Construction of cell factory capable of efficiently converting l-tryptophan into 5-hydroxytryptamine

Background l-Tryptophan (l-Trp) derivatives such as 5-hydroxytryptophan (5-HTP) and 5-hydroxytryptamine (5-HT), N-Acetyl-5-hydroxytryptamine and melatonin are important molecules with pharmaceutical interest. Among, 5-HT is an inhibitory neurotransmitter with proven benefits for treating the symptoms of depression. At present, 5-HT depends on plant extraction and chemical synthesis, which limits its mass production and causes environmental problems. Therefore, it is necessary to develop an efficient, green and sustainable biosynthesis method to produce 5-HT. Results Here we propose a one-pot production of 5-HT from l-Trp via two enzyme cascades for the first time. First, a chassis cell that can convert l-Trp into 5-HTP was constructed by heterologous expression of tryptophan hydroxylase from Schistosoma mansoni (SmTPH) and an artificial endogenous tetrahydrobiopterin (BH4) module. Then, dopa decarboxylase from Harminia axyridis (HaDDC), which can specifically catalyse 5-HTP to 5-HT, was used for 5-HT production. The cell factory, E. coli BL21(DE3)△tnaA/BH4/HaDDC-SmTPH, which contains SmTPH and HaDDC, was constructed for 5-HT synthesis. The highest concentration of 5-HT reached 414.5 ± 1.6 mg/L (with conversion rate of 25.9 mol%) at the optimal conditions (substrate concentration,2 g/L; induced temperature, 25℃; IPTG concentration, 0.5 mM; catalysis temperature, 30℃; catalysis time, 72 h). Conclusions This protocol provided an efficient one-pot method for converting. l-Trp into 5-HT production, which opens up possibilities for the practical biosynthesis of natural 5-HT at an industrial scale. Supplementary Information The online version contains supplementary material available at 10.1186/s12934-022-01745-0.

Background l-Tryptophan (l-Trp) is an essential amino acid, and its derivatives, such as l-5-hydroxytryptophan (5-HTP), 5-hydroxytryptamine , N-acetyl-5-hydroxytryptamine and melatonin, have gained increasing attention due to their direct health and medical benefits, and their production of other valuable molecules as key biosynthetic precursors [1][2][3][4]. Among these, 5-HT is a major neurotransmitter that is naturally present in animals and plants [5]. Similar to the catecholamines, dopamine, epinephrine, and norepinephrine, 5-HT modulates the activity of the nervous system and plays a significant role in the coordination of movement and the regulation of mood. As well as scavenging harmful free radicals, 5-HT also plays roles in behaviour management, sleep cycles, appetite and liver regeneration [6].
Currently, the main supply of 5-HT depends on the extraction from mammals (e.g. buffalo, rat, and pig) or plants (safflower, sea buckthorn, and Moringa Oleifera seeds) [7]. However, the low 5-HT yield from natural extraction and the insufficient supply of raw materials

Open Access
Microbial Cell Factories *Correspondence: kqchen@njtech.edu.cn 1 State Key Laboratory of Materials-Oriented Chemical Engineering, Nanjing Tech University, Nanjing 211816, China Full list of author information is available at the end of the article result in demands that exceed the supply. With the rapid development of biotechnology, biological approaches show great potential for synthesizing natural and nonnatural molecules. Previous studies have shown that 5-HT can be produced from l-Trp in the nervous system of human and animals via two enzymatic steps: l-Trp is hydroxylated to 5-HTP by pterin-dependent tryptophan hydroxylase (TPH), then converted to 5-HT by tryptophan decarboxylase (TDC), which is one of the Aromatic amino acid decarboxylase [8]. However, the protocol has several issues, including the problems that TPH possesses poor stability and the expensive cofactor, pterin, is required [5,9]. Aromatic amino acid decarboxylases (AAADs) play important roles in the key step of 5-HTP conversion to 5-HT. Previously, tryptophan decarboxylase (TDC) was regarded as an enzyme that catalyses non-specific reactions in l-Trp or 5-HTP (prefer to l-Trp), with the result that a one-pot cascade reaction cannot be performed and has therefore never been reported [10,11] (All the methods are shown in Table 1).
Dopa decarboxylase (DDC), another AAAD, is able to convert l-dopa to dopamine [12]. In addition, previous study reported that DDC can catalyse 5-HTP into 5-HT without the ability of converting l-Trp [13,14]. However, almost studies have focused on the characterization of the enzyme involved in the biosynthesis of dopamine, and systematic studies for the production of 5-HT have not been conducted [7,15].
In this study, we proposed a one-pot enzymatic catalytic approach for converting l-Trp into 5-HT using a whole-cell factory for the first time. First, artificial endogenous cofactor tetrahydrobiopterin (BH 4 ) and tetrahydromonapterin (MH 4 ) modules were constructed and compared to catalyse l-trp into 5-HTP using a tryptophan hydroxylase (TPH) from Schistosoma mansoni with high stability and catalytic ability. Then, a DDC from Harmonia axyridis that specifically catalyses 5-HTP to produce 5-HT was introduced to the chassis cell to form the cell factory. The optimal conditions of the recombinant cell for 5-HT production were also studied.

