A method for screening high-activity tryptophanase based on random forest and its application

Abstract

The present invention discloses a method and application for screening highly active tryptophanase based on random forests. The present invention efficiently obtains potential highly active tryptophanase encoding genes based on the random forest method, obtains cells containing recombinant tryptophanase through heterologous expression, and applies them to the splitting of DL-tryptophan to efficiently prepare high-purity D-tryptophan. This solves the problems of low flux, high cost, large error, poor accuracy, and complex process in existing tryptophanase screening methods. The tryptophanase gene from Morganella morganii screened by the present invention is heterologously expressed, and the resulting recombinant strain is capable of more efficiently expressing tryptophanase. Whole cells containing the recombinant tryptophanase are used for the splitting of DL-tryptophan. The hydrolysis rate of L-tryptophan reaches 99.36%, and the ee value of the obtained product D-tryptophan reaches 99.23%.

Description

(1) Technical field

[0001] The present invention belongs to the field of biotechnology, and particularly relates to a method for screening highly active tryptophanase based on random forests, and application of the screened tryptophanase in splitting DL-tryptophan to prepare high-purity D-tryptophan. (2) Background technology

[0002] Tryptophanase (TPase, EC 4.1.99.1), also known as tryptophan indole lyase, is a pyridoxal phosphate-dependent hydrolase that hydrolyzes L-tryptophan into indole, pyruvate, and ammonia via an α-elimination mechanism. Its catalytic mechanism primarily involves the following steps: First, a lysine residue in the active site of the enzyme protein binds to the coenzyme pyridoxal phosphate to form an internal aldimine (I); in the presence of the substrate L-tryptophan, this is converted to an external aldimine (III) via a geminal diamine structure (II); α-C deprotonation produces a quinone intermediate (IV); The synergistic effects of indole protonation and C-C bond cleavage result in the loss of the side chain indole to form an α-aminoacrylate intermediate (VI); the α-aminoacrylate intermediate is further attacked by the ε-amino group of the lysine residue bound to the coenzyme, losing pyruvate and ammonia to form the holoenzyme. Based on this principle, tryptophanase has been successfully applied to the resolution of racemic DL-tryptophan, enabling the biocatalytic preparation of D-tryptophan. D-tryptophan is a non-protein-active amino acid with unique physiological properties. It can be used as a non-nutritive sweetener, feed additive, and plant growth enhancer in the food, feed, and agricultural industries. In the pharmaceutical industry, D-tryptophan is primarily used in the synthesis of peptides to extend the half-life of peptide drugs and reduce their side effects. Its derivative, D-tryptophan methyl ester hydrochloride, is also an important precursor for the production of oncolytic drugs. To date, methods for preparing D-tryptophan include chemical and enzymatic methods. Chemical methods suffer from complex procedures, low yields, and environmental pollution. In comparison, enzymatic synthesis offers simplicity, high specificity, mild conditions, and environmental friendliness, making it more suitable for large-scale industrial production. However, screening methods for tryptophanase, used in enzymatic preparation techniques, suffer from low throughput, high cost, large errors, poor accuracy, and complex processes.

[0003] To date, tryptophanases have been found in most Gram-negative bacteria, a few Gram-positive bacteria, and fungi. However, the diversity of microorganisms and the randomness and limitations of traditional screening methods have resulted in low efficiency in identifying novel, highly efficient tryptophanase genes. Recent advances in computer technology have opened up new possibilities for deep mining of microbial genetic resources. Among them, random forests (RFs) are an ensemble learning algorithm widely used in classification and regression tasks. They improve decision accuracy by constructing multiple decision trees and combining their predictions. In medicine, RFs have been widely applied to disease subtype prediction, biomarker discovery, drug target identification, gene function prediction, and gene expression data analysis.

