Pyrrolysyl-tRNA Synthetase Activity can be Improved by a P188 Mutation that Stabilizes the Full-Length Enzyme

Abstract

The amber suppression-based noncanonical amino acid (ncAA) mutagenesis technique has been widely used in both basic and applied research. So far more than two hundred ncAAs have been genetically encoded by amber codon in both prokaryotes and eukaryotes using wild-type and engineered pyrrolysyl-tRNA synthetase (PylRS)-tRNA^Pyl (PylT) pairs. Methanosarcina mazei PylRS (MmPylRS) is arguably one of two most used PylRS variants. However, it contains an unstable N-terminal domain that is usually cleaved from the full-length protein during expression and therefore leads to a low enzyme activity. We discovered that the cleavage takes place after A189 and this cleavage is inhibited when MmPylRS is co-expressed with Ca. Methanomethylophilus alvus tRNA^Pyl (CmaPylT). In the presence of CmaPylT, MmPylRS is cleaved after an alternative site K110. MmPylRS is active toward CmaPylT. Its combined use with CmaPylT leads to enhanced incorporation of N^ε-Boc-lysine (BocK) at amber codon. To prevent MmPylRS from cleavage after A189 in the presence of its cognate M. mazei tRNA^Pyl (MmPylT), we introduced mutations at P188. Our results indicated that the P188G mutation stabilizes MmPylRS. We showed that the P188G mutation in wild-type MmPylRS or its engineered variants allows enhanced incorporation of BocK and other noncanonical amino acids including N^ε-acetyl-lysine when they are co-expressed with MmPylT.

Graphical Abstract

Introduction

Pyrrolysine (Pyl) is the 22^nd proteinogenic amino acid that is genetically encoded by the amber UAG codon. Its genetic encoding requires pyrrolysyl-tRNA synthetase (PylRS) and its cognate tRNA tRNA^Pyl (PylT).^1–3 The Pyl incorporation system was discovered in 2002 and has been developed as the arguably most used system for undergoing amber suppression-based noncanonical amino acid (ncAA) mutagenesis in cells.^4–8 So far more than 200 different ncAAs have been genetically encoded using wild-type and engineered PylRS-PylT pairs.⁹ These ncAAs have a large variety of chemical functionalities and some of them are posttranslationally modified amino acids.^{6, 10–29} Applications of these genetically encoded ncAAs include the synthesis of proteins with posttranslational modifications for their functional investigations and the engineering of proteins for various purposes such as the synthesis of antibody-drug conjugates.

The widely used PylRS strains for genetic code expansion are from two archaebacteria, Methanosarcina mazei and Methanosarcina bakeri. They share 83% sequence identity.³⁰ However, both PylRSs are unstable due to an unstable N-terminal domain (NTD) that separates from the C-terminal catalytical domain (CTD). NTD is known to recognize PylT for improved binding.^{31, 32} However, its easy aggregation leads to an insoluble full-length PylRS or is cleaved from CTD that itself is not catalytically active.³² To overcome this instability issue, several evolution strategies have been used to optimize PylRS. For example, a more stable and active tetra-mutation IPYE clone was identified for M. bakeri PylRS (MbPylRS) and a tri-mutation R3–11 clone was introduced for M. mazei PylRS (MmPylRS)^{33, 34}. An extreme solution to this problem is the total elimination of NTD, a phenomenon that occurs naturally in Ca. Methanomethylophilus alvus. Ca. M. alvus PylRS (CmaPylRS) has only a CTD domain that shares 80% similarity with the MmPylRS CTD.^35–38

