The Nothapodytes nimmoniana genome provides insights into the independent evolution of camptothecin biosynthesis in asterids

The Nothapodytes nimmoniana genome provides insights into the independent evolution of camptothecin biosynthesis in asterids | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article The Nothapodytes nimmoniana genome provides insights into the independent evolution of camptothecin biosynthesis in asterids Xiaofeng Shen, Xiaofeng Shen, Zhijing Guan, Sijie Sun, Ke Du, and 23 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5971869/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Nothapodytes nimmoniana is known to produce the highest content of the anticancer compound camptothecin (CPT) in the plant kingdom. We present the chromosome-level allotetraploid genome of N. nimmoniana , marking the first genome sequence from the order Icacinales. This 5-Gb genome encodes 92,630 genes, with subgenome B exhibiting dominant gene expression. Through genome mining, we identified and characterized four key enzymes involved in CPT biosynthesis, revealing that N. nimmoniana shares a similar prestrictosidine pathway with most monoterpene indole alkaloid-producing plants. Notably, homoeologous pairs of all characterized enzymes maintained their functions across both subgenomes, suggesting that gene duplication from allotetraploidization likely enhances CPT production in this species. Phylogenetic and syntenic analyses revealed that strictosidine synthase and strictosamide epoxidase were independently recruited in N. nimmoniana , Camptotheca acuminata , and Ophiorrhiza pumila , supporting the hypothesis that CPT biosynthesis evolved independently at least three times within the asterid clade. Biological sciences/Molecular biology/Transcriptomics Biological sciences/Evolution/Molecular evolution Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Camptothecin (CPT) is a plant alkaloid with potent anticancer activity that is primarily achieved by impeding cancer cell replication through the inhibition of DNA topoisomerase I 1 . Currently, at least four CPT derivatives—hydroxycamptothecin, irinotecan, topotecan and belotecan—have been successfully used in clinical treatments. These derivatives are synthesized predominantly through the modification of CPT obtained from plant sources. Nothapodytes nimmoniana , a member of the Icacinaceae family, is recognized as the most potent CPT producer 2 . However, owing to the increasing global demand for CPT in the pharmaceutical industry, N. nimmoniana faces the risk of unsustainable harvesting practices. Therefore, elucidating the biosynthetic pathway of CPT is essential for addressing the growing demand through synthetic biology approaches and conserving this endangered species. CPT is synthesized via the monoterpene indole alkaloid (MIA) biosynthetic pathway. Strictosidine serves as the common intermediate for most MIAs and is formed by the condensation of tryptamine and secologanin. CPT is derived from strictosidine in Ophiorrhiza pumila 3 . In contrast, in C. acuminata , an alternative biosynthetic pathway has evolved due to the lack of loganic acid O-methyltransferase (LAMT), in which strictosidinic acid, rather than strictosidine, is used as the intermediate for CPT biosynthesis 4 (Supplementary Fig. 1). Six enzymes [geraniol synthase (GES) 5 , 10-hydroxygeraniol oxidoreductase (10HGO) 6 , iridodial synthase (IS) 7 , 7-deoxyloganic acid 7-hydroxylase (7-DLH) 8 , secologanin acid synthase (SLAS) 9 , strictosidine synthase (STR) 3 , and strictosamide epoxidase (SEO) (CYPC71BE206) 10 ] from C. acuminata and four enzymes [LAMT 3 , secologanin synthase (SLS) 3 , STR 11 , and SEO (CYP716E111) 12 ] from O. pumila have been identified as being involved in CPT biosynthesis. However, investigations into CPT biosynthesis in N. nimmoniana remain limited, with only Nn SLS (CYP72A1) characterized in this species 13 . CPT has been identified in more than forty plant species, approximately half of which belong to the order Icacinales 14 . While high-quality genomes of C. acuminata 4 and O. pumila 15 have been reported, no CPT-producing species from Icacinales have been sequenced. In this study, we sequenced and assembled the chromosome-level allotetraploid genome of N. nimmoniana , providing the first genome sequence from Icacinales. We characterized LAMT, SLS, STR, and SEO enzymes in the genome and identified at least one functional copy of each enzyme in each subgenome, suggesting that allotetraploidization likely contributes to the high CPT content in this species. Additionally, phylogenetic analysis revealed that the STR and SEO enzymes have evolved independently in N. nimmoniana , C. acuminata , and O. pumila . This genomic resource provides a foundation for elucidating the phylogenetic position of Icacinales and investigating the evolution of CPT biosynthesis within asterids. Furthermore, this study advances our understanding of the CPT biosynthetic pathway, enhances the feasibility of CPT production via synthetic biology, and supports the potential for increasing the CPT content in N. nimmoniana through genome-assisted breeding strategies. Results Genome assembly and annotation of N. nimmoniana Flow cytometry analysis revealed that the genome size of N. nimmoniana was approximately 5 Gb, whereas 17-mer distribution analysis revealed a genome size of 5,046.54 Mb with a heterozygosity rate of 0.23% (Supplementary Fig. 2). The N. nimmoniana genome was assembled using PacBio HiFi reads (~ 171 Gb, 34x coverage) (Supplementary Table 1) with Hifiasm (v0.16.0) 16 . The assembled genome was 5,075.69 Mb in size and contained 732 contigs with a contig N50 of 159.63 Mb, which closely matches the estimated genome size (Supplementary Table 2). Based on chromosome counts from karyotype analysis (Supplementary Fig. 3), these contigs were further anchored onto twenty-eight pseudochromosomes using Hi-C data (~ 509 Gb, 100x coverage) (Supplementary Fig. 4 and Fig. 1 ), which covered 98.93% of the assembled sequence. The final assembled genome spanned a total length of 5,075.71 Mb, with a scaffold N50 of 183.65 Mb. To evaluate genome accuracy, Illumina PE reads were mapped to the assembly, achieving a mapping rate of 99.51%, while the percentage of RNA-seq reads from various tissues was greater than 94% (Supplementary Table 1). Benchmarking Universal Single-Copy Orthologs (BUSCO) analysis revealed 95.70% (2,030 of 2,121) complete core genes, indicating high completeness of the N. nimmoniana genome assembly (Supplementary Table 3). In the N. nimmoniana genome, 56.76% of the sequences were identified as transposable elements (TEs). The majority of the TEs (53.46%) were long terminal repeat retrotransposons (LTR-RTs), with the Gypsy and Copia superfamilies comprising 29.49% and 16.92%, respectively (Supplementary Table 4). A total of 92,630 protein-coding gene models were predicted by integrating de novo gene prediction, transcriptome-based gene prediction, and protein-based homology searches. More than 94% of these genes were functionally annotated using databases such as NR, UniProt, Swiss-Prot, Pfam, GO, and KEGG (Supplementary Table 5). Additionally, 683 miRNAs, 5,537 tRNAs, 27,760 rRNAs, and 33,348 snRNAs were annotated. Subgenome characteristics Syntenic analysis of the N. nimmoniana genome suggested that this species may have undergone polyploidization (Fig. 1 ). Principal component analysis (PCA) based on TE profiles clearly separated the chromosomes into two distinct groups, and with SUBPHASER 17 , we successfully assigned 28 pseudochromosomes into two subgenomes, designated subgenome A and subgenome B, indicating that N. nimmoniana is an allotetraploid (Fig. 2 a, c and Supplementary Fig. 5). Subgenome A is approximately 334 Mb larger than subgenome B is, largely owing to differences in repeat sequence content (Supplementary Table 6). Subgenome B contains significantly lower levels of TEs, particularly within the Gypsy and Copia superfamilies (Supplementary Table 7). Additionally, subgenome B presented greater gene density than subgenome A did (Supplementary Fig. 6). A pairwise whole-genome comparison revealed 10,721,604 single-nucleotide polymorphisms (SNPs), 1,227,690 insertions/deletions (indels), 600 inversions, 14,174 translocations, and 20,713 duplications between the subgenomes, indicating substantial structural differences between homeologous chromosomes (Fig. 2 c and Supplementary Table 8). Gene expression dominance in one subgenome is commonly observed across polyploid species. To examine whether this trend is present in the N. nimmoniana genome, we compared the transcriptional levels of homoeologous gene pairs from the two subgenomes across eight different tissues (Fig. 2 d). Among the 29,549 homoeologous gene pairs, 23,879 pairs (~ 81%) presented more than a twofold difference in expression in at least one tissue, with subgenome B consistently showing a greater number of highly expressed homoeologs than subgenome A did (Supplementary Table 9). Notably, 3,994 homoeologous gene pairs demonstrated extreme expression divergence (greater than a 32-fold difference), with subgenome B containing 180 more highly expressed homoeologs than subgenome A did. These findings suggest a subtle but clear bias in homoeolog expression dominance toward subgenome B. Additionally, we calculated the TE density around homoeologous genes with expression differences, revealing that dominant genes (DGs) had significantly lower surrounding TE density than did submissive genes (SGs) within each subgenome (Fig. 2 e). These findings suggest that TEs may negatively regulate the expression of submissive genes in N. nimmoniana , which is consistent with findings in Brachypodium hybridum 18 . Taken together, the observed lower TE content, greater gene density, and greater number of highly expressed homoeologs in subgenome B than in subgenome A suggest that subgenome B is the dominant subgenome. Evolution of the N. nimmoniana genome N. nimmoniana is the first sequenced species in the order Icacinales and is positioned at the base of lamiids. The genome assembly presented here will serve as a valuable resource for advancing research on the phylogeny and genome evolution of Icacinales. The evolutionary dynamics of gene families were investigated by comparing the N. nimmoniana genome to those of 12 representative plant species, including four MIA-producing species: O. pumila , C. acuminata , C. roseus , and G. sempervirens (Fig. 3 a). In total, 38,555 genes from the A subgenome and 38,571 genes from the B subgenome were assigned to 17,210 and 17,258 gene families, respectively. Among these, 458 gene families were unique to the A subgenome, whereas 489 were specific to the B subgenome (Fig. 3 a). Furthermore, we identified 5,975 gene families that expanded (2,887 in subgenome A and 3,088 in subgenome B) and 2,455 gene families that contracted (1,272 in subgenome A and 1,183 in subgenome B) following the divergence of the two subgenomes (Fig. 3 a). Phylogenetic and molecular dating analyses of 493 shared single-copy genes revealed that the ancestors of N. nimmoniana subgenomes A and B diverged from the common ancestor of Gentianales and Lamiales approximately 102.0 million years ago (Fig. 3 a). Polyploidization, or whole-genome duplication (WGD), results in the retention of hundreds to thousands of duplicate genes, which provides evolutionary potential for the development of novel gene functions. To investigate WGD events in the N. nimmoniana genome, we identified syntenic blocks within each subgenome. In total, 1,018 and 989 syntenic blocks, containing 22,728 and 22,325 paralogous gene pairs, were identified in the A and B subgenomes, respectively. The density distribution of the synonymous substitution rate (Ks) between collinear paralogous genes revealed peaks for the A (Ks = 0.191) and B (Ks = 0.206) subgenomes, suggesting that a WGD event occurred approximately 33.87–36.59 Mya. Two prominent peaks were observed in the comparison between the A and B subgenomes of N. nimmoniana (Fig. 3 b). These peaks correspond to a speciation event (Ks = 0.075, ~ 13.43 Mya) and an ancestral WGD event (Ks = 0.186, ~ 30.58 Mya), suggesting that the WGD event occurred in the common ancestor of these subgenomes prior to their evolutionary separation (Fig. 3 b). Dot plots (Fig. 2 c) illustrated paralogs inherited from whole-genome triplication (WGT)-γ in the grape genome, revealing a 3–3 diagonal relationship. Similarly, 4–4 diagonal relationships were observed within subgenomes A and B (Fig. 3 c), as well as between the two subgenomes (Supplementary Fig. 7), indicating the retention of two successive WGD events in each subgenome. Additionally, analysis of gene duplication types using MCScanX revealed that WGD/segmental duplications are the most prevalent in both subgenomes (Supplementary Fig. 8), suggesting that the retention of whole-genome duplicated genes plays a crucial role in gene expansion in N. nimmoniana . Characterization of the enzymes involved in CPT biosynthesis in N. nimmoniana NnCYP72A1 (accession no. AQW38832.1), a secologanin synthase, catalyzes the conversion of loganin to secologanin and is, to date, the only enzyme characterized in camptothecin (CPT) biosynthesis in N. nimmoniana . In this study, we identified two candidate secologanin synthases, a homoeologous gene pair (NnSLS1: Nni_ChrA05G28720.1 and NnSLS2: Nni_ChrB05G22310.1), which shares high sequence identity with NnCYP72A1 (98.66% and 96.74%, respectively) and has the highest expression level in roots (Fig. 4 a). To elucidate their catalytic roles, recombinant NnSLS1 and NnSLS2 were expressed in the Saccharomyces cerevisiae strain WAT11 (Supplementary Fig. 9). After microsomes were prepared, in vitro enzymatic assays were performed using loganin as the substrate. Both NnSLS1 and NnSLS2 successfully catalyzed the conversion of loganin to secologanin, as confirmed by UPLC‒MS/MS analysis, which matched the retention time and MS/MS spectra of an authentic standard (Fig. 4 b, c). Using loganic acid as a substrate resulted in no conversion to secologanic acid (Supplementary Fig. 10), which is in agreement with the established role of most SLS enzymes. An exception is found in C. acuminata , where loganic acid is converted to secologanic acid by CaSLASs (Supplementary Fig. 1). Given that NnSLS1 and NnSLS2 utilize loganin as a substrate, we inferred that loganic acid is likely converted to loganin by an LAMT positioned immediately upstream of SLS in the CPT biosynthetic pathway in N. nimmoniana . LAMTs belong to the plant SABATH (salicylic acid/benzoic acid/theobromine) methyltransferase family, which catalyzes the SAM (S-adenosyl-methionine)-dependent methylation of a variety of substrates in plants, such as plant hormones and other small molecules. We identified 20 full-length SABATH methyltransferase genes across both subgenomes, and phylogenetic analysis clustered NnLAMT1 (Nni_ChrA07G12930.1) and NnLAMT2 (Nni_ChrB07G25210.1), a homoeologous gene pair, with the previously characterized OpLAMT 3 and CrLAMT 19 (Supplementary Fig. 11). Expression profiling across eight tissues revealed high coexpression of NnLAMT1/2 with NnSLS1/2, with peak expression in roots (Fig. 4 d and Supplementary Fig. 12). Recombinant NnLAMT1/2, expressed in Escherichia coli , demonstrated the methylation of loganic acid to loganin, which was consistent with CrLAMT activity (Fig. 4 e, f). Additionally, subcellular localization analysis of NnLAMT2-GFP in Nicotiana benthamiana revealed nucleocytosolic localization, which aligns with CrLAMT findings 20 (Fig. 4 g). STR is a key enzyme involved in the biosynthesis of monoterpenoid indole alkaloids (MIAs), which catalyze the condensation of secologanin with tryptamine to form strictosidine. We identified 30 full-length candidate STR genes from the N. nimmoniana genome. Coexpression analysis revealed that six putative STR genes (three homologous gene pairs) were strongly coexpressed with the characterized NnLAMTs and NnSLSs, with the highest expression observed in the roots of N. nimmoniana (Fig. 5 a and Supplementary Fig. 13). Five putative NnSTRs were successfully cloned, and transient expression experiments in N. benthamiana leaves indicated that NnSTR1 and NnSTR3 could catalyze the coupling of secologanin and tryptamine to generate strictosidine (Fig. 5 a, b, c, compared with CrSTR). Additionally, using secologanic acid and tryptamine as substrates led to the production of strictosidinic acid (Supplementary Fig. 14), a substrate promiscuity also observed in OpSTR 3 . Cytochrome P450 enzymes (CYP450s) play a critical role in the epoxidation of strictosamide. Recently, CaCYP71BE206 and OpCYP716E111 were identified as strictosamide epoxidases (SEOs) involved in CPT biosynthesis. Genome mining of N. nimmoniana revealed 344 full-length CYP450s across 48 families, with sixteen exhibiting strong coexpression with characterized NnLAMTs, NnSLSs, and NnSTRs (Supplementary Fig. 15). Fourteen candidate NnCYPs were successfully cloned, and functional analysis using a transient expression system identified three CYP450s—Nni_ChrB12G01000.1 (NnCYP76B), Nni_ChrA12G09540.1 (NnCYP76B), and Nni_ChrB05G34420.1 (NnCYP72A)—as SEOs, which were named NnSEO1-3 (Fig. 5 d, e, f). Among these, NnSEO1 and NnSEO2 form a pair of homoeologous genes. Notably, for all the characterized enzymes in this study, at least one functional copy was retained in each subgenome following allotetraploidization, with all copies exhibiting the highest expression levels in the roots of N. nimmoniana . In addition, candidate genes involved in the prestrictosidine pathway and CPT modification were identified through homology searches and further refined based on their coexpression relationships with the characterized genes (Supplementary Table 10 and Supplementary Fig. 16). Among these candidates, all enzymes except geraniol synthase and 10-hydroxygeraniol oxidoreductase were found to contain at least one homoeologous pair. These findings indicate that allotetraploidization may significantly increase CPT biosynthesis in N. nimmoniana . Independent evolution of CPT biosynthesis Strictosidine synthase has been previously identified in distantly related plants, including C. roseus and Rauvolfia serpentina (Apocynaceae), Gelsemium sempervirens (Gelsemiaceae), O. pumila (Rubiaceae), and C. acuminata (Nyssaceae). While these enzymes belong to the same family, they exhibit