[0004] Implementing the random forest strategy requires constructing a random forest model. The first step is to determine the number of trees in the forest (n_estimators). If the number of trees is too small, underfitting is likely, while if it is too large, the computational effort increases exponentially. Furthermore, once the number of trees reaches a threshold, increasing the number of trees further does not significantly improve the model. The second step is to perform self-sampling on the data, repeatedly randomly extracting samples from the sample with replacement to train each tree. This increases data diversity and allows each tree to learn from the data from different perspectives. (Random forests are so called because they incorporate randomness into their construction. Self-sampling data and constructing decision trees based on the new dataset are manifestations of this randomness.) The third step is to construct a decision tree based on the new dataset.

[0005] When selecting highly active tryptophanase, not only the sample source but also the number of amino acids and conserved regions must be considered. Selecting microbial samples from different sources will lead to different results in the selection of RF sample sets, as well as the selection of training and test sets. Therefore, how to quickly and cost-effectively screen for highly active tryptophanase becomes a challenge that needs to be addressed. (3) Summary of the invention

[0006] The present invention aims to provide a method and application for screening highly active tryptophanase based on random forests. The present invention uses the random forest method to efficiently obtain potential highly active tryptophanase encoding genes, obtains cells containing recombinant tryptophanase through heterologous expression, and applies the cells to DL-tryptophan splitting to efficiently prepare high-purity D-tryptophan. This method solves the problems of low throughput, high cost, large error, poor accuracy, and complex process in existing tryptophanase screening methods.

[0007] The technical solution adopted in the present invention is:

[0008] The present invention provides a method for screening high-activity tryptophanase based on random forest, which comprises the following steps:

[0009] (1) Data preparation:

[0010] Search the amino acid sequences of tryptophanase from bacteria from NCBI and randomly select 100 tryptophanase amino acid sequences;

[0011] (2) Feature setting:

[0012] a. Protein size: The candidate protein is >400 amino acids long;

[0013] b. Two essential conserved sequences for tryptophanase (i.e., only an enzyme containing both sequences is considered a tryptophanase): the first segment is KKD, where the second K is a necessary site for binding to the coenzyme (pyridoxal phosphate); the second segment is any one of the seven conserved regions TDS, TYT, TYE, RHXT, QTHXD, XRD, or EXXR (where X is any amino acid);

[0014] (3) Decision tree splitting:

[0015] Step (1) randomly selects 100 tryptophanase amino acid sequences and uses the scikit-learn learning toolkit in PyCharm to simulate. During the decision tree formation process, each node is split according to the characteristics of step (2) until it can no longer be split;

[0016] Parameter settings of the learning toolkit scikit-learn: the number of estimators (n_estimators) is set to 10, the splitting criterion (gini) is selected, the maximum depth (max_depth) is set to None, the maximum number of features (max_features) is set to square root (sqrt), the random state (random_state) is set to None, the bootstrap sampling (bootstrap) is set to True, and the number of jobs (n_job) is set to None;

[0017] (4) The amino acid sequence obtained in step (3) was compared with the tryptophanase (ACE63255.1) derived from Escherichia coli K12 to select the amino acid sequence with the highest homology.

[0018] The present invention provides a tryptophanase screened by the method, wherein the amino acid sequence of the tryptophanase is shown in SEQ ID NO. 1. The nucleotide sequence of the edited gene is shown in SEQ ID NO. 2.

[0019] The present invention also provides an application of the tryptophanase screened by the method in splitting DL-tryptophan. The application method comprises the following steps: using wet bacteria produced by induced expression of tryptophan-producing recombinant genetically engineered bacteria as a catalyst, DL-tryptophan as a substrate, and a pH 6-8 buffer as a reaction medium to form a conversion system, and carrying out the reaction at 30-50° C. and 100-300 rpm (preferably 200 rpm) to obtain a reaction solution containing D-tryptophan.

[0020] Furthermore, in the transformation system, the wet bacteria are added at a concentration of 1-21 g / L (preferably 16 g / L); and the substrate is added at a concentration of 1-10 g / L (preferably 4 g / L).

[0021] Furthermore, the conversion system uses pyridoxal phosphate as an auxiliary agent, and the amount of pyridoxal phosphate added is 1-10 mM (preferably 6 mM).

[0022] Furthermore, the reaction medium is preferably 0.12 M potassium phosphate buffer at pH 7.5.

