Buffers
Table of Contents
Resuspension buffer: 100 mM sodium acetate, 50 mM NaCl, 0.1 mM EDTA, pH 5.0; deacylation buffer: 50 mM bicine pH 9.6, 1 mM EDTA; Buffer D: 50 mM sodium acetate pH 5, 150 mM NaCl, 10 mM MgCl2, 0.1 mM EDTA; hybridization buffer: 10 mM Tris-HCl, 25 mM NaCl, pH 7.4; Klenow fragment exo− master mix with Cy5-11-dCTP (KMM–Cy5): 17 μl water, 5 μl 10x NEBuffer 2.0, 1 μl Klenow fragment exo− (NEB), 1 dNTPS-dCTP (10 mM), 1 μl Cy5-11-dCTP (20 μM); orange loading dye: 8 M urea, orange G; Klenow fragment exo− master mix mini (KMM -mini): 3.35 μl water, 1.2 μl 10x NEBuffer 2.0, 0.1 μl Klenow fragment exo−, 0.25 dNTPS (10 mM); Klenow fragment exo− master mix with Cy5-11-dCTP mini (KMM–Cy5-mini): 2.1 μl water, 1.2 μl 10x NEBuffer 2.0, 0.2 μl Klenow fragment exo−, 0.5 dNTPS (5 mM, without dCTP), 1 μl Cy5-11-dCTP (5 μM); Klenow fragment exo− master mix with Bio-11-dCTP (KMM-Bio): 17 μl water, 5 μl 10x NEBuffer 2.0, 1 μl Klenow fragment exo−, 1 dNTPS-dCTP (10 mM), 1 μl Bio-11-dCTP (20 μM); washing buffer: 10 mM Tris-HCl, 150 mM LiCl, 1 mM EDTA, 0.05% (v/v) Tween-20, pH 7.5; binding buffer: 20 mM Tris-HCl, 1 M LiCl, 2 mM EDTA, pH 7.5; formamide loading buffer (FLB): 90% formamide; SDS lysis buffer (SLB): 100 mM sodium acetate, 50 mM NaCl, 0.1 mM EDTA, pH 5.0, 1% (v/v) sodium dodecyl sulphate (SDS); alkaline washing buffer (AWB): 25 mM NaOH, 4 mM EDTA, 0.05% (v/v) Tween-20; reverse transcription hybridization mix (RHM): 1 μl DNA primer (2 μM), 1 μl 10 mM dNTPs, 1 μl 10x hybridization buffer, 10 μl water; reverse transcription master mix (RMM): 4 μl SSIV buffer, 1 μl RNAse Out, 1 μl SSIV reverse transcriptase, 1 μl 0.1 M dithiothreitol (DTT); acidic washing buffer 1(aWB1): 100 mM sodium acetate pH 5; acidic washing buffer 1 plus Tween-20 (aWB1-T): 100 mM sodium acetate pH 5, 0.01% (v/v) Tween-20; acidic washing buffer 2 (aWB2): 20 mM sodium acetate pH 5.
Media
SOC: Super Optimal Broth plus 20 mM glucose; 2xYT-s: Yeast Extract Tryptone supplemented with 75 μg ml−1 spectinomycin; 2xYT-s-t: Yeast Extract Tryptone supplemented with 75 μg ml−1 spectinomycin and 10 μg ml−1 tetracycline; 2xYT-s-ap: Yeast Extract Tryptone supplemented with 75 μg ml−1 spectinomycin and 50 μg ml−1 apramycin; 2xYT-am: Yeast Extract Tryptone supplemented with 75 μg ml−1 ampicillin.
Chemicals
NcAA 1 and 2 were purchased from Bachem. NcAA 4 was purchased from Fluorochem. NcAA 5 was purchased from Ambeed. NcAAs and ncMs 6, 8, 14, S1, S4 and S5 were purchased from Enamine. NcAA 7 was purchased from aaBlocks. NcAAs and ncMs 9, 10, 12 and S3 were purchased from Merck. NcMs 13, 15, 16, 17, 18, 20 and S7 were purchased from BLD. NcM S2 was purchased from Advanced ChemBlock. NcM 19 and S6 were purchased from AstaTech. NcAAs and ncMs 3 and S8 were synthesized as previously described3, and ncAA 11 was custom synthesized as previously described54. NcMs S1 and S5 were Boc-deprotected in concentrated HCl in dioxane.
