Simulation method for developing inhibitors based on the mechanism of action of antibiotics with metallo-beta-lactamases

Abstract

The present application belongs to the field of biotechnology, and particularly relates to a simulation method for developing inhibitors based on the action mechanism of antibiotics and metal beta-lactamase, comprising the following steps: obtaining a three-dimensional structure model of a complex crystal of metal beta-lactamase and antibiotics and a structure formula of an inhibitor; predicting and building a model of a complex of metal beta-lactamase and the inhibitor based on the structure of the complex of metal beta-lactamase and antibiotics; and analyzing the action sites and rules of metal beta-lactamase, antibiotics and the inhibitor. The method explores the action mechanism and action sites of metal beta-lactamase and antibiotics, finds the action rules of metal beta-lactamase and beta-lactam antibiotics in catalytic reactions, and provides a new idea and direction for the research of new inhibitors.

Description

Technical Field

[0001] This invention belongs to the field of biotechnology, specifically relating to a simulation method for developing inhibitors based on the mechanism of action of antibiotics and metallo-β-lactamases. Background Technology

[0002] In 2017, the WHO listed carbapenem-resistant Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacteriaceae as "extremely hazardous" among the superbugs posing the greatest threat to human health. Carbapenem antibiotics are hailed as the "last line of defense for human health," but infection by bacteria resistant to these drugs leaves clinical treatment largely ineffective. The mechanism by which these bacteria develop carbapenem resistance is the presence of genes encoding metallo-β-lactamases, enzymes that catalyze the hydrolysis of antibiotics. Currently, the most threatening type of metallo-β-lactamase to humans is New Delhi metallo-β-lactamase (NDM-1), because it can efficiently hydrolyze β-lactam antibiotics, and the gene encoding NDM is located on a plasmid, allowing for rapid spread among bacteria and a very high probability of mutation.

[0003] Currently, research on inhibitors all targets the NDM-1 active site. This is because during the NDM-1 hydrolysis of the substrate, the substrate reacts with the NDM-1 active site Zn. 2+ Following coordination, a reaction occurs, and the active site, the location of the hydrolysis reaction, is crucial in the NDM-1 structure. There are two main development routes for inhibitors: targeting and stripping or replacing the active site Zn. 2+ These inhibitors work by disrupting the structure of the NDM-1 active site or by mimicking the binding of substrate intermediates to the NDM-1 active site, thus preventing the substrate from entering the active site. Although there are many reports on NDM-1 inhibitors, there are currently no effective inhibitors available for clinical use, mainly because these inhibitors are either highly toxic to humans or have poor antibacterial effects. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a method and application for developing inhibitors based on the mechanism of action of antibiotics and metallo-β-lactamases.

[0005] To achieve the above-mentioned objectives, this application provides a method for developing inhibitors based on the mechanism of action of antibiotics and metallo-β-lactamases, comprising the following steps:

[0006] 1) Obtain the three-dimensional crystal structure model of the complex of metallo-β-lactamase and antibiotic and the inhibitor structure formula;

[0007] 2) Predict and construct a model of the metallo-β-lactamase-inhibitor complex to obtain the structure of the metallo-β-lactamase-inhibitor complex;

[0008] 3) Based on the complex structure, analyze the interaction patterns of antibiotics, potential inhibitors and proteins using molecular dynamics simulation methods.

[0009] Optionally, the method for predicting and constructing the metallo-β-lactamase-inhibitor complex model includes: based on the electron cloud density structure of the metallo-β-lactamase-penicillin V complex obtained by X-ray diffraction, using the Coot software in the CCP4 software package, constructing the inhibitor structure at the vacant electron cloud density of the active center of the metallo-β-lactamase crystal structure, and using the Desmond software package to minimize the energy of the complex structure model.

[0010] Optionally, the simulation method further includes performing molecular dynamics simulations on the complex model of metal β-lactamase with antibiotics and inhibitors in the three-dimensional crystal structure model.

[0011] Optionally, the molecular dynamics simulation includes:

[0012] Molecular dynamics simulations were performed on the simulation system using the Molecular Dynamics module in the Desmond software package under the NVT and NPT systems.

