Cloning and characterization of a tyrosine decarboxylase involved in the biosynthesis of galanthamine in Lycoris aurea

Plant materials and growth conditions

L. aurea were planted in a nursery at the Institute of Botany, Jiangsu Province and Chinese Academy of Sciences, Nanjing, China. The flowers and scapes of L. aurea were collected in August while bulbs, roots, seeds, and leaves of L. aurea were collected in October. For the treatment experiments, the seeds of L. aurea were planted in an incubator with 16/8 h (22/18 °C) day/night regimes at 100 µmol m⁻²s⁻¹ irradiation. Eight month-old seedlings were imposed for 0, 6, 12 and 24 h with 100 µM MeJA in 0.02% DMSO, or with 0.02% Dimethyl sulfoxide (DMSO) as the vehicle control, respectively, as previously described by Ma et al. (2016). Then each treated sample was collected for further analyses.

Isolation and cloning of LaTYDC1 cDNA

By BLAST against the transcriptome database of L. aurea (Wang et al., 2013; Wang et al., 2017), using the relevant primers LaTYDC-F and LaTYDC-R (Table S1) for running PCR, the full length sequence of LaTYDC1 was identified. The accession number of LaTYDC1 is MG932082.2.

Total RNA extraction, cDNA first-strand synthesis and qRT-PCR

100 mg of fresh tissue were fully ground in liquid nitrogen and 1 ml RNAiso was added following the RNA extraction protocol described in the instructions (Takara, Dalian, China). After the extracted RNA was measured using a UV spectrophotometer and 1% (w/v) agarose gel electrophoresis, they were used as templates to create the cDNA transcribed in reverse as previously described by Xu et al. (2016). For the qRT-PCR tests, the DNA sequence of primers of LaTYDC1 and LaTIP41 are listed in Table S1. The qRT-PCR program started at 95 °C for 5 min, then took place in a condition of 95 °C for 15 s, 56 °C for 15 s, 72 °C for 20 s, for 40 cycles. The L. aurea TIP41 gene was applied as a reference gene (Ma et al., 2016, the sequence of TIP41 is shown in Supplemental Information 1).

miRNA first-strand synthesis and qRT-PCR for microRNA

Reverse transcription was carried out with Mir-X™ miRNA First-Strand Synthesis kit based on a SYBR qRT-PCR user manual for cDNA synthesis (Takara, Dalian, China). Real-time fluorescence quantitative PCR was performed using qTPWER (Analytik Jena, Jena, Germany). The L. aurea U6 gene was used for normalization. The qRT-PCR protocol was as follows: 95 °C for 30 s, 40 cycles of 95 °C for 5 s, 56 °C for 15 s, 72 °C for 20 s. In order to verify the specificity of the primers, a melting-curve was also analyzed. Each qPCR was repeated three times.

Analysis of RNA ligase-mediated-rapid amplification of cDNA ends (RLM-RACE)

Total RNA was extracted and purified from the leaves of L. aurea as previously described, followed with further purification with the TRIzol™ Plus RNA Purification Kit (Invitrogen, Carlsbad, CA, USA). The purified RNA was amplified for the RLM-RACE experiment using a First Choice™ RLM-RACE Kit (Invitrogen, Carlsbad, CA, USA), following the method described by Shen et al. (2014). The ligation of 5′ RACE adapter and oligo RNA were performed following the manual of the kit. By using the gene specific primers listed in Table S1 with the 5′ RACE outer or inner primer, nested PCR was performed. The PCR products were agarose gel purified and cloned into pGEM-T vector (Promega, Madison, WI, USA) for sequencing.

Subcellular localization

The complete open reading frame fragment of LaTYDC1 was cloned into the expression vector pAN580 and followed by green fluorescent protein (GFP). The construct was transformed into Arabidopsis protoplasts as previously described (Zhang et al., 2014). After incubation for 16 h, the expressing cells were observed under an LSM 780 confocal laser scanning microscope (Zeiss, Oberkochen, Germany) with the 488 nm excitation and 505–530 nm band-pass emission filter (Xu et al., 2017).