DNA oligonucleotides
See Supplementary Data 2.
DNA constructs
See Supplementary Data 2 for all constructs.
DNA construct cloning
Standard cloning was performed by Gibson assembly using NEBuilder HiFi DNA Assembly Master Mix (NEB) according to manufacturer’s guidelines. Libraries were generated by enzymatic inverse PCR, as previously described55,56. In brief, a template plasmid was amplified by PCR using two primers (see primer list) containing degenerate codons at desired mutagenesis sites and a BsaI cleavage site. In the case of custom mixes, primers containing different codons were manually mixed and used for PCR reactions. PCR products were gel purified and digested using BsaI and DpnI. Subsequently, samples were purified, ligated using T4 DNA Ligase, and transformed into electrocompetent E. coli DH10β cells ensuring a minimal transformation efficiency of 109. Individual colonies (>10) were evaluated using Sanger sequencing for quality control of the library assembly. Total plasmid DNA was prepared from the resulting culture, sequenced as a bulk using Sanger sequencing and used for subsequent experiments.
General protocols
Isolation and oxidation of tRNAs (protocol A)
This protocol was used to isolate tRNAs from 1–10 ml of cell culture. Chemically competent DH10β cells were transformed with a pMB1 plasmid encoding a PylRS variant and a tRNA, or a circularly permutated split tRNA, and rescued in 1 ml of SOC for 1 h at 37 °C, 700–1,000 rpm. Cells were transferred into selective 2xYT-s medium and grown overnight. Overnight cultures were diluted in a ratio between 1:20 and 1:40 and grown to OD600 of 0.5–1. Cells were centrifuged at 4,200 rcf at 4 °C for 12 min, taken up in 200 µl resuspension buffer, transferred to a 96-well plate and centrifuged at 4,200 rcf at 4 °C for 12 min. Cells were resuspended in 135 µl resuspension buffer and 15 µl liquid phenol was added. Cells were lysed by shaking at 650 rpm for 20 min, and then centrifuged at 4,200 rcf at 4 °C for 20 min; the cell lysate was added to 40 µl chloroform, and the resulting suspension mixed by pipetting up and down. The mixture was centrifuged at 4,200 rcf at 4 °C for 10 min, and 115 µl of the aqueous layer were transferred into 6 µl 0.1 M NaIO4. The isolated RNA was oxidized for 1 h on ice, and the oxidation reaction was quenched by addition of 8 µl of 0.1 M DTT. tRNAs were purified using the ZR-96 Oligo Clean & Concentrator from Zymo Research. In brief, 250 µl oligo binding buffer was added to the oxidation reaction, subsequently 400 µl isopropanol was added, and the mixture transferred to a 96-well silica column plate. The plate was centrifuged for 2 min, 4,200 rcf at room temperature, and 800 µl of oligo wash buffer was added. The plate was centrifuged for 2 min, 4,200 rcf at room temperature, aerated, and centrifuged for another 4 min, 4,200 rcf at room temperature. Finally, the RNA was eluted in either 14 µl water, when the samples were not processed, or in 50 µl water, when the samples were further deacylated, by centrifugation for 4 min, 4,200 rcf at room temperature.
Isolation and oxidation of tRNAs (protocol B)
The volumes described in this protocol were used to isolate tRNAs from 5–25 ml of cell culture as previously described22,57. In brief, cells were grown as described in protocol A, washed with 800 µl resuspension buffer and transferred to a 1.5 ml Eppendorf tube. Cells were taken up in 225 µl resuspension buffer and 25 µl of liquid phenol was added. Cells were lysed by vortexing for one minute and incubation, with head-over-tail rotation, for 20 min. Lysed cells were centrifuged for 15 min, 20,000 rcf at room temperature, the cell lysate was added to 250 µl chloroform, the samples were vortexed for one minute and then centrifuged, 10 min, 20,000 rcf at room temperature. Two-hundred microlitres of the aqueous layer was transferred into 10 µl 0.1 M NaIO4 and the RNA was oxidized for 1 h on ice. Finally, the oxidized RNA was added to 440 µl ethanol and precipitated for at least 20 min at −20 °C. The samples were centrifuged for 25 min, 20,000 rcf at 4 °C and aspirated. RNA pellets were dried for 10 min at room temperature and dissolved in water or buffer D.
