Dramatic events in ciliate evolution: alteration of UAA and UAG termination codons to glutamine codons due to anticodon mutations in two Tetrahymena tRNAs

The three major glutamine tRNAs of Tetrahymena thermophila were isolated and their nucleotide sequences determined by post-labeling techniques. Two of these tRNAs^Gln show unusual codon recognition: a previously isolated tRNA^Gln_UmUA and a second species with CUA in the anticodon (tRNA^Gln_CUA). These two tRNAs recognize two of the three termination codons on natural mRNAs in a reticulocyte system. tRNA^Gln_UmUA reads the UAA codon of α-globin mRNA and the UAG codon of tobacco mosaic virus (TMV) RNA, whereas tRNA^Gln_CUA recognizes only UAG. This indicates that Tetrahymena uses UAA and UAG as glutamine codons and that UGA may be the only functional termination codon. A notable feature of these two tRNAs^Gln is their unusually strong readthrough efficiency, e.g. purified tRNA^Gln_CUA achieves complete readthrough over the UAG stop codon of TMV RNA. The third major tRNA^Gln of Tetrahymena has a UmUG anticodon and presumably reads the two normal glutamine codons CAA and CAG. The sequence homology between tRNA^Gln_UmUG and tRNA^Gln_UmUA is 81%, whereas that between tRNA^Gln_CUA and tRNA^Gln_UmUA is 95%, indicating that the two unusual tRNAs^Gln evolved from the normal tRNA^Gln early in ciliate evolution. Possible events leading to an altered genetic code in ciliates are discussed.     

Glutamine is incorporated at the nonsense codons UAG and UAA in a suppressor-free Escherichia coli strain

Readthrough of the nonsense codons UAG, UAA, and UGA is seen in Escherichia coli strains lacking tRNA suppressors. Earlier results indicate that UGA is miscoded by tRNA(Trp). It has also been shown that tRNA(Tyr) and tRNA(Gln) are involved in UAG and UAA decoding in several eukaryotic viruses as well as in yeast. Here we have investigated which amino acid(s) is inserted in response to the nonsense codons UAG and UAA in E. coli. To do this, the stop codon in question was introduced into the staphylococcal protein A gene. Protein A binds to IgG, which facilitates purification of the readthrough product. We have shown that the stop codons UAG and UAA direct insertion of glutamine, indicating that tRNA(Gln) can read the two codons. We have also confirmed that tryptophan is inserted in response to UGA, suggesting that it is read by tRNA(Trp).     

Yeast suppressors of UAA and UAG nonsense codons work efficiently in vitro via tRNA

A cell-free protein-synthesizing system, containing an S-100 fraction from yeast, ribosomal subunits from Krebs ascites cells, and ribosome initiation factors from rabbit reticulocytes, translates yeast, adenovirus, and rabbit globin messenger RNAs and the RNA from bacteriophage Qβ. An amber mutation in the Qβ synthetase gene is suppressed in vitro if the S-100 fraction is from yeast strains carrying amber suppressor mutations. Suppressor SUP6-2 gives 16% suppression, and the recessive lethal suppressor RL-1 gives 50% suppression. Extracts from strain FM6, which has the ochre suppressor SUP4-1, give a longer protein product from the normal synthetase gene of Qβ with an efficiency of 63%. This implies that UAA is the terminator for the synthetase gene, and that synthesis of this read through protein can be used as an assay for ochre suppression. Suppression in each of these cases is mediated by tRNA, since purified tRNA is the only fraction from suppressing strains that is required in an otherwise nonsuppressing cell-free system.     

Replacement and insertion of nucleotides at the anticodon loop of E. coli tRNAMetf by ligation of chemically synthesized ribooligonucleotides.

