Novel PET hydrolase designed based on deep learning algorithm and preparation method thereof

Abstract

The invention discloses a PET hydrolase designed by a deep learning algorithm and a preparation method thereof, and relates to the technical field of biologication.The preparation method comprises the steps that polyester hydrolase PHL7 is selected, a protein crystal structure is learned through the deep learning algorithm InstructPLM, three-dimensional structure modeling is conducted through Esmfold, candidate protein and PET are subjected to molecular docking, and molecular dynamics simulation is conducted. Screening out protein which is well combined with a substrate; performing expression and purification; amino acid sequences of the hydrolase HN and / or the hydrolase JW are respectively shown as SEQ ID NO: 1 and / or SEQ ID NO: 2; the invention also relates to application of the novel PET hydrolase in hydrolysis of PET. Through the screening process, the PET hydrolase can be simply, rapidly and efficiently screened from the candidate sequences generated in a large scale, and the source of the PET hydrolase is expanded.

Description

Technical Field

[0001] The present invention relates to the field of biotechnology, and in particular to a novel PET hydrolase designed based on a deep learning algorithm and a preparation method thereof. Background Art

[0002] Polyethylene terephthalate (PET) is a key polymer material widely used in the manufacture of beverage bottles, packaging films and textiles. Due to its excellent durability, PET plays an important role in many aspects of human society. However, the chemical stability of PET also leads to its accumulation in the natural environment, posing a serious threat to the ecosystem. In order to solve this problem, biodegradation technology based on PET hydrolase has been widely studied and has shown great potential. Compared with traditional treatment methods, biodegradation can reduce the adverse effects on the environment while avoiding the use of harmful chemical reagents and high-demand reaction vessels. In addition, the reaction conditions of the biodegradation method are relatively mild, and PET can be degraded into small molecules and recycled, which is both environmentally friendly and economically meaningful.

[0003] To date, about 94 enzymes with PET degradation capabilities have been discovered and identified. However, the application of PET hydrolases still faces some challenges, such as insufficient stability to pH and temperature ranges, slow reaction rates, and the inability to directly use untreated post-consumer plastics. Therefore, further enzyme engineering research is needed to improve the degradation efficiency of these enzymes or to find new PET hydrolases.

[0004] At present, almost all PET hydrolases are isolated from environmental microbial communities, and the efficiency of finding PET hydrolases with good functions in nature is very low. The enzyme engineering modification of existing PET hydrolases is also close to the limit. With the rapid development of computer technology and artificial intelligence, the exploration of enzyme sequence space is constantly expanding. It can be used for de novo design of proteins based on the requirements of protein structure and function through computational means, and then design new PET hydrolases, which greatly broadens the source and diversity of PET hydrolases and is expected to provide more reliable and sustainable technical solutions for industrial-scale PET waste biodegradation.

[0005] Therefore, technicians in this field are committed to developing a fast, efficient and convenient method for designing new PET hydrolases and screening new PET hydrolases, which may not only improve the efficiency of PET recycling, but also help promote the development of protein design. Summary of the invention

[0006] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is how to develop a rapid, efficient and convenient method for designing a novel PET hydrolase and screen out the novel PET hydrolase.

[0007] To achieve the above object, the present invention provides a method for designing a novel PET hydrolase based on a deep learning algorithm, comprising the following steps:

[0008] Step 1: Analyze the characteristics of enzymes with PET hydrolysis activity through literature research, select polyester hydrolase PHL7, and obtain the protein crystal structure and corresponding amino acid sequence of the hydrolase PHL7 through the PDB database;

[0009] Step 2: Use the deep learning algorithm InstructPLM to learn the protein crystal structure of the hydrolase PHL7 obtained in step 1 and generate sequences in batches to screen out candidate sequences;

[0010] Step 3: Use Esmfold to perform three-dimensional structural modeling on the candidate sequences obtained in step 2, and retain the structures with higher confidence as candidate proteins;

[0011] Step 4: Molecular docking of the candidate protein obtained in step 3 with PET to obtain the enzyme-ligand complex structure;

[0012] Step 5: Perform molecular dynamics simulation on the composite structure obtained in step 4 to screen out proteins that bind well to the substrate;

[0013] Step 6: Express and purify the protein screened in step 5.