tRNA deacylation
45 μl of isolated RNA was added to 5 µl 10 x deacylation buffer and tRNAs were deacylated for 36 min at 42 °C. The deacylation reaction was quenched by addition of 6 µl 3 M sodium acetate, and tRNAs were purified using the ZR-96 Oligo Clean & Concentrator from Zymo Research, as described in protocol A for the isolation and oxidation of tRNAs (with the exception of using 100 µl of oligo binding buffer, instead of 250 µl). Deacylated tRNAs were eluted in 14 µl water.
Fluoro-tREX
This protocol was used to run the experiments shown in Fig. 2c and Supplementary Fig. 4. RNA concentrations were adjusted to the lowest common denominator and 10 µl of RNA was added to 2.5 µl 10x hybridization buffer, 11.5 µl water and 1 µl Cy3-labelled extension primer (2 µM; note that probes for tREX-based approaches are described as primers throughout the methods even though they template extension and are not themselves extended). The DNA primer was hybridized at 65 °C for 5 min, before addition of 25 µl KMM–Cy5 and extension at 37 °C for 6 min. Samples were purified using the 10 µg NEB Monarch RNA clean-up Kit (NEB) and eluted in 12 µl water. 12 µl of orange loading dye was added, the samples were loaded onto a Novex TBE 6 M urea 10 or 15% PAGE gel (Invitrogen) and run for 36 min in 0.5x Tris-borate-EDTA (TBE) buffer at 270 V. Gels were imaged on an Amersham Typhoon Biomolecular Imager (GE) using the Cy3 and Cy5 emission filters. Then gels were stained with SYBR Gold (Invitrogen) and imaged again using the same filters.
Mini-fluoro-tREX
Unless stated otherwise, all fluoro-tREX experiments were run with the mini-fluoro-tREX protocol. Six microlitres of RNA was added to 0.5 µl 10x hybridization buffer, and 0.5 µl Cy3-labelled extension primer (2 µM). The DNA primer was hybridized at 65 °C for 5 min, before addition of 5 µl KMM–Cy5-mini and extension at 37 °C for 6 min. Samples were analysed as described for fluoro-tREX.
Mini-tREX
This protocol was adapted from Cervettini et al.22. Total tRNA was isolated as described in protocol A and deacylated. One to two micrograms of RNA was diluted into a total volume of 6 µl and added to 0.5 µl 10x hybridization buffer, and 0.5 µl Cy5-labelled extension primer (2 µM). The DNA primer was hybridized at 65 °C for 5 min, before addition of 5 µl KMM-mini and extension at 37 °C for 6 min. Samples were analysed by running a 15% acrylamide 1x TBE gel (running 200 V, 40–80 min).
Bio–tREX
RNA concentrations were adjusted to match the lowest concentration in the samples being compared. Ten microlitres of RNA was added to 2.5 µl 10x hybridization buffer, 11.5 µl water and 1 µl extension primer (2 µM). The DNA probe was hybridized at 65 °C for 5 min, before addition of 25 µl KMM-bio and extension at 37 °C for 6 min. 10 µl of Dynabeads MyOne Streptavidin C1 magnetic beads (Invitrogen) per reaction were washed 3 times with 200 µl washing buffer, resuspended in 50 µl binding buffer; the beads were added to the extension reaction and binding was performed for at least 30 min at 4 °C, with head-over-tail rotation. The supernatant was removed, and the beads were washed 4 times with 200 µl washing buffer. The washed beads were resuspended in 10 µl FLB and heated to 98 °C for 3 min to release the tRNAs. Beads were removed and the supernatant was directly loaded onto a Novex TBE 6 M urea, 10% or 15% PAGE gel (Invitrogen) and run for 36 min in 0.5x TBE at 270 V. Gels were stained with SYBR Gold (Invitrogen) and imaged on an Amersham Typhoon Biomolecular Imager (GE) using the Cy2 emission filter.
