Screening method and application of multi-drug-resistant efflux pump AcrAB-TolC inhibitor

Abstract

The invention discloses a screening method and application of a multi-drug-resistant efflux pump AcrAB-TolC inhibitor, and the screening method comprises the following steps: 1) obtaining an AcrB protein sequence through an NCBI database, carrying out homologous modeling by using a SWISS MODE website, and optimizing by using GalaxyRefine to obtain an AcrB protein model; 2) simultaneously verifying binding sites obtained from software generation and literature, and evaluating the advantages and disadvantages of a binding pocket according to the obtained 2D structural diagram to obtain an AcrB protein binding pocket; 3) aiming at the determined AcrB protein model and AcrB protein binding pocket, selecting a classic ZINC15 natural compound database to carry out high-throughput screening so as to obtain preliminarily screened small molecules; and 4) carrying out ADMET screening and drug-likeness screening on the preliminarily screened small molecules to finally obtain the target inhibitor. A compound 5-methyl-3-hexene-2-ketone is successfully screened through the method provided by the invention, and the inhibitor can significantly reduce the efflux effect of an efflux pump AcrAB-TolC on antibacterial drugs, thereby enhancing the antibacterial activity of antibiotics.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of drug screening and new use of drugs, and particularly relates to a screening method of a multi-drug resistant efflux pump AcrAB-TolC inhibitor and application thereof. BACKGROUND

[0002] The aggravation of bacterial multi-drug resistance (MDR) poses a serious challenge to global public health. Infection with multi-drug resistant bacteria not only leads to an increase in mortality, but also significantly prolongs hospitalization time, increases treatment costs and the complexity of infection control. Efflux pumps are one of the important mechanisms of bacterial drug resistance, which can actively efflux antibiotics and reduce the concentration of drugs in cells, thereby making many effective antibiotics ineffective. Efflux pump AcrAB-TolC is an important multi-drug efflux pump system in Gram-negative bacteria, which widely exists in Enterobacteriaceae bacteria, including Escherichia coli, Salmonella and Klebsiella pneumoniae, etc. Developing effective efflux pump inhibitors to restore the antibacterial activity of antibiotics is one of the key strategies to solve the problem of bacterial drug resistance. Existing efflux pump inhibitors such as PAβN have problems such as high toxicity and poor in vivo stability, which limit their clinical application. Natural products become potential candidates due to their structural diversity and low toxicity. SUMMARY

[0003] The application aims to provide a screening method of a multi-drug resistant efflux pump AcrAB-TolC inhibitor and application thereof. The application aims to discover new inhibitors to restore antibiotic efficacy through virtual screening combined with experimental verification, and to provide new strategies and methods for clinical treatment of multi-drug resistant bacterial infections.

[0004] Technical solution: In order to achieve the above application purpose, the technical solution adopted by the application is as follows:

[0005] In a first aspect, the application provides a screening method of a multi-drug resistant efflux pump AcrAB-TolC inhibitor, comprising the following steps:

[0006] 1) Obtain the AcrB protein sequence through the NCBI database, perform homology modeling using the SWISS MODE website and optimize using GalaxyRefine to obtain the AcrB protein model;

[0007] 2) Verify the binding sites obtained from software and literature at the same time, evaluate the advantages and disadvantages of the binding pocket according to the obtained 2D structure diagram, and obtain the AcrB protein binding pocket;

[0008] 3) Select the classic ZINC15 natural compound database for high-throughput screening according to the determined AcrB protein model and AcrB protein binding pocket, and obtain the preliminary screening small molecules;

[0009] 4) The primary screening small molecules are subjected to ADMET screening and drug-like screening, and finally the target inhibitors are obtained.

[0010] As a specific embodiment, in step 2), the determination of the AcrB protein binding pocket is verified by hydroxyl amide compounds according to the literature and the automatically generated binding sites by software.

[0011] As a specific embodiment, in step 2), three AcrB protein binding pockets are determined, and the key amino acids are as follows:

[0012] Binding pocket (1): MET573, MET575, GLN577, PHE617, THR624, MET662, PHE664, PHE666, ARG717, ASN719;

[0013] Binding pocket (2): Ser79, Thr91, Ser134, Ser135, LYS292, PHE617, LEU674, ASP681;

[0014] Binding pocket (3): Ser79, Thr 91, Ser134, Ser135, Ser462, Phe 617, Phe 666, Leu674, Ala 677, Asp 681;

[0015] As a specific embodiment, in step 3), the screening criteria of ZINC15 natural compound database is: Affinity≤-7kcal / mol.

[0016] As a specific embodiment, in step 4):

[0017] The criteria for ADMET screening are: water solution solubility is 4; blood brain barrier penetration is 1; intestinal absorption is 1; no hepatotoxicity; no binding to plasma proteins; no inhibition of CYP2D6; no skin irritation;

[0018] The criteria for drug-like screening are based on the five rules of Lipinski, which are: no more than 5 hydrogen bond donors; no more than 10 hydrogen bond acceptors; no more than 5 lipophilic water partition coefficients (LogP); molecular weight below 500; no more than 10 rotatable bonds.

[0019] In a second aspect, the application provides the use of the multiple drug resistance efflux pump AcrAB-TolC inhibitors obtained by the screening method in the preparation of a drug for inhibiting bacterial drug resistance.

