|Home | About | Journals | Submit | Contact Us | Français|
Unnatural amino acids can be genetically incorporated into proteins in live cells by using an orthogonal tRNA/aminoacyl-tRNA synthetase pair. Here we describe a method to efficiently express the orthogonal tRNA and synthetase in Saccharomyces cerevisiae, which enables unnatural amino acids to be genetically incorporated into target proteins in yeast with high efficiency. We also describe the use of a yeast strain deficient in the nonsense-mediated mRNA decay, which further increases the unnatural amino acid incorporation efficiency when a stop codon is used to encode the unnatural amino acid. These strategies will facilitate the investigation of proteins and their related biological processes in yeast by exploiting the novel properties afforded by unnatural amino acids.
To genetically incorporate an unnatural amino acid into proteins in yeast, an orthogonal tRNA/aminoacyl-tRNA synthetase pair needs to be expressed (1, 2). This tRNA/synthetase pair does not cross-talk with endogenous tRNA/synthetase pairs and is engineered to be specific for the desired unnatural amino acid. One challenge is to efficiently express the orthogonal tRNA in yeast (1, 3). Most orthogonal tRNAs used in yeast are derived from bacteria. However, bacteria and yeast differ significantly in tRNA transcription and processing (4) (Fig. 1a). Bacterial tRNAs are transcribed by the sole RNA polymerase (Pol) through promoters upstream of the tRNA gene. The transcription of yeast (and other eukaryotic) tRNAs by Pol III depends principally on promoter elements within the tRNA known as the A- and B-box (4). The A- and B-box identity elements are conserved among eukaryotic tRNAs, but are lacking in many bacterial tRNAs. In addition, bacterial tRNA genes encode full tRNA sequences, whereas eukaryotic tRNAs have the 3′-CCA trinucleotide enzymatically added after transcription (4). Therefore, transplanting bacterial tRNA directly into the tRNA gene cassette in yeast does not generate functional tRNA.
We developed a general method to express bacterial tRNAs in yeast (3), which involves placing an external Pol III promoter containing the consensus eukaryotic A- and B-box sequences upstream of the target bacterial tRNA gene (Fig. 1b). The 3′-CCA trinucleotide of the tRNA is excluded, and the tRNA(−CCA) gene is followed by a 3′-flanking sequence of an endogenous yeast tRNA. A primary RNA consisting of the promoter and the tRNA is transcribed, and the promoter is cleaved posttranscriptionally to yield the mature tRNA. Two yeast Pol III promoters, the RPR1 promoter and the SNR52 promoter, have been shown to efficiently drive the expression of Escherichia coli tRNAs in yeast. The expressed E. coli tRNA is six- to ninefold more active in translation than the same tRNA transcribed by using the SUP4 5′-flanking sequence. Alternative methods using a strong Pol II promoter with tandem tRNA repeats (5) or the yeast tRNAArg fused upstream of the target tRNA (6, 7) have also been developed. We will focus on the SNR52/RPR1 promoter method here, as this method works with different E. coli tRNAs and has been reproducibly used in different laboratories (8, 9). A similar approach involving the use of type-3 Pol III promoters also works efficiently in mammalian cells (10).
The amber stop codon, UAG, is often introduced into the gene of interest to specify the site at which the unnatural amino acid is to be incorporated. The Nonsense-mediated mRNA decay (NMD) pathway mediates the rapid degradation of mRNAs that contain premature stop codons in yeast (11), whereas no such pathway exists in E. coli. When stop codons are used to encode unnatural amino acids, NMD could result in a shorter lifetime for the target mRNA and thus a lower protein yield in yeast. We generated an NMD-deficient yeast strain (LWUPF1Δ) by knocking out the upf1 gene from the yeast genome, and found that this strain indeed increases the unnatural amino acid incorporation efficiency in comparison to the wild-type (wt) yeast (3).
To demonstrate this method, we describe here the procedures to incorporate a fluorescent unnatural amino acid 2-amino-3-(5-(dimethylamino) napththalene-1-sulfonamido) propanoic acid (abbreviated as DanAla) into the green fluorescent protein (GFP) at position 39. As shown in Fig. 2, the orthogonal E. coli leucyl amber suppressor tRNA () will be expressed using the SNR52 promoter, and the orthogonal DanAla-specific synthetase (DanAlaRS) (12) will be expressed using the GPD promoter. The target GFP gene with a UAG codon at site 39 will be expressed using the ADH1 promoter in another plasmid. A His6 tag is appended to the C-terminus of GFP for Western detection and affinity purification. The two plasmids will be co-transformed into the wt or LWUPF1Δ yeast strain. The incorporation of DanAla into GFP will be verified with the generation of GFP fluorescence and quantified using flow cytometry or Western blot. DanAla-containing GFP proteins will be extracted from cells and purified using Ni–NTA chromatography.