Expression and purification of recombinant LaTYDC1

Heterologous expression and protein purification of recombinant LaTYDC1 were carried out as described by Zhang et al. (2011) with some modification. The full length of LaTYDC1 without the termination codon was cloned into the expression vector pGEX4T-1, followed by GST tag. This recombinan expression plasmid was confirmed by sequencing before being introduced into E. coli BL21 (DE3). The transformed bacteria were grown at 37 °C with shaking (180 rpm) in 500 mL of Luria-Bertani medium (Kawalleck et al., 1993) containing ampicillin (50 µg mL⁻¹). When the OD600 of the bacteria reached 0.6∼0.8, isopropyl thio-b-D-galactopyranoside (IPTG) at a final concentration of 0.5 mM was added and the incubation conditions were adjusted to 16 °C, 100 rpm. After incubation for 20 h, cells were harvested by centrifugation (4,000 rpm) for 10 min, suspended in 10 ml of 50 mM Tris-HCl (pH 8.0), and then sonicated on ice for 30 min. The homogenate was centrifuged at 12,000 rpm for 20 min at 4 °C, and then the supernatant was passed through a column of high affinity GST purification medium (GenScript, Nanjing, China). The column was washed with 50 mM potassium phosphate buffer (pH 8.0), then the fusion protein was eluted by elution buffer, which contains 50 mM Tris-HCl (pH 8.0) and 0.01 M reduced glutathione.

Enzyme activity and substrate-specificity analyses

The catalytic activity of the purified GST-LaTYCD1 protein were analyzed followed the method of Lehmann & Pollmann (2009) with some modification. 150 µL mixture containing 15 µg fusion protein, 10 mM pyridoxal-5-phosphate, 15 µL of 20 mM L-tyrosine or L-Dopa and 50 mM Tris-HCl (pH 8.0), and was incubated at 37 °C for 40 min. The reaction was stopped by adding 300 µL of methanol or boiling. Then the mixture of reaction was clarified by passing it through a 0.22 µm pore size nylon membrane filter.

Tyramine and dopamine were analyzed by High Performance Liquid Chromatography (HPLC, Shimadzu LC-20AT, Tokyo, Japan) on a reverse phase column (InertSustain C18, 5 µm, 4.6 mm × 250 mm). To measure tyramine, the solvent composition was: mobile phase A, water with 0.1% phosphoric acid; mobile phase B, 100% acetonitrile. The eluent gradient was: 10% B to 15% B in 25 min, from 15% B to 100% B in 1 min, kept for 10 min, from 100% B to 10% B in 1 min, kept for 15 min, 0.6 mL/min flow rate. The injection volume was 10 µL, and tyramine was detected at 275 nm. To measure dopamine, the solvent composition was: mobile phase A, water with 0.1% formic acid; mobile phase B, 100% acetonitrile. The eluent gradient was: 10% B to 15% B in 15 min, from 15% B to 100% B in 5 min, kept for 5 min, from 100% B to 10% B in 5 min, kept for 10 min, 0.6 mL/min flow rate. The injection volume was 10 µL, and dopamine was detected at 275 nm. Quantification of the target compound was performed using the calibration curves. Varying substrate concentration was used to determine the Km and Vmax values (Kang et al., 2007).

Determination of tyramine in plants by HPLC

Fresh plant materials were weighted and totally ground. 1 g of the ground sample was extracted with 25 mL of methanol in a glass tube, and all of these tubes were placed in an ultrasonic bath for 8 h (Zhou & Su, 2008). The supernatants were clarified by passing through a 0.22 µm pore size NC filter and tyramine was detected by HPLC.

To analyze the tyramine of the extracts by HPLC, the solvent composition was: mobile phase A, water with 0.3% phosphoric acid; mobile phase B, 100% acetonitrile. The eluent gradient was: 5% B for 15 min, then the eluent from 5% B to 100% B in 5 min, kept for 10 min, then from 100% B to 5% B in 5 min, kept for 10 min, 0.8 mL/min flow rate. The injection volume was 10 µL, and tyramine was detected at 275 nm. Quantification of tyramine was performed using the calibration curves.

