Method for screening compound for treating atrial fibrillation and application

Abstract

The invention belongs to the field of medicinal chemistry, and particularly relates to a method for screening a compound for treating atrial fibrillation and application. Comprising the following steps: (1) selection and pretreatment of a CaMK II delta crystal structure; (2) selecting the center of a butt-joint box and setting the size of the butt-joint box to generate the butt-joint box; (3) selecting and downloading a to-be-screened compound library; (4) performing docking calculation by adopting a molecular docking algorithm in molecular docking software according to the docking box and the to-be-screened compound library, and retaining the first 10% of compounds with higher docking scores; (5) treating the compounds which are 10% before docking scoring by using a Lipinski class drug five rule, and excluding the compounds which are low in drug-likeness; and (6) carrying out ADMET property prediction on the compounds conforming to the drug-like V rule, and screening the active compounds under the conditions. According to the invention, on the basis of analysis and virtual screening of AI on a large amount of compound data, the compounds obakurol A and urolithin C with high affinity binding characteristics with atrial fibrillation related target CaMKII delta are obtained, and calcium ion homeostasis in myocardial cells can be effectively regulated.

Description

Technical Field

[0001] The present invention belongs to the field of medicinal chemistry, and in particular relates to a method for screening compounds for treating atrial fibrillation and its application. Background Art

[0002] Atrial fibrillation (AF) is a common cardiac arrhythmia that affects a large number of people worldwide, with its prevalence increasing significantly with age. AF is a serious health risk, not only affecting the heart's normal rhythm and pumping function but also causing a range of serious complications. Studies have shown that AF increases the incidence of ischemic stroke by nearly fivefold, while significantly increasing the risk of heart failure, myocardial infarction, chronic kidney disease, cognitive impairment, and dementia, severely reducing patients' quality of life and increasing mortality and healthcare costs.

[0003] In drug development, target identification is crucial. CaMKⅡδ plays a key role in the excitation-contraction coupling of cardiomyocytes and is closely related to the maintenance of normal cardiac rhythm. Studies have found that the activity of CaMKⅡδ in the myocardial tissue of patients with atrial fibrillation is significantly higher than normal levels. Its excessive activation can lead to an imbalance in calcium homeostasis in cardiomyocytes, causing delayed afterdepolarization and triggered activity. It also regulates gene expression and promotes cardiomyocyte hypertrophy and fibrosis. It plays a key role in the occurrence and development of atrial fibrillation and is an important target for the treatment of atrial fibrillation.

[0004] Traditional drug screening methods are time-consuming, labor-intensive, and costly, making them inadequate for the rapid discovery of effective drugs. The development of artificial intelligence (AI) technology has shown tremendous potential in drug development. AI technology can rapidly process and analyze vast amounts of data. By learning and predicting the relationship between compound structure and activity, it can efficiently screen for compounds with potential bioactivity, significantly shortening the drug development cycle and improving R&D efficiency. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for screening compounds for treating atrial fibrillation and its application.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] A method for screening a compound for treating atrial fibrillation, comprising the following steps:

[0008] (1) Obtain the crystal structure of the CaMKⅡδ-small molecule complex from the protein structure database and preprocess the crystal structure of the CaMKⅡδ-small molecule complex;

[0009] (2) Using molecular docking software, select the docking box center and set the docking box size to generate the docking box;

[0010] (3) Select and download the compound library to be screened based on the natural product compound database;

[0011] (4) Based on the docking box and the library of compounds to be screened, the molecular docking algorithm in the molecular docking software was used to perform docking calculations, and the top 10% of compounds with higher docking scores were retained to preliminarily exclude compounds with weaker binding ability to the CaMKⅡδ active pocket;

[0012] (5) The top 10% of the docking-scoring compounds were processed using Lipinski's five drug-like rules to exclude compounds with low drug-like properties and obtain compounds that meet the five drug-like rules;

[0013] (6) ADMET properties were predicted for compounds that met the five drug-like rules, and active compounds that simultaneously met the three conditions of TPSA < 140, hERG < 0.7, and H-HT < 0.7 were screened.