Prepare all solutions with ultrapure water and use analytical grade reagents unless indicated otherwise.
A plasmid containing the 2µ ori, LEU2, Ampr, the ColE1 ori, and MCS is used as the backbone to construct the pGFP-39TAG plasmid (3).
The following procedures are applicable to both wt yeast and the NMD-deficient LWUPF1Δ strain. To determine if the LWUPF1Δ strain would be helpful for your target protein expression, see Note 10.
We thank Dr. Vicki Lundblad and members of the Lundblad lab for providing reagents and advice on yeast protocols. This work was supported by CIRM (RN1-00577-1) and NIH (1DP2OD004744).
1Some unnatural amino acids may be difficult to dissolve; lower the stock concentration to 200 mM or 100 mM when necessary. Solubility is also dependent on the purity of the unnatural amino acid. If racemic mixture of unnatural amino acids is used, the concentration of the effective l-amino acid will be 50% of the calculated value. Use optical pure l-amino acids whenever possible.
2Amino acid dropout supplements can be purchased from Clontech (Catalog number 630417). If purchasing from other suppliers, change the amount accordingly by following the product information.
3The agar will not fully dissolve until it is autoclaved.
4Prepare agar plates in advance. Unsleeve to dry the plates at room temperature for 1 day prior to plating.
5Store the boiled carrier DNA at −20°C, which can be reboiled and reused for three times without loss of activity.
6Prepare these solutions fresh prior to use.
7The sequence for the SNR52 promoter is the following: tctttgaaaagataatgtatgattatgctttcactcatatttatacagaaacttgatgtttt ctttcgagtatatacaaggtgattacatgtacgtttgaagtacaactctagattttgtagtg ccctcttgggctagcggtaaaggtgcgcattttttcacaccctacaatgttctgttcaaaaga ttttggtcaaacgctgtagaagtgaaagttggtgcgcatgtttcggcgttcgaaact tctccgcagtgaaagataaatgatc. The sequence for the is the following: GCCCGGATGGTGGAATCGGTAGACACAAGGGATTCTAAATCCCTCGGCGTTCGCGCTGTGCGGGTTCAAGTCCCGCTCCGGGTA. Note the underlined anticodon CTA, which recognizes the UAG amber codon. The 3′-CCA trinucleotide of the tRNA is not included in the plasmid. The 3′-flanking sequence of the SUP4 is the following: TTTTTTTGTTTTTTATGTCT. We avoided introducing a restriction enzyme site between the SNR52 promoter and the tRNA because such mutations may impair the promoter strength and/or hamper the generation of the correct 5′ end of the tRNA. If a new tRNA does not show activity after being expressed using the SNR52 promoter, check if there is any mutation in the promoter and the tRNA.
8Plasmids pSNR- -DanAlaRS and pSNR- -TyrRS harbor the orthogonal E. coli leucyl and tyrosyl amber suppressor tRNA, respectively, and are available from the Wang group (http://wang.salk.edu) upon request. When incorporating unnatural amino acids using orthogonal tRNA/synthetase pairs, the synthetase is evolved to be specific for different unnatural amino acids, but the orthogonal tRNA does not need to be changed and works with all these mutant synthetases. Therefore, mutant synthetases evolved from the E. coli TyrRS are all used with the E. coli tyrosyl amber suppressor tRNA (), and mutant synthetases evolved from the E. coli LeuRS are all used with the . To make an orthogonal tRNA/synthetase expression plasmid for the incorporation of a different unnatural amino acid, one just needs to replace the synthetase gene without recloning the tRNA gene. tRNA genes are generally difficult to be cloned, and the PstI and SalI sites are no longer unique in the final plasmid. The DanAlaRS and TyrRS gene can be swapped for other mutant synthetase genes using the SpeI (N-terminus) and XhoI (C-terminus) unique sites.
9To express your gene of interest, any plasmid with LEU2 and 2µ ori can be used. A strong promoter such as the ADH1 promoter used here is preferred for high expression level. Through the above cloning procedure, we built in a unique SalI site at the N-terminus and a BamHI site at the C-terminus of the GFP gene in the pGFP-39TAG. This plasmid is also available through the Wang group (http://wang.salk.edu) as a convenient fluorescent positive control to monitor unnatural amino acid incorporation (see Subheadings 3.5 and 3.6) and to facilitate the cloning of your genes of interest.