Phylogenetic analysis

Phylogenetic tree analysis of LaTYDC1 and TYDCs proteins from different species by using Maximum likelihood method. The phylogenetic tree constructed using MEGA5.5 software comparsion of amino acid sequences of plant EgTYDC1 (Erythranthe guttata, XP_012827457.1); NnTYDC2 (Nelumbo nucifera, XP_010245171.1); DcTYDC2(Dendrobium catenatum, PKU83762.1); QsTYDC1(Quercus suber, SXP_023871760.1); AcTYDC (Aristolochia contorta, DQ986331.1); AsTYDC2 (Apostasia shenzhenica, PKA66030.1); AcTYDC2 (Ananas comosus, OAY76026.1); RcTYDC1 (Rosa chinensis, XP_024161133.1); JrTYDC1 (Juglans regiatyrosine, XP_018819733.1); LaTYDC1(Lycoris aurea, MG932082.1); NpTYDC1 (Narcissus pseudonarcissus, AUG71932.1); PbTYDC1 (Pyrus bretschneideri, XP_009377476.1); NpTYDC2 (Narcissus pseudonarcissus, AUG71933.1); RsTYDC (Rhodiola sachalinensis, ABF06560.1); OsTYDC (Oryza sativa, XP_015644906.1); PcTYDC (Petroselinum crispum, AAA33862.1); PaTYDC1(Prunus avium, XP_021812417.1); PsTYDC7 (Papaver somniferum, AF025434.1); AmTYDC (Argemone mexicana, ACJ76782.1); AtTYDC (Arabidopsis thaliana, NP_001190862.1); TcTYDC (Theobroma cacao, EOX96928.1).

Statistical analysis

Statistical analysis was performed using SPSS 13.0, and differences were analyzed with one-way ANOVA test or t-test. Statistical significance was assumed at P < 0.05.

MeJA treatment resulted in more tyramine accumulation in L. aurea plants

Secondary metabolite accumulation is tightly controlled by Jasmonate (JA) signaling (Wasternack, 2007), and our previous results show that MeJA induces the accumulation of galanthamine in Lycoris chinensis Traub (Mu et al., 2009). In this study, 100 µM MeJA were used to treat the L. aurea seedlings for 24 h, then the contents of tyramine in the roots were measured by HPLC. As the results show in Fig. 2, the seedlings accumulated 31.59 µmol g⁻¹ tyramine in the roots after 6 h of MeJA treatment, when the control sample accumulate 19.32 µmol g⁻¹ tyramine. After 12 h of MeJA treatment, tyramine content of the treated sample was 30.31 µmol g⁻¹ , almost two-folds the amount in the control samples. After 24 h, the content of tyramine in experimental seedlings decreased to 7.7 µmol g⁻¹, almost one half of the control samples.

Characterization of LaTYDC1

LaTYDC1 covers an open reading frame representing 511 amino acids with a predicted molecular weight of 56.7 kDa. Phylogenetic analysis of LaTYDC along with TYDCs form other species was carried out using maximum likelihood method. We conferred this gene of interest as LaTYDC1, which is placed near NpTYDC1 and another TYDC from L. aurea, and shares a high similarity with NpTYDC2 (Fig. 3A). In addition, the sequence alignment of these TYDCs indicated that LaTYDC1obviously shared the conserved domain with other TYDC proteins, and the deduced amino acid sequence of LaTYDC1 showed 47–89.67% shared identity with those from other plant origins (Fig. S1). LaTYDC1 expression in root, leaf and flower were 40, 10, and 15-fold greater, respectively, than that in bulb (Fig. 3B). At the beginning, six hours after 100 µM MeJA treatment, the accumulation of LaTYDC1 significantly increased almost 200 folds when compared with the vehicle control samples, and sharply decreased when compared with the expression level of samples 6 h after MeJA treatment. The similar results were observed after 24 h treatment (Fig. 3C).

