Use of formylation defective initiator tRNAs in the study of significance of formylation in initiation of protein synthesis in Escherichia coli

Abstract

Initiation of protein synthesis is the major regulatory step in the translation of a messenger RNA. Initiation occurs with formylmethionine in eubacteria as opposed to methionine in archaea and eukaryotes. However, the requirement of this additional step of formylation in eubacteria is not completely understood.

A mismatch at the 1–72 position of the eubacterial initiator tRNAs is a major determinant for its recognition by formyltransferase. This mismatch is also important in preventing the binding of the initiator tRNA to EF-Tu, and in preventing wasteful hydrolysis of formylated initiator tRNA by peptidyl-tRNA hydrolase (PTH).

The initiator tRNA with CUA anticodon (U35A36 mutation) initiates protein synthesis from the chloramphenicol acetyltransferase (CAT) reporter mRNA containing UAG as the initiation codon. Using this in vivo assay system, a number of initiator tRNA mutants were characterized. Of these, the G72/U35A36 tRNA (extremely poor substrate for formyltransferase) and the G72G73/U35A36 tRNA (cannot be formylated in vitro) do not initiate protein synthesis.

We have used these formylation-defective initiator tRNAs to analyze the significance of formylation in the initiation of protein synthesis. The approach involved isolation and characterization of: (I) intragenic and (II) extragenic suppressors for these tRNAs.

I) Characterization of intragenic suppressor: A spontaneous intragenic suppressor carrying a C1 to T1 mutation (U1G72/U35A36) was isolated which rescued the G72/U35A36 tRNA. The U1G72/U35A36 tRNA was formylated in vivo. The presence of U1G72/U35A36 tRNA in formylated form explained its activity in initiation and highlighted the significance of formylation in initiation of protein synthesis. Biochemical analyses using U1G72/U35A36 tRNA showed that the mismatch at the top of the acceptor stem of the initiator tRNA, which was thus far thought to be the hallmark of the resistance of this tRNA against PTH, is not sufficient in conferring resistance to PTH. Both U1G72/U35A36 and the U35A36 initiator tRNA^fMet are substrates for PTH when aminoacylated with glutamine. However, when aminoacylated with methionine, the wild-type initiator tRNA showed complete resistance to PTH whereas U1G72 tRNA showed decreased hydrolysis by PTH. This clearly showed that not only the mismatch at 1:72 position, but also the amino acid attached to the tRNA is important in conferring resistance to hydrolysis by PTH.

Furthermore, we have studied the effect of overexpression of PTH and/or IF2 on the path followed by the initiator tRNA in protein synthesis. The PTH overproduction imparted elongation activity to the initiator tRNAs. However, simultaneous overproduction of IF2 annulled this enhancement in elongation activity. These studies show that the relative levels of various proteins that interact with the initiator tRNAs determine the fate of initiator tRNA in protein synthesis. These studies may provide an important clue to understand the dual function of a single tRNAMet in initiation and elongation in the mitochondria of various organisms.

II) Analysis of Extragenic Suppressors:

A) E. coli cells carrying the formylation-defective tRNA and the CAT^aml genes were plated on chloramphenicol-containing plates, directly or after treatment with MNNG. The extragenic suppressors were selected from the Cm^R colonies by use of a genetic screen. To avoid selection of revertants in the CAT or the tRNA genes, only those mutants of E. coli, plasmids from which did not confer Cm resistance to wild-type E. coli cells, were selected. E. coli mutants, thus selected, were cured of their resident plasmid(s). Subsequent to plasmid curing, all isolates showed Cm^s phenotype. Of these, only isolates in which Cm phenotype was restored upon transformation of the CAT^aml and the tRNA-containing plasmid were short-listed for further studies.

The four extragenic suppressors thus isolated rescued both the G72/U35A36 and G72G73/U35A36 tRNAs, suggesting that the extragenic suppressors were not strictly allele-specific. Further, overexpression of PTH in all these suppressors resulted in 2–3 fold decrease in initiation activity of G72/U35A36 and G72G73/U35A36 tRNAs. As both of these tRNAs possess a C1-G72 base pair, their formylated forms are substrates for PTH, these results suggested that the formylation-dependent initiation mechanism contributed to the initiation activities of G72/U35A36 and G72G73/U35A36 tRNAs in the suppressors. Detailed biochemical analyses ruled out mutations in aminoacyl-tRNA synthetase(s), formyltransferase, and PTH genes. Subsequently, genetic mapping experiments using P1 transductions localized the site of the suppressor mutation to be in the vicinity of the IF2 gene (71.5 min).

B) Formylation-independent suppressors: The balance of initiation activity seen in the presence of PTH overexpression in the suppressors described above could have been due to a formylation-independent mechanism. To investigate this phenomenon further, we isolated another class of suppressors. Isolation of these suppressors utilized constitutive overproduction of PTH to eliminate formylation-dependent initiation, as well as the G72G73/U35A36 tRNA, which cannot be formylated in vitro. Four suppressors were selected for further characterization. The initiation activity of these suppressors was 15–60% of the U35A36 tRNA, which is 3–4 fold higher than that shown by the suppressors described earlier. Transductions using P1 revealed that the suppressor mutation is located neither in the IF2 locus (71.5 min) nor in the formyltransferase gene locus (74 min).

III) Initiation with formylation-defective tRNAs with or without rT modification at position 54: Earlier studies using Streptococcus faecalis and E. coli suggested that absence of rT modification at position 54 in the initiator tRNAs could facilitate formylation-independent initiation. However, technical difficulties did not permit detailed analysis of those mutants at that time. The availability of the formylation-defective tRNAs and the extragenic suppressors isolated in this study enabled us to address the role of lack of rT modification in formylation-independent initiation afresh.

The initiation activity of the formylation-defective tRNAs was also examined in E. coli strains defective in m5U54-methyltransferase (trmA–). Both of the formylation-defective initiator tRNAs (G72/U35A36 and G72G73/U35A36) failed to initiate. This showed that the absence of rT is not sufficient for initiation independent of formylation. Further, the base modification analysis of G72G73/U35A36 tRNA from the second class of extragenic suppressors, which appear to initiate independent of formylation, did not show any lack of rT modification at position 54. Taken together, these analyses demonstrate that lack of rT modification is neither essential nor sufficient for formylation-independent initiation of protein synthesis in E. coli.