressing the amber and ochre suppressors derived from human serine tRNA (Table 7; line 6). As expected, no FLuc activity was detected in cells not expressing E. coli GlnRS (line 2) or only one of the suppressor tRNAs (lines 4 and 5).

    Table 7. Concomitant suppression of amber and ochre codons, and amber and opal codons in HEK293T cells

    Similarly, the hsup2/C32A38am and hsup2/C32A38op tRNAs were used for concomitant suppression of amber and opal codons using the RLucFLuc.Y70opY165am reporter gene. In this case also, co-expression of the two tRNAs resulted in a significant level of FLuc activity of 1.7 x 106 RLU/μg (Table 7, line 9). Little activity was detected in cells not expressing E.coli GlnRS (line 8) or only one of the suppressor tRNAs (lines 10 and 11). These results clearly show that the newly generated amber, ochre and opal suppressor tRNAs derived from E.coli tRNAGln, hsup2/C32A38am, hsup2/C32A38oc, and hsup2/C32A38op, fulfill the requirements of high activity and specificity for their cognate codons necessary for site-specific incorporation of one or two unnatural amino acids into proteins in a mammalian system.

    Concomitant suppression of three different termination codons in RLucFLuc mRNA

    The availability of a complete set of orthogonal amber, ochre and opal suppressor tRNAs enabled us to question whether it would be possible to concomitantly suppress three different termination codons in a mRNA. Accordingly, the E.coli tRNAGln derived amber, ochre and opal suppressors were transfected into HEK293T cells along with the RLucFLuc.Y70ocY165amQ283op reporter gene. In a parallel experiment, human serine amber, ochre and opal suppressor tRNAs were also used. Table 8 summarizes the data on FLuc activities in extracts of transfected cells. It can be seen that the E.coli tRNAGln derived amber, ochre and opal suppressors can suppress all three termination codons in the reporter mRNA (Table 8, line 3). Suppression is dependent upon expression of E.coli GlnRS (Table 8, compare lines 2 and 3) and upon the presence of all three suppressor tRNAs (data not shown). As expected, FLuc activity is lower when suppressor tRNAs are used to suppress three different termination codons instead of two (compare FLuc activity in Table 8, line 3 to Table 7, line 3).

    Table 8. Concomitant suppression of amber, ochre and opal codons in HEK293T cells

    FLuc activity in extracts of cells transfected with all three E.coli tRNAGln derived suppressors is about 25% of that obtained with the human tRNASer derived suppressors (Table 8, compare lines 3 and 4). One possible reason for this is that the E.coli GlnRS activity in transfected cells becomes limiting, particularly since these suppressor tRNAs are known to be poor substrates for E.coli GlnRS and now three glutamine-accepting suppressor tRNAs are overexpressed to significant levels while E.coli GlnRS remains constant throughout the experiment. In contrast, the anticodon sequences in the human tRNASer derived suppressors are not important for their aminoacylation by human seryl-tRNA synthetase (33,34). Thus, it might be possible to increase the efficiency of suppression of ochre and opal codons by increasing the levels of expression of E.coli GlnRS in transfected cells. Another possibility would be to use mutant forms of E.coli GlnRS that have increased activity towards suppressor tRNAs similar to an approach taken by Yokoyama and co-workers for Methanococcus jannaschii TyrRS (35).

    DISCUSSION

    We have shown that amber, ochre and opal suppressor tRNAs derived from E.coli tRNAGln suppress amber, ochre and opal codons, respectively, in mammalian cells. The suppressor tRNAs, expressed in mammalian cells, are specific for their cognate codons and their activity in suppression is totally dependent upon expression of E.coli GlnRS. Thus, all three suppressor tRNAs are completely orthogonal in mammalian cells. Previously, we described orthogonal 21st aaRS-amber suppressor tRNA pair for use in mammalian cells based on E.coli GlnRS and amber suppressor tRNAs derived from E.coli tRNAGln (15) and the human initiator tRNAMet (36). Other laboratories have described a different orthogonal 21st aaRS-amber suppressor tRNA pair (3) and, recently, an orthogonal 21st aaRS-opal suppressor tRNA pair (11) for use in mammalian cells. The current work provides the first example of a complete set of orthogonal 21st aaRS-amber, ochre and opal suppressor tRNA pairs including the first example of a 21st aaRS-ochre suppressor tRNA pair. To the best of our knowledge, this work also demonstrates, for the first time, suppression of three different termination codons in a mRNA.