[0023] Furthermore, the tryptophan recombinant genetic engineering bacteria were constructed as follows:

[0024] (1) Using the genome of Morganella CICC 22599 as a template, the TPase target gene fragment was cloned by PCR with the help of upstream and downstream primers;

[0025] Upstream:

[0026] 5'-CAGTGGTGGTGGTGGTGGTG CTCGAG TTACATGACAGGTTTCAGGCGGG-3'(XhoⅠ);

[0027] Downstream:

[0028] 5'-AGCAAATGGGTCGCGGATCC GAATTC ATGAAACGTATTCCAGAACCGTT CCG-3'(EcoRⅠ);

[0029] (2) The plasmid pET-28a(+) and the target gene fragment were digested with XhoⅠ and EcoRⅠ, respectively, and seamlessly cloned and connected using a kit to obtain the recombinant expression vector pET28a(+)-TPase;

[0030] (3) The recombinant expression vector pET28a(+)-TPase was heat-shock transformed into Escherichia coli DH5α competent cells, the plasmid was extracted, and then transferred into Escherichia coli BL21 competent cells to construct Escherichia coli BL21-pET28a(+)-TPase.

[0031] Furthermore, the wet cells were prepared as follows: a single colony of Escherichia coli BL21-pET28a(+)-TPase was picked up in LB liquid medium containing 50 mg / L Kan resistance, and cultured overnight at 37°C and 200 rpm; the inoculum was transferred to a new LB liquid medium at a volume concentration of 1%, and cultured at 37°C and 200 rpm on a shaker until the OD 600 The value was 0.6, IPTG was added with a final concentration of 0.1 mM, and cultured at 25 ° C and 200 rpm for 12 h. The induced bacterial liquid was centrifuged at 8000 rpm and 4 ° C for 10 min to collect the wet bacteria.

[0032] Compared with the prior art, the beneficial effects of the present invention are mainly reflected in:

[0033] (1) The present invention provides a new tryptophanase screening method for splitting DL-tryptophan, which improves the efficiency and accuracy of tryptophanase screening, provides a reference for the screening and preparation of other enzymes, and overcomes the problems of existing methods based on high-throughput fluorescence screening or high-throughput screening based on biosensors, such as high difficulty in experiments, expensive instruments, and long experimental cycles.

[0034] (2) The tryptophanase gene from Morganella morganii screened by the present invention was heterologously expressed, and the resulting recombinant strain was able to express tryptophanase more efficiently. Whole cells containing the recombinant tryptophanase were used to split DL-tryptophan, and the hydrolysis rate of L-tryptophan reached 99.36%, and the ee value of the obtained product D-tryptophan reached 99.23%. (IV) Description of the accompanying drawings

[0035] Figure 1 , random forest output results.

[0036] Figure 2 , tryptophanase amino acid sequence alignment results; 1: Escherichia coli K12; 2: Morganella; 3: Photorhabdus; 4: Citrobacter cosei; 5: Dickie's bacillus; 6: Klebsiella; 7: Yersinia enterocolitica; 8: Enterobacter sakazakii; 9: Xenorhabdus burnetii; 10: Shewanella; 11: Achromobacter; 12: Aeromonas hydrophila; 13: Eilishella; 14: Limnocodia.

[0037] Figure 3 , Gel electrophoresis of the tryptophanase PCR amplification product in Example 2; M: Marker; 1: tnaA.

[0038] Figure 4, Agarose gel electrophoresis diagram of the plasmid after double enzyme digestion (XhoⅠ and EcoRI) during the construction of the recombinant engineering bacteria for heterologous expression of tryptophanase in Example 3; M: Marker; 1: DH5α-pET 28a(+)-TPase; 2: BL21-pET 28a(+)-TPase; 3: tnaA.

[0039] Figure 5 , Example 4 Tryptophanase protein SDS-PAGE results; M: Marker; 1: supernatant; 2: precipitate; 3: pure enzyme TPase.

[0040] Figure 6 , Example 4 Tryptophanase protein Native-PAGE result diagram; M: Marker; 1: pure enzyme TPase.

