Interpretation of incomplete aminoacylation of yeast tRNA-Phe: evidence for a phenylalanyl-tRNA synthetase catalyzed transconformation of tRNA-Phe

Aminoacyl-tRNA synthesis: divergent routes to a common goal

Aminoacyl-tRNAs are key components in protein synthesis. They are formed directly by correct acylation of tRNA (by aminoacyl-tRNA synthetases) or indirectly by tRNA-dependent transformation of misacylated tRNAs. The accuracy of aminoacyl-tRNA synthesis is enhanced by a number of further protein-RNA or protein-protein interactions, some of which are restricted to Archaea, and might reflect adaptation mechanisms to diverse conditions.

An example of non-conservation of oligomeric structure in prokaryotic aminoacyl-tRNA synthetases. Biochemical and structural properties of glycyl-tRNA synthetase from Thermus thermophilus

Glycyl-tRNA synthetase (Gly-tRNA synthetase) from Thermus thermophilus was purified to homogeneity and with high yield using a five-step purification procedure in amounts sufficient to solve its crystallographic structure [Logan, D.T., Mazauric, M.-H., Kern, D. & Moras, D. (1995) EMBO J. 14, 4156-4167]. Molecular-mass determinations of the native and denatured protein indicate an oligomeric structure of the alpha 2 type consistent with that found for eukaryotic Gly-tRNA synthetases (yeast and Bombyx mori), but different from that of Gly-tRNA synthetases from mesophilic prokaryotes (Escherichia coli and Bacillus brevis) which are alpha 2 beta 2 tetramers. N-terminal sequencing of the polypeptide chain reveals significant identity, reaching 50% with those of the eukaryotic enzymes (B. mori, Homo sapiens, yeast and Caenorhabditis elegans) but no significant identity was found with both alpha and beta chains of the prokaryotic enzymes (E. coli, Haemophilus influenzae and Coxiella burnetii) albeit the enzyme is deprived of the N-terminal extension characterizing eukaryotic synthetases. Thus, the thermophilic Gly-tRNA synthetase combines strong structural homologies of eukaryotic Gly-tRNA synthetases with a feature of prokaryotic synthetases. Heat-stability measurements show that this synthetase keeps its ATP-PPi exchange and aminoacylation activities up to 70 degrees C. Glycyladenylate strongly protects the enzyme against thermal inactivation at higher temperatures. Unexpectedly, tRNA(Gly) does not induce protection. Cross-aminoacylations reveal that the thermophilic Gly-tRNA synthetase charges heterologous E. coli tRNA(gly(GCC)) and tRNA(Gly(GCC)) and yeast tRNA(Gly(GCC)) as efficiently as T. thermophilus tRNA(Gly). All these aminoacylation reactions are characterized by similar activation energies as deduced from Arrhenius plots. Therefore, contrary to the E. coli and H. sapiens Gly-tRNA synthetases, the prokaryotic thermophilic enzyme does not possess a strict species specificity. The results are discussed in the context of the three-dimensional structure of the synthetase and in the view of the particular evolution of the glycinylation systems.

Two kinetically distinct tRNAile isoacceptors in Escherichia coli C6

The isoleucine acceptance of tRNA from Escherichia coli C6 was previously shown to be influenced by the synthetase level (Marashi, F. and Harris, C.L. 1977. Biochim. Biophys. Acta 477, 84-88). We show here that the increased acceptance observed at higher enzyme levels is accompanied by an increase in the aminoacylation of one tRNAile species. Hence, tRNAile, a minor species at low enzyme levels, is a major isoacceptor after full aminoacylation. The two major isoleucine species have been purified using a combination of BD-cellulose, DEAE-Sephadex A-50 and methylated albumin kieselguhr chromatography. tRNAile (1511 pmoles ile/A260 of tRNA) was found to be slowly acylated, with 2a Vmax one-seventh that observed with tRNAil3le (1475 pmoles ile/A260). Two-dimensional TLC analysis of RNase T2 digests revealed differences in nucleotide content between the purified tRNAs. These results are discussed in terms of the presence of slow and fast tRNAile species in E. coli.

Incomplete aminoacylation of tRNALeu catalyzed in vitro by leucyl-tRNA synthetase from Escherichia coli B

The extent of esterification of [14C] leucine into Escherichia coli B tRNALeu apparently depends on the concentration of leucyl-tRNA synthetase. The effect is more pronounced at pH 9.0 than at pH 7.4. When reciprocals of leucyl-tRNA concentration at plateau [aa-tRNA]-1 are plotted against reciprocals of initial velocities vo-1 of aminoacylations a straight line is obtained with a slope equal to the rate constant of non-enzymatic deacylation of leucyl-tRNA. Factors which change the stability of leucyl-tRNA, e.g. pH and temperature, also change the shape of the function [aa-tRNA]-1 vs. vo-1. The data are consistent with the idea that the rate constant of spontaneous deacylation of aminoacyl-tRNA is the factor which accounts for the dependence of the level of aminoacylation on initial velocity of aminoacylation.

