The C. acuminata seeds collection, seedlings growth, total RNA extraction and isolation, and cDNA library construction were performed on the basis of the reported procedures (Yang et al., 2019; Awadasseid et al., 2020).

The amino acid residue sequences of CaTDC1 (GenBank: U73656) (López-Meyer and Nessler, 1997), CaTDC2 (GenBank: U73657) (López-Meyer and Nessler, 1997), and a putative CaTDC (GenBank: KU842377) (Sadre et al., 2016) were retrieved from the National Center for Biotechnology Information (NCBI) database. The annotated TDCs Caa_locus_20066 and Caa_locus_5306 were retrieved from the C. acuminata transcriptome dataset downloaded from http://medicinalplantgenomics.msu.edu (Góngora-Castillo et al., 2012). The annotated TDCs Cac_g018974.t1 and Cac_g023139.t1 were retrieved from the C. acuminata genome dataset downloaded from http://medicinalplantgenomics.msu.edu by Zhao et al. (2017). The annotated TDCs CacGene21142.t1 and CacGene13733.t1 were retrieved from the C. acuminata genome dataset downloaded from https://doi.org/10.6084/m9.figshare.12570599 by Kang et al. (2021).

Multiple amino acid residue sequences alignment was performed by using Clustal Omega software.¹ The results of the multiple amino acid residue sequences alignment of the aforementioned TDCs were presented in Figure 2.

According to the multiple amino acid residue sequences alignment results (Figure 2), the specific primers CaTDC3-F (ATGGGTAGCCTTGATTCCAATTACG, 5′→ 3′) and CaTDC3-R (TCAATCCTCTTTCAGGAGAACATCC, 5′→ 3′) were designed to clone the whole open reading frame of the target TDC. All primers were synthesized, purified, and authenticated by TsingKe Biological Technology (Chengdu) Co., Ltd. Using the previously prepared cDNA mixture (Yang et al., 2019; Awadasseid et al., 2020) as template and CaTDC3-F and -R as primers, the open reading frame of CaTDC3 was obtained from the PCR amplification reaction by following the cycling conditions: 1 cycle of 95°C for 2 min and 35 cycles of 95°C for 20 s, 55°C for 20 s, 72°C for 1 min followed by a final extension at 72°C for 5 min. The amplification products were gel-purified and ligated into the pGM-T vector [Tiangen Biotech (Beijing) Co., Ltd, China]. The constructs were chemically transformed into E. coli DH5α competent cells. The following DNA sequencing was performed in both directions by TsingKe Biological Technology (Chengdu) Co., Ltd. The nucleotide sequence was analyzed by using the similarity search BLAST program.

The properties of CaTDC3 were estimated by using the online ExPASy ProtParam tools.² To assess the evolutionary relationship between CaTDC3 and other TDC homologs from different plant species, CaTDC3 was set as query to search the NCBI database by using BLASTp searches and the amino acid residue sequences of the functionally characterized TDCs were retrieved from the NCBI database. The amino acid residue sequences of TDCs were aligned by using the Clustal W program.³ MEGA 7 software was employed to construct a phylogenetic tree with neighbor-joining method. Bootstrap analysis with 1,000 replicates was conducted to obtain the confidence levels for the branches.

The crystal structure of C. roseus TDC (CrTDC, PDB: 6EEW) (Torrens-Spence et al., 2020) was used as a template to build the three-dimension structure of CaTDC3 by using the SWISS-MODEL toolkit from ExPASy web server.⁴ Based on the predicted three-dimension structure, the structurally and functionally important regions of CaTDC3 were identified from the deduced amino acid residue sequence of CaTDC3 by using AutoDock 1.5.7.⁵ The number of points in x-, y-, and z-dimension are 50 and the spacing is 0.375. The x, y, and z center are −14.917, −8.839, and −20.646, respectively. Genetic algorithm was selected for the search parameters and the numbers of runs are 50. The genetic algorithm (4.2) was selected for output. The conformation 4 was selected for further analyses. Simulations with bonded L-tryptophan were performed with PyMOL (2.5.2) molecular graphics system.

