Figure 1. A ribosome in translation and glycine aminoacyl-tRNAs in E. coli. An mRNA sequence is shown with a single ribosome particle engaged in the process of translation. In this snapshot, the growing polypeptide chain is attached to its C-terminal tRNA located in the P-site adjacent to the next aminoacyl-tRNA located in the A-site. The K12 strain of E. coli has six different genes encoding glycine-tRNAs, illustrated here in their aminoacylated form with the anticodon sequence shown paired with the respective glycine codon that is recognized by each tRNA. There are four genes in E. coli that encode copies of the Gly tRNA with a 3¢-CCG-5¢ anticodon loop coloured green (the number of genes is indicated as a superscript number above the tRNAs shown). The Gly tRNAs with 3¢-CCG-5¢ anticodons can read both the GGC (pink) and GGU (gold) mRNA codons, since the G in the wobble position of the anticodon loop is capable of reading both the C or U in the complementary position within the codon (star). The frequency with which the four different glycine codons are found in the ORFs of E. coli are listed below each of the codons for all genomic ORFs, or for only the ORFs of highly expressed proteins (HX). Website sources of information used in this figure are located at: http://lowelab.ucsc.edu/GtRNAdb/Esch_coli_K12 and http://www.kazusa.or.jp/codon.

Figure 2. Effects of synonymous sequence differences on protein production. Top Panel: Effect of synonymous codon usage on the kinetics of protein production. Two synonymous mRNAs are shown with multiple ribosome complexes in the elongation phase of translation. One mRNA contains two rarely used (GGA and GGG) codons for glycine (top panel), while the other mRNA contains two commonly used (GGU and GGC) codons for glycine (bottom panel). The differential codon usage is anticipated to affect the kinetics of translation such that the mRNA with preferred codon usage results in a higher overall steady state protein production level. Lower Panel: Effect of synonymous codon usage on protein function. Two synonymous mRNAs are shown with multiple ribosome complexes in the elongation phase of translation. Synonymous changes for Gly and Ile codons do not affect the overall level of protein production for the entire mRNA, but do affect the final ligand specificity of the resulting protein. In this case, differential codon usage is anticipated to affect the specific kinetics of translation at crucial points in the folding of the protein, resulting in a different folding pattern and ligand specificity of the final protein product (Kimchi-Sarfaty et al., 2007).

Figure 3. Engineering synonymous ‘silent’ changes in gene sequences for improved protein production. Top Panel: One possible DNA sequence encoding an expressed protein is shown as the top sequence. The bottom sequence is an alternative DNA sequence with synonymous codon sequence changes (lower case) that achieve the desired engineering features in either the DNA (restriction site removal or introduction) or the RNA (RNase removal, ACA removal, cryptic Shine–Dalgarno removal, ambush stop codon introduction and hairpin removal). DNA/RNA elements removed from the gene are highlighted in red. Sequence changes that result in the introduction of a new sequence element are shown in green, while changes that eliminate sequence elements are shown in blue. Bottom Panel: An mRNA is shown with multiple ribosome complexes in the elongation phase of translation along the mRNA. When the ribosome translates through a ‘slippery’ nucleotide repeat sequence it has an increased propensity to shift its reading frame to the –1 frame where it encodes a mistranslated amino acid sequence at its C-terminus until it reaches a stop codon in the –1 frame. Engineering of silent/synonymous changes to the gene sequence can eliminate the slippery nucleotide repeat sequence and also introduce a stop codon in the –1 reading frame that would be encountered by a frameshifted ribosome sooner than would otherwise happen in the native RNA sequence. The combined effect of eliminating the ‘slippery’ repeat site and introduction of an ‘ambush’ stop signal should improve overall normal protein production.

Figure 4. Gene synthesis by PCR from overlapping oligonucleotides in combination with mismatch specific endonuclease elimination of mutant strands. This figure illustrates a generic PCR-based gene synthesis protocol starting with overlapping complementary oligonucleotides and following the steps described in the figure itself and in the body text of this article.