Construction of chassis cells for 5-HTP production
To produce 5-HT from l-Trp, a chassis cell that could efficiently convert l-Trp to the intermediate product 5-HTP, needed to be constructed first. Previous studies have reported that TPH from S. mansoni (SmTPH) possesses good activity and stability compared that of other resources [16]; therefore, SmTPH was chosen for converting l-Trp into 5-HTP in this study. However, SmTPH is a cofactor-dependent enzyme and requires the expensive cofactor biopterin [9,17]. To solve this issue, synthesis and regeneration cofactor tetrahydrobiopterin (BH 4 ) and tetrahydromonapterin (MH 4 ) modules were constructed in the chassis cell containing SmTPH [1,18]. As shown in Fig. 1, the 5-HTP concentrations in the cells containing BH 4 and MH 4 increased until 60 h of fermentation, then decreased with the further increase of time. The observed decrease in 5-HTP concentration may have been due to the oxidation and degradation of 5-HTP as the fermentation time increased [5]. The highest concentrations of 5-HTP were 0.93 g/L and 0.34 g/L at 60 h with BH 4 and MH 4 , respectively. Zhang et al. reported that the production of 5-HTP from SmTPH using BH 4 as a cofactor was higher than that when using MH 4 as a cofactor in S. cerevisiae, which correlated with the findings of our study [18]. Therefore, the chassis cell containing the synthesis and regeneration cofactor BH 4 and SmTPH was used in the next experiment.

Expression and purification of dopa decarboxylase HaDDC and its catalytic specificity towards L-Trp and 5-HTP
Dopa decarboxylase from Drosophila melanogaster (drDDC) has attracted significant attention, due to its activity for both 5-HTP and l-DOPA but with no activity to l-Trp [19][20][21]. In this study, we used the DDC from Harmonia axyridis (HaDDC, GenBank: AMQ 13055.1) for 5-HT synthesis.
To evaluate the catalytic specificity of purified HaDDC, l-Trp and 5-HTP were used as substrates respectively. As shown in Fig. 2a, no 5-HT was released from l-Trp within 4 h or longer (data not shown), indicating that the recombinant HaDDC had no activity toward l-Trp. Meanwhile, as shown in the Fig. 2c, there was also no tryptamine found from l-Trp. It was clearly that 5-HT was released within 4 h by HaDDC with 5-HTP as a substrate in the Fig. 2b. These results showed that HaDDC exhibited strict substrate specificity similar to that displayed by drDDC from D. melanogaster [7,15]. According to the results shown in Table 2, HaDDC exhibited a high efficiency on 5-HTP with a yield of 88.45% at 4 h, which was much higher than that of TDC (16.04%) from Catharanthus roseus which was higher than TDC reported previously [6,11]. Therefore, HaDDC from H. axyridis was chosen for converting 5-HTP into 5-HT (Additional file 1: Figs. S1-S7, Table S1).

