English

• Camptothecin, a special pentacyclic monoterpene indole alkaloid (MIA) with pyrroloquinoline scaffold, is one of the most important plant-derived anticancer agents from the stems and woods of Camptotheca acuminata [1]. As a DNA topoisomerase I inhibitor, it can prevent the religation of DNA strands and inhibit the growth and reproduction of tumor cells [2–5]. Its well-known derivatives topotecan and irinotecan have been approved for clinical use against a variety of malignant tumors like lung, colon and rectum cancers [6]. With the increasing demand in clinic, many chemical synthesis methods of camptothecin and its derivatives were reported [7,8], while the long synthetic routes and high cost hamper their practical application. Direct extraction and isolation from plants is still the major source of camptothecin [9], however, the low content in plants and complex isolation methods make it difficult to meet the clinical demand. Synthetic biology, based on the complete elucidation of biosynthetic pathway, is considered as an efficient, environmentally friendly and promising approach, and has been applied to construct platforms producing many natural medicines in microbial factories [10–12]. However, elucidation of camptothecin biosynthesis is still challenging and difficult due to its enigmatic formation processes and the corresponding enzymes in the downstream biosynthetic pathway.

  The upstream biosynthetic pathway of camptothecin has been well elucidated. Generally, tryptophan generated from shikimate pathway is converted to tryptamine by tryptophan decarboxylase (TDC) (Fig. 1A-a) [13]. Geranyl diphosphate (GPP) is originated from mevalonic acid (MVA) or methylerythritol 4-phosphate (MEP) pathway, and produces secologanin or secologanic acid by a series of enzymes (Fig. 1A-b) [14–24]. Afterwards, tryptamine and secologanin (or secologanic acid) are condensed to strictosidine (or strictosidinic acid) by strictosidine synthase (STR) (or strictosidinic acid synthase (STRAS)) (Fig. 1A-c) [23,25]. While the downstream biosynthetic pathway of camptothecin includes the complex processes: (a) Condensation to form D ring from strictosidine or strictosidinic acid; (b) oxidative rearrangement of B and C rings from indole to quinoline moiety; (c) reductions of C17, C18 and C19; (d) aromatization of D ring; (e) hydrolysis of glycosides; (f) oxidation of C21-OH and hydroxylation of C20 (Fig. 1B). Whereas most of the formation processes in the downstream biosynthetic pathway remain ambiguous and are speculated just based on LC-MS/MS analyses [26] and the few isolated intermediates such as pumiloside (1) and (3S)-deoxypumiloside (2) (Fig. 1B) [27,28]. Moreover, the genes and their encoding enzymes responsible for these downstream reactions are rarely reported. Until recently, only two enzymes converting strictosamide (3) to strictosamide epoxide in the downstream biosynthetic pathway were reported [29,30], and other studies mostly focused on the post-modification of camptothecin [31–33]. Therefore, identification of their biosynthetic genes/enzymes is of high importance for deciphering and further constructing the biosynthetic pathway of camptothecin.

  Figure 1

  

  Figure 1.  The proposed biosynthetic pathway of camptothecin. (A) The upstream biosynthetic pathway of camptothecin: (a) The formation of tryptamine, (b) the formation of secologanin or secologanic acid, (c) the formation of strictosidine or strictosidinic acid. (B) The proposed downstream biosynthetic pathway of camptothecin. The functionally characterized enzymes are presented on solid arrows. TDC: tryptophan decarboxylase; GPP: geranyl diphosphate; GES: geraniol synthase; G10H: geraniol 10-hydroxylase; 10HGO: 10-hydroxygeraniol oxidase; IS: iridodial synthase; 7DLS: 7-deoxyloganetic acid synthase; 7DLGT: 7-deoxyloganetic acid glucosyltransferase; 7DLH: 7-deoxyloganic acid-7-hydroxylase; LAMT: loganic acid O-methyltransferase; SLS: secologanin synthase; SLAS: secologanic acid synthase; STR: strictosidine synthase; STRAS: strictosidinic acid synthase; CYP450: including CYP71BE206 from C. acuminata and OpCYP716E111 from Ophiorrhiza pumila.