low amino acid sequence identity across species from different families (Supplementary Table 11). Notably, the two NnSTRs characterized in this study share only 38.6% amino acid sequence similarity. Genome-wide phylogenetic analysis revealed that both NnSTRs are grouped into two adjacent clades: NnSTR1 clusters with CaSTRs, whereas NnSTR3 groups with STRs from Gentianales species (Fig. 6 a and Supplementary Fig. 17), and both clades include species that have been clearly confirmed to lack strictosidine derivatives. These two clades likely originated from an ancient gene duplication event. Collinearity analysis suggested that this duplication event occurred as a tandem duplication (named AncSTR-like1 and AncSTR-like2) in the last common ancestor of asterid plants (Fig. 6 b and Supplementary Fig. 18). Among the modern plants examined, only N. nimmoniana and O. pumila retain the tandem duplication state of the two ancestral copies. Specifically, NnSTR1 and CaSTRs were derived from AncSTR-like1, whereas NnSTR3 and STRs from Gentianales originated from AncSTR-like2 (Fig. 6 b). Positive selection analysis indicated that the average rate of nonsynonymous substitutions significantly increased in both the AncSTR-like1 (ω(AncSTR-like1) = 1.93165) and AncSTR-like2 (ω(AncSTR-like2) = 6.58244) branches compared with the branch that predated the duplication event (ω₀ = 0.20768), indicating that both branches underwent strong positive Darwinian selection immediately following the duplication event (Supplementary Fig. 19). These findings imply that STR function was promoted after the divergence of AncSTR-like1 and AncSTR-like2, which underwent at least two independent evolutionary events. To investigate the catalytic mechanisms of NnSTRs, 3D protein structure modeling and molecular docking analyses were conducted. Both NnSTRs displayed a conserved six-bladed, four-stranded β-propeller fold structure, which is consistent with the crystal structures previously reported for RsSTR (PDB code: 2FP8), CrSTR (PDB code: 6ZEA), and OpSTR (PDB code: 6S5M) (Supplementary Fig. 20). This structural conservation underscores the evolutionary stability of the STR fold across asterids. Molecular docking analysis suggested that distinct STRs utilize different catalytic residues to initiate the Pictet–Spengler reaction. For example, D140 in NnSTR1 and Y121 in NnSTR3 were predicted to donate a proton to the amine group of tryptamine, thereby facilitating its reaction with the aldehyde group of secologanin to form strictosidine (Fig. 6 c, d and Supplementary Fig. 21). In contrast, D296 was identified as the likely catalytic residue in CacSTR1 (Supplementary Fig. 22). Site-directed mutagenesis of NnSTR1 (D140A) and NnSTR3 (Y121K and E164A) resulted in mutants incapable of producing strictosidine when transiently expressed in N. benthamiana (Figs. 6 e, f), confirming the critical roles of these residues in STR activity. Notably, previous studies reported that glutamic acid serves as the catalytic residue in RsSTR 21 and OpSTR 22 , suggesting that STRs from Gentianales species share a common ancestral catalytic mechanism. These findings indicate that while STRs within asterids have independently evolved distinct active sites to catalyze the condensation of secologanin and tryptamine, they rely on a conserved chemical logic to facilitate this reaction. SEO catalyzes the conversion of strictosamide into strictosamide epoxide, thereby shunting the poststrictosidine biosynthetic flux toward the CPT pathway. SEO activity has been characterized in four CYP450 families across three distantly related plants: the CYP71 family in C. acuminata (CYP71BE206), the CYP716 family in O. pumila (CYP716E111), and the CYP76 and CYP72 families in N. nimmoniana (this study). This finding demonstrates the recruitment of distinct CYP families to catalyze identical biosynthetic steps. The phylogenetic analysis of NnSEOs, OpCYP716E111, CaCYP71BE206, and other well-characterized enzymes (Supplementary Table 12) from the same subfamilies revealed that each SEO is more closely related to other functional enzymes than to one another (Fig. 7 ). These findings strongly suggest that SEO activity evolved independently in N. nimmoniana (Icacinaceae), C. acuminata (Nyssaceae), and O. pumila (Rubiaceae). Additionally, NnSEO3, Nn7DLH-like proteins, and NnSLSs belonging to the CYP72A subfamily were grouped into three distinct subclades. We propose that the emergence of 7DLH, SLS, and SEO activities likely resulted from an ancient gene duplication event within NnCYP72A. On the basis of phylogenetic analyses and synonymous substitution rates (Ks values of 1.61–1.65), this duplication event is estimated to have occurred approximately 124–127 Mya, predating the evolutionary divergence of asterid species. Discussion As one of the most widely used anticancer drugs, CPT was first isolated from C. acuminata . To date, CPT has been identified in 43 angiosperm species, with over 93% of these confined to three orders within the asterid clade: Icacinales (49%), Gentianales (37%), and Cornales (7%) 14 . Among these, N. nimmoniana is the most potent natural producer of CPT. Both C. acuminata and N. nimmoniana serve as major plant sources for the production of CPT and its derivatives. In this study, we sequenced and assembled a chromosome-level genome of N. nimmoniana , representing the first genome sequence from the order Icacinales. This genomic resource provides valuable insights into the biosynthesis of CPT and its evolutionary history. The enzymatic activities of LAMT, SLS and STR were identified in N. nimmoniana , suggesting that this species shares a similar prestrictosidine pathway with major MIA-producing plants. Although both NnSTRs can catalyze the conversion of secologanic acid and tryptamine into strictosidinic acid, this reaction is likely a side activity, comparable to that observed in OpSTR . The presence of LAMT and SLS enzymes in N. nimmoniana suggests that the primary metabolic flux is directed toward strictosidine production, as reported in O. pumila . In C. acuminata , however, loganic acid is converted to secologanic acid by SLAS, which is subsequently coupled with tryptamine by strictosidine acid synthase to form strictosidine acid. This divergence may be exceptional, representing adaptive evolution following the loss of LAMT activity in C. acuminata . Interestingly, three enzymes, two belonging to the CYP76B subfamily and one to the CYP72A subfamily, were identified as SEOs in N. nimmoniana . A similar phenomenon has been reported in the spiroketal steroid biosynthesis of Paris polyphylla , where the conversion of an unstable intermediate to diosgenin is mediated by three enzymes from the CYP94D and CYP72A subfamilies 23 . The presence of multiple enzymes that catalyze the same reaction within a species may facilitate fine-tuned regulation of the reaction intensity in a tissue- or cell-specific manner, as evidenced by their distinct expression patterns. N. nimmoniana , as an allotetraploid, exhibits subtle subgenome dominance. However, our analysis revealed that all known genes involved in CPT biosynthesis, with the exception of GES and 10HGO , retain homoeologous gene pairs. Moreover, these homoeologous gene pairs demonstrate high levels of expression. These observations suggest that the increased gene dosage resulting from allotetraploidization may play a pivotal role in the elevated CPT content observed in this species. In addition, these findings also indicate that N. nimmoniana represents a relatively recent allotetraploid lineage that is currently in the process of diploidization and the progressive establishment of more pronounced subgenome dominance. CPT and all MIAs are synthesized via the iridoid biosynthetic pathway. Iridoids are widely distributed across the asterid clade, where their production is regarded as a synapomorphic trait. However, CPT and MIAs are predominantly restricted to species within the orders Icacinales, Gentianales, and Cornales. To investigate the origin and evolution of enzymes specific to the CPT branch pathway, we conducted phylogenetic and syntenic analyses. Our results revealed that STRs and SEOs were independently recruited in N. nimmoniana , C. acuminata , and O. pumila . Additionally, 3D protein structure reconstruction and molecular docking analyses demonstrated that the STRs in these species evolved distinct active sites independently. These findings suggest that CPT biosynthesis has evolved independently multiple times within the asterid clade. Furthermore, given the widespread production of MIAs among species in Icacinales, Gentianales, and Cornales, it is plausible that STR, as the first enzyme involved in MIA biosynthesis, is monophyletic within each of these orders. Methods Plant material and sequencing The material for genome sequencing was obtained from Fujian Institute of Subtropical Botany (24.52318°N, 118.11537°E) (Supplementary Fig. 23), which was introduced from Lanyu County, Taiwan Province. Young leaves of an individual plant were used for DNA extraction via the Qiagen DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. Two paired-end (PE150) libraries with average insert sizes of 350 bp were constructed and sequenced on the Illumina NovaSeq 6000 platform according to the manufacturer’s instructions (Illumina Inc., USA). Seven HiFi libraries were constructed and sequenced on the Pacific Biosciences Sequel II platform. Four Hi-C libraries were prepared and sequenced on the NovaSeq 6000 platform. Eight tissues from the same plant used for genome sequencing, namely, the root, stem, leaf, petiole, flower bud, flower, seed and seed stalk, with three replicates, were prepared for RNAseq. Karyotyping and genome size estimation For karyotyping, mitotic chromosome slides were prepared as previously described with some modifications 24 . The fresh root tips of N. nimmoniana seedlings were pretreated with aqueous solution saturated with p-dichlorobenzene at room temperature (RT) for 2.5 hours, fixed in 3:1 (v/v) ethanol:acetic acid for 2 hours and stored in 70% ethanol at 4°C until further use. After being washed with distilled water, the root tips were hydrolyzed with 1 mol/L hydrochloric acid at 45°C for 45 minutes. After washing, the digested root tips were stained with Carbol-Fuchsin solution. The traditional squash was performed, and the chromosome number was observed via a Nikon Eclipse E600FN microscope with a Spot RTKE (Diagnostic Instruments) CCD camera. Genome size estimation of N. nimmoniana by flow cytometry followed the method described by Zhixiang Liu 25 et al. with Pepper ( Capsicum annuum ) used as a reference. The relative DNA content of the isolated nuclei was analyzed via a BD FACSCalibur system (Becton Dickinson, New Jersey, USA). The data were acquired and processed by BD FACS DIVA software (v 7.0). The 1N pepper genome was 3,349 Mb 26 ; thus, the N. nimmoniana genome size was ~ 4.998 Gb. The Illumina short PE150 reads (30X coverage) were subjected to genome size and heterozygosity estimation. The 17-mer frequency of short reads was calculated using JellyFish 27 , and the genome size and heterozygosity were subsequently calculated on the basis of the k-mer frequency results. De novo genome assembly and assessment The HiFi reads were generated using CCS and assembled by hifiasm (v0.16.0) 16 with default parameters. The Hi-C reads were processed and aligned to Hi-Fi contigs by Juicer and then clustered into chromosomes via the 3D-DNA package (v180419) 28 with default parameters. We performed manual correction and validation using Juicebox (v1.11.08) to obtain the final chromosome-level nuclear genome sequences. The quality of the final genome sequences was assessed based on multiple datasets. The Illumina PE reads were mapped to the assembly using BWA-MEM (v0.7.16) 29 with default parameters. The RNA-seq data were mapped to the genome using HISAT2 (v2.1.0) 30 . The completeness of the genome assemblies was evaluated using BUSCO (v4.1.2) 31 with the Embryophyta odb10 database. Repeat and gene annotations Repeat sequences were identified using a three-step method. First, the de novo repeat sequence library was established using LTR_FINDER 32 and RepeatModeler 33 . The resulting library was combined with the RepBase database ( https://www.girinst.org/repbase/ ) and then used to mask the genome with RepeatMasker (version 4.1.0) 34 . Second, RepeatProteinMask was used to search against the TE protein database. In addition, TRF ( https://tandem.bu.edu/trf/trf.html ) was used for tandem repeat identification. The results were combined, and duplications were removed to form the final set of repeat sequence annotations. For gene model prediction, three ab initio gene prediction tools were used on the basis of the statistical characteristics of the genomic sequence: AUGUSTUS (v3.1, https://github.com/Gaius-Augustus/Augustus ), GlimmerHMM (v1.2, https://ccb.jhu.edu/software/glimmerhmm ) and SNAP (v2006, https://github.com/KorfLab/SNAP ). Homology-based prediction was performed using BLAST (v2.13.0) and GeneWise ( https://www.ebi.ac.uk/Tools/psa/genewise/ ). The PASA pipeline (v2.0.2, https://github.com/PASApipeline/PASApipeline ) was used to generate the RNA-seq evidence prediction. Finally, all the gene structures predicted by these three methods were integrated into a nonredundant gene set using EVidenceModeler (EVM, v1.1.1, https://github.com/EVidenceModeler/EVidenceModeler ). The resulting gene models were functionally annotated by integrating the annotation information from the NCBI nonredundant protein database (NR, https://www.ncbi.nlm.nih.gov/protein ), SwissProt ( https://www.uniprot.org/ ), KEGG ( https://www.genome.jp/kegg/ ), and InterPro ( https://www.ebi.ac.uk/interpro/ ). To annotate the noncoding RNAs, tRNA genes were identified by tRNAscan-SE ( https://lowelab.ucsc.edu/tRNAscan-SE/ ), whereas ribosomal RNAs (rRNAs) were predicted via BLASTN searches against rRNA sequences with an E value cutoff of 1 × 10 − 10 . Using the covariance model of the Rfam database, INFERNAL (v1.1.4, http://eddylab.org/infernal/ ) was applied to predict miRNAs and snRNAs in the genome. Assigning chromosome assemblies to subgenomes and comparative subgenomic analysis The subgenome-phasing algorithm, SubPhaser (v1.2) 17 , was employed to assign homoeologous chromosomes into two subgenomes. Additionally, subgenomes in N. nimmoniana were distinguished on the basis of TE characteristics between homoeologous chromosomes. A matrix summarizing the copy number of each TE family was built on the 28 chromosomes of N. nimmoniana . This matrix was used for the PCA in the R program. The two subgenomes were subjected to pairwise comparisons using Minimap2 (v2.28) 35 . Syri (v1.7.0) 36 was used to identify synteny and structural rearrangements. Homoeolog expression dominance analysis The RNA-Seq reads were mapped to the genome with HISAT2 (v2.1.0), and the gene expression levels for each RNA-seq sample were estimated using StringTie (v2.2.1) by calculated fragments per kilobase of transcript per million mapped reads (FPKM) values. The homoeolog pairs between two subgenomes were identified via MCScan (Python version) ( https://github.com/tanghaibao/jcvi/wiki ) in ‘-full’ mode to assume 1-to-1 quota synteny blocks. The genome-wide transcriptional levels of subgenomes A and B were reflected in the gene expression levels of homoeologous gene pairs. The log2-fold change in the FPKM values between homoeologous gene pairs was calculated to measure the expression bias. This assessment identified gene pairs exhibiting differential expression exceeding a twofold change threshold as dominant gene pairs. Among these pairs, the genes with relatively high expression levels were designated dominant genes, whereas their counterparts with low expression were categorized as submissive genes. Conversely, the remaining syntenic gene pairs that did not exhibit dominance were classified as neutral genes. Genome evolution Gene clusters associated with 13 other plant species were identified using OrthoFinder (v2.5.4) 37 . Protein and coding sequences from 493 single-copy orthologous gene clusters were used to construct phylogenetic relationships and estimate divergence times. Alignments from MAFFT (v7.471) 38 were converted to coding sequences. IQ-TREE (v2.0.3) 39 was used to construct the phylogenetic tree. The Bayesian relaxed molecular clock (BRMC) approach was used to estimate the species divergence time using the MCMCTree program, which is in the PAML package (v4.9j) 40 . Published species divergence times downloaded from the TimeTree database 41 were used to calibrate the divergence times. The synonymous mutation rate ( ks ) was calculated using the HKY mode, and the WGD time point in each subgenome of N. nimmoniana was estimated according to ks via the method described by Vanneste et al. 42 . Screening of candidate genes involved in CPT biosynthesis Sequences containing PF03492 (SAM-dependent carboxyl methyltransferase), PF00067 (cytochrome P450), or PF03088 (strictosidine synthase) domains were scanned via the hmmsearch program from the HMMER package (v3.3). The hits were then filtered for full-length sequences using a BLAST search against OMT, CYP450, or STR protein sequences from Arabidopsis and sequences from these three families downloaded from SwissProt. The remaining sequences were structurally corrected according to the N. nimmonana transcriptome data with Apollo. Clustering and visualization of gene expression patterns were performed using TBtools. Genes whose expression patterns were consistent with those of the characterized genes or whose expression levels were high in roots were considered candidate genes involved in CPT biosynthesis and were selected for subsequent experiments. Phylogenetic analysis of STR and SEO To infer phylogenetic trees, we analyzed sequences of STR enzymes and STR family sequences from 26 angiosperm species, SEO enzymes and other well-characterized enzymes, including CYP71D, CYP72A, CYP716E, and CYP76B. Sequence alignments were first generated using MAFFT v7.453 and refined through trimAl (V1.5.rev0). The best-fit substitution model for multiple sequence alignment was selected using IQ-TREE (v2.0.3) with ‘-mf’ parameters. The maximum-likelihood gene trees were reconstructed using RAxML-NG (V1.2.2) and visualized using the R package ggtree. Adaptive evolution analysis of functional STRs after gene duplication was