Insertion of the four major nucleotides at the 5'-side of the anticodon triplet of E. coli tRNAMetf was performed by joining of the half molecules obtained by limited digestion with RNase A and the chemically synthesized tetranucleotide pN-C-A-U using RNA ligase. Insertion of U-U at the 5'-side or A and A-A at the 3'-side of the anticodon were also performed using U-U-C-A-U, C-A-U-A and C-A-U-A-A. The constant U next to the 5'-side of the anticodon was replaced with A and C by ligation of A-C-A-U and C-C-A-U to the 5'-half molecule which had been treated with periodate plus lysine, followed by joining to the 3'-half. These modified tRNAs were tested for their ability to accept methionine with the methionyl-tRNA synthetase of E. coli. The affinity of these analogs for the synthetase decreased more extensively when the insertion was at the 3'-side of the anticodon triplet. Insertion of mononucleotides at the 5'-side or replacement of the constant U next to the 5'-side of the anticodon did not affect aminoacylation drastically. This may mean that the 3'-side of the anticodon loop of tRNA is one of the major recognition sites for the methionyl-tRNA synthetase.     

Modification of the amino acid acceptor stem of E. coli tRNAMetf by ligation of chemically synthesized ribooligonucleotides

The single-stranded region of the amino acid acceptor stem corresponding to the 3'-end of E. coli tRNAMetf was replaced by ligation of chemically synthesized ribooligonucleotides, in order to change the length of the single-stranded CCA terminus. The chemically synthesized ribooligomers, CCA, ACCA, AACCA and CAACCA, were ligated to nuclease-treated E. coli tRNAMetf, which lacked the ACCA sequence at the 3'-end. The methionine acceptor activities of these modified tRNAs were examined using E. coli methionyl-tRNA synthetase. Ligation of the chemically synthesized pentamer (AACCA) to the acceptor terminus restored the methionine acceptor activity, whereas ligation of the hexamer (CAACCA) or trimer (CCA) to the acceptor terminus did not Modification of the acceptor terminus had no effect on the formylation of accepted methionine.     

Mutational alterations of tryptophan-specific transfer RNA that generate translation suppressors of the UAA, UAG and UGA nonsense codons

The recessive lethal amber suppressor su⁺7(UAG-1) in Escherichia coli inserts glutamine in response to the UAG codon. The genetic analysis presented in this paper shows that the su^?7 precursor allele can give rise to suppressors of the UGA codon as well as of the UAG codon. This observation suggests that the su^?7 gene normally codes for transfer RNA^Trp, a tRNA whose anticodon can be modified by single base changes to forms that can translate either UAG or UGA. The chemical findings presented in the accompanying paper (Yaniv et al., 1974) are wholly in accord with this interpretation. Thus, a single base substitution in the anticodon sequence of a tRNA can affect both the coding specificity of the molecule and also the amino acid acceptor specificity.     

Base substitutions in the wobble position of the anticodon inhibit aminoacylation of E. coli tRNAfMet by E. coli Met-tRNA synthetase

Derivatives of E. coli tRNAfMet containing single base substitutions at the wobble position of the anticodon have been enzymatically synthesized in vitro. The procedure involves excision of the normal anticodon, CAU, by limited digestion of intact tRNAfMet with RNase A. RNA ligase is then used to join each of four trinucleotides, NAU, to the 5' half molecule and to subsequently link the 3' and modified 5' fragments to regenerate the anticodon loop. Synthesis of intact tRNAfMet containing the anticodon CAU by this procedure yields a product which is indistinguishable from native tRNAfMet with respect to its ability to be aminoacylated by E. coli methionyl-tRNA synthetase. Substitution of any other nucleotide at the wobble position of tRNAfMet drastically impairs the ability of the synthetase to recognize the tRNA. Measurement of methionine acceptance in the presence of high concentrations of pure enzyme has established that the rate of aminoacylation of the AAU, GAU and UAU anticodon derivatives of tRNAfMet is four to five orders of magnitude slower than that of the native or synthesized tRNA containing C as the wobble base. In addition, the inactive tRNA derivatives fail to inhibit aminoacylation of normal tRNAfMet, indicating that they bind poorly to the enzyme. These results support a model involving direct interaction between Met-tRNA synthetase and the C in the wobble position during aminoacylation of tRNAfMet.     