One-pot catalysis for 5-HT production
The enzyme cascade of purified SmTPH and HaDDC was investigated both in vitro and in vivo. The catalysis processing of l-Trp by recombinant SmTPH and HaDDC in vitro showed that 5-HT production increased along with increasing time before 48 h, then decreased with   the further increase of time (Fig. 3). Meanwhile, the concentration of l-Trp decreased continuously. The highest concentration of 5-HT was 352.4 mg/L, with a yield of 17.62%, at 48 h, which was similar to the two steps to produce 5-HT [6].  Fig. 4b, the l-Trp concentration decreased continuously from 2.00 to 0.899 g/L within 96 h. Meanwhile, the 5-HT concentration increased as time increased and reached the highest concentration (0.3136 g/L) at 84 h, which was 2.04-fold higher than that obtained from the two-stage fermentation process by CtAAAH and TDC [6]. However, the concentration of 5-HT decreased after 84 h and the fermentation broth turned dark, which was similar to the results reported previously [22]. It may because 5-HT was unstable. During the reaction process, oxidation reaction occurred and degradation products generated, resulting in the decrease of 5-HT and colour changed.
In addition, l-Trp could be also degraded by other tryptophanases in E.coli and caused the broth turned dark [22][23][24].

Optimisation of one-pot cascade in vivo
To determine the optimal conditions for 5-HT production, a number of catalysis conditions were investigated. The effect of l-Trp concentration on 5-HT production was illustrated in Fig. 5a. 5-HT concentration increased with l-Trp concentration from 0.5 to 2.0 g/L before 48 h, followed by decreasing slightly. However, as the concentration of l-Trp increased to 2.5 g/L, the 5-HT concentration lower than that from 1.0, 1.5, 2.0 g/L l-Trp. Therefore, 2.0 g/L l-Trp was chosen for subsequent experiments.
The effect of induction temperature on the production of 5-HT is shown in Fig. 5b. The production of 5-HT increased quickly as the induction temperature increased from 18 to 25 ℃, then decreased with the temperature increased from 25 to 37 ℃. Therefore, the optimal induction temperature for the production of 5-HT was 25 ℃.
The effects of concentration of IPTG on 5-HT synthesis were investigated (Fig. 5c). 5-HT concentration increased with the increase in concentration of IPTG from 0.025 to 0.05 mM, followed by a decrease at higher IPTG concentrations. The results suggested that 0.05 mM was the optimal IPTG concentration.
As shown in Fig. 5d, an increase in the concentration of 5-HT in the broth appeared at 8 h to 12 h during incubation with E. coli BL21(DE3)△tnaA/BH 4 /HaDDC-SmTPH, then decreased. Thus, the optimal time for induction was 12 h. The relationship between 5-HT concentration and catalysis temperature was presented in Fig. 5e. 5-HT production from l-Trp increased as the catalysis temperature increased from 25 to 30 ℃, and the maximum concentration of 5-HT (0.4145 ± 1.6 g/L, reduced L-Trp was 1.13 g/L) was reached at 30℃. However, with the temperature increased from 30 to 40℃, the yield of the product decreased rapidly.
Based on these results, the optimal conditions for 5-HT production of one-pot catalysis using E. coli BL21(DE3)△tnaA/BH 4 /HaDDC-SmTPH were 12 h for the expression of proteins at 25 ℃ with 0.05 mM IPTG, and the optimal conditions for catalysis were 30 ℃ for 48 h. The highest yield of 5-HT from l-trp was 25.9 mol%.

Conclusions
In this study, a cell factory capable of efficiently converting l-Trp into 5-HT was successfully constructed. First, a chassis cell that contained a screened artificial endogenous BH4 module as a cofactor and tryptophan hydroxylase from Schistosoma mansoni (SmTPH) was constructed, which could convert L-Trp into 5-HTP efficiently. The dopa decarboxylase from Harminia axyridis (HaDDC) that can specific catalyse 5-HTP to 5-HT was chosen for 5-HT production. Finally, the two enzymes were cascaded in vivo for converting L-Trp to 5-HT, which lead to a yield of 414.5 ± 1.6 mg/L at the optimal conditions. This study provides a novel and important foundation for artificial cell factory synthesis of 5-HT.

Plasmid and strain construction
E.coli Trans1T1 was used for plasmid construction and propagation, and E. coli BL21(DE3) was used for protein expression. As tryptophanase A (tnaA) gene, which encodes tryptophanase, is necessary to accelerate degrade l-Trp and 5-HTP into indole and 5-hydroxy indole in E.coli, respectively. Therefore, tnaA gene should be knocked out in E. coli BL21(DE3) to ensure the biosynthesis of 5-HTP or 5-HT, using the previously established CRISPR/Cas9 protocol [1,25]. The expression vectors pCDFDuet-1 and pET-28a ( +) were supplied by Novagen Co., Ltd.