To further explore the function of LaTYDC1, we transiently expressed the LaTYDC1-GFP (green fluorescent protein) fusion protein in the Arabidopsis protoplast. As the green fluorescence signal was observed at the cytoplasm, it showed that LaTYDC1 localizes at the cytoplasm (Fig. 3D).

Enzymatic characterization of LaTYDC1

To examine the enzymatic function of LaTYDC1, we expressed the LaTYDC1 protein fused with a 26.8 kDa GST tag in E. coli BL21. As shown in Fig. 4, after purification using a high affinity GST purification column, the LaTYDC1-GST fusion protein was checked by SDS-PAGE, and the fusion protein showed a predictable molecular weight (MW) of 83.5 kDa (Fig. 4A).

Assuming that LaTYDC1 has reductase activity, we were interested in testing whether the fusion protein could catalyze tyrosine into tyramine conversion. HPLC analysis was then carried out with the samples for the enzyme assay, in which the purified LaTYDC1 fusion protein uses tyrosine as substrate. The result shown in Fig. 4B demonstrated that the fusion LaTYDC1 was able to specifically catalyze the synthesis of tyramine, since the product with an identical peak in regards to retention time was the same as the corresponding tyramine standard (Fig. 4B). In addition, this product was determined by GC-MS analysis, and it was confirmed that the product is tyramine (Fig. 4C). Moreover, the optimal temperature for LaTYDC1 enzyme activity is 50 °C; when the temperature is below 30 °C or above 55 °C, enzyme activity rapidly declines (Fig. S2A). The optimum pH of LaTYDC1 is 8.5; when the pH value is above 9, enzyme activity sharply declines (Fig. S2B). Furthermore, the substrate affinity was determined by calculating the K_m we used tyrosine and 3,4-dihydroxy-L-phenylalanine (Dopa) as the substrate for LaTYDC1. The K_m for tyrosine is 602.34 ± 38 µM, and the K_m for dopa is 966.35 ± 79 µM. The Vmax is 3.19 ± 0.51 (nkat mg⁻¹) and 1.24 ± 0.095 (nkat mg⁻¹), respectively (Table 1).

Identification of LaTYDC1-ralated miRNA in L. aurea

The sudden change of LaTYDC1 expression under MeJA attracted our attention. MiRNAs play post-transcriptional regulatory roles in plants, and are involved in the regulation of secondary metabolite biosynthesis in many plants (Zhao et al., 2012). To identify the genes and miRNA that participate in the JA signal transduction pathway and are involved in galanthamine biosynthesis, we searched the response miRNA sequencing database of L. aurea (Xu et al., 2016). We identified that miR396 is responsive to MeJA treatment, and the predicted target is LaTYDC1. The cleavage site was confirmed by degradome data and is shown in Fig. 5A. qRT-PCR were used to determine the expression level of miR396 in different tissues. The results showed that the expression of miR396 was the lowest in roots, and was mainly expressed in leaves. The expression of miR396 in root, leaf and flower were respectively 2, 17, and 4-fold greater than that in the bulb (Fig. 5B). This co-expression analysis also showed that miR396 was negatively correlated with the target LaTYDC1. As miRNA paired with LaTYDC1 mRNA to direct transcriptional suppression, we did qRT-PCR to determine miR396 expression under MeJA treatment, and the expression of miR396 was repressed about one half of the vehicle sample. Twelve hours after MeJA treatment, the expression of miR396 increased ∼3 folds (Fig. 5C). Combined with the repressed expression of LaTYDC1, the reversely correlated expression relationship suggests that miR396 may target LaTYDC1. RLM-RACE experiments were carried out to confirm the cleavage, and the cleavage sites were mapped in flanking sequence of the complementary region (Fig. 5D), the DNA sequences in the RLM-RACE experiments were listed in Supplemental Information 2).