10The NMD-deficient LWUPF1Δ strain is available at the Wang group (http://wang.salk.edu) upon request. To determine whether the LWUPF1Δ strain will help the expression, check the location of the UAG codon in your gene of interest. NMD in yeast shows a polar effect of nonsense codon positions (13). The steady-state mRNA level is reduced by NMD more significantly when the nonsense codon is closer to the 5′ end than to the 3′ end of an mRNA. Consistently, the increase of unnatural amino acid incorporation efficiency in the LWUPF1Δ strain correlates with the position of the UAG codon: more than a twofold increase is measured when the amber codon is within the N-terminal two thirds of the gene, whereas no significant increase is detected when it is within the C-terminal fourth of the coding region.
11For the LWUPF1Δ strain, add 0.5 mg/mL G418 into the YPD medium to keep the selective pressure. If your yeast strain has additional marker or plasmid, use the appropriate SD/drop out medium to keep selective pressure.
12Different yeast strains grow at different rates. If colonies are small, or if you are inoculating a larger volume, use several colonies.
13For the highest transformation efficiency, use competent cells within 1 h of their preparation.
14The concentration of the plasmid is an important factor for the transformation efficiency. The smaller is the volume ratio of DNA mixture to competent cells, the higher will be the transformation efficiency. The total volume of the DNA mixture less than 10 µL is preferable.
15The volume of the competent cells should ≥10× volume of the DNA mixture.
16The PEG/LiAc solution should be freshly prepared before use.
17In order to get single colonies, spread 190 µL suspension on one plate; then add 180 µL fresh 1× TE to the suspension left and spread on another plate.
18The pGFP-39TAG can be used as the reporter plasmid to quickly verify the incorporation of an unnatural amino acid by the orthogonal tRNA/synthetase on agar plates, and to quantify the incorporation efficiency of the unnatural amino acid into GFP by using flow cytometry. Green fluorescence of GFP should be detected on cells when the unnatural amino acid is added to the agar plate. In the absence of the unnatural amino acid, no green fluorescence will be detected. To determine the incorporation efficiency, follow the procedures in Subheading 3.6. Measure the mean fluorescence intensity of cells transformed with pSNR- -DanAlaRS and pGFP-39TAG grown in 1 mM of DanAla (Int1), and those grown in the absence of DanAla (Int2). Also measure the mean intensity of cells transformed with pSNR- - LeuRS and pGFP-39TAG (Int3) and of cells transformed with pGFP-39TAG alone (Int4). Here Leu is incorporated by the orthogonal E. coli /LeuRS pair through amber suppression. The ratio defined by (Int1−Int2)/(Int3−Int4) will determine the relative incorporation efficiency of DanAla to Leu. To obtain the net incorporation efficiency of DanAla, measure the mean intensity (Int5) of cells transformed with pSNR- -DanAlaRS and pGFP (a plasmid identical to pGFP-39TAG except that the 39TAG is reverted to wt tyrosine codon TAC). The net DanAla incorporation efficiency is defined by (Int1−Int2)/(Int5−Int2). Note that this GFP reporter can be generally used to evaluate the incorporation of many unnatural amino acids, as the 39TAG site is permissive for GFP fluorescence. Also note that the incorporation efficiency of an unnatural amino acid can be protein-dependent and site-dependent.
19Western blot analysis can also be used to determine the incorporation efficiency of the unnatural amino acid into the target protein. Use densitometry to measure the intensities of target protein bands and use the loading control for sample normalization. If the orthogonal tRNA/synthetase for your unnatural amino acid of interest has not been fully characterized before, it is also necessary to perform mass spectrometric analysis of the purified target protein with the unnatural amino acid incorporated (14). Tandem mass spectrometric analysis of protease digested peptides will determine the identity of the amino acid incorporated at the UAG site. A semiquantitative estimation of the incorporation fidelity can be obtained from the peptide intensities (5). Mass analysis of the intact protein will also reveal if common amino acids are incorporated at the UAG site and if there is any misincorporation at other sites in the target protein.
20Alternatively, 10 µL of supernatant similarly prepared from cells transformed with pGFP can be used here.
21If a primary antibody against the target protein is available, it can be used here for Western detection in replacement of the Penta·His antibody.
22The incubation time for expression is protein dependent. We highly recommend a time course experiment to determine the optimal expression time for your target protein. Western blot analysis of cell lysates can be used to monitor target protein expression level conveniently.