[0014] Furthermore, step 3 also includes:

[0015] Step 3.1, using the SoluProt v1.0 algorithm to predict the protein solubility of the sequences generated in step 2, and screening out sequences with higher predicted solubility;

[0016] Step 3.2: Calculate the protein charge corresponding to the sequence with higher solubility obtained in step 3.1, and select the sequence with less negative charge as the candidate sequence.

[0017] Furthermore, step 4 also includes globally docking the candidate protein and the PET molecule using AMDock software, screening the candidate protein that can be docked into the correct pocket, and obtaining the enzyme-ligand complex structure of the candidate protein.

[0018] Furthermore, step 5 also includes:

[0019] Step 5.1, using Amber software, the enzyme-ligand complex structure is placed in a cubic water box constructed by the TIP3P model, and counterions are added to obtain an enzyme-protein system in the solution;

[0020] Step 5.2, performing energy minimization optimization on the enzyme-protein system in the solution obtained in step 5.1, removing obvious unreasonable factors in the system, and obtaining an energy-minimized enzyme-protein system;

[0021] Step 5.3, constraining the protein backbone atoms of the enzyme protein system with minimized energy in step 5.2, specifically: gradually raising the temperature from 0°C to 338°C to obtain the constrained enzyme protein system;

[0022] Step 5.4, performing a 1 ns pre-equilibrium kinetic simulation on the constrained enzyme-protein system obtained in step 5.3 to obtain an enzyme-protein system after pre-equilibrium kinetic simulation;

[0023] Step 5.5: The enzyme-protein system after the pre-equilibrium kinetic simulation obtained in step 5.4 is subjected to three independent isothermal kinetic simulations of 50 ns under the NPT ensemble.

[0024] Furthermore, the counter ion in step 5.1 is: Na + and Cl - .

[0025] Furthermore, step 5.5 also includes: visualizing the enzyme protein structure change information of the enzyme protein system during the kinetic simulation process, and selecting a stable structure; a stable structure is: the distance between the catalytic residues Ser and His and the distance between the substrate and the catalytic residue Ser are both less than

[0026] Furthermore, step 6 also includes: codon optimization of the protein screened out in step 5 through an Escherichia coli overexpression system, and then inserting the gene and histidine tag into the NdeI / XhoI site of the plasmid pET-26b+, and introducing the plasmid into E. coli Bl21 (DE3) for cell culture at 37°C; after the OD reaches 0.4-0.6, adding isopropyl thiogalactoside with a final concentration of 0.3mM and culturing at 30°C for 5 hours, and then centrifuging to obtain a cell solution; then lysing, eluting and concentrating the cell solution at 4°C, and finally purifying it by two steps of nickel column affinity chromatography and desalting column chromatography.

[0027] Furthermore, the activity of the new PETPET hydrolase was identified, and the reaction products of PET hydrolase enzymatic hydrolysis of PET film were analyzed by high performance liquid chromatography.

[0028] The present invention also discloses a novel PET hydrolase, hydrolase HN and / or hydrolase JW, prepared by a method for designing a novel PET hydrolase based on a deep learning algorithm; the amino acid sequence of hydrolase HN is shown in SEQ ID NO: 1, and the amino acid sequence of hydrolase JW is shown in SEQ ID NO: 2.

[0029] The invention also discloses application of a novel PET hydrolase in hydrolyzing PET.

[0030] In preferred embodiment 1 of the present invention, the design and screening process of PET hydrolase is described in detail;

[0031] In another preferred embodiment 2 of the present invention, the expression and purification process of the designed PET hydrolase is described in detail;

[0032] In another preferred embodiment 3 of the present invention, the activity identification process of the designed PET hydrolase is described in detail.

