Linker and N-Terminal Domain Engineering of Pyrrolysyl-tRNA Synthetase for Substrate Range Shifting and Activity Enhancement

Abstract

The Methanosarcina mazei pyrrolysyl-tRNA synthetase (PylRS)⋅tRNA^Pyl pair can be used to incorporate non-canonical amino acids (ncAAs) into proteins at installed amber stop codons. Although engineering of the PylRS active site generates diverse binding pockets, the substrate ranges are found similar in charging lysine and phenylalanine analogs. To expand the diversity of the ncAA side chains that can be incorporated via the PylRS⋅tRNA^Pyl pair, exploring remote interactions beyond the active site is an emerging approach in expanding the genetic code research. In this work, remote interactions between tRNA^Pyl, the tRNA binding domain of PylRS, and/or an introduced non-structured linker between the N- and C-terminus of PylRS were studied. The substrate range of the PylRS⋅tRNA^Pyl pair was visualized by producing sfGFP-UAG gene products, which also indicated amber suppression efficiencies and substrate specificity. The unstructured loop linking the N-terminal and C-terminal domains (CTDs) of PylRS has been suggested to regulate the interaction between PylRS and tRNA^Pyl. In exploring the detailed role of the loop region, different lengths of the linker were inserted into the junction between the N-terminal and the C-terminal domains of PylRS to unearth the impact on remote effects. Our findings suggest that the insertion of a moderate-length linker tunes the interface between PylRS and tRNA^Pyl and subsequently leads to improved suppression efficiencies. The suppression activity and the substrate specificity of PylRS were altered by introducing three mutations at or near the N-terminal domain of PylRS (N-PylRS). Using a N-PylRS⋅tRNA^Pyl pair, three ncAA substrates, two S-benzyl cysteine and a histidine analog, were incorporated into the protein site specifically.

Keywords: amber codon suppression; linker engineering; non-canonical amino acids; pyrrolysyl-tRNA synthetase; tRNA binding domain.

FIGURE 1

Protein designs of evolved MmPylRS and their linker engineering. (A) Illustration of the interaction between MmPylRS and tRNA^Pyl. R61K/H63Y/S193R mutations (labeled in open red circles) are distributed on the tRNA binding domain of MmPylRS NTD or CTD (labeled as filled blue circle). The detailed interactions from X-ray crystal structures in PylRS-NTD/tRNA (Suzuki et al., 2017) and PylRS-CTD/tRNA (Nozawa et al., 2009) are indicated in Figures 2, 3. tRNA 2D topology and sequence are labeled in green filled circles. The orange dashed line denotes the inserted linker between P149 and A150. (B) Engineered MmPylRS constructs used in this work. N-PylRS represents MmPylRS with R61K/H63Y/S193R mutations; PylRS-D1 represents MmPylRS with insertion of the spacer containing the stop and the start codons between P149 and A150; PylRS-L1 indicates the insertion of SGGGGS linker between P149 and A150; PylRS-L2 for S(GGGGS)₂ linker insertion; PylRS-L3 for S(GGGGS)₃ linker insertion.

SCHEME 1

Chemical structures of the non-canonical amino acids described in this study. ncAA 1: N^ε-(tert-butoxycarbonyl)-L-lysine (BocK); 2: N^ε- (alloyloxycarbonyl)-L-lysine (AlloK); 3: S-(4-methylbenzyl)-L-cysteine (MbzlC); 4: S-(4-methoxylbenzyl)-L-cysteine (MeObzlC); 5: N-π-methyl-L-histidine (3MeH); 6: N^ε-(carbobenzyloxy)-L-lysine (CbzK); 7: N^ε-(2-chlorocarbobenzyloxy)-L-lysine (ClCbzK); 8: N^ε-(carbobenzyloxy)-D-lysine (DCbzK); 9: N^ε-(2-chlorocarbobenzyloxy)- D-lysine (DClCbzK); 10: (R)-2-amino-3-(2-benzyloxycarbonylaminoethylselanyl)propanoic acid (SeCbzK); 11: (S)-6-(benzyloxycarbonylamino)-2-hydroxygexanoic acid (CbzKOH).

FIGURE 2

Structure of MmPylRS NTD complex with tRNA^Pyl. R61 and H63 located in the PylRS NTD; the yellow dashed lines represent potential hydrogen-bonding interactions. Three such interactions are illustrated between the side chain of R61 and R52 within PylRS NTD and the phosphodiester backbone of 58A and 59U in tRNA^Pyl. One hydrogen-bonding interaction was found between the side chain of H63 and K85 within the PylRS NTD. The structure is based on the PDB entry 5UD5.

FIGURE 3

Structure of MmPylRS CTD and its superimposition with the DhPylRS CTD/tRNA^Pyl complex. (A) S193 is located in the tRNA binding domain of MmPylRS CTD (green). (B) The superimposed structure of MmPylRS CTD (green) and DhPylRS CTD⋅DhtRNA^Pyl complex (cyan). The structures are based on the PDB entry 2Q7H and 2ZNI.

FIGURE 4

The sfGFP-UAG2 and sfGFP-UAG27 gene suppression efficiencies of PylRS enzyme variants. Incorporation efficiencies of PylRS variants (Figure 1B) as measured by fluorescence intensities of sfGFP with an amber mutation at position 2 (A) and 27 (B). The proteins were expressed in 1 mM ncAA and IPTG in GMML medium at 37°C for 12 h. The cells were excited at 485 nm and the fluorescence intensities were detected at 535 nm. The cell density was monitored by absorbance at 595 nm. “C” indicates the Control experiments of cells with the supplement of 1 mM IPTG; 1–5 denote the supplement of 1 mM IPTG and ncAA 1–5 (Scheme 1). The background signals from cells without adding IPTG were subtracted from each group. The error bars represent the standard deviation of sfGFP production from four repeated experiments.

