Charging of tRNA with non-natural amino acids at high pressure Malgorzata Giel-Pietraszuk and Jan Barciszewski Institute of Bioorganic Chemistry, Polish Academy of Sciences, Poznan, Poland Site-specific incorporation of non-natural amino acids into proteins is an increasingly emerging field because of the application of non-natural amino acids as biophysical probes in structure–function studies. Moreover, modified peptides may be key pharmaceuticals for the treatment of a variety of dis- eases [1]. Among these compounds are protease inhibitors, a classic example of which are the HIV protease inhibitors [2,3]. Replacement of methionine with selenomethionine has been used extensively for phase determination in protein crystallography, and the exchange of 4-fluorotryptophan for tryptophan has been used in NMR analysis. Studies on the function and properties of proteins require mutants containing amino acid analogues, for example thiopr- oline, at multiple sites that do not influence protein function, including immunogenicity, but may serve as promising vehicles for targeted drug delivery [4]. Replacement of leucine residues with 5,5,5-trifluoro- leucine at d-positions of the leucine GCN4-zipper peptide increases the thermal stability of the coiled- coil structure [5]. Several strategies have been used to introduce non- natural amino acids into proteins [1,6]. One of the first was the derivatization of amino acids at reactive side chains, for example, conversion of Lys to N e -acetyl lysine. Chemical preparation provides a straightfor- ward method for the incorporation of non-natural amino acids using solid-phase peptide synthesis, but for technical reasons, it remains restricted to small peptides [7–9]. Development of enzymatic and native chemical ligations allows us to obtain larger proteins [10]. General in vitro methods of site-specific incorpor- ation of the desired amino acid into a protein are based on chemically charged suppressor tRNA, used in a translation system [11]. Over 100 non-natural amino acids have been introduced into proteins of varying size [12]. The utility of mischarged tRNAs has Keywords high pressure; non-natural amino acids; tRNA charging Correspondence Jan Barciszewski, Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12 ⁄ 14, 61-704 Poznan, Poland Fax: +49 61 852 05 32 Tel: +48 61 852 85 03 (ext. 132) E-mail: Jan.Barciszewski@ibch.poznan.pl (Received 3 October 2005, revised 25 April 2006, accepted 9 May 2006) doi:10.1111/j.1742-4658.2006.05312.x We show a simple and reliable method of tRNA aminoacylation with natural, as well as non-natural, amino acids at high pressure. Such specific and noncognate tRNAs can be used as valuable substrates for protein engineering. Aminoacylation yield at high pressure depends on the chem- ical nature of the amino acid used and it is up to 10%. Using CoA, which carries two potentially reactive groups -SH and -OH, as a model com- pound we showed that at high pressure amino acid is bound preferentially to the hydroxyl group of the terminal ribose ring. Abbreviations AARS, aminoacyl–tRNA synthetase; aa-tRNA, aminoacyl-tRNA; Cl-Phe, p-chloro-phenylalanine; Cl-Tyr, 3-chloro-tyrosine; DOPA, 3,4-dihydroxyphenylalanine; D-Orn, D-ornithine; L-Orn, L-ornithine; Orn-Ado, adenosyl-ornithine; p-Cl-Phe-Ado, adenosyl-p-chlorophenylalanine; Phe-Ado, adenosyl-phenylalanine; PPO, 2,5-diphenyloxazone. 3014 FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS been expanded by developing chemical acylation of the unprotected dinucleotide pCpA, followed by enzymatic ligation to the 3¢-terminus of truncated tRNA using T4 RNA ligase. This approach has a low acylation yield of 3–4% [13,14]. An improved version of the method, based on the acylation of fully protected 5¢pCpCpA resulted in 26% charging [15]. Other meth- ods of tRNA aminoacylation with non-natural amino acids take advantage of ribozymes or appropriately mutated aminoacyl-tRNA synthetases (AARS) charg- ing tRNA, having specific codons consisting of four or five bases [16–20]. These methods, in contrast to enzy- matic aminoacylation, which is generally limited to natural amino acids or their analogues, enables the acylation of tRNAs with any non-native amino acid [16–21]. We have previously shown that yeast tRNA Phe can be charged with phenylalanine at high pressure without a specific AARS and the product, Phe– tRNA Phe , was the correct substrate for protein biosyn- thesis [22,23]. Here, we show that aminoacylation of tRNA at high pressure may be used