MiRNA (micro Ribonucleic Acid) marker for liver injury caused by antituberculous drugs and application of miRNA marker

Abstract

According to the invention, 23 key miRNAs are identified in patients with liver injury caused by taking antituberculous drugs. Compared with patients who cannot cause liver injury by taking antituberculous drugs, the miRNAs are obviously differentially expressed in the patients who cannot cause liver injury by taking antituberculous drugs. Furthermore, five miRNAs are selected from the key miRNAs to construct a diagnosis model for drug-induced liver injury caused by tuberculosis and antituberculosis drugs. The model can accurately diagnose tuberculosis, and also can accurately diagnose drug-induced liver injury caused by antituberculosis drugs.

Description

Technical Field

[0001] The present invention relates to the fields of transcriptomics, bioinformatics and disease diagnosis, and in particular to the application of a miRNA marker for liver damage induced by anti-tuberculosis drugs in diagnosis. Background Art

[0002] Drug-induced liver injury (DILI) caused by anti-tuberculosis drugs is a serious problem that can lead to acute liver failure or even death in severe cases. The risk of liver injury is particularly increased with the use of first-line anti-tuberculosis drugs (isoniazid, pyrazinamide), which not only affects patients' quality of life but can also lead to treatment interruptions, changes in treatment regimens, and the development of drug resistance (Liu, Y. et al., Chem Biol Interact, 2023.376:p.110439). Therefore, early and specific identification of anti-tuberculosis (TB)-induced DILI to prevent severe liver injury or even liver failure and to enable early intervention is of great clinical significance. Currently, the diagnosis of DILI caused by anti-tuberculosis drugs relies primarily on exclusionary diagnosis. However, based on diagnostic criteria, a definitive diagnosis can be made in the majority of cases. However, since anti-TB drugs are usually used in combination with multiple drugs, and often with concomitant medication, it is difficult to determine which anti-TB drug is causing DILI when liver function damage occurs during anti-TB treatment, making it difficult to stop or change the drug early (Arbex, MA et al., "J Bras Pneumol", 2010.36(5):p.626-40).

[0003] Glutamate dehydrogenase (GLDH), ferrochelatase (FECH), and liver enzymes are considered to be potential biomarkers of liver damage caused by anti-tuberculosis drugs. Among them, the combined detection of GLDH and FECH can improve the specificity and accuracy of diagnosis and reduce the false positive rate (He, B. et al., Br J Clin Pharmacol, 2023.89(10): p.3092-3104.); liver enzymes and other liver function indicators will increase with the prolonged action of anti-tuberculosis drugs. Although these indicators can help determine the severity and prognosis of TB-DILI, they usually change only when liver tissue damage reaches a certain level, and cannot be used for early diagnosis of TB-DILI. Therefore, finding earlier, more liver-specific, and more sensitive TB-DILI biomarkers is a hot topic in current research.

[0004] Studies have found that changes in multiple microRNAs can be detected early in animal models of liver damage induced by anti-tuberculosis drugs, suggesting that they can be used as biomarkers for the diagnosis of TB-DILI (Singh, AK et al., Drug Discov Today, 2021.26(5):p.1245-1255; Liang, S. et al., Front Immunol, 2022.13:p.987018). MicroRNAs are secreted from liver tissue into the blood in the form of exosomes, which are tissue-specific and have high detection sensitivity. When the liver is damaged, the content and types of microRNAs contained in liver tissue and blood circulation can change. Therefore, exosomal microRNAs become a more ideal biomarker. Important microRNA biomarkers of DILI include microRNA-122, but it is also elevated in DILI caused by other drugs, so it is not possible to judge based solely on changes in microRNA-122. It is likely that a specific microRNA spectrum is needed to diagnose TB-DILI.

[0005] In summary, there is an urgent need in the art for a microRNA profile for diagnosing liver damage induced by anti-tuberculosis drugs and a diagnostic system constructed based on the microRNA profile. Summary of the Invention

[0006] The purpose of the present invention is to provide a microRNA marker for diagnosing liver damage caused by anti-tuberculosis drugs and a diagnostic system constructed based on the microRNA spectrum.