    Our finding that the E.coli tRNAGln derived amber, ochre and opal suppressors can be used to concomitantly suppress two or even three termination codons in a mRNA opens up the possibility of site-specific incorporation of two or even three different unnatural amino acids into one or more proteins in mammalian cells. For example, the E.coli GlnRS and the tRNAGln derived ochre suppressor tRNA system could be combined with either the E.coli TyrRS-Bacillus stearothermophilus (B.st.) tRNATyr derived amber suppressor (3) and/or the Bacillus subtilis (B.s.) TrpRS-B.s. tRNATrp derived opal suppressor system (11). Yokoyama and co-workers have described mutants of E.coli TyrRS that aminoacylate the B.st. tRNATyr derived amber suppressor with iodotyrosine (3,37) and Schultz and co-workers have, recently, described a mutant of B.s. TrpRS that aminoacylates the B.s. tRNATrp derived opal suppressor with 5'-hydroxytryptophan (11). Attempts at isolating mutants of E.coli GlnRS that aminoacylate the E.coli tRNAGln derived suppressor tRNAs with analogs of glutamine are underway in our laboratory. The overall suppression efficiencies for concomitant suppression of two or three different termination codons by amber, ochre and opal suppressor tRNAs appear to be within the expected range of multiplying the efficiencies of the individual suppression events. The efficiencies of suppressor tRNAs derived from the E.coli tRNAGln are somewhat lower than efficiencies of suppressor tRNAs derived from the human tRNASer, probably reflecting the depletion of E.coli GlnRS in the presence of two or three glutamine-accepting tRNAs.

    An alternative approach in using mutant 21st aaRSs to aminoacylate the corresponding suppressor tRNAs with unnatural amino acids in vivo involves the import into mammalian cells of suppressor tRNAs that are aminoacylated in vitro with an unnatural amino acid. Because each suppressor tRNA molecule works only once to insert the unnatural amino acid into a protein, this approach is more useful for producing smaller amounts of a protein, whose properties and function can be analyzed directly in vivo. The isolation of a complete set of orthogonal suppressor tRNAs and the finding that these suppressor tRNAs are specific for their cognate codons, open up the possibility of aminoacylation of each of the suppressor tRNAs with a different unnatural amino acid and their import into mammalian cells for the site-specific incorporation of two or three unnatural amino acids into proteins.

    The incorporation of two different unnatural amino acids into a protein using two different suppressor tRNAs involves an mRNA carrying two different termination codons within its open reading frame. This approach poses no particular problems in terms of the suppressor tRNAs also reading through the normal termination signal at the end of the reading frame. For example, the firefly luciferase gene used in this work contains, at the end of the reading frame, an ochre codon followed by UUC and then an amber codon. Therefore, a combination of opal and amber suppressor tRNAs can be used to incorporate two different unnatural amino acids into the protein without the suppressor tRNAs also reading through the normal termination codons, the ochre codon acting as a barrier in this case. However, the use of a mRNA carrying three different termination codons in the open reading frame requires strategies for preventing the readthrough of normal termination codon(s) at the end of the reading frame by the three suppressor tRNAs. Our finding that suppression of the ochre codon is the weakest (Tables 2, 3 and 4) of the three termination codons, suggests that the use of a gene carrying tandem ochre termination codons at the end of the reading frame would minimize any significant readthrough of the termination codons beyond the end of the mRNA. Parenthetically, we note that under the conditions used, there is no significant readthrough of cellular protein genes as indicated by the lack of any deleterious effects on cell viability.

