Rational Design and Modification of NphB for Cannabinoids Biosynthesis

2.2. Engineering of NphB for Cannabinoids Biosynthesis

Based on the docking results of the enzyme with the substrates, 54 residues within a 6 Å range of oli and GPP were selected for preliminary analysis as part of the substrate-binding pocket. Conservation analysis led to the exclusion of 10 conserved residues from mutation considerations (Figure 2A). In the reaction between oli and GPP, the pyrophosphate moiety of GPP engages in hydrogen bonding with the side chains of residues K119, N173, Y175, Y216, R228, and K284, thereby securing its position (Figure 2C). The prenyl chain of GPP is embedded in a hydrophobic environment, surrounded by hydrophobic amino acid residues. Olivetol forms hydrogen bonds with Y288 and is stabilized by π-π stacking with the phenyl ring of F213 (Figure 2D). While residues critical for stabilizing GPP’s pyrophosphate group were excluded from the mutagenesis targets to avoid any destabilization of GPP, Y288, and F213, impacting both GPP and olivetol, are still considered viable targets for mutational analysis. To further narrow down the target sites, virtual saturation mutagenesis was performed on the residues within the substrate-binding pocket using Rosetta 2022.08. The change in binding energy for each mutant was used to assess their potential impact on enzyme activity (Figure 2B). A decrease in binding energy indicates a more stable substrate–enzyme complex, which is favorable for catalysis. Consequently, the sites with increased binding energy among all 20 residues were eliminated. Considering the aforementioned criteria, we identified V47, V49, F107, A108, S212, F213, S214, Q161, F176, S177, T269, V271, C286, Y288, and H290 for our mutational experiments.

We designed a dual-site combinatorial saturation screening for efficient screening of candidate residues. Considering that the selected mutation residues are either adjacent or very close in spatial position, this basis was used for the dual-site combinatorial saturation screening. We first used a pyrophosphate assay kit for high-throughput preliminary screenings, then cultured and purified clones showing elevated fluorescence values. We then used the purified enzymes to confirm their activity. The first round of screening identified the mutant WT-Q161R/S214H (named W1) as showing the most significant increase in CBG catalytic activity, boosting CBG production from 2.42 mg/L to 16.23 mg/L. Interestingly, the WT-S212A/F213S mutant exhibited good catalytic activity for CBGVA, possibly due to the smaller side chains of the mutated residues allowing the reaction chamber to accommodate larger substrates. Subsequent screenings continued based on W1 and WT-S212A/F213S, focusing on sites with beneficial mutations from the first round. Further mutations based on W1 led to strains W2 (W1-V49C/H290D) and W5 (W1-F176D/S177T/Y288V). Both strains achieved nearly identical CBG yields of approximately 18.62 mg/L. Further mutations based on the W2 strain led to the creation of strain W3 (W2-V47D/G286C), which marked a significant milestone in our research. This strain increased CBG yield to 25.56 mg/L (Figure 3A), a tenfold increase from the original wild-type strain, and exhibited a remarkable 10.3-fold increase in catalytic efficiency for CBGA synthesis (Figure 3B) and a 20.8-fold increase for CBGV production (Figure 3C). Additionally, the mutants W2 and W6 (W2-V47C/G286C) also showed significant improvements, with CBGV production increasing by 17.8 and 14.6 times, respectively. Furthermore, in the biosynthesis of CBGVA through the combination of oa3 and GPP, the W4 (WT-S212A/F213S-Q161R/Y288V) mutant has an 9.3-fold increase in enzyme activity compared to the wild-type (Figure 3D).

4.2. Evolution of NphB Enzyme Family

To identify homologous sequences, we used the aromatic prenyltransferase Orf2 from Streptomyces sp. strain CL190 as a query template in the National Center for Biotechnology Information (NCBI) Protein Data Bank, retrieving 100 sequences. Multiple sequence alignment was performed with ClustalW, followed by Gblock extraction of conserved regions (Figure S7). Phylogenetic analysis was then conducted in MEGA to construct a gene tree, enabling selection of 16 enzyme candidates for subsequent experimentation.