For the heterologous overexpression of CaTDC3 in E. coli BL21(DE3), the specific primes CaTDC3-BN-F (CGGGATCCATGGGTAGCCTTGATT, 5′→ 3′) and CaTDC3-BN-R (ATAGTTTAGCGGCCGCTCAATCCTCTT TCAG, 5′→ 3′) were designed and synthesized to amplify the whole open reading frame of CaTDC3 from the above cloned full-length cDNA of CaTDC3 by using a TransStart^® FastPfu DNA polymerase [TransGen Biotech (Beijing) Co., Ltd., China] with the following cycling conditions: 1 cycle of 95°C for 1 min and 35 cycles of 95°C for 20 s, 62°C for 20 s, 72°C for 30 s followed by a final extension at 72°C for 10 min. The PCR amplification products were gel-purified, digested with BamHI and NotI, and ligated into the pET28a vector that was digested with the same endonucleases to obtain the recombinant plasmid pET28a-CaTDC3. E. coli DH5α competent cells were chemically transformed with the expression construct pET28a-CaTDC3. The nucleotide sequence was sequenced and analyzed by following the procedures mentioned above.

The verified recombinant plasmid pET28a-CaTDC3 was chemically transformed into E. coli BL21(DE3) competent cells to generate the recombinant strain which was grown on Luria-Bertani (10 g L^–1 of tryptone, 5 g L^–1 of yeast extract, and 10 g L^–1 of NaCl) plates supplemented with 50 μg ml^–1 of kanamycin. A single colony of the recombinant strain was inoculated into 5 ml of Luria-Bertani broth containing kanamycin (50 μg ml^–1) and the resulting mixture was incubated overnight at 37°C, 180 rpm in a shaking incubator. An aliquot (500 μl) was inoculated into 500 ml of Luria-Bertani broth supplemented with the same antibiotics and incubated at 37°C, 180 rpm. When the optical density (OD_{600 nm}) of the culture reached 0.6–0.8, 1 mM isopropyl β-D-L-thiogalactopyranoside (Sangon Biotech, Shanghai, China) was added into the culture to induce the overexpression of recombinant CaTDC3. The resulting mixture was incubated at 16°C, 150 rpm for 18 hrs. The cells were harvested by centrifugation at 4°C, 4,000 rpm for 15 min. The cell pellets were washed twice by PBS buffer (20 mM NaH₂PO₄, 500 mM NaCl, 10% glycerol, 10 mM imidazole, 5 mM β-mercaptoethanol, pH 7.4) and re-suspended in the same buffer containing 1 mg L^–1 of lysozyme and 1 mM phenylmethylsulfonyl fluoride. The resulting solution was sonicated in an ice-bath at a 10 s interval until the mixture was homogeneous. The supernatant soluble fraction was recovered by centrifugation at 4°C, 12,000 rpm for 30 min. The supernatant was filtered through a 0.45 μm PES membrane and the resulting filtrate was loaded onto a gravity column of nickel nitrilotriacetic acid resin (Ni-NTA, Sangon Biotech, Shanghai, China) pre-equilibrated with binding buffer (20 mM NaH₂PO₄, 500 mM NaCl, pH 7.4). The binding buffer with different concentration of imidazole (10, 50, and 250 mM) was employed to elute the column. Aliquots of each fraction were analyzed on a 12% SDS-PAGE. Fractions containing His₆-tagged CaTDC3 were pooled, desalted, and concentrated by using an Amicon Ultra centrifugal filter MWCO 30 kDa (Merck Millipore Ltd., United States) with dialysis buffer (50 mM NaH₂PO₄, 150 mM NaCl, pH 7.6). The purified protein was stored in dialysis buffer containing 20% glycerol at −80°C for further usage. The purified protein samples were analyzed on 12% SDS-PAGE and the concentration was estimated by using the ε_{280 nm} = 84,060 M^–1 cm^–1 for CaTDC3 calculated from ExPASy ProtParam.

The catalytic activity of CaTDC3 was measured as described previously (Liu et al., 2012). Briefly, the reaction was performed in 100 μl of PBS buffer (50 mM, pH 7.6) containing 4 μM PLP, 5 mM L-tryptophan, and 16 μM CaTDC3. The reaction mixtures were incubated at 37°C for 5, 10, and 20 min. At each time point the mixture was quenched with 100 μL of chill methanol to precipitate the protein. All precipitates were removed by centrifugation at 12,000 rpm for 10 min. The resulting supernatant was subjected to analyze by high-performance liquid chromatography (HPLC)-diode array detector (DAD) that was equipped with a C₁₈ column (250 mm × 4.6 mm, 5 μm, Altima). The mobile phase consisted of solvent A (CH₃CN) and solvent B (H₂O containing phosphoric acid, pH 2.3) followed a gradient elution program (0 min, 5% A; 16 min, 50% A; 18 min, 95% A; and 23 min, 95% A) at a flow rate of 1 ml min^–1 at room temperature, monitored by a DAD at 280 nm.

All enzymatic reactions were performed in triplicate and each reaction was initiated by the addition of CaTDC3.

To determine the optimal pH value of the reaction buffer for CaTDC3-catalyzed decarboxylation reaction, the enzymatic reactions were performed in various buffer systems, including AcOH-NaOAc buffer (50 mM, pH 5.0), Na₂HPO₄-NaH₂PO₄ buffer (50 mM, pH 6.5–7.6), Tris–HCl buffer (50 mM, pH 8.0–8.5), and Glycine-NaOH buffer (50 mM, pH 9.0–10.0), at 37°C. The effects of the reaction temperature on the catalytic activity of CaTDC3 were evaluated by incubation of the enzymatic reaction mixtures at 15, 25, 32, 37, 42, 50, and 60°C, respectively, with a constant pH 7.6. Under the optimal reaction temperature and buffer, the time course experiments with varying enzyme concentrations demonstrated that CaTDC3 (0.16 μM) exhibits an initial linear reaction rate within 10 min.

The calibration curve for tryptamine was established to quantitate the formation of tryptamine by HPLC-DAD analyses. Briefly, the standard tryptamine solutions with 0, 0.5, 1, 2.5, 5, 10, 25, 50, and 100 μM were prepared respectively and subjected to HPLC-DAD analyses by using the methods described above. The calibration curve was made by following a linear fit for the relationship of the specific concentration of tryptamine versus the corresponding peak area integral at 280 nm. Limits of determination and quantification were determined as signal/noise = 3 and 10, respectively. The kinetics parameters of CaTDC3 were determined under the aforementioned optimal reaction conditions. The enzymatic reaction was performed in 100 μL of PBS buffer (50 mM NaH₂PO₄, pH 7.6) containing 4 μM PLP, 0.16 μM CaTDC3, and L-tryptophan with different concentrations (10, 25, 50, 100, 250, and 500 μM). The kinetic constants were calculated with non-linear regression analysis using Origin 9.0 software.

To investigate the substrate scope and selectivity of CaTDC3, 28 L-tryptophan derivatives and analogs were introduced as potential substrates under the optimal reaction conditions determined for CaTDC3-catalyzed L-tryptophan decarboxylation. The fluoro-, chloro-, bromo-, methyl-, methoxy-, and 7-aza-L-tryptophan analogs were purchased from Amatek Scientific, Suzhou, China. L-Tryptophan, 5-hydroxy-L-tryptophan, N^α -acetyl-L-tryptophan, and L-tryptophan methyl ester were purchased from Aladdin Industrial Corporation, Shanghai, China. 6-Hydroxy-L-tryptophan, 1-thio-L-tryptophan, 7-aza-tryptamine, 5-chloro-tryptamine hydrochloride, and 5-bromo-tryptamine hydrochloride were purchased from Bide Pharmatech Ltd., Shanghai, China. 5-Hydroxy-tryptamine was purchased from Macklin Biochemical Co., Ltd., Shanghai, China. 5-Fluoro-tryptamine hydrochloride was purchased from Titan scientific Co., Ltd., Shanghai, China. Kynuramine dihydrobromide was purchased from Sigma-Aldrich, American. The enzymatic and control reaction assays were conducted in duplicated. The HPLC-DAD analyses were performed on the basis of the aforementioned analytic methods.