Northern blotting
tRNAs were isolated following the general protocol A or B, omitting the oxidation by NaIO4. Two to three micrograms of RNA was loaded onto an acidic urea PAGE gel (9% acrylamide (19:1), 100 mM sodium acetate pH 5, 8 M urea) and the gel was run for 12–16 hours, using 100 mM sodium acetate as running buffer, at 6 W constant power. The gel was stained with SYBR Gold (Invitrogen) to identify the tRNAs and an appropriate section of the gel was cut and blotted using iBlot DNA Transfer Stack (Invitrogen) with the iBlot Dry Blotting System. The tRNAs were cross-linked to the membrane (Stratalinker UV Crosslinker 2400), which was subsequently blocked in Ambion ULTRAhyb-Oligo buffer (Invitrogen) for 30 min. The biotinylated DNA probe was added to a final concentration of 0.2 µg ml−1 and hybridized overnight at 37 °C and 160 rpm. The membrane was washed 3 times with 20 ml 0.5x TBE buffer and transferred into 15 ml Odyssey blocking buffer for 20 min before addition of IRDye 800CW Streptavidin (LI-COR) to a final concentration of 0.2 µg ml−1. Finally, the membrane was washed 3 times with 20 ml 0.5x TBE and imaged on an Amersham Typhoon Biomolecular Imager (GE) using the IR long range emission filter.
mRNA extraction and oxidation (A)
The volumes given are suited for 2 to 3 ml of cell culture and were adjusted proportionally when required. Chemically competent BL21 cells were transformed with a pColE1 plasmid encoding the stmRNA construct, which was under the control of a T7 promoter and T7 terminator, rescued in SOC, shaken at 220 rpm, 37 °C for 1 h, diluted into 2xYT-am and grown overnight. The overnight cultures were diluted in a ratio of 1:20 into 2xYT-am in absence or presence of the ncM and grown at 37 °C, 220 rpm to an OD600 of 0.5–0.8. PylRS production was induced by addition of Isopropyl β-d-1-thiogalactopyranoside (IPTG) to a final concentration of 1 mM and cells were grown for 20 min, 220 rpm at 37 °C.
Cells were centrifuged at 4,200 rcf at 4 °C for 12 min, resuspended in 800 µl resuspension buffer, transferred to a 96-well plate and centrifuged at 4,200 rcf, 4 °C for 12 min. Subsequently, the procedure outlined in the user manual of the Agencourt RNAdvance Cell v2 RNA isolation kit (Beckman) was followed. In brief, 200 µl LBE containing 10 µl proteinase K were used to lyse cells at room temperature for 1 h. 244 μl BBC beads were mixed with 266 µl isopropanol, added to the lysate, and the RNA was bound to the beads for 10 min at room temperature. Beads were washed 3 times with 200 µl 80% ethanol, after the final wash the beads were carefully dried, and the RNA was eluted in 80 µl water.
70 μl of the RNA solution was added to 40 µl resuspension buffer and 7.5 µl of 0.1 M NaIO4. The oxidation was run on ice for 1 h and quenched with 10.5 µl 100 mM DTT. To the oxidized RNA, a mix of 1.5 µl of 1.6 M Na2CO3 and 16.5 µl DNAse I buffer (Ambion DNAse I, Thermofisher) was added and the samples were resuspended. Subsequently, 18 µl of DNAse I was added, and the RNA was incubated at 37 °C for 30 min. The digestion was cooled on ice and 300 µl Agencourt RNAClean XP beads (Beckman) were added and the RNA bound for 10 min at room temperature. The beads were washed 3 times with 80% ethanol. After the final wash the beads were carefully dried, and the RNA was eluted in 25 µl water.