to prepare aminoacyl-tRNA (aa-tRNA) using any natural or non-natural amino acid. Using CoA, we also show that amino acid binds specifically to the ribose ring at high pressure. Applica- tion of MS provides evidence that charging occurs at the hydroxyl group of the 3¢-end ribose. Results TRNA Phe aminoacylation with non-natural amino acids at high pressure Aminoacylation of tRNA Phe using natural and non- cognate amino acids was carried out at 6 kbar as described in Experimental procedures. Charging of tRNA Phe with 3-chloro-tyrosine (Cl-Tyr), l-ornithine (l-Orn), d-ornithine (d-Orn), 3,4-dihydroxyphenylala- nine (DOPA) and p-chloro-phenylalanine (Cl-Phe) was analysed using PAGE (Fig. 1A,B). The amounts of amino acid incorporated into 1600 pmol of yeast tRNA Phe , estimated on the basis of imagequant, were 40, 80, 96, 72 and 144 pmol, respectively (Table 1). The amounts of Val–tRNA Val and Leu–tRNA Val cal- culated from the scintillation measurement of gel slices, obtained after fluorography (Fig. 1C), were 123 and 60 pmol per 1600 pmol of Escherichia coli tRNA Val (Table 1). The yield of the yeast tRNA Phe charging with natural amino acids is shown in Table 1. Time-dependent aminoacylation of tRNA with tryptophan showed that the best result, 96 pmol of Tyr per 1600 pmol of tRNA Phe , was obtained after 30 min of pressure at 6 kbar (Fig. 2). The yield of charging of crude tRNA from wheatgerm with lysine at high pressure was 333 pmol Lys per 40 lg tRNA, and in the enzymatic reaction was 133 pmol (Fig. 3). 1234 123 12 3 4 56 aa-tRNA Phe tRNA Phe aa-tRNA Phe aa-tRNA Val tRNA Phe A B C Fig. 1. Detection of aa-tRNA using acidic ⁄ urea gel electrophoresis. Reactions were performed as described in the Experimental proce- dures. Aminoacylation of tRNA with different amino acids carried at 6 kbar for 6 h. (A) Aminoacylation of yeast tRNA Phe . 1, Control [5¢- 32 P]tRNA Phe at 6 kbar for 6 h in a reaction buffer without amino acid; 2, control [5¢- 32 P]tRNA Phe incubated at normal pressure for 6 h with Cl-Tyr; 3, [5¢- 32 P]tRNA Phe with Cl-Tyr; 4, [5¢- 32 P]tRNA Phe with L-Orn, [5¢- 32 P]tRNA Phe with D-Orn, [5¢- 32 P]tRNA Phe with DOPA. (B) Aminoacylation of yeast tRNA Phe .1,[5¢- 32 P]tRNA Phe not treated at high pressure; 2, [5¢- 32 P]tRNA Phe at 6 kbar for 5 h in reaction buffer without amino acid; and 3, [5¢- 32 P]tRNA Phe with Cl-Phe; 4, [5¢- 32 P]tRNA Phe with Phe. (C) Aminoacylation of E. coli tRNA Val .1, Control [ 14 C]Val-tRNA Val from T. thermophilus aminoacylated enzy- matically, 2, [ 14 C]Leu-tRNA Val ;3,[ 14 C]Val-tRNA Val acylated under high pressure. Bands were visualized by fluorography. M. Giel-Pietraszuk and J. Barciszewski Charging of tRNA with non-natural amino acids FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS 3015 HPLC-MS analysis Charging of tRNA with non-natural amino acids was also confirmed using HPLC-MS analysis. The HPLC chromatogram of aa-tRNA, partially hydrolysed with RNaseA, showed a peak at a retention time of 9.99 min; this was identified using ESI-MS as adenosyl-phenyl- alanine (Phe-Ado) (Fig. 4A). Two major signals at m ⁄ z ¼ 415 and 437 corresponded to [M + 1] + and [M + Na] + ions, respectively. The ESI-MS spectrum of tRNA aminoacylated with Chl-Phe showed signals at m ⁄ z ¼ 415 and 457, corresponding to [M + 1] + and [M + Na] + of Phe-Ado, respectively, and m ⁄ z ¼ 450 corresponding to [M + 1] + of adenosyl-p-chloroph- enylalanine (p-Cl-Phe-Ado) (Fig. 4B). The strongest sig- nal on the ESI-MS spectrum, m ⁄ z ¼ 419, recorded for aminoacylation of tRNA with ornithine, originated from adenosyl-ornithine (Orn-Ado), whereas the signals at m ⁄ z ¼ 331 and 389 derived from its decomposition products. The first corresponded to fragmentation of a five-membered ring sugar by releasing 29 mass units, and the second by breaking the C–C bond between ribose and the methyl group (Fig. 4C) [24,25]. Activity of high pressure-charged tRNA in protein biosynthesis Activity of [ 14 C]Phe–tRNA Phe aminoacylated under high pressure has been checked previously in an in vitro translation assay using poly-(U)-programmed ribo- somes [23]. Here we also show that [ 14 C]Val–tRNA Val prepared at high pressure was active in an in vitro transcription ⁄ translation assay. This means that high pressure-charged tRNA is a good substrate in protein synthesis (Fig. 5). Aminoacylation of CoA at high pressure The data clearly show that high pressure induces acyla- tion of the ribose OH group. In order to check Table 1. Yield of aminoacylation of