  DownLoad: Full-Size Img PowerPoint

  Due to the absence of glucosyl group in camptothecin, the hydrolysis of glycoside intermediates is necessary in the biosynthesis of camptothecin (Fig. 1B), however, the responsible enzyme and its substrate(s) remain unknown to date. Herein, we functionally identify a key glycoside hydrolase (CaGH1) from C. acuminata that catalyzes pumiloside (1), (3S)-deoxypumilosde (2) and strictosamide (3) to produce the corresponding aglycones, respectively. The unusual stereoselective catalytic activity of CaGH1 towards (3S)-biosynthetic intermediates is also explored. These findings provide new insights into the biosynthesis of camptothecin, especially the deglycosylation step.

  Using crude protein extracts from plants to perform biochemical assays in vitro is helpful for mining enzymes and the corresponding substrates, especially for those catalyzing unknown reaction types in the biosynthesis of natural products with complex structures [34–36]. Thus, to search for potential enzymes in the downstream biosynthetic pathway of camptothecin, we first prepared the crude protein extracts from young leaves of C. acuminata for in vitro assays. When the crude protein lysates were incubated with pumiloside (1), the putative key intermediate in camptothecin biosynthesis [27,28], to our delight, 1 was completely consumed and five new product peaks (i–v) were observed in high-performance liquid chromatography (HPLC) chromatograms (Fig. 2A). The MS spectra revealed the molecular weights of these products were all 162 Da less than that of 1 (Fig. 2B) with the similar ultraviolet (UV) spectra to that of 1 (Fig. S1 in Supporting information), suggesting that all of these products were the deglycosylated derivatives of 1. And, pumiloside aglycone (1a) was speculated to be the direct product of 1 (Fig. 2C). Considering the unstability of newly formed hemiacetal unit in E ring of 1a, these five product forms were possibly generated by spontaneous opening of E ring and the subsequent tautomerization of enols and aldehydes (Fig. 2D), though the specific chemical structures of i–v were still uncharacterized. These results indicated the presence of enzymes with glycoside hydrolysis function towards 1 in the crude protein extracts, and that 1 might be an intermediate substrate for the glycoside hydrolysis in the biosynthesis of camptothecin.

  Figure 2

  

  Figure 2.  Activity analysis of crude protein extracts from C. acuminata leaves. (A) HPLC chromatograms (240 nm) of C. acuminata crude protein extracts with 1 as a substrate. (a) Crude protein extracts + 1 incubated in sodium phosphate buffer (PB, pH 7.5); (b) boiled crude protein extracts + 1 incubated in PB (pH 7.5). (B) Typical positive MS spectra of products i–v and substrate 1. (C) Conversion of 1 to 1a by crude protein extracts. (D) Possible chemical structures and the interconversion of five products of crude protein extracts with 1 as a substrate reacting in PB. UV spectra of the products i–v and substrate 1 are shown in Fig. S1.

  DownLoad: Full-Size Img PowerPoint

  Given that the deglycosylation was generally catalyzed by glycoside hydrolases (glycosidases) in the biosynthesis of other MIAs [37–40], the enzymes in this family were firstly considered as candidates for further analysis. Thus, a transcriptome database of C. acuminata was established using the same plant materials for crude protein extraction to mine glycoside hydrolase genes (CaGHs). Since the crude protein lysates showed obvious deglycosylating activity, the genes coding CaGHs might exhibit high expression levels in this transcriptome. Accordingly, top 11 full-length CaGH transcripts (CaGH1–CaGH11) with fragments per kilobase of transcript per million mapped reads (FPKM) greater than 10 as candidate genes were obtained from the transcriptomic dataset (Table S2 in Supporting information). To identify the enzymes responsible for deglycosylation of pumiloside (1), the above 11 genes were individually expressed in Escherichia coli. After purification of the recombinant protein by affinity chromatography, 1 was incubated with the enzymes. The biochemical assays revealed that four enzymes (CaGH1–CaGH4) could convert 1 to the five products (i–v), which were consistent with the result of C. acuminata leaf crude protein reaction (Figs. 2A and 3A, Fig. S2 in Supporting information). Among them, CaGH1 exhibited the highest activity, and was thus selected for further investigations.

