RNA uridylyltransferases

The terminal RNA uridylyltransferases (TUTases) catalyze transfer of UMP residues to the 3' hydroxyl group of RNA. These activities are widespread among eukaryotes and appear to be involved in a variety of RNA-processing pathways. Recent studies of RNA editing in trypanosomatids have provided the first insights into the biological functions of RNA uridylyltransferases, which had eluded biochemical identification despite 30-year-old evidence of such activities in mammals and plants. Comparative sequence analysis of trypanosomal TUTases and their homologs revealed by large-scale genomic projects demonstrates a significant level of biochemical and structural diversity between putative uridylyltransferases. The conserved catalytic domain has acquired additional protein modules and appears to have adapted to perform functionally distinct tasks of guided U-insertion into mRNA and constrained addition of an oligo[U] tail to guide RNAs. Here I discuss the current knowledge of this novel enzyme family and possible roles of RNA uridylylation in the regulation of gene expression.

Isolation and characterization of a U-specific 3'-5'-exonuclease from mitochondria of Leishmania tarentolae

We have purified a 3'-5'-exoribonuclease from mitochondrial extract of Leishmania tarentolae over 4000-fold through six column fractionations. This enzyme digested RNA in a distributive manner, showed a high level of specificity for 3'-terminal Us, and was blocked by a terminal dU; there was slight exonucleolytic activity on a 3'-terminal A or C but no activity on a 3'-terminal G residue. The enzyme preferred single-stranded 3'-oligo(U) overhangs and did not digest duplex RNA. Two other 3'-5'-exoribonuclease activities were also detected in the mitochondrial extract, one of which was stimulated by a 3'-phosphate and the other of which degraded RNAs with a 3'-OH to mononucleotides in a processive manner. The properties of the distributive U-specific 3'-5'-exoribonuclease suggest an involvement in the U-deletion RNA editing reaction that occurs in the mitochondrion of these cells.

The mitochondrial RNA ligase from Leishmania tarentolae can join RNA molecules bridged by a complementary RNA

A biochemical characterization was performed with a partially purified RNA ligase from isolated mitochondria of Leishmania tarentolae. This ligase has a K(m) of 25 +/- 0.75 nM and a V(max) of 1.0 x 10(-4) +/- 2.4 x 10(-4) nmol/min when ligating a nicked double-stranded RNA substrate. Ligation was negatively affected by a gap between the donor and acceptor nucleotides. The catalytic efficiency of the circularization of a single-stranded substrate was 5-fold less than that of the ligation of a nicked substrate. These properties of the mitochondrial RNA ligase are consistent with an expected in vivo role in the process of uridine insertion/deletion RNA editing, in which the mRNA cleavage fragments are bridged by a cognate guide RNA.

Are tRNAs imported into the mitochondria of kinetoplastid protozoa as 5'-extended precursors?

All mitochondrial tRNAs in kinetoplastid protozoa are encoded in nuclear DNA and transported into the mitochondrion (Simpson et al., Nucl Acids Res 1989;17:5427-5445; Hancock and Hajduk, J Biol Chem 1990;265:19208-19215). It has been proposed that tRNAs in these cells are imported into the mitochondrion as 5'-extended precursors which are processed by a mitochondrial RNase P-like activity (Hancock et al., J Biol Chem 1992;267:23963-23971). We have examined this hypothesis by cloning and sequencing primer extension products of mitochondrial tRNAs from Leishmania tarentolae and Trypanosoma brucei, and have found that these are derived from circularized mature tRNA molecules. We suggest that these molecules are produced by the endogenous RNA ligase activity (Bakalara et al., J Biol Chem 1989;264:18679-18686) either in vivo or during mitochondrial isolation. We did not obtain any evidence for the existence of high molecular weight precursors of mitochondrial tRNAs. This negative result is consistent with previous in vivo transfection studies with both L. tarentolae (Lima and Simpson, RNA 1996;2:429-440) and T. brucei (Hauser and Schneider, EMBO J 1995;14:4212-4220; Schneider et al., Mol Cell Biol 1994;14:2317-2322), in which mitochondrial targeting of plasmid-expressed tRNAs was independent of the presence of 5'-flanking sequences. We conclude that the hypothesis for 5'-extended tRNA precursors in kinetoplastid mitochondrial importation remains to be verified.