[0014] In step (1), the pre-processing is specifically as follows:

[0015] After obtaining the crystal structure of the CaMKⅡδ-small molecule complex, the crystal structure of the CaMKⅡδ-small molecule complex is input into molecular modeling software, and the CaMKⅡδ protein as the macromolecular target is subjected to three-dimensional structural analysis in the molecular modeling software, crystallization water molecules in the protein structure are removed, and hydrogen atoms are added to the entire protein molecule to generate structural data;

[0016] The structural data processed by the molecular modeling software were input into the molecular docking software, and the molecular docking software was used to perform energy minimization calculations on the protein structure, so that the total energy of the system was reduced to the local minimum, and the energy-stable and preliminarily optimized CaMKⅡδ-small molecule complex crystal structure was obtained.

[0017] In step (2), a docking box is generated by molecular docking software, the position of the docking box is set with the ligand small molecule in the pre-processed CaMKⅡδ-small molecule complex crystal structure as the center, and the box size is set according to the size of the ligand small molecule in the CaMKⅡδ-small molecule complex crystal structure.

[0018] In step (3), the mol2 file of the natural product component is downloaded from the natural product compound database, and the structure is converted into an sdf structure using chemical informatics software to obtain M molecular structures; which are then merged with K molecules in the natural product compound database to obtain N molecular structures as a compound library for virtual screening.

[0019] One method screened and obtained compounds for treating atrial fibrillation: phellodendronol A and urolithin C; the compound structures are:

[0020]

[0021] An application of the compound, wherein phellodendronol A and urolithin C are used to prepare a drug for treating atrial fibrillation.

[0022] The drug for treating atrial fibrillation is a potent inhibitor of CaMKⅡδ, a target related to atrial fibrillation.

[0023] The atrial fibrillation drug of the present invention is prepared by mixing chloroquine A / urolithin C with a pharmaceutically acceptable carrier to prepare a clinically acceptable injection, oral preparation, transdermal absorption preparation, and mucosal absorption preparation.

[0024] The atrial fibrillation suppressing drug of the present invention can inhibit the excessive activation of the atrial fibrillation-related target CaMKⅡδ, can effectively regulate the calcium ion homeostasis in myocardial cells, and can significantly shorten the duration of atrial fibrillation.

[0025] The beneficial effects of the present invention are:

[0026] 1. This invention, through AI-based analysis and virtual screening of a large amount of compound data, identified compounds such as phellodendron A and urolithin C that exhibit high-affinity binding to the atrial fibrillation-associated target CaMKIIδ, effectively regulating calcium homeostasis in cardiomyocytes. Compared to random compound screening methods used in existing techniques, this method is more targeted and efficient.

[0027] 2. This invention uses AI virtual screening platform to perform molecular docking and affinity scoring on natural compound library. Maestro software, which has powerful molecular simulation and analysis capabilities, can accurately predict the binding mode and affinity of compounds with CaMKⅡδ targets, greatly improving screening efficiency and accuracy compared to traditional manual screening.

[0028] 3. The present invention conducts drug-likeness analysis and ADMET prediction on the screened compounds. Drug-likeness analysis is performed based on Lipinski's five rules, and the ADMETlab3.0 platform is used to evaluate indicators such as topological polar surface area (TPSA), cardiotoxicity (hERG), and hepatotoxicity (H-HT). Compounds with drug potential are further screened from a large number of virtual screening compounds, avoiding failures in subsequent R&D due to poor drug properties, saving R&D costs and time.

[0029] 4. The compounds screened by this invention, chalcedinol A and urolithin C, are used to treat atrial fibrillation by regulating calcium homeostasis within cardiomyocytes and improving cardiac electrophysiological properties. In cell-based experiments, they effectively reduced calcium overload within cardiomyocytes; in animal models, administration of the compounds significantly reduced the duration of atrial fibrillation, providing a new drug option for the treatment of atrial fibrillation and potentially addressing the limitations of existing treatments. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 These are the specific values of phellodendronate A and urolithin C in the drug-like five rules and ADMET-related parameters in Example 1 of the present invention.

[0031] Figure 2 The inhibitory effects of phellodendron A and urolithin C on calcium overload in neonatal rat atrial myocytes in Example 3 of the present invention are as follows: (A) the inhibitory effect of phellodendron A on calcium overload in neonatal rat atrial myocytes; (B) the inhibitory effect of urolithin C on calcium overload in neonatal rat atrial myocytes.