mRNA extraction and oxidation (B)
A similar protocol to the one outlined in procedure A was followed, but the RNA was isolated using acid phenol/chloroform extraction. In brief, BL21 E. coli harbouring stmRNAs were grown in 5 ml of 2xYT-am in the presence or absence of the ncM until an OD600 0.5–0.9. At this point stmRNA expression was induced by addition of 1 M IPTG to a final concentration of 1 mM. After 20 min, cells were collected by centrifugation at 4,200 rcf for 12 min at 4 °C. Cell pellets were resuspended in 800 µl resuspension buffer, transferred into 1.5 ml Eppendorf tubes, and centrifuged for 3 min at 4,200 rcf at room temperature. The supernatant was removed, and the pellets were resuspended in 500 µl SLB. 500 µl of acid-phenol was rapidly added and the tubes were vortexed for 1 min and centrifuged for 6 min, 21,000 rcf at room temperature. 450 μl of the aqueous layer was recovered and 50 µl of 2.5 M KCl was added. Acid-phenol chloroform extraction was repeated, retrieving 400 µl of the aqueous layer. 400 μl of chloroform was added and the samples were vortexed for 1 min followed by centrifugation at 21,000 rcf 6 min at room temperature. 300 μl of the aqueous layer was recovered, and chloroform extraction was repeated with 300 µl chloroform. 200 µl of the aqueous phase were transferred to new 1.5 ml Eppendorf tubes and 10 µl of 0.1 M NaIO4 was added. The oxidation reaction was run for 1 h on ice. 440 μl of ethanol was added and RNA was precipitated at −20 °C for at least 20 min. The RNA was pelleted by centrifugation at 21,000 rcf at 4 °C for 30 min. The supernatant was removed, and the pellets air-dried for 10 min. The RNA was resuspended in 50 µl water. 30 μl of each RNA sample was digested in 120 µl 1x Ambion DNAse I buffer including 12 µl Ambion DNAse I enzyme. Samples were purified with 50 µg Monarch RNA clean-up Kit (NEB) and eluted into 20 µl water. We note, that when the RNA is isolated by acid-phenol chloroform extraction, an additional deacylation step in deacylation buffer is required, to measure acylation of stmRNAs with non-α amino acid monomers.
Fluoro-mREX
RNA concentration for all samples were adjusted to match the lowest concentration in the samples being compared. Six to twelve micrograms of RNA was added to a mixture of 1 µl DNA probe (2 µM), 2.5 µl 10x hybridization buffer and water (added to a final volume of 25 µl). The primer was annealed at 65 °C for 5 min. 25 µl KMM–Cy5 was added. The primer was extended for 6 min, 37 °C. Samples were purified using the 10 µg NEB Monarch RNA clean-up Kit (NEB) and eluted in 12 µl water. 12 μl of orange loading dye was added to each sample. Gel electrophoresis was conducted using 1% agarose gels (cast using NorthernMax MOPS running buffer) in NorthernMax MOPS running buffer at 135 V, 42 min. Gels were stained with SYBR Gold (Invitrogen) and imaged on an Amersham Typhoon Biomolecular Imager (GE) using the Cy2 and Cy5 emission filter.
Bio–mREX
The RNA concentration of all samples, after RNA extraction and oxidation (Methods), was adjusted to match the lowest concentration in the samples being compared.
To perform the pulldown, 6–12 µg of RNA was added to a mixture of 1 µl DNA probe (2 µM), 2.5 µl 10x hybridization buffer and water (added to a final volume of 25 µl). The primer was annealed at 65 °C for 5 min. 25 µl KMM-bio were added. The primer was extended for 6 min, 37 °C. Ten microlitres of Dynabeads MyOne C1 streptavidin beads (Invitrogen) were washed 2 times with washing buffer, and resuspended in 50 µl binding buffer. Resuspended beads were added to the extension reaction, and the biotinylated stmRNAs were bound to the beads for 1 h, 4 °C, with head-over-tail rotation. The beads were washed on a magnetic stand with 3 × 200 µl washing buffer, two times 200 µl AWB, one time 200 µl washing buffer, one time 200 µl water and were finally resuspended in 13 µl of RHM. After the AWB wash and after the final wash with washing buffer, the beads were transferred into new plastic tubes. The primer was annealed at 65 °C for 5 min. 7 µl of RMM was added and the RNA reverse transcribed at 50 °C for 10 min. One microlitre of RNAse H was added and the mixture heated to 37 °C for 15 min and 98 °C for 3 min to release the cDNA from the beads. Finally, the cDNA was separated from the beads and either used for quantification by qPCR, NGS, or as a template for further cloning.