Generally the aforementioned L-tryptophan derivatives and analogs were employed as substrates in the CaTDC3-catalyzed decarboxylation reactions, the enzymatic products were identified by comparison of their HPLC retention time and UV profile with those of the authentic standards. Then the individual product in the enzymatic reaction was collected and characterized by high resolution mass spectrometry (HRMS) data analysis. The pooled product was directly analyzed on a Bruker microTOF-Q mass spectrometer (Bremen, Germany) equipped with an electrospray ionization interface (ESI). The HRMS(ESI) was operated in positive ion mode and the spectra were collected in the enhanced full mass scan mode from m/z 50–1,500.

For HPLC-DAD analysis of the enzymatic reaction with 7-aza-L-tryptophan as substrate, the mobile phase consisted of solvent A (CH₃CN) and solvent B (H₂O containing-phosphoric acid, pH 2.3) followed a gradient elution program (0 min, 5% A; 16 min, 50% A; 18 min, 95% A; and 23 min, 95% A) at a flow rate of 0.5 ml min^–1, monitored by a DAD at 280 nm. When 5-hydroxy- and 6-hydroxy-L-tryptophans were used as substrate, the HPLC-DAD analyses were performed with the mobile phase consisted of solvent A (CH₃OH) and solvent B (H₂O containing-phosphoric acid, pH 2.3) followed a gradient elution program (0 min, 5% A; 20 min, 20% A; 23 min, 95% A; and 27 min, 95% A) at a flow rate of 1.0 ml min^–1, monitored by a DAD at 280 nm. The enzymatic reaction with L-tryptophan methyl ester as substrate was analyzed by HPLC-DAD equipped with an Agilent ZORBAX SB C₁₈ analytic column (250 mm × 4.6 mm, 5 μm). The mobile phase consisted of solvent A (CH₃CN) and solvent B (H₂O containing-phosphoric acid, pH 2.3) followed a gradient elution program (0 min, 5% A; 16 min, 50% A; 18 min, 95% A; and 20 min, 95% A) at a flow rate of 1.0 ml min^–1, monitored by a DAD at 280 nm.

To characterize the structure of the product from the CaTDC3-catalyzed decarboxylation reaction using 1-thio-L-tryptophan as substrate, 2.2 mg of 1-thio-L-tryptophan was added into 2 ml of NaH₂PO₄-NaOH buffer (50 mM, pH 7.6) containing CaTDC3 (3 μM) and PLP (4 μM). The reaction was incubated at 37°C for 48 h. The reaction mixture was extracted with EtOAc (2 ml) three times. EtOAc was removed to give the crude product. HRMS(ESI) m/z 178.0698 [M + H]⁺; ¹H NMR (400 MHz, CDCl₃₎ δ 8.03 (1H, dd, J = 7.0, 1.0 Hz), 7.92 (1H, dd, J = 7.3, 1.2 Hz), 7.56 (1H, td, J = 7.5, 1.2 Hz), 7.52 (1H, td, J = 7.5, 1.0 Hz), 7.32 (1H, s), 3.26 (2H, t, J = 6.3 Hz), and 3.18 (2H, t, J = 6.3 Hz).

Multiple amino acid residue sequences alignment of the functionally characterized CaTDCs, i.e., CaTDC1 and CaTDC2 (López-Meyer and Nessler, 1997), a putative CaTDC (GenBank: KU842377) (Sadre et al., 2016), and the annotated CaTDCs from the C. acuminata transcriptome (Góngora-Castillo et al., 2012) and genome (Zhao et al., 2017; Kang et al., 2021) datasets showed these TDCs could be classified into two groups in view of more than 60 amino acid residues difference highlighted in yellow for Group I and in cyan for Group II, respectively (Figure 2). CaTDC1, Caa_locus_20066, Cac_g018974.t1, CacGene21142.t1, and KU842377 were classified into Group I, while CaTDC2, Caa_locus_5306, Cac_g023139.t1, CacGene13733.t1 were grouped in Group II (Figure 2). In Group I, Cac_g018974.t1 and KU842377 showed identical amino acid residue sequences (Figure 2). Caa_locus_20066 displayed 7 different amino acid residues at 261-267, while CacGene21142.t1 had a different C-terminus at 484-576 (Figure 2). Closest analyses of their amino acid residue sequences in Group I demonstrated that the aforementioned four TDCs must be the same TDC that has 13 amino acid residues (in red color) different from CaTDC1 (Figure 2). CaTDC2 has 6 different amino acid residues (in green color), comparing with others in Group II (Figure 2).

The above mentioned multiple amino acid residue sequences alignment of the functionally characterized CaTDCs and the annotated TDCs from C. acuminata transcriptome and genome datasets implied the presence of 13 different amino acid residues in Group I CaTDCs (Figure 2). To further verify the presence of different amino acid residues and clarify the catalytic function of CaTDC in Group I, we cloned a TDC-encoding gene from C. acuminata on the basis of the aforementioned alignment results (Figure 2). The cloned gene is an open reading frame of 1,509 bp, encoding a TDC with 502-amino acid residues that is identical to those of KU842377. It was designated CaTDC3 (GenBank: ON964510) in view of 13 and 66 different amino acid residues, respectively, from functionally characterized CaTDC1 and CaTDC2 (Figure 2).

CaTDC3 was predicted to have a calculated molecular mass of 55.6 kDa and an isoelectric point of 6.56. Similarity search showed that CaTDC3 shares 54–97% amino acid residue identities with the functionally characterized plant TDCs (Figure 3). Based on the catalytic mechanisms of plant TDCs (López-Meyer and Nessler, 1997; Liu et al., 2012), Lys319 is most likely to be the key catalytic amino acid in CaTDC3 (Figure 3). The phylogenetic tree analysis showed that the plant aromatic amino acid decarboxylases should be grouped into three groups depending on the substrate specificity for tryptophan, tyrosine, and phenylalanine, respectively (Figure 4A). CaTDC3 falls into the clade of TDCs in which TDCs from C. roseus (CrTDC) (De Luca et al., 1989), R. verticillata (RvTDC) (Liu et al., 2012), Gelsemium sempervirens (GsTDC) (Franke et al., 2019), and O. pumila (OpTDC) (Yamazaki et al., 2003) had been functionally characterized to be involved in the biosynthesis of monoterpene indole alkaloids, indicating CaTDC3 has similar catalytic function (Figure 4A).

The crystal structure of CrTDC with 2.05 angstrom resolution (Torrens-Spence et al., 2020) was employed to build the three-dimension structure of CaTDC3. As depicted in Figure 4B, CaTDC3 is present in a dimeric form, in which the catalytically active domain consisted of Thr205, Thr167, Ser316, His318, and Lys319 (Figure 4C).

To clarify the catalytic decarboxylation function of CaTDC3 (Figure 5A), the entire open reading frame of CaTDC3 was subcloned into a pET28a vector to generate an N-terminal His₆-tagged expression construct. The intact CaTDC3 was heterologously overexpressed in E. coli BL21(DE3) and the recombinant CaTDC3 was purified to homogeneity (Supplementary Figure 1) by using Ni-NTA affinity chromatograph.

The enzymatic decarboxylation activity assays were conducted by incubation of L-tryptophan, PLP, and CaTDC3. HPLC-DAD analyses showed the presence of an enzymatic reaction product in the CaTDC3-catalyzed reaction mixture (Figure 5B, panel II), compared with the control experiment using the boiled CaTDC3 as catalyst (Figure 5B, panel III). The aforementioned product was demonstrated to have identical HPLC retention time (Figure 5B, panel II) and UV profile (Supplementary Figure 2A, panel I) to those of the authentic tryptamine (Figure 5B, panel IV and Supplementary Figure 2A, panel II). The protonated molecule at m/z 161.1082 ([M + H]⁺) in the HRMS(ESI) of the enzymatic product (Supplementary Figure 2B) confirmed that the CaTDC3-catalyzed reaction product is tryptamine, the decarboxylation product of L-tryptophan.

The recombinant CaTDC3 exhibited a maximum catalytic activity when the reaction was performed in the reaction buffer with pH = 7.6 (Figure 5C) and the reaction temperature was set at 37°C (Figure 5D). Under the aforementioned optimal reaction conditions, CaTDC3 showed an apparent K_m value of 48 ± 2 μM and V_max value of 4.10 ± 0.07 μM min^–1 for L-tryptophan (Figure 5E).

The predicted three-dimension structure of CaTDC3 (Figures 4B,C) was employed to perform molecular docking using L-tryptophan as substrate (Figure 6A). The catalytically active pocket of CaTDC3 was composed of amino acid residues from two monomers, one in yellow and another in light gray (Figure 6A). The molecular docking results implied that L-tryptophan analogs with the substituent groups at indole ring and the side chain might be recognized by CaTDC3 in view of a relative big spatial volume of the catalytic pocket (Figures 6A,B). Thus, CaTDC3 may catalyze the decarboxylation of L-tryptophan analogs.

Plant TDCs are members of aromatic amino acid decarboxylases family (Figure 4A), which implied that some TDCs might recognize other aromatic amino acids and catalyze the corresponding decarboxylation reaction. Herein, various L-tryptophan derivatives and analogs were employed to investigate the catalytic promiscuity of CaTDC3 toward these substrates.

Carefully checking the three-dimension structure of CaTDC3 docked with L-tryptophan and PLP suggested that the substitution groups located at the aromatic ring and the side-chain might be recognized by CaTDC3 (Figure 6). The hydroxylated analog 5-hydroxy-L-tryptophan (Figure 8A) was employed to testify the aforementioned hypothesis. The HPLC-DAD analyses indicated that CaTDC3 can recognize 5-hydroxy-L-tryptophan and afford a product with retention time at 11.61 min (Figure 8B, panel II), comparing with the control experimental (Figure 8B, panel III). The product possesses an identical retention time (Figure 8B, panel II) and a UV profile (Supplementary Figure 4A, panel I) with those of 5-hydroxytryptamine (Figure 8B, panel IV and Supplementary Figure 4A, panel II). The enzymatic product showed the protonated molecule at m/z 177.1026 ([M + H]⁺) in the HRMS(ESI) (Supplementary Figure 4B), which confirmed that the CaTDC3-catalyzed reaction product is 5-hydroxy-tryptamine, the decarboxylation product of 5-hydroxy-L-tryptophan.

6-Hydroxy-L-tryptophan (Figure 8A) was verified to be recognized by CaTDC3 and the decarboxylation product, i.e., 6-hydroxy-tryptamine, was detected in the reaction mixture (Supplementary Figure 5).

Encouraged by the above results, we employed 10 halogenated L-tryptophans, including 4- F-, 5- F-, 6- F-, 7- F-, 4- Cl-, 5- Cl-, 6- Cl-, 7- Cl-, 5- Br-, and 6-Br-L-tryptophans (Figure 8A), to investigate the substrate scope of CaTDC3-catalyzed decarboxylation reaction. The corresponding decarboxylated products, i.e., halogenated tryptamines, were confirmed by HPLC-DAD and HRMS(ESI) data analyses from the reaction mixtures (Figures 8C–E and Supplementary Figures 6–15).