[0020] In a third aspect, the present application provides use of 5-methyl-3-hexen-2-one in preparation of a drug for inhibiting bacterial drug resistance mediated by efflux pump AcrAB-TolC. The 5-methyl-3-hexen-2-one is obtained by the screening method described above.

[0021] As a specific embodiment, the 5-methyl-3-hexen-2-one can reduce the efflux effect of bacterial efflux pump AcrAB-TolC on antibacterial drugs, and enhance the antibacterial activity of the antibacterial drugs.

[0022] As a specific embodiment, the bacterial drug resistance is the drug resistance of Escherichia coli clinical strains CM2, CM4, CM5, CS5L, CS6L, CS9F or CS18F to florfenicol.

[0023] In a fourth aspect, the present application provides use of 5-methyl-3-hexen-2-one in combination with an antibacterial drug in preparation of a drug for treating drug-resistant bacterial infection mediated by efflux pump AcrAB-TolC, wherein the antibacterial drug is florfenicol.

[0024] As a specific embodiment, the drug-resistant bacteria are clinical strains CM2, CM4, CM5, CS5L, CS6L, CS9F or CS18F resistant to florfenicol.

[0025] Advantages: Compared with the prior art, the method of the present application obtains an inhibitor capable of inhibiting efflux pump AcrAB-TolC through computer modeling, determination of the pocket, and database screening. A compound T is successfully screened by the method of the present application, and the chemical name of the compound T is 5-methyl-3-hexen-2-one (molecular formula: C7H 12 O, molecular weight 110.15). The inhibitor can significantly reduce the efflux effect of efflux pump AcrAB-TolC on antibacterial drugs, thereby enhancing the antibacterial activity of the antibiotics. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 AcrB protein homology modeling model.

[0027] Figure 2 AcrB protein model Ramachandran plot.

[0028] Figure 3 Molecular docking results of the binding site (1) and antibiotics.

[0029] Figure 4 Molecular docking results of the binding site (2) and antibiotics.

[0030] Figure 5 Molecular docking results of the binding site (3) and antibiotics.

[0031] Figure 6 Molecular docking results for the binding site (1) with inhibitors.

[0032] Figure 7 Molecular docking results for the binding site (1) with inhibitors.

[0033] Figure 8 FIC index for checkerboard susceptibility screening.

[0034] Figure 9 Time-kill curve for florfenicol and T used in combination against CM5.

[0035] Figure 10 Standard curve for florfenicol accumulation test.

[0036] Figure 11 Accumulation test for florfenicol and T used in combination against ATCC 25922. Note: **P<0.05.

[0037] Figure 12 Accumulation test for florfenicol and T used in combination against CM5. Note: **P<0.01.

[0038] Figure 13 Toxic effect of T on Vero cells.

[0039] Figure 14 Hemolytic effect of T on red blood cells.

[0040] Figure 15 Survival curve for CM5 infected by Galleria mellonella.

[0041] Figure 16 Treatment survival curve for Galleria mellonella.

[0042] Figure 17 Bacterial load determination for Galleria mellonella infected by bacteria. Note: A: bacterial load in Galleria mellonella at 24h; B: bacterial load in Galleria mellonella at 48h; ***P<0.001; ****P<0.0001. DETAILED DESCRIPTION

[0043] The present application is further illustrated by the following examples. These examples are purely illustrative and are not to be taken as limiting the application. The application is further described below with reference to the accompanying drawings and examples.

[0044] Example 1 Method for screening of multiple drug resistant efflux pump AcrAB-TolC inhibitors

[0045] 1. Protein model building and optimization

[0046] The AcrB protein sequence was obtained from the NCBI database, and a protein model was established using SWISS MODEL. After importing multiple protein model files into the online tool GalaxyRefine, the input model was first preprocessed, including removing water molecules, repairing incorrect side chain conformations, etc., to ensure the integrity and usability of the input model. During the optimization process, GalaxyRefine performs energy evaluation and structure stability analysis for each conformation, including calculating the total energy, evaluating the reliability and compliance of the conformation, and analyzing structural features such as hydrogen bonds and disulfide bonds. These evaluation results help determine the most stable and reliable conformation, ensuring that the optimized structure is closer to the ideal state in terms of energy. Finally, a high-quality AcrB screening model is obtained, as shown in Figure 1 Model evaluation shows that 99.9% of the residues are in the Ramachandran plot allowed region, as shown in Figure 2 which verifies the reasonableness of the model.

[0047] 2. Determination of binding pockets and pocket screening

[0048] Binding pockets are the regions where small molecules interact with proteins, and their identification and screening are crucial for subsequent molecular docking. Based on the various antibacterial drugs (florfenicol, chloramphenicol, ciprofloxacin, erythromycin, gentamicin, rifampicin, and tigecycline) expelled by AcrB, we obtained the three-dimensional structures of the corresponding small molecules in the NCBI database and performed docking using AutoDockVina to obtain the binding pockets. By obtaining multiple binding pockets after docking different drugs, we evaluated the docking results and determined the binding pocket for subsequent screening of AcrB protein inhibitors based on the analysis of the number of carbon-hydrogen bonds between the protein and the antibiotics, i.e., site (1). We performed molecular docking of site (1) with the more mature AcrAB-TolC inhibitor BDM91288, NMP, and Ar5 reported in the literature to verify the effectiveness of the binding pocket again, as shown in Figure 7 .