Structure and aminoacylation capacities of tRNA transcripts containing deoxyribonucleotides

The contribution of the ribose 2'-hydroxyls to RNA structure and function has been analyzed, but still remains controversial. In this work, we report the use of a mutant T7 RNA polymerase as a tool in RNA studies, applied to the aspartate and methionine tRNA aminoacylation systems from yeast. Our approach consists of determining the effect of substituting natural ribonucleotides by deoxyribonucleotides in RNA and, thereby, defining the subset of important 2'-hydroxyl groups. We show that deoxyribose-containing RNA can be folded in a global conformation similar to that of natural RNA. Melting curves of tRNAs, obtained by temperature-gradient gel electrophoresis, indicate that in deoxyribo-containing molecules, the thermal stability of the tertiary network drops down, whereas the stability of the secondary structure remains unaltered. Nuclease footprinting reveals a significant increase in the accessibility of both single- and double-stranded regions. As to the functionality of the deoxyribose-containing tRNAs, their in vitro aminoacylation efficiency indicates striking differential effects depending upon the nature of the substituted ribonucleotides. Strongest decrease in charging occurs for yeast initiator tRNA(Met) transcripts containing dG or dC residues and for yeast tRNA(Asp) transcripts with dU or dG. In the aspartate system, the decreased aminoacylation capacities can be correlated with the substitution of the ribose moieties of U11 and G27, disrupting two hydrogen bond contacts with the synthetase. Altogether, this suggests that specific 2'-hydroxyl groups in tRNAs can act as determinants specifying aminoacylation identity.

Importance of structural features for tRNA(Met) identity

We showed previously that the tRNA tertiary structure makes an important contribution to the identity of yeast tRNA(Met) (Senger B, Aphasizhev R, Walter P, Fasiolo F, 1995, J Mol Biol 249:45-58). To learn more about the role played by the tRNA framework, we analyzed the effect of some phosphodiester cleavages and 2'OH groups in tRNA binding and aminoacylation. The tRNA is inactivated provided the break occurs in the central core region responsible for the tertiary fold or in the anticodon stem/loop region. We also show that, for tRNA(Met) to bind, the anticodon loop, but not the anticodon stem, requires a ribosephosphate backbone. A tertiary mutant of yeast tRNA(Met) involving interactions from the D- and T-loop unique to the initiator species fails to be aminoacylated, but still binds to yeast methionyl-tRNA synthetase. In the presence of 10 mM MgCl2, the mutant transcript has a 3D fold significantly stabilized by about 30 degrees C over a wild-type transcript as deduced from the measure of their T(m) values. The k(cat) defect of the tRNA(Met) mutant may arise from a failure to overcome an increase of the free energetic cost of distorting the more stable tRNA structure and/or a tRNA based MetRS conformational change required for formation of transition state of aminoacylation.

Conservation in evolution for a small monomeric phenylalanyl-tRNA synthetase of the tRNA(Phe) recognition nucleotides and initial aminoacylation site