[0041] Figure 7 , diagram of the catalytic mechanism of tryptophanase.

[0042] Figure 8 , Optimization of DL-tryptophanase conversion conditions in Example 5; 1: Effect of reaction time on TPase reaction; 2: Effect of bacterial concentration on TPase reaction; 3: Effect of reaction temperature on TPase reaction; 4: Effect of pyridoxal phosphate concentration on TPase reaction.

[0043] Figure 9 , HPLC chart of the conversion liquid in step 5 in Example 5; 1: conversion liquid under the optimal conditions of Example 5; 2: DL-tryptophan standard 4 g / L; 3: L-tryptophan standard 2 g / L; 4: D-tryptophan standard 2 g / L. (V) Specific implementation methods

[0044] The present invention is further described below with reference to specific embodiments, but the protection scope of the present invention is not limited thereto:

[0045] The culture medium used in the embodiments of the present invention is as follows:

[0046] NB medium: 0.5 g peptone, 0.3 g beef extract, 0.5 g sodium chloride, 100 mL distilled water, pH 7.0.

[0047] LB medium: peptone 1.0 g, sodium chloride 1.0 g, yeast extract 0.5 g, distilled water 100 mL.

[0048] Example 1: Screening for high-activity tryptophanase based on random forest

[0049] 1. Data preparation:

[0050] Tryptophanase amino acid sequences from bacteria were searched from NCBI, and 100 tryptophanase amino acid sequences were randomly selected.

[0051] 2. Feature settings:

[0052] a. Protein size: The candidate protein is >400 amino acids long;

[0053] b. Two essential conserved sequences for tryptophanase (i.e., only an enzyme containing both sequences is considered a tryptophanase): the first segment is KKD, where the second K is the necessary site for binding to the coenzyme (pyridoxal phosphate); the second segment is any one of the seven conserved regions TDS, TYT, TYE, RHXT, QTHXD, XRD, or EXXR (where X is any amino acid);

[0054] 3. Decision tree splitting:

[0055] https: / / scikit-learn.org / stable / modules / generated / sklearn.ensemble.RandomForestClassifier.html. We used the scikit-learn learning toolkit in PyCharm for simulation. The parameter settings are shown in Table 1. During the decision tree formation process, each node was split according to the features in step 2 until it could not be split any further.

[0056] Table 1 Random Forest Parameters

[0057]

[0058]

[0059] 4. Build a decision tree according to steps 1 to 3 to form a random forest. Put the 100 amino acid sequences randomly selected in step 1 into the random forest and let each decision tree in the random forest make judgments and classifications. The screening results are shown in Figure 1 : 98 amino acid sequences were found that met feature a (protein length > 400 amino acids), and 2 amino acid sequences did not meet feature b. 98 sequences were found that met feature b, and the sequences with a number of occurrences ≥ 3 were selected. Among them, 13 sequences had a number of occurrences ≥ 3, and 85 sequences had a number of occurrences < 3 (as shown in Table 2).

[0060] 5. The 13 amino acid sequences obtained in step 4 were compared with the tryptophanase reported to be derived from Escherichia coli K12 (ACE63255.1, Deeley MC, Yanofsky C. Nucleotide sequence of the structural gene for tryptophanase of Escherichia coli K-12. [J]. Journal of Bacteriology, 1981, 147 (3): 787-96.) (the results are shown in Figure 5). Figure 2 As shown). According to the amino acid sequence comparison, it was found that the tryptophanase from Morganella had the highest homology, with a homology of 58.32% (as shown in Table 2), and its amino acid sequence is SEQ ID NO.1 (Gene ID: WP_015422483.1). The genus Morganella currently consists of a single species (Morganella morganii) with two subspecies, namely Morganella morganii and Morganella serrata (Liu H, Zhu J, Hu Q, et al. Morganella morganii, a non-negligent opportunistic pathogen [J]. International Journal of Infectious Diseases, 2016, 50 (C): 10-17.) Morganella morganii with the strain number CICC 22599 was selected through the China Industrial Microbiological Culture Collection Administration Center. To verify whether Morganella morganii contains tryptophanase activity, it was first cultured in NB medium at 37°C for 12 h, centrifuged, and 0.1 g of wet cells was added to 10 mL of 0.12 mM potassium phosphate buffer, pH 7.5, containing DL-tryptophan (0.02 M) and pyridoxal phosphate (0.06 mM). The reaction was incubated at 37°C overnight. 5% p-dimethylaminobenzaldehyde was added to the resulting fermentation broth, and the solution changed color, indicating that Morganella morganii did contain the tryptophanase-encoding gene and that the encoded tryptophanase had the activity to split DL-tryptophan.