[0049] 3. Establishment of small molecule database and molecular docking

[0050] (1) To facilitate subsequent docking, we established a small molecule ligand database based on the 270,549 small molecule structures in the natural product library downloaded from the ZINC15 database.

[0051] (2) In molecular docking, we used the Vina (v.1.1.2) software to handle tasks. This software can automatically calculate the energy of small molecules based on their structure and force field parameters. Based on this, Vina will refer to the active pocket of the protein model and perform rotation and translation operations to adjust the position of the small molecule, and then search for all possible binding site regions around it.

[0052] (3) To facilitate subsequent analysis, we first converted all structural file formats to PDBQT format. Subsequently, by calculating the van der Waals forces, electrostatic interactions, and hydrogen bonds between the protein-molecule at the site, and other key factors, the calculation results of the function score were obtained.

[0053] (4) In the process of molecular docking, we used a semi-flexible docking method, i.e., AcrAB-TolC was set as a rigid body, while the small molecule ligand was treated as a flexible body. In this process, all rotatable bonds in the ligand were sampled to fully explore its conformational space. In addition, the remaining parameters were kept at their default settings to ensure the stability and reliability of the docking process.

[0054] After this round of screening, a total of 111,159 potential compounds were obtained. It is worth noting that the docking score of Vina uses Affinity, which is used to represent the strength of the binding force. The smaller the value, the stronger the binding force. According to previous experience, when Affinity≤-7 kcal / mol, it can be determined that the binding force is strong. In view of this, for the 100,994 potential compounds obtained from the docking results, we sorted them according to the Affinity score and removed the small molecule ligands with a score greater than -7, finally 92,145 compounds successfully entered the subsequent screening link.

[0055] 4. ADMET screening

[0056] (1) In the process of drug screening and optimization, ADMET (absorption, distribution, metabolism, excretion, and toxicity) property evaluation is a key link to reduce research and development risks. By systematically analyzing the pharmacokinetic properties (such as absorption efficiency, metabolic stability) and toxicity characteristics (such as hepatotoxicity, cardiotoxicity) of candidate compounds, molecules with high drugability and controllable safety can be preferentially screened, thereby reducing the elimination rate of candidate drugs due to pharmacokinetic defects or toxicity in the clinical stage.

[0057] (2) To achieve the above goal, this study uses the ADMET prediction module of the GHDDI-AIDD platform to evaluate the drugability of the compounds screened by molecular docking. The screening parameters focus on the following six core indicators: water solubility (LogS≥-4.5), blood-brain barrier permeability (BBB score≤0.3), CYP2D6 inhibition risk (probability≤50%), hepatotoxicity (alarm negative), intestinal absorption rate (HIA≥80%), and plasma protein binding rate (PPB≤90%) to comprehensively predict the in vivo behavior and potential risks of the compounds. After this round of screening operation, 91,917 compounds that do not meet the requirements were effectively filtered out, and finally 228 potential lead compounds were successfully obtained.

[0058] 5. Drug-like screening

[0059] (1) Drug-like analysis is based on Lipinski's Rule of Five (RO5), which predicts the potential of oral absorption and bioavailability of a compound through four indicators: molecular weight (≤500 Da), octanol-water partition coefficient (LogP≤5), number of hydrogen bond donors (≤5) and acceptors (≤10).

[0060] (2) Through drug-like evaluation, candidate molecules meeting the drug-likeness criteria can be systematically screened to ensure acceptable pharmacokinetic properties (such as oral absorption rate, membrane permeability) and low toxicity risk, thereby reducing the failure rate of subsequent development.

[0061] (3) Based on the SwissADME platform, the drug-likeness of potential lead compounds screened by molecular docking is verified, focusing on RO5 parameters, polar surface area and synthetic score (SAscore≤4.5), and finally retaining candidate molecules with both activity and drug-likeness. Based on the 228 compounds obtained after drug-like screening, further drug-like screening was carried out, and 222 compounds that did not meet the above rules were filtered out, and finally 7 small molecules were successfully obtained, as shown in Table 1.

[0062] The chemical structures of these small molecules obtained through ZINC website are as follows. Among them, ZINC5355869, ZINC5193911 and ZINC100018180 have moderate drug-likeness and high safety. Among them, ZINC5193911 and ZINC100018180 can be found in the source, considering the cost problem, finally ZINC100018180, i.e. 5-methyl-3-hexen-2-one, is selected as the potential inhibitor for subsequent antibacterial activity evaluation, and it is named as T.

[0063]

[0064] Table 1 Small molecule score

[0065]

[0066] Example 2 Study on the inhibition of bacterial drug resistance of small molecule compound T

[0067] 1. Drug sensitivity test

[0068] (1) The streak test strains CM2, CM4, CM5, CS5L, CS6L, CS9F, CS18F and ATCC 25922 were inoculated on LB agar plates and cultured at 37°C for 10h, then single colonies were picked and inoculated in 3mL fresh LB broth and cultured at 37°C for 5h.

[0069] (2) Take 100 μL of bacterial solution and add it to 900 μL of fresh LB broth, mix well, then take 20 μL of the bacterial solution and add it to 20 mL of MHB broth in a 90 mm plate to dilute it 1,000 times, and use an eight-row pipette to mix well.