[0013] Optionally, the simulation method further includes: verifying the interaction mechanism and rules between metallo-β-lactamases and antibiotics and inhibitors obtained through simulation through enzyme kinetic experiments.

[0014] Optionally, the enzyme kinetic experiment includes: preparing solutions of metallo-β-lactamase with antibiotics and inhibitors, monitoring the hydrolysis reaction rates of antibiotics and inhibitors respectively, fitting the Michaelis equation with substrate concentration as the x-axis and the initial reaction rate of the enzyme-catalyzed reaction as the y-axis, and calculating the Michaelis constant, maximum reaction rate and catalytic constant through the Michaelis equation.

[0015] Optionally, the molar ratio of the metallo-β-lactamase to the antibiotic is 1:100 to 50000, and the molar ratio of the metallo-β-lactamase to the inhibitor is 1:100 to 50000.

[0016] In another aspect, this application provides the application of the simulation method described above in screening inhibitors of metallo-β-lactamases.

[0017] In another aspect, this application provides the use of inhibitors of metallo-β-lactamases at the Phe70 site and / or Trp93 site in the preparation of products for the prevention and / or treatment of diseases caused by superbugs.

[0018] Optionally, the inhibitor can interact with the Phe70 site and / or the Trp93 site to reduce the activity of metallo-β-lactamases.

[0019] The beneficial effects of this invention are as follows:

[0020] This invention explores the mechanism of action of metallo-β-lactamases with existing antibiotics and inhibitors, and discovers the main action sites (i.e., Phe70 and Trp93) and action patterns of metallo-β-lactamases with antibiotics and inhibitors in catalytic reactions, providing new directions and ideas for the research on targeted design of competitive inhibitors of metallo-β-lactamases. Attached Figure Description

[0021] Figure 1 The image shows the pET28a-NDM-1 plasmid.

[0022] Figure 2 The structural diagrams of four penicillin antibiotics with different R1 groups are shown, from left to right: ampicillin, penicillin G, penicillin V, and oxacillin;

[0023] Figure 3 The crystal morphology of the complex is shown in a, b, c, and d, which are crystal morphology diagrams of NDM-1-ampicillin, NDM-1-penicillin G, NDM-1-penicillin V, and NDM-1-oxacillin, respectively.

[0024] Figure 4 The crystal structures are complexes. a, b, c, and d are the crystal structures of NDM-1-ampicillin, NDM-1-penicillin G, NDM-1-penicillin V, and NDM-1-oxacillin, respectively. The protein is represented by a cartoon model, the zinc ion by a spherical model, and the antibiotic by a stick model.

[0025] Figure 5 The structure of inhibitor compound 1;

[0026] Figure 6 The structure of the complex formed by compound 1 and NDM-1;

[0027] Figure 7 For molecular dynamics models, a, b, c, d, and e are the molecular dynamics models of NDM-1 ampicillin, NDM-1 penicillin G, NDM-1-penicillin V, NDM-1-oxacillin, and NDM-1-compound 1, respectively.

[0028] Figure 8 To simulate the RMSD curves of the system, black represents NDM-1, gray represents antibiotics, and a, b, c, d, and e are the RMSD curves of NDM-1-ampicillin, NDM-1-penicillin G, NDM-1-penicillin V, NDM-1-oxacillin, and NDM-1-compound 1, respectively.

[0029] Figure 9To simulate the RMSF curves of the system, NDM-1 is on the left and the antibiotics are on the right. a, b, c, d, and e are the RMSF curves of NDM-1-ampicillin, NDM-1-penicillin G, NDM-1-penicillin V, NDM-1-oxacillin, and NDM-1-compound 1, respectively.

[0030] Figure 10 The ratio of the inhibitor side chain (R1) to the NDM-1 residues formed during the kinetic simulation is given, where Amp represents ampicillin, PG represents penicillin G, PV represents penicillin V, C1 represents compound 1, and Oxa represents oxacillin.