Temperature jump relaxation studies on the interactions between transfer RNAs with complementary anticodons. The effect of modified bases adjacent to the anticodon triplet

We have used the temperature-jump relaxation technique to determine the kinetic and thermodynamic parameters for the association between the following tRNAs pairs having complementary anticodons: tRNA(Ser) with tRNA(Gly), tRNA(Cys) with tRNA(Ala) and tRNA(Trp) with tRNA(Pro). The anticodon sequence of E. coli tRNA(Ser), GGA, is complementary to the U*CC anticodon of E. coli tRNA(Gly(2] (where U* is a still unknown modified uridine base) and A37 is not modified in none of these two tRNAs. E. coli tRNA(Ala) has a VGC anticodon (V is 5-oxyacetic acid uridine) while tRNA(Cys) has the complementary GCA anticodon with a modified adenine on the 3' side, namely 2-methylthio N6-isopentenyl adenine (mS2i6A37) in E. Coli tRNA(Cys) and N6-isopentenyl adenine (i6A37) in yeast tRNA(Cys). The brewer yeast tRNA(Trp) (anticodon CmCA) differs from the wild type E. coli tRNA(Trp) (anticodon CCA) in several positions of the nucleotide sequence. Nevertheless, in the anticodon loop, only two interesting differences are present: A37 is not modified while C34 at the first anticodon position is modified into a ribose 2'-O methyl derivative (Cm). The corresponding complementary tRNA is E.coli tRNA(Pro) with the VGG anticodon. Our results indicate a dominant effect of the nature and sequence of the anticodon bases and their nearest neighbor in the anticodon loop (particularly at position 37 on the 3' side); no detectable influence of modifications in the other tRNA stems has been detected. We found a strong stabilizing effect of the methylthio group on i6A37 as compared to isopentenyl modification of the same residue. We have not been able so far to assess the effect of isopentenyl modification alone in comparison to unmodified A37. The results obtained with the complex yeast tRNA(Trp)-E.coli tRNA(Pro) also suggest that a modification of C34 to Cm34 does not significantly increase the stability of tRNA(Trp) association with its complementary anticodon in tRNA(Pro). The observations are discussed in the light of inter- and intra-strand stacking interactions among the anticodon triplets and with the purine base adjacent to them, and of possible biological implications.     

Stability of the unique anticodon loop conformation of E.coli tRNAfMet.

Initiator tRNAs have an anticodon loop conformation distinct from that of elongation tRNAs as detected by susceptibility to S1 nuclease. We now find the anticodon loop conformation of E. coli tRNAfMet to be stable under different salt conditions as detected by using S1 nuclease as a structural probe. In contrast, a conformational change is observed in the T- and D- loop of this tRNA in the absence of added Mg2+. This change can be suppressed by spermine. Even under those conditions effecting a change in T- and D- loop conformation, the anticodon loop does not change. This suggests that the conformational shift is controlled by Mg2+ and restricted to the D- and T- loop region only without affecting the anticodon domain. The use of S1 nuclease as a conformational probe requires the use of kinetic studies to determine the initial cleavage sites. Thus, the use of a strong inhibitor which immediately stops the action of this nuclease is necessary. ATP is shown to be such an inhibitor.     

Modification of specific lysine residues in E. coli methionyl-tRNA synthetase by crosslinking to E. coli formylmethionine tRNA

A protein affinity labeling derivative of E. coli tRNAfMet has been prepared which carries an average of one reactive side chain per molecule, distributed over four structural regions. Each side chain contains a disulfide bond capable of reaction with cysteine residues and an N-hydroxysuccinimide ester group capable of coupling to lysine epsilon-amino groups in proteins. Reaction of the modified tRNA with E. coli methionyl-tRNA synthetase leads to crosslinking only by reaction with lysine residues in the protein. Examination of the tRNA present in the crosslinked complex reveals that the enzyme is coupled to side chains attached to the 5' terminal nucleotide, the dihydrouridine loop, the anticodon and the CCA sequence. Digestion of the crosslinked enzyme with trypsin followed by peptide mapping reveals that the major crosslinking reactions occur at four specific lysine residues, with minor reaction at two additional sites. Native methionyl-tRNA synthetase contains 90 lysine residues, 45 in unique sequences of the dimeric alpha 2 enzyme. Crosslinking of the protein to different regions in tRNAfMet thus occurs with the high degree of selectivity necessary for use in determining the peptide sequences which are near specific nucleotide sequences of tRNA bound to the protein.     