[0033] The beneficial technical effects of the present invention are as follows:

[0034] The present invention utilizes a deep learning algorithm to design a novel PET hydrolase, thereby overcoming the problem of low efficiency in finding a PET hydrolase with good functions in nature. The deep learning algorithm also expands the sequence space and expands the source of PET hydrolases.

[0035] The screening process provided by the present invention can quickly and efficiently screen a large number of generated sequences, and quickly screen the sequences according to the net charge number of the protein, the predicted solubility and the affinity with the substrate, and finally efficiently screen out two PET hydrolases with good hydrolysis activity on PET membranes.

[0036] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a comparison diagram of the tertiary structure of PHL7 of a preferred embodiment 1 of the present invention and the designed PET hydrolase (HN, JW);

[0038] Figure 2 This is a protein gel image of the PET hydrolase (HN, JW) designed in a preferred embodiment 2 of the present invention;

[0039] Figure 3 This is a product diagram of a preferred embodiment 3 of the present invention in which HN and JW hydrolyze a PET film for 48 hours. DETAILED DESCRIPTION

[0040] The following describes several preferred embodiments of the present invention with reference to the drawings in the specification, so that the technical content is clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the protection scope of the present invention is not limited to the embodiments mentioned in the text.

[0041] Example 1: Design and screening of PET hydrolases

[0042] Polyester hydrolase PHL7 has excellent hydrolysis ability for amorphous PET film at 70°C. The present invention uses the crystal structure of polyester hydrolase PHL7 as the initial structure and uses the published InstructPLM inverse folding algorithm to learn the protein structure to generate a large number of candidate sequences. The three-dimensional structure of the hydrolase PHL7 is shown in Figure 1 As shown, part A is the structure 7NEI of PHL 7 in the PDB database; part B is the structural model predicted by HN; and part C is the structural model predicted by JW.

[0043] The protein solubility prediction of PHL7 and candidate sequences was performed using the published SoluProt v1.0 software. The predicted solubility score of PHL7 was 0.636, and sequences with predicted solubility scores lower than 0.64 were removed from the candidate sequences. The net charge of PHL7 and the candidate sequences was calculated. PHL7 carried 5 negative charges, and candidate sequences with positive charge, no charge, and negative charge less than or equal to 5 were retained. Esmfold was used to perform three-dimensional structural modeling on the candidate sequences, and candidate sequences and corresponding structures with predicted local distance difference test (pLDDT) greater than or equal to 85 were retained. AMDock software was used to globally dock the candidate proteins with 4PET molecules, and candidate proteins whose substrates could be docked to the correct pockets were screened and their complex structures were obtained. The complex structure of the enzyme and substrate was placed in a cubic water box constructed by the TIP3P model using Amber software, and counterions (Na + , Cl - ), which is conducive to maintaining the overall electrical neutrality of the system and simulating the environment of a real enzyme protein solution. The enzyme protein system in the solution is then energy minimized and optimized to remove obvious unreasonable factors in the system. The protein skeleton atoms in the system are then constrained and the temperature is gradually raised from 0°C to a high temperature environment (338°C), which is conducive to maintaining the rationality of the overall structure. The system is then subjected to a 1ns pre-equilibrium kinetic simulation, which is conducive to the early relaxation of the structure. Finally, each system is subjected to three independent isothermal kinetic simulations of 50ns under the NPT ensemble, which is conducive to the full relaxation and optimization of the three-dimensional structure of the protein. Finally, the enzyme protein structural change information (RMSD) during the kinetic simulation is visualized, and a more stable structure is selected. The distance between the catalytic residues Ser and His and the distance between the substrate and the catalytic residue Ser are calculated, and the two distances in the simulation are selected to be less than The candidate proteins obtained in the previous step were subjected to 100 ns molecular dynamics simulation with the amorphous PET film mimics using Gromacs software to calculate the interaction energy and interaction area between the two candidate proteins and the PET film.