tRNA with different amino acids. Amino acid pmole amino acid per 1600 pmole tRNA tRNA Phe (yeast) crude tRNA (wheatgerm) tRNA Val (E. coli) 3-Chlorotyrosine 40 (2.5%) a Chlorophenylanine 144 (9%) a L-Ornithine 80 (5%) a D-Ornithine 96 (6%) a DOPA 72 (4.5%) a Phenylalanine [160 (10%) a ] 116 b 247 b – Tyrosine 114 b (7.1%) 163 b – Tryptophan 109 b (6.8%) 173 b – Arginine 96 b (6.0%) 170 b – Lysine 95 b (5.9%) 204 b – Histidine 82 b (5.1%) – – Valine 74 b (4.6%) 134 b 123 c Leucine 71 b (4.4%) 124 b 60 c Glycine 59 b (3.6%) – Glutamic acid 58 b (3.6%) – Methionine 54 b (3.3%) – Alanine 52 b (3.5%) – a IMAGEQUANT measurement of [5¢- 32 P]tRNA separated on acidic ⁄ urea PAGE. b Filter binding assay of[ 3 H] or[ 14 C]-amino acids. c Scintillation counting of gel slabs containing [ 3 H] or [ 14 C]-amino acids. A B Fig. 2. Aminoacylation of yeast tRNA Phe with [ 3 H]Trp at 6 kbar pres- sure as a function of (A) tRNA concentration and (B) time. Fig. 3. High-pressure aminoacylation of tRNA crude from wheat- germ with [ 14 C] Lys at 6 kbar (r) in a control experiment, enzymat- ic charging with crude aa-tRNA synthetase was carried out at ambient pressure (n). Charging of tRNA with non-natural amino acids M. Giel-Pietraszuk and J. Barciszewski 3016 FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS whether the hydroxyl group of ribose is preferentially acylated, we used CoA as a model. CoA and trypto- phan were subjected to a pressure of 10 kbar overnight followed by TLC. Bands of free substrates and amino- acyl–CoA were visualized using UV light and then stained with ninhydrin. The yield of this reaction was  8% (Fig. 6). Analysis of dephosphorylated CoA (CoA[OH]), aminoacylated with different amino acids carried out using TLC (Figs 7 and 8) showed the fol- lowing reaction yields: 50, 90, 32, 23 and 28% for Ala, Gly, Val, Phe and Lys, respectively. Tryptophan bound to CoA[OH] and acetyl-CoA[OH] with yields of 17 and 29%, respectively. The aminoacylation of CoA[OH] with tryptophan was essentially completed in 3 h and, after that, a slow decrease in product concentration was observed (Fig. 9A). The pressure dependence of CoA aminoacylation was linear (Fig. 9B). Discussion The preparation of tRNA charged with non-natural amino acid is a critical step in the synthesis of modified protein. All methods of preparing aa-tRNA charged with non-native amino acids are complicated and time- consuming [1–19]. In this study, we developed a general method of tRNA aminoacylation using any amino acid. 0 10 20 20 40 60 [ 14 C]Val incorporated [pmole] time [min] Fig. 5. In vitro transcription ⁄ translation assay. Analysis of 14 C-labelled Val incorporation into protein was carried out by scintil- lation counting of trichloroacetic acid-insoluble material. A B C D Fig. 4. HPLC-MS analysis of an aminoacylation reaction of yeast tRNA Phe with different amino acids carried out at 6 kbar for 5 h. Sig- nals at (A) m ⁄ z ¼ 415 and 437 correspond to [M + H] + and [M + Na] + of the Phe-Ade, respectively; (B) m ⁄ z ¼ 415 and 437 correspond to [M + H] + and [M + Na] + of the Phe-Ade, respectively, m ⁄ z ¼ 450 to [M + H] + of the p-Chl-Phe-Ade; (C) m ⁄ z ¼ 419 to [M + H] + of the Orn-Ade; (D) other signals correspond to the disintegration products Orn-Ade formula showing disintegration products [24,25]. M. Giel-Pietraszuk and J. Barciszewski Charging of tRNA with non-natural amino acids FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS 3017 For that purpose, we used a high-pressure technique. High pressure is currently used in many areas of bio- technology. Its mechanism of action includes the deci- sive role of water structure [26,27]. High pressure allows preparation of aa-tRNA in one step, without an enzyme or additional modification of the tRNA molecule. We have previously shown that tRNA Phe could be charged with Phe at high pressure without the need for a specific aa-tRNA synthetase. Aminoacylation occurred only at the 3¢-end of tRNA [22] and pressure-aminoacylated Phe–tRNA Phe was a normal substrate for peptide syn- thesis on the ribosome [23]. We tested our method using Cl-Tyr, Cl-Phe, l-Orn, d-Orn and DOPA. Aminoacyl–tRNA formation analysed on acidic PAGE showed small shift in 32 P-labelled tRNA (Fig. 1A,B). For comparison of the results, we used tRNA Val charged with [ 14 C]Val and [ 14 C]Leu, and analysed them using acidic PAGE visu- alized with fluorography (Fig. 1C). The results, estima- Fig. 7. TLC of CoA[OH] aminoacylation carried out at 6 kbar for 6 h in buffer: 