[0032] Figure 3 The effects of the two compounds in Example 4 of the present invention on the duration of atrial fibrillation in an animal experimental model are shown; (A) the effect of phellodendron A on the duration of atrial fibrillation in rats; and (B) the effect of urolithin C on the duration of atrial fibrillation in rats. DETAILED DESCRIPTION

[0033] The following examples are used to illustrate the technical solutions of the present invention, but the scope of protection of the present invention is not limited thereto. Unless otherwise specified, the experimental methods adopted are conventional technical means in the field.

[0034] The screening method of the present invention comprises the following steps:

[0035] (1) Selection and preprocessing of CaMKⅡδ crystal structure: The crystal structure of CaMKⅡδ-small molecule complex was obtained from the Protein Data Bank database, and the crystal structure of CaMKⅡδ-small molecule complex was preprocessed.

[0036] After obtaining the crystal structure of the CaMKⅡδ-small molecule complex, the crystal structure of the CaMKⅡδ-small molecule complex was input into Discovery Studio software (molecular modeling software), and a three-dimensional structural analysis of the CaMKⅡδ protein as a macromolecular target was performed in the Discovery Studio software. The crystal water molecules in the protein structure were removed, and hydrogen atoms were added to the entire protein molecule. The structural data processed by the Discovery Studio software was input into Maestro software (molecular docking software), and the structure was subjected to energy minimization calculations using the Protein Preparation module in the Maestro software to reduce the total energy of the system (complex composed of CaMKⅡδ and small molecules) to a local minimum, thereby obtaining a preliminary optimized CaMKⅡδ-small molecule complex structure that is energy-stable.

[0037] Protein Data Bank, PDB code is 5VLO, resolution is

[0038] (2) Select the docking box center and set the docking box size to generate the docking box;

[0039] The Receptor Grid Generation module in Maestro was used to generate the docking box. The position of the docking box was set with the ligand small molecule in the crystal structure of the CaMKⅡδ-small molecule complex as the center, and the box size was set with the ligand small molecule in the crystal structure of the CaMKⅡδ-small molecule complex as the reference.

[0040] (3) Select and download the library of compounds to be screened; select a natural compound library covering a variety of chemical structures and active ingredients as the object of virtual screening.

[0041] We downloaded mol2 files of natural product compounds from the TCMBank natural product database and converted them into SDF files using Openbabel, resulting in 75,639 molecular structures. These structures were then merged with 16,627 molecules from the commercial natural product database obtained from TargetMol. Repeated compounds in the two libraries were removed, resulting in a library of 91,383 molecular structures for virtual screening.

[0042] (4) Using the high-precision molecular docking algorithm in Maestro software (such as Glide's standard precision docking mode SP or high-precision docking mode XP, etc.) to perform docking calculations, compounds with weak binding ability to the CaMKⅡδ active pocket were initially excluded, and the top 10% of compounds with high docking scores were retained;

[0043] During the docking process, the compounds in the library of compounds to be screened downloaded in step (3) are docked into the docking box (active pocket) in sequence, and the binding ability of each compound to the docking box (active pocket) is scored.

[0044] (5) The top 10% of the docking-scoring compounds were processed using Lipinski's five drug-like rules to exclude compounds with low drug-like properties and obtain compounds that met the five drug-like rules;

[0045] (6) ADMET properties of compounds that meet the five drug-like rules are predicted, and compounds with good ADMET properties are screened, that is, compounds that simultaneously meet the three conditions of TPSA < 140, hERG < 0.7, and H-HT < 0.7.

[0046] The activity of the compounds screened by the above screening method was verified by the following method:

[0047] (1) The CaMKⅡδ inhibitory activity of the screened active compounds was evaluated using time-resolved fluorescence resonance energy transfer (TR-FRET).

[0048] (2) Using the calcium overload model of neonatal rat atrial myocytes as the research object, the inhibitory effect of the compound on calcium overload in neonatal rat atrial myocytes was evaluated by observing the changes in free calcium in atrial myocytes before and after administration.

[0049] (3) Using the atrial fibrillation rat model as the research object, the therapeutic effect of the compound on atrial fibrillation was evaluated by calculating the changes in the duration of atrial fibrillation before and after administration.

[0050] Example 1: AI-driven virtual screening

[0051] Virtual screening process

[0052] 1. Selection and preprocessing of the CaMKⅡδ crystal structure: The crystal structure of CaMKⅡδ (PDB: 5VLO) was obtained from the Protein Data Bank (PDB). The structure was preprocessed using Discovery Studio and Maestro software.