[0070] (3) Use an eight-row pipette to extract the bacterial solution from the plate and add it to a 96-well plate, 50 μL per well, up to the 11th column, and add MHB blank broth to the 12th column as a negative control.

[0071] (4) Add 10 μL of FFC to the first column of the 96-well plate to make the final concentration 512 μg / mL, mix well, then take 50 μL of the mixture to the next column for dilution. After dilution to the 10th column, leave 11 columns as positive controls.

[0072] (5) Seal the 96-well plate with sealing film and place it in a 37°C incubator for 20 h.

[0073] (6) Observe the results and record the minimum inhibitory concentration, the results are shown in Table 2.

[0074] Table 2 Drug Sensitivity Test Results

[0075]

[0076]

[0077] 2. Checkerboard Drug Sensitivity Test

[0078] (1) Draw a line test on the LB agar plate with strains CM2, CM4, CM5, CS5L, CS6L, CS9F, CS18F and ATCC 25922, incubate at 37°C for 10 h, then pick a single colony into 3 mL of fresh LB broth and incubate at 37°C for 5 h.

[0079] (2) Take 100 μL of bacterial solution and add it to 900 μL of fresh LB broth, mix well, then take 20 μL of the bacterial solution and add it to 20 mL of MHB broth in a 90 mm plate to dilute it 1,000 times, and use an eight-row pipette to mix well.

[0080] (3) Use an eight-row pipette to extract the bacterial solution from the plate and add it to a 96-well plate, 50 μL per well, a total of 11 columns.

[0081] (4) Add 10 μL of FFC to the first column of the 96-well plate to make the final concentration 512 μg / mL, mix well, then take 50 μL of the mixture to the next column for dilution. After dilution to the 10th column, leave 11 columns as positive controls.

[0082] (5) In 96-well plates, add corresponding from high to low concentration of T from top to bottom of each layer, a total of 7 rows and 11 columns. At this time, the 12th column of the 96-well plate is the T single-drug positive control, and the 8th row is the florfenicol positive control.

[0083] (6) Seal the 96-well plate with sealing film and put it into a 37℃ incubator for 24h.

[0084] (7) Observe the results and record the MIC.

[0085] (8) Calculate the FIC index of the bacteriostatic concentration: FIC = MIC(A combination) / MIC(antibiotic alone) + MIC(B combination) / MIC(compound alone).

[0086] (9) Result determination: FIC <0.5 is synergistic effect; when FIC is between 0.5-1, it is additive effect; when FIC is greater than 1 and less than 2, it is irrelevant effect; when FIC >2, it is antagonistic effect.

[0087] The results are as follows Figure 8 , the combination of compound T and florfenicol has synergistic effect, and the FIC of CM2, CM4, CM5, CS5L, CS6L, CS9F, CS18F and ATCC 25922 is 0.3125, 0.375, 0.25, 0.375, 0.1875, 0.375, 0.5 and 0.5, respectively, <0.5, with synergistic effect, which reduces the MIC of florfenicol by 4, 4, 8, 4, 16, 8, 4 times, respectively; the combination with chloramphenicol has FIC of 0.125, 0.5, 0.5, 0.375, 1, 0.375, 1.5, 0.75 for the above strains, respectively, among which the FIC of CS6L and ATCC 25922 is <1, which has additive effect; the combination with ciprofloxacin, meropenem, gentamicin, amikacin and tetracycline has antagonistic effect. The results show that the combination of T and florfenicol can achieve better effect.

[0088] 3. Time-kill curve

[0089] (1) Inoculate CM5 on LB agar plates and incubate at 37℃ for 8h;

[0090] (2) Pick single colonies into LB broth and incubate to the logarithmic growth phase;

[0091] (3) Set 6 groups, respectively, florfenicol concentration is 1 / 4 MIC, T concentration is 1 / 4 MIC and 1 / 8 MIC, florfenicol and 1 / 4 MIC T combination group, florfenicol and 1 / 8 MIC T combination group, blank control group without adding antibiotic. Add 30mL LB broth to each group, add 1×10 7The bacteria solution was diluted to 1.5x108CFU / mL, and the set antibiotic concentration was added. Each group was placed in a 37℃ shaking incubator, and 1 mL of bacteria solution was taken at 0, 4, 8, 12, and 24 h to measure OD 540 . According to the OD, three gradients were diluted, 100 μL of bacteria solution from each gradient was spread on a plate (three parallel settings), and it was placed in a 37℃ constant temperature incubator for 10 h. The number of colonies was observed and counted, and a time-kill curve was drawn.

[0092] (4) Result determination: compared with each drug alone, the average bacterial growth of the combination of the two drugs was reduced by ≥2 Log 10 CFU / mL, which was determined as synergistic effect; the average bacterial growth was reduced by 1-2 Log 10 CFU / mL, which was determined as additive effect; the average bacterial growth was reduced by ≤1 Log 10 CFU / mL, which was determined as no effect; and the average bacterial growth was increased by ≥2 Log 10 CFU / mL, which was determined as antagonistic effect.