[0044] Example 2: Expression and purification of the designed PET hydrolase

[0045] The two candidate proteins obtained in Example 1 were codon optimized through the E. coli overexpression system, and then the gene and 6 histidine tags were inserted into the NdeI / XhoI site of pET26b, and the plasmid was introduced into E. coli Bl21 (DE3) for cell culture at 37°C. After the OD reached 0.4-0.6, isopropylthiogalactoside (IPTG) was added at a final concentration of 0.3mM and cultured at 30°C for 5 hours, followed by centrifugation. The cell solution was then lysed, eluted and concentrated at 4°C, and finally the protein was obtained after two-step purification by nickel column affinity chromatography and desalting column chromatography. The protein gel images of the two candidate proteins (HN, JW) are shown below. Figure 2 As shown, channel 1 is a protein marker; channel 2 is the supernatant after HN crushing and centrifugation; channel 3 is the precipitate after HN crushing and centrifugation; channel 4 is the impurity protein eluted by HN; channel 5 is the target protein HN eluted by the nickel column; channel 6 is a protein marker; channel 7 is the supernatant after JW crushing and centrifugation; channel 8 is the precipitate after JW crushing and centrifugation; channel 9 is the impurity protein eluted by JW; channel 10 is the target protein JW eluted by the nickel column.

[0046] Example 3: Activity identification of the designed PET hydrolase

[0047] The protein obtained in Example 2 was reacted with the amorphous PET film purchased from GoodFellow at 30°C to 60°C and pH 7 to pH 8. Samples were taken at 24 hours and 48 hours of incubation, diluted 10 times with water and then diluted 2 times with acetonitrile. The substrate solution without enzyme treatment was used as a control. The reaction products were analyzed by high performance liquid chromatography (HPLC) using an Agilent 1260 Infinity II system equipped with an Agilent Poroshell 120EC-C18 column (150×4.6mm, 4μm) and a detection wavelength of 280nm. The column oven was set at 30°C, with eluent A (H2O containing 0.1% [v / v] trifluoroacetic acid) and eluent B (acetonitrile) as the mobile phase, and the following gradient was maintained for 20 min: 0-4 min, 5% B; 4-16 min, 5%-30% B; 16-18 min, 30% B; 18-20 min, 5% B, the flow rate was fixed at 0.7 ml / min, and the column temperature was 30°C. HN and JW hydrolyzed PET film (purchased from GoodFellow, catalog number es30-fm-000145) in 1M pH 8 phosphate buffer system for 48 hours. Figure 3 As shown, part A is the yield of hydrolysis products after JW with a protein concentration of 300 nM was incubated with PET film at different pH and temperature for 48 hours; part B is the yield of hydrolysis products formed after HN with different concentrations and PET film were incubated at 55°C for 48 hours; the ordinate is the concentration of MHET and TPA in the reaction solution after the reaction; the error bars represent the standard error of three independent experiments. Figure 3 It can be seen that the hydrolysis ability of JW is significantly weaker than that of HN, and the optimum temperature of JW is 45°C, and its hydrolysis ability at pH 8 is greater than that at pH 7 and pH 7.5. The hydrolysis ability of HN changes significantly with the increase of its protein concentration from 300nM to 900nM. When its concentration increases from 900nM to 1500nM, the total amount of hydrolysate changes very little, but the content of MHET in the hydrolysate decreases.

[0048] The preferred specific embodiments of the present invention are described in detail above. It should be understood that ordinary technicians in the field can make many modifications and changes based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by technicians in the technical field based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for designing a novel PET hydrolase based on a deep learning algorithm, characterized in that: The method comprises the following steps: Step 1, analyzing the characteristics of enzymes with PET hydrolysis activity through literature research, selecting polyester hydrolase PHL7, and obtaining the protein crystal structure and corresponding amino acid sequence of the hydrolase PHL7 through the PDB database; Step 2: Use the deep learning algorithm InstructPLM to learn the protein crystal structure of the hydrolase PHL7 obtained in step 1 and generate sequences in batches to screen out candidate sequences; Step 3: Use Esmfold to perform three-dimensional structural modeling on the candidate sequences obtained in step 2, and retain the structures with higher confidence as candidate proteins; Step 4: Molecular docking of the candidate protein obtained in step 3 with PET to obtain the enzyme-ligand complex structure; Step 5: Perform molecular dynamics simulation on the composite structure obtained in step 4 to screen out proteins that bind well to the substrate; Step 6: Express and purify the protein screened in step 5.

2. The method according to claim 1, characterized in that The step 3 also includes: Step 3.1, using the SoluProt v1.0 algorithm to predict the protein solubility of the sequence generated in step 2, and screening out sequences with higher predicted solubility; Step 3.2, calculate the protein charge corresponding to the sequence with higher solubility obtained in step 3.1, and screen out the sequence with less negative charge as the candidate sequence.

3. The method according to claim 1, characterized in that The step 4 also includes globally docking the candidate protein and the PET molecule using AMDock software, screening the candidate protein that can be docked to the correct pocket, and obtaining the enzyme-ligand complex structure of the candidate protein.

4. The preparation method according to claim 1, characterized in that: The step 5 also includes: Step 5.1, placing the enzyme-ligand complex structure in a cubic water box constructed by the TIP3P model using the Amber software, and adding counter ions to obtain an enzyme-protein system in the solution; Step 5.2, performing energy minimization optimization on the enzyme-protein system in the solution obtained in step 5.1, removing obvious unreasonable factors in the system, and obtaining an energy-minimized enzyme-protein system; Step 5.3, constraining the protein backbone atoms of the enzyme protein system with minimized energy in the system 5.2 described in step 5. Specifically, gradually raising the temperature from 0° C. to 338° C. to obtain a constrained enzyme protein system; Step 5.4, performing a 1 ns pre-equilibrium kinetic simulation on the constrained enzyme-protein system obtained in step 5.3 to obtain an enzyme-protein system after pre-equilibrium kinetic simulation; Step 5.5: The enzyme-protein system after the pre-equilibrium kinetic simulation obtained in step 5.4 is subjected to three independent isothermal kinetic simulations of 50 ns under the NPT ensemble.

5. The method according to claim 4, characterized in that The counter ion in step 5.1 is: Na + and Cl - .

6. The method according to claim 5, characterized in that The step 5.5 also includes: visualizing the enzyme protein structure change information of the enzyme protein system during the kinetic simulation process, and selecting a stable structure; the stable structure is: the distance between the catalytic residues Ser and His and the distance between the substrate and the catalytic residue Ser are both less than 7. The method according to claim 1, characterized in that The step 6 also includes: codon optimization of the protein screened out in the step 5 through an Escherichia coli overexpression system, then inserting the gene and the histidine tag into the NdeI / XhoI site of the plasmid pET-26b+, introducing the plasmid into E. coli Bl21 (DE3) and culturing the cells at 37°C; after the OD reaches 0.4-0.6, adding isopropyl thiogalactoside with a final concentration of 0.3 mM and culturing at 30°C for 5 hours, and then centrifuging to obtain a cell solution; then lysing, eluting and concentrating the cell solution at 4°C, and finally purifying it by two steps of nickel column affinity chromatography and desalting column chromatography.

8. The method according to claim 1, characterized in that The method further comprises identifying the activity of the PET hydrolase, wherein the identification adopts high performance liquid chromatography to analyze the reaction product of the PET hydrolase enzymatically decomposing the PET film.

9. The novel PET hydrolase prepared by the method for designing a novel PET hydrolase based on a deep learning algorithm according to any one of claims 1 to 8, characterized in that: The novel PET hydrolase is hydrolase HN and / or hydrolase JW; the amino acid sequence of the hydrolase HN is shown in SEQ ID NO: 1, and the amino acid sequence of the hydrolase JW is shown in SEQ ID NO:

2.

10. Use of the novel PET hydrolase as claimed in claim 9 in hydrolyzing PET.

Images