[0053] After obtaining the crystal structure of the CaMKⅡδ-small molecule complex, the crystal structure of the CaMKⅡδ-small molecule complex was input into Discovery Studio software (molecular modeling software), and a three-dimensional structural analysis of the CaMKⅡδ protein as a macromolecular target was performed in the Discovery Studio software. The crystal water molecules in the protein structure were removed, and hydrogen atoms were added to the entire protein molecule. The structural data processed by the Discovery Studio software was input into Maestro software (molecular docking software), and the structure was subjected to energy minimization calculations using the Protein Preparation module in the Maestro software to reduce the total energy of the system (complex composed of CaMKⅡδ and small molecules) to a local minimum, thereby obtaining a preliminary optimized CaMKⅡδ-small molecule complex structure that is energy-stable.

[0054] Protein Data Bank, PDB code is 5VLO, resolution is

[0055] 2. Select the docking box center and set the docking box size to generate the docking box;

[0056] The docking box was generated using the Receptor Grid Generation module in Maestro. The position of the docking box was set with the ligand small molecule in the crystal structure of the CaMKⅡδ-small molecule complex as the center, and the box size was set with the ligand small molecule size in the crystal structure of the CaMKⅡδ-small molecule complex as the reference.

[0057] 3. Database selection: A natural compound library covering a variety of chemical structures and active ingredients is selected as the object of virtual screening.

[0058] We downloaded mol2 files of natural product compounds from the TCMBank natural product database and converted them into SDF files using Openbabel, resulting in 75,639 molecular structures. These structures were then merged with 16,627 molecules from the commercial natural product database obtained from TargetMol. Repeated compounds in the two libraries were removed, resulting in a library of 91,383 molecular structures for virtual screening.

[0059] 4. Initial screening: Use The company's Maestro software was used for virtual screening, selecting the top 10% of compounds with the highest docking scores to preliminarily exclude compounds with weaker binding to the CaMKⅡδ active pocket. High-precision molecular docking algorithms within Maestro (such as Glide's standard precision docking mode SP or high precision docking mode XP) were used for docking calculations to preliminarily exclude compounds with weaker binding to the CaMKⅡδ active pocket and retain the top 10% of compounds with higher docking scores.

[0060] During the docking process, the compounds in the library of compounds to be screened downloaded in step (3) are docked into the docking box (active pocket) in sequence, and the binding ability of each compound to the docking box (active pocket) is scored.

[0061] 5. Drug-likeness filtering: The top 10% of the compounds with the highest docking scores were processed using the Lipinski five-drug-like rule (MW < 500; nHA ≤ 10; nHD ≤ 5; LogP ≤ 5; nRot ≤ 10), and molecules that did not meet the drug-like properties were eliminated to obtain compounds that met the five-drug-like rule.

[0062] 6. ADMET prediction: ADMETlab3.0 was used to predict the pharmacokinetic properties of the compounds, including topological polar surface area TPSA <140, intestinal absorption cardiotoxicity hERG <0.7, and hepatotoxicity H-HT <0.7, etc., to understand the absorption, distribution, metabolism, excretion and toxicity of the compounds in the body. Two compounds, urolithin A and urolithin C, were screened and obtained. These two compounds performed well in terms of the five drug-like rules and ADMET-related parameters, see Figure 1 , and has the value of further research and development as a potential drug.

[0063] Example 2: Evaluation of CaMKIIδ inhibitory activity in vitro

[0064] The inhibitory activity of chloroquine A (commercially available) and urolithin C (commercially available) on CaMKIIδ was evaluated using time-resolved fluorescence resonance energy transfer (TR-FRET). The specific process is as follows:

[0065] First, full-length GST-tagged human CaMKIIδ (purchased from CarnaBiosciences, purity >85% and activity of 5860 nmol / min / mg) was mixed with terbium (Tb 3 +)-labeled anti-GST antibody (donor) and BODIPY-labeled probe ligand (receptor) were mixed in kinase buffer (50mM HEPES pH 7.6, 10mM MgCl2, 1mM EGTA, 0.01% Brij-35, 0.1mM DTT), where HEPES maintained pH stability and MgCl2 was added. 2+ As a kinase cofactor, EGTA chelates free calcium ions to control activation conditions, and DTT maintains the protein in a reduced state. After the protein-probe complex is incubated on ice in the dark for 30 minutes to form a binding system, 8 concentration gradients of test compounds (4-fold dilution, 100nM→0.0061nM) are added and incubated at room temperature for 1 hour. TR-FRET signals are then detected using an EnVision microplate reader (PerkinElmer): Tb is excited by a 340nm laser. 3 +, its energy is transferred to BODIPY (emission 520nm), short-lived background fluorescence is eliminated by time delay, and it is detected simultaneously by CFP495 (Tb) and 520nm (BODIPY) dual channels.