    The orthogonality of the E.coli tRNAGln derived amber suppressor was described previously (15). However, the orthogonality of the tRNAGln derived ochre and opal suppressors was not necessarily expected. In particular, the opal suppressor tRNA, which has C35 in the middle of the anticodon sequence (Figure 1), could have been a substrate for one of the mammalian aaRSs, including TrpRS, which uses C35 as an important identity determinant. For example, in E.coli, S?ll, Inokuchi and co-workers (38) have shown that the E.coli tRNAGln derived opal suppressor is a substrate for E.coli TrpRS and that this opal suppressor tRNA inserts predominantly tryptophan into proteins. Our finding that the E.coli tRNAGln derived opal suppressor tRNA is not a substrate for mammalian TrpRS (Table 4) or any other mammalian aaRS (Figure 5C), indicates that the requirements in the substrate tRNA for mammalian TrpRS are quite different from those of E.coli TrpRS. This conclusion is basically in agreement with reports that B.s. tRNATrp is a poor substrate in vitro for yeast and mammalian TrpRS (39–41)

    In bacteria, ochre suppressor tRNAs also suppress amber codons (42–44), whereas in eukaryotes, to the extent that they have been studied, ochre suppressor tRNAs are specific for the ochre codon (6,9,45). This is commonly ascribed to Wobble pairing (46) between the modified U, the first nucleotide in the anticodon of the ochre suppressor tRNA and G, the third nucleotide of the amber codon UAG. Nevertheless, the finding in this work, that the most active E.coli tRNAGln derived ochre suppressor, when expressed in mammalian cells, is still specific for the ochre codon (Table 6) is noteworthy, since the same tRNA, when expressed in E.coli, suppresses the amber codon quite well in E.coli (Figure 6). Whether the specificity of the ochre suppressor tRNA, expressed in mammalian cells, is due to a different base modification of the U at the Wobble position 34, or whether the eukaryotic ribosome is inherently more restrictive in translation of the amber codon by an ochre suppressor tRNA, remains to be seen.

    In attempts to isolate suppressor tRNAs with a range of activity in suppression, we introduced mutations in the anticodon loop of the suppressor tRNAs, in addition to those mutations in the anticodon sequence necessary to obtain the various suppressors. These mutations in the anticodon loop increased the activity of the suppressor tRNAs significantly. The most active suppressor tRNAs contained the C32 and A38 mutations, thus, showing a anticodon loop sequence found in most eukaryotic elongator tRNAs (47). Yarus, Soll and co-workers have described similar, but not identical, mutants by genetic selection in E.coli (48) and by swapping the restriction fragments comprising the anticodon loop and the proximal anticodon stem base pair between amber suppressor tRNA genes derived from E.coli tRNATrp and tRNAGln (24,49). Yarus and co-workers also found that the tRNAGln mutant carrying the C32A38 changes in the anticodon loop was the most active in suppression among the tRNAGln derived amber suppressors in E.coli (24).

    The primary reason for the increased activity of the E.coli tRNAGln derived amber, ochre and opal suppressors also carrying the C32A38 mutations, in mammalian cells is most likely due to increased activity at the ribosomal level. For example, in the case of the amber suppressor tRNA, where there is a 36-fold increase in the activity of the most active mutant, the tRNAs are all aminoacylated to approximately the same levels (87–95%) and there is at the most a 2–2.5 fold difference in steady-state levels of the suppressor tRNAs (Figure 5A). Similarly, for the mutants derived from the ochre and opal suppressor tRNAs, while there is an 1.5-fold difference in the extent of aminoacylation of one of the tRNAs (Figure 5B and C) and a 2–2.5 fold difference in the steady-state levels of some of the tRNAs, these differences cannot account for the 156 and 200-fold increase in activity, respectively, of the ochre and opal suppressor tRNAs carrying the C32A38 mutations.