0.1 M imidazole–HCl pH 6.6, 20 mM MgCl 2 ,10mM EDTA. The TLC plate was developed in butanol-1 ⁄ acetic acid ⁄ water (1:1:1 v ⁄ v ⁄ v) and visualized with a 0.1% ethanolic solution of ninhydrin. Lanes are as follows: (1) CoA[OH], (2) CoA[OH] + Ala, (3) Ala, (4) CoA[OH] + Gly, (5) Gly, (6) CoA[OH] + Phe, (7) Phe, (8) CoA[OH] + Val, (9) Val. Position of CoA[OH], in circles, was visual- ized under UV light. The arrows show the position of the products. Fig. 6. Aminoacylation of CoA[OH] with [ 14 C]Trp at high pressure. Graphic representation shows the distribution of radioactivity on a TLC plate: (d)CoA+[ 14 C]Trp, (n) CoA[OH] + [ 14 C]Trp, (m) acetyl- CoA + [ 14 C] Trp, (r) acetyl-CoA[OH] + [ 14 C]Trp. The signal at posi- tion 7 corresponds to Trp-CoA, at position 9 free Trp was detected. The reaction was analysed by TLC on cellulose with fluorescence indicator F254 and developed in an isobutyric acid solution, the TLC plate was cut into pieces as shown in the left-hand panel and the radioactivity was counted in scintillator solvent using Beckmann Apparatus LS 5000 TA. The position of the substrates was visual- ized under UV light. A B [cpm] [cpm] Lys [3’OH]CoA Lys-[3’OH]CoA Phe 0 1000 2000 3000 4000 60005000 0 1500 3000 4500 6000 90007500 [3’OH]CoA Phe-[3’OH]CoA Fig. 8. Aminoacylation of CoA[OH] with (A) [ 14 C]Lys and (B) [ 14 C]Phe. Reactions were carried out at 10 kbar pressure for 12 h. Aminoacyl-CoA[OH] was separated from free amino acids using TLC cellulose F. Diagrams show the distribution of radioactivity on the TLC plate measured in scintillator solvent. Charging of tRNA with non-natural amino acids M. Giel-Pietraszuk and J. Barciszewski 3018 FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS ted using the imagequant calculation, as well as by scintillation measurements of bands corresponding to [ 14 C-aa]tRNA Val , were similar. The yield of tRNA charging at high pressure with non-natural amino acids was between 2.5 and 9%, similar to data obtained for aminoacylation with natural amino acids (Table 1). Analysis of tRNA Phe and tRNA Val charging with a series of natural amino acids showed that the best yields were obtained for aromatic amino acids, but aminoacylation using amino acids with an aliphatic side chain was less efficient (Table 1). This can be explained by chemical activation of the carbonyl group by aromatic moiety. This observation is consistent with the suggestion that the aromatic ring of some amino acids is stabilized by association with an adenine ring. A similar effect was observed for Phe-AMP ester [28,29]. Synthesis of Trp–tRNA Phe at high pressure, measured as a function of tRNA Phe concentration, showed the highest yield after 30 min (Fig. 2A) [30]. Longer incubation decreased the amount of product (Fig. 2B). Aminoacylation of crude tRNA with Lys at high pressure was approximately 2.5 times higher compared with the enzymatic reaction, which was due to misacylation (Fig. 3, Table 1). To obtain more data on tRNA charging at the 3¢- end, we performed MS analysis of a product after lim- ited hydrolysis of aa-tRNA with RNaseA. MS analysis showed that the signals corresponded to Phe-Ade, p-Cl- Phe-Ade and l-Orn-Ade (Fig. 4). In the spectrum for l-Orn-Ade, in addition to the highest peak, other signals were observed. One of them, at m ⁄ z ¼ 331, suggests that the 2¢-OH group becomes esterified (Fig. 4D). The high-pressure aminoacylation occurred preferentially at the OH group of the terminal ribose ring. The 3¢-phosphate-free CoA molecule carried two potentially reactive sites, a thiol group and a 2¢-or3¢- OH group of ribose and, owing to this, we found it to be a very good substrate for high-pressure aminoacyla- tion (Figs 6–9). It has previously been reported that the thiol group of CoA can be acylated by AARS [31]. Furthermore, it was shown that AARSs are able to utilize noncognate amino acids in the aminoacylation of CoA, and in the acylation of mini helix of RNA [32]. The equilibrium of CoA acylation was shifted towards an aa-S-CoA formation [33]. In the case of high-pressure induced aminoacylation of CoA, we observed that the -OH group was acylated preferen- tially. The [HS]CoA acylation yield with Trp was 8%, whereas for [HS]CoA[OH] and [acetyl-S]CoA[OH] it was 17 and 29%, respectively. The detailed mechanism for the aminoacylation reaction of tRNA at high pressure remains unknown. Recently, we obtained new information about the con- formation of tRNA at elevated pressure [26]. It is known that high pressure lowers