[0066] Inhibition rate = (T-μ L ) / (μ H -μ L )×100%, where T is the test hole signal, μ H and μ L The IC values are the mean values of 0% and 100% inhibition controls, respectively. 50 The values were fitted by the four-parameter logistic equation of XLfit software, and the IC values of the two compounds, urolithin C and phloxetine A, were calculated. 50 The values were 7.6 nM and 8.2 nM, respectively, both less than 10 nM, indicating that urolithin C and chloroquine A are potent inhibitors of CaMKⅡδ.

[0067] Example 3: Verification of the effects of compounds on cardiomyocytes at the cellular level

[0068] Neonatal rat atrial myocytes were cultured for 96 hours and treated with 1.0 μmol / L of the calcium ion introduction agent ionomycin to establish a calcium overload model. The atrial myocytes were loaded with the calcium ion indicator Fluo-3 / AM, and the changes in free calcium in the atrial myocytes were observed under a laser confocal microscope. The experiment was divided into four groups: the intervention group was pretreated with phellodendron A or urolithin C (0.25 μmol / L (low dose), 0.5 μmol / L (medium dose), 1.0 μmol / L (high dose)) for 30 minutes and then stimulated with ionomycin; the calcium overload group was only given ionomycin; the control group was pretreated with 1.0 μmol / L of the CaMKIIδ inhibitor KN93 (commercially available) for 30 minutes and then stimulated with ionomycin. The inhibitory effect on calcium overload in neonatal rat atrial myocytes was shown in Figure 3. Figure 2 .

[0069] Depend on Figure 2 As can be seen, ionomycin significantly increased intracellular calcium fluorescence, but pre-treatment with varying concentrations of urolithin C or urolithin A significantly reduced the ionomycin-induced increase in fluorescence intensity. Therefore, urolithin A and urolithin C can reduce ionomycin-induced calcium overload in rat atrial myocytes. (Inhibition rate = (fluorescence value of the calcium overload group - fluorescence value of the compound-treated calcium overload group) / fluorescence value of the calcium overload group × 100%).

[0070] Example 4: Verification of the therapeutic effect of the compound on atrial fibrillation at the animal level

[0071] Sixty clean-grade SD rats weighing 220-300g were randomly divided into a blank group, an atrial fibrillation group (AF), and a low-, medium-, or high-dose group (4.2, 12.6, or 37.8 mg / kg / d) of phellodendron A or urolithin C. Half of the rats were male and half were female, with 12 rats in each group. An AF animal model was established by tail vein injection of an ACh-CaCl2 mixture. Lead II electrocardiograms were recorded using a biosignal acquisition and analysis system. After successful modeling, each treatment group was given phellodendron A or urolithin C orally for 4 weeks. The duration of AF was recorded. Figure 3 .

[0072] Depend on Figure 3 As can be seen, compared with the blank group, the duration of atrial fibrillation in each model group was significantly prolonged; compared with the atrial fibrillation group, the duration of atrial fibrillation in each treatment group was significantly shortened, and gradually decreased with increasing dose. These results indicate that urolithin A and urolithin C can effectively inhibit the duration of atrial fibrillation in animal experimental models and have a good therapeutic effect on atrial fibrillation.

[0073] As shown above, in cell experiments, urolithin C and urolithin A can effectively reduce calcium overload in myocardial cells. In animal models, administration of urolithin C and urolithin A can significantly shorten the duration of atrial fibrillation.

