Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. Sep 2007; 189(17): 6494–6496.
Published online Jul 6, 2007. doi:  10.1128/JB.00622-07
PMCID: PMC1951902

Genetic Code Ambiguity Confers a Selective Advantage on Acinetobacter baylyi[down-pointing small open triangle]


A primitive genetic code, composed of a smaller set of amino acids, may have expanded via recursive periods of genetic code ambiguity that were followed by specificity. Here we model a step in this process by showing how genetic code ambiguity could result in an enhanced growth rate in Acinetobacter baylyi.

The genetic code likely evolved from a primitive, smaller group of amino acids to the modern complement of 20 amino acids via a series of ambiguous intermediate codes (7, 17). According to this hypothesis, each codon block would originally have included a greater number of codons that recursively cycled between ambiguity and specific coding, until eventually the modern codon assignments evolved. This is similar to the ambiguous intermediate theory regarding codon reassignments in genetic codes of modern organisms (15, 16). This theory posits that new interpretations of the code occur via ambiguous reading of specific codons. It has been experimentally demonstrated with bacteriophage that adaptation to an ambiguous genetic code is possible (2), and several species of Candida have been shown to ambiguously interpret the CUG codon as Leu and Ser (recently reviewed in reference 10).

The NUN codon block, which is thought to have originally encoded only valine, is one example in which ambiguity is thought to have been important in codon assignments. This codon block incorporated leucine, isoleucine, and methionine into the genetic code, each in successively smaller codon blocks. With the exception of phenylalanine, the NUN-encoded amino acids are structurally similar. (Phenylalanine may have “captured codons” from the NUN block, as a later addition to the genetic code [17].) Therefore, the cognate aminoacyl-tRNA synthetases would naturally be more permissive towards these related amino acids. Indeed, wild-type isoleucyl-tRNA synthetase (IleRS) activates valine at a frequency of ~1:200, compared with the general error level of translation of ~1:3,000 (8, 13, 14). However, genetic code fidelity is maintained by a distinct hydrolytic editing domain in IleRS, which clears misactivated or mischarged Val-tRNAIle (5, 6).

In previous work, we explored growth characteristics of Escherichia coli that carries a mutant IleRS that can no longer edit mischarged tRNA (1, 11). Eleven codons in the editing center of the chromosomal copy of ileS were mutated to encode alanine. The mutant enzyme, IleRSAla, was stable and fully active for aminoacylation and could not clear Val-tRNAIle. The mutant allele proved mildly detrimental under most growth conditions.

While the wild-type strain maintained a growth rate advantage, under certain conditions excess Val resulted in an increased yield of viable, editing-deficient cells. However, effects based on yield result in a linear response based on the concentration of nutrients. By comparison, a growth rate effect results in an exponential difference in population sizes. Indeed, the exponential amplification of small differences in growth rates (rather than yields) between genetic variants is the theoretical basis for natural selection. Therefore, we set out to determine whether a growth rate advantage could be obtained for the editing-deficient strain. Towards this end, we considered that a mutation that allows an aminoacyl-tRNA synthetase to expand its substrate range may permit the organism to enter exponential growth, particularly when the cognate amino acid is limiting but the new amino acid is in excess. (The new amino acid would also have to be sufficiently similar to the native amino acid in order for the proteins being synthesized to fold correctly and retain function.) By analogy, adaptation of clonal E. coli to glucose-limiting conditions resulted in two lineages: one that specialized in growth on glucose and one that grew particularly well on acetate, a product of glucose metabolism (12, 18). Thus, we hypothesized that conditions might exist where an ambiguous genetic code would result in a growth rate advantage (rather than a growth yield advantage) by supplementing a limiting amino acid with a structural analog.

Isogenic strains of Acinetobacter baylyi were constructed essentially as described previously (9) to carry E. coli ileS alleles that were either wild type (ileSEc) or editing defective (ileSEc, Ala). The native ileS allele and a crucial gene (ilvC) in the isoleucine-leucine-valine biosynthetic pathway were deleted from both A. baylyi strains. (The deletion of ilvC enabled us to exogenously control the supply of branched-chain amino acid.) A. baylyi was grown in microplate wells, and growth curves were generated using a microplate reader to determine the growth rates in parallel for wild-type and editing-defective strains under various conditions of amino acid concentrations in minimal medium, MSglc (1, 11). A systematic, recursive screen was deployed to identify those conditions that would confer a growth rate advantage on A. baylyi having an ambiguous genetic code. (A. baylyi is naturally highly competent and recombinogenic in culture, especially at the onset of the logarithmic phase of growth. This strain was used in order to allow the potential for recombination in planned, future evolutionary experiments.)

When both isoleucine and valine were limiting (Ile = 30 μM; Val = 50 μM), growth rates were low and equivalent between editing-defective and wild-type strains (Fig. (Fig.1A)1A) (Leu was maintained at 50 μM throughout these experiments). However, when the Val concentration was increased 10-fold (Ile = 30 μM; Val = 500 μM), the editing-defective strain improved its doubling time from ~3.3 h to ~2.3 h. Titrating Ile back into the medium (Ile = 70 μM; Val = 500 μM) increased the growth rate of the wild-type strain to give the higher growth achieved by the editing-defective strain. That Val failed to give a growth rate advantage when in such large excess (Ile = 70 μM; Val = 500 μM) shows that the active site of IleRS retained significant discrimination.

FIG. 1.
Growth rate advantage and enhanced misincorporation of valine due to genetic code ambiguity in A. baylyi. Growth conditions were systematically varied to determine whether certain concentrations of isoleucine and valine could result in a growth rate advantage ...

While the observed change in the growth rate may signify a growth rate advantage stemming from genetic code ambiguity, the disparity in concentrations of Ile and Val may activate permeases that better discriminate in favor of Ile, thereby maintaining a more balanced intracellular concentration. In this scenario, the growth rate advantage from excess valine may ultimately result from more-efficient “scavenging” of limiting isoleucine in the medium. Scavenging has been observed in E. coli as an evolved response to long-term growth in medium with a skewed ratio of tryptophan and its structural analog, 4-fluorotryptophan (4).

“Scavenging” can be tested by measuring the relative levels of the two amino acids that are actually incorporated into proteins in the editing-defective and wild-type strains (3, 4). If the incorporated amino acids are present at constant levels regardless of the amino acid concentration in the medium, then discrimination and scavenging are more likely. However, if increased incorporation of Val into proteins is observed and this correlates with the higher growth rate, then scavenging is ruled out.

The amino acid composition of the cellular proteome was therefore determined (3). When the growth rates of the bacteria were equivalent, the amounts of valine in the proteome relative to the total valine plus isoleucine were the same between strains (Fig. (Fig.1B).1B). However, when valine was in excess over isoleucine in the medium, the valine content of the total protein expressed by both strains increased, but the difference was 2.5-fold greater for the editing-defective strain (Fig. (Fig.1B).1B). In parallel with the loss of a growth rate advantage, when the Ile concentration was raised from 30 μM to 70 μM (Fig. (Fig.1A),1A), the excess incorporation of Val dropped back to normal (Fig. (Fig.1B).1B). These results suggest that Val may be substituting for limiting Ile, thereby granting a growth rate advantage.

Genetic code ambiguity can result in a growth rate advantage in correlation with a change in the amino acid content of the proteome. Access to new amino acids, especially to substitute for limiting amino acids, may have been the selective pressure by which the genetic code expanded during the origin and early evolution of living systems. This access could have resulted in the selective advantage required for the evolution of the genetic code in primitive organisms. These results develop the applicability of the ambiguous intermediate hypothesis of genetic code evolution to include the expansion of a primitive, smaller genetic code to the modern set of 20 amino acids encoded in modern organisms.


This work was supported by National Institutes of Health grant GM23562, by a fellowship from the National Foundation for Cancer Research, and by National Science Foundation grant MCB-0128901.


[down-pointing small open triangle]Published ahead of print on 6 July 2007.


1. Bacher, J. M., V. de Crécy-Lagard, and P. R. Schimmel. 2005. Inhibited cell growth and protein functional changes from an editing-defective tRNA synthetase. Proc. Natl. Acad. Sci. USA 102:1697-1701. [PMC free article] [PubMed]
2. Bacher, J. M., J. J. Bull, and A. D. Ellington. 2003. Evolution of phage with chemically ambiguous proteomes. BMC Evol. Biol. 3:24. [PMC free article] [PubMed]
3. Bacher, J. M., and A. D. Ellington. 2007. Global incorporation of unnatural amino acids in Escherichia coli. Methods Mol. Biol. 352:23-34. [PubMed]
4. Bacher, J. M., and A. D. Ellington. 2001. Selection and characterization of Escherichia coli variants capable of growth on an otherwise toxic tryptophan analogue. J. Bacteriol. 183:5414-5425. [PMC free article] [PubMed]
5. Baldwin, A. N., and P. Berg. 1966. Transfer ribonucleic acid-induced hydrolysis of valyladenylate bound to isoleucyl ribonucleic acid synthetase. J. Biol. Chem. 241:839-845. [PubMed]
6. Eldred, E. W., and P. R. Schimmel. 1972. Rapid deacylation by isoleucyl transfer ribonucleic acid synthetase of isoleucine-specific transfer ribonucleic acid aminoacylated with valine. J. Biol. Chem. 247:2961-2964. [PubMed]
7. Jordan, I. K., F. A. Kondrashov, I. A. Adzhubei, Y. I. Wolf, E. V. Koonin, A. S. Kondrashov, and S. Sunyaev. 2005. A universal trend of amino acid gain and loss in protein evolution. Nature 433:633-638. [PubMed]
8. Loftfield, R. B., and D. Vanderjagt. 1972. The frequency of errors in protein biosynthesis. Biochem. J. 128:1353-1356. [PMC free article] [PubMed]
9. Metzgar, D., J. M. Bacher, V. Pezo, J. Reader, V. Döring, P. Schimmel, P. Marlière, and V. de Crécy-Lagard. 2004. Acinetobacter sp. ADP1: an ideal model organism for genetic analysis and genome engineering. Nucleic Acids Res. 32:5780-5790. [PMC free article] [PubMed]
10. Miranda, I., R. Silva, and M. A. S. Santos. 2006. Evolution of the genetic code in yeasts. Yeast 23:203-213. [PubMed]
11. Pezo, V., D. Metzgar, T. L. Hendrickson, W. F. Waas, S. Hazebrouck, V. Döring, P. Marlière, P. Schimmel, and V. De Crécy-Lagard. 2004. Artificially ambiguous genetic code confers growth yield advantage. Proc. Natl. Acad. Sci. USA 101:8593-8597. [PMC free article] [PubMed]
12. Rozen, D. E., D. Schneider, and R. E. Lenski. 2005. Long-term experimental evolution in Escherichia coli. XIII. Phylogenetic history of a balanced polymorphism. J. Mol. Evol. 61:171-180. [PubMed]
13. Schmidt, E., and P. Schimmel. 1994. Mutational isolation of a sieve for editing in a transfer RNA synthetase. Science 264:265-267. [PubMed]
14. Schmidt, E., and P. Schimmel. 1995. Residues in a class I tRNA synthetase which determine selectivity of amino acid recognition in the context of tRNA. Biochemistry 34:11204-11210. [PubMed]
15. Schultz, D. W., and M. Yarus. 1994. Transfer RNA mutation and the malleability of the genetic code. J. Mol. Biol. 235:1377-1380. [PubMed]
16. Schultz, D. W., and M. Yarus. 1996. On malleability in the genetic code. J. Mol. Evol 42:597-601. [PubMed]
17. Trifonov, E. N. 2004. The triplet code from first principles. J. Biomol. Struct. Dyn. 22:1-11. [PubMed]
18. Tyerman, J., N. Havard, G. Saxer, M. Travisano, and M. Doebeli. 2005. Unparallel diversification in bacterial microcosms. Proc. Biol. Sci. 272:1393-1398. [PMC free article] [PubMed]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...