[0061] Table 2 Output results and homology comparison results

[0062]

[0063]

[0064] Example 2: Primer design and target gene amplification

[0065] 1. Primer design

[0066] According to the gene sequence of tryptophanase from Morganella morganii CICC 22599 screened in Example 1, SEQ ID NO. 2 (Gene ID: 69679949), primers were designed by snap gene. The primer details are as follows:

[0067] Upstream:

[0068] 5'-CAGTGGTGGTGGTGGTGGTG CTCGAG TTACATGACAGGTTTCAGGCGGG-3'(XhoⅠ);

[0069] Downstream:

[0070] 5'-AGCAAATGGGTCGCGGATCC GAATTC ATGAACGTATTCCAGAACCGTT CCG-3'(EcoRⅠ).

[0071] 2. Extraction of Morganella morganii genomic DNA

[0072] The genomic DNA of Morganella morganii CICC 22599 was extracted using a DNA extraction kit.

[0073] 3. PCR cloning of tryptophanase gene

[0074] Using the genomic DNA extracted in step 2 as a template, the tryptophanase gene was cloned by PCR using the primers designed in step 1.

[0075] PCR reaction system (total volume 50 μL): 2 μL each of upstream and downstream primers (10 μM), 2 μL of Morganella morganii genomic template, 25 μL of DNA polymerase premix (Primer Star Max Mix), and 19 μL of nuclease-free water.

[0076] PCR reaction program: preliminary denaturation: 95 °C for 5 min, complete denaturation at 95 °C for 15 s, annealing: 55 °C for 15 s, extension: 72 °C for 90 s, 30 cycles, re-extension: 72 °C for 5 min, reduce to 4 °C for insulation.

[0077] The PCR amplification product (tnaA) was detected by 1% agarose gel electrophoresis. Figure 3 As shown, the order of loading from left to right is the marker (100-5000 bp) and the PCR amplification product (tnaA) recovered from gel excision. This product is consistent with the size of the tryptophanase gene (tnaA, approximately 1398 bp) in the GENBANK database, confirming the identification of the tryptophanase gene (tnaA). The gene fragment was extracted and sequenced to determine the nucleotide sequence shown in SEQ ID NO. 2.

[0078] Example 3: Construction of recombinant engineering bacteria expressing heterologous tryptophanase

[0079] 1. The PCR product of Example 2 was recovered using an agarose gel recovery kit, and the insert fragment (tnaA) was extracted using the SanPrep column plasmid DNA small amount extraction kit method (see gel electrophoresis). Figure 4 middle lane 3).

[0080] 2. Digest the pET-28a(+) expression vector with XhoⅠ and EcoRI, and use the SanPrep column-based PCR product purification kit to recover the linearized plasmid.

[0081] 3. The pET-28a(+) linearized plasmid and the insert fragment (tnaA) were seamlessly cloned according to the SeamLess cloning MasterMix kit method. The reaction system used is shown in Table 3.

[0082] 4. Transform the ligation product into E. coli DH5α competent cells to construct DH5α-pET 28a(+)-TPase cloning strain. Extract the plasmid and perform double enzyme digestion with XhoⅠ and EcoRI on gel electrophoresis. Figure 4 Lane 1 was used for sequencing to verify the accuracy of the gene. The pET28a(+)-TPase plasmid was extracted and transformed into E. coli BL21 competent cells to construct E. coli BL21-pET28a(+)-TPase tryptophanase heterologous expression engineering bacteria. The plasmid was extracted and double-digested with XhoⅠ and EcoRI by gel electrophoresis. Figure 4 Lane 2 was cloned and sequenced to confirm that the engineered bacteria were successfully constructed.