the pH of water. Because of this, the carbonyl group of the amino acid becomes protonated [27], which creates a positively charged carbon reactive towards nucleophilic attack by the ribose -OH group. Such acylation does not occur at normal pressure or at high pressure without imidaz- ole, which is a commonly occurring group in the active centres of many enzymes and plays an important role in electron transfer. Imidazole catalyses the aminoacyl transfer from adenylate anhydride to the 2¢OH groups along the RNA backbone [34]. The nitrogen of imidaz- ole attracts a proton from the hydroxyl group, which facilitates nucleophilic attack (Fig. 10). The entire pro- cess is induced by high pressure and does not proceed without it. We showed that high pressure influences the conformation of tRNA because of rearrangements in the structure of water [26]. These changes most probably create a binding pocket anchoring side chain in the amino acid, which brings the substrates closer. In summary, we have shown that the high pressure method could be used to prepare aa-tRNA in one step, A B Fig. 9. (A) Time-dependent aminoacylation of CoA with Trp at 10 kbar pressure. (B) Pressure-dependent aminoacylation of CoA[OH]. M. Giel-Pietraszuk and J. Barciszewski Charging of tRNA with non-natural amino acids FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS 3019 without any additional substrate modification. In addi- tion, we showed that the terminal OH group is acylat- ed preferentially. Experimental procedures Transfer RNA Yeast tRNA Phe , E. coli tRNA Val , CoA and acetyl-CoA were purchased from Sigma (St Louis, MO, USA). Crude tRNA from wheatgerm and AARS were purified by us [35,36]. Ala, Gly, Arg, Glu, His, Leu, Lys, Met, Phe, Tyr, Cl-Phe, Cl-Tyr, l-Orn and d-Orn were purchased from Sigma, and DOPA was from Behringwerke AG (Marburg, Germany). Uniformly radiolabelled amino acids [ 14 C]Phe (360 mCiÆmmol )1 ), [ 14 C]Gly (235 mCiÆmmol )1 ), [ 14 C]Arg (210 mCiÆmmol )1 ), [ 14 C]Tyr (250 mCiÆmmol )1 ), [ 14 C]Lys (235 mCiÆmmol )1 ), [ 14 C]Leu (230 mCiÆmmol )1 ) and [ 14 C]His (215 mCiÆmmol )1 ) were from UVVVR (Pra- gue, Czech Republic); l-[5- 3 H]Trp (28 CiÆmmol )1 ), G-[ 3 H]Glu (53 CiÆmmol )1 ), [ 3 H]Met (15 CiÆmmol )1 ) and [ 14 C]Val (45 mCiÆmmol )1 ) were purchased from Amersham Pharmacia (Little Chalfont, UK). Aminoacylation of tRNA at high pressure Aminoacylation of tRNAs was carried out at 6 kbar for 5 h in a mixture containing 1 lm [5¢- 32 P]tRNA Phe (1–5 lm tRNA Phe , 0.1 lm tRNA Val or 40–50 lg of crude tRNA), and 0.1 mm of nonlabelled amino acid (or radioactively labelled amino acid mixed with nonlabelled tRNA to obtain the desired specific activity per mmol), 0.1 m imidaz- ole–HCl buffer pH 6.6, 20 mm MgCl 2 , and 1 mm M EDTA. The solutions were pressured in 35 lLor1mL Teflon vessels placed in high-pressure cell (Unipress, War- saw, Poland). After pressuring, aa-tRNA was precipitated with ethanol, dried and dissolved in water. Detection of aa-tRNA using acidic/urea gels electrophoresis aa-tRNA was purified from free tRNA on a 6.5% poly- acrylamide gel (19:1 acrylamide ⁄ bisacrylamide v ⁄ v) with 8 m urea in 0.1 m sodium acetate buffer, pH 5.0. Electro- phoresis was carried out at 600 V until the Bromophenol blue reached the bottom of the gel [37]. [5¢- 32 P]-tRNA Phe was identified by autoradiography. Gels were exposed overnight at )70 °C to X-ray films in a cassette with an intensifying screen. Distribution of aa-tRNAs labelled with [ 14 C] was detected by fluorography. After elec- trophoresis the gel was treated with dimethylsulfoxide for 20 min in order to remove water, and soaked with 10% 2,5- diphenyloxazone (PPO) in dimethylsulfoxide for 2 h. Excess PPO was removed with water, and the gel was dried and exposed to X-ray film in a cassette with an intensifying screen at )70 °C for 48 h. Radioactivity was measured by scintilla- tion counting of individual gel slices [38,39]. Charging the efficiency of crude tRNA was monitored using the filter binding method [40]. Enzymatic aminoacylation of tRNA Five, 10, 15 and 20 lg of crude tRNA from wheatgerm were dissolved in a buffer containing 50 lL of 0.1 m Tris ⁄ HCl, pH 7.5, containing 0.01 m MgCl 2 ,4mm b-mercaptoethanol, 2 mm ATP and 2 lm [ 14 C]-lysine. After 20 min of incubation with crude plant AARS at 37 °C, the reaction mixture was spotted onto Whatmann 3 mm filter paper, washed once in 10% ice-cold trichloroacetic acid, twice in 5% trichloroacetic acid and, finally, with ethanol [41,42]. The radioactivity