[0031] Figure 11 The data are the Michaelis constants for the hydrolysis of five substrates by NDM-1.

[0032] Figure 12 Catalytic constant data for the hydrolysis of five substrates by NDM-1;

[0033] Figure 13 Here is the general structural formula of the modified compound. Detailed Implementation

[0034] The examples provided are for illustrative purposes only and are not intended to limit the scope of the invention. Therefore, non-essential modifications and adjustments made to the embodiments by those skilled in the art based on the above description are still within the protection scope of this invention.

[0035] I. Crystal structure analysis of the complex and design and construction of the inhibitor structural model

[0036] (1) Expression and purification of NDM-1

[0037] The amino acid gene information from NDM-129 to 270 was cloned into the pET28a plasmid vector to construct the expression plasmid (e.g., Figure 1 As shown), the specific steps are the same as described in the literature: FENG H, LIU X, WANG S, Wei L, et al. The mechanism of NDM-1-catalyzed carbapenem hydrolysis is distinct from that of penicillin orcephalosporin hydrolysis[J]. Nat Commun, 2017, 8(1): 2242.

[0038] Then, the obtained expression plasmid was transformed into the engineered Escherichia coli BL21(DE3). Specifically, the engineered Escherichia coli BL21(DE3) was mixed with the NDM-1 plasmid, incubated on ice for 30 min, then heat-shocked in a water bath at 35-50℃ for 1-10 min, followed by an ice bath for 5 min, and the plasmid was transformed into the engineered bacteria.

[0039] Next, non-resistant LB liquid medium was added to the engineered bacteria and activated in a shaker at 37°C. The culture was then evenly spread onto resistant LB nutrient agar plates. After bacterial growth, single colonies were picked and cultured in resistant LB liquid medium. When the OD600 of the bacterial culture reached 0.4–0.8, an appropriate concentration of IPTG was added to induce the engineered bacteria to express NDM-1 protein. After protein expression was completed, the engineered bacteria were disrupted using a high-pressure cell disruptor to release NDM-1 into the solution. The solution was then enriched by sequentially passing NDM-1 through affinity chromatography, ion exchange chromatography, and gel filtration chromatography to remove impurities and obtain an NDM-1 solution with a purity of over 98% and a concentration >20 mg / ml.

[0040] (2) Culture of NDM-1 and penicillin antibiotic complex crystals

[0041] To study the interaction mechanism between proteins and their substrates, it is necessary to obtain crystals of the complex. A co-crystallization method is used, in which a protein solution and a substrate solution are mixed and crystallized, allowing the two to interact in solution first. Then, a precipitant is added, causing them to co-aggregate and form crystals under the action of the precipitant.

[0042] The inhibitor in this application is a β-lactam inhibitor. Existing β-lactam antibiotics mainly include penicillin antibiotics, cephalosporin antibiotics, and carbapenem antibiotics. This application takes penicillin antibiotics as an example to explore the mechanism and pattern of action between metallo-β-lactamases and penicillin antibiotics and their inhibitors. The general structural formula of penicillin antibiotics is as follows: Figure 13 As shown;

[0043] like Figure 13 As shown, this application selected four different penicillin antibiotics: ampicillin (Amp), penicillin G (PG), penicillin V (PV), and oxacillin (Oxacillin, Oxa) as research subjects (e.g. Figure 2As shown), where R1 is phenyl. NDM-1 was prepared as a 10-50 mg / mL solution, and the antibiotic as a 7-50 mg / mL solution. These were mixed at a molar ratio of 1:1 to 1:10 to obtain the mother liquor. The precipitant consisted of 0.05-10 mol / L Succinate (pH 5.0-7.5) and 15%-45% PEG3350. After mixing the mother liquor and precipitant, the mixture was incubated in a constant temperature dark room for 2-30 days to obtain the complex crystals (e.g., Figure 3 (As shown).