To determine the number of molecules in the input for the stmRNA pulldown, extracted and oxidized RNA samples, in 13 µl of RHM, were reverse transcribed using the same procedure as for the pulled-down stmRNA. The percentage of the input stmRNA molecules recovered in the pulldown was determined from the number of molecules before and after the pulldown, as determined by qPCR ((number of molecules after pulldown/number of molecules in input) × 100).
qPCR of cDNA from bio–mREX
qPCR reactions were run in triplicate for each bio–mREX sample and were composed of 2 µl of cDNA, 10 µl PowerUp SYBR Green Master Mix (Applied Biosystems), 0.4 µl of each primer and 7.2 µl water. A standard was generated by PCR of the MmPylRS gene and quantified using a Qubit 2 Fluorometer (Life Technologies) and the Qubit 1x dsDNA HS Assay Kit (Invitrogen). A five-step fivefold serial dilution was used to generate a qPCR standard curve. This allowed calculation of qPCR efficiency and the number of molecules in each sample. qPCR was run on a ViiA 7 Real-Time PCR System (Applied Biosystems) using the standard supplier protocol for SYBR Green (Invitrogen).
Preparation of cDNA from bio–tREX for NGS
Half of the cDNA from the 20 µl reverse transcription reaction from bio–tREX was added into a PCR mix containing 25 µl Q5 High-Fidelity 2x Master Mix, 12 µl water and 2 µl of a 10 µM predefined mix of indexing primers (see Supplementary Data 2). A standard PCR program with 29 amplification cycles and an annealing temperature of 60 °C was used. Extension times were adapted to the amplicon length according to the manufacturer’s guidelines. DNA was bound to 100 µl of Agencourt AMPure XP (Beckman) for 10 min and the beads were washed 3 times with 200 µl 80% ethanol. Beads were dried and DNA was eluted in 25 µl water. DNA concentrations were measured using Qubit 2 Fluorometer (Life Technologies) and the Qubit 1x dsDNA HS Assay Kit (Invitrogen) and 80 ng of each amplicon were combined into the NGS library. The combined library was diluted in HT1 Hybridization Buffer (Illumina) to a concentration of 2 nM. PhiX (Illumina) was added to increase the diversity of the library at a 20% molar ratio. 12 µl of the library was added to 18 µl HT1 Hybridization Buffer (Illumina) and 20 µl of the diluted mixture was used for NGS analysis.
Cloning of cDNA from bio–tREX for further evolution
Half of the cDNA from the 20-µl reverse transcription reaction, from bio–tREX, was added into a PCR mix containing 25 µl Q5 High-Fidelity 2x Master Mix, 12 µl water and 2 µl of a 10 µM predefined mix of golden gate assembly primers (see Supplementary Data 2). A touchdown PCR program was used. The initial annealing temperature of 65 °C was decreased over 10 cycles by 0.5 °C per cycle. Subsequently, 20 regular cycles using an annealing temperature of 58 °C were performed. Extension times were adapted to the amplicon length according to the manufacturer’s guidelines. DNA was bound to 100 µl of Agencourt AMPure XP (Beckman) for 10 min and the beads were washed 3 times with 200 µl 80% ethanol. Beads were dried and DNA was eluted in 25 µl water. The amplicon was then cloned into a new pColE1 backbone, which was previously amplified by Golden Gate primers (Supplementary Data 2), by two-piece Golden Gate assembly according to New England Biolabs’ guidelines.
NGS data analysis
NGS was performed on a MiSeq system (in the case of the test evolution with library 1 and substrate 2) or a NextSeq2000 system (in all other cases). The resulting cDNA from tRNA display was amplified using oligos NGS A(1-8) and NGS_B(1-8) (Supplementary Data 2) containing different combination of Nextera sequencing barcodes via PCR. Samples were purified, quantified, and combined in equimolar amounts. Paired end reads were first paired using PEAR58, and aligned to a reference sequence of MmPylRS using Bowtie259. The relevant library positions were extracted and translated to amino acids, and resulting variants were counted using R script. Subsequent operations were performed using the frequency of each variant in each library which was computed as the count value divided by the total number of counts of that library. Using R script, enrichment and selectivity scores were calculated for all variants as follows. First, variants that were only present in all positive replicates were considered (tables were merged using AND operator). Assuming that highly enriched sequences could potentially not be covered in the negative and the input samples but may still be of interest, the negative and the naïve replicates were merged to the positive table using an OR operator. A placeholder value of 0.95 counts was adopted in cases where a replicate did not cover a specific variant. The resulting dataset was used to calculate mean enrichments in the presence and in the absence of the ncAA or ncM, computed as the quotient of the mean frequencies in one condition and the input condition. The resulting positive and negative enrichments were used to calculate the selectivity value for each variant (equivalent to the quotient of positive and negative frequencies). For further analysis, variants were filtered using an empirically determined threshold value for the normalized standard deviation of the positive frequency (dispersion error in the plus substrate condition).