performed using the PAML program codeml according to the method described by Bielawski 43 . The null model (the one-ratio model) assumes the same ω ratio (nonsynonymous/synonymous nucleotide substitution rate ratio, dN/dS) for all branches. Nested models are constructed based on the assumption that selective constraints change following gene duplication. Model R2 (two-ratio model) assumes two independent ω ratios: one ratio for all branches predating a duplication event and a second for all branches postdating the tandem duplication event of the STR ancestor. Model R3 assumes three independent ω ratios: one for all branches predating the duplication event, a second for the branches immediately following the duplication event, and a third for all subsequent branches. Model free ratios assume an independent ω ratio for each branch. A likelihood ratio test (LRT) of the one-ratio model with Model R2/R3/free ratios was used to examine the difference between average selective constraints before and after a tandem duplication event of the STR ancestor. Protein structure prediction and molecular docking The putative protein structures of functional STRs were predicted by AlphaFold3. The chemical structures of tryptamine and secologanin were downloaded from PubChem ( https://pubchem.ncbi.nlm.nih.gov/ ) and saved in mol2 format using Open Bable GUI for molecular docking as ligands. Molecular docking of NnSTR1, NnSTR3 and CacSTR with tryptamine and secologanin was performed using AutoDock Vina 1.2.5 44, 45 . PyMOL 3.0 was used for viewing the molecular interactions and image processing. The 2D images of the docking results are presented by a ligand interaction diagram model in Schrödinger suite 2023. Gene cloning Total RNA was extracted from N. nimmoniana roots, C. roseus leaves, and O. pumila leaves using an RNAprep Pure Plant Plus Kit (TIANGEN, catalog no. DP411). First-strand cDNA was synthesized from total RNA using the PrimeScript™ IV 1st strand cDNA Synthesis Mix (TaKaRa, catalog no. 6215A) according to the manufacturer’s protocol. The full-length CDSs of candidate NnSLS s, NnSTR s, and NnSEO s and CrLAMT , CrSTR , OpCYP716E111 , were amplified using PrimeSTAR® GXL DNA Polymerase (TaKaRa, catalog no. 6215A), and the primers used are listed in Supplementary Table 13. NnLAMT1 and NnLAMT2 were synthesized and inserted into the Nde I/Xho I sites of pET-28a vector (GENEWIZ, Beijing, China). Functional characterization of NnLAMT The plasmids pET28a-NnLAMT1 and pET28a-LAMT2 were subsequently transformed into E. coli BL21 (DE3). Single colonies for each construct were inoculated in 10 mL of liquid Luria Bertani (LB) medium containing 50 mg/L kanamycin, followed by cultivation at 37°C with shaking at 220 rpm for 12 hours. The cultures were then transferred into 1 L of fresh liquid LB medium and grown until the optical density at 600 nm (OD 600 ) reached 0.6. Protein expression was induced by adding isopropyl β-d-thiogalactoside (IPTG) to a final concentration of 0.02 mM, followed by incubation at 16°C with shaking at 110 rpm for 18 hours. The recombinant His-fusion proteins were purified using Ni-agarose resin (Mei5bio, catalog no. MF205-01) according to the manufacturer’s instructions. The target proteins were desalted using a PD-10 column. The protein concentration was determined via a BCA protein quantification kit (Vazyme, catalog no. E112-01). A total of 100 µL of the reaction mixture containing 20 µg of protein, 2 mM loganic acid and 1 mM S-adenosyl-L-methionine (SAM) was incubated at 30°C for 180 min. The reaction was stopped by adding 100 µL of methanol and vortexing for 5 min. After centrifugation and filtration, 1 µL of the enzyme reaction mixture was detected via LC‒MS/MS (ACQUITY UPLC I-Class System with Xevo G2 Q-TOF (Waters, Milford, MA, USA)). A UPLC BEH C18 column (1.7 µm, 2.1×100 mm, Waters) was used. The mobile phase consisted of 0.1% formic acid aqueous solution (A) and acetonitrile (B). The gradient was as follows: 5% B solution, 0–3 min; 5–30% B, 3–12 min; 30–95% B, 12–15 min; 95% B, 15–18 min; 95–5% B, 18–21 min; and 5% B, 21–24 min. The conditions of the mass spectrum detector were as follows: cone hole voltage, 40 V; capillary voltage, 3 kV; desolvent temperature, 250°C; ion source temperature, 100°C; desolvent gas flow rate, 600 L/Hr; cone hole flow rate, 50 L/Hr; collision energy, 6 kV; and positive ion electrospray mode. MassLynx 4.1 software was used for data acquisition and analysis. Enzyme assay of NnSLS The candidate NnSLS and CrSLS genes were subsequently cloned and inserted into the BamH I/Xho I sites of the yeast expression pYES2-CT vector using a ClonExpress II One Step Cloning Kit (Vazyme, catalog no. C112). The recombinant plasmids were subsequently introduced into the S. cerevisiae strain WAT11. Positive transformants were screened on solid SD-Ura medium (SD dropout medium without uracil) containing 20 g/L glucose and then cultured in 10 ml of liquid SD-Ura medium until the OD 600 reached 1.5-2. The cells were subsequently centrifuged at 4,000 rpm for 10 min and washed three times with ddH 2 O to remove glucose residue. The cell precipitates were then transferred into YPGal medium containing 2% galactose to induce the expression of the target protein. The microsomes were prepared as previously reported by Yang 9 , the culture was centrifuged, and the cell pellets were washed twice with TEK buffer (50 mM Tris-HCl, 1 mM EDTA, 100 mM KCl, pH 8.0). The cells were resuspended in TES buffer (50 mM Tris-HCl, 1 mM EDTA, 0.6 M sorbitol, and 1 M DTT, pH 8.0) and then disrupted using a high-pressure homogenizer. The homogenate was subsequently centrifuged at 11,000 rpm for 30 min twice. The supernatant was centrifuged at 1,000,000 rpm for 90 min to obtain the microsomal fraction. The resulting microsomal fraction was dissolved in TEG buffer (50 mM Tris-HCl, 1 mM EDTA, 20% glycerol, pH 8.0). For the enzyme activity assay, 100 µL of microsomal suspension was mixed with 500 µM NADPH and 1 mM substrate (loganin or loganic acid). The mixture was incubated at 30°C for 2 hours, and the reaction was terminated with 100 µL of methanol. After centrifugation and filtration, 5 µL of the enzyme reaction mixture was detected with a Thermo Scientific Orbitrap Exploris 120 (Fisher Scientific, Waltham, MA, USA). A Waters ACQUITY UPLC BEH C18 column (1.7 µm, 2.1×100 mm) was used. The mobile phase consisted of 0.1% formic acid aqueous solution (A) and acetonitrile (B), with a flow rate of 300 µL/min. The gradient conditions were as follows: 0–3 min, 5% B; 3–12 min, 5–30% B; 12–15 min, 30–95% B; 15–18 min, 95% B; 18–21 min, 95–5% B; and 21–24 min, 5% B. The MS parameters in ESI-positive mode were as follows: sheath gas, 50 psi; aux gas, 10 psi; sweep gas, 0 psi; ion transfer tube temperature, 320°C; vaporizer temperature, 320°C; and positive ion, 3.5 kV. Qualitative analysis was performed using the XCalibur™ (v. 4.4.16.14) software. Functional characterization of NnSTR and NnSEO in Nicotiana benthamiana The candidate NnSTR and NnSEO genes were ligated into the pEAQ-HT-DEST1 plasmid using the Gateway cloning technology. The recombinant plasmids were subsequently transformed into Agrobacterium tumefaciens (GV3101). Positive transformants were screened on solid LB media supplemented with 50 µg/mL kanamycin and 25 µg/mL gentamycin and then cultured in liquid LB media supplemented with the same antibiotics until the OD 600 reached 0.9. The cells were subsequently centrifuged at 3,000 rpm at 4°C for 10 min and resuspended in MMA buffer (10 mM MES, 30 mM glucose, 10 mM MgCl 2 , 100 µM acetosyringone) with OD 600 = 0.6. After incubation at room temperature for 2 h, A. tumefaciens suspensions were infiltrated into N. benthamiana leaves using a syringe. After 3 days, 0.3 mM substrates were infiltrated into previously generated Agrobacterium- infiltrated leaves for an additional 2 days. The infiltrated N. benthamiana leaves were collected for metabolite analysis via LC‒MS. Subcellular localization The NnLAMT2 gene was cloned and inserted into the SpeI site of the pCAMBIA1302 vector. The recombinant plasmids were transiently expressed in N. benthamiana leaves. N. benthamiana leaf protoplasts were isolated as described by Vivien Rolland 46 . Briefly, the cut leaf strips (0.5–1 mm) were immersed in an enzyme mixture (1.5% cellulase R-10, 0.3% macerozyme R-10, 0.4 M mannitol, 0.2 M MES, 0.1% BSA and 10 mM CaCl 2 , 0.2 M KCl pH 5.7) at 25°C and 50 rpm for 3 hours. After passing through 70-µm strainers, the protoplasts were centrifuged at 100 × g for 5 min and washed once with protoplasting solution without enzymes. The protoplasts were stained with 3 volumes of DAPI staining solution (Biosharp, catalog no. BL105A) for 3 min‒5 min. Fluorescence imaging was performed via a Leica SP8 laser scanning confocal microscope. Declarations Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Data availability The genome used in this study has been deposited in China National Center for Bioinformation (CNCB) under BioProject number PRJCA033896. Competing interests The authors declare no competing interests. Author contributions Y.L., C.S. and S.C. conceived and designed the project. X.S., Z.G., K.D., Y.Z. and L.Q. performed the experiments. X.S., Y.L., S.S., C.Z., G.S., S.G., J.X., W.C. and L.L. analyzed the data. X.S., Z.G. and C.S. wrote the manuscript draft. R.S. and Z.H. provided plant samples. X.S., C.S., B.G., L.Y., J.W., L.S., L.X., W.S., Z.X., X.L., V.C. and S.C. revised the manuscript. Acknowledgments This work was supported by National Key Research and Development Program (NO. 2024YFD2100700) and the CAMS Innovation Fund for Medical Sciences (CIFMS) [grant number 2021-I2M-1-032]. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results. References Liu YQ, et al. Perspectives on biologically active camptothecin derivatives. Med. Res. Rev. 35 , 753–789 (2015). Mingzhang A, et al. Camptothecin distribution and content in Nothapodytes nimmoniana. Nat. Prod. Commun. 6 , 197–200 (2011). Yang M, et al. Divergent camptothecin biosynthetic pathway in Ophiorrhiza pumila . BMC Biol. 19 , 122 (2021). Kang M, et al. A chromosome-level Camptotheca acuminata genome assembly provides insights into the evolutionary origin of camptothecin biosynthesis. Nat. Commun. 12 , 3531 (2021). Yang L, et al. A homomeric geranyl diphosphate synthase-encoding gene from Camptotheca acuminata and its combinatorial optimization for production of geraniol in Escherichia coli . J. Ind. Microbiol. Biotechnol. 44 , 1431–1441 (2017). Awadasseid A, et al. Characterization of Camptotheca acuminata 10-hydroxygeraniol oxidoreductase and iridoid synthase and their application in biological preparation of nepetalactol in Escherichia coli featuring NADP+ - NADPH cofactors recycling. Int. J. Biol. Macromol. 162 , 1076–1085 (2020). Sadre R, et al. Metabolite Diversity in Alkaloid Biosynthesis: A Multilane (Diastereomer) Highway for Camptothecin Synthesis in Camptotheca acuminata . Plant cell 28 , 1926–1944 (2016). Liu Z, et al. Catalytic selectivity and evolution of cytochrome P450 enzymes involved in monoterpene indole alkaloids biosynthesis. Physiol. Plant 176 , e14515 (2024). Yang Y, et al. Bifunctional Cytochrome P450 Enzymes Involved in Camptothecin Biosynthesis. ACS Chem. Biol. 14 , 1091–1096 (2019). Pu X, et al. Proteomics-Guided Mining and Characterization of Epoxidase Involved in Camptothecin Biosynthesis from Camptotheca acuminata . ACS Chem. Biol. 18 , 1772–1785 (2023). Yamazaki Y, et al. Camptothecin biosynthetic genes in hairy roots of Ophiorrhiza pumila : cloning, characterization and differential expression in tissues and by stress compounds. Plant Cell Physiol. 44 , 395–403 (2003). Zhang T, et al. Chemoproteomics reveals the epoxidase enzyme for the biosynthesis of camptothecin in Ophiorrhiza pumila . J. Integr. Plant Biol. 66 , 1044–1047 (2024). Rather GA, et al. Molecular characterization and overexpression analyses of secologanin synthase to understand the regulation of camptothecin biosynthesis in Nothapodytes nimmoniana (Graham.) Mabb. Protoplasma 257 , 391–405 (2020). Pu X, et al. Possible clues for camptothecin biosynthesis from the metabolites in camptothecin-producing plants. Fitoterapia 134 , 113–128 (2019). Rai A, et al. Chromosome-level genome assembly of Ophiorrhiza pumila reveals the evolution of camptothecin biosynthesis. Nat. Commun. 12 , 405 (2021). Cheng H, et al. Haplotype-resolved de novo assembly using phased assembly graphs with hifiasm. Nat. Methods 18 , 170–175 (2021). Jia K-H, et al. SubPhaser: a robust allopolyploid subgenome phasing method based on subgenome-specific k-mers. New Phytol. 235 , 801–809 (2022). Mu W, et al. Subgenomic Stability of Progenitor Genomes During Repeated Allotetraploid Origins of the Same Grass Brachypodium hybridum . Mol. Biol. Evol. 40 , msad259 (2023). Petronikolou N, et al. Loganic Acid Methyltransferase: Insights into the Specificity of Methylation on an Iridoid Glycoside. ChemBioChem 19 , 784–788 (2018). Guirimand G, et al. The subcellular organization of strictosidine biosynthesis in Catharanthus roseus epidermis highlights several trans-tonoplast translocations of intermediate metabolites. FEBS J. 278 , 749–763 (2011). Ma X, et al. The structure of Rauvolfia serpentina strictosidine synthase is a novel six-bladed beta-propeller fold in plant proteins. Plant Cell 18 , 907–920 (2006). Eger E, et al. Inverted Binding of Non-natural Substrates in Strictosidine Synthase Leads to a Switch of Stereochemical Outcome in Enzyme-Catalyzed Pictet–Spengler Reactions. J. Am. Chem. Soc. 142 , 792–800 (2020). Christ B, et al. Repeated evolution of cytochrome P450-mediated spiroketal steroid biosynthesis in plants. Nat. Commun. 10 , 3206 (2019). Kynast RG, et al. Chromosome behavior at the base of the angiosperm radiation: karyology of Trithuria submersa (Hydatellaceae, Nymphaeales). Am. J. Bot. 101 , 1447–1455 (2014). Liu Z, et al. Genome size estimation of Chinese cultured Artemisia annua L. J. Plant Biol. Crop Res. 1 , 1002 (2018). Qin C, et al. Whole-genome sequencing of cultivated and wild peppers provides insights into Capsicum domestication and specialization. Proc. Natl Acad. Sci. USA 111 , 5135–5140 (2014). Marcais G, Kingsford C. A fast, lock-free approach for efficient parallel counting of occurrences of k-mers. Bioinformatics 27 , 764–770 (2011). Dudchenko O, et al. De novo assembly of the Aedes aegypti genome using Hi-C yields chromosome-length scaffolds. Science 356 , 92–95 (2017). Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25 , 1754–1760 (2009). Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nat. Methods 12 , 357–360 (2015). Simao FA, et al. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics 31 , 3210–3212 (2015). Xu Z, Wang H. LTR_FINDER: an efficient tool for the prediction of full-length LTR retrotransposons. Nucleic. Acids Res. 35 , W265-W268 (2007). Flynn JM, et al. RepeatModeler2 for automated genomic discovery of transposable element families. Proc Natl Acad Sci U S A 117 , 9451–9457 (2020). Tarailo-Graovac M, Chen N. Using RepeatMasker to identify repetitive elements in genomic sequences. Curr. Protoc. Bioinformatics Chap. 4 , 4.10.11–14.10.14 (2009). Li H. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics 34 , 3094–3100 (2018). Goel M, et al. SyRI: finding genomic rearrangements and local sequence differences from whole-genome assemblies. Genome Biol 20 , 277 (2019). Emms DM, Kelly S. OrthoFinder: phylogenetic orthology inference for comparative genomics. Genome Biol. 20 , 238 (2019). Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol. Biol. Evol. 30 , 772–780 (2013). Nguyen LT, et al. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol Biol Evol 32 , 268–274 (2015). Yang Z. PAML 4: phylogenetic analysis by maximum likelihood. Mol. Biol. Evol. 24 , 1586–1591 (2007). Kumar S, et al. TimeTree 5: An Expanded Resource for Species Divergence Times. Mol. Biol. Evol. 39 , (2022). Vanneste K, et al. Analysis of 41 plant genomes supports a wave of successful genome duplications in association with the Cretaceous-Paleogene boundary. Genome Res. 24 , 1334–1347 (2014). Bielawski JP, Yang Z. Maximum likelihood methods for detecting adaptive evolution after gene duplication. J. Struct. Funct. Genomics 3 , 201–212 (2003). Eberhardt J, et al. AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings. J. Chem. Inf. Model 61 , 3891–3898 (2021). Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 31 , 455–461 (2010). Rolland V. Determining the Subcellular Localization of Fluorescently Tagged Proteins Using Protoplasts Extracted from Transiently Transformed Nicotiana benthamiana Leaves. Methods Mol. Biol. 1770 , 263–283 (2018). Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryTables.xlsx Dataset 1 Supplementaryfiles.docx The Nothapodytes nimmoniana genome provides insights into the independent evolution of camptothecin biosynthesis in asterids Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5971869","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":412813270,"identity":"e5c25ec4-6b8d-406f-8c4a-4803d0ca746e","order_by":0,"name":"Xiaofeng Shen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIiWNgGAWjYDAC5gNsQNKGB8hqYOAhSgtbAkhLGlAxI2laDjMQr8XgGPuzBx93nJfRnd3Y+OENg508A/vZA3i1SLbxmBvOPHObx+zOwWbJOQzJhg08eQl4tfDL97BJ87YBtdxIbJDmYWBOYJDgMcDvFTb2Z9J/286BtDT/5mGoJ6yFn43BTJqx7QBISxvQlsOEtQD9YibZ25YM8kub5RyD44ZtPDn4tYBCTOJnm5292e3mwzfeVFTL87Ofwa8FASTAJgB9R6R6mJZRMApGwSgYBVgAACINPLVqFG1KAAAAAElFTkSuQmCC","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Union Medical College","correspondingAuthor":true,"prefix":"","firstName":"Xiaofeng","middleName":"","lastName":"Shen","suffix":""},{"id":412813271,"identity":"dc54dcbc-2e74-4e8f-9304-55f4dca0f1cd","order_by":1,"name":"Xiaofeng Shen","email":"","orcid":"","institution":"The Institute of Chinese Materia Medica, China Academy of Chinese Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xiaofeng","middleName":"","lastName":"Shen","suffix":""},{"id":412813272,"identity":"0890ff54-846c-460c-b9a2-158ed6be901f","order_by":2,"name":"Zhijing Guan","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Zhijing","middleName":"","lastName":"Guan","suffix":""},{"id":412813273,"identity":"333f6a17-7597-4683-8eaa-2e9a3b30c872","order_by":3,"name":"Sijie Sun","email":"","orcid":"https://orcid.org/0000-0002-4786-6819","institution":"National Genomics Data Center, Beijing Institute of Genomics, Chinese Academy of Sciences and China National Center for Bioinformation","correspondingAuthor":false,"prefix":"","firstName":"Sijie","middleName":"","lastName":"Sun","suffix":""},{"id":412813274,"identity":"9dc4440d-6b31-45a0-a466-3b1018dffe94","order_by":4,"name":"Ke Du","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Unio","correspondingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Du","suffix":""},{"id":412813275,"identity":"9427879f-bdcc-422b-a508-394a8d2bfdeb","order_by":5,"name":"Yue Zhao","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Unio","correspondingAuthor":false,"prefix":"","firstName":"Yue","middleName":"","lastName":"Zhao","suffix":""},{"id":412813276,"identity":"ceddd014-2bd3-4446-b73c-7b2ade3f0af6","order_by":6,"name":"Zhaojiu Yan","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Unio","correspondingAuthor":false,"prefix":"","firstName":"Zhaojiu","middleName":"","lastName":"Yan","suffix":""},{"id":412813277,"identity":"e5d38c0a-f97b-4ae6-aaaa-bfce5b405bd0","order_by":7,"name":"Chuyi Zhang","email":"","orcid":"","institution":"Chinese Academy of Medical Science \u0026 Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Chuyi","middleName":"","lastName":"Zhang","suffix":""},{"id":412813278,"identity":"5cbf3111-753b-4d6a-9c1a-7d49163f290a","order_by":8,"name":"Li Qiu","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Unio","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Qiu","suffix":""},{"id":412813279,"identity":"74e55b97-4e52-4428-9bd6-7966b24353c5","order_by":9,"name":"Lingling Yun","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Unio","correspondingAuthor":false,"prefix":"","firstName":"Lingling","middleName":"","lastName":"Yun","suffix":""},{"id":412813280,"identity":"26d4428b-d04a-4a63-98b7-fd911f60fec9","order_by":10,"name":"Ruichi Shen","email":"","orcid":"","institution":"Fujian Institute of Subtropical Botany, Fujian Key Laboratory of Physiology and Biochemistry for Subtropical Plant","correspondingAuthor":false,"prefix":"","firstName":"Ruichi","middleName":"","lastName":"Shen","suffix":""},{"id":412813281,"identity":"d4b91f74-9235-4e95-b428-81f29684613f","order_by":11,"name":"Bao-Lin Guo","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Bao-Lin","middleName":"","lastName":"Guo","suffix":""},{"id":412813282,"identity":"75c30256-198f-4748-b0d4-9065e309aab4","order_by":12,"name":"Guoan Shen","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Science\u0026Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Guoan","middleName":"","lastName":"Shen","suffix":""},{"id":412813283,"identity":"7a69531e-2097-451e-881e-6bea3cba78ec","order_by":13,"name":"Xiwen Li","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Science\u0026Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Xiwen","middleName":"","lastName":"Li","suffix":""},{"id":412813284,"identity":"c62ae666-d9a0-4cb1-8648-8941c510077c","order_by":14,"name":"Jianhe Wei","email":"","orcid":"","institution":"Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Jianhe","middleName":"","lastName":"Wei","suffix":""},{"id":412813285,"identity":"6ffe753d-8881-4db8-92a9-536e207efecf","order_by":15,"name":"Linchun Shi","email":"","orcid":"","institution":"Chinese Academy of Medical Science \u0026 Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Linchun","middleName":"","lastName":"Shi","suffix":""},{"id":412813286,"identity":"3ae22c7d-3a64-4474-854c-0d22f1278eb2","order_by":16,"name":"Zhigang Hu","email":"","orcid":"","institution":"Hubei University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Zhigang","middleName":"","lastName":"Hu","suffix":""},{"id":412813287,"identity":"32a1e276-40ba-425a-b975-2c0f8b958468","order_by":17,"name":"Li Xiang","email":"","orcid":"","institution":"Institute of Chinese Materia Medica, China Academy of Chinese Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Xiang","suffix":""},{"id":412813288,"identity":"2f7947c3-e098-42a2-afd6-00ba01ce9d3c","order_by":18,"name":"Jiang Xu","email":"","orcid":"","institution":"Institute of Chinese Materia Medica, China Academy of Chinese Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Jiang","middleName":"","lastName":"Xu","suffix":""},{"id":412813289,"identity":"2baa789c-e779-4712-bb72-9492d3f735e1","order_by":19,"name":"Shuai Guo","email":"","orcid":"","institution":"Institute of Chinese Materia Medica, China Academy of Chinese Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Shuai","middleName":"","lastName":"Guo","suffix":""},{"id":412813290,"identity":"4a14140f-88e6-430a-9bd1-1878aeb00223","order_by":20,"name":"Sun Wei","email":"","orcid":"","institution":"China Academy of Chinese Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Sun","middleName":"","lastName":"Wei","suffix":""},{"id":412813291,"identity":"566066b5-2448-4e79-b11c-60491aedd3f9","order_by":21,"name":"Zhichao Xu","email":"","orcid":"https://orcid.org/0000-0003-1753-5602","institution":"College of Life Science, Northeast Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Zhichao","middleName":"","lastName":"Xu","suffix":""},{"id":412813292,"identity":"b8547ba1-19bf-4d2c-bb10-01cc9bf1de76","order_by":22,"name":"Wei Chen","email":"","orcid":"","institution":"Institute of Herbgenomics, Chengdu University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Chen","suffix":""},{"id":412813293,"identity":"7e92cb33-1b38-425d-8919-e36dd50c39e9","order_by":23,"name":"Liang Leng","email":"","orcid":"https://orcid.org/0000-0002-9629-6808","institution":"Institute of Herbgenomics, Chengdu University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Liang","middleName":"","lastName":"Leng","suffix":""},{"id":412813294,"identity":"5ecbc2f6-89a0-4323-aa02-414d3f0993d7","order_by":24,"name":"Vincent Courdavault","email":"","orcid":"https://orcid.org/0000-0001-8902-4532","institution":"Université de Tours","correspondingAuthor":false,"prefix":"","firstName":"Vincent","middleName":"","lastName":"Courdavault","suffix":""},{"id":412813295,"identity":"fc24013b-497c-41e0-9912-508c75e92f05","order_by":25,"name":"Ying Li","email":"","orcid":"","institution":"Chinese Academy of Medical Science \u0026 Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Ying","middleName":"","lastName":"Li","suffix":""},{"id":412813296,"identity":"798d98d9-b857-4ad6-9b68-7615bbcc338a","order_by":26,"name":"Shilin Chen","email":"","orcid":"https://orcid.org/0000-0002-0449-236X","institution":"Institute of Herbgenomics, Chengdu University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Shilin","middleName":"","lastName":"Chen","suffix":""},{"id":412813297,"identity":"7c28003f-5ef0-4c78-8867-59c2b193fea6","order_by":27,"name":"Chao Sun","email":"","orcid":"https://orcid.org/0000-0001-7096-5033","institution":"Institute of Medicinal Plant Development","correspondingAuthor":false,"prefix":"","firstName":"Chao","middleName":"","lastName":"Sun","suffix":""}],"badges":[],"createdAt":"2025-02-06 09:15:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5971869/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5971869/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":76007214,"identity":"d3d3c53d-9725-4d53-9248-732caa2185b2","added_by":"auto","created_at":"2025-02-11 11:34:20","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":4417004,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOverview of the \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eN. nimmoniana\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e genome. \u003c/strong\u003eFrom the outer tracks to the inner tracks, the Fig. displays the following: (a) chromosome ideogram, with a scale label indicating 1 Mb; (b) repeat content, calculated as repeat sequence length per 200 Kb; (c) gene density, shown as the number of genes per 200 Kb; (d) Gypsy element density across chromosomes, shown as Gypsy length per 200 Kb; (e) Copia element density across chromosomes, shown as Copia length per 200 Kb; (f) gene expression levels, represented as the average reads per kilobase per million mapped reads (RPKM) per 200 Kb; and (g) synteny blocks, illustrating conserved regions across the genome.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/247433df464c330646bfa532.png"},{"id":76008065,"identity":"cb765cff-e610-4ae2-b6f4-baf2b99ec861","added_by":"auto","created_at":"2025-02-11 11:42:22","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1807370,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSubgenome phasing and characteristics of the \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eN. nimmoniana\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e genome\u003c/strong\u003e. \u003cstrong\u003ea\u003c/strong\u003e Phasing of the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome using SUBPHASER, organized from outer to inner circles as follows: (1) subgenome assignments determined by k-means clustering; (2) enrichment of subgenome-specific k-mers, where areas matching subgenome colors show significant enrichment; (3) normalized proportion of subgenome-specific k-mers; (4–5) absolute count of each subgenome-specific k-mer set; (6) density of LTR-RTs, with colors corresponding to enriched subgenome-specific k-mers and gray denoting nonspecificLTR-RTs; (7) homologous blocks. All the statistics (2–6) are computed in 1-Mb sliding windows. \u003cstrong\u003eb\u003c/strong\u003e PCA indicating the origin of homoeologous chromosomes in \u003cem\u003eN. nimmoniana\u003c/em\u003e on the basis of TE profiles. \u003cstrong\u003ec\u003c/strong\u003e Structural variations, including SNPs, indels, inversions, translocations, and duplications, identified between the two subgenomes. \u003cstrong\u003ed\u003c/strong\u003eGenome-wide expression histograms of homoeologous genes across eight \u003cem\u003eN. nimmoniana\u003c/em\u003e tissues, where log₂(FPKM_A/FPKM_B) illustrates the expression divergence of homoeologous gene pairs, and \"n\" indicates the number of dominant genes in subgenomes A and B. \u003cstrong\u003ee\u003c/strong\u003e TE density comparisons in homoeologous gene pairs grouped by expression bias type (Dom: dominant; Sub: submissive), where * indicates \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05 and *** indicates \u003cem\u003eP\u003c/em\u003e\u0026lt; 0.001 (paired Wilcoxon test).\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/b9a068aa8a360a218ae9a366.png"},{"id":76007200,"identity":"283bd95a-3d0a-4cbb-ac51-7b9c39c0d6d0","added_by":"auto","created_at":"2025-02-11 11:34:19","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":343450,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvolution of the \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eN. nimmoniana\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e genome.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e Phylogenetic tree of \u003cem\u003eN. nimmoniana\u003c/em\u003e and 12 other angiosperm species, with the indicated WGT and WGD events. \u003cstrong\u003eb\u003c/strong\u003e Distribution of synonymous substitution rates (\u003cem\u003eKs\u003c/em\u003e) for paralogs within the \u003cem\u003eN. nimmoniana\u003c/em\u003e subgenomes, \u003cem\u003eC. acuminata\u003c/em\u003e, and \u003cem\u003eVitis vinifera\u003c/em\u003e, as well as the \u003cem\u003eKs\u003c/em\u003e distribution for orthologs between the two \u003cem\u003eN. nimmoniana\u003c/em\u003e subgenomes. \u003cstrong\u003ec\u003c/strong\u003e Dot plot showing syntenic blocks within the \u003cem\u003eN. nimmoniana\u003c/em\u003e subgenomes and the \u003cem\u003eV. vinifera\u003c/em\u003e genome.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/9316047034b1a2af9adeebae.png"},{"id":76007198,"identity":"bc17fbaf-c4d1-4fe9-911e-ca97feb1f029","added_by":"auto","created_at":"2025-02-11 11:34:19","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1205531,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFunctional analysis of NnSLS and NnLAMT. a\u003c/strong\u003e The catalytic functions of SLS in the conversion of loganin to secologanin. \u003cstrong\u003eb\u003c/strong\u003e UPLC‒MS extracted ion chromatograms (EIC) ([M + H]\u003csup\u003e+ \u003c/sup\u003e= 389.14 ± 0.05 for secologanin) showing the \u003cem\u003ein vitro\u003c/em\u003e activity of NnSLS1, NnSLS2 and CrSLS when loganin was used as the substrate. \u003cstrong\u003ec \u003c/strong\u003eMS/MS fragmentation patterns for the product peak at the corresponding retention time detected in the secologanin standard, CrSLS, NnSLS1, and NnSLS2. \u003cstrong\u003ed\u003c/strong\u003e Catalytic functions of LAMT in the conversion of loganin from loganic acid. \u003cstrong\u003ee\u003c/strong\u003e EIC ([M + H]\u003csup\u003e+\u003c/sup\u003e = 391.16 ± 0.05 for loganin) showing the \u003cem\u003ein vitro\u003c/em\u003e activity of NnLAMT1, NnLAMT2 and CrLAMT with loganic acid as substrates. \u003cstrong\u003ef\u003c/strong\u003e MS/MS fragmentation patterns for the product peak at the corresponding retention time detected in the loganin standard, CrLAMT, NnLAMT1, and NnLAMT2. \u003cstrong\u003eg\u003c/strong\u003e Subcellular localization of NnLAMT2 in \u003cem\u003eN. benthamiana\u003c/em\u003e leaf protoplasts. The expression value for each gene is represented on a log\u003csub\u003e10\u003c/sub\u003e(FPKM+1) scale for eight tissues: root (R), fruit stalk (FrS), fruit (Fr), flower bud (FlB), flower (Fl), stem (S), petiole (P), and leaf (L). Low to high expression is indicated by the change in color from blue to red.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/02d63e4ad20136998c2050a3.png"},{"id":76007199,"identity":"b5b9069c-b119-4423-bfed-a02b69a47185","added_by":"auto","created_at":"2025-02-11 11:34:19","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":266188,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFunctional analysis of NnSTR and NnSEO. a\u003c/strong\u003e The catalytic functions of STR in the formation of strictosidine from tryptamine and secologanin. \u003cstrong\u003eb\u003c/strong\u003e UPLC‒MS EIC of \u003cem\u003eN. benthamiana\u003c/em\u003e leaf extracts after expression of NnSTR1, NnSTR3 and CrSTR using tryptamine and secologanin as substrates. All three enzymes resulted in the production of strictosidine ([M + H]\u003csup\u003e+ \u003c/sup\u003e= 531.23 ± 0.05). \u003cstrong\u003ec \u003c/strong\u003eMS/MS fragmentation patterns of the product peak at the corresponding retention time detected in the strictosidine standard, CrSTR, NnSTR1, and NnSTR3. \u003cstrong\u003ed\u003c/strong\u003e Catalytic functions of NnSEO in the formation of strictosamide epoxide from strictosamide. \u003cstrong\u003ee\u003c/strong\u003e UPLC‒MS EIC of \u003cem\u003eN. benthamiana\u003c/em\u003e leaf extracts after expression of OpSEO, NnSEO1, NnSEO2, and NnSEO3 using strictosamide as the substrate. All four enzymes resulted in the production of strictosamide epoxide ([M + H]\u003csup\u003e+ \u003c/sup\u003e= 515.20 ± 0.05). \u003cstrong\u003ef \u003c/strong\u003eMS/MS fragmentation patterns for the product peak at the corresponding retention time detected in OpSEO, NnSEO1, NnSEO2, and NnSEO3. The expression value for each gene is represented on a log\u003csub\u003e10\u003c/sub\u003e(FPKM+1) scale for eight tissues: root (R), fruit stalk (FrS), fruit (Fr), flower bud (FlB), flower (Fl), stem (S), petiole (P), and leaf (L). Low to high expression is indicated by the change in color from blue to red.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/b5bb49bfa1a49b7b07759dbb.png"},{"id":76007222,"identity":"93a184ec-cec7-4e14-aa6f-970ecab95ad7","added_by":"auto","created_at":"2025-02-11 11:34:37","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1731792,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvolution of STRs. a\u003c/strong\u003ePhylogenetic tree of STRs across 26 angiosperm species. Characterized STR enzymes are marked with red dots. The detailed phylogenetic tree is available in Supplementary Fig. 16. \u003cstrong\u003eb\u003c/strong\u003e Synteny analysis of CrSTR, OpSTRs, GsSTR, NnSTRs, and CaSTRs in \u003cem\u003eC. roseus\u003c/em\u003e, \u003cem\u003eO. pumila\u003c/em\u003e,\u003cem\u003e G. sempervirens\u003c/em\u003e,\u003cem\u003e N. nimmoniana\u003c/em\u003e and\u003cem\u003e C. acuminata\u003c/em\u003e.\u003cstrong\u003e \u003c/strong\u003eA fully annotated synteny analysis is available in Supplementary Fig. 18. \u003cstrong\u003ec-d\u003c/strong\u003e Molecular docking of tryptamine and secologanin with the predicted structures of NnSTR1 and NnSTR3. \u003cstrong\u003ee-f\u003c/strong\u003e UPLC-MS EICs corresponding to the transient expression of NnSTR1, NnSTR3 and the mutants in \u003cem\u003eN. benthamiana\u003c/em\u003e.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/97c75b59b6c0e78f9dc1a7f2.png"},{"id":76007196,"identity":"cff6eed8-41cb-471b-a239-3732177e7bc7","added_by":"auto","created_at":"2025-02-11 11:34:19","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":122287,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMaximum-likelihood phylogenetic analysis of SEO and other well-characterized enzymes from CYP72A, CYP71D, CYP76B, and CYP716E\u003c/strong\u003e. The colored dots indicate enzymes involved in MIA biosynthesis, with red dots specifically representing SEOs participating in CPT biosynthesis. Details of the selected CYPs are provided in Supplementary Table 12.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/7668322a64c7574fdc47ad67.png"},{"id":76008336,"identity":"54fa9adb-4841-48a8-818d-165c12e60654","added_by":"auto","created_at":"2025-02-11 11:50:40","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":19043815,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/c07817e4-b80c-4331-835f-b2d9c1b92b32.pdf"},{"id":76007216,"identity":"617fdb51-aaa0-4479-ac42-c0b373d4f77c","added_by":"auto","created_at":"2025-02-11 11:34:21","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":49474,"visible":true,"origin":"","legend":"Dataset 1","description":"","filename":"SupplementaryTables.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/a49507d8aff3b054831cdb89.xlsx"},{"id":76008064,"identity":"9ace9e56-5ad2-41c6-a6f8-cafca7a9c80c","added_by":"auto","created_at":"2025-02-11 11:42:21","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":124701446,"visible":true,"origin":"","legend":"The Nothapodytes nimmoniana genome provides insights into the independent evolution of camptothecin biosynthesis in asterids","description":"","filename":"Supplementaryfiles.docx","url":"https://assets-eu.researchsquare.com/files/rs-5971869/v1/c22af5f69de10067acb5935c.