In comparison to MmPylRS and MbPylRS, CmaPylRS has much improved solubility. When expressed in Escherichia coli, full-length MmPylRS was only observed in the cell lysate pellet whereas CmaPylRS was primarily in the cell lysate supernatant (supplementary S1). Due to this feature, there has been a trend to switch from using MmPylRS and MbPylRS to using CmaPylRS to conduct genetic code expansion research. However, the majority of ncAAs are genetically encoded using engineered MmPylRS and MbPylRS variants. Mutations in these engineered MmPylRS and MbPylRS variants cannot be simply transferred to CmaPylRS to afford variants for the genetic incorporation of corresponding ncAAs. We have introduced mutations that were found in MmAcKRS1 and MmBuKRS into CmaPylRS to generate corresponding CmaAcKRS and CmaBuKRS. However, afforded CmaAcKRS and CmaBuKRS clones were not able to mediate the incorporation of N^ε-acetyl-L-lysine (AcK) and N^ε-crotonyl-L-lysine (CrK), respectively (supplementary S2). Therefore, finding a way to stabilize MmPylRS and MbPylRS is still in need to improve current existing systems for the genetic incorporation of ncAAs. Chin et al has recently shown that CmaPylRS can be combined with MmPylRS for the genetic incorporation of two different ncAAs at two separate codons when they are used together with an evolved Ca. M. alvus PylT (CmaPylT) that doesn’t cross-interact with MmPylRS and M. mazei PylT (MmPylT) that doesn’t cross-interact with CmaPylRS.³⁹ This type of combined uses of PylRS clones will also demand existing MmPylRS and MbPylRS clones to be improved. In this work, we wish to report our finding that a P188G mutation in MmPylRS can significantly stabilize MmPylRS for improved ncAA incorporation.

Results and Discussions

CmaPylT enhances MmPylRS-mediated amber suppression.

We started our research by testing the orthogonality of MmPylRS and CmaPylRS toward each other’s cognate PylTs. Our original purpose was to examine whether these independent pairs can be combined to genetically encode two different ncAAs. To test orthogonality of MmPylRS toward CmaPylT, we constructed a plasmid pEVOL-MmPylRS-CmaPylT that encodes MmPylRS and CmaPylT (Fig. 1A). This plasmid was used along with a reporter plasmid pBAD-sfGFP134TAG that encodes superfolder green fluorescent protein (sfGFP) with an amber mutation at the 134^th amino acid coding position (sfGFP134TAG) to transform E. coli Top10 cells. Transformed cells were growth in the presence of 1 mM N^ε-Boc-lysine (BocK). Without amber suppression, the sfGFP134TAG gene was expressed as a truncated protein with a premature termination at the amber codon site. Otherwise, full-length sfGFP that was quantified by measuring fluorescence was expressed to indicate amber suppression levels. After 6 h expression, cells were collected and lysed to quantify expressed full-length sfGFP. A similar experiment in which a plasmid pEVOL-MmPylRS-MmPylT that encodes MmPylRS and MmPylT was used as a positive control. MmPylRS showed an amber suppression effect in the presence of CmaPylT slightly more than twice better than that in the presence of MmPylT (Fig. 1B). To explore whether this enhanced expression can be transferred to other MmPylRS variants, we paired CmaPylT with engineered MmPylRS variants including MmAcKRS2, MmAcdKRS, MmAznLRS, and MmBuKRS (Mm was added in the names for these variants to indicate their origin) that were previously evolved for the genetic incorporation of AcK, N^ε-(4-azidobenoxycarbonyl)-δ,ε-dehydrolysine (AcdK), azidonorleucine (AznL), and N^ε-crotonyl-lysine (CrK) respectively.^{13, 17, 18} Plasmids similar to pEVOL-MmPylRS-CmaPylT that encode different MmPylRS variants and CmaPylT were constructed. They were used along with pBAD-sfGFP134TAG to transform Top10 cells for the expression of full-length sfGFP in the presence of 1 mM corresponding ncAA. Their corresponding MmPylRS variant-MmPylT pairs were used as positive controls. Among four ncAAs, only AcdK and CrK showed enhanced incorporation when their specific MmPylRS variants were paired with CmaPylT. AcK and AznL exhibited reduced incorporation (Fig. 1C). Therefore, it is evident that the binding of CmaPylT and its aminoacylation by MmPylRS are two intricately related processes. The binding of a ncAA and its catalysis to form an aminoacyl-CmaPylT might require unique structural dynamics in the binding between MmPylRS and CmaPylT that varies between different ncAAs. Therefore, Not all engineered MmPylRS variants work well with CmaPylT.

Figure 1.

(A) Clove leaf structures of MmPylT and CmaPylT. (B) The amber suppression efficiency of MmPylRS in the presence of CmaPylT is significantly better than that in the presence of MmPylT (p-value < 0.05). (C) CmaPylT doesn’t lead to improved activities in all engineered MmPylRS variants. 1 mM ncAA was used to induce full-length sfGFP expression (p-value < 0.05 except the one from AznLRS-CmaPylT in the presence of AznL was 0.3).

CmaPylT prevents MmPylRS from its cleavage after A189.