tRNA pulldown and ncM identification by LC–MS
tRNAs were isolated from 8 ml of cells following the general protocol B omitting the oxidation by NaIO4. The RNA pellet was resuspended in 90 µl buffer D and RNA concentrations adjusted to match the lowest concentration in the samples being compared. A volume of 0.5 µl of biotinylated DNA probe (100 µM) was added to the RNA and the DNA probe was hybridized at 65 °C for 5 min. 40 μl of Streptavidin Dynabeads MyOne C1 (Invitrogen), were washed twice with buffer D-T (buffer D with 0.05% (v/v) Tween), and added into 10 µl buffer D. The beads were added to hybridization reaction and the probe was bound to the beads for minimally 30 min at 4 °C with head-over-tail rotation. The samples were washed 3 times with 200 µl of acW1-T, twice with 200 µl of acW1, 3 times with 200 µl of acW2 and once with 200 µl water, all on a magnetic stand. 24 µl of deacylation buffer was added and the beads, which were incubated at 42 °C for one hour. 12 µl of the deacylation mix was added to 3 µl 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC, 3 mg ml−1 in acetonitrile) and the reaction incubated at 55 °C for 15 min. Samples were analysed on an Agilent Technologies 6130 Quadrupole LC–MS using single ion monitoring.
GFP(150X)–His6 and GFP(3X)–His6 fluorescence assay
Chemically competent DH10β cells harbouring a p15A plasmid encoding GFP(150TAG)His6 or GFP(3TAG)His6 (two versions of the p15A plasmid with either a tetracycline or apramycin resistance cassette, which led to similar results, were used interchangeably) were transformed with a pMB1 plasmid encoding MmPylRS, or a mutant thereof, and M. mazei \({{\rm{tRNA}}}_{{\rm{C}}{\rm{U}}{\rm{A}}}^{{\rm{P}}{\rm{y}}{\rm{l}}}\), rescued in SOC and grown overnight in 2xYTs-t or 2xYTs-ap. 20 μl of the overnight culture was diluted into 480 µl 2xYTs-t or 2xYTs-ap containing 0.2 l-arabinose in presence and absence of 2 to 4 mM of the respective ncM in a 96-well-plate format. Cells were grown for 16–20 h at 37 °C at 700 rpm The plates were centrifuged for 12 min, 4,200 rcf at 4 °C and the cells resuspended in 150 µl PBS. One-hundred microlitres of the resuspended cells were transferred into a Costar 96-well flat bottom plate and the OD600, and GFP fluorescence was measured using a PHERAstar FS plate reader.
GFP(150X)His6 isolation for mass spectrometry analysis
Three replicates of the protein produced as described above were combined in a 1.5 ml Eppendorf tube, centrifuged at 4,200 rcf for 3 min, frozen at −20 °C, thawed and resuspended in 150 µl BugBuster (Millipore). Cells were lysed for 1 h with head-over-tail rotation. Lysed cells were centrifuged for 20 min, 20 000 rcf, at 4 °C and the lysate was added to 20 µl of Ni-NTA beads. GFP(150X)–His6 was bound to the beads for 20 min at room temperature with head-over-tail rotation. The beads were washed 6 times with 60 µl 30 mM imidazole in PBS, and the protein was eluted with 5 times 30 µl 300 mM imidazole in PBS.
For lower-activity mutants for ncM 13, 5–15 ml of cell culture was used for protein production. The volumes of BugBuster were adjusted proportionally, all other volumes were kept the same.
Mass spectroscopy
ESI–MS and MS/MS were performed as previously described29,33.