[0093] As shown in Figure 9 , florfenicol (128 μg / mL) and T (1.5625 mg / mL) alone had no bactericidal effect on CM5; T (3.125 mg / mL) alone had a certain bactericidal effect; florfenicol (128 μg / mL) and T (1.5625 mg / mL) in combination had no obvious bactericidal effect; but when florfenicol (128 μg / mL) and T (3.125 mg / mL) were combined, the bacteria decreased by 4.22 log within 24 h, showing obvious bactericidal effect.

[0094] 4. Florfenicol accumulation test

[0095] (1) Inoculate CM5 and ATCC 25922 on LB agar plates and incubate at 37℃ for 8 h;

[0096] (2) Pick single colonies and inoculate in LB broth (conical flask 300 mL) and incubate at 37℃, 180 rpm overnight;

[0097] (3) Weigh 15 mL empty tube;

[0098] (4) Collect bacterial cells (transfer to 50 mL centrifuge tube at 4℃, 4,000 x g, centrifuge for 5 min);

[0099] (5) Wash the bacterial cells once with PBS (pH 7.2) (4℃, 4,000 x g, centrifuge for 5 min);

[0100] (6) Resuspend the bacterial cells in PBS to a concentration of OD 600 = 20, at least 3.5 mL;

[0101] (7) The above bacterial solution was cultured at 37°C under normal gas conditions for 10 min;

[0102] (8) 0.5 mL of the bacterial suspension was taken as a blank control, and FFC was added to the remaining 3 mL of the bacterial solution to a final concentration of 128 μg / mL;

[0103] (9) After 15 min of FFC addition, another group was added with T (50 mg / mL, to a final concentration of 3.125 mg / mL) and carbonyl cyanide m-chloro phenyl hydrazone (CCCP) (10 mM, to a final concentration of 100 μM), and 0.5 mL of the bacterial solution was taken out at 20, 25, and 30 min, respectively, and 2.5 mL of pre-cooled PBS (pre-cooled in -20°C for more than 90% ice) was added for dilution at 1 min before the countdown, and 0.5 mL of the bacterial solution was taken out into a 15 mL centrifuge tube (weighed and covered with an empty tube), and centrifuged at 4°C at 6,000 x g for 5 min;

[0104] (10) After centrifugation, the supernatant was removed and placed in an oven, and the moisture was dried for 10 min to obtain the net weight of the bacterial precipitate (total weight minus the weight of the empty tube);

[0105] (11) The bacterial precipitate was resuspended in 1.5 mL of deionized water, and the bacterial cells were lysed by repeated freezing and thawing using liquid nitrogen and a 65°C water bath, with temperature change every 3 min, for 3-5 times;

[0106] (12) After centrifugation, the supernatant was taken and centrifuged at 4°C at 6,000 x g for 15 min;

[0107] (13) After centrifugation of the supernatant, filtration was performed using a 0.22 μm filter membrane to filter into a liquid sample bottle, and standard samples of florfenicol at concentrations of 16, 8, 4, 2, 1, and 0.5 μg / mL were prepared and added to the sample bottle after filtration;

[0108] (14) 35% aqueous acetonitrile and methanol were prepared, filtered using an electric filter, 0.22 μm, and finally cleaned using ultrasonic waves;

[0109] (15) The sample was detected using a liquid chromatograph, ultraviolet light at 224 nm, sample injection amount of 20 μL, flow rate of 0.8 mL / min, and column temperature of 30°C;

[0110] (16) The concentration was calculated according to the standard curve based on the detection results of the florfenicol standard sample, and the results were expressed as: ng (florfenicol) / mg (wet weight of bacterial cells), and analyzed using the T-tset method.

[0111] The standard curve was plotted based on the detection results of the florfenicol standard sample by high-performance liquid chromatography, Figure 10The drug concentration showed a tendency of exponential increase in the corresponding test period, which was consistent with the expectation of the experiment.

[0112] As Figure 11 shown, the accumulation concentration of florfenicol in ATCC 25922 increased with time after the addition of T and CCCP. At 25 min, the accumulation concentration of the T group was higher than that of the CCCP group, and at 30 min, the trend was consistent with that at 25 min, and the florfenicol accumulation amount increased. The experiment showed that the combination of florfenicol and T could increase the accumulation of florfenicol in the bacteria, and had the effect of inhibiting the efflux of florfenicol by the efflux pump.

[0113] The results of the florfenicol accumulation test of CM5 are shown in Figure 12 After the addition of T and CCCP, the trend was consistent with the results of ATCC 25922, and the florfenicol accumulation amount of CM5 was more obviously improved. At 20 min, the florfenicol accumulation concentration of the T combination group was lower than that of the CCCP combination group, and at 25 min and 30 min, the florfenicol accumulation concentration of the T combination group was obviously improved.