To investigate the role of CmaPylT in the MmPylRS-mediated amber suppression enhancement, we lysed cells that expressed MmPylRS in the presence of either CmaPylT or MmPylT and analyzed soluble proteins by SDS-PAGE (Figure 2A). In the presence of MmPylT, a truncated soluble MmPylRS fragment whose molecular weight was around 30 kDa was clearly visible. However, this soluble MmPylRS fragment disappeared when CmaPylT was co-expressed. In the presence of CmaPylT, a soluble protein band with a molecular weight around 38 kDa appeared in the SDS-PAGE gel. We purified both soluble proteins and characterized their molecular weights using electrospray ionization mass spectrometry (ESI-MS). The deconvoluted molecular weights of the two proteins are 30,673.3 and 38,335.0 Da, respectively (Supplementary Fig. S3). MmPylRS is known to undergo hydrolysis in E. coli and only its CTD is soluble. These two determined molecular weights match well with MmPylRS fragments aa190-454 (theoretical molecular weight: 30672.2 Da) and aa111-454 (theoretical molecular weight: 38,336.7 Da). The MmPylRS NTD domain is aa1-189. This domain is known to bind PylT but how this domain folds is only partially studied. Although the crystal structure of aa1-100 has been determined,³¹ the structure of aa101-189 is unknown. Our results indicate that aa111-189 is highly likely an independently folded domain and this domain interacts directly with CmaPylT. This interaction stabilizes the MmPylRS CTD fusion with aa111-189 and consequently contributes to a strong activity of MmPylRS in the presence of CmaPylT. However, how exactly CmaPylT prevents the cleavage of MmPylRS after A189 needs to be further investigated.

Figure 2.

(A) MmPylRS showed different truncated fragments in the presence of MmPylT and CmaPylT. Arrows indicate truncated fragments aa190-454 and aa111-454. (B) Different MmPylRS clones used in combination with CmaPylT for testing amber suppression efficiency. Split means that CTD and NTD were expressed separately (p-value < 0.05). (C) Amber suppression efficiency of different MmPylRS clones in the presence of CmaPylT. (D) Purified full-length sfGFP that was expressed when different MmPylRS clones were combined with CmaPylT. 1 mM BocK was used to induce full-length sfGFP expression.

Full-length MmPylRS is essential for an enhanced activity.

To examine effects of different domains on the MmPylRS activity, we constructed plasmids encoding MmPylRS fragments and CmaPylT. Three designs were conducted. One contained just the MmPylRS CTD, the second had two fragments aa1-110 and aa111-454 expressed separately, and the third encoded independently expressed aa111-454 and aa1-110 that was fused with sfGFP with a Y66F mutation (Fig. 2B). A previous publication has shown that two split fragments from a PylRS chimera can be co-expressed to improve genetic incorporation of a ncAA.⁴⁰ We hoped to recapitulate this result. Two independently expressed CTD and NTD are also observed in Desulfitobacterium hafinience and other organisms.³⁰ In these organisms, CTD, NTD and PylT form a tertiary complex and NTD is required for the in vivo activity of the whole complex. sfGFP is known to improve solubility and folding of its fused proteins.^{41, 42} The purpose of its fusion to aa1-110 was to improve the solubility of aa1-110 in E. coli so that a more soluble complex between aa111-454 and aa1-110 could be generated to improve genetic incorporation of a ncAA. To prevent fluorescence of the fused sfGFP from interfering with the characterization of fluorescence-based amber suppression levels, we mutated its active site Y66 to F, which eliminates its fluorescence.⁴³ aa1-110 instead of aa1-189 was used since CmaPylT stabilizes the MmPylRS CTD fused with aa110-aa189. The plasmid that encoded MmPylRS and MmPylT was used as a control. These plasmids along with pBAD-sfGFP134TAG were used to transform E. coli Top10 cells. Transformed cells were grown in the presence of 1 mM BocK for the characterization of amber suppression levels. As shown in Figure 2C, removing the NTD domain from MmPylRS totally killed its amber suppression activity even in the presence of CmaPylT. No expressed full-length sfGFP was detected (supplementary Fig. S4). Cells expressing split NTD and CTD domains resumed partially amber suppression activity. Although this activity was close to the activity from the MmPylRS-MmPylT pair, it was only about 40% of that from the MmPylRS-CmaPylT pair. The fusion of sfGFP-Y66F to NTD improved the overall amber suppression activity. However, this improved activity was only about 60% of that from the MmPylRS-CmaPylT pair. Collectively, our results approve that the expression of full-length MmPylRS is critical for a high amber suppression activity even when it is coupled with CmaPylT. Since we did not observe an obvious band for full-length MmPylRS in the presence of either MmPylT or CmaPylT in our SDS-PAGE analysis of the cell lysate supernatants as shown in Figure 2A, we suspected that full-length MmPylRS in the presence of CmaPylT might aggregate during the cell lysis process. To confirm this, we directly dissolved cells expressing MmPylRS and CmaPylT in the SDS-PAGE sample loading buffer, boiled the sample and analyzed proteins in whole cells by SDS-PAGE. As shown in Supplementary Fig. S5, the SDS-PAGE gel showed a heavy band corresponding to full-length MmPylRS. However, this band was not observable in cells expressing MmPylRS and MmPylT. Therefore, it is clear that CmaPylT stabilizes full-length MmPylRS. This stabilization likely contributes to the overall high activity of MmPylRS in its mediated BocK incorporation at amber codon in the presence of CmaPylT. How exactly CmaPylT stabilizes MmPylRS will demand more structural analysis of the interactions between them.