[0083] Table 3 Seamless cloning reaction system

[0084]

[0085] Example 4: Heterologous expression of recombinant tryptophanase and enzyme activity determination

[0086] 1. Pick a single colony of E. coli BL21-pET 28a(+)-TPase from the plate and culture it in 50 mL LB liquid medium containing Kan resistance (50 mg / L) at 37°C with shaking at 200 rpm overnight.

[0087] 2. Transfer the inoculum to a new 100 mL LB liquid medium at a volume concentration of 1% and culture at 37°C and 200 rpm in a shaking incubator until the OD 600 The value was 0.6, IPTG was added to a final concentration of 0.1 mM, and the culture was induced at 25°C and 200 rpm for 12 h.

[0088] 3. Cell disruption: Centrifuge the induced bacterial solution at 8000 rpm at 4°C for 10 min. Remove the supernatant and resuspend the wet cells in 0.1 mM sodium phosphate buffer, pH 8. Ultrasonicate at 360W for 20 min on ice, with 2-s intervals between 2s and 2-s intervals. Centrifuge the ultrasonic cell disruption solution at 8000 rpm at 4°C for 10 min to obtain the supernatant (i.e., crude enzyme solution) and the precipitate.

[0089] 4. Extraction of pure enzyme: The nickel column was balanced with a binding buffer (20 mM sodium phosphate buffer containing a final concentration of 0.5 M NaCl) containing a final concentration of 20 mM imidazole (pH = 7.4). The crude enzyme solution was passed through a 0.22 μm aqueous filter membrane and added to the nickel column for equilibration for 30 to 60 min. The TPase was then gradient eluted with a binding buffer containing a final concentration of 50 mM imidazole (pH = 7.4) for 20 column volumes, followed by gradient elution with a binding buffer containing a final concentration of 150 mM imidazole (pH = 7.4) for 10 column volumes, and then gradient elution with a binding buffer containing a final concentration of 250 mM imidazole (pH = 7.4) for 10 column volumes and the eluate was collected. The collected eluate was filtered through an ultrafiltration membrane with a molecular weight cutoff of 50 KDa, and the concentrate was collected to obtain pure TPase enzyme.

[0090] 5. Detect the expression of target protein by SDS-PAGE gel electrophoresis: Prepare 10% SDS-PAGE gel, take 18μL of sample (supernatant, precipitate from step 3, pure enzyme from step 4), mix with 2μL of 10× Loading Buffer, boil at 100℃ for 5min, and take 15μL for sample loading. When the protein is in the stacking gel, the voltage is 80V, and when it is in the separation gel, the voltage is 120V. After the electrophoresis is completed, remove the stacking gel part, stain with Coomassie Brilliant Blue for 30min, and then decolorize until the bands are clear. The protein electrophoresis results are as follows Figure 5 As shown, the protein was expressed soluble and the size of the tryptophanase subunit was determined to be approximately 52 kDa.

[0091] 6. Native-PAGE gel electrophoresis: The operation steps are basically the same as those of SDS-PAGE. The difference is that when making the Native-PAGE gel, no denaturants such as SDS and mercaptoethanol are added. The gel is placed on ice and run at a constant voltage of 100V for 20 minutes, followed by 160V for 80 minutes. The electrophoresis results are as follows: Figure 6 The size of tryptophanase was determined to be approximately 312 KDa. Combined with the SDS-PAGE results, it was confirmed that the tryptophanase obtained in the present invention was a homohexamer.

[0092] 7. Determination of enzyme activity: Add 0.1 g of wet cells to 10 mL of 0.12 mM (pH 7.5) potassium phosphate buffer, then add 0.2 M reduced glutathione, 0.06 mM pyridoxal phosphate, 0.25 mg / mL bovine serum albumin, and 0.02 M DL-tryptophan at a final concentration. After reacting at 37°C for 1 h, add 10 mL of toluene, and then add 100 mL of an aqueous solution containing 5% by volume p-dimethylaminobenzaldehyde and 5% by volume n-butyl sulfate as a mixed colorimetric solution to terminate the reaction and form a rose-red color of varying shades. After 30 min, quantitatively determine the absorbance at 570 nm (tryptophanase decomposes L-tryptophan into indole, pyruvate, and ammonia. Indole can react with p-dimethylaminobenzaldehyde to form a deep red indole derivative by Ehrlich reagent).

[0093] Definition of enzyme activity: Under certain reaction conditions, at 37°C, the amount of enzyme required to catalyze the formation of 0.01 μmol of indole per minute is 1 U.

[0094] The enzyme activity of the wet cells of Morganella morganii CICC 22599 prepared using steps 1-3 was 0.75 U / mg, and the enzyme activity of the wet cells of Escherichia coli BL21-pET 28a(+)-TPase was 1.58 U / mg.

[0095] Example 5: Optimization of catalytic conditions for whole-cell genetically engineered bacteria containing tryptophanase TPase

[0096] Reference Figure 7 The catalytic mechanism of tryptophanase was investigated by screening the following conditions for the splitting of DL-tryptophan by E. coli BL21-pET 28a(+)-TPase:

[0097] 1. Reaction time

[0098] The volume of the reaction system is 10 mL: DL-tryptophan is dissolved in 0.12 mM, pH = 7.5 potassium phosphate buffer to a final concentration of 4 g / L, and 16 g / L of Escherichia coli BL21-pET 28a(+)-TPase wet bacteria prepared by the method of Example 4 and 6 mM pyridoxal phosphate are added in sequence to form a 10 mL reaction system, and the reaction is carried out at 35°C and 200 rpm for 2 to 10 hours (2, 4, 6, 8, 10 hours). After the reaction is completed, the transformation liquid is centrifuged at 8000 rpm for 10 minutes, the cells are removed, and the supernatant is ice-bathed for 15 minutes. The peak areas of D-tryptophan (D-Trp) and L-tryptophan (L-Trp) before and after transformation (substrate is added, enzyme has not yet been added) and after transformation are determined by HPLC, and the corresponding contents are calculated based on the standard curve of peak area and concentration of D-tryptophan and L-tryptophan under the same conditions. The results are as follows Figure 8 As shown in Figure 1, the optimal reaction time is 8 hours.

[0099] HPLC detection conditions: chromatographic column: InertSustain C18; mobile phase: aqueous phase (0.375 mM L-phenylalanine and 0.075 mM copper sulfate): organic phase (methanol) = 76:24, v / v; flow rate: 1 mL / min; temperature: 25°C; injection volume: 20 μL.

[0100] Conversion rate = (amount of L-tryptophan substance consumed / amount of L-tryptophan substance added) * 100%;

[0101] ee=[(A D -A L ) / (A D +A L )]*100%, A D : Peak area of ​​D-Trp, A L : Peak area of ​​L-Trp

[0102] 2. Bacteria concentration

[0103] The wet cell concentration in the reaction system of step 1 was changed to 1, 6, 11, 16, 21 g / L, the reaction time was changed to 8 h, and other operations were the same. The results are shown in the figure. Figure 8 The optimal bacterial concentration is 16 g / L.

[0104] 3. Reaction temperature

[0105] The reaction time in step 1 was changed to 8 h, the reaction temperature was 30, 35, 40, 45, 50 ° C, and other operations were the same. The results are shown in the figure. Figure 8 In 3, the optimal reaction temperature is 35℃.

[0106] 4. Pyridoxal phosphate concentration

[0107] The reaction time in step 1 was changed to 8 h, the concentration of pyridoxal phosphate was changed to 0, 2, 4, 6, 8, 10 mM, and other operations were the same. The results are shown in the table. Figure 8 In 4, the optimal pyridoxal phosphate concentration was 6 mM.

