Enhancing Metabolic Engineering in Medicinal Plants Through Prime Editing

Tanshinone and phenolic acid are key therapeutic compounds in the medicinal plant Salvia miltiorrhiza, while rutin is the major bioactive metabolite in the medicinal plant Fagopyrum dibotrys. However, their natural levels in cultivated varieties remain low, limiting their pharmacological potential. Enhancing metabolite accumulation through the modification of upstream transcription factors and key biosynthetic enzymes has shown promise (Deng et al. 2020). Compounding the challenge, medicinal plants like S. miltiorrhiza are perennials with long life cycles, making conventional breeding inefficient. While CRISPR-Cas9 has facilitated gene knockout strategies in medicinal plants, precise base editing technologies remain underutilised (Das et al. 2024). Prime editing (PE) has emerged as a powerful tool for introducing targeted nucleotide changes; thus, offering a promising route for molecular breeding (Anzalone et al. 2019). While their application in medicinal plants remains largely unexplored. To address these limitations in genome engineering of medicinal plants, we develop the prime editor in medicinal plants. Our previous work demonstrated that N-terminus M-MLV RT fused to Cas9 nickase plus epegRNA prime editor (NEPE) is highly efficient in rice (Zhong et al. 2024), indicating that the N-terminal fusion of M-MLV reverse transcriptase is particularly well-suited for plant applications. Building on this, and with optimization of the expression system, we further adapted the prime editor for medicinal plants, termed MediPlant-NEPE (Figure 1a; Figure S1). We first targeted SmMYB36 and SmTCP15 in S. miltiorrhiza to test the efficiency of MediPlant-NEPE and investigate their roles in flower pigmentation. Targeted C>T conversion at Q22 in SmMYB36 and a G>T conversion at E24 in SmTCP15 induced premature stop codons. Following Agrobacterium-mediated transformation, positive transgenic lines were obtained with mutation frequencies of 21.21% for SmMYB36 and 21.21% for SmTCP15 (biallelic mutation rates of 15.15% and 12.12%, respectively; Figure 1b,d; Table S1). SmMYB36 biallelic mutants exhibited lighter flowers, while SmTCP15 biallelic mutants showed deeper coloration (Figure 1c,e). HPLC analysis revealed significant changes in anthocyanin content: SmMYB36 mutants had a drastic decrease in delphinidin and a 30% reduction in cyanidin and pelargonidin, while SmTCP15 mutants showed a 1.66-fold increase in delphinidin and a 1.43-fold increase in cyanidin (Figure 1f; Figure S2a,b). Unlike previous reports (Hsu et al. 2024), SmMYB36 mutants did not exhibit albino flowers, likely due to reduced SmANS expression (Figure S2c). In contrast, SmTCP15 mutants had a 2.3- to 2.6-fold increase in SmANS expression (Figure S2d). This study provides new insights into the role of SmTCP15 as a negative regulator of anthocyanin biosynthesis and demonstrates the potential of prime editing for fine-tuning phenotypic traits. We next applied MediPlant-NEPE to enhance secondary metabolite production in S. miltiorrhiza by targeting SmbZIP1, a key regulator of tanshinone and phenolic acid biosynthesis. A GTG>TAA mutation was introduced at residue V43, generating a stop codon, resulting in a mutation frequency of 26.47% and a biallelic mutation rate of 17.65% (Figure 1g). Homozygous mutants displayed deeper root colour (Figure 1h; Figure S3a) and increased tanshinone content (Figure 1i; Figure S3b). Specifically, cryptotanshinone and tanshinone IIA levels increased by 1.86-fold and 1.40-fold, respectively, while tanshinone I showed no significant change. The phenolic acid content was decreased in SmbZIP1 mutants (Figure S3c,d), consistent with previous reports (Deng et al. 2020). We also targeted SmRAS in hairy roots to enhance rosmarinic acid and salvianolic acid B production (Figure S4a). The G49A mutation, generated by a G>C conversion, achieved an editing efficiency of 39.47% and a biallelic mutation rate of 15.79% (Figure 1j). Two heterozygous lines (#HM04-02, #HM04-03) and