We previously showed that yeast mitochondrial phenylalanyl-tRNA synthetase (MSF protein) is evolutionarily distant to the cytoplasmic counterpart based on a high degree of divergence in protein sequence, molecular mass, and quaternary structure. Using yeast cytoplasmic tRNA(Phe) which is efficiently aminoacylated by MSF protein, we report here the tRNA(Phe) primary site of aminoacylation and the identity determinants for MSF protein. As for the cytoplasmic phenylalanyl-tRNA synthetase (Sampson, J. R., Di Renzo, A. B., Behlen, L. S., & Uhlenbeck, O. C. (1989) Science 243, 1363-1366), MSF protein recognizes nucleotides from the anticodon and the acceptor end including base A73 and, as shown here, adjacent G1-C72 base pair or at least C72 base. This indicates that the way of tRNA(Phe) binding for the two phenylalanine enzymes is conserved in evolution. However, tRNA(Phe) tertiary structure seems more critical for the interaction with the cytoplasmic enzyme than with MSF protein, and unlike cytoplasmic phenylalanyl-tRNA synthetase, the small size of the monomeric MSF protein probably does not allow contacts with residue 20 at the top corner of the L molecule. We also show that MSF protein preferentially aminoacylates the terminal 2'-OH group of tRNA(Phe) but with a catalytic efficiency for tRNA(Phe)-CC-3'-deoxyadenosine reduced 100-fold from that of native tRNA(Phe), suggesting a role of the terminal 3'-OH in catalysis. The loss is only 1.5-fold when tRNA(Phe)-CC-3'-deoxyadenosine is aminoacylated by yeast cytoplasmic PheRS (Sprinzl, M., & Cramer, F. (1973) Nature 245, 3-5), indicating mechanistic differences between the two PheRS's active sites for the amino acid transfer step.

The presence of a D-stem but not a T-stem is essential for triggering aminoacylation upon anticodon binding in yeast methionine tRNA

Dissection of the yeast cytoplasmic initiator tRNA(Met) into two helical domains, the T psi C acceptor and anticodon minihelices, failed to show anminoacylation and binding of the acceptor minihelix by the yeast methionyl-tRNA synthetase (MetRS) even in the presence of the anticodon minihelix. In contrast, based on the measure of the inhibition constant Ki, the anticodon minihelix carrying the methionine anticodon CAU is specifically bound to the synthetase and with an affinity comparable to that of the full-length tRNA. The yeast tRNA(Met) acceptor and anticodon minihelices were covalently linked using the central core sequences of either bovine mitochondrial tRNA(Ser) (AGY) lacking a D-stem or initiator tRNA(Met) from Caenorhabditis elegans lacking a T-stem. Based on modeling studies of analogous constructs performed by others, we assume that the folding and distance between the anticodon and acceptor ends of these hybrid tRNAs are identical to that of canonical tRNA. The three-quarter molecule, which includes the T-stem, has aminoacylation activity significantly more than an acceptor minihelix, while the acceptor stem/anticodon-D stem biloop has near wild-type aminoacylation activity. These results suggest that the high selectivity of the anticodon bases in tRNA(Met) depends upon the tRNA L-shape conformation and the presence of a D-arm. Protein contacts with the D-arm phosphate backbone are required for connecting anticodon recognition with the active site. These interactions probably contribute to fine tune the position of the acceptor end in the active site, allowing entry into the transition state of aminoacylation upon anticodon binding. The importance of an L structure for recognition of tRNA(Met) by yeast MetRS was also deduced from mutations of tertiary interactions known to play a general role in tRNA folding.

Random-splitting of tRNA transcripts as an approach for studying tRNA-protein interactions

Location of phosphodiester bonds essential for aminoacylation of bovine tRNA(Trp) was identified using a randomly cleaved transcript synthesized in vitro. It was found that cleavage of phosphodiester bonds after nucleotides in positions 21, 22, 36-38, 57-59, 62 and 64 were critical for aminoacylation capacity of tRNA(Trp)-transcript. These cleavage sites were located in the regions of tRNA molecule protected by the cognate synthetase against chemical modification and in the regions presumably outside the contact area as well. These results indicate that for maintenance of aminoacylation ability the intactness of the certain regions of the tRNA backbone structure is necessary. Random splitting of non-modified RNA with alkali followed by separation of active and inactive molecules and identification of cleavage sites developed in this work may become a general approach for studying the role of RNA covalent structure in its interaction with proteins.