of aa-tRNA was measured by scintillation counting [43]. Valyl-tRNA Val from Thermus thermophilus (220 cpmÆpmole )1 ) used as a control was a gift from M. Sprinzl (Bayreuth University, Germany). Coupled in vitro transcription/translation The in vitro translation reaction was based on an E. coli S30 lysate (strain D10) and was performed as described previ- Fig. 10. Putative mechanism of tRNA aminoacylation at high pres- sure. In the first step, high pressure induces a lowering of pH and protonation of amino acid. A proton from the 2¢-or3¢-OH group is transferred to imidazole and a lone oxygen pair attack activates the carbon of the amino acid. Releasing of the high pressure causes dehydration of the intermediate product and aa-tRNA formation. Charging of tRNA with non-natural amino acids M. Giel-Pietraszuk and J. Barciszewski 3020 FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS ously [44]. Translation was carried out for 20, 40 and 80 min at 37 °C in a 240 mL reaction mixture containing the follow- ing components: 50 mm Hepes-KOH pH 7.6, 70 mm CH 3 COOK, 30 mm NH 4 Cl, 14 mm MgCl 2 , 0.1 mm EDTA, 0.02% NaN 3 ,40lg[ 14 C]Val-tRNAVal (18 500 c.p.m. ⁄ A 260 ), 0.2 mm of each amino acid (Val omitted), 1 mm each of ATP and GTP, 0.5 mm each of CTP and UTP, 30 mm phospho- enolpyruvate, 10 mm acetyl phosphate, 4% poly(ethylene glycol) 2000, 20 lgÆmL )1 rifampicin, 0.1 mgÆmL )1 total E. coli tRNA, 0.1 mm folinic acid, 100 unitsÆmL )1 RNase inhibitor, 26% (v ⁄ v) S30, 0.2–0.6 lm mRNA and 5 lm anti- ssrA oligonucleotide, 1 lgÆmL )1 leupeptin, 2 lgÆmL )1 aproti- nin, 1 lgÆmL )1 pepstatin, 500 unitsÆmL )1 T7 phage RNA polymerase, and 0.5–2 nm of a covalently closed plasmid. The incorporation of l-[ 14 C]Val into the synthesized proteins was determined by liquid scintillation counting of the trichlo- roacetic acid-insoluble material as described previously [44]. HPLC/ESI/MS analysis Ten micrograms of aa-tRNA obtained at high pressure were purified from free amino acid on a Sephadex G-75 column and subsequently digested at 37 °C for 15 min with 1 unit of RNaseA in 0.1 m imidazole–HCl buffer pH 6.6 containing 20 mm MgCl 2 and 1 mm EDTA. The hydrolysed tRNA was separated by HPLC ⁄ ESI ⁄ MS on Waters ⁄ Micromass ZQ mass spectrometer (Manchester, UK). A sample was injected using an autosampler onto a Waters Nova Pak C 18 RP-18 column (150 · 3.9 mm) at a flow rate of 0.5 mLÆmin )1 and eluted with a gradient of solvent A (95% water, 5% aceto- nitrile v ⁄ v) and solvent B (45% water, 50% methanol, 5% acetonitrile: v ⁄ v ⁄ v). A linear gradient from 100% A to 100% B in (A + B) within 10 min was applied, followed by iso- cratic elution for 20 min with 100% methanol. The source temperature was 120 °C and the desolvation temperature was 300 °C. Nitrogen was used as the nebulizing and desol- vation gas at flow rates of 100 and 600 LÆh )1 , respectively. Dephosphorylation of CoA and acetyl-CoA CoA or acetyl-CoA (4 mmol) was dephosphorylated for 1 h at 37 °C with 1 unit of nuclease P1 (Sigma) in 50 m m ammo- nium acetate buffer, pH 5.3, in a total volume of 10 lL. Dephosphorylated CoA (CoA[OH], acetyl-CoA, acetyl- CoA[OH]) was purified on a cellulose F plate (Merck, Darm- stadt, Germany) in a solvent containing isobutyric acid ⁄ 25% ammonium hydroxide ⁄ water (15:0.25:7.25 v ⁄ v ⁄ v). The spots corresponding to CoA[OH] and acetyl-CoA[OH] were visualized at UV light, scraped out and eluted with water. Aminoacylation of CoA at high pressure Aminoacylation of CoA was carried out at 10 kbar for 16 h in a mixture containing 1 mm CoA ( acetyl -CoA, CoA[OH], acetyl-CoA[OH]), 10 mm labelled amino acid in the buffer used for the tRNA aminoacylation. Trp-CoA purified on a TLC plate was dissolved in 20 lL of 0.1 m Tris ⁄ HCl pH 8.2 and left at room temperature for 14 h. Random deacylation was observed on TLC. TLC of aminoacyl-CoA The aminoacyl-CoA[OH] (or aminoacyl-acetyl-CoA[OH]) was analysed by TLC on cellulose F (Merck) in solvent containing isobutyric acid ⁄ ammonium hydroxide ⁄ water (15:0.25:7.25 v ⁄ v ⁄ v). The positions of free CoA were visual- ized under UV, amino acid with ninhydrin staining, [ 14 C]-labelled amino acids were located by measuring of the samples in scintillation counter (Beckman, Fullerton, CA, USA). For that purpose, the TLC plate was cut into pieces and the amount of radioactivity was measured using scintilla- tion counting [43]. Each reaction was repeated five times and the per cent yield of aminoacylation and standard deviation were calculated based on five independent measurements. Acknowledgements We thank Ms Sylwia Dolecka and Ms Ewa Powalska for their assistance in laboratory work. We thank also Prof Dr Volker A. Erdmann and Dr Torsten Lamla from the Free University in Berlin for help in carrying out the transcription ⁄ translation assay and Prof Math- ias Sprinzl from Bayreuth University for providing us with [ 14 C]Val-tRNA Val . References 1 Hendrickson TL, Crecy-Lagard V & Schimmel P (2004) Incorporation of non-natural amino acids into proteins. Annu Rev Biochem 73, 147–176. 