Pyrrolysine-Inspired in Cellulo Synthesis of an Unnatural Amino Acid for Facile Macrocyclization of Proteins

Abstract

Macrocyclization has been touted as an effective strategy to enhance the in vivo stability and efficacy of protein therapeutics. Herein, we describe a scalable and robust system based on the endogenous biosynthesis of a noncanonical amino acid coupled to the pyrrolysine translational machinery for the generation of lasso-grafted proteins. The in cellulo biosynthesis of the noncanonical amino acid d-Cys-ε-Lys was achieved by hijacking the pyrrolysine biosynthesis pathway, and then, its genetical incorporation into proteins was performed using an optimized PylRS/tRNA^Pyl pair and cell line. This system was then applied to the structurally inspired cyclization of a 23-mer therapeutic P16 peptide engrafted on a fusion protein, resulting in near-complete cyclization of the target cyclic subunit in under 3 h. The resulting cyclic P16 peptide fusion protein possessed much higher CDK4 binding affinity than its linear counterpart. Furthermore, a bifunctional bicyclic protein harboring a cyclic cancer cell targeting RGD motif on the one end and the cyclic P16 peptide on the other is produced and shown to be a potent cell cycle arrestor with improved serum stability.

Figure 1

Optimization of the d-Cys-ε-Lys readthrough system. (a) Chemical structure of d-Cys-ε-Lys. (b) mCherry readthrough assay comparing the original and the optimized UAG readthrough system. The lysine codon at position 55 in the mCherry gene was mutated to TAG, and the medium was supplemented with 0, 2, or 5 mM d-Cys-ε-Lys. Data represent mean fluorescence intensity ± standard error of the mean (n = 3). WT: wild type. M15: tRNA^M15. U: unmodified pPylST. tL: modified pPylST with the two T7lac promoters replaced by P_tac and PL_lacO1, respectively. R2: Rosetta 2 (DE3). C321: C321.ΔA.M9adapted. (c) Comparison of the original UAG readthrough system and the optimized one for d-Cys-ε-Lys incorporation into different recombinant proteins. X represents d-Cys-ε-Lys. CaM is short for calmodulin.

Figure 2

Engineering PylC for in vivo synthesis of d-Cys-ε-Lys. (a) Schematic representation of the enzymatic function of engineered PylC. (b) Screening of putative PylC mutants that could recognize d-cysteine and catalyze the production of d-Cys-ε-Lys based on mCherry fluorescence. Saturation mutagenesis was performed on four residues: S177, E179, D233, and T256 of PylC. Data represent mean fluorescence intensity ± standard error of the mean (n = 3). (c) Comparison of the wild type PylC (PylC^WT) and the evolved PylC mutant (PylC^NPSV) in the production of d-Cys-ε-Lys for its incorporation into UAG-containing proteins at different concentrations of d-cysteine. Sample readthrough was benchmarked against the readthrough protein produced by exogenous supplementation of 4 mM d-Cys-ε-Lys.

Figure 3

Computational analysis of cyclized P16 peptide. (a) Amino acid sequences of cyclic P16 peptide (cycP16p) showing site of d-Cys-ε-Lys incorporation (denoted by X) and linear P16p. The black bracket denotes d-Cys-ε-Lys linkage-mediated cyclization. (b) Model of cycP16p interacting with CDK6. The structure of p16p (cyan) and CDK6 (gray) complex was derived from PDB: 1BI7. CDK6-interacting residues were labeled with their numbers in P16 protein. (c) Comparison of MD simulation results between linear P16p (LinP16p) (left) and cycP16p (right). Figures are superposition of 10 rounds of LinP16p (left) and CycP16p (right) MD simulations at 10 ns. (d) RMSD and (e) RMSF profiles of LinP16p and CycP16p in 300 ns MD simulation. Results were calculated based on backbone atoms. Error bars present standard deviation (SD) of 3 replicate simulation runs and were plotted as shaded area in the RMSD profile.

Figure 4

Intramolecular cyclization of GFP-P16p. (a) Schematic illustration of the protein construct GFP-X-P16p-intein-CBD-His₇ and mechanism of d-Cys-ε-Lys-based protein cyclization. (b) SDS–PAGE of reaction samples taken at different time points during the cyclization of GFP-X-P16p. Shown are gels stained by coomassie blue (upper) and detected by in-gel GFP fluorescence (lower). (c) Deconvoluted mass spectrum of GFP-cycP16p obtained by ESI-Orbitrap mass spectrometry.

Figure 5

Cyclic P16p exhibits higher CDK4-binding affinity than its linear counterpart. Binding curves from MST assays of MBP-P16p and MBP-cycP16p with GST-CDK4 are shown. Error bars present standard deviation (SD) of three replicate measurements.

Figure 6

Model of the bicyclic cycRGD-mCh-cycP16p protein. RGD motif (marine blue) fused to the N-terminus of mCherry (gray) was cyclized by the formation of the disulfide bond between two cysteines. P16 peptide (cyan) fused to the C-terminus of mCherry was cyclized by the d-Cys-ε-Lys-based cyclization method.

Figure 7

Effects of cycRGD-mCh-cycP16p on MCF-7 cells. (a) MCF-7 cells were exposed to different treatments for 24 h followed by cell cycle analysis on a BD flow cytometer. 10 nM actinomycin was included as the positive control. (b) MCF-7 cells after 24 h treatment with different peptides. Cell numbers were normalized to the PBS control group. Data are presented as the mean ± SD (n = 3). Statistical significances versus cycRGD-mCh-P16p group were shown. (c) Percent of arrested MCF-7 cells at the G0/G1 phase after exposure to different treatments at different time points. Error bars present standard deviation (SD) of three replicate measurements. P values are calculated by one-way ANOVA test. *p < 0.05, ***p < 0.001, ****p < 0.0001, ns, not significant. (d) Western-blot analysis of the phosphorylation status of Rb in MCF-7 cells exposed to different treatments using anti-pRb antibodies.