4.3. Cloning, Expression and Purification of Enzymes

The genes were synthesized by Wuhan Genecreate Bioengineering Co., Ltd. (Wuhan, China) and subcloned into the pET-28a(+) vector via one-step clone assembly. All plasmids were transformed into E. coli BL21(DE3), and a single colony was inoculated into 5 mL LB (with 100 μg/mL kanamycin) and grown overnight (220 rpm, 37 °C). This starter culture was diluted into 800 mL 2xYT (with 100 μg/mL kanamycin) and incubated (220 rpm, 37 °C) until an OD₆₀₀ 0.6–0.8 was attained. Protein expression was then induced by adding isopropyl β-D-1-thiogalactopyranoside (IPTG) to a final concentration of 0.5 mM, followed by incubation at 16 °C for 16–20 h. The induced culture was centrifuged (5500 rpm, 10 min, 4 °C), the supernatant discarded, and the cell pellet resuspended in lysis buffer (50 mM Tris, 150 mM NaCl, pH 8.0). The suspension underwent high-pressure homogenization (800–1000 bar, 4 °C, 3–5 min) and was clarified by centrifugation (8000 rpm, 1 h, 4 °C). The resulting supernatant was subjected to immobilized nickel affinity chromatography, washing with buffer (50 mM Tris pH 8.0, 150 mM NaCl, 20 mM imidazole) to remove impurities, and eluting the target protein with buffer (50 mM Tris pH 8.0, 150 mM NaCl, 200 mM imidazole). The eluate was concentrated and stored at −80 °C.

4.4. Catalytic Conditions for CBG Synthesis by NphB Homologous Sequences

To evaluate the catalytic proficiency of NphB homologs, reactions were initiated by combining enzymes and cofactors in 50 μL volumes, with final concentrations of 0.5 mM GPP, 1 mM olivetol, 5 mM MgCl₂, and 50 mM Tris (pH 8.0), and 1.0 mg/mL NphB. Following 8 h incubation at 30 °C, reactions were terminated via addition of 100 μL ethyl acetate, shaking (5–10 s), and centrifuged (12,000× g, 3 min) to recover the organic fraction in fresh tubes. The aqueous fraction underwent identical ethyl acetate extraction and pooling of the combined organic layers, drying under vacuum centrifugation, and reconstituting in 50 μL methanol for high performance liquid chromatography (HPLC) analysis. By employing the method of controlled variables, the optimal metal ions, optimal metal ion concentration, optimal temperature, and optimal pH for NphB were tested.

4.5. Modification of Prenyltransferase NphB

To probe olivetol binding and its impacts on NphB activity, in silico molecular docking studies were undertaken utilizing Rosetta Ligand Dock. The crystal structure of NphB complexed with substrate (PDB ID: 1ZB6) was obtained from the RCSB Protein Data Bank. The native ligands GSPP and 1,6-dihydroxynaphtalene were removed, leaving the position of the metal ions intact. Using Rosetta, GPP and olivetol were docked into the catalytic center. Rosetta relax was then used to generate 200 models and the lowest energy structure was selected for further docking to optimize the complex (Figure S8). The olivetol substrate structure was downloaded from Zinc and prepared via Schrödinger conformational searches, furnishing a sdf file. The Rosetta molfile_to_params script enabled sdf to params file format conversion alongside the generation of a multi-constant image file. Following ligand and receptor processing, docking was executed with the Rosetta (rosetta_scripts.mpi.linuxgccrelease) application utilizing StartFrom and Transform protocols and 10,000 models. Resulting complexes were ranked by total score and binding energy, with the top 10% visually inspected to eliminate unreasonable poses. The lowest energy binding mode was selected as the final docked complex. Subsequent point mutations employed Rosetta Cartesian_ddG calculations, initially performing two rounds of relax on the complex (200 models each) and utilizing the lowest energy structure for further refinement of all residues within 5 Å of the substrate. The refined complex then underwent saturation mutagenesis of these proximal residues to all other 19 amino acids, evaluating the change in binding energy. Finally, mutants exhibiting both favorable binding energy changes and large magnitude effects were experimentally validated.