2 Abdel-Rahman HM, Al-Karamany GS, El-Koussi NA, Youssef AF & Kiso Y (2002) HIV protease inhibitors: peptidomimetic drugs and future perspectives. Curr Med Chem 9, 1905–1922. 3 Kiso Y, Matsumoto H, Mizumoto S, Kimura T, Fujiw- ara Y & Akaji K (1999) Small dipeptide-based HIV protease inhibitors containing the hydroxymethylcarbo- nyl isoster as an ideal transition-state mimic. Biopoly- mers 51, 51–58. 4 Budisa N, Minks C, Medrano FJ, Lutz J, Huber R & Moroder L (1998) Residue-specific bioincorporation of non-natural, biologically active amino acids into pro- teins as possible drug carriers: structure and stability of the per-thiaproline mutant of annexin V. Proc Nat Acad Sc USA 95, 455–459. 5 Tang Y, Ghirlanda G, Vaidehi N, Kua J, Mainz DT, Goddard WA III, DeGrado WF & Tirrell DA (2001) M. Giel-Pietraszuk and J. Barciszewski Charging of tRNA with non-natural amino acids FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS 3021 Stabilization of coiled-coil peptide domains by introduction of trifluoroleucine. Biochemistry 40, 2790– 2796. 6 Rothschild KJ & Gite S (1999) tRNA-mediated protein engineering. Cur Opin Biotech 10, 64–70. 7 Krieg UC, Walter P & Johnson AE (1986) Photocross- linking of the signal sequence of nascent preprolactin to the 54-kilodalton polypeptide of the signal recognition particle. Proc Natl Acad Sci USA 83, 8604–8608. 8 Krieg UC, Johnson AE & Walter P (1989) Protein translocation across the endoplasmic reticulum mem- brane: identification by photocross-linking of a 39-kD integral membrane glycoprotein as part of a putative translocation tunnel. J Cell Biol 109, 2033–2043. 9 Thrift RN, Andrews DW, Walter P & Johnson AE (1991) A nascent membrane protein is located adjacent to ER membrane proteins throughout its integration and translation. J Cell Biol 112, 809–821. 10 Dawson PE, Muir TW, Clark-Lewis I & Kent SB (1994) Synthesis of proteins by native chemical ligation. Science 266, 776–779. 11 Noren CJ, Anthony-Cahill SJ, Griffith MC & Schultz PG (1989) A general method for site-specific incorpora- tion of unnatural amino acids into proteins. Science 244, 182–188. 12 Anthony-Cahill SJ, Griffith MC, Noren CJ, Suich DJ & Schultz PG (1989) Site-specific mutagenesis with unna- tural amino acids. Trends Biochem Sci 14, 400–403. 13 Heckler TG, Chang LH, Zama Y, Naka T, Chorghade MS & Hecht SM (1984) T4 RNA ligase mediated pre- paration of novel ’chemically misacylated’ tRNAPheS. Biochemistry 23, 1468–1473. 14 Resto E, Iida A, Van Cleve MD & Hecht SM (1992) Amplification of protein expression in a cell free system. Nucleic Acids Res 20, 5979–5983. 15 Hagen MD, Scalfi-Happ C, Happ E & Chladek S (1987) New methodology for the synthesis of 2¢(3¢)-O- aminoacyl oligoribonucleotides related to the 3¢-termi- nus of aa-tRNA. Nucleic Acids Symp Series 18, 285– 288. 16 Lohse PA & Szostak JW (1996) Ribozyme-catalysed amino-acid transfer reactions. Nature 381, 442–444. 17 Lee N, Bessho Y, Wei K, Szostak JW & Suga H (2000) Ribozyme-catalyzed tRNA aminoacylation. Nat Struct Biol 1, 28–33. 18 Liu DR, Magliery TJ, Pastrnak M & Schultz PG (1997) Engineering a tRNA and aminoacyl-tRNA synthase for the site-specific incorporation of unnatural amino acids into proteins in vivo. Proc Natl Acad Sci USA 94, 10092–10097. 19 Wang L & Schultz PG (2001) A general approach for the generation of orthogonal tRNAs. Chem Biol 9, 883– 890. 20 Hohsaka T, Ashizuka Y & Sisido M (1999) Incorpora- tion of two nonnatural amino acids into proteins through extension of the genetic code. Nucleic Acids Symp Series 42, 79–80. 21 Hohsaka T, Ashizuka Y, Murakami H & Sisido M (2001) Five-base codons for incorporation of nonnatural amino acids into proteins. Nucleic Acids Res 29, 3646– 3651. 22 Krzy_zaniak A, Barciszewski JSa, an ´ ski P & Jurczak J (1994) The non-enzymatic specific aminoacylation of transfer RNA at high pressure. Int J Biol Macromol 16, 153–158. 