[0044] (3) Data collection on the crystal structure of the complex

[0045] Four composite crystals were cryopreserved in liquid nitrogen and sent to the Shanghai Synchrotron Radiation Facility. Data collection was performed using a BL18U1 linear synchrotron radiation instrument, with the crystals positioned 350 mm from the detector and the X-ray wavelength being [missing value]. Temperature 100K, exposure time 0.5s per frame, swing angle 1°, 360° data collection per set, resulting in 360 complete diffraction images per set, with all data resolutions within [specific range missing]. The crystals were found to be in space group P212121 after XDS processing. Subsequently, the three-dimensional structural model was analyzed using molecular substitution, with small molecule structures constructed by manually generating dictionary files and importing them into Coot.

[0046] (4) Analysis of the crystal structure of the complex

[0047] First, the collected crystal structure diffraction data was processed using XDS software. The diffraction data was transformed into indexed data for structure analysis through indexing, parameter correction, integration, and normalization. After obtaining the indexed data, the number of asymmetric units in each unit cell of the crystal was determined using the Matthews Probability Calculator server. Then, the phaser-MR plugin under the PHENIX software package was used to perform phase analysis of the structural data using molecular substitution, obtaining the composite structure model. Finally, the composite model was refined using coot software and the refinement plugin under the PHENIX software package until the composite structure model parameter R was corrected. free With R work To achieve the ideal value (R-work value is approximately one-tenth of the crystal resolution, R...) free R work Large, but the difference is within 0.3); finally, the complex structure is obtained (e.g., Figure 4 (As shown).

[0048] (5) Inhibitor design and structural model building

[0049] The structure of penicillin V was modified by replacing its R1 group (phenyl) with a naphthyl group to enhance its aromaticity, resulting in compound 1 (abbreviated as C1). Figure 5 (As shown). Based on the diffraction data of the NDM-1 and penicillin V complex structure, a structural model of the NDM-1 and compound 1 complex was predicted and constructed, resulting in the structural model of the NDM-1 and compound 1 complex (as shown). Figure 6 (As shown).

[0050] Specifically, the following steps are taken: using Chemdraw software to generate a Smile file of the inhibitor structure, and then obtaining a dictionary (.cif) file of the inhibitor structure. In the Coot software, based on the crystal structure of the complex of metallo-β-lactamase and penicillin V, the dictionary file of the inhibitor structure is imported into the metallo-β-lactamase structure.

[0051] Then, based on the electron cloud density structure of the metallo-β-lactamase-penicillin V complex obtained by X-ray diffraction, the inhibitor structure was constructed at the vacant electron cloud density of the active center of the metallo-β-lactamase crystal structure using the Coot software in the CCP4 package, and the energy of the complex structure model was minimized using the Desmond package.

[0052] II. Molecular Dynamics Simulations Based on Complex Structures

[0053] (1) Construction of the simulation system model

[0054] Using the above five complex structures as initial structural models, initial three-dimensional molecular dynamics models of the complex structures and water molecules, sodium ions, and chloride ions were constructed respectively using the system builder module in the Desmond molecular dynamics simulation software package (e.g., Figure 7 (As shown).

[0055] Specifically, the complex structure file was read into Maestro software. After initializing and optimizing the complex structure, the Desmond software package was called, and the system builder module was used to build the initial structural model for kinetic simulation. The complex structure was placed in a water tank, and 0.15M NaCl was added to simulate physiological saline. The water molecules were modeled using the SPC model, and a fixed amount of Na was added. + or Cl - Using the neutralization system's charge, the distance between the protein structure and the tank boundary is... Periodic boundary conditions are used.

[0056] (2) Model optimization

[0057] The initial molecular dynamics model was minimized using the Minimization module in the Desmond software package to obtain a stable model. Then, the Molecular Dynamics module in the Desmond software package was used to slowly heat the simulation system to 300K in an NVT system.

[0058] (3) Molecular dynamics simulation

[0059] Using the Molecular Dynamics module in the Desmond software package, molecular dynamics simulations were performed on the simulated system under NPT conditions of 300K and 1 bar for 100 ns each, and the motion trajectory files of the complex system were obtained.