Protein expression, purification and crystallization
Chemically competent DH10β cells harbouring a p15A plasmid encoding GFP(150TAG)His6 and a pMB1 plasmid encoding MmPylRS(13_1) and MmtRNAPyl were transformed, rescued in SOC and grown overnight in 2xYTs-ap. 10 ml of the overnight culture was diluted into 1 l 2xYTs-ap containing 0.2% l-arabinose in presence of 5 mM 13.
Bacterial pellet of 1 l expression culture of GFP(150(S)β3mBrF)–His6 was lysed by sonication, centrifuged at 142,000 rcf for 30 min and supernatant bound to Ni-NTA beads (Qiagen). Beads were washed three times before protein was eluted and further purified by gel filtration using a Superdex 75 HiLoad 26/60 pg column (GE Healthcare) in 25 mM Tris pH 7.4, 200 mM NaCl and 0.06% NaN3. The purified protein was concentrated using Vivaspin 20, 10,000 MWCO (Sartorius) to 6 mg ml−1. Sample was Trypsin digested with Sequencing Grade Modified Trypsin (Promega) in a 50:1 ratio. Sample was incubated for 1 h at 37 °C, centrifuged at 21,000 rcf for 10 min before plating in crystal trays. Crystallization trials with multiple commercial crystallization kits were performed in 96-well sitting-drop vapour diffusion plates (Molecular Dimensions) at 18 °C and set up with a Mosquito HTS robot (TTP Labtech). Drop ratios of 0.2 μl protein solution plus 0.2 μl reservoir solution were used for screening. The only useful dataset was collected from a crystal collected from the Fusion screen (Molecular Dimensions)60 with following composition: 37.5% PEG 3350/PEG 1 K/MPD (1:1:1), 0.1 M Bicine/Trizma pH 8.5, 0.8% (w/v) Morpheus III Alkaloids and 0.12 M Morpheus Alcohols. Crystals were collected and flash frozen in liquid nitrogen.
Diffraction data collection, processing and structure solution
Diffraction data were collected at the ESRF on beamline ID23-2 at an energy of 14.2 keV. Data were processed with XDS61 via the pipeline autoProc (Global Phasing)62. The structure was solved by molecular replacement with MolRep63 using the homologue model PDB 2B3P. Iterative building was performed with Coot64, refinement with REFMAC565, and validation with Molprobity66. Figures of the structure were prepared with PyMOL (PyMOL Molecular Graphics System, Schrödinger).
Selection for ncAAs
The selection was performed as described in Supplementary Fig. 11. RNA was isolated and oxidized as described in general procedure A. Bio–mREX was performed as specified in the general procedure. After the first round of selection, the new libraries were assembled from the amplified cDNA as described above. After the second round of selection the NGS samples were prepared from the isolated cDNA as described above, the NGS was run using a P2 600 cycles cartridge, and the data were analysed as specified above.
Selection for ncMs
The selections were performed as described in Supplementary Figs. 22 and 35. RNA was isolated and oxidized as described in general procedure A. Bio–mREX was performed as specified in the general procedure. The NGS samples were prepared from the isolated cDNA as described above, the NGS run using a P1 600 cycles cartridge, and the data were analysed as specified above.
Selection for substrate 13 using a random mutagenesis library
The concentrations of the pMB1 plasmids encoding PylRS hits 13_1 and 13_2 were measured by Qubit 2 Fluorometer (Life Technologies) and the Qubit 1x dsDNA HS Assay Kit (Invitrogen) and the plasmids combined in equimolar amounts. The combined plasmids were used for an error prone PCR of the active site of PylRS using golden gate primers (Supplementary Data 2) and the GeneMorph II kit (Agilent) at conditions leading to the maximal number of random mutations. The amplicons were cloned into a new pColE1 backbone by two-piece Golden Gate assembly according to NEB (New England Biolabs) guidelines. The selection was performed as outlined in Supplementary Fig. 33. RNA was isolated and oxidized as described in general procedure A. Bio–mREX was performed as specified in the general procedure. The NGS samples were prepared from the isolated cDNA as described above, the NGS was run using a P2 600 cycles cartridge, and the data were analysed as specified above.
Statistics
Graphpad Prism version 9 was used to generate all bar graphs in this study. For sequence alignments and further processing of the NGS data custom R scripts were used.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.