K110A, P188Q and P188G enhance the activity of MmPylRS.

Since the activity enhancement of MmPylRS by CmaPylT cannot be simply transferred to engineered MmPylRS variants, we sought other ways to improve activity of engineered MmPylRS variants that can be used together with MmPylT. Our investigation of MmPylRS stabilization by CmaPylT showed that MmPylRS is cleaved after K110 and A189 in E. coli. These two cleavages are presumably done by E. coli proteases. K110 is a lysine, which is a typical protease recognition site.⁴⁴ A189 is not a unique protease recognition site.⁴⁵ However, its N-terminal proximal residue is P188. Proline and glycine are known as protein secondary structure breakers.⁴⁶ P188 Mutations likely generate large structure changes that disrupt protease recognition at A189. Therefore, we decided to conduct mutagenesis at K110 and P188. To neutralize the charge of K110, we introduced a K110A mutation and for P188 we introduced two mutations P188G and P188Q. The P188G mutation was introduced with the purpose to maintain a possible turn structure and P188Q was introduced to improve the solubility of the P188-containing peptide. Clones with three separate single mutations and that with double mutations including K110A/P188G and K110A/P188Q were constructed. These mutated clones were coupled with MmPylT to test their amber suppression activity. For comparison, we only constructed the K110A mutant clone of MmPylRS for its coupled use with CmaPylT. Only the K110A mutation was introduced because CmaPylT prevents the MmPylRS cleavage after A189. Plasmids coding these mutants along with either MmPylT or CmaPlyT were built. There were coupled with pBAD-sfGFP134TAG to transform E. coli Top10 cells. Transformed cells were grown in the presence of 1 mM BocK for the characterization of overall full-length sfGFP expression levels. The K110A mutation improved the activity of MmPylRS in the presence of CmaPylT for about 10% but this improvement was not significant (Supplementary Fig. S6). An SDS-PAGE analysis of cells expressing MmPylRS K110A and CmaPylT showed an enhanced level of full-length MmPylRS in the cell lysate pellet compared to cells expressing wild-type MmPylRS and CmaPylT (Supplementary Fig. S7). Cells expressing MmPylRS K110A and CmaPylT had also close to non-visible truncated MmPylRS fragement aa111-454 in the cell lysate supernatant. On the contrary, cells expressing wild-type MmPylRS and CmaPylT had a high level of truncated MmPylRS fragment aa111-454 (Supplementary Fig. S7). In the presence of MmPylT, all three mutations significantly improved the activity of MmPylRS (Figure 3A). Although clones with double mutations showed higher activities than wild-type MmPylRS, their activities were lower than that from single mutation clones. It is obvious that mutations at the two cleavage sites influence the MmPylRS activity. Although our results showed that tested mutations at two sites cannot be synergistically used likely due to their combined disruption to the overall protein folding, it is highly possible that other combinations of mutations at two sites will lead to better activities in afforded enzyme variants than single mutation clones.

Figure 3.

(A) Amber suppression efficiency for different MmPylRS mutants in the presence of MmPylT (p-value < 0.05). (B) Transfer of the P188G mutation to MmAcKRS1, MmAcdKRS and MmAznLRS to improve their amber suppression efficiency in the presence of MmPylT. 1 mM ncAA was used for expression (p-value < 0.05).

The enzyme activity enhancement effect of P188G can be transferred to engineered MmPylRS variants.