[0060] (4) Analysis of simulation results

[0061] The Simulation Interactions Diagram module within the Desmond software package was used to analyze data obtained from molecular dynamics simulations. Information such as the NDM-1 RMSD curve, antibiotic RMSD curve, NDM-1 RMSF curve, antibiotic RMSF curve, and the interaction between NDM-1 and the antibiotic were calculated during the simulation. The RMSD curves of the simulation system are shown below. Figure 8 As shown, the RMSF curve of the simulated system is as follows: Figure 9 As shown, the proportion of π-π interactions formed between the antibiotic side chain (R1) and NDM-1 residues during the kinetic simulation is as follows: Figure 10 As shown.

[0062] from Figure 8 The RMSD results show that all five simulation systems reached a stable state during the simulation process, indicating that the results are reliable. Figure 9 The RMSF results show that the NDM-1 loop 3 structure and the antibiotic R1 group undergo significant conformational changes during the simulation, indicating a strong interaction between the two. Figure 10 Interaction data show that, except for oxacillin, the phenyl groups on the R1 groups of the other four antibiotics formed strong π-π interactions with the Phe70 residue on NDM-1 loop 3. The proportions of time spent in the simulation duration due to π-π interactions between the antibiotic side chains and Phe70 were as follows: oxacillin 0.3%, penicillin G 5.6%, ampicillin 13.5%, penicillin V 39.3%, and the compound 144.6%. Based on this, it is speculated that the stronger the aromaticity of the substrate R1 group, the stronger the interaction with Phe70, leading to an increased affinity of the antibiotic for NDM-1 (Michaelis constant K). mThe smaller the value, the slower the apparent rate of NDM-1 hydrolysis of the antibiotic (catalytic constant k). On the other hand, it is not conducive to the removal of the antibiotic hydrolysis products from the active site. cat (The smaller the value). Without considering compound 1, the π-π interaction between the phenyl side chain of penicillin V and NDM-1 is the strongest, its affinity for NDM-1 should be the highest, and the catalytic constant of penicillin V catalyzed by NDM-1 should be the lowest. After modifying the R1 group of penicillin V to a naphthyl group to obtain compound 1, due to the enhanced aromaticity of the R1 group, the π-π interaction between the naphthyl side chain of compound 1 and NDM-1 is enhanced, its affinity for NDM-1 should be the highest, and the catalytic constant of compound 1 catalyzed by NDM-1 should be the lowest.

[0063] III. Verification of Simulation Results Based on Enzyme Kinetics Experiments

[0064] NDM-1 was prepared into solutions ranging from 1 nmol / L to 2000 nmol / L. The molar ratio of NDM-1 to antibiotics was mixed between 1:100 and 1:50000. The reaction rate of antibiotic hydrolysis was monitored using a multi-functional microplate reader. The Michaelis-Menten equation was obtained by fitting the substrate concentration [s] to the x-axis and the initial reaction rate v0 to the y-axis.

[0065]

[0066] The Michaelis constant K is obtained by calculating using the Michaelis equation. m With the maximum reaction rate V max k cat Through V max Dividing by the enzyme concentration [E] yields:

[0067]

[0068] Specific values ​​are as follows: Figure 11 and Figure 12 As shown.

[0069] Since oxacillin is not a good substrate for metallo-β-lactamases, when compound 1 is not considered, the enzyme kinetic parameters of NDM-1 catalyzing different antibiotics, except for oxacillin, are consistent with our molecular dynamics simulation predictions. Penicillin V has the highest affinity for NDM-1 and the lowest catalytic constant.

[0070] After modifying the R1 group of penicillin V to a naphthyl group to obtain compound 1, enzyme kinetic parameters showed that compound 1 had a significantly increased affinity for NDM-1 and a significantly decreased catalytic constant. m With k catThe reductions were all statistically significant. These results indicate that modifying the phenyl group of the antibiotic R1 group to a more aromatic naphthyl group significantly enhances the π-π interaction between the substrate and NDM-1, increasing its affinity and reducing the rate of product detachment from the active site, thus decreasing the catalytic constant. This demonstrates that compound 1 is an effective competitive inhibitor of NDM-1, significantly inhibiting its catalytic activity. These results provide a new direction for the design of inhibitors of metallo-β-lactamases, including NDM-1. The above results demonstrate the reliability of the aforementioned molecular dynamics simulation scheme and results, indicating consistency between theory and practice.