4.6. Screening of Prenyltransferase Saturated Mutant Library

To screen the initial NphB mutant library, we designed double-site combinatorial saturation mutagenesis primers based on the selected mutation target sites, considering that the chosen mutated residues are adjacent or in close spatial proximity. Variants were constructed using these primers and then transformed into BL21(DE3). Single colonies were inoculated into 96-well plates containing 1 mL LB medium per well supplemented with 100 μg/mL kanamycin and grown overnight (37 °C, 750 rpm). The next day, cultures were diluted to 1% into fresh 96-well plates (1 mL 2YT medium plus 100 μg/mL kanamycin per well) and incubated (37 °C, 750 rpm) until an OD₆₀₀ 0.4–0.6. Protein expression was induced by adding 0.5 mM IPTG followed by 14–16 h at 16 °C. After centrifugation (3500 rpm, 5 min) to harvest cells, pellets were washed in buffer (50 mM Tris pH 8.0, 150 mM NaCl), lysed, and 30 μL magnetic beads were added per well to immobilize proteins. Beads were washed in buffer before adding the reaction components (0.5 mM olivetol, 0.25 mM geranyl pyrophosphate (GPP), 5 mM MgCl₂, 50 mM Tris pH 8.0, 40 μL final volume) and incubating at 30 °C for 2 h. Preliminary screening relied on a pyrophosphate fluorescence kit, measuring the catalytic activity of the mutants in the initial screening by detecting the fluorescence value of the residual pyrophosphate groups after the reaction. If no mutants with fluorescence values higher than the wild-type were detected in the initial screening, it was considered that no positive mutations occurred at that site. Variants with increased fluorescence were selected as the focus mutation library for scale-up cultivation. The 800 mL expression and purification of NphB variants were performed as previously described. The enzymatic assay consisted of 0.25 mM GPP, 0.5 mM olivetol, 5 mM MgCl₂, 50 mM Tris pH 8.0, and 3 mg/mL NphB in a total volume of 40 μL, incubated at 30 °C for 2 h. The reaction was quenched with ethyl acetate, reconstituted in 50 μL methanol, and analyzed by HPLC as previously detailed.

4.7. In Vitro Synthesis of CBG, CBGA, CBGV and CBGVA

To evaluate the catalytic proficiency of NphB mutants, reactions were initiated by combining enzymes and cofactors in 50 μL volumes, with final concentrations of 0.25 mM GPP, 0.5 mM olivetol (oa/ol3/oa3), 5 mM MgCl₂, and 50 mM Tris (pH 8.0), and 1.0 mg/mL NphB. Following 6 h incubation at 30 °C, reactions were terminated via addition of 100 μL ethyl acetate, shaking (5–10 s), and centrifuged (12,000× g, 3 min) to recover the organic fraction in fresh tubes. The aqueous fraction underwent identical ethyl acetate extraction and pooling of the combined organic layers, drying under vacuum centrifugation, and reconstituting in 50 μL methanol for high performance liquid chromatography (HPLC) analysis (Figure S9). CBG quantitation relied on reverse phase fractionation over a C18 column (250 mm × 4.6 mm × 5 μm) using an Alliance Waters e2695 HPLC system with 1 mL/min flow rate. Gradient elution employed water + 0.1% formic acid (solvent A) and acetonitrile + 0.1% formic acid (solvent B) mobile phases, holding B at 70% for 1 min followed by a 6 min increase to 77%, holding for 10 min, and re-equilibration to 70% B, totaling 26 min.

This work has received support from Jiaxing Synbiolab Technology Co., Ltd. We would like to thank all the staff in Jiaxing Synbiolab Technology Co., Ltd. for their support and help in this project.