To explore whether we can transfer the activity improvement effect of the P188 mutation to engineered MmPylRS variants, we introduced P188G into MmAcKRS1, MmAcdKRS and MmAznLRS, the three MmPylRS clones that we have frequently used to express proteins with posttranslational lysine modifications. Plasmids that encode these clones with the P188G mutation and MmPylT were constructed. They were used together with pBAD-sfGFP134TAG to transform E. coli Top10 cells. Transformed cells were grown in the presence of 1 mM corresponding ncAAs. As shown in Figure 3B, the P188G mutation led to 3-fold improvement of AcK incorporation by MmAcKRS1. The genetic incorporation of AcK by amber codon has been broadly used for the characterization of posttranslational lysine acetylation functions. However, the approach has suffered from a low incorporation yield. To improve AcK incorporation, we have typically used 5 or 10 mM AcK for expression.^{47, 48} The dramatic improvement of AcK incorporation by a single P188G mutation will likely promote broader adoption of the technique by removing a significant hurdle. We have also tested the synthesis of histone H3 K36ac using the MmAcKRS1 P188G clone in combination with MmPylT. In comparison to almost non-detectable H3 K36ac that was expressed using the original MmAcKRS1, MmAcKRS1 P188G led to a high expression level of H3 K36ac in the presence of 5 mM AcK (Supplementary Fig. S8). The P188G mutation did not change the MmAcdKRS activity very much. It led to an arguably same activity as the original MmAcdKRS. Since MmAcdKRS is an efficient enzyme, this is not a surprise. The P188G mutation led to a better performing MmAznLRS. However, the activity improvement in MmAznLRS is not as significant as that was shown in MmAcKRS1. The aminoacylation catalysis requires coordination between aminoacyl-tRNA synthetase and several substrates including ATP, amino acid and tRNA. How exactly a mutation will influence interactions between the enzyme and three substrates and the catalytic process itself is difficult to perceive. However, based on all our collected data, we can conclude that the P188G mutation will likely improve activity of an MmPylRS variant although exact effects are hard to predict.

Conclusions

CmaPylT stabilizes MmPylRS in E. coli by preventing its cleavage after A189 although MmPylRS is still prone to the hydrolysis after K110. Mutations at K110 and P188 in general improve the activity of MmPylRS in the presence of MmPylT. The P188G mutation can also be potentially transferred to engineered MmPylRS variants to improve their activities for the coupled use with MmPylT.

Materials and Methods

Construction of plasmids encoding wild-type MmPylRS or a variant together with CmaPylT

The pEVOL-MmPylRS-CmaPylT plasmid was constructed by replacing CmaPylRS in the pEVOL-CmaPylRS-CmaPylT plasmid with megaWhop PCR. The MmPylRS fragment was amplified from the pEVOL-MmPylRS-mmPylT plasmid using two primers MmPylRS-for and MmPylRS-rev (supplementary table 1). The amplified MmPylRS fragment was then used as a primer to amplify pEVOL-CmaPylRS-CmaPylT. Plasmids pEVOL-MmAcKRS1-CmaPylT, pEVOL-MmAcdKRS-CmaPylT, and pEVOL-MmAznLRS-CmaPylT were cloned in the same way.

Construction of plasmids encoding MmPylRS CTD, split MmPylRS CTD and NTD, and split MmPylRS CTD and sfGFP-Y66F-NTD

The split construct was made by PCR amplification of the whole construct pEVOL-MmPylRS-CmaPylT using two primers split-for and split-rev primer pairs that added a stop codon, a ribosome binding site, and a start codon between MmPylRS 1–110 and 111–454. The sfGFPY66F-NTD split construct was then made using megawhop PCR from the construct coding split CTD and NTD. The sfGFP fragment was amplified from the plasmid pBAD-sfGFP using a primer pair sfGFP-for and sfGFP-rev. The amplified fragment was then used as a primer to amplify the split construct. A site directed mutagenesis was performed to the sequence-confirmed sfGFP-NTD split construct using two primers Y66F-for and Y66F-rev to generate a non-fluorescent sfGFP-Y66F-NTD. The plasmid encoding only the CTD domain was generated by PCR of pEVOL-MmPylRS-CmaPylT using a primer pair CTD-for and CTD-rev to remove the whole NTD region.