  Figure 3

  

  Figure 3.  Functional characterization of CaGH1. (A) HPLC chromatograms (240 nm) of purified recombinant CaGH1 with 1 as a substrate. (a) CaGH1 + 1 incubated in sodium phosphate buffer (pH 7.5, PB); (b) boiled CaGH1 + 1 incubated in PB (pH 7.5); (c) CaGH1 + 1 incubated in Tris–HCl buffer (pH 7.5); (d) boiled CaGH1 + 1 incubated in Tris–HCl buffer (pH 7.5). (B) (+)-High-resolution electrospray ionization mass spectrometry ((+)-HR-ESI-MS) spectrum of product 1b. (C) Enzymatic conversion from 1 to 1a by CaGH1 and spontaneous conversion from 1a to 1b in Tris–HCl buffer. (D) Proposed formation mechanism from pumiloside aglycone (1a) to 1b in Tris–HCl buffer.

  DownLoad: Full-Size Img PowerPoint

  To further confirm the function of CaGH1, the incubation of 1 with this enzyme was scaled up for the preparation and structural characterization of the products i–v. However, we failed to purify any single product for nuclear magnetic resonance (NMR) analysis after numerous efforts, possibly because of their unstability and interconversion as described above (Fig. 2D). Very fortunately, when we tried to investigate the influence on the enzyme activity with various buffers, only one stable product (1b, Fig. 3A) was observed in Tris–HCl buffer. According to its high-resolution mass spectrometry (HRMS) data ([M + H]⁺ calcd. for [C₂₄H₂₆N₃O₅]⁺, 436.1867; found, 436.1876), this product was 85 Da larger than those of products i–v (Figs. 2B and 3B). In detail, compared to the unstable deglycosylated derivatives i–v (C₂₀H₁₈N₂O₄), it (C₂₄H₂₅N₃O₅) had four more carbon atoms and one more nitrogen atom. It was supposed that the glucosyl group of 1 was firstly hydrolyzed by CaGH1 to form i–v, and a Tris (C₄H₁₁NO₃) unit was introduced into i–v to generate the compound 1b. To confirm this deduction, a scale-up reaction was conducted in Tris–HCl buffer and 0.3 mg of the product 1b was obtained. The structure of 1b was characterized as a pumiloside aglycone Tris-derivative (1b, Figs. 3B and C, Figs. S9–S16 and Table S4 in Supporting information) as supposed on the basis of HRMS and 1D/2D NMR analyses. Accordingly, we proposed a possible formation mechanism from pumiloside aglycone (1a) to 1b (Fig. 3D). 1a is unstable in Tris–HCl buffer and its dialdehyde form (1a-1) easily generates. The reaction of the primary amine in Tris with aldehyde leads to the formation of a Schiff base (1a-2), and the same reaction occurs from 1a-3 to 1a-4. Afterwards, the stable product 1b is yielded via a Mannich reaction. Thus, the structure of unstable product was indirectly but unambiguously determined through its stable derivative, and CaGH1 has now been functionally identified as a glycoside hydrolase that can catalyze pumiloside (1), the key biosynthetic intermediate of camptothecin, to hydrolyze the glucosyl group.

  Subsequently, the biochemical properties of CaGH1 were explored using 1 as a substrate (Figs. S3 and S4 in Supporting information). The optimal temperature and pH value of CaGH1 were determined to be 45 ℃ and 4.5, respectively. And CaGH1 is divalent metal cation-independent; CaGH1 has K_m = 269.0 µmol/L and k_cat/K_m = 15.6 L mmol⁻¹ min⁻¹ towards 1. Phylogenetic analysis of CaGH1 with other characterized plant glycoside hydrolases showed that CaGH1 is grouped in the same cluster with the reported alkaloid glucosidases which are involved in the biosynthesis of other MIAs (Fig. S5 in Supporting information) [37–40].