[0114] 5. Cytotoxicity test

[0115] (1) Cell recovery: African monkey kidney cells (Vero) were taken out of the liquid nitrogen tank, and then immediately placed in a 37°C water bath for thawing until completely melted. Then, the cells were gently transferred to a 10 mL EP tube containing 6 mL of medium using a pipette, and centrifuged at room temperature (1,200 rpm, 3 min). The supernatant was carefully discarded, and 2 mL of new DMEM medium (containing 10% BI serum and 1% double antibody) was added, and the cells were uniformly resuspended by gentle blowing. Then, the cells in the EP tube were completely transferred to a culture dish pre-added with 6 mL of medium, and the culture dish was gently shaken to evenly distribute the cells on the bottom of the bottle. Finally, the culture dish was placed in a cell incubator (5% CO2, 37°C) for further culture, and the cells in the culture dish were drawn an 8-shaped line to shake again, and the cells were grown;

[0116] (2) After the cells were grown adherently, the medium in the culture dish was discarded, and then washed twice with 2 mL of PBS, and then the PBS was gently aspirated, and 6 mL of medium was added for further culture. After 24 h, the culture solution was changed in time, and the culture was continued under the above culture conditions;

[0117] (3) Passage: through the microscope, cells adhere to grow to about 80%, discard the culture medium, 2 mL PBS rinse twice, after cleaning, pour off the PBS, add 1 mL of trypsin (along the wall) gently shake, digestion, wait for 2-3 minutes (digestion time control), observed under a microscope, round cells, add 2 mL of 10% fetal bovine serum medium to the culture dish to stop digestion, gently blow down the cells, blow and turn the dish, finally transferred to 10 mL of EP tube. 1,200 rpm, centrifugation 3 min, discard the supernatant, add 1 mL of medium, blow evenly;

[0118] (4) Take 500 μL of cell suspension into the culture dish with 8-9 mL of medium (can be one transmission three) draw 8 characters mix, put into 37℃ incubator culture;

[0119] (5) Cell counting: take the cell suspension, dilute the corresponding multiple, take 10 μL to the counting plate hole, diagonal line counting (4 regions or 2 regions);

[0120] (6) Pave 96-well plate: if 10,000 cells per well, 100 μL of cell suspension per well, set 3-6 repeated holes, 96-well plate only take the middle 60 holes, calculate when can be calculated as 70 holes (7 mL);

[0121] (7) Assuming the cell counting value is X, 70 x 1 x 1,000 = X x Y (cell suspension volume μL). Assuming Y is 166 μL of cell suspension, take 166 μL of suspension + 6.834 mL of medium. Cell suspension cell number / mL = cell counting average value X x 10 4 ;

[0122] (8) Drug: aspirate the supernatant medium on the 96-well plate, dilute the drug (directly dilute with the base culture medium) into different concentrations (100 mg / mL, 50 mg / mL, 25 mg / mL, 12.5 mg / mL, 6.25 mg / mL, 3.125 mg / mL, 1.5625 mg / mL, 0.78125 mg / mL), after 18-24 h, the cells adhere to the wall, add 100 μL of different concentrations of drug in the form of liquid change, each group of 3 repeated holes, 37℃ incubator culture;

[0123] (9) Add CCK-8 and measure OD value: after 24 h or so after adding the drug, add 10 μL of CCK-8 solution, gently knock the edge of the culture plate to promote mixing, pay attention to avoid the generation of bubbles (bubbles may affect the accuracy of subsequent absorbance detection), then put into 37℃ incubator for 1-4 h, use the enzyme label instrument to measure OD 450 nm value, calculate the cell survival rate, the calculation method is: cell survival rate (%) = (drug group OD450 Medium group OD 450 ) / (Cell group OD 450 Medium group OD 450 ) x 100%.

[0124] As Figure 13 , the cell survival rate was 87% when T (0.9 mg / mL) was used alone in Vero cells, and the cell survival rate reached 100% when it was used in combination with florfenicol, with a 13% increase in cell survival rate. When the concentration was lower than 0.9 mg / mL, the cell survival rate reached 100% when used in combination, and the cytotoxicity was negligible. This indicates that T has good safety.

[0125] 6. Red blood cell hemolysis test

[0126] (1) Preparation of red blood cell (RBC) suspension: 5 mL of defibrillated sheep blood was centrifuged at 3,500 rpm for 10 min at 4°C, the supernatant was removed, and the precipitate was gently blown and washed with PBS, then centrifuged at 3,500 rpm for 10 min at 4°C. 300 μL of red blood cell precipitate was added to 5,700 μL of PBS to prepare a 5% red blood cell suspension.

[0127] (2) Take 5% red blood cell suspension, 100 μL in 96-well plate, add 100 mg / mL (final concentration) compound 100 μL in the first column, add 50 mg / mL in the second column, and so on to the 10th column, add 100 μL / well of PBS solution as negative control in the 11th column, and add 100 μL of 0.1% Triton X-100 as positive control in the last column, with three parallel controls.

[0128] (3) After placing at 37°C for 1 h, centrifuge at 3,500 rpm for 10 min, take 100 μL of supernatant in a new 96-well plate, measure the absorbance at 540 nm, and calculate the red blood cell hemolysis rate, which is calculated as follows: Hemolysis rate (%) = (Drug group OD 540 -5% blank red blood cell suspension OD 540 ) / (0.1% Triton X-100 OD 540 -5% blank red blood cell suspension OD 540 ) x 100%.

[0129] As Figure 14 shown, when the concentration of T was 25 mg / mL, a slight hemolytic phenomenon occurred, with a hemolysis rate of 0.99%; while at a concentration of 12.5 mg / mL and below, the hemolysis rate was 0. Therefore, it can be concluded that when the concentration is 12.5 mg / mL or below, T will not cause hemolysis, and has high safety.