Construction of plasmids containing K110A, P188G, P118Q and double mutations in MmPylRS

Plasmid pEVOL-MmPylRS-K110A was made by PCR using a primer pair CTD-for and K110A-rev from a template pEVOL-MmPylRS-MmPylT. The pEVOL-MmPylRS-P188G-MmPylT and pEVOL-MmPylRS-P188Q-MmPylT was made from pEVOL-MmPylRS-MmPylT using primer pairs p188G-for/p188G-rev and p188Q-for/P188Q-rev, respectively. Double mutations K110A/P188Q or P188G were constructed in a same way. The P188G mutated plasmids of MmAcKRS1, MmAcdKRS, and MmAznLRS were made from their corresponding constructs pEVOL-MmAcKRS1-MmPylT, pEVOL-MmAcdKRS-MmPylT and pEVOL-MmAznLRS-MmPylT by site-directed mutagenesis using two primers P188G-for and P188G-rev.

Purification of MmPylRS

Transformed cells were grown in 10 mL 2YT media (Ampicillin and Chloramphenicol) at 37 °C overnight. The overnight culture was inoculated into 1L of 2YT and grown at 37 °C until OD₆₀₀ reached 0.6. The expression of MmPylRS was induced by the addition of 0.2 % arabinose for 6 h at 37 °C. Cells were collected by centrifuging at 6K rpm for 15 min. The cell pellet was resuspended in lysis buffer (50 mM HEPES, 500 mM NaCl, 10 % glycerol and 1 mM PMSF at pH 7.5). Suspended cells were lysed by sonication with 1 sec on and 4 sec off for a total of 5 min. The lysate was removed by centrifugation at 16K rpm for 30 min at 4 °C. Ammonium sulfate precipitation was performed for the supernatant. Precipitation from 0% to 40% was collected by centrifuging at 16K rpm for 30 min. The pellet was dissolved to buffer A (50 mM HEPES, 100 mM NaCl and 10% glycerol at pH 7.5) and dialyzed against buffer A overnight. Solubilized pellet was then filtrated through 0.45 μm filter and loaded onto a source 15Q column for purification by FPLC. The protein was eluted with a gradient from buffer A to buffer B (50 mM HEPES, 1 M NaCl and 10% glycerol at pH 7.5). The desired peak was collected, concentrated, and loaded onto a superdex 200 Increase 10/300 GL column for further purification.

Quantification of amber codon suppression

Transformed cells were grown in 6 mL 2YT media at 37 °C until they reached OD₆₀₀ as 0.6. Cells were then were induced with 0.2% arabinose and split to two 3 mL cultures induced with and without 1 mM ncAA. After 6 h of expression at 37 °C, cells were collected. 25 μL of the culture was diluted to 100 μL on a 96 well plate. OD₆₀₀ was measured in a BioTek synergy neo2 plate reader. The remaining cells were spun down with 4K rpm for 15 min. The pellet was collected and resuspend in 0.5 mL lysis buffer (50 mM Tris, 50 mM NaCl, 1 mM PMSF and 1 mg/mL lysozyme at pH 8). Cells were lysed by freezing and thawing between liquid nitrogen and 42 °C water bath with 20 sec vortexing after thawing. The lysate was clarified by centrifugation at 14K rpm for 30 min. 100 uL of the supernatant was transferred to a Greinier half area black 96-well plate. The fluorescence of sfGFP was quantified using a BioTek synergy neo2 plate reader with 485 nm excitation and 510 emission. The supernatant was collected and incubated with 250 uL Ni-NTA resins for 10 min. The Ni resins were washed with 1 mL of wash buffer (50 mM Tris, 50 mM NaCl and 30 mM imidazole at pH8). The bound sfGFP was then eluted with elution buffer (50 mM Tris, 50 mM NaCl and 300 mM imidazole at pH8) after rocking in room temperature for 10 min. The eluent was then analyzed with SDS-PAGE.

Highlights.

Methanosarcina mazei pyrrolysyl-tRNA synthetase (PylRS) displays enhanced activity when it is coupled with Ca. Methanomethylophilus alvus tRNA^Pyl. This enhanced activity is due to the stabilization effect of Ca. M. alvus tRNA^Pyl toward M. mazei PylRS. It prevents M. mazei PylRS from cleavage after A189. By introducing P188 mutations to M. mazei PylRS, we showed that activity of M. mazei PylRS and its engineered variants can be generally improved when they are coupled with the cognate tRNA^Pyl from M. mazei. This discovery will be generally useful to develop systems for enhanced incorporation of noncanonical amino acids at amber codon.