23 Krzy_zaniak A, Salan ´ ski P, Twardowski T, Jurczak J & Barciszewski J (1998) tRNA aminoacylated at high pres- sure is a correct substrate for protein biosynthesis. Bio- chem Mol Biol Int 45, 489–500. 24 Harvey DJ (2000) Collision-induced fragmentation of underivatized N-linked carbohydrates ionized by elec- trospray. J Mass Spektr 35, 1178–1190. 25 Barciszewski J, Siboska G, Pedersen BO, Clark BFC & Rattan S (1996) Evidence for presence of kinetin in DNA and cell extracts. FEBS Lett 393, 197–200. 26 Giel-Pietraszuk M & Barciszewski J (2005) A nature of conformational changes of yeast tRNAPhe. High hydrostatic pressure effects. Int J Biol Macromol 37 , 109–114. 27 Giel-Pietraszuk M, Gdaniec Z, Brukwicki T & Barcis- zewski J (2006) Molecular mechanism of high pressure action on lupanine. J Mol Struct, in press. 28 Wickramasinghe NSMD, Staves MP & Lacey JC Jr (1991) Stereoselective, nonenzymatic, intramolecular transfer of amino acids. Biochemistry 30, 2768–2772. 29 Lacey JC Jr, Mullins DW & Watkins CL (1986) Alipha- tic amino acid side chains associate with the ‘face’ of the adenine ring. J Biomol Struct Dyn 3, 783–793. 30 Bonnet J & Ebel JP (1972) Interpretation of incomplete reactions in tRNA aminoacylation. Aminoacylation of yeast tRNA Val with yeast valyl-tRNA synthetase. Eur J Biochem 31, 335–344. 31 Buerle T & Pichersky E (2002) Enzymatic synthesis and purification of aromatic coenzyme A esters. Anal Chem 302, 305–312. 32 Jakubowski H (1998) Aminoacylation of coenzyme A and pantetheine by aminoacyl-tRNA synthetases: poss- ible link between noncoded and coded peptide synthesis. Biochemistry 37, 5147–5153. 33 Jakubowski H (2000) Amino acid selectivity in the ami- noacylation of coenzyme A and RNA minihelices by aminoacyl-tRNA synthetases. J Biol Chem 275, 34845– 34848. 34 Weber AL & Lacey JC Jr (1975) Aminoacyl transfer from adenylate anhydride to polyribonucleotides. J Mol Evol 6, 309–320. 35 Barciszewska M, Joachimiak A & Barciszewski J (1989) The initiator transfer ribonucleic acids from yellow lupin seeds. Correction of the nucleotide sequence and crystallisation. Phytochemistry 28, 2039–2043. Charging of tRNA with non-natural amino acids M. Giel-Pietraszuk and J. Barciszewski 3022 FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS 36 Pulikowska J, Barciszewska M, Barciszewski J, Joachi- miak A, Rafalski AJ & Twardowski T (1979) Effect of elastase on elongation factor 1 from wheat germ. Biochem Biophys Res Commun 91, 1011–1017. 37 Vershney U, Lee Ch, P & RajBhandary UL (1991) Direct analysis of aminoacylation levels of tRNAs in vitro. J Biol Chem 266, 24712–24717. 38 Bonner WM & Laskey RA (1974) A film detection method for tritium-labelled proteins and nucleic acids in polyacrylamide gels. Eur J Biochem 46, 83–88. 39 Laskey RA & Mills AD (1975) Quantitative film detec- tion of 3 H and 14 C in polyacrylamide gels fluorography. Eur J Biochem 56, 335–341. 40 Draper DE, Deckman I & Vartikar J (1988) Physical studies of ribosomal protein – RNA interactions. Methods Enzymol 164, 203–209. 41 Joachimiak A, Barciszewski J, Twardowski T, Bar- ciszewska M & Wiewio ´ rowski M (1978) Purification and properties of methionyl-tRNA-synthetase from yellow lupine seeds. FEBS Lett 93, 51–54. 42 Gamian A, Krzyzaniak A, Barciszewska MZ, Gaw- ronska I & Barciszewski J (1991) Specific incorporation of glycine into bacterial lipopolysaccharide. Novel func- tion of specific transfer ribonucleic acids. Nucleic Acids Res 19, 6021–6025. 43 Kinjo M, Ishigami M, Hasegawa T & Nagano K (1984) Differential coupling efficiency of chemically activated amino acid to tRNA. J Mol Evol 20, 59–65. 44 Lamla T, Stiege S & Erdman VA (2002) An improved protein bioreactor efficient product isolation during in vitro protein biosynthesis via affinity tag. Mol Cel Proteom 1, 466–471. M. Giel-Pietraszuk and J. Barciszewski Charging of tRNA with non-natural amino acids FEBS Journal 273 (2006) 3014–3023 ª 2006 The Authors Journal compilation ª 2006 FEBS 3023 . at high pressure. Applica- tion of MS provides evidence that charging occurs at the hydroxyl group of the 3¢-end ribose. Results TRNA Phe aminoacylation with non-natural amino acids at high pressure Aminoacylation. high pressure with non-natural amino acids was between 2.5 and 9%, similar to data obtained for aminoacylation with natural amino acids (Table 1). Analysis of tRNA Phe and tRNA Val charging with. method of tRNA aminoacylation with natural, as well as non-natural, amino acids at high pressure. Such specific and noncognate tRNAs can be used as valuable substrates for protein engineering. Aminoacylation