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"The Nothapodytes nimmoniana genome provides insights into the independent evolution of camptothecin biosynthesis in asterids","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCamptothecin (CPT) is a plant alkaloid with potent anticancer activity that is primarily achieved by impeding cancer cell replication through the inhibition of DNA topoisomerase I\u003csup\u003e1\u003c/sup\u003e. Currently, at least four CPT derivatives\u0026mdash;hydroxycamptothecin, irinotecan, topotecan and belotecan\u0026mdash;have been successfully used in clinical treatments. These derivatives are synthesized predominantly through the modification of CPT obtained from plant sources. \u003cem\u003eNothapodytes nimmoniana\u003c/em\u003e, a member of the Icacinaceae family, is recognized as the most potent CPT producer\u003csup\u003e2\u003c/sup\u003e. However, owing to the increasing global demand for CPT in the pharmaceutical industry, \u003cem\u003eN. nimmoniana\u003c/em\u003e faces the risk of unsustainable harvesting practices. Therefore, elucidating the biosynthetic pathway of CPT is essential for addressing the growing demand through synthetic biology approaches and conserving this endangered species.\u003c/p\u003e \u003cp\u003eCPT is synthesized via the monoterpene indole alkaloid (MIA) biosynthetic pathway. Strictosidine serves as the common intermediate for most MIAs and is formed by the condensation of tryptamine and secologanin. CPT is derived from strictosidine in \u003cem\u003eOphiorrhiza pumila\u003c/em\u003e\u003csup\u003e3\u003c/sup\u003e. In contrast, in \u003cem\u003eC. acuminata\u003c/em\u003e, an alternative biosynthetic pathway has evolved due to the lack of loganic acid O-methyltransferase (LAMT), in which strictosidinic acid, rather than strictosidine, is used as the intermediate for CPT biosynthesis\u003csup\u003e4\u003c/sup\u003e (Supplementary Fig.\u0026nbsp;1). Six enzymes [geraniol synthase (GES)\u003csup\u003e5\u003c/sup\u003e, 10-hydroxygeraniol oxidoreductase (10HGO)\u003csup\u003e6\u003c/sup\u003e, iridodial synthase (IS)\u003csup\u003e7\u003c/sup\u003e, 7-deoxyloganic acid 7-hydroxylase (7-DLH)\u003csup\u003e8\u003c/sup\u003e, secologanin acid synthase (SLAS)\u003csup\u003e9\u003c/sup\u003e, strictosidine synthase (STR)\u003csup\u003e3\u003c/sup\u003e, and strictosamide epoxidase (SEO) (CYPC71BE206)\u003csup\u003e10\u003c/sup\u003e] from \u003cem\u003eC. acuminata\u003c/em\u003e and four enzymes [LAMT\u003csup\u003e3\u003c/sup\u003e, secologanin synthase (SLS)\u003csup\u003e3\u003c/sup\u003e, STR\u003csup\u003e11\u003c/sup\u003e, and SEO (CYP716E111)\u003csup\u003e12\u003c/sup\u003e] from \u003cem\u003eO. pumila\u003c/em\u003e have been identified as being involved in CPT biosynthesis. However, investigations into CPT biosynthesis in \u003cem\u003eN. nimmoniana\u003c/em\u003e remain limited, with only \u003cem\u003eNn\u003c/em\u003eSLS (CYP72A1) characterized in this species\u003csup\u003e13\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCPT has been identified in more than forty plant species, approximately half of which belong to the order Icacinales\u003csup\u003e14\u003c/sup\u003e. While high-quality genomes of \u003cem\u003eC. acuminata\u003c/em\u003e\u003csup\u003e4\u003c/sup\u003e and \u003cem\u003eO. pumila\u003c/em\u003e\u003csup\u003e15\u003c/sup\u003e have been reported, no CPT-producing species from Icacinales have been sequenced. In this study, we sequenced and assembled the chromosome-level allotetraploid genome of \u003cem\u003eN. nimmoniana\u003c/em\u003e, providing the first genome sequence from Icacinales. We characterized LAMT, SLS, STR, and SEO enzymes in the genome and identified at least one functional copy of each enzyme in each subgenome, suggesting that allotetraploidization likely contributes to the high CPT content in this species. Additionally, phylogenetic analysis revealed that the STR and SEO enzymes have evolved independently in \u003cem\u003eN. nimmoniana\u003c/em\u003e, \u003cem\u003eC. acuminata\u003c/em\u003e, and \u003cem\u003eO. pumila\u003c/em\u003e. This genomic resource provides a foundation for elucidating the phylogenetic position of Icacinales and investigating the evolution of CPT biosynthesis within asterids. Furthermore, this study advances our understanding of the CPT biosynthetic pathway, enhances the feasibility of CPT production via synthetic biology, and supports the potential for increasing the CPT content in \u003cem\u003eN. nimmoniana\u003c/em\u003e through genome-assisted breeding strategies.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003e\u003cstrong\u003eGenome assembly and annotation of\u003c/strong\u003e \u003cstrong\u003eN. nimmoniana\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eFlow cytometry analysis revealed that the genome size of \u003cem\u003eN. nimmoniana\u003c/em\u003e was approximately 5 Gb, whereas 17-mer distribution analysis revealed a genome size of 5,046.54 Mb with a heterozygosity rate of 0.23% (Supplementary Fig.\u0026nbsp;2). The \u003cem\u003eN. nimmoniana\u003c/em\u003e genome was assembled using PacBio HiFi reads (~\u0026thinsp;171 Gb, 34x coverage) (Supplementary Table\u0026nbsp;1) with Hifiasm (v0.16.0)\u003csup\u003e16\u003c/sup\u003e. The assembled genome was 5,075.69 Mb in size and contained 732 contigs with a contig N50 of 159.63 Mb, which closely matches the estimated genome size (Supplementary Table\u0026nbsp;2). Based on chromosome counts from karyotype analysis (Supplementary Fig.\u0026nbsp;3), these contigs were further anchored onto twenty-eight pseudochromosomes using Hi-C data (~\u0026thinsp;509 Gb, 100x coverage) (Supplementary Fig.\u0026nbsp;4 and Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e), which covered 98.93% of the assembled sequence. The final assembled genome spanned a total length of 5,075.71 Mb, with a scaffold N50 of 183.65 Mb. To evaluate genome accuracy, Illumina PE reads were mapped to the assembly, achieving a mapping rate of 99.51%, while the percentage of RNA-seq reads from various tissues was greater than 94% (Supplementary Table\u0026nbsp;1). Benchmarking Universal Single-Copy Orthologs (BUSCO) analysis revealed 95.70% (2,030 of 2,121) complete core genes, indicating high completeness of the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome assembly (Supplementary Table\u0026nbsp;3).\u003c/p\u003e\n\u003cp\u003eIn the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome, 56.76% of the sequences were identified as transposable elements (TEs). The majority of the TEs (53.46%) were long terminal repeat retrotransposons (LTR-RTs), with the \u003cem\u003eGypsy\u003c/em\u003e and \u003cem\u003eCopia\u003c/em\u003e superfamilies comprising 29.49% and 16.92%, respectively (Supplementary Table\u0026nbsp;4). A total of 92,630 protein-coding gene models were predicted by integrating \u003cem\u003ede novo\u003c/em\u003e gene prediction, transcriptome-based gene prediction, and protein-based homology searches. More than 94% of these genes were functionally annotated using databases such as NR, UniProt, Swiss-Prot, Pfam, GO, and KEGG (Supplementary Table\u0026nbsp;5). Additionally, 683 miRNAs, 5,537 tRNAs, 27,760 rRNAs, and 33,348 snRNAs were annotated.\u003c/p\u003e\n\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n\u003ch2\u003eSubgenome characteristics\u003c/h2\u003e\n\u003cp\u003eSyntenic analysis of the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome suggested that this species may have undergone polyploidization (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Principal component analysis (PCA) based on TE profiles clearly separated the chromosomes into two distinct groups, and with SUBPHASER\u003csup\u003e17\u003c/sup\u003e, we successfully assigned 28 pseudochromosomes into two subgenomes, designated subgenome A and subgenome B, indicating that \u003cem\u003eN. nimmoniana\u003c/em\u003e is an allotetraploid (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ea, c and Supplementary Fig.\u0026nbsp;5). Subgenome A is approximately 334 Mb larger than subgenome B is, largely owing to differences in repeat sequence content (Supplementary Table\u0026nbsp;6). Subgenome B contains significantly lower levels of TEs, particularly within the \u003cem\u003eGypsy\u003c/em\u003e and \u003cem\u003eCopia\u003c/em\u003e superfamilies (Supplementary Table\u0026nbsp;7). Additionally, subgenome B presented greater gene density than subgenome A did (Supplementary Fig.\u0026nbsp;6). A pairwise whole-genome comparison revealed 10,721,604 single-nucleotide polymorphisms (SNPs), 1,227,690 insertions/deletions (indels), 600 inversions, 14,174 translocations, and 20,713 duplications between the subgenomes, indicating substantial structural differences between homeologous chromosomes (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ec and Supplementary Table\u0026nbsp;8).\u003c/p\u003e\n\u003cp\u003eGene expression dominance in one subgenome is commonly observed across polyploid species. To examine whether this trend is present in the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome, we compared the transcriptional levels of homoeologous gene pairs from the two subgenomes across eight different tissues (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ed). Among the 29,549 homoeologous gene pairs, 23,879 pairs (~\u0026thinsp;81%) presented more than a twofold difference in expression in at least one tissue, with subgenome B consistently showing a greater number of highly expressed homoeologs than subgenome A did (Supplementary Table\u0026nbsp;9). Notably, 3,994 homoeologous gene pairs demonstrated extreme expression divergence (greater than a 32-fold difference), with subgenome B containing 180 more highly expressed homoeologs than subgenome A did. These findings suggest a subtle but clear bias in homoeolog expression dominance toward subgenome B. Additionally, we calculated the TE density around homoeologous genes with expression differences, revealing that dominant genes (DGs) had significantly lower surrounding TE density than did submissive genes (SGs) within each subgenome (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ee). These findings suggest that TEs may negatively regulate the expression of submissive genes in \u003cem\u003eN. nimmoniana\u003c/em\u003e, which is consistent with findings in \u003cem\u003eBrachypodium hybridum\u003c/em\u003e\u003csup\u003e18\u003c/sup\u003e. Taken together, the observed lower TE content, greater gene density, and greater number of highly expressed homoeologs in subgenome B than in subgenome A suggest that subgenome B is the dominant subgenome.\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eEvolution of the\u003c/strong\u003e \u003cstrong\u003eN. nimmoniana\u003c/strong\u003e \u003cstrong\u003egenome\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003e\u003cem\u003eN. nimmoniana\u003c/em\u003e is the first sequenced species in the order Icacinales and is positioned at the base of lamiids. The genome assembly presented here will serve as a valuable resource for advancing research on the phylogeny and genome evolution of Icacinales. The evolutionary dynamics of gene families were investigated by comparing the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome to those of 12 representative plant species, including four MIA-producing species: \u003cem\u003eO. pumila\u003c/em\u003e, \u003cem\u003eC. acuminata\u003c/em\u003e, \u003cem\u003eC. roseus\u003c/em\u003e, and \u003cem\u003eG. sempervirens\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ea). In total, 38,555 genes from the A subgenome and 38,571 genes from the B subgenome were assigned to 17,210 and 17,258 gene families, respectively. Among these, 458 gene families were unique to the A subgenome, whereas 489 were specific to the B subgenome (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ea). Furthermore, we identified 5,975 gene families that expanded (2,887 in subgenome A and 3,088 in subgenome B) and 2,455 gene families that contracted (1,272 in subgenome A and 1,183 in subgenome B) following the divergence of the two subgenomes (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ea). Phylogenetic and molecular dating analyses of 493 shared single-copy genes revealed that the ancestors of \u003cem\u003eN. nimmoniana\u003c/em\u003e subgenomes A and B diverged from the common ancestor of Gentianales and Lamiales approximately 102.0\u0026nbsp;million years ago (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ea).\u003c/p\u003e\n\u003cp\u003ePolyploidization, or whole-genome duplication (WGD), results in the retention of hundreds to thousands of duplicate genes, which provides evolutionary potential for the development of novel gene functions. To investigate WGD events in the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome, we identified syntenic blocks within each subgenome. In total, 1,018 and 989 syntenic blocks, containing 22,728 and 22,325 paralogous gene pairs, were identified in the A and B subgenomes, respectively. The density distribution of the synonymous substitution rate (Ks) between collinear paralogous genes revealed peaks for the A (Ks\u0026thinsp;=\u0026thinsp;0.191) and B (Ks\u0026thinsp;=\u0026thinsp;0.206) subgenomes, suggesting that a WGD event occurred approximately 33.87\u0026ndash;36.59 Mya. Two prominent peaks were observed in the comparison between the A and B subgenomes of \u003cem\u003eN. nimmoniana\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eb). These peaks correspond to a speciation event (Ks\u0026thinsp;=\u0026thinsp;0.075, ~\u0026thinsp;13.43 Mya) and an ancestral WGD event (Ks\u0026thinsp;=\u0026thinsp;0.186, ~\u0026thinsp;30.58 Mya), suggesting that the WGD event occurred in the common ancestor of these subgenomes prior to their evolutionary separation (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eb).\u003c/p\u003e\n\u003cp\u003eDot plots (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ec) illustrated paralogs inherited from whole-genome triplication (WGT)-\u0026gamma; in the grape genome, revealing a 3\u0026ndash;3 diagonal relationship. Similarly, 4\u0026ndash;4 diagonal relationships were observed within subgenomes A and B (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003ec), as well as between the two subgenomes (Supplementary Fig.\u0026nbsp;7), indicating the retention of two successive WGD events in each subgenome. Additionally, analysis of gene duplication types using MCScanX revealed that WGD/segmental duplications are the most prevalent in both subgenomes (Supplementary Fig.\u0026nbsp;8), suggesting that the retention of whole-genome duplicated genes plays a crucial role in gene expansion in \u003cem\u003eN. nimmoniana\u003c/em\u003e.\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eCharacterization of the enzymes involved in CPT biosynthesis in\u003c/strong\u003e \u003cstrong\u003eN. nimmoniana\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eNnCYP72A1 (accession no. AQW38832.1), a secologanin synthase, catalyzes the conversion of loganin to secologanin and is, to date, the only enzyme characterized in camptothecin (CPT) biosynthesis in \u003cem\u003eN. nimmoniana\u003c/em\u003e. In this study, we identified two candidate secologanin synthases, a homoeologous gene pair (NnSLS1: Nni_ChrA05G28720.1 and NnSLS2: Nni_ChrB05G22310.1), which shares high sequence identity with NnCYP72A1 (98.66% and 96.74%, respectively) and has the highest expression level in roots (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea). To elucidate their catalytic roles, recombinant NnSLS1 and NnSLS2 were expressed in the \u003cem\u003eSaccharomyces cerevisiae\u003c/em\u003e strain WAT11 (Supplementary Fig.\u0026nbsp;9). After microsomes were prepared, \u003cem\u003ein vitro\u003c/em\u003e enzymatic assays were performed using loganin as the substrate. Both NnSLS1 and NnSLS2 successfully catalyzed the conversion of loganin to secologanin, as confirmed by UPLC‒MS/MS analysis, which matched the retention time and MS/MS spectra of an authentic standard (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eb, c). Using loganic acid as a substrate resulted in no conversion to secologanic acid (Supplementary Fig.\u0026nbsp;10), which is in agreement with the established role of most SLS enzymes. An exception is found in \u003cem\u003eC. acuminata\u003c/em\u003e, where loganic acid is converted to secologanic acid by CaSLASs (Supplementary Fig.\u0026nbsp;1).\u003c/p\u003e\n\u003cp\u003eGiven that NnSLS1 and NnSLS2 utilize loganin as a substrate, we inferred that loganic acid is likely converted to loganin by an LAMT positioned immediately upstream of SLS in the CPT biosynthetic pathway in \u003cem\u003eN. nimmoniana\u003c/em\u003e. LAMTs belong to the plant SABATH (salicylic acid/benzoic acid/theobromine) methyltransferase family, which catalyzes the SAM (S-adenosyl-methionine)-dependent methylation of a variety of substrates in plants, such as plant hormones and other small molecules. We identified 20 full-length SABATH methyltransferase genes across both subgenomes, and phylogenetic analysis clustered NnLAMT1 (Nni_ChrA07G12930.1) and NnLAMT2 (Nni_ChrB07G25210.1), a homoeologous gene pair, with the previously characterized OpLAMT \u003csup\u003e3\u003c/sup\u003e and CrLAMT\u003csup\u003e19\u003c/sup\u003e (Supplementary Fig.\u0026nbsp;11). Expression profiling across eight tissues revealed high coexpression of NnLAMT1/2 with NnSLS1/2, with peak expression in roots (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ed and Supplementary Fig.\u0026nbsp;12). Recombinant NnLAMT1/2, expressed in \u003cem\u003eEscherichia coli\u003c/em\u003e, demonstrated the methylation of loganic acid to loganin, which was consistent with CrLAMT activity (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ee, f). Additionally, subcellular localization analysis of NnLAMT2-GFP in \u003cem\u003eNicotiana benthamiana\u003c/em\u003e revealed nucleocytosolic localization, which aligns with CrLAMT findings\u003csup\u003e20\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eg).