  It has been assumed that the hydrolysis of glycoside occurs after the aromatization of D ring of (3S)-deoxypumiloside (2) [26]. However, no glycoside hydrolases together with their native substrates in camptothecin biosynthesis have been experimentally identified to date. The above results revealed that CaGH1 can use pumiloside (1) as a substrate, suggesting that the deglycosylation could occur before the aromatization of D ring, which is different from the previous hypothesis. To comprehensively uncover the deglycosylation process in the biosynthetic pathway of camptothecin, we further explored the substrate scope of CaGH1. To obtain more potential downstream glycoside intermediates in camptothecin biosynthesis, the methanolic crude extracts from the one-year-old C. acuminata seedlings were analyzed by HPLC-HRMS. The possible downstream glycoside substrates of CaGH1 include two chemical structure types: indole alkaloids with 6/5/6/6/6 ring system and quinoline alkaloids with 6/6/5/6/6 ring system (Figs. 1D and 4A). After careful analysis for the crude extracts, the candidate glycoside substrates (3S)-deoxypumiloside (2) and (3R)-deoxypumiloside (2′ , the diastereoisomer of 2), together with pumiloside (1), strictosamide (3) and vincosamide (3′ , the diastereoisomer of 3) were detected in the crude extracts (Fig. 4A and Fig. S6 in Supporting information). Therefore, 2 and 2′ were isolated from 1.34 kg one-year-old seedlings of C. acuminata, and structurally characterized by HRMS and NMR (Fig. 4A; Figs. S6, S17–S22 and Table S5 in Supporting information) [41].

  Figure 4

  

  Figure 4.  Exploring the substrate scope and stereospecificity of CaGH1. (A) The chemical structures of substrates (2, 2′, 3 and 3′) used in this experiment. (B) HPLC chromatograms (240 nm) of purified recombinant CaGH1 with 2, 2′, 3 and 3′ incubated in Tris–HCl buffer (pH 7.5), respectively. (a) CaGH1 + 2; (b) boiled CaGH1 + 2; (c) CaGH1 + 2′; (d) boiled CaGH1 + 2′; (e) CaGH1 + 3; (f) boiled CaGH1 + 3; (g) CaGH1 + 3′; (h) boiled CaGH1 + 3′. (C) (+)-HR-ESI-MS spectra of products 2b and 3b. (D) Enzymatic conversion from 2 or 3 to 2a or 3a by CaGH1 and spontaneous conversion from 2a or 3a to 2b or 3b in Tris–HCl buffer, respectively.

  DownLoad: Full-Size Img PowerPoint

  When the two pairs of diastereoisomers 2 and 2′, 3 and 3′ were in hand, the substrate scope and stereospecificity of CaGH1 were further investigated. The incubations of the two pairs of substrates with CaGH1 in Tris–HCl buffer individually were performed and analyzed by HPLC-HRMS. Interestingly, only 2 and 3 with 3S configuration could be converted to the corresponding products, while no product was observed using 2′ and 3′ with 3R configuration as substrates (Fig. 4B). The HRMS spectra showed that the molecular weights of the enzymatic products of 2 and 3 were consistent with (3S)-deoxypumiloside aglycone and strictosamide aglycone Tris-derivatives (2b and 3b), respectively (Fig. 4C). We proposed that 2b and 3b might generate through the similar speculated reaction mechanism as that of 1a to 1b (Figs. 3D and 4D, Fig. S7 in Supporting information). Collectively, the above results showed that the deglycosylation could occur in both quinoline-type and indole-type glycosides with 3S configuration in the biosynthesis of camptothecin, further supporting that quinoline alkaloids (1 and 2) and indole alkaloid (3) are the putative biosynthetic intermediates of camptothecin [18,23,42]. Additionally, the deglycosylating activities of CaGH2–CaGH11 were also tested with 2, 2′, 3, and 3′ as substrates in Tris–HCl buffer individually, and the biochemical assays showed that none of the enzymes could hydrolyze the glucosyl groups of these compounds.

  Although aglycones 1a, 2a and 3a were not isolated from plants before, probably due to their unstability, the three aglycones might also be the possible intermediates in planta in the biosynthesis of camptothecin. Furthermore, the biosynthetic pathways of natural products like huperzine A [43] and paclitaxel [44,45] often exhibit network ones rather than single linear ones. Possibly, one enzyme with substrate promiscuity can catalyze the same reaction towards diverse substrates in such complex pathways. Thus, CaGH1 may function at different stages in camptothecin biosynthesis, and a possible nonlinear downstream biosynthetic pathway of camptothecin is proposed, including the conversion from both quinoline and indole intermediates to camptothecin (Fig. S8 in Supporting information).