    Finally, our finding that the E.coli tRNAGln derived amber, ochre and opal suppressor tRNAs are dependent upon expression of E.coli GlnRS, can be used to regulate the suppression of the amber, ochre and opal codons in mammalian cells by regulating the expression of E.coli GlnRS. Cell lines carrying such inducible suppressor tRNA function would be useful for a variety of purposes, for example, the propagation of animal viruses carrying nonsense mutations in their genes. Suppressor tRNAs have been used for diphtheria toxin-mediated ablation of photoreceptor cells in Drosophila (50) and toxin-mediated ablation dependent upon suppressor tRNA function has also been suggested as a possibility for cancer therapy (51). Moreover, the finding that nonsense mutations in genes cause a variety of human genetic diseases (such as muscular dystrophy, Xeroderma pigmentosum, etc.), has led to suggestions of suppressor tRNAs as a tool in gene therapy (52,53). Unfortunately, constitutive expression of suppressor tRNAs is usually detrimental to mammalian cells and to Drosophila (12,13). There is, therefore, an important need for regulated expression of suppressor tRNA function in these cells. Previously, we described tetracycline-regulated suppression of amber codons in mammalian cells by regulating the expression of E.coli GlnRS (54). This system can now be extended to obtain tetracycline-regulated suppression of ochre and opal codons in mammalian cells.

    ACKNOWLEDGEMENTS

    We thank Annmarie McInnis for patience and care in the preparation of this manuscript. Work in our laboratory is supported by grants GM 067741 from the National Institutes of Health and DAAD 19-99-1-0300 and W911NF-04-1-0353 from the US Army Research Office.

    REFERENCES

    K?hrer,C. and RajBhandary,U.L. ( (2004) ) Proteins carrying one or more unnatural amino acids. In Ibba,M., Francklyn,C. and Cusack,S. (eds), Aminoacyl-tRNA Synthetases. Landes Bioscience, Georgetown, TX.

    Wang,L., Brock,A., Herberich,B. and Schultz,P.G. ( (2001) ) Expanding the genetic code of Escherichia coli. Science, , 292, , 498–500.

    Sakamoto,K., Hayashi,A., Sakamoto,A., Kiga,D., Nakayama,H., Soma,A., Kobayashi,T., Kitabatake,M., Takio,K., Saito,K., Shirouzu,M., Hirao,I. and Yokoyama,S. ( (2002) ) Site-specific incorporation of an unnatural amino acid into proteins in mammalian cells. Nucleic Acids Res., , 30, , 4692–4699.

    Chin,J.W., Cropp,T.A., Anderson,J.C., Mukherji,M., Zhang,Z. and Schultz,P.G. ( (2003) ) An expanded eukaryotic genetic code. Science, , 301, , 964–967.

    Nowak,M.W., Kearney,P.C., Sampson,J.R., Saks,M.E., Labarca,C.G., Silverman,S.K., Zhong,W., Thorson,J., Abelson,J.N., Davidson,N. et al. ( (1995) ) Nicotinic receptor binding site probed with unnatural amino acid incorporation in intact cells. Science, , 268, , 439–442.

    K?hrer,C., Xie,L., Kellerer,S., Varshney,U. and RajBhandary,U.L. ( (2001) ) Import of amber and ochre suppressor tRNAs into mammalian cells: a general approach to site-specific insertion of amino acid analogues into proteins. Proc. Natl Acad. Sci. USA, , 98, , 14310–14315.

    Ilegems,E., Pick,H.M. and Vogel,H. ( (2002) ) Monitoring mis-acylated tRNA suppression efficiency in mammalian cells via EGFP fluorescence recovery. Nucleic Acids Res., , 30, , e128, 1–6.

    Monahan,S.L., Lester,H.A. and Dougherty,D.A. ( (2003) ) Site-specific incorporation of unnatural amino acids into receptors expressed in mammalian cells. Chem. Biol., , 10, , 573–580.

    K?hrer,C., Yoo,J., Bennett,M., Schaack,J. and RajBhandary,U.L. ( (2003) ) A possible approach to site-specific insertion of two different unnatural amino acids into proteins in mammalian cells via nonsense suppression. Chem. Biol., , 10, , 1095–1102.