FIGURE 5

The sfGFP-UAG2 and sfGFP-UAG27 gene suppression efficiencies of ZRS enzyme variants. Incorporation efficiencies of ZRS variants (Figure 1B) as measured by the fluorescence intensities of sfGFP with amber mutation at position 2 (A) and 27 (B). The proteins were expressed in 1 mM ncAA and IPTG in GMML medium at 37°C for 12 h. The cells were excited at 485 nm and the fluorescence intensities were detected at 535 nm. The cell density was monitored by absorbance at 595 nm. “C” indicate the Control experiments of cells with the supplement of 1 mM IPTG; 3, 4, 6–11 denote the supplement of 1 mM IPTG and ncAA 3, 4, 6–11 (Scheme 1). Error bars represent the standard deviation of sfGFP production from four repeated experiments.

FIGURE 6

sfGFP production by ZRS variants and mass characterization. (A) Amber suppression of the sfGFP-UAG27 gene (N-ZRS⋅tRNA^Pyl pair) and the sfGFP-UAG2 gene (ZRS-D1⋅tRNA^Pyl pair) product with ncAA 3–4 and 6–11. The sfGFP proteins were produced in E. coli BL21 (DE3) coding N-ZRS⋅tRNA^Pyl or ZRS-D1⋅tRNA^Pyl pair with the supplement of 1 mM IPTG and ncAAs in GMML medium at 37°C for 12 h. The whole-cell lysate was analyzed by SDS-PAGE and western blotting by anti-His tag antibody indicated as α-His6X. (B) ESI-MS determination of sfGFP-UAG27 proteins with ncAA 6, 7, and 11. Full-length sfGFP-6 and sfGFP-7 proteins were produced by N-ZRS⋅tRNA^Pyl pair in E. coli BL21 (DE3) with the supplement of 1 mM IPTG and ncAA 6 or 7 in LB medium at 37°C for 12 h. Full-length sfGFP-11 proteins were produced with the same condition but with 1 mM ncAA 11 and in GMML minimal medium. The calculated molecular masses of sfGFP-7 are 28,121 and 27,989 Da (–Met); the observed molecular masses are 28,120 and 27,989 Da (–Met). The calculated molecular masses of sfGFP-6 are 28,085 and 27,954 Da (–Met); the observed molecular masses are 28,085 and 27,955 Da (–Met). The calculated molecular masses of sfGFP-11 are 28,086, 27,955 (–Met), and 25,320 Da (truncated sfGFP at 27 position); the observed molecular masses are 27,955 (–Met) and 27,820 Da (without Cbz group at 27 position and N-terminal Met residues) and 25,036 Da. The detailed electrospray and deconvoluted mass spectra are shown in Supplementary Figures S4–S6. ESI-MS determination of sfGFP-UAG2 protein. sfGFP-11* with ncAA 11 (ZRS-D1⋅tRNA^Pyl pair) is shown in Supplementary Figure S7.

FIGURE 7

Deconvoluted ESI-MS spectra of sfGFP containing ncAAs at position 27. Full-length sfGFP-3, sfGFP-4, and sfGFP-5 were expressed using the N-PylRS⋅tRNA^Pyl pair in the presence of ncAA 3, 4 (2 mM) or 5 (1 mM), respectively, in GMML medium (sfGFP-3 and sfGFP-4) or LB medium (sfGFP-5). The purified protein samples were subjected to ESI-MS for identification of the incorporation of defined ncAA. Deconvoluted ESI-MS of ncAA 3–5 incorporated sfGFP at position 27 are shown above. sfGFP-X in which X represents one of the ncAA 3–5 was incorporated into sfGFP at position 27. The calculated molecular masses of sfGFP-3 are 28,030 and 27,899 Da (-Met); the observed molecular masses are 28,030 and 27,899 Da (-Met). The calculated molecular masses of sfGFP-4 are 28,046 and 27,915 Da (-Met); the observed molecular masses are 28,046 and 27,915 Da (-Met). The calculated molecular masses of sfGFP-5 are 27,974 and 27,843 Da (-Met); the observed molecular masses are 27,974 and 27,843 Da (-Met). The detailed electrospray and deconvoluted mass spectra are shown in Supplementary Figures S1–S3.

FIGURE 8

MALDI-TOF-MS/MS analysis of sfGFP-3 and sfGFP-5. The sfGFP-3 and sfGFP-5 proteins were in-gel digested with trypsin and subjected to MALDI-TOF-MS/MS analysis. (A) The tandem mass spectrum of the X(3)SVR; X denotes a fragment from sfGFP-3, with ncAA 3 incorporated at position 27. The calculated molecular mass of the X(3)SVR peptide fragment is 567.284 Da and the actual molecular mass is 568.327 Da. Full-length sfGFP-3 was expressed using N-PylRS in the presence of 2 mM ncAA 3 in GMML medium. (B) The tandem mass spectrum of the X(5)SVR; X denotes a fragment from sfGFP-5 with ncAA 5 incorporated at position 27. The calculated molecular mass of the X(5)SVR peptide fragment is 511.287 Da and the actual molecular mass is 512.609 Da. Full-length sfGFP-5 was produced using N-PylRS⋅tRNA^Pyl pair in the presence of 1 mM ncAA 5 in LB medium.