one homozygous line (#HM04-06) were selected for further analysis (Figure 1k). HPLC confirmed that the homozygous mutant #HM04-06 exhibited a 1.98-fold increase in rosmarinic acid and a 1.80-fold increase in salvianolic acid B (Figure 1l). All homozygous mutants showed a 1.93-fold increase in rosmarinic acid and a 2.07-fold increase in salvianolic acid B (Figure S4b–e). Molecular docking revealed that the G49A mutation minimises the enzyme pocket size and creates new interactions with the substrate and product (Figure S4f). These results demonstrate that SmRASG49A enhances both rosmarinic acid and salvianolic acid B content. We further applied the MediPlant-NEPE system to enhance rutin content in the wild species F. dibotrys, aiming to facilitate its rapid domestication as a medicinal plant. The UFGT3 gene, a key enzyme in rutin biosynthesis (He et al. 2022) (Figure S5a) Amino acid substitutions at positions 120 (Q120H) and 170 (D170G) of the FdUFGT3 protein were predicted to enhance enzymatic activity and increase rutin production (Figure S5b). Using MediPlant-NEPE, we introduced these mutations, achieving editing efficiencies of 12.00% for Q120H, 8.00% for D170G, and 12.00% for the double mutant with both Q120H and D170G (Figure 1m). Homozygous mutant lines were selected for further analysis (Figure 1n). HPLC analysis showed a decrease in quercetin levels across all three lines (Figure 1o; Figure S5c), with the double mutant exhibiting a 2.75-fold increase in rutin content, while single mutations showed limited improvement. These results demonstrate that MediPlant-NEPE can efficiently enhance secondary metabolite production through enzyme engineering. Molecular docking revealed that the mutations altered the FdUFGT3 enzyme pocket, with the single mutation causing quercetin and UDP-glucose to bind in separate pockets (Figure S5d). In contrast, the double mutation enlarged the enzyme pocket to accommodate both substrates and the product rutin (Figure S5e). These findings highlight the potential of MediPlant-NEPE for improving metabolite production in medicinal plants. Overall, MediPlant-NEPE provides a proof-of-concept for prime editing in the medicinal plants S. miltiorrhiza and F. dibotrys. Our results demonstrate its potential for broader application in other medicinal species and highlight its promise for molecular breeding, enhancement of secondary metabolites, and germplasm innovation, thereby accelerating the development of improved cultivars and shortening breeding cycles for economically and therapeutically valuable plants. Z. Zhong. and L.Z. designed the experiments. H.Y., X.F. and X.Z. generated all the constructs. X.W. and W.Z. generated stable transgenic lines and analysed the plants. Z. Zhang., Y.J. and R.Y. conducted the HPLC assays. Z. Zhong., L.Z. and H.Y. wrote the paper with input from other authors. All authors read and approved the final manuscript. This work was supported by the National Science Foundation of China (award no. 32301245), the China National Postdoctoral Program for Innovative Talents (award no. BX20240244), Sichuan Science and Technology Program (award no. 2024JDRC0010), and Sichuan Agriculture University startup funds to Z. Zhong. This research also received financial support from the National modern agricultural industry technology system Sichuan innovation team (SCCXTD-2024-19) and the Featured Medicinal Plants Sharing and Service Platform of Sichuan Province. This work was supported by the National Natural Science Foundation of China, 32301245. China National Postdoctoral Program for Innovative Talents, BX20240244. Sichuan Science and Technology Program, 2024JDRC0010. National modern agricultural industry technology system Sichuan innovation team, SCCXTD-2024-19. The authors declare no conflicts of interest. The data that supports the findings of this study are available in the Supporting Information of this article. Appendix S1: pbi70532-sup-0001-AppendixS1.xlsx.zip. Appendix S2: pbi70532-sup-0002-AppendixS2.docx. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.