The effect of cofactor BH 4 and MH 4 synthesis and regeneration pathways on the conversion of L-Trp into 5-HTP
The fermentation broth of E. coli BL21(DE3)△tnaA/ BH 4 /SmTPH and E. coli BL21(DE3)△tnaA/ MH 4 /SmTPH were centrifuged at 4000 rpm for 10 min, and the cells were respectively transferred to M9Y media containing 2 g/L of l-Trp. Sampling and measuring were conducted at 12-h intervals.

Purification of E. coli BL21(DE3)△tnaA/HaDDC and its catalysis specificity towards L-Trp and 5-HTP
The cells harvested from cultures were washed, resuspended in 50 mM HEPES (pH 7.0), and lysed at 4 ℃ by JY92-IIN ultrasonication (Ningbo Xinzhi Biotechnology, Ltd., Ningbo, China). The lysates were centrifuged at 12,000 rpm at 4 ℃ for 20 min, and the supernatant was used as a crude enzyme solution. The recombinant HaDDC was purified using a fast protein liquid chromatography (FPLC) system (GE AKTA Pure 150; General Electric Co., Fairfield, America) with a Ni-nitrilotriacetic acid affinity chromatography (Ni-NTA) column (His Trap ™ FF5 mL) according to the manufacturer's instructions. Protein concentrations were determined at 595 nm using the Bradford method with bovine serum albumin as the standard [28]. All protein samples were analysed by reductive SDS-PAGE with 20 mM β-Mercapto ethanol incubation. A premixed protein marker (Takara Biotechnology Co., Ltd., Nanjing, China) containing 180-, 140-, 100-, 75-, 60-and 45-kDa proteins was used as the molecular mass standard. The method of purification for the recombinant protein SmTPH was same as that for HaDDC.
Optimisation of the production of 5-HT from L-Trp using E.coli BL21(DE3)△tnaA /BH 4 /HaDDC-SmTPH whole-cell factory Cells were transferred to 100 mL of M9Y medium in a 500 mL flask and shaken at 200 rpm in various conditions. In this study, single-factor experiments using five predominant factors (substrate concentration, IPTG, induced time, induced temperature and catalysis temperature) were conducted to investigate the effects of these factors on the production of 5-HT [29]. Investigation of the effect of each factor on the production of 5-HT was carried out based on changes only to that factor, while keeping other variables constant.
The investigation was carried out at varying concentrations of L-Trp (0.5, 1.0,1.

Analytical method
The cell growth density at 600 nm (OD 600 ) was measured using an UV1000D ultraviolet-visible spectrophotometer (AOE Instruments, Shanghai Co., Ltd). High-performance liquid chromatographic (HPLC) analysis of the L-Trp, 5-HTP and 5-HT contents was performed on an Agilent 1260 series LC system (Agilent Technologies, Santa Clara, CA, USA) with an ultraviolet detector reading at 276 nm [1]. The samples were filtered through Millex-LG filter units (Millipore, Billerica, MA, USA) prior to HPLC analysis. Separation of samples was achieved using a reverse phase Agilent TC-C18 column (5 × 4.6 mm × 250 mm; Agilent) with a constant flow rate of 1 mL/min at 25 ℃ and an injection volume of 10 μL. The mobile phase consisted of methanol (and potassium phosphate buffer (10 mM, pH 6.5). The substrate and product concentration were measured by HPLC against the l-Trp, 5-HTP and 5-HT standards using a calibration curve. Tryptamine standard was eluted via column chromatography (TC18, 5 × 4.6 mm × 250 mm; Agilent) in the following gradients: methanol (solvent A) and 0.1% formic acid in water (solvent B): 0-15 min 10% A; 15-16 min, 55% A; 16-22 min, 100% A; 22-26 min, 10% A; the flow rate was 0.8 mL/min, and the reaction absorption was recorded at 280 nm.
• fast, convenient online submission

Learn more biomedcentral.com/submissions
Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: