Machine learning gene mining method and phosphinothricin dehydrogenase mutant for amino translocation
11781117 · 2023-10-10
Assignee
Inventors
- Yaping XUE (Hangzhou, CN)
- Feng CHENG (Hangzhou, CN)
- Dongyang Wu (Hangzhou, CN)
- Shuping ZOU (Hangzhou, CN)
- Jianmiao XU (Hangzhou, CN)
- Yuguo Zheng (Hangzhou, CN)
Cpc classification
C12Y104/01003
CHEMISTRY; METALLURGY
C12Y104/01004
CHEMISTRY; METALLURGY
International classification
Abstract
Disclosed are a machine learning gene mining method and a phosphinothricin dehydrogenase mutant for amino translocation. The phosphinothricin dehydrogenase mutant for amino translocation is obtained by mutation of a wild-type phosphinothricin dehydrogenase with an amino acid sequence as shown in SEQ ID No.2 at one of the following sites: (1) E263D-K134R-H96A-R290V; (2) E263D-K134R-H96A; (3) E263D-K134R; (4) E263D; (5) E263N; (6) E263C; and (7) E263G. The present invention utilizes the site-saturation mutagenesis technology to mutate a phosphinothricin dehydrogenase gene as shown in SEQ ID No. 1, finds that the 263rd, 134th, 290th and 290th positions are the key sites affecting enzyme activity and stereoselectivity, and obtains a mutant with enzyme activity and ee value much higher than those of the parent phosphinothricin dehydrogenase.
Claims
1. A phosphinothricin dehydrogenase mutant for amino translocation with the amino acid sequence of wild-type phosphinothricin dehydrogenase from Pseudomonas hunanensis as shown in SEQ ID NO: 2 except having one of the following sites of mutation: (1) E263D-K134R-H96A-R290V; (2) E263D-K134R-H96A; (3) E263D-K134R; (4) E263D; (5) E263N; (6) E263C; or (7) E263G.
2. A gene encoding the phosphinothricin dehydrogenase mutant for amino translocation according to claim 1.
3. A genetically recombinant bacterium comprising a host cell and a target gene transformed into the host cell, wherein the target gene comprises the gene according to claim 2.
4. The genetically recombinant bacterium according to claim 3, wherein the genetically recombinant bacterium further comprises a gene encoding a glucose dehydrogenase transformed into the host cell.
5. The phosphinothricin dehydrogenase mutant for amino translocation according to claim 1, wherein the site of mutation is: E263D-K134R-H96A-R290V.
6. A method for preparing L-phosphinothricin, by reacting 2-carbonyl-4-(hydroxymethylphosphono)butyric acid as a substrate as catalyzed by a catalyst in the presence of an inorganic amino donor, a coenzyme circulation system and a corresponding co-substrate of the coenzyme circulation system to obtain L-phosphinothricin; wherein the catalyst is one of: (1) the phosphinothricin dehydrogenase mutant for amino translocation according to claim 1; and (2) a genetically recombinant bacterium producing the phosphinothricin dehydrogenase mutant for amino translocation according to claim 1, or a crude enzyme liquid containing the phosphinothricin dehydrogenase mutant obtained by lysis of the genetically recombinant bacterium.
7. The method according to claim 6, wherein the coenzyme circulation system is at least one of: (1) a formate dehydrogenase coenzyme circulation system comprising a formate dehydrogenase, formate and a coenzyme; (2) a glucose dehydrogenase coenzyme circulation system comprising a glucose dehydrogenase, glucose and a coenzyme; and (3) an alcohol dehydrogenase coenzyme circulation system comprising an alcohol dehydrogenase, isopropanol and a coenzyme.
8. The method according to claim 7, wherein the coenzyme circulation system is a formate dehydrogenase coenzyme circulation system comprising a formate dehydrogenase, formate and a coenzyme.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1)
(2)
(3)
(4)
(5)
DESCRIPTION OF THE EMBODIMENTS
Example 1
Machine Learning (Random Forest Strategy) Gene Mining of Phosphinothricin Dehydrogenase Gene
(6) (1) Establishment of decision tree: 100,000 sequences were randomly selected from an NCBI microbial gene bank; for these 100,000 sequences, the machine learning tool kit scikit-learn was configured using the random forest, with n_estimator parameter set as 1000, and the remaining parameters as default values: 10 samples were randomly selected with replacement (select one sample at a time at random, and then return to continue the selection), and the selected 10 samples were used to train a decision tree as the samples at a root node of the decision tree.
(7) (2) Setting features (attributes): (a) protein size: length of candidate proteins (300-500 amino acids); (b) necessary characteristic sequences at both ends of phosphinothricin dehydrogenase (i.e., only the enzyme containing both sequences can be a phosphinothricin dehydrogenase): a first segment was GGGKGG, and a second segment was one of VVTG, FVTG, VLTG, VFTG, FITG, FFTG, VVFG, FVFTG, VLFG, VFFG, FLFG, and FFFG.
(8) (3) Decision tree splitting: when each sample had the above three attributes, when a node of the decision tree needed to be split, one attribute was randomly selected from the three attributes. Each node in the decision tree formation process was split according to the step (2) (until it could no longer be split).
(9) (4) 246 decision trees were established according to the steps 1 to 3 to form a random forest. Then, all genes in the NCBI microbial gene bank were put into the forest to be judged and classified by each decision tree in the forest respectively, and the tree with the largest number of genes was selected.
(10) (5) The amino acid sequences of these genes were aligned with a known phosphinothricin dehydrogenase (WP_060477601.1). A phosphinothricin dehydrogenase (PhPPTDH, Genbank: WP_179026919.1, with an amino acid sequence as shown in SEQ ID No. 2, derived from Pseudomonas hunanensis) with the highest sequence similarity was selected.
Example 2
Construction of Genetically Recombinant Bacterium
(11)
(12) 1. Construction of Genetically Recombinant Bacterium with Phosphinothricin Dehydrogenase
(13) The phosphinothricin dehydrogenase gene (nucleotide sequence as shown in SEQ ID No.1, amino acid sequence as shown in SEQ ID No. 2) was used to construct an expression vector pETDuet-PhPPTDH to transform E. coli, thereby obtaining an original strain E. coli BL21(DE3)/pETDuet-PhPPTDH.
(14) 2. Construction of Genetically Recombinant Bacterium with Coupled Phosphinothricin Dehydrogenase and Glucose Dehydrogenase
(15) A glucose dehydrogenase (GDH) gene (Genbank: KM817194.1) derived from Exiguobacterium sibiricum was selected to perform whole-gene synthesis to construct a genetically engineered strain E. coli BL21(DE3)/pETDuet-PhPPTDH-EsGDH. The plasmid profile of the obtained coupled PPTDH and EsGDH expression vector is shown in
Example 3
(16) The library of phosphinothricin dehydrogenase mutants was prepared by three rounds of site-saturation mutagenesis with primer designs shown in Table 1.
(17) In the first round, with a vector pETDuet-PhPPTDH as a template, and the site-directed mutagenesis primers E263X-F and E263X-R in Table 1 as primers (wherein in the degenerate bases involved in the primer sequence, N represents A, C, G or T; K represents G or T; M represents A or C), PCR was conducted to mutate the glutamic acid at position 263 of the phosphinothricin dehydrogenase amino acid sequence as shown in SEQ ID No. 2 into other natural amino acids, followed by transformation, plate coating, and screening of dominant strains to obtain phosphinothricin dehydrogenase mutants PhPPTDH-E263D, PhPPTDH-E263N, PhPPTDH-E263C, and PhPPTDH-E263G, and recombinant plasmids pETDuet-PhPPTDH-E263D-EsGDH, pETDuet-PhPPTDH-E263N-EsGDH, pETDuet-PhPPTDH-E263C-EsGDH, and pETDuet-PhPPTDH-E263G-EsGDH. Among them, PhPPTDH-E263D and the recombinant plasmid pETDuet-PhPPTDH-E263D-EsGDH had the highest activity.
(18) In the second round, with the recombinant plasmid pETDuet-PhPPTDH(E263D)-EsGDH obtained in the first round as a template and the site-directed mutagenesis primers K134X-F and K134X-R in Table 1 as primers, PCR was conducted to mutate the valine at position 134 into other natural amino acids, followed by transformation and plate coating to obtain an optimal phosphinothricin dehydrogenase mutant PhPPTDH-E263D-K134R, and the recombinant plasmid obtained by mutation was pETDuet-PhPPTDH(E263D-K134R)-EsGDH.
(19) In the third round, with the recombinant plasmid pETDuet-PhPPTDH(E263D-K134R)-EsGDH obtained in the second round as a template and the site-directed mutagenesis primers H96X-F and H96X-R in table 1 as primers, PCR was conducted to mutate the histidine at position 96 into other natural amino acids, followed by transformation and plate coating to obtain the optimal phosphinothricin dehydrogenase dominant mutant PhPPTDH(E263D-K134R-H96A), and the obtained recombinant plasmid was pETDuet-PhPPTDH(E263D-K134R-H96A)-EsGDH.
(20) In the fourth round, with the recombinant plasmid pETDuet-PhPPTDH-E263D-K134R-H96A-EsGDH obtained in the third round as a template, and the site-directed mutagenesis primers R290X-F and R290X-R in Table 1 as primers, PCR was conducted to mutate the arginine at position 290 into other natural amino acids, followed by transformation and plate coating to obtain the optimal phosphinothricin dehydrogenase dominant mutant PhPPTDH(E263D-K134R-H96A-R290V), and the obtained recombinant plasmid was pETDuet-PhPPTDH(E263D-K134R-H96A-R290V)-EsGDH.
(21) TABLE-US-00001 TABLE 1 Primer designs for phosphinothricin dehydro- genase site-directed mutagenesis Mutation Primer site name Primer sequence (5′-3′) E263X E263X-F GTCTGACTCCnnkGGCACCTTGTACGCTG E263X-R TACAAGGTGCCmnnGGAGTCAGACAGCGA K134X K134X-F CGACCCTnnkGGCAAGAGCGACGCTGAAG K134X-R GTCGCTCTTGCCmnnAGGGTCGAAGTCCG H96X H96X-F TGCGTTTCnnkCCGTCGGTTAACCTCAGC H96X-R ACCGACGGmnnGAAACGCAGCCCGCCCTT R290X R290X-F GCGCGGCnnkATCAGCGAGCTGGCCGGGC R290X-R TCGCTGATGCGGCCmnnCTTGACGTTCTT
(22) The mutation PCR system (50 μL) consisted of 25 μL of 2-fold Phanta Max buffer, 1 μL of dNTPs, 1 μL of each of upper and lower primers for mutation, 1 μL of template, 1 μL of Phanta Super-Fidelity DNA polymerase, and ddH.sub.2O making up to 50 μL.
(23) PCR conditions: pre-denaturation at 95° C. for 5 min; 30 cycles: 90° C. for 30 seconds, 58-60° C. for 30 seconds, 72° C. for 7 min; and final elongation at 72° C. for 10 min.
(24) The PCR results were verified by DNA agarose gel electrophoresis, the PCR products were subjected to template digestion by Dpn I enzyme at 37° C. and 160 rpm for 15 min, and purified with a purification kit. E. coli BL21(DE3) competent cells were prepared and transformed with the PCR product through heat-shock, cultured at 37° C. and 160 rpm for 1 hour, and coated on an LB plate containing 50 μg/mL ampicillin resistance to culture upside down at 37° C. overnight.
Example 4
Construction of a Gene Library of the Recombinant Enzyme (PhPPTDH)
(25) The competent cells of E. coli BL21(DE3) (Invitrogen) stored at −80° C. were placed in an ice bath at 0° C. for 10 min, and then 5 μL of the expression vector pETDuet-1-PhPPTDH-EsGDH with the phosphinothricin dehydrogenase mutant and the glucose dehydrogenase was added thereto in a super clean bench. The mixture was placed in an ice bath at 0° C. for 30 min, heat-shocked in a water bath at 42° C. for 90 seconds, and placed in an ice bath at 0° C. for 2 min. 600 μL of LB culture medium was added thereto, and the mixture was then shaking-cultured at 37° C. and 200 rpm for 1 hour. The mixture was coated on an LB plate containing 50 μg/ml ampicillin resistance to culture at 37° C. for 8-12 hours, thereby obtaining a recombinant E. coli BL21(DE3)/pETDuet-1-PhPPTDH-EsGDH containing the mutated phosphinothricin dehydrogenase.
(26) Preparation of competent cells in the example: The E. coli BL21(DE3) strain preserved in a glycerol tube was obtained from a refrigerator at −80° C., streaked on an antibiotic-free LB plate to culture at 37° C. for 10 hours to obtain single colonies. The single colonies on the LB plate were picked, inoculated into a test tube containing 5 mL of LB culture medium to culture at 37° C. and 180 rpm for 9 hours. Then, 200 μL of the bacterial liquid was taken from the test tube and inoculated into 50 mL of LB culture medium to culture at 37° C. and 180 rpm until OD.sub.600 reached 0.4-0.6. The bacterial liquid was pre-cooled on ice, put into a sterilized centrifuge tube, placed on ice for 10 min, and then centrifuged at 4° C. and 5,000 rpm for 10 min. The supernatant was poured out (contamination should be prevented), and the precipitated cells were re-suspended in pre-cooled 0.1 mol/L aqueous CaCl.sub.2 solution and placed on ice for 30 min, and then centrifuged at 4° C. and 5,000 rpm for 10 min. The supernatant was discarded and the precipitated cells were re-suspended in pre-cooled 0.1 mol/L aqueous CaCl.sub.2 solution containing 15% glycerol. 100 μL of the re-suspended cells were aliquoted into a sterilized 1.5 mL centrifuge tube and stored in a refrigerator at −80° C. for later use.
Example 5
Induced Expression of Parent and Mutant Phosphinothricin Dehydrogenase
(27) The original strain pETDuet-PhPPTDH-EsGDH of the first step of Example 1 was inoculated into an LB liquid medium containing ampicillin with a final concentration of 50 μg/mL to culture at 37° C. for 8 hours, and then inoculated at a volume fraction of 2% (v/v) into a fresh LB liquid medium containing ampicillin with a final concentration of 50 μg/mL to culture at 37° C. and 180 rpm for 2 hours. Then, IPTG with a final concentration of 0.1 mM was added to the culture liquid to culture at 28° C. for 14 hours. The mixture was then centrifuged at 4° C. and 8,000 rpm for 10 min, thereby obtaining the corresponding wet cells. The cells obtained above produced corresponding proteins, which can be used for preparing pure protein enzyme liquid, or can be used for catalysis in the form of crude enzyme liquid to obtain L-PPT.
Example 6
Screening of Phosphinothricin Dehydrogenase Mutant Library
(28) The wet cells of mutant strains with induced expression were placed in a refrigerator at −20° C. to freeze overnight, re-suspended at an amount of 40 g/L total cells in a 100 mM phosphate buffer at pH 7.4, and allowed to melt on ice to obtain a crude enzyme liquid of the mutant strain. Under the same conditions, a crude enzyme liquid of the original strain was prepared using the original strain instead of the mutant strain. 200 mM glucose and 240 mM ammonium sulfate were dissolved in 10 mL of phosphate buffer (100 mM) at pH 8.5. 500 μL of the original solution was added to 500 μL of the crude enzyme solution of the different mutants from Example 2 (1 g of wet cells in 10 mL of a 100 mM phosphate buffer at pH 7.4), 1 mM NAD was added, the mixture was shaken at 35° C. and 600 rpm for 10 min, and then the reaction was terminated by adding 10 μL of 6 M HCl. The product concentration was detected by HPLC, and the dominant mutants were screened using the product L-phosphinothricin and e.e. value as indicators. The experimental results are shown in Table 2. Among others, E. coli BL21 (DE3)/PhPPTDH(E263D-K134R-H96A-R290V)-EsGDH was the most dominant strain, and the PHPPTDH gene of this dominant strain had a four-site mutation E263D-K134R-H96A-R290V.
(29) Chiral analysis and concentration analysis of the products in the example were performed by pre-column derivatization high performance liquid chromatography, which specifically consisted of:
(30) (1) Chromatographic conditions: column model: QS-C18, 5 μm, 4.6×250 mm; mobile phase: 50 mM ammonium acetate solution:methanol=10:1; fluorescence detection wavelength: λ.sub.ex=340 nm, λ.sub.em=455 nm; flow rate: 1 mL/min; column temperature: 30° C., L-PPT retention time: 11 min, and D-PPT retention time: 13.4 min.
(31) (2) Derivatization reagent: 0.1 g of o-phthalaldehyde and 0.12 g of N-acetyl-L-cysteine were weighed separately and dissolved in 10 mL of ethanol, 40 mL of 0.1 mol/L boric acid buffer (pH 9.8) was added and the mixture was shaken to fully dissolve and then stored in a refrigerator at 4° C. for later use (no more than 4 days).
(32) (3) Derivatization reaction and HPLC detection: Ultra-pure water was used to make up to 1 mL, i.e., the reaction mixture was diluted 10 times. The diluted sample was subjected to derivatization treatment. To 200 μL of the diluted reaction mixture, 400 μL of the derivatization reagent was added for derivatization at 30° C. for 5 min, and then 400 μL of ultra-pure water was added to make up to 1 mL. The mixture was centrifuged at 12000 rpm for 1 min. The supernatant was passed through a 0.22 μM microfiltration membrane as a liquid sample, and detected for PPO, L-PPT, D-PPT, and e.e. value by HPLC.
(33) TABLE-US-00002 TABLE 2 Catalytic performance and stereoselectivity of PhPPTDH and mutants thereof Corresponding Mutated Relative PhPPTDH bases and enzyme Enzyme e.e. mutant for corresponding activity activity value strain amino acids (%) (U/g) (%) PhPPTDH — .sup. 100.sup.a 32.75 99.5 PhPPTDH(E263D) GAA.fwdarw.GAT(E263D) 600 196.7 99.5 PhPPTDH(E263N) GAA.fwdarw.TTA(E263N) 371 121.65 99.5 PhPPTDH(E263C) GAA.fwdarw.ACA(E263C) 463 151.55 99.5 PhPPTDH(E263G) GAA.fwdarw.CCT(E263G) 484 158.45 99.5 PhPPTDH(E263D- GAA .fwdarw. GAT(E263D) 492 165.7 99.5 K134R) AAG.fwdarw.CGA(K134R) PhPPTDH(E263D- GAA .fwdarw. GAT(E263D) 609 205.2 99.5 K134R-H96A) AAG.fwdarw.CGA(K134R) CAC.fwdarw.GCA(H96A) PhPPTDH(E263D- GAA .fwdarw. GAT(E263D) 666 224.5 99.5 K134R-H96A- AAG.fwdarw.CGA(K134R) R290V) CAC.fwdarw.GCA(H96A) CGC.fwdarw.GTA(R290V) PhPPTDH — .sup. 100.sup.b 137.3 99.5 PhPPTDH(E263D) GAA.fwdarw.GAT(E263D) 224 307.2 99.5 PhPPTDH(E263N) GAA.fwdarw.TTA(E263N) 220 305.4 99.5 PhPPTDH(E263C) GAA.fwdarw.ACA(E263C) 209 287.2 99.5 PhPPTDH(E263G) GAA.fwdarw.CCT(E263G) 224 306.5 99.5 PhPPTDH(E263D- GAA .fwdarw. GAT(E263D) 208 285.8 99.5 K134R) AAG-CGA(K134R) PhPPTDH(E263D- GAA .fwdarw. GAT(E263D) 215 295.3 99.5 K134R-H96A) AAG.fwdarw.CGA(K134R) CAC.fwdarw.GCA(H96A) PhPPTDH(E263D- GAA .fwdarw. GAT(E263D) 319 437.6 99.5 K134R-H96A- AAG.fwdarw.CGA(K134R) R290V) CAC.fwdarw.GCA(H96A) CGC.fwdarw.GTA(R290V) .sup.aUnder the standard conditions, with NAD as the coenzyme, the enzyme activity of PhPPTDH was 100%. .sup.bUnder the standard conditions, with NADP as the coenzyme, the enzyme activity of PhPPTDH was 100%.
Example 7
Asymmetric Reductive Amination of Low Concentration of 2-carbonyl-4-(hydroxymethylphosphinyl)-butyric Acid (PPO) Using Phosphinothricin Dehydrogenase Mutant Coupled with Glucose Dehydrogenase
(34) 1 g of E. coli BL21 (DE3)/PhPPTDH(E263D-K134R-H96A-R290V)-EsGDH wet cells prepare by the method of Example 6 were re-suspended with 50 mL of phosphate buffer (100 mM) at pH 7.4, and 2-carbonyl-4-(hydroxymethylphosphinyl)-butyric acid with a final concentration of 100 mM, glucose with a final concentration of 110 mM, and ammonium sulfate with a final concentration of 120 mM were added to form a reaction system of 50 mL to react at 35° C. and a magnetic stirring speed of 600 rpm, and ammonia was fed to maintain the pH of the reaction mixture at 7.4. The production of the product L-phosphinothricin during the reaction was detected by the liquid phase method shown in Example 4, and the reaction progress curve is shown in
Example 8
Asymmetric Reductive Amination of High Concentration of 2-carbonyl-4-(hydroxymethylphosphinyl)-butyric Acid (PPO) Using Phosphinothricin Dehydrogenase Mutant Coupled with Glucose Dehydrogenase
(35) 1 g of E. coli BL21 (DE3)/PhPPTDH(E263D-K134R-H96A-R290V)-EsGDH wet cells prepare by the method of Example 6 were re-suspended with 100 mL of phosphate buffer (100 mM) at pH 7.4, and 2-carbonyl-4-(hydroxymethylphosphinyl)-butyric acid with a final concentration of 500 mM, glucose with a final concentration of 510 mM, and ammonium sulfate with a final concentration of 520 mM were added to form a reaction system of 50 mL to react at 35° C. and a magnetic stirring speed of 600 rpm, and ammonia was fed to maintain the pH of the reaction mixture at 7.4. The production of the product L-phosphinothricin during the reaction was detected by the liquid phase method shown in Example 4, and the reaction progress curve is shown in
Comparative Example 1
Asymmetric Reductive Amination of Low Concentration of 2-carbonyl-4-(hydroxymethylphosphinyl)-butyric Acid (PPO) Using Wild-Type Phosphinothricin Dehydrogenase Coupled with Glucose Dehydrogenase
(36) 1 g of E. coli BL21 (DE3)/PhPPTDH-EsGDH wet cells prepare by the method of Example 2 were re-suspended with 50 mL of phosphate buffer (100 mM) at pH 7.4, and 2-carbonyl-4-(hydroxymethylphosphinyl)-butyric acid with a final concentration of 100 mM, glucose with a final concentration of 110 mM, and ammonium sulfate with a final concentration of 120 mM were added. The reaction system was 50 mL, the temperature was 35° C., and the magnetic stirring speed was 600 rpm. The production of the product L-phosphinothricin during the reaction was detected by the liquid phase method shown in Example 4. After 4 hours of reaction, the substrate conversion was only 64%.
Comparative Example 2
Asymmetric Reductive Amination of High Concentration of 2-carbonyl-4-(hydroxymethylphosphinyl)-Butyric Acid (PPO) Using Wild-Type Phosphinothricin Dehydrogenase Coupled with Glucose Dehydrogenase
(37) 1 g of E. coli BL21 (DE3)/PhPPTDH-EsGDH wet cells prepare by the method of Example 2 were re-suspended with 100 mL of phosphate buffer (100 mM) at pH 7.4, and 2-carbonyl-4-(hydroxymethylphosphinyl)-butyric acid with a final concentration of 500 mM, glucose with a final concentration of 510 mM, and ammonium sulfate with a final concentration of 520 mM were added. The reaction system was 50 mL, the temperature was 35° C., and the magnetic stirring speed was 600 rpm. Ammonia was fed to maintain the pH of the reaction mixture at 7.4. The production of the product L-phosphinothricin during the reaction was detected by the liquid phase method shown in Example 4. After 4 hours of reaction, the substrate conversion was only 35%.