\u003c/p\u003e\n\u003cp\u003eSTR is a key enzyme involved in the biosynthesis of monoterpenoid indole alkaloids (MIAs), which catalyze the condensation of secologanin with tryptamine to form strictosidine. We identified 30 full-length candidate STR genes from the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome. Coexpression analysis revealed that six putative STR genes (three homologous gene pairs) were strongly coexpressed with the characterized NnLAMTs and NnSLSs, with the highest expression observed in the roots of \u003cem\u003eN. nimmoniana\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ea and Supplementary Fig.\u0026nbsp;13). Five putative NnSTRs were successfully cloned, and transient expression experiments in \u003cem\u003eN. benthamiana\u003c/em\u003e leaves indicated that NnSTR1 and NnSTR3 could catalyze the coupling of secologanin and tryptamine to generate strictosidine (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ea, b, c, compared with CrSTR). Additionally, using secologanic acid and tryptamine as substrates led to the production of strictosidinic acid (Supplementary Fig.\u0026nbsp;14), a substrate promiscuity also observed in OpSTR\u003csup\u003e3\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eCytochrome P450 enzymes (CYP450s) play a critical role in the epoxidation of strictosamide. Recently, CaCYP71BE206 and OpCYP716E111 were identified as strictosamide epoxidases (SEOs) involved in CPT biosynthesis. Genome mining of \u003cem\u003eN. nimmoniana\u003c/em\u003e revealed 344 full-length CYP450s across 48 families, with sixteen exhibiting strong coexpression with characterized NnLAMTs, NnSLSs, and NnSTRs (Supplementary Fig.\u0026nbsp;15). Fourteen candidate NnCYPs were successfully cloned, and functional analysis using a transient expression system identified three CYP450s\u0026mdash;Nni_ChrB12G01000.1 (NnCYP76B), Nni_ChrA12G09540.1 (NnCYP76B), and Nni_ChrB05G34420.1 (NnCYP72A)\u0026mdash;as SEOs, which were named NnSEO1-3 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003ed, e, f). Among these, NnSEO1 and NnSEO2 form a pair of homoeologous genes.\u003c/p\u003e\n\u003cp\u003eNotably, for all the characterized enzymes in this study, at least one functional copy was retained in each subgenome following allotetraploidization, with all copies exhibiting the highest expression levels in the roots of \u003cem\u003eN. nimmoniana\u003c/em\u003e. In addition, candidate genes involved in the prestrictosidine pathway and CPT modification were identified through homology searches and further refined based on their coexpression relationships with the characterized genes (Supplementary Table\u0026nbsp;10 and Supplementary Fig.\u0026nbsp;16). Among these candidates, all enzymes except geraniol synthase and 10-hydroxygeraniol oxidoreductase were found to contain at least one homoeologous pair. These findings indicate that allotetraploidization may significantly increase CPT biosynthesis in \u003cem\u003eN. nimmoniana\u003c/em\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eIndependent evolution of CPT biosynthesis\u003c/h3\u003e\n\u003cp\u003eStrictosidine synthase has been previously identified in distantly related plants, including \u003cem\u003eC. roseus\u003c/em\u003e and \u003cem\u003eRauvolfia serpentina\u003c/em\u003e (Apocynaceae), \u003cem\u003eGelsemium sempervirens\u003c/em\u003e (Gelsemiaceae), \u003cem\u003eO. pumila\u003c/em\u003e (Rubiaceae), and \u003cem\u003eC. acuminata\u003c/em\u003e (Nyssaceae). While these enzymes belong to the same family, they exhibit low amino acid sequence identity across species from different families (Supplementary Table\u0026nbsp;11). Notably, the two NnSTRs characterized in this study share only 38.6% amino acid sequence similarity. Genome-wide phylogenetic analysis revealed that both NnSTRs are grouped into two adjacent clades: NnSTR1 clusters with CaSTRs, whereas NnSTR3 groups with STRs from Gentianales species (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ea and Supplementary Fig.\u0026nbsp;17), and both clades include species that have been clearly confirmed to lack strictosidine derivatives. These two clades likely originated from an ancient gene duplication event. Collinearity analysis suggested that this duplication event occurred as a tandem duplication (named AncSTR-like1 and AncSTR-like2) in the last common ancestor of asterid plants (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eb and Supplementary Fig.\u0026nbsp;18). Among the modern plants examined, only \u003cem\u003eN. nimmoniana\u003c/em\u003e and \u003cem\u003eO. pumila\u003c/em\u003e retain the tandem duplication state of the two ancestral copies. Specifically, NnSTR1 and CaSTRs were derived from AncSTR-like1, whereas NnSTR3 and STRs from Gentianales originated from AncSTR-like2 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eb). Positive selection analysis indicated that the average rate of nonsynonymous substitutions significantly increased in both the AncSTR-like1 (\u0026omega;(AncSTR-like1)\u0026thinsp;=\u0026thinsp;1.93165) and AncSTR-like2 (\u0026omega;(AncSTR-like2)\u0026thinsp;=\u0026thinsp;6.58244) branches compared with the branch that predated the duplication event (\u0026omega;₀ = 0.20768), indicating that both branches underwent strong positive Darwinian selection immediately following the duplication event (Supplementary Fig.\u0026nbsp;19). These findings imply that STR function was promoted after the divergence of AncSTR-like1 and AncSTR-like2, which underwent at least two independent evolutionary events.\u003c/p\u003e\n\u003cp\u003eTo investigate the catalytic mechanisms of NnSTRs, 3D protein structure modeling and molecular docking analyses were conducted. Both NnSTRs displayed a conserved six-bladed, four-stranded \u0026beta;-propeller fold structure, which is consistent with the crystal structures previously reported for RsSTR (PDB code: 2FP8), CrSTR (PDB code: 6ZEA), and OpSTR (PDB code: 6S5M) (Supplementary Fig.\u0026nbsp;20). This structural conservation underscores the evolutionary stability of the STR fold across asterids. Molecular docking analysis suggested that distinct STRs utilize different catalytic residues to initiate the Pictet\u0026ndash;Spengler reaction. For example, D140 in NnSTR1 and Y121 in NnSTR3 were predicted to donate a proton to the amine group of tryptamine, thereby facilitating its reaction with the aldehyde group of secologanin to form strictosidine (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ec, d and Supplementary Fig.\u0026nbsp;21). In contrast, D296 was identified as the likely catalytic residue in CacSTR1 (Supplementary Fig.\u0026nbsp;22). Site-directed mutagenesis of NnSTR1 (D140A) and NnSTR3 (Y121K and E164A) resulted in mutants incapable of producing strictosidine when transiently expressed in \u003cem\u003eN. benthamiana\u003c/em\u003e (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003ee, f), confirming the critical roles of these residues in STR activity. Notably, previous studies reported that glutamic acid serves as the catalytic residue in RsSTR\u003csup\u003e21\u003c/sup\u003e and OpSTR\u003csup\u003e22\u003c/sup\u003e, suggesting that STRs from Gentianales species share a common ancestral catalytic mechanism. These findings indicate that while STRs within asterids have independently evolved distinct active sites to catalyze the condensation of secologanin and tryptamine, they rely on a conserved chemical logic to facilitate this reaction.\u003c/p\u003e\n\u003cp\u003eSEO catalyzes the conversion of strictosamide into strictosamide epoxide, thereby shunting the poststrictosidine biosynthetic flux toward the CPT pathway. SEO activity has been characterized in four CYP450 families across three distantly related plants: the CYP71 family in \u003cem\u003eC. acuminata\u003c/em\u003e (CYP71BE206), the CYP716 family in \u003cem\u003eO. pumila\u003c/em\u003e (CYP716E111), and the CYP76 and CYP72 families in \u003cem\u003eN. nimmoniana\u003c/em\u003e (this study). This finding demonstrates the recruitment of distinct CYP families to catalyze identical biosynthetic steps. The phylogenetic analysis of NnSEOs, OpCYP716E111, CaCYP71BE206, and other well-characterized enzymes (Supplementary Table\u0026nbsp;12) from the same subfamilies revealed that each SEO is more closely related to other functional enzymes than to one another (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e). These findings strongly suggest that SEO activity evolved independently in \u003cem\u003eN. nimmoniana\u003c/em\u003e (Icacinaceae), \u003cem\u003eC. acuminata\u003c/em\u003e (Nyssaceae), and \u003cem\u003eO. pumila\u003c/em\u003e (Rubiaceae). Additionally, NnSEO3, Nn7DLH-like proteins, and NnSLSs belonging to the CYP72A subfamily were grouped into three distinct subclades. We propose that the emergence of 7DLH, SLS, and SEO activities likely resulted from an ancient gene duplication event within NnCYP72A. On the basis of phylogenetic analyses and synonymous substitution rates (Ks values of 1.61\u0026ndash;1.65), this duplication event is estimated to have occurred approximately 124\u0026ndash;127 Mya, predating the evolutionary divergence of asterid species.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eAs one of the most widely used anticancer drugs, CPT was first isolated from \u003cem\u003eC. acuminata\u003c/em\u003e. To date, CPT has been identified in 43 angiosperm species, with over 93% of these confined to three orders within the asterid clade: Icacinales (49%), Gentianales (37%), and Cornales (7%)\u003csup\u003e14\u003c/sup\u003e. Among these, \u003cem\u003eN. nimmoniana\u003c/em\u003e is the most potent natural producer of CPT. Both \u003cem\u003eC. acuminata\u003c/em\u003e and \u003cem\u003eN. nimmoniana\u003c/em\u003e serve as major plant sources for the production of CPT and its derivatives. In this study, we sequenced and assembled a chromosome-level genome of \u003cem\u003eN. nimmoniana\u003c/em\u003e, representing the first genome sequence from the order Icacinales. This genomic resource provides valuable insights into the biosynthesis of CPT and its evolutionary history.\u003c/p\u003e \u003cp\u003eThe enzymatic activities of LAMT, SLS and STR were identified in \u003cem\u003eN. nimmoniana\u003c/em\u003e, suggesting that this species shares a similar prestrictosidine pathway with major MIA-producing plants. Although both NnSTRs can catalyze the conversion of secologanic acid and tryptamine into strictosidinic acid, this reaction is likely a side activity, comparable to that observed in \u003cem\u003eOpSTR\u003c/em\u003e. The presence of LAMT and SLS enzymes in \u003cem\u003eN. nimmoniana\u003c/em\u003e suggests that the primary metabolic flux is directed toward strictosidine production, as reported in \u003cem\u003eO. pumila\u003c/em\u003e. In \u003cem\u003eC. acuminata\u003c/em\u003e, however, loganic acid is converted to secologanic acid by SLAS, which is subsequently coupled with tryptamine by strictosidine acid synthase to form strictosidine acid. This divergence may be exceptional, representing adaptive evolution following the loss of LAMT activity in \u003cem\u003eC. acuminata\u003c/em\u003e. Interestingly, three enzymes, two belonging to the CYP76B subfamily and one to the CYP72A subfamily, were identified as SEOs in \u003cem\u003eN. nimmoniana\u003c/em\u003e. A similar phenomenon has been reported in the spiroketal steroid biosynthesis of \u003cem\u003eParis polyphylla\u003c/em\u003e, where the conversion of an unstable intermediate to diosgenin is mediated by three enzymes from the CYP94D and CYP72A subfamilies\u003csup\u003e23\u003c/sup\u003e. The presence of multiple enzymes that catalyze the same reaction within a species may facilitate fine-tuned regulation of the reaction intensity in a tissue- or cell-specific manner, as evidenced by their distinct expression patterns.\u003c/p\u003e \u003cp\u003e \u003cem\u003eN. nimmoniana\u003c/em\u003e, as an allotetraploid, exhibits subtle subgenome dominance. However, our analysis revealed that all known genes involved in CPT biosynthesis, with the exception of \u003cem\u003eGES\u003c/em\u003e and \u003cem\u003e10HGO\u003c/em\u003e, retain homoeologous gene pairs. Moreover, these homoeologous gene pairs demonstrate high levels of expression. These observations suggest that the increased gene dosage resulting from allotetraploidization may play a pivotal role in the elevated CPT content observed in this species. In addition, these findings also indicate that \u003cem\u003eN. nimmoniana\u003c/em\u003e represents a relatively recent allotetraploid lineage that is currently in the process of diploidization and the progressive establishment of more pronounced subgenome dominance.\u003c/p\u003e \u003cp\u003eCPT and all MIAs are synthesized via the iridoid biosynthetic pathway. Iridoids are widely distributed across the asterid clade, where their production is regarded as a synapomorphic trait. However, CPT and MIAs are predominantly restricted to species within the orders Icacinales, Gentianales, and Cornales. To investigate the origin and evolution of enzymes specific to the CPT branch pathway, we conducted phylogenetic and syntenic analyses. Our results revealed that STRs and SEOs were independently recruited in \u003cem\u003eN. nimmoniana\u003c/em\u003e, \u003cem\u003eC. acuminata\u003c/em\u003e, and \u003cem\u003eO. pumila\u003c/em\u003e. Additionally, 3D protein structure reconstruction and molecular docking analyses demonstrated that the STRs in these species evolved distinct active sites independently. These findings suggest that CPT biosynthesis has evolved independently multiple times within the asterid clade. Furthermore, given the widespread production of MIAs among species in Icacinales, Gentianales, and Cornales, it is plausible that STR, as the first enzyme involved in MIA biosynthesis, is monophyletic within each of these orders.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003ePlant material and sequencing\u003c/h2\u003e \u003cp\u003eThe material for genome sequencing was obtained from Fujian Institute of Subtropical Botany (24.52318\u0026deg;N, 118.11537\u0026deg;E) (Supplementary Fig.\u0026nbsp;23), which was introduced from Lanyu County, Taiwan Province. Young leaves of an individual plant were used for DNA extraction via the Qiagen DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) following the manufacturer\u0026rsquo;s instructions. Two paired-end (PE150) libraries with average insert sizes of 350 bp were constructed and sequenced on the Illumina NovaSeq 6000 platform according to the manufacturer\u0026rsquo;s instructions (Illumina Inc., USA). Seven HiFi libraries were constructed and sequenced on the Pacific Biosciences Sequel II platform. Four Hi-C libraries were prepared and sequenced on the NovaSeq 6000 platform. Eight tissues from the same plant used for genome sequencing, namely, the root, stem, leaf, petiole, flower bud, flower, seed and seed stalk, with three replicates, were prepared for RNAseq.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eKaryotyping and genome size estimation\u003c/h2\u003e \u003cp\u003eFor karyotyping, mitotic chromosome slides were prepared as previously described with some modifications\u003csup\u003e24\u003c/sup\u003e. The fresh root tips of \u003cem\u003eN. nimmoniana\u003c/em\u003e seedlings were pretreated with aqueous solution saturated with p-dichlorobenzene at room temperature (RT) for 2.5 hours, fixed in 3:1 (v/v) ethanol:acetic acid for 2 hours and stored in 70% ethanol at 4\u0026deg;C until further use. After being washed with distilled water, the root tips were hydrolyzed with 1 mol/L hydrochloric acid at 45\u0026deg;C for 45 minutes. After washing, the digested root tips were stained with Carbol-Fuchsin solution. The traditional squash was performed, and the chromosome number was observed via a Nikon Eclipse E600FN microscope with a Spot RTKE (Diagnostic Instruments) CCD camera. Genome size estimation of \u003cem\u003eN. nimmoniana\u003c/em\u003e by flow cytometry followed the method described by Zhixiang Liu\u003csup\u003e25\u003c/sup\u003e et al. with Pepper (\u003cem\u003eCapsicum annuum\u003c/em\u003e) used as a reference. The relative DNA content of the isolated nuclei was analyzed via a BD FACSCalibur system (Becton Dickinson, New Jersey, USA). The data were acquired and processed by BD FACS DIVA software (v 7.0). The 1N pepper genome was 3,349 Mb\u003csup\u003e26\u003c/sup\u003e; thus, the \u003cem\u003eN. nimmoniana\u003c/em\u003e genome size was ~\u0026thinsp;4.998 Gb. The Illumina short PE150 reads (30X coverage) were subjected to genome size and heterozygosity estimation. The 17-mer frequency of short reads was calculated using JellyFish\u003csup\u003e27\u003c/sup\u003e, and the genome size and heterozygosity were subsequently calculated on the basis of the k-mer frequency results.\u003c/p\u003e \u003cp\u003e \u003cb\u003eDe novo\u003c/b\u003e \u003cb\u003egenome assembly and assessment\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe HiFi reads were generated using CCS and assembled by hifiasm (v0.16.0)\u003csup\u003e16\u003c/sup\u003e with default parameters. The Hi-C reads were processed and aligned to Hi-Fi contigs by Juicer and then clustered into chromosomes via the 3D-DNA package (v180419)\u003csup\u003e28\u003c/sup\u003e with default parameters. We performed manual correction and validation using Juicebox (v1.11.08) to obtain the final chromosome-level nuclear genome sequences.\u003c/p\u003e \u003cp\u003eThe quality of the final genome sequences was assessed based on multiple datasets. The Illumina PE reads were mapped to the assembly using BWA-MEM (v0.7.16)\u003csup\u003e29\u003c/sup\u003e with default parameters. The RNA-seq data were mapped to the genome using HISAT2 (v2.1.0)\u003csup\u003e30\u003c/sup\u003e. The completeness of the genome assemblies was evaluated using BUSCO (v4.1.2)\u003csup\u003e31\u003c/sup\u003e with the Embryophyta odb10 database.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eRepeat and gene annotations\u003c/h3\u003e\n\u003cp\u003eRepeat sequences were identified using a three-step method. First, the \u003cem\u003ede novo\u003c/em\u003e repeat sequence library was established using LTR_FINDER\u003csup\u003e32\u003c/sup\u003e and RepeatModeler\u003csup\u003e33\u003c/sup\u003e. The resulting library was combined with the RepBase database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.girinst.org/repbase/\u003c/span\u003e\u003cspan address=\"https://www.girinst.org/repbase/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and then used to mask the genome with RepeatMasker (version 4.1.0)\u003csup\u003e34\u003c/sup\u003e. Second, RepeatProteinMask was used to search against the TE protein database. In addition, TRF (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://tandem.bu.edu/trf/trf.html\u003c/span\u003e\u003cspan address=\"https://tandem.bu.edu/trf/trf.