[0130] 7. Determination of lethal dose of Galleria mellonella infected by CM5 strain

[0131] CM5 was inoculated in LB broth and cultured at 37℃, 200rpm for 4h to make the bacteria grow to logarithmic phase (10 8 CFU / mL), 4,000xg, 10min to collect the bacteria, resuspend the bacteria in PBS, collect the bacteria again and discard the supernatant, then dilute the bacteria solution 10 times to 10 7 CFU / mL, 10 6 CFU / mL, 10 5 CFU / mL, 10 4 CFU / mL.

[0132] Galleria mellonella were randomly divided into 6 groups, 10 in each group, and injected with 10μL bacteria solution with a concentration of 10 8 CFU / mL, 10 7 CFU / mL, 10 6 CFU / mL, 10 5 CFU / mL, 10 4 CFU / mL in the left hind leg using a microsyringe, and the infected larvae were placed in sterile plates. At the same time, two groups of controls were set up, one group was injected with 0.9% saline and the other group was not treated as a blank control group. The plates were incubated at 37℃ in the dark for 120h, and the survival of Galleria mellonella was observed every 24h. When the number of dead larvae in the control group was no more than 2, the experimental results were reliable. The optimal infection dose was the amount of bacteria that caused 80% mortality of the larvae. According to the number of surviving and dead larvae within 120h, the survival curve was drawn using Graphpad software and the results were analyzed.

[0133] The survival curve of Galleria mellonella infected by CM5 strain is shown in Figure 15 When the injected bacteria solution reached 1x10 8 CFU, the larvae all died within 48h; when the injected bacteria solution reached 1x10 7 CFU, the larvae reached 80% mortality within 72h; when the injected bacteria solution reached 1x10 6 CFU and 1x10 5 CFU, the mortality of the larvae reached 40% and 10% respectively within 72h; when the injected bacteria solution reached 1x10 4 CFU, the survival rate of the larvae reached 90% within 120h. According to the survival curve of the Galleria mellonella larvae infection model, it can be determined that the 80% lethal dose of bacteria solution concentration of CM5 to the larvae within 120h is 1x10 7 CFU / mL, and the sub-lethal dose of bacteria solution concentration is 1x10 6 CFU / mL.

[0134] 8. Treatment of infected Galleria mellonella model with combination drugs

[0135] According to the lethal dose results of the wax moth, the infection and treatment bacterial liquid concentration was determined to be 10 7 CFU / mL.

[0136] The wax moths were divided into 6 groups, 10 in each group, and each group was injected according to the concentration determined above, and 2h later, drug treatment was given. Florfenicol was given at a dose of 20mg / kg, T was given at a dose of 20mg / kg, and the combined drug group was given florfenicol and T at the same time. A negative control group was set up, one group was given an equal volume of normal saline, and one group was not treated. The larvae were placed in a 37°C incubator in the dark for 120h. The survival of the wax moths was observed every 24h, and the number was counted. According to the number of survival and death of the larvae within 120h, the survival curve was drawn using Graphpad software and the results were analyzed.

[0137] The survival curve of the wax moth larvae infected with the CM5 strain and treated with drugs is shown in Figure 16 After the wax moth larvae were infected with the strain and treated with drugs, the survival rate of the combined treatment group of florfenicol (20mg / kg) and T (20mg / kg) was significantly higher than that of the single drug group, indicating that the combined use of florfenicol and T had a significant antibacterial effect in vivo.

[0138] 9. Determination of bacterial load in wax moth infected in vivo

[0139] The wax moths were divided into 5 groups, 30 in each group, and each group was injected according to the concentration determined above, and 2h later, drug treatment was given. Florfenicol was given at a dose of 20mg / kg, T was given at a dose of 20mg / kg, and the combined drug group was given florfenicol and T at the same time. A negative control group was set up, one group was given an equal volume of normal saline, and one group was not treated. The larvae were placed in a 37°C incubator in the dark. 24h sampling was taken once every day, 8 live larvae were randomly selected from each group into a 2mL EP tube, 1mL PBS and 5-8 steel balls were added, and it was placed in a tissue grinder, 90Hz, 30s grinding for 3-5 times. 100μL was taken into 900μL PBS and diluted by 6 times in sequence, 100μL was taken from each gradient and plated (3 parallels were set), and after overnight culture in a 37°C incubator, the number of bacterial colonies was observed and counted. The operation of the previous day was repeated at 48h, and the graph was drawn using Graphpad software.

[0140] The results of the bacterial load in the wax moth larvae infected with the CM5 strain and treated with drugs in vivo are shown in Figure 17As shown in the figure, the bacterial load in the G. mellonella group in the florfenicol and T combination group was significantly lower than that in the PBS group, the florfenicol alone group, and the T group at 24 hours. This suggests that T can effectively increase the sensitivity of florfenicol to E. coli and reduce the efflux pump's effect on florfenicol efflux. At 48 hours after administration, the bacterial load in the single-drug group was still significantly higher than that in the combination group, demonstrating the drug's persistent inhibitory effect.