  In this work, based on plant crude protein activity-guided enzyme verification, transcriptome-based gene mining, biochemical assay in vitro and rational derivatization of unstable enzymatic products, a novel glycoside hydrolase (CaGH1) involved in the biosynthesis of camptothecin was discovered and functionally characterized from C. acuminata. The investigation of substrate scope revealed that the deglycosylation could occur in both quinoline-type and indole-type glycoside intermediates in the biosynthesis of camptothecin. Moreover, this work also demonstrated that CaGH1 could stereo-specifically catalyze the deglycosylation of the biosynthetic intermediates with 3S configuration. Therefore, these findings not only provide new insights into the role of glycoside hydrolase in camptothecin biosynthesis, also lays the foundation for further elucidating the complete biosynthetic pathway of camptothecin.

  Declaration of competing interest

  The authors declare that they have no competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Xincheng Sun: Writing – original draft, Investigation, Formal analysis, Data curation. Yuxin Wang: Writing – original draft, Investigation, Formal analysis, Data curation. Changkang Li: Writing – review & editing, Investigation, Formal analysis. Ridao Chen: Writing – review & editing, Formal analysis. Kebo Xie: Writing – review & editing, Formal analysis. Jimei Liu: Writing – review & editing, Formal analysis. Songyang Sui: Writing – review & editing, Formal analysis. Yaotian Han: Writing – review & editing, Formal analysis. Dawei Chen: Writing – review & editing, Validation, Formal analysis, Conceptualization. Jungui Dai: Writing – review & editing, Validation, Supervision, Project administration, Funding acquisition, Formal analysis, Data curation, Conceptualization.

  Acknowledgments

  This work was financially supported by the National Key Research and Development Program of China (No. 2020YFA0908000), and CAMS Innovation Fund for Medical Sciences (No. CIFMS-2023-I2M-2–006).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.110895.

  
  
• 1. [1]

     M.E. Wall, M.C. Wani, C.E. Cook, et al., J. Am. Chem. Soc. 88 (1966) 3888–3890. doi: 10.1021/ja00968a057

  2. [2]

     Y.H. Hsiang, R. Hertzberg, S. Hecht, L.F. Liu, J. Biol. Chem. 260 (1985) 14873–14878.

  3. [3]

     E. Kjeldsen, J.Q. Svejstrup, I.I. Gromova, J. Alsner, O. Westergaard, J. Mol. Biol. 228 (1992) 1025–1030.

  4. [4]

     X. Wu, M. Liu, C. Zheng, et al., Chin. Chem. Lett. 34 (2023) 107590.

  5. [5]

     H. Li, M. Zhang, J. He, et al., Chin. Chem. Lett. 36 (2025) 110615. doi: 10.1016/j.cclet.2024.110615

  6. [6]

     A.L. Demain, P. Vaishnav, Microb. Biotechnol. 4 (2011) 687–699. doi: 10.1111/j.1751-7915.2010.00221.x

  7. [7]

     W. Du, Tetrahedron 59 (2003) 8649–8687.

  8. [8]

     L. Song, Z. Lv, K. Zhang, et al., Asian J. Org. Chem. 11 (2022) e202200515.

  9. [9]

     A. Lorence, C.L. Nessler, Phytochemistry 65 (2004) 2735–2749.

  10. [10]

      J. Zhang, L.G. Hansen, O. Gudich, et al., Nature 609 (2022) 341–347. doi: 10.1038/s41586-022-05157-3

  11. [11]

      Y. Liu, X. Li, S. Sui, et al., Acta Pharm. Sin. B 13 (2023) 1771–1785.

  12. [12]

      R. Yan, B. Xie, K. Xie, et al., Nat. Commun. 15 (2024) 3539.

  13. [13]

      M. López-Meyer, C.L. Nessler, Plant J. 11 (1997) 1167–1175. doi: 10.1046/j.1365-313x.1997.11061167.x

  14. [14]

      F. Chen, W. Li, L. Jiang, et al., J. Ind. Microbiol. Biotechnol. 43 (2016) 1281–1292. doi: 10.1007/s10295-016-1802-2

  15. [15]

      G. Collu, N. Unver, A.M.G. Peltenburg-Looman, et al., FEBS Lett. 508 (2001) 215–220.

  16. [16]

      Y. Sun, H. Luo, Y. Li, et al., BMC Genom. 12 (2011) 533.

  17. [17]