Two L-tryptophan analogs with larger substituent groups at C-5, i.e., 5-methyl- and 5-methoxy-L-tryptophans (Figure 8A), were used as substrate, respectively, in the CaTDC3-catalyzed decarboxylation reactions. However, the corresponding decarboxylation products could not be detected in the reaction mixture (Supplementary Figure 16).

Two TDC-encoding genes, CaTDC1 and CaTDC2 (López-Meyer and Nessler, 1997), had been cloned from C. acuminata, a camptothecin-producing plant. They share 81% nucleotide sequence identity and 84% amino acid residue identity (López-Meyer and Nessler, 1997). CaTDC1 was developmentally expressed at different levels in all tissues of C. acuminata and it was clarified to provide tryptamine for biosynthesis of camptothein under normal growth conditions (López-Meyer and Nessler, 1997). CaTDC2 was specifically induced when C. acuminata plant tissues were treated with elicitors (López-Meyer and Nessler, 1997). The previous DNA gel blot analysis had implied a third TDC when C. acuminata DNA was digested with EcoR I (López-Meyer and Nessler, 1997). Herein, we mined and cloned CaTDC3, a TDC-encoding gene from C. acuminata. CaTDC3 shares 97 and 85% amino acid residue identities with CaTDC1 and CaTDC2, respectively (Figures 1, 2). It is in good accordance with previous transcriptome and genome sequencing results (Góngora-Castillo et al., 2012; Zhao et al., 2017; Kang et al., 2021) and DNA gel blot analysis (López-Meyer and Nessler, 1997).

Recombinant CaTDC3 was clarified to catalyze an efficient decarboxylation conversion of L-tryptophan into tryptamine (Figure 5). The apparent K_m and k_cat values of recombinant CaTDC3 are 48 μM and 25.6 min^–1, respectively, for L-tryptophan (Figure 5). CaTDC3’s K_m is closest to that of CrTDC (51.7 μM), whereas the k_cat value for CaTDC3 is higher than that of CrTDC (k_cat 5.1 min^–1) (Runguphan et al., 2010). OpTDC (0.72 mM) (Yamazaki et al., 2003), OsTDC_AK31 (0.69 mM) (Kang et al., 2007), and RvTDC (2.89 mM) (Liu et al., 2012) showed higher K_m values than that of CaTDC3. Taken together, CaTDC3 has greater efficiency in decarboxylation for L-tryptophan than other plant TDCs.

Multiple amino acid residue sequences alignment results showed that CaTDC1, CaTDC2, and CaTDC3 contain same conserve amino acid residues, identical catalytic amino acid residues, and similar three-dimensional structure, comparing with those of CrTDC (Figure 3). Phylogenetic tree analyses revealed that CaTDC1, CaTDC2, and CaTDC3 are very close to each other (Figure 4A). These results implied that they might display similar catalytic activity and substrate scope.

CaTDC3 shows strict stereoselectivity for substrate in view of that only L-tryptophan could be recognized by CaTDC3 and afford the corresponding decarboxylative product (Figure 7). The conclusion was further supported by the following CaTDC3-catalyzed decarboxylation reactions using hydroxylated, halogenated, and C^β -methylated L-tryptophans as substrates and the related decarboxylation products were detected (Figures 8–10). Additionally, CaTDC3 could recognized L-kynurenine and catalyze its decarboxylation to generate kynuramine (Figure 11). While D-kynurenine could not be accepted by CaTDC3 (Supplementary Figure 24).

Homology modeling and molecular docking (Figure 6) revealed a relative big spatial volume of the catalytic pocket presented in CaTDC3, which implied that CaTDC3 might recognize the L-tryptophan analogs with the appropriate substituent groups at the indole ring and the side chain on the basis of the substrate basic requirement experimental results (Figure 7).