Modification of L-isoleucyl-tRNA synthetase with L-isoleucyl-bromomethyl ketone. The effect of the catalytic steps

The rapidly reacting cysteine-sulfhydryl group of L-isoleucyl-tRNA synthetase has been specifically alkylated with L-isoleucyl-bromomethyl ketone [Rainey, P., Holler, E. & Kula, M.-R. (1976) Eur. J. Biochem. 63, 419-426]. We have now investigated the catalytic and substrate binding properties of the modified protein by radioactive and fluorescence techniques. The rate constants for the transfer of AMP and isoleucine from the protein - adenylate complex to form ATP or Ile-tRNAIle were only 3% of those for native enzyme, whereas the rate constant for the formation of adenylate was essentially unchanged. The tendency to form synthetase - substrate complexes remained almost unchanged with the exception of L-isoleucine which exhibited a 20-fold reduction. Similarly, complex formation of L-isoleucinol together with its synergistic coupling to complex formation of ATP was partially inhibited. The results rule out the essential participation of the rapidly alkylatable cysteine-sulfhydryl group during catalysis.

Isoleucyl-tRNA synthetase inactivation and the extent of aminoacylation of tRNAIle from Escherichia coli

A difference in isoleucine acceptance between normal and sulfur-deficient tRNA from Escherichia coli C6 (rel-, met-, cys-) was eliminated when more isoleucyl-tRNA synthetase was added at the reaction plateau. Enzymatic deacylation was similar for both tRNAs. These results suggest that enzyme inactivation caused a premature reaction plateau which was not predicted by the rates of acylation and deacylation.

Interpretation of tRNA-mischarging kinetics

Incorrect tRNA aminoacylation reactions are characterized by very slow reaction rates and by the fact that in most cases they are incomplete. In a previous study some of us explained the incompleteness of the correct aminoacylation reactions of tRNA, which can be encountered under certain experimental conditions (for instance low enzyme concentration or high ionic strength) by an equilibrium between the aminoacylation and the deacylation reactions [J. Bonnet and J.P. Ebel (1972) Eur. J.Biochem. 31, 335-344]. In the present report we bring evidence that the incorrect valylation of yeast tRNAfMet by yeast valyl-tRNA synthetase studied under standard experimental conditions, can also be described by a kinetic rate law including the rate equations of the aminoacylation and of the various deacylation reactions. In particular we show that the incomplete mischarging plateaus reflect the existence of an equilibrium between the valylation reaction on the one hand and the spontaneous and enzymic deacylation of valyl-tRNAfMet and the reverse of the valylation reaction on the other hand. However, when the valyl-tRNA synthetase concentration is not very high the reverse reaction of the amino-acylation does not play a predominant part in the establishment of the plateau. These interpretations have been extended to other mischarging systems: valylation of yeast tRNAPhE by yeast valyl-tRNA synthetase and mischarging of tRNAfMet and tRNA2Val from yeast by yeast phenylalanyl-tRNA synthetase. Unusual mischarging kinetics have been discussed. Furthermore, and as in correct systems, we found that during the mischarging of tRNAfMet one ATP is hydrolyzed per tRNA charged with valine. We conclude that the correct and the incorrect amino-acylation of tRNA behave kinetically in a similar way.

Catalytic activation of transfer ribonucleic acid by a mammalian protein

A tRNA activator has been isolated from mammalian organs which increases the capability of tRNA to accept certain amino acids through the action of mammalian aminoacyl-tRNA synthetases. This activity may be separated from the aminoacyl-tRNA synthetases for isoleucine, lysine, serine, and methionine by fractionation of liver or pancreas cytosol with ammonium sulfate or by chromatography over Sephadex G-200. The tRNA activating material is nondialyzable and is destroyed by trypsin or short heating. It acts catalytically. A molecular weight of approximately 45,000 was obtained by chromatography of tRNA activator on a calibrated Sephadex G-150 column. Activator increases acceptance of yeast tRNA for the amino acids isoleucine, leucine, lysine, serine, and methionine. It shows higher activity on liver tRNAMet f, tRNAMet m, and tRNALys than on unfractionated liver tRNA. Removal of protein from mammalian tRNA by extra phenol extractions, chromatography, or proteinase treatment increases its response to activator.

Proton nuclear magnetic resonance of spin-labeled Escherichia coli tRNAf1MET

Thiouridine at position 8 (s4U8) of tRNAf1Met was spin-labeled with the nitroxide free radical, N-(1-oxyl-2,2,5,5-Tetramethyl-3-pyrrolidinyl) bromacetamide, for proton nuclear magnetic resonance spectroscopic studies. The well-resolved methyl peak of ribothymidine is unperturbed, but the peak tentatively assigned to the C-5 methylene group of dihydrouridine is considerably broadened in spin-labeled tRNAf1Met. Of the approximately 27 slowly exchanging protons observed in the region between 11 and 15 ppm downfield from 4,4-dimethyl-4-silapentane-1-sulfonic acid, the equivalent of about five protons apparently disappeared in spin-labeled tRNAf1Met. The well-resolved single proton at 14.8 ppm was missing not only in the paramagnetic species, but also in the diamagnetic reduced form of spin-labeled tRNAf1Met, and was unequivocally identified as a hydrogen bond involving s4U8 by comparison of several forms of tRNAf1Met specifically modified at s4U. Evidence that the perturbation of a second single proton resonance at 14.6 ppm (shift and broadening) is coupled to the loss of a tertiary hydrogen bond involving residue 8, arises from the same modified forms. The resolved resonances in the methyl and N-H regions, particularly the resonance at 14.6 ppm as well as the four N-bonded proton resonances at higher field which are broadened solely due to their proximity to the unpaired electron of the spin label, provide specific indicators of the geometry of tRNAf1Met structure in solution. Their observability by nuclear magnetic resonance spectroscopy opens up the possibility of monitoring distance changes among the base residues of spin-labeled tRNAf1Met upon its interaction with aminoacyl-tRNA synthetase and other enzymes.