      A. Awadasseid, W. Li, Z. Liu, et al., Int. J. Biol. Macromol. 162 (2020) 1076–1085.

  18. [18]

      R. Sadre, M. Magallanes-Lundback, S. Pradhan, et al., Plant Cell 28 (2016) 1926–1944. doi: 10.1105/tpc.16.00193

  19. [19]

      V. Salim, B. Wiens, S. Masada-Atsumi, F. Yu, V. De Luca. Phytochemistry 101 (2014) 23–31.

  20. [20]

      K. Asada, V. Salim, S. Masada-Atsumi, et al., Plant Cell 25 (2013) 4123–4134. doi: 10.1105/tpc.113.115154

  21. [21]

      Y. Yang, W. Li, J. Pang, et al., ACS Chem. Biol. 14 (2019) 1091–1096. doi: 10.1021/acschembio.8b01124

  22. [22]

      M. Kang, R. Fu, P. Zhang, et al., Nat. Commun. 12 (2021) 3531.

  23. [23]

      M. Yang, Q. Wang, Y. Liu, et al., BMC Biol. 19 (2021) 122.

  24. [24]

      J.C. Miller, A.J. Hollatz, M.A. Schuler, Phytochemistry 183 (2021) 112626.

  25. [25]

      Y. Yamazaki, H. Sudo, M. Yamazaki, N. Aimi, K. Saito, Plant Cell Physiol. 44 (2003) 395–403.

  26. [26]

      X. Pu, C. Zhang, H. Gao, et al., J. Chromatogr. A 11 (2020) 41.

  27. [27]

      N. Aimi, M. Nishimura, A. Miwa, et al., Tetrahedron Lett. 30 (1989) 4991–4994.

  28. [28]

      B.K. Carte, C. DeBrosse, D. Eggleston, et al., Tetrahedron 46 (1990) 2747–2760.

  29. [29]

      X. Pu, M. Wang, M. Chen, et al., ACS Chem. Biol. 18 (2023) 1772–1785. doi: 10.1021/acschembio.3c00222

  30. [30]

      T. Zhang, Y. Wang, S. Wu, et al., J. Integr. Plant Biol. 66 (2024) 1044–1047. doi: 10.1111/jipb.13594

  31. [31]

      V. Salim, A.D. Jones, D. DellaPenna, Plant J. 95 (2018) 112–125. doi: 10.1111/tpj.13936

  32. [32]

      T.M. Nguyen, T. Nguyen, Y.Y. Leung, et al., Commun. Chem. 4 (2021) 177.

  33. [33]

      Y. Chen, J.P. Huang, Y.J. Wang, et al., J. Integr. Plant Biol. 66 (2024) 1158–1169. doi: 10.1111/jipb.13649

  34. [34]

      L. Gao, C. Su, X. Du, et al., Nat. Chem. 12 (2020) 620–628. doi: 10.1038/s41557-020-0467-7

  35. [35]

      Z. Fan, A. Jaisi, Y. Chen, et al., ACS Catal. 11 (2021) 8818–8828. doi: 10.1021/acscatal.1c01514

  36. [36]

      R.S. Nett, E.S. Sattely, J. Am. Chem. Soc. 143 (2021) 19454–19465. doi: 10.1021/jacs.1c08659

  37. [37]

      A. Geerlings, M.M. Ibañez, J. Memelink, R. van Der Heijden, R. Verpoorte, J. Biol. Chem. 275 (2000) 3051–3056.

  38. [38]

      H. Warzecha, I. Gerasimenko, T.M. Kutchan, J. Stöckigt, Phytochemistry 54 (2000) 657–666.

  39. [39]

      I. Gerasimenko, Y. Sheludko, X. Ma, J. Stöckigt, Eur J Biochem 269 (2002) 2204–2213.

  40. [40]

      Y. Wu, C. Liu, A. Koganitsky, F.L. Gong, S. Li, Angew. Chem. Int. Ed. 62 (2023) e202307995.

  41. [41]

      M. Kitajima, S. Yoshida, K. Yamagata, et al., Tetrahedron 58 (2002) 9169–9178.

  42. [42]

      C.R. Hutchinson, A.H. Heckendorf, J.L. Straughn, P.E. Daddona, D.E. Cane, J. Am. Chem. Soc. 101 (1979) 3358–3369. doi: 10.1021/ja00506a037

  43. [43]

      R.S. Nett, Y. Dho, C. Tsai, et al., Nature 624 (2023) 182–191. doi: 10.1038/s41586-023-06716-y

  44. [44]

      C. Li, X. Yin, S. Wang, et al., Angew. Chem. Int. Ed. 63 (2024) e202407070.

  45. [45]

      J.C. Liu, R.D.L. Peña, C. Tocol, E.S. Sattely, Nat. Commun. 15 (2024) 1419.