Twelve hydroxylated and halogenated L-tryptophans (Figure 8) can be converted into the corresponding tryptamines by the catalysis of CaTDC3. However, L-tryptophan with larger groups substituted at C-5, i.e., 5-methyl- and 5-methoxy-L-tryptophan, cannot be converted by CaTDC3 (Figure 8). The molecular docking of CaTDC3 with 5-methoxy-L-tryptophan (Figure 12B) gave higher binding energy than that of CaTDC3 with 5-hydroxy-L-tryptophan (Figure 12A), indicating larger substituent groups at the indole ring could not be accepted.

The integration of halogen is a frequent modification for specialized metabolites and the resulting halogenated products display a wide range of biological activities and well pharmacological properties, including anticancer and antibiotic properties (Neumann et al., 2008; Runguphan et al., 2010). For instance, chlorinated alkaloids were generated from introduction of chlorine into tryptophan by halogenase and the following integration into monoterpene indole alkaloids (Runguphan et al., 2010). The halogenated tryptamines such as ¹⁸F-tryptamine, a potential radiopharmaceutical for monitoring of abnormal brain states occurring in diverse diseases, are extremely useful in nuclear medicine (Dragulska and Kańska, 2014). Here we showed that all halogenated L-tryptophans could be converted into halogenated tryptamine by CaTDC3 (Figure 8), indicating its potential usage in metabolic engineering for pharmaceutically important metabolites. The representative molecular docking of CaTDC3 with 6-F-L-tryptophan (Figure 12C), 6-Cl-L-tryptophan (Figure 12D), and 6-Br-L-tryptophan (Figure 12E), respectively, supported the above mentioned experimental results.

Molecular docking of CaTDC3 with C^β -methyl-L-tryptophan (Figure 12F) suggested the binding energy is −6.86 kcal mol^–1 and CaTDC3 may catalyze decarboxylation of the aforementioned L-tryptophan analogs. The experimental results (Figure 9) revealed that both (R)- and (S)-C^β -methyl-L-tryptophans were decarboxylated by CaTDC3 efficiently.

1-Thio-L-tryptophan and 7-aza-L-tryptophan, two unnatural L-tryptophans substituted by hetero atoms, could be recognized by CaTDC3 and the related decarboxylation products were obtained successfully (Figure 10). Molecular docking of CaTDC3 with 1-thio-L-tryptophan (Figure 12G) and 7-aza-L-tryptophan (Figure 12H), respectively, suggested the binding energy are −6.53 and −7.51 kcal mol^–1. Thus, the NH group of the indole ring and the C atom on the aromatic ring of L-tryptophan might be replaced by appropriate hetero atoms and the corresponding hetero-tryptamines could be obtained from CaTDC3-catalyzed decarboxylation reactions.

CaTDC3 cannot catalyze the decarboxylation of L-phenylalanine, L-tyrosine, D,L-phenylglycine, L-homophenylalanine, and L-DOPA (Figure 7). However, L-kynurenine, an L-tryptophan metabolic degradation product, could be decarboxylated by CaTDC3 (Figure 10). It seems that the consistency of the indole ring of L-tryptophan has little effect on CaTDC3-catalyzed decarboxylation reaction. Closest checking suggested that the structure and conformation of L-kynurenine are similar to those of L-tryptophan (Figures 6A, 12I).

Both chemical and biological approaches had been developed to prepare tryptamine and its analogs (Righi et al., 2012; McDonald et al., 2019; Wang et al., 2019). The present CaTDC3 shows high efficiency in catalytic decarboxylation and displays substrate promiscuity for a wide range of L-tryptophan analogs. Thus, CaTDC3 is an alternative TDC for metabolic engineering and synthetic biological applications (Runguphan et al., 2010; Kalb et al., 2016).