[0074] The present invention uses AI technology to design and screen effective compounds for treating atrial fibrillation, which has significant advantages in many aspects. In terms of therapeutic efficacy, urolithin A and urolithin C can precisely target CaMKⅡδ, a key target in atrial fibrillation, and strongly regulate calcium ion homeostasis in cardiomyocytes. They can reduce calcium overload in cell experiments and shorten the duration of atrial fibrillation attacks in animal experiments, greatly alleviating atrial fibrillation symptoms. In terms of R&D efficiency, AI technology can process massive amounts of data in a very short time, greatly shortening the R&D cycle compared to traditional screening, which takes months or even years. In addition, it combines drug-like analysis and ADMET prediction to accurately screen compounds, avoiding ineffective research, reducing resource waste, and accelerating the R&D process. In terms of drug properties, AI screening and related evaluations have shown that the compounds have suitable drug properties based on Lipinski's five rules, which are beneficial for absorption, distribution, metabolism, and excretion in the body. At the same time, ADMET prediction can pre-evaluate toxicity and ensure drug efficiency and safety. In terms of industrial value, it provides a new option for atrial fibrillation treatment and can be combined with existing methods to improve efficacy. AI-based screening technology opens up new R&D paths to promote industrial technology upgrades. The simple and efficient preparation method facilitates large-scale production and reduces costs, playing a significant role in promoting the development of the atrial fibrillation treatment industry.

Claims

1. A method for screening a compound for treating atrial fibrillation, characterized in that: The following steps are involved: (1) Obtain the crystal structure of the CaMKⅡδ-small molecule complex from the protein structure database and preprocess the crystal structure of the CaMKⅡδ-small molecule complex; (2) Using molecular docking software, select the docking box center and set the docking box size to generate the docking box; (3) Select and download the compound library to be screened based on the natural product compound database; (4) Based on the docking box and the library of compounds to be screened, the molecular docking algorithm in the molecular docking software was used to perform docking calculations, and the top 10% of compounds with higher docking scores were retained to preliminarily exclude compounds with weaker binding ability to the CaMKⅡδ active pocket; (5) The top 10% of the docking-scoring compounds were processed using Lipinski's five drug-like rules to exclude compounds with low drug-like properties and obtain compounds that meet the five drug-like rules; (6) ADMET properties were predicted for compounds that met the five drug-like rules, and active compounds that simultaneously met the three conditions of TPSA < 140, hERG < 0.7, and H-HT < 0.7 were screened.

2. The method for screening a compound for treating atrial fibrillation according to claim 1, characterized in that: In step (1), the pre-processing is specifically as follows: After obtaining the crystal structure of the CaMKⅡδ-small molecule complex, the crystal structure of the CaMKⅡδ-small molecule complex is input into molecular modeling software, and the CaMKⅡδ protein as the macromolecular target is subjected to three-dimensional structural analysis in the molecular modeling software, crystallization water molecules in the protein structure are removed, and hydrogen atoms are added to the entire protein molecule to generate structural data; The structural data processed by the molecular modeling software were input into the molecular docking software, and the molecular docking software was used to perform energy minimization calculations on the protein structure, so that the total energy of the system was reduced to the local minimum, and the energy-stable and preliminarily optimized CaMKⅡδ-small molecule complex crystal structure was obtained.

3. The method for screening a compound for treating atrial fibrillation according to claim 1, characterized in that: In step (2), a docking box is generated by molecular docking software, the position of the docking box is set with the ligand small molecule in the pre-processed CaMKⅡδ-small molecule complex crystal structure as the center, and the box size is set according to the size of the ligand small molecule in the CaMKⅡδ-small molecule complex crystal structure.

4. The method for screening a compound for treating atrial fibrillation according to claim 1, characterized in that: In step (3), the mol2 file of the natural product component is downloaded from the natural product compound database, and the structure is converted into an sdf structure using chemical informatics software to obtain M molecular structures; which are then merged with K molecules in the natural product compound database to obtain N molecular structures as a compound library for virtual screening.

5. A compound obtained by screening according to the method of claim 1: phellodendronol A and urolithin C for treating atrial fibrillation; the compound structure is:

6. Use of the compound according to claim 5, characterized in that: Huangbai alcohol A and urolithin C are used for preparing medicines for treating atrial fibrillation.

7. The use according to claim 6, characterized in that The drug for treating atrial fibrillation is a potent inhibitor of CaMKⅡδ, a target related to atrial fibrillation.

8. The use according to claim 6 or 7, characterized in that The atrial fibrillation drug is prepared by mixing chloroquine A / urolithin C with a pharmaceutically acceptable carrier to prepare clinically acceptable injections, oral preparations, transdermal absorption preparations, and mucosal absorption preparations.

Images