• 

  Figure 1  The proposed biosynthetic pathway of camptothecin. (A) The upstream biosynthetic pathway of camptothecin: (a) The formation of tryptamine, (b) the formation of secologanin or secologanic acid, (c) the formation of strictosidine or strictosidinic acid. (B) The proposed downstream biosynthetic pathway of camptothecin. The functionally characterized enzymes are presented on solid arrows. TDC: tryptophan decarboxylase; GPP: geranyl diphosphate; GES: geraniol synthase; G10H: geraniol 10-hydroxylase; 10HGO: 10-hydroxygeraniol oxidase; IS: iridodial synthase; 7DLS: 7-deoxyloganetic acid synthase; 7DLGT: 7-deoxyloganetic acid glucosyltransferase; 7DLH: 7-deoxyloganic acid-7-hydroxylase; LAMT: loganic acid O-methyltransferase; SLS: secologanin synthase; SLAS: secologanic acid synthase; STR: strictosidine synthase; STRAS: strictosidinic acid synthase; CYP450: including CYP71BE206 from C. acuminata and OpCYP716E111 from Ophiorrhiza pumila.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Activity analysis of crude protein extracts from C. acuminata leaves. (A) HPLC chromatograms (240 nm) of C. acuminata crude protein extracts with 1 as a substrate. (a) Crude protein extracts + 1 incubated in sodium phosphate buffer (PB, pH 7.5); (b) boiled crude protein extracts + 1 incubated in PB (pH 7.5). (B) Typical positive MS spectra of products i–v and substrate 1. (C) Conversion of 1 to 1a by crude protein extracts. (D) Possible chemical structures and the interconversion of five products of crude protein extracts with 1 as a substrate reacting in PB. UV spectra of the products i–v and substrate 1 are shown in Fig. S1.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Functional characterization of CaGH1. (A) HPLC chromatograms (240 nm) of purified recombinant CaGH1 with 1 as a substrate. (a) CaGH1 + 1 incubated in sodium phosphate buffer (pH 7.5, PB); (b) boiled CaGH1 + 1 incubated in PB (pH 7.5); (c) CaGH1 + 1 incubated in Tris–HCl buffer (pH 7.5); (d) boiled CaGH1 + 1 incubated in Tris–HCl buffer (pH 7.5). (B) (+)-High-resolution electrospray ionization mass spectrometry ((+)-HR-ESI-MS) spectrum of product 1b. (C) Enzymatic conversion from 1 to 1a by CaGH1 and spontaneous conversion from 1a to 1b in Tris–HCl buffer. (D) Proposed formation mechanism from pumiloside aglycone (1a) to 1b in Tris–HCl buffer.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Exploring the substrate scope and stereospecificity of CaGH1. (A) The chemical structures of substrates (2, 2′, 3 and 3′) used in this experiment. (B) HPLC chromatograms (240 nm) of purified recombinant CaGH1 with 2, 2′, 3 and 3′ incubated in Tris–HCl buffer (pH 7.5), respectively. (a) CaGH1 + 2; (b) boiled CaGH1 + 2; (c) CaGH1 + 2′; (d) boiled CaGH1 + 2′; (e) CaGH1 + 3; (f) boiled CaGH1 + 3; (g) CaGH1 + 3′; (h) boiled CaGH1 + 3′. (C) (+)-HR-ESI-MS spectra of products 2b and 3b. (D) Enzymatic conversion from 2 or 3 to 2a or 3a by CaGH1 and spontaneous conversion from 2a or 3a to 2b or 3b in Tris–HCl buffer, respectively.

  下载: 全尺寸图片 幻灯片
•