DNA complementary to rabbit globin mRNA made by E. coli polymerase I

Incubation of liver microsomes with cytochrome $\text{b}{}_5$, purified after solubilization with detergents, caused an effective incorporation of the cytochrome into the microsomal membranes. The incorporated cytochrome was reducible by NADH and could not be removed by repeated washing with 0.3 M KCl or 10 mM EDTA. The incorporation was much more efficient at 37°C than at 0°C. Trypsin-solubilized cytochrome $\text{b}{}_5$, which lacks the hydrophobic tail of the native protein, could not be inserted into the membranes. These findings confirm the view that the hydrophobic tail of the cytochrome molecule is responsible for its tight binding to the microsomal membranes.     

The anticodon of the maize chloroplast gene for tRNA Leu UAA is split by a large intron.

The maize chloroplast gene encoding tRNA Leu UAA has been sequenced. It contains a 458 base pair intron between the first and second bases of the anticodon. The tRNA is 88 nucleotides long (the 3'-terminal CCA sequence included which, however, is not encoded by the gene) and differs in only four nucleotides (modified nucleotides are not considered) from the corresponding isoacceptor from bean chloroplasts. The unusual position of the intron in this maize chloroplast tRNA gene suggests a splicing model different from that generally accepted for eukaryotic split tRNA genes.     

Mutation in the D arm enables a suppressor with a CUA anticodon to read both amber and ochre codons in Escherichia coli

Su9 of Escherichia coli differs from tRNATrp by only a G to A transition in the D arm, yet has an enhanced ability to translate UGA by an unusual C X A wobble pairing. In order to examine the effects of this mutation on translation of the complementary and wobble codons in vivo, we constructed the gene for an amber (UAG) suppressing variant of Su9, trpT179, by making the additional nucleotide change required for an amber suppressor anticodon. The resultant suppressor tRNA, Su79, is a very strong amber suppressor. Furthermore, the D arm mutation enables Su79 to suppress ochre (UAA) codons by C X A wobble pairing. These data demonstrate that the effect of the D arm mutation on wobble pairing is not restricted to a CCA anticodon. The effect extends to the CUA anticodon of Su79, thereby creating a new type of ochre suppressor. The new coding activity of Su79 cannot be explained by alterations in the level of aminoacylation, steady-state tRNA concentration, or nucleotide modification. The A24 mutation could permit unorthodox wobble pairings by generally enhancing tRNA efficiency at all codons or by altering codon specificity.     

Investigation of the base-pairing structure of the anticodon hairpin from E. coli initiator tRNA by high-resolution nmr

The high-resolution nmr spectrum of the anticodon hairpin from E. coli tRNA^fMet has been obtained at a number of different temperatures. The positions of the resonances from interior Watson-Crick base pairs are well accounted for (within 0.1 ppm) by a semi-empirical ring current shift theory, but the terminal base pairs are susceptible to the exact orientation of adjacent bases in single-stranded regions. From a careful examination of the exact way in which resonances disappear at elevated temperatures, we conclude that melting in the nmr experiments occurs when the lifetime of a base pair is reduced to several milliseconds. On the basis of these experiments we are able to assign an nmr T_m to each individual base pair and these should be useful in interpreting the melting behavior of the intact molecule. An “extra” resonance is observed at ～11.3 ppm and, on the basis of its position and temperature sensitivity, it is tentatively assigned to the ring nitrogen proton of a “protected” U residue in the anticodon loop. A strong preference for stacking of a nonbase-paired A residue on an adjacent GC base pair is observed even at temperatures in excess of 52°C.