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was used for tandem repeat identification. The results were combined, and duplications were removed to form the final set of repeat sequence annotations.\u003c/p\u003e \u003cp\u003eFor gene model prediction, three \u003cem\u003eab initio\u003c/em\u003e gene prediction tools were used on the basis of the statistical characteristics of the genomic sequence: AUGUSTUS (v3.1, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/Gaius-Augustus/Augustus\u003c/span\u003e\u003cspan address=\"https://github.com/Gaius-Augustus/Augustus\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), GlimmerHMM (v1.2, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://ccb.jhu.edu/software/glimmerhmm\u003c/span\u003e\u003cspan address=\"https://ccb.jhu.edu/software/glimmerhmm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and SNAP (v2006, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/KorfLab/SNAP\u003c/span\u003e\u003cspan address=\"https://github.com/KorfLab/SNAP\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Homology-based prediction was performed using BLAST (v2.13.0) and GeneWise (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ebi.ac.uk/Tools/psa/genewise/\u003c/span\u003e\u003cspan address=\"https://www.ebi.ac.uk/Tools/psa/genewise/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The PASA pipeline (v2.0.2, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/PASApipeline/PASApipeline\u003c/span\u003e\u003cspan address=\"https://github.com/PASApipeline/PASApipeline\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was used to generate the RNA-seq evidence prediction. Finally, all the gene structures predicted by these three methods were integrated into a nonredundant gene set using EVidenceModeler (EVM, v1.1.1, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/EVidenceModeler/EVidenceModeler\u003c/span\u003e\u003cspan address=\"https://github.com/EVidenceModeler/EVidenceModeler\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The resulting gene models were functionally annotated by integrating the annotation information from the NCBI nonredundant protein database (NR, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/protein\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/protein\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), SwissProt (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.uniprot.org/\u003c/span\u003e\u003cspan address=\"https://www.uniprot.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), KEGG (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.genome.jp/kegg/\u003c/span\u003e\u003cspan address=\"https://www.genome.jp/kegg/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and InterPro (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ebi.ac.uk/interpro/\u003c/span\u003e\u003cspan address=\"https://www.ebi.ac.uk/interpro/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). To annotate the noncoding RNAs, tRNA genes were identified by tRNAscan-SE (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://lowelab.ucsc.edu/tRNAscan-SE/\u003c/span\u003e\u003cspan address=\"https://lowelab.ucsc.edu/tRNAscan-SE/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), whereas ribosomal RNAs (rRNAs) were predicted via BLASTN searches against rRNA sequences with an E value cutoff of 1 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;10\u003c/sup\u003e. Using the covariance model of the Rfam database, INFERNAL (v1.1.4, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://eddylab.org/infernal/\u003c/span\u003e\u003cspan address=\"http://eddylab.org/infernal/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was applied to predict miRNAs and snRNAs in the genome.\u003c/p\u003e\n\u003ch3\u003eAssigning chromosome assemblies to subgenomes and comparative subgenomic analysis\u003c/h3\u003e\n\u003cp\u003eThe subgenome-phasing algorithm, SubPhaser (v1.2)\u003csup\u003e17\u003c/sup\u003e, was employed to assign homoeologous chromosomes into two subgenomes. Additionally, subgenomes in \u003cem\u003eN. nimmoniana\u003c/em\u003e were distinguished on the basis of TE characteristics between homoeologous chromosomes. A matrix summarizing the copy number of each TE family was built on the 28 chromosomes of \u003cem\u003eN. nimmoniana\u003c/em\u003e. This matrix was used for the PCA in the R program. The two subgenomes were subjected to pairwise comparisons using Minimap2 (v2.28)\u003csup\u003e35\u003c/sup\u003e. Syri (v1.7.0)\u003csup\u003e36\u003c/sup\u003e was used to identify synteny and structural rearrangements.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eHomoeolog expression dominance analysis\u003c/h2\u003e \u003cp\u003eThe RNA-Seq reads were mapped to the genome with HISAT2 (v2.1.0), and the gene expression levels for each RNA-seq sample were estimated using StringTie (v2.2.1) by calculated fragments per kilobase of transcript per million mapped reads (FPKM) values. The homoeolog pairs between two subgenomes were identified via MCScan (Python version) (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/tanghaibao/jcvi/wiki\u003c/span\u003e\u003cspan address=\"https://github.com/tanghaibao/jcvi/wiki\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) in \u0026lsquo;-full\u0026rsquo; mode to assume 1-to-1 quota synteny blocks. The genome-wide transcriptional levels of subgenomes A and B were reflected in the gene expression levels of homoeologous gene pairs. The log2-fold change in the FPKM values between homoeologous gene pairs was calculated to measure the expression bias. This assessment identified gene pairs exhibiting differential expression exceeding a twofold change threshold as dominant gene pairs. Among these pairs, the genes with relatively high expression levels were designated dominant genes, whereas their counterparts with low expression were categorized as submissive genes. Conversely, the remaining syntenic gene pairs that did not exhibit dominance were classified as neutral genes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eGenome evolution\u003c/h2\u003e \u003cp\u003eGene clusters associated with 13 other plant species were identified using OrthoFinder (v2.5.4)\u003csup\u003e37\u003c/sup\u003e. Protein and coding sequences from 493 single-copy orthologous gene clusters were used to construct phylogenetic relationships and estimate divergence times. Alignments from MAFFT (v7.471)\u003csup\u003e38\u003c/sup\u003e were converted to coding sequences. IQ-TREE (v2.0.3)\u003csup\u003e39\u003c/sup\u003e was used to construct the phylogenetic tree. The Bayesian relaxed molecular clock (BRMC) approach was used to estimate the species divergence time using the MCMCTree program, which is in the PAML package (v4.9j)\u003csup\u003e40\u003c/sup\u003e. Published species divergence times downloaded from the TimeTree database\u003csup\u003e41\u003c/sup\u003e were used to calibrate the divergence times. The synonymous mutation rate (\u003cem\u003eks\u003c/em\u003e) was calculated using the HKY mode, and the WGD time point in each subgenome of \u003cem\u003eN. nimmoniana\u003c/em\u003e was estimated according to \u003cem\u003eks\u003c/em\u003e via the method described by Vanneste et al.\u003csup\u003e42\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eScreening of candidate genes involved in CPT biosynthesis\u003c/h2\u003e \u003cp\u003eSequences containing PF03492 (SAM-dependent carboxyl methyltransferase), PF00067 (cytochrome P450), or PF03088 (strictosidine synthase) domains were scanned via the hmmsearch program from the HMMER package (v3.3). The hits were then filtered for full-length sequences using a BLAST search against OMT, CYP450, or STR protein sequences from Arabidopsis and sequences from these three families downloaded from SwissProt. The remaining sequences were structurally corrected according to the \u003cem\u003eN. nimmonana\u003c/em\u003e transcriptome data with Apollo. Clustering and visualization of gene expression patterns were performed using TBtools. Genes whose expression patterns were consistent with those of the characterized genes or whose expression levels were high in roots were considered candidate genes involved in CPT biosynthesis and were selected for subsequent experiments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003ePhylogenetic analysis of STR and SEO\u003c/h2\u003e \u003cp\u003eTo infer phylogenetic trees, we analyzed sequences of STR enzymes and STR family sequences from 26 angiosperm species, SEO enzymes and other well-characterized enzymes, including CYP71D, CYP72A, CYP716E, and CYP76B. Sequence alignments were first generated using MAFFT v7.453 and refined through trimAl (V1.5.rev0). The best-fit substitution model for multiple sequence alignment was selected using IQ-TREE (v2.0.3) with \u0026lsquo;-mf\u0026rsquo; parameters. The maximum-likelihood gene trees were reconstructed using RAxML-NG (V1.2.2) and visualized using the R package ggtree.\u003c/p\u003e \u003cp\u003eAdaptive evolution analysis of functional STRs after gene duplication was performed using the PAML program codeml according to the method described by Bielawski\u003csup\u003e43\u003c/sup\u003e. The null model (the one-ratio model) assumes the same ω ratio (nonsynonymous/synonymous nucleotide substitution rate ratio, dN/dS) for all branches. Nested models are constructed based on the assumption that selective constraints change following gene duplication. Model R2 (two-ratio model) assumes two independent ω ratios: one ratio for all branches predating a duplication event and a second for all branches postdating the tandem duplication event of the STR ancestor. Model R3 assumes three independent ω ratios: one for all branches predating the duplication event, a second for the branches immediately following the duplication event, and a third for all subsequent branches. Model free ratios assume an independent ω ratio for each branch. A likelihood ratio test (LRT) of the one-ratio model with Model R2/R3/free ratios was used to examine the difference between average selective constraints before and after a tandem duplication event of the STR ancestor.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eProtein structure prediction and molecular docking\u003c/h2\u003e \u003cp\u003eThe putative protein structures of functional STRs were predicted by AlphaFold3. The chemical structures of tryptamine and secologanin were downloaded from PubChem (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pubchem.ncbi.nlm.nih.gov/\u003c/span\u003e\u003cspan address=\"https://pubchem.ncbi.nlm.nih.gov/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and saved in mol2 format using Open Bable GUI for molecular docking as ligands. Molecular docking of NnSTR1, NnSTR3 and CacSTR with tryptamine and secologanin was performed using AutoDock Vina 1.2.5\u003csup\u003e44, 45\u003c/sup\u003e. PyMOL 3.0 was used for viewing the molecular interactions and image processing. The 2D images of the docking results are presented by a ligand interaction diagram model in Schr\u0026ouml;dinger suite 2023.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eGene cloning\u003c/h2\u003e \u003cp\u003eTotal RNA was extracted from \u003cem\u003eN. nimmoniana\u003c/em\u003e roots, \u003cem\u003eC. roseus\u003c/em\u003e leaves, and \u003cem\u003eO. pumila\u003c/em\u003e leaves using an RNAprep Pure Plant Plus Kit (TIANGEN, catalog no. DP411). First-strand cDNA was synthesized from total RNA using the PrimeScript\u0026trade; IV 1st strand cDNA Synthesis Mix (TaKaRa, catalog no. 6215A) according to the manufacturer\u0026rsquo;s protocol. The full-length CDSs of candidate \u003cem\u003eNnSLS\u003c/em\u003es, \u003cem\u003eNnSTR\u003c/em\u003es, and \u003cem\u003eNnSEO\u003c/em\u003es and \u003cem\u003eCrLAMT\u003c/em\u003e, \u003cem\u003eCrSTR\u003c/em\u003e, \u003cem\u003eOpCYP716E111\u003c/em\u003e, were amplified using PrimeSTAR\u0026reg; GXL DNA Polymerase (TaKaRa, catalog no. 6215A), and the primers used are listed in Supplementary Table\u0026nbsp;13. \u003cem\u003eNnLAMT1\u003c/em\u003e and \u003cem\u003eNnLAMT2\u003c/em\u003e were synthesized and inserted into the Nde I/Xho I sites of pET-28a vector (GENEWIZ, Beijing, China).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eFunctional characterization of NnLAMT\u003c/h2\u003e \u003cp\u003eThe plasmids pET28a-NnLAMT1 and pET28a-LAMT2 were subsequently transformed into \u003cem\u003eE. coli\u003c/em\u003e BL21 (DE3). Single colonies for each construct were inoculated in 10 mL of liquid Luria Bertani (LB) medium containing 50 mg/L kanamycin, followed by cultivation at 37\u0026deg;C with shaking at 220 rpm for 12 hours. The cultures were then transferred into 1 L of fresh liquid LB medium and grown until the optical density at 600 nm (OD\u003csub\u003e600\u003c/sub\u003e) reached 0.6. Protein expression was induced by adding isopropyl β-d-thiogalactoside (IPTG) to a final concentration of 0.02 mM, followed by incubation at 16\u0026deg;C with shaking at 110 rpm for 18 hours. The recombinant His-fusion proteins were purified using Ni-agarose resin (Mei5bio, catalog no. MF205-01) according to the manufacturer\u0026rsquo;s instructions. The target proteins were desalted using a PD-10 column. The protein concentration was determined via a BCA protein quantification kit (Vazyme, catalog no. E112-01). A total of 100 \u0026micro;L of the reaction mixture containing 20 \u0026micro;g of protein, 2 mM loganic acid and 1 mM S-adenosyl-L-methionine (SAM) was incubated at 30\u0026deg;C for 180 min. The reaction was stopped by adding 100 \u0026micro;L of methanol and vortexing for 5 min. After centrifugation and filtration, 1 \u0026micro;L of the enzyme reaction mixture was detected via LC‒MS/MS (ACQUITY UPLC I-Class System with Xevo G2 Q-TOF (Waters, Milford, MA, USA)). A UPLC BEH C18 column (1.7 \u0026micro;m, 2.1\u0026times;100 mm, Waters) was used. The mobile phase consisted of 0.1% formic acid aqueous solution (A) and acetonitrile (B). The gradient was as follows: 5% B solution, 0\u0026ndash;3 min; 5\u0026ndash;30% B, 3\u0026ndash;12 min; 30\u0026ndash;95% B, 12\u0026ndash;15 min; 95% B, 15\u0026ndash;18 min; 95\u0026ndash;5% B, 18\u0026ndash;21 min; and 5% B, 21\u0026ndash;24 min. The conditions of the mass spectrum detector were as follows: cone hole voltage, 40 V; capillary voltage, 3 kV; desolvent temperature, 250\u0026deg;C; ion source temperature, 100\u0026deg;C; desolvent gas flow rate, 600 L/Hr; cone hole flow rate, 50 L/Hr; collision energy, 6 kV; and positive ion electrospray mode. MassLynx 4.1 software was used for data acquisition and analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eEnzyme assay of NnSLS\u003c/h2\u003e \u003cp\u003eThe candidate \u003cem\u003eNnSLS\u003c/em\u003e and \u003cem\u003eCrSLS\u003c/em\u003e genes were subsequently cloned and inserted into the BamH I/Xho I sites of the yeast expression pYES2-CT vector using a ClonExpress II One Step Cloning Kit (Vazyme, catalog no. C112). The recombinant plasmids were subsequently introduced into the \u003cem\u003eS. cerevisiae\u003c/em\u003e strain WAT11. Positive transformants were screened on solid SD-Ura medium (SD dropout medium without uracil) containing 20 g/L glucose and then cultured in 10 ml of liquid SD-Ura medium until the OD\u003csub\u003e600\u003c/sub\u003e reached 1.5-2. The cells were subsequently centrifuged at 4,000 rpm for 10 min and washed three times with ddH\u003csub\u003e2\u003c/sub\u003eO to remove glucose residue. The cell precipitates were then transferred into YPGal medium containing 2% galactose to induce the expression of the target protein. The microsomes were prepared as previously reported by Yang\u003csup\u003e9\u003c/sup\u003e, the culture was centrifuged, and the cell pellets were washed twice with TEK buffer (50 mM Tris-HCl, 1 mM EDTA, 100 mM KCl, pH 8.0). The cells were resuspended in TES buffer (50 mM Tris-HCl, 1 mM EDTA, 0.6 M sorbitol, and 1 M DTT, pH 8.0) and then disrupted using a high-pressure homogenizer. The homogenate was subsequently centrifuged at 11,000 rpm for 30 min twice. The supernatant was centrifuged at 1,000,000 rpm for 90 min to obtain the microsomal fraction. The resulting microsomal fraction was dissolved in TEG buffer (50 mM Tris-HCl, 1 mM EDTA, 20% glycerol, pH 8.0). For the enzyme activity assay, 100 \u0026micro;L of microsomal suspension was mixed with 500 \u0026micro;M NADPH and 1 mM substrate (loganin or loganic acid). The mixture was incubated at 30\u0026deg;C for 2 hours, and the reaction was terminated with 100 \u0026micro;L of methanol. After centrifugation and filtration, 5 \u0026micro;L of the enzyme reaction mixture was detected with a Thermo Scientific Orbitrap Exploris 120 (Fisher Scientific, Waltham, MA, USA). A Waters ACQUITY UPLC BEH C18 column (1.7 \u0026micro;m, 2.1\u0026times;100 mm) was used. The mobile phase consisted of 0.1% formic acid aqueous solution (A) and acetonitrile (B), with a flow rate of 300 \u0026micro;L/min. The gradient conditions were as follows: 0\u0026ndash;3 min, 5% B; 3\u0026ndash;12 min, 5\u0026ndash;30% B; 12\u0026ndash;15 min, 30\u0026ndash;95% B; 15\u0026ndash;18 min, 95% B; 18\u0026ndash;21 min, 95\u0026ndash;5% B; and 21\u0026ndash;24 min, 5% B. The MS parameters in ESI-positive mode were as follows: sheath gas, 50 psi; aux gas, 10 psi; sweep gas, 0 psi; ion transfer tube temperature, 320\u0026deg;C; vaporizer temperature, 320\u0026deg;C; and positive ion, 3.5 kV. Qualitative analysis was performed using the XCalibur\u0026trade; (v. 4.4.16.14) software.