[0071] In summary, starting from the three-dimensional molecular dynamics model of the NDM-1-antibiotic complex obtained from the crystal structure analysis, and using the Desmond kinetics software package, the study of the interaction between residues surrounding the NDM-1 active site and the antibiotic side chain (R1) shows that NDM-1 residues Phe70 and Trp93 are key residues in the dynamic interaction with the antibiotic side chain (R1). The stronger the aromaticity of the antibiotic side chain (R1), the easier it is to form π-π and hydrophobic interactions with key NDM-1 residues. Combined with enzyme kinetic experiments, it was found that the stronger the interaction, the higher the affinity of the antibiotic for NDM-1, and the lower the catalytic constant of NDM-1 in hydrolyzing the antibiotic. This finding provides a direction for designing novel inhibitors, namely, improving the interaction between the antibiotic side chain (R1) and residues surrounding the NDM-1 active site. NDM-1 residues Phe70 and Trp93 are the main interaction sites with the antibiotic side chain (R1).

[0072] Specifically, the general structural formula of the modified compound is as follows: Figure 13 As shown, where,

[0073] X is an oxygen atom, a sulfur atom, an imine group, a hydrocarbon group, or a hydrocarbon-derived group, such as X being a carbon atom, an oxygen atom, a sulfur atom, an imine group, a carbonyl group, an alkoxy group, or an alkylamine group;

[0074] R1 is a polyphenyl group, a phenyl heterocyclic aryl group, a cyclic olefin, or a derivative thereof. For example, R1 is a naphthyl group, anthracene group, or a benzo[5] nitrogen-containing heterocycle.

[0075] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for developing inhibitors based on the mechanism of action of antibiotics and metallo-β-lactamases, characterized in that, Includes the following steps: 1) Obtain the three-dimensional crystal structure model of the complex of metallo-β-lactamase and antibiotic and the structural formula of the inhibitor; 2) Predict and construct a model of the metallo-β-lactamase-inhibitor complex to obtain the structure of the metallo-β-lactamase-inhibitor complex; 3) Based on the complex structure, analyze the interaction patterns of antibiotics, potential inhibitors, and proteins using molecular dynamics simulation methods; The method for predicting and constructing the metallo-β-lactamase-inhibitor complex model includes: based on the electron cloud density structure of the metallo-β-lactamase-penicillin V complex obtained by X-ray diffraction, using the Coot software in the CCP4 software package, constructing the inhibitor structure at the vacant electron cloud density of the active center of the metallo-β-lactamase crystal structure, and using the Desmond software package to minimize the energy of the complex structure model.

2. The simulation method according to claim 1, characterized in that, Also includes: Molecular dynamics simulations were performed on the complex models of metallo-β-lactamases with antibiotics and inhibitors in the three-dimensional crystal structure model.

3. The simulation method according to claim 2, characterized in that, The molecular dynamics simulations include: Molecular dynamics simulations were performed on the simulation system using the Molecular Dynamics module in the Desmond software package under the NVT and NPT systems.

4. The simulation method according to claim 1, characterized in that, Also includes: Enzyme kinetic experiments were conducted on metallo-β-lactamases with antibiotics and inhibitors to verify the results of molecular dynamics simulations.

5. The simulation method according to claim 4, characterized in that, The enzyme kinetics experiment includes: Metallo-β-lactamases were prepared into solutions with antibiotics and inhibitors, and the hydrolysis rates of antibiotics and inhibitors catalyzed by metallo-β-lactamases were monitored. The Michaelis equation was obtained by fitting the substrate concentration as the x-axis and the initial reaction rate of the enzyme-catalyzed reaction as the y-axis. The Michaelis constant, the maximum reaction rate, and the catalytic constant were calculated using the Michaelis equation.

Images