    Anderson,J.C., Wu,N., Santoro,S.W., Lakshman,V., King,D.S. and Schultz,P.G. ( (2004) ) An expanded genetic code with a functional quadruplet codon. Proc. Natl Acad. Sci. USA., , 101, , 7566–7571.

    Zhang,Z., Alfonta,L., Tian,F., Bursulaya,B., Uryu,S., King,D.S. and Schultz,P.G. ( (2004) ) Selective incorporation of 5-hydroxytryptophan into proteins in mammalian cells. Proc. Natl Acad. Sci. USA., , 101, , 8882–8887.

    Hudziak,R.M., Laski,F.A., RajBhandary,U.L., Sharp,P.A. and Capecchi,M.R. ( (1982) ) Establishment of mammalian cell lines containing multiple nonsense mutations and functional suppressor tRNA genes. Cell, , 31, , 137–146.

    Laski,F.A., Ganguly,S., Sharp,P.A., RajBhandary,U.L. and Rubin,G.M. ( (1989) ) Construction, stable transformation, and function of an amber suppressor tRNA gene in Drosophila melanogaster. Proc. Natl Acad. Sci. USA., , 86, , 6696–6698.

    Bennett,M. and Schaack,J. ( (2003) ) Development of a dual-luciferase fusion gene as a sensitive marker for site-directed DNA repair strategies. J. Gene Med., , 5, , 723–732.

    Drabkin,H.J., Park,H.J. and RajBhandary,U.L. ( (1996) ) Amber suppression in mammalian cells dependent upon expression of an Escherichia coli aminoacyl-tRNA synthetase gene. Mol. Cell. Biol., , 16, , 907–913.

    Capone,J.P., Sharp,P.A. and RajBhandary,U.L. ( (1985) ) Amber, ochre and opal suppressor tRNA genes derived from a human serine tRNA gene. EMBO J., , 4, , 213–221.

    Kowal,A.K., K?hrer,C. and RajBhandary,U.L. ( (2001) ) Twenty-first aminoacyl-tRNA synthetase-suppressor tRNA pairs for possible use in site-specific incorporation of amino acid analogues into proteins in eukaryotes and in eubacteria. Proc. Natl Acad. Sci. USA., , 98, , 2268–2273.

    Varshney,U., Lee,C.P. and RajBhandary,U.L. ( (1991) ) Direct analysis of aminoacylation levels of tRNAs in vivo. Application to studying recognition of Escherichia coli initiator tRNA mutants by glutaminyl-tRNA synthetase. J. Biol. Chem., , 266, , 24712–24718.

    Mayer,C., K?hrer,C., Kenny,E., Prusko,C. and RajBhandary,U.L. ( (2003) ) Anticodon sequence mutants of Escherichia coli initiator tRNA: effects of overproduction of aminoacyl-tRNA synthetases, methionyl-tRNA formyltransferase, and initiation factor 2 on activity in initiation. Biochemistry, , 42, , 4787–4799.

    Rould,M.A., Perona,J.J., S?ll,D. and Steitz,T.A. ( (1989) ) Structure of E.coli glutaminyl-tRNA synthetase complexed with tRNAGln and ATP at 2.8 A resolution. Science, , 246, , 1135–1142.

    Jahn,M., Rogers,M.J. and S?ll,D. ( (1991) ) Anticodon and acceptor stem nucleotides in tRNAGln are major recognition elements for E.coli glutaminyl-tRNA synthetase. Nature, , 352, , 258–260.

    Mangroo,D., Limbach,P.A., McCloskey,J.A. and RajBhandary,U.L. ( (1995) ) An anticodon sequence mutant of Escherichia coli initiator tRNA: possible importance of a newly acquired base modification next to the anticodon on its activity in initiation. J. Bacteriol., , 177, , 2858–2862.

    Yarus,M. ( (1982) ) Translational efficiency of transfer RNAs: uses of an extended anticodon. Science, , 218, , 646–652.