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFunctional characterization of NnSTR and NnSEO in\u003c/b\u003e \u003cb\u003eNicotiana benthamiana\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe candidate NnSTR and NnSEO genes were ligated into the pEAQ-HT-DEST1 plasmid using the Gateway cloning technology. The recombinant plasmids were subsequently transformed into \u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003e (GV3101). Positive transformants were screened on solid LB media supplemented with 50 \u0026micro;g/mL kanamycin and 25 \u0026micro;g/mL gentamycin and then cultured in liquid LB media supplemented with the same antibiotics until the OD\u003csub\u003e600\u003c/sub\u003e reached 0.9. The cells were subsequently centrifuged at 3,000 rpm at 4\u0026deg;C for 10 min and resuspended in MMA buffer (10 mM MES, 30 mM glucose, 10 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 100 \u0026micro;M acetosyringone) with OD\u003csub\u003e600\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.6. After incubation at room temperature for 2 h, \u003cem\u003eA. tumefaciens\u003c/em\u003e suspensions were infiltrated into \u003cem\u003eN. benthamiana\u003c/em\u003e leaves using a syringe. After 3 days, 0.3 mM substrates were infiltrated into previously generated \u003cem\u003eAgrobacterium-\u003c/em\u003einfiltrated leaves for an additional 2 days. The infiltrated \u003cem\u003eN. benthamiana\u003c/em\u003e leaves were collected for metabolite analysis via LC‒MS.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eSubcellular localization\u003c/h2\u003e \u003cp\u003eThe NnLAMT2 gene was cloned and inserted into the SpeI site of the pCAMBIA1302 vector. The recombinant plasmids were transiently expressed in \u003cem\u003eN. benthamiana\u003c/em\u003e leaves. \u003cem\u003eN. benthamiana\u003c/em\u003e leaf protoplasts were isolated as described by Vivien Rolland\u003csup\u003e46\u003c/sup\u003e. Briefly, the cut leaf strips (0.5\u0026ndash;1 mm) were immersed in an enzyme mixture (1.5% cellulase R-10, 0.3% macerozyme R-10, 0.4 M mannitol, 0.2 M MES, 0.1% BSA and 10 mM CaCl\u003csub\u003e2\u003c/sub\u003e, 0.2 M KCl pH 5.7) at 25\u0026deg;C and 50 rpm for 3 hours. After passing through 70-\u0026micro;m strainers, the protoplasts were centrifuged at \u003cem\u003e100 \u0026times; g\u003c/em\u003e for 5 min and washed once with protoplasting solution without enzymes. The protoplasts were stained with 3 volumes of DAPI staining solution (Biosharp, catalog no. BL105A) for 3 min‒5 min. Fluorescence imaging was performed via a Leica SP8 laser scanning confocal microscope.\u003c/p\u003e \u003c/div\u003e "},{"header":"Declarations","content":"\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eReporting summary\u003c/h2\u003e \u003cp\u003eFurther information on research design is available in the Nature Portfolio Reporting Summary linked to this article.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eData availability\u003c/h2\u003e \u003cp\u003eThe genome used in this study has been deposited in China National Center for Bioinformation (CNCB) under BioProject number PRJCA033896.\u003c/p\u003e \u003c/div\u003e\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor contributions\u003c/h2\u003e \u003cp\u003eY.L., C.S. and S.C. conceived and designed the project. X.S., Z.G., K.D., Y.Z. and L.Q. performed the experiments. X.S., Y.L., S.S., C.Z., G.S., S.G., J.X., W.C. and L.L. analyzed the data. X.S., Z.G. and C.S. wrote the manuscript draft. R.S. and Z.H. provided plant samples. X.S., C.S., B.G., L.Y., J.W., L.S., L.X., W.S., Z.X., X.L., V.C. and S.C. revised the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgments\u003c/h2\u003e \u003cp\u003eThis work was supported by National Key Research and Development Program (NO. 2024YFD2100700) and the CAMS Innovation Fund for Medical Sciences (CIFMS) [grant number 2021-I2M-1-032]. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e Liu YQ, \u003cem\u003eet al.\u003c/em\u003e Perspectives on biologically active camptothecin derivatives. \u003cem\u003eMed. Res. Rev.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 753\u0026ndash;789 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Mingzhang A, \u003cem\u003eet al.\u003c/em\u003e Camptothecin distribution and content in Nothapodytes nimmoniana. \u003cem\u003eNat. Prod. Commun.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 197\u0026ndash;200 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Yang M, \u003cem\u003eet al.\u003c/em\u003e Divergent camptothecin biosynthetic pathway in \u003cem\u003eOphiorrhiza pumila\u003c/em\u003e. \u003cem\u003eBMC Biol.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 122 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Kang M, \u003cem\u003eet al.\u003c/em\u003e A chromosome-level Camptotheca acuminata genome assembly provides insights into the evolutionary origin of camptothecin biosynthesis. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 3531 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Yang L, \u003cem\u003eet al.\u003c/em\u003e A homomeric geranyl diphosphate synthase-encoding gene from \u003cem\u003eCamptotheca acuminata\u003c/em\u003e and its combinatorial optimization for production of geraniol in \u003cem\u003eEscherichia coli\u003c/em\u003e. \u003cem\u003eJ. Ind. Microbiol. Biotechnol.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e, 1431\u0026ndash;1441 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Awadasseid A, \u003cem\u003eet al.\u003c/em\u003e Characterization of Camptotheca acuminata 10-hydroxygeraniol oxidoreductase and iridoid synthase and their application in biological preparation of nepetalactol in Escherichia coli featuring NADP+ - NADPH cofactors recycling. \u003cem\u003eInt. J. Biol. Macromol.\u003c/em\u003e \u003cb\u003e162\u003c/b\u003e, 1076\u0026ndash;1085 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Sadre R, \u003cem\u003eet al.\u003c/em\u003e Metabolite Diversity in Alkaloid Biosynthesis: A Multilane (Diastereomer) Highway for Camptothecin Synthesis in \u003cem\u003eCamptotheca acuminata\u003c/em\u003e. \u003cem\u003ePlant cell\u003c/em\u003e \u003cb\u003e28\u003c/b\u003e, 1926\u0026ndash;1944 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Liu Z, \u003cem\u003eet al.\u003c/em\u003e Catalytic selectivity and evolution of cytochrome P450 enzymes involved in monoterpene indole alkaloids biosynthesis. \u003cem\u003ePhysiol. Plant\u003c/em\u003e \u003cb\u003e176\u003c/b\u003e, e14515 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Yang Y, \u003cem\u003eet al.\u003c/em\u003e Bifunctional Cytochrome P450 Enzymes Involved in Camptothecin Biosynthesis. \u003cem\u003eACS Chem. Biol.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 1091\u0026ndash;1096 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Pu X, \u003cem\u003eet al.\u003c/em\u003e Proteomics-Guided Mining and Characterization of Epoxidase Involved in Camptothecin Biosynthesis from \u003cem\u003eCamptotheca acuminata\u003c/em\u003e. \u003cem\u003eACS Chem. Biol.\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e, 1772\u0026ndash;1785 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Yamazaki Y, \u003cem\u003eet al.\u003c/em\u003e Camptothecin biosynthetic genes in hairy roots of \u003cem\u003eOphiorrhiza pumila\u003c/em\u003e: cloning, characterization and differential expression in tissues and by stress compounds. \u003cem\u003ePlant Cell Physiol.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e, 395\u0026ndash;403 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Zhang T, \u003cem\u003eet al.\u003c/em\u003e Chemoproteomics reveals the epoxidase enzyme for the biosynthesis of camptothecin in \u003cem\u003eOphiorrhiza pumila\u003c/em\u003e. \u003cem\u003eJ. Integr. Plant Biol.\u003c/em\u003e \u003cb\u003e66\u003c/b\u003e, 1044\u0026ndash;1047 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Rather GA, \u003cem\u003eet al.\u003c/em\u003e Molecular characterization and overexpression analyses of secologanin synthase to understand the regulation of camptothecin biosynthesis in \u003cem\u003eNothapodytes nimmoniana\u003c/em\u003e (Graham.) Mabb. \u003cem\u003eProtoplasma\u003c/em\u003e \u003cb\u003e257\u003c/b\u003e, 391\u0026ndash;405 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Pu X, \u003cem\u003eet al.\u003c/em\u003e Possible clues for camptothecin biosynthesis from the metabolites in camptothecin-producing plants. \u003cem\u003eFitoterapia\u003c/em\u003e \u003cb\u003e134\u003c/b\u003e, 113\u0026ndash;128 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Rai A, \u003cem\u003eet al.\u003c/em\u003e Chromosome-level genome assembly of \u003cem\u003eOphiorrhiza pumila\u003c/em\u003e reveals the evolution of camptothecin biosynthesis. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 405 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Cheng H, \u003cem\u003eet al.\u003c/em\u003e Haplotype-resolved de novo assembly using phased assembly graphs with hifiasm. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e, 170\u0026ndash;175 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Jia K-H, \u003cem\u003eet al.\u003c/em\u003e SubPhaser: a robust allopolyploid subgenome phasing method based on subgenome-specific k-mers. \u003cem\u003eNew Phytol.\u003c/em\u003e \u003cb\u003e235\u003c/b\u003e, 801\u0026ndash;809 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Mu W, \u003cem\u003eet al.\u003c/em\u003e Subgenomic Stability of Progenitor Genomes During Repeated Allotetraploid Origins of the Same Grass \u003cem\u003eBrachypodium hybridum\u003c/em\u003e. \u003cem\u003eMol. Biol. Evol.\u003c/em\u003e \u003cb\u003e40\u003c/b\u003e, msad259 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Petronikolou N, \u003cem\u003eet al.\u003c/em\u003e Loganic Acid Methyltransferase: Insights into the Specificity of Methylation on an Iridoid Glycoside. \u003cem\u003eChemBioChem\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 784\u0026ndash;788 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Guirimand G, \u003cem\u003eet al.\u003c/em\u003e The subcellular organization of strictosidine biosynthesis in \u003cem\u003eCatharanthus roseus\u003c/em\u003e epidermis highlights several trans-tonoplast translocations of intermediate metabolites. \u003cem\u003eFEBS J.\u003c/em\u003e \u003cb\u003e278\u003c/b\u003e, 749\u0026ndash;763 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Ma X, \u003cem\u003eet al.\u003c/em\u003e The structure of \u003cem\u003eRauvolfia serpentina\u003c/em\u003e strictosidine synthase is a novel six-bladed beta-propeller fold in plant proteins. \u003cem\u003ePlant Cell\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e, 907\u0026ndash;920 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Eger E, et al. Inverted Binding of Non-natural Substrates in Strictosidine Synthase Leads to a Switch of Stereochemical Outcome in Enzyme-Catalyzed Pictet\u0026ndash;Spengler Reactions. J. Am. Chem. Soc. \u003cb\u003e142\u003c/b\u003e, 792\u0026ndash;800 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Christ B, \u003cem\u003eet al.\u003c/em\u003e Repeated evolution of cytochrome P450-mediated spiroketal steroid biosynthesis in plants. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 3206 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Kynast RG, \u003cem\u003eet al.\u003c/em\u003e Chromosome behavior at the base of the angiosperm radiation: karyology of \u003cem\u003eTrithuria submersa\u003c/em\u003e (Hydatellaceae, Nymphaeales). \u003cem\u003eAm. J. Bot.\u003c/em\u003e \u003cb\u003e101\u003c/b\u003e, 1447\u0026ndash;1455 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Liu Z, \u003cem\u003eet al.\u003c/em\u003e Genome size estimation of Chinese cultured \u003cem\u003eArtemisia annua\u003c/em\u003e L. J. \u003cem\u003ePlant Biol. Crop Res.\u003c/em\u003e \u003cb\u003e1\u003c/b\u003e, 1002 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Qin C, \u003cem\u003eet al.\u003c/em\u003e Whole-genome sequencing of cultivated and wild peppers provides insights into Capsicum domestication and specialization. \u003cem\u003eProc. Natl Acad. Sci. USA\u003c/em\u003e \u003cb\u003e111\u003c/b\u003e, 5135\u0026ndash;5140 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Marcais G, Kingsford C. A fast, lock-free approach for efficient parallel counting of occurrences of k-mers. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 764\u0026ndash;770 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Dudchenko O, \u003cem\u003eet al.\u003c/em\u003e De novo assembly of the Aedes aegypti genome using Hi-C yields chromosome-length scaffolds. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e356\u003c/b\u003e, 92\u0026ndash;95 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 1754\u0026ndash;1760 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 357\u0026ndash;360 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Simao FA, \u003cem\u003eet al.\u003c/em\u003e BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, 3210\u0026ndash;3212 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Xu Z, Wang H. LTR_FINDER: an efficient tool for the prediction of full-length LTR retrotransposons. \u003cem\u003eNucleic. Acids Res.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, W265-W268 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Flynn JM, \u003cem\u003eet al.\u003c/em\u003e RepeatModeler2 for automated genomic discovery of transposable element families. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cb\u003e117\u003c/b\u003e, 9451\u0026ndash;9457 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Tarailo-Graovac M, Chen N. Using RepeatMasker to identify repetitive elements in genomic sequences. \u003cem\u003eCurr. Protoc. Bioinformatics Chap.\u0026nbsp;4\u003c/em\u003e, 4.10.11\u0026ndash;14.10.14 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Li H. Minimap2: pairwise alignment for nucleotide sequences. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, 3094\u0026ndash;3100 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Goel M, \u003cem\u003eet al.\u003c/em\u003e SyRI: finding genomic rearrangements and local sequence differences from whole-genome assemblies. \u003cem\u003eGenome Biol\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e, 277 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Emms DM, Kelly S. OrthoFinder: phylogenetic orthology inference for comparative genomics. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e, 238 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. \u003cem\u003eMol. Biol. Evol.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 772\u0026ndash;780 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Nguyen LT, \u003cem\u003eet al.\u003c/em\u003e IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. \u003cem\u003eMol Biol Evol\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e, 268\u0026ndash;274 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Yang Z. PAML 4: phylogenetic analysis by maximum likelihood. \u003cem\u003eMol. Biol. Evol.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 1586\u0026ndash;1591 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Kumar S, \u003cem\u003eet al.\u003c/em\u003e TimeTree 5: An Expanded Resource for Species Divergence Times. \u003cem\u003eMol. Biol. Evol.\u003c/em\u003e \u003cb\u003e39\u003c/b\u003e, (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Vanneste K, \u003cem\u003eet al.\u003c/em\u003e Analysis of 41 plant genomes supports a wave of successful genome duplications in association with the Cretaceous-Paleogene boundary. \u003cem\u003eGenome Res.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 1334\u0026ndash;1347 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Bielawski JP, Yang Z. Maximum likelihood methods for detecting adaptive evolution after gene duplication. \u003cem\u003eJ. Struct. Funct. Genomics\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, 201\u0026ndash;212 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Eberhardt J, \u003cem\u003eet al.\u003c/em\u003e AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings. \u003cem\u003eJ. Chem. Inf. Model\u003c/em\u003e \u003cb\u003e61\u003c/b\u003e, 3891\u0026ndash;3898 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. \u003cem\u003eJ. Comput. Chem.\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, 455\u0026ndash;461 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e Rolland V. Determining the Subcellular Localization of Fluorescently Tagged Proteins Using Protoplasts Extracted from Transiently Transformed \u003cem\u003eNicotiana benthamiana\u003c/em\u003e Leaves. \u003cem\u003eMethods Mol. Biol.\u003c/em\u003e \u003cb\u003e1770\u003c/b\u003e, 263\u0026ndash;283 (2018).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5971869/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5971869/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e \u003cem\u003eNothapodytes nimmoniana\u003c/em\u003e is known to produce the highest content of the anticancer compound camptothecin (CPT) in the plant kingdom. We present the chromosome-level allotetraploid genome of \u003cem\u003eN. nimmoniana\u003c/em\u003e, marking the first genome sequence from the order Icacinales. This 5-Gb genome encodes 92,630 genes, with subgenome B exhibiting dominant gene expression. Through genome mining, we identified and characterized four key enzymes involved in CPT biosynthesis, revealing that \u003cem\u003eN. nimmoniana\u003c/em\u003e shares a similar prestrictosidine pathway with most monoterpene indole alkaloid-producing plants. Notably, homoeologous pairs of all characterized enzymes maintained their functions across both subgenomes, suggesting that gene duplication from allotetraploidization likely enhances CPT production in this species. Phylogenetic and syntenic analyses revealed that strictosidine synthase and strictosamide epoxidase were independently recruited in \u003cem\u003eN. nimmoniana\u003c/em\u003e, \u003cem\u003eCamptotheca acuminata\u003c/em\u003e, and \u003cem\u003eOphiorrhiza pumila\u003c/em\u003e, supporting the hypothesis that CPT biosynthesis evolved independently at least three times within the asterid clade.\u003c/p\u003e","manuscriptTitle":"The Nothapodytes nimmoniana genome provides insights into the independent evolution of camptothecin biosynthesis in asterids","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-02-11 11:33:44","doi":"10.21203/rs.3.rs-5971869/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"1bc02eb4-77bb-4ace-a777-e4ed0eb30bbd","owner":[],"postedDate":"February 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":44025394,"name":"Biological sciences/Molecular biology/Transcriptomics"},{"id":44025395,"name":"Biological sciences/Evolution/Molecular evolution"}],"tags":[],"updatedAt":"2025-03-12T15:15:10+00:00","versionOfRecord":[],"versionCreatedAt":"2025-02-11 11:33:44","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5971869","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5971869","identity":"rs-5971869","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}