[0108] 5. Conversion process under optimal conditions

[0109] The volume of the reaction system is 10 mL: DL-tryptophan is dissolved in 0.12 mM, pH = 7.5 potassium phosphate buffer to a final concentration of 4 g / L, and 16 g / L of Escherichia coli BL21-pET 28a(+)-TPase wet cells prepared by the method of Example 4 and 6 mM pyridoxal phosphate are added in sequence to form a 10 mL reaction system, and the reaction is carried out at 35°C and 200 rpm for 8 hours. After the reaction is completed, the transformation liquid is centrifuged at 8000 rpm for 10 minutes, and after removing the cell precipitate, the transformation liquid supernatant is ice-bathed for 15 minutes. The peak areas of D-tryptophan and L-tryptophan in the supernatant are determined by HPLC, and the corresponding contents are calculated based on the standard curve of peak area and concentration of D-tryptophan and L-tryptophan under the same conditions. Using 4 g / L DL-tryptophan standard aqueous solution, 2 g / L L-tryptophan standard aqueous solution, and 2 g / L D-tryptophan standard aqueous solution as detection controls, the tryptophanase conversion rate and D-tryptophan ee value are calculated, and the results are as follows: Figure 9 As shown, the results showed that the tryptophanase conversion rate reached 99.36% and the ee value of the product reached 99.23%.

Claims

1. A method for screening high-activity tryptophanase based on random forest, characterized in that: The method comprises the following steps: (1) Data preparation: Search the amino acid sequences of tryptophanase from bacteria from NCBI and randomly select 100 tryptophanase amino acid sequences; (2) Feature setting: a. Protein size: The candidate protein is >400 amino acids long; b. Two essential conserved sequences of tryptophanase: the first segment is KKD, where the second K is the necessary site for binding to the coenzyme pyridoxal phosphate; the second segment is any one of the seven conserved regions TDS, TYT, TYE, RHXT, QTHXD, XRD, and EXXR, where X is any amino acid; (3) Decision tree splitting: Step (1) Randomly select 100 tryptophanase amino acid sequences and use the learning toolkit scikit-learn in PyCharm to simulate. During the decision tree formation process, each node is split according to the characteristics of step (2) until it can no longer be split; Parameter settings for the scikit-learn learning toolkit: set the number of estimators to 10, select Gini impurity as the splitting criterion, set the maximum depth to None, set the maximum number of features to square root, set the random state to None, set the bootstrap sampling to True, and set the number of jobs to None; (4) Align the amino acid sequence obtained in step (3) with the amino acid sequence of tryptophanase ACE63255.1 derived from Escherichia coli K12 to select the amino acid sequence with the highest homology.

2. Use of the tryptophanase screened by the method of claim 1 in the separation of DL-tryptophan to prepare high-purity D-tryptophan.

3. The use according to claim 2, characterized in that The application method comprises the following steps: using wet cells of recombinant genetically engineered bacteria expressing tryptophan through induction as a catalyst, DL-tryptophan as a substrate, and a pH 6-8 buffer as a reaction medium to form a conversion system, and performing the reaction at 30-50° C. and 100-300 rpm to obtain a reaction solution containing D-tryptophan.

4. The use according to claim 3, characterized in that In the transformation system, the wet bacteria are added at a concentration of 1-21 g / L; and the substrate is added at a concentration of 1-10 g / L.

5. The use according to claim 3, characterized in that The transformation system uses pyridoxal phosphate as an auxiliary agent, and the addition amount of the pyridoxal phosphate is 1-10 mM.

6. The use according to claim 3, characterized in that The wet bacteria were prepared as follows: a single colony of Escherichia coli BL21-pET28a(+)-TPase was picked and placed in an LB liquid culture medium containing 50 mg / L of Kan resistance, and cultured overnight at 37°C and 200 rpm; the inoculum was transferred to a new LB liquid culture medium at a volume concentration of 1%, and cultured on a shaker at 37°C and 200 rpm until the OD600 value was 0.6, IPTG was added at a final concentration of 0.1 mM, and cultured on a shaker at 25°C and 200 rpm for 12 hours. The induced bacterial liquid was centrifuged at 8000 rpm and 4°C for 10 minutes to collect the wet bacteria.

Images