[0141] In summary, this study, based on homology modeling and molecular docking techniques, screened a novel natural product small molecule, T (5-methyl-3-hexen-2-one), and evaluated its potential as an efflux pump inhibitor through in vitro and in vivo experiments. The results showed that the combination of T and florfenicol reduced the MIC of drug-resistant strains by 4-16 times; time-kill curve results showed that the combination of florfenicol (128 μg / mL) and T (3.125 mg / mL) reduced bacterial counts by 4.22 log within 24 hours. 10 CFU / mL, demonstrating a significant bactericidal effect. Safety test results showed that when T was used at a concentration below 12.5 mg / mL, it did not cause hemolysis of erythrocytes, and when the concentration was below 0.9 mg / mL, it had almost no effect on Vero cells, indicating that T has a high safety profile. Accumulation test results showed that the combination of florfenicol and T increased the accumulation of florfenicol in bacteria and inhibited the efflux of florfenicol by the AcrAB-TolC efflux pump. In vivo tests showed that the 80% lethal dose of CM5 against greater wax moth larvae within 120 hours was 1×10 7 CFU / mL. The combined use of florfenicol (20 mg / kg) and T (20 mg / kg) increased the cure rate against G. mellonella by 40%, demonstrating significant in vivo antibacterial activity. Furthermore, at 24 hours, the bacterial load in the florfenicol-T combination group was significantly lower than that in the PBS, florfenicol, and T groups, and the trend was consistent with that observed at 48 hours, indicating that T can effectively enhance the sensitivity of E. coli to florfenicol. This study provides theoretical and experimental evidence for the development of low-toxic, highly effective efflux pump inhibitors.

[0142] The above describes the implementation mode of the present invention in detail with reference to the accompanying drawings and specific embodiments. However, the present invention is not limited to the above implementation mode. Various changes can be made within the scope of knowledge possessed by ordinary technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A method for screening inhibitors of the multidrug resistance efflux pump AcrAB-TolC, characterized in that: The following steps are involved: 1) The AcrB protein sequence was obtained from the NCBI database, homology modeling was performed using the SWISS MODE website, and the model was optimized using GalaxyRefine to obtain the AcrB protein model; 2) The binding sites generated by the software and obtained from the literature were simultaneously verified, and the quality of the binding pocket was evaluated based on the obtained 2D structure diagram to obtain the AcrB protein binding pocket; 3) Based on the determined AcrB protein model and AcrB protein binding pocket, high-throughput screening was performed using the classic ZINC15 natural compound database to obtain preliminary screened small molecules; 4) The initially screened small molecules are subjected to ADMET screening and drug-like screening to ultimately obtain the target inhibitors.

2. The method for screening inhibitors of the multidrug resistance efflux pump AcrAB-TolC according to claim 1, characterized in that In step 2), the binding pockets of the AcrB protein are determined by using hydroxyamide compounds to verify the advantages and disadvantages of these pockets based on the binding sites automatically generated by the literature and software.

3. The method for screening inhibitors of the multidrug resistance efflux pump AcrAB-TolC according to claim 1, characterized in that: In step 2), three AcrB protein binding pockets were identified, and their key amino acids are as follows: Binding pocket (1): MET573, MET575, GLN577, PHE617, THR624, MET662, PHE664, PHE666, ARG717, ASN719; Binding pocket (2): Ser79, Thr91, Ser134, Ser135, LYS292, PHE617, LEU674, ASP681; Binding pocket (3): Ser79, Thr 91, Ser134, Ser135, Ser462, Phe 617, Phe 666, Leu674, Ala677, Asp 681.

4. The method for screening inhibitors of the multidrug resistance efflux pump AcrAB-TolC according to claim 1, characterized in that: In step 3), the screening criteria of the ZINC15 natural compound database is: Affinity ≤ -7 kcal / mol.

5. The method for screening inhibitors of the multidrug resistance efflux pump AcrAB-TolC according to claim 1, characterized in that: In step 4): The ADMET screening criteria are: aqueous solubility of 4; blood-brain barrier penetration of 1; intestinal absorption of 1; no hepatotoxicity; no plasma protein binding; no CYP2D6 inhibition; and no skin irritation. The criteria for drug-likeness screening are based on Lipinski's rule of five, which are: no more than 5 hydrogen bond donors; no more than 10 hydrogen bond acceptors; the lipid-water partition coefficient (LogP) is no more than 5; the molecular weight is below 500; and the number of rotatable bonds is no more than 10.

6. Use of a multidrug-resistant efflux pump AcrAB-TolC inhibitor obtained by screening according to any one of claims 1 to 5 in the preparation of a drug for inhibiting bacterial drug resistance.

7. Application of 5-methyl-3-hexen-2-one in the preparation of drugs that inhibit bacterial resistance mediated by the efflux pump AcrAB-TolC.

8. The use according to claim 7, characterized in that The 5-methyl-3-hexene-2-one can reduce the efflux effect of the bacterial efflux pump AcrAB-TolC on antibacterial drugs and enhance the antibacterial activity of the antibacterial drugs.

9. The use according to claim 7, characterized in that The bacterial drug resistance is the drug resistance of Escherichia coli clinical strains CM2, CM4, CM5, CS5L, CS6L, CS9F or CS18F to florfenicol.

10. Use of 5-methyl-3-hexen-2-one in combination with an antibacterial drug in the preparation of a drug for treating drug-resistant bacterial infections mediated by the efflux pump AcrAB-TolC, characterized in that: The antibacterial drug is florfenicol.

Images