    Yarus,M., Cline,S., Raftery,L., Wier,P. and Bradley,D. ( (1986) ) The translational efficiency of tRNA is a property of the anticodon arm. J. Biol. Chem., , 261, , 496–505.

    Agris,P.F. ( (2004) ) Decoding the genome: a modified view. Nucleic Acids Res., , 32, , 223–238.

    Colby,D.S., Schedl,P. and Guthrie,C. ( (1976) ) A functional requirement for modification of the wobble nucleotide in the anticodon of a T4 suppressor tRNA. Cell, , 9, , 449–463.

    Ericson,J.U. and Bj?rk,G.R. ( (1991) ) tRNA anticodons with the modified nucleoside 2-methylthio-N6-(4-hydroxyisopentenyl)adenosine distinguish between bases 3' of the codon. J. Mol. Biol., , 218, , 509–516.

    Bj?rk,G.R. ( (1995) ) Biosynthesis and function of modified nucleosides. In S?ll, D. and RajBhandary,U.L. (eds), tRNA: Structure, Biosynthesis, and Function. American Society for Microbiology, Washington DC, pp.165–205.

    Motorin,Y., Bec,G., Tewari,R. and Grosjean,H. ( (1997) ) Transfer RNA recognition by the Escherichia coli delta2-isopentenyl-pyrophosphate:tRNA delta2-isopentenyl transferase: dependence on the anticodon arm structure. RNA, , 3, , 721–733.

    Smith,D., Breeden,L., Farrell,E. and Yarus,M. ( (1987) ) The bases of the tRNA anticodon loop are independent by genetic criteria. Nucleic Acids Res., , 15, , 4669–4686.

    Koski,R.A., Clarkson,S.G., Kurjan,J., Hall,B.D. and Smith,M. ( (1980) ) Mutations of the yeast SUP4 tRNATyr locus: transcription of the mutant genes in vitro. Cell, , 22, , 415–425.

    Hamada,M., Sakulich,A.L., Koduru,S.B. and Maraia,R.J. ( (2000) ) Transcription termination by RNA polymerase III in fission yeast. A genetic and biochemically tractable model system. J. Biol. Chem., , 275, , 29076–29081.

    Achsel,T. and Gross,H.J. ( (1993) ) Identity determinants of human tRNASer: sequence elements necessary for serylation and maturation of a tRNA with a long extra arm. EMBO J., , 12, , 3333–3338.

    Heckl,M., Busch,K. and Gross,H.J. ( (1998) ) Minimal tRNASer and tRNASec substrates for human seryl-tRNA synthetase: contribution of tRNA domains to serylation and tertiary structure. FEBS Lett., , 427, , 315–319.

    Kobayashi,T., Nureki,O., Ishitani,R., Yaremchuk,A., Tukalo,M., Cusack,S., Sakamoto,K. and Yokoyama,S. ( (2003) ) Stuctural basis for orthogonal tRNA specificities of tyrosyl-tRNA synthetases for genetic code expansion. Nature Struct. Biol., , 10, , 425–432.

    Drabkin,H.J., Estrella,M. and RajBhandary,U.L. ( (1998) ) Initiator–elongator discrimination in vertebrate tRNAs for protein synthesis. Mol. Cell. Biol., , 18, , 1459–1466.

    Kiga,D., Sakamoto,K., Kodama,K., Kigawa,T., Matsuda,T., Yabuki,T., Shirouzu,M., Harada,Y., Nakayama,H., Takio,K., Hasegawa,Y., Endo,Y., Hirao,I. and Yokoyama,S. ( (2002) ) An engineered Escherichia coli tyrosyl-tRNA synthetase for site-specific incorporation of an unnatural amino acid into proteins in eukaryotic translation and its application in a wheat germ cell-free system. Proc. Natl Acad. Sci. USA, , 99, , 9715–9720.

    Rogers,M.J., Adachi,T., Inokuchi,H. and S?ll,D. ( (1992) ) Switching tRNAGln identity from glutamine to tryptophan. Proc. Natl Acad. Sci. USA, , 89, , 3463–3467.

    Xue,H., Shen,W., Giege,R. and Wong,J.T. ( (1993) ) Identity elements of tRNATrp: identification and evolutionary conservation. J. Biol. Chem., , 268, , 9316–9322.

    Xu,F., Chen,X., Xin,L., Chen,L., Jin,Y. and Wang,D. ( (2001) ) Species-specific differences in the operational RNA code for aminoacylation of tRNATrp. Nucleic Acids Res., , 29, , 4125–4133.

    Guo,Q., Gong,Q., Tong,K.L., Vestergaard,B., Costa,A., Desgres,J., Wong,M., Grosjean,H., Zhu,G., Wong,J.T. and Xue,H. ( (2002) ) Recognition by tryptophanyl-tRNA synthetases of discriminator base on tRNATrp from three biological domains. J. Biol. Chem., , 277, , 14343–14349.

    Brenner,S. and Beckwith,J.R. ( (1965) ) Ochre mutants, a new class of suppressible nonsense mutants. J. Mol. Biol., , 13, , 629–637.

    Raftery,L.A., Egan,J.B., Cline,S.W. and Yarus,M. ( (1984) ) Defined set of cloned termination suppressors: In vivo activity of isogenetic UAG, UAA, and UGA suppressor tRNAs. J. Bacteriol., , 158, , 849–859.

    Eggertsson,G. and S?ll,D. ( (1988) ) Transfer ribonucleic acid-mediated suppression of termination codons in Escherichia coli. Microbiol. Rev., , 52, , 354–374.

    Capone,J.P., Sedivy,J.M., Sharp,P.A. and RajBhandary,U.L. ( (1986) ) Introduction of UAG, UAA, and UGA nonsense mutations at a specific site in the Escherichia coli chloramphenicol acetyltransferase gene: use in measurement of amber, ochre, and opal suppression in mammalian cells. Mol. Cell. Biol., , 6, , 3059–3067.

    Crick,F.H. ( (1966) ) Codon–anticodon pairing: the wobble hypothesis. J. Mol. Biol., , 19, , 548–555.

    Marck,C. and Grosjean,H. ( (2002) ) tRNomics: analysis of tRNA genes from 50 genomes of Eukarya, Archaea, and Bacteria reveals anticodon-sparing strategies and domain-specific features. RNA, , 8, , 1189–1232.

    Bradley,D., Park,J.V. and Soll,L. ( (1981) ) tRNA2Gln Su+2 mutants that increase amber suppression. J. Bacteriol., , 145, , 704–712.

    Yarus,M., McMillan,C.III, Cline,S., Bradley,D. and Snyder,M. ( (1980) ) Construction of a composite tRNA gene by anticodon loop transplant. Proc. Natl Acad. Sci. USA, , 77, , 5092–5096.

    Kunes,S. and Steller,H. ( (1991) ) Ablation of Drosophila photoreceptor cells by conditional expression of a toxin gene. Genes Dev., , 5, , 970–983.

    Robinson,D.F. and Maxwell,I.H. ( (1995) ) Suppression of single and double nonsense mutations introduced into the diphtheria toxin A-chain gene: a potential binary system for toxin gene therapy. Hum. Gene Ther., , 6, , 137–143.

    Temple,G.F., Dozy,A.M., Roy,K.L. and Kan,Y.W. ( (1982) ) Construction of a functional human suppressor tRNA gene: an approach to gene therapy for beta-thalassaemia. Nature, , 296, , 537–540.

    Atkinson,J. and Martin,R. ( (1994) ) Mutations to nonsense codons in human genetic disease: implications for gene therapy by nonsense suppressor tRNAs. Nucleic Acids Res., , 22, , 1327–1334.

    Park,H.J. and RajBhandary,U.L. ( (1998) ) Tetracycline-regulated suppression of amber codons in mammalian cells. Mol. Cell. Biol., , 18, , 4418–4425.(Caroline K?hrer, Eric L. Sullivan and Ut)