[0007] In a first aspect, the present invention provides a use of a miRNA or a detection reagent thereof, (i) as a marker for detecting drug-induced liver injury caused by taking anti-tuberculosis drugs; and / or (ii) for preparing a diagnostic reagent or kit for detecting drug-induced liver injury caused by taking anti-tuberculosis drugs;

[0008] Wherein, the miRNA is selected from the following group:

[0009] (A) any one marker selected from A1 to A5, or a combination thereof: (A1) hsa-miR-151a-3p; (A2) hsa-miR-22-3p; (A3) hsa-miR-148a-3p; (A4) hsa-miR-122-5p; (A5) hsa-miR-223-3p;

[0010] (B) any one marker selected from B1 to B18, or a combination thereof: (B1) hsa-miR-4433b-5p; (B2) hsa-miR-151a-5p; (B3) hsa-miR-151b; (B4) hsa-miR-548j-5p; (B5) hsa-miR-192-5p; (B6) hsa-miR-146a-5p; (B7) hsa-miR-370-3p; (B8) hsa-miR-744-5p; (B9) hsa-miR -382-5p; (B10)hsa-miR-335-5p; (B11)hsa-miR-1307-3p; (B12)hsa-miR-409-3p; (B13)hsa-let-7d-3p; (B14)h sa-miR-629-5p; (B15) hsa-miR-379-5p; (B16) hsa-miR-432-5p; (B17) hsa-miR-134-5p; (B18) hsa-miR-199b-3p;

[0011] (C) A combination of one or more markers from A1 to A5 and one or more markers from B1 to B18.

[0012] In another preferred embodiment, the miRNA is selected from the group consisting of hsa-miR-151a-3p, hsa-miR-22-3p, hsa-miR-148a-3p, hsa-miR-122-5p, hsa-miR-223-3p, hsa-miR-192-5p, or a combination thereof.

[0013] In another preferred embodiment, the miRNA is selected from the group consisting of hsa-miR-151a-3p, hsa-miR-22-3p, hsa-miR-148a-3p, hsa-miR-122-5p, hsa-miR-223-3p, or a combination thereof.

[0014] In another preferred embodiment, the miRNA is exosomal miRNA, preferably serum exosomal miRNA.

[0015] In another preferred embodiment, the miRNA is derived from mammals.

[0016] In another preferred embodiment, the miRNA is derived from rodents, primates or humans.

[0017] In another preferred embodiment, the miRNA is derived from tuberculosis patients taking anti-tuberculosis drugs.

[0018] In another preferred embodiment, the anti-tuberculosis drugs include: isoniazid and rifampicin.

[0019] In another preferred embodiment, the detection is the detection of an in vitro sample.

[0020] In another preferred embodiment, the in vitro sample includes: a blood sample, a serum sample, or a combination thereof.

[0021] In another preferred embodiment, the diagnostic reagent includes: a miRNA-specific binding molecule, a miRNA-specific antibody, a miRNA-specific primer, a miRNA probe or a chip.

[0022] In another preferred embodiment, the diagnostic agent is coupled to or carries a detectable label.

[0023] In another preferred embodiment, the detectable label is selected from the following group: a chromophore, a chemiluminescent group, a fluorophore, an isotope, an enzyme, or a combination thereof.

[0024] In another preferred embodiment, the diagnostic reagents include: antibodies, primers, probes, sequencing libraries, nucleic acid chips or protein chips.

[0025] In another preferred embodiment, the diagnostic reagent further comprises a pharmaceutically acceptable carrier, diluent or excipient.

[0026] In another preferred embodiment, the kit further comprises a label or instructions.

[0027] In another preferred embodiment, the label or instructions indicate that the kit is used to detect drug-induced liver injury caused by taking anti-tuberculosis drugs.

[0028] A second aspect of the present invention provides a kit for detecting drug-induced liver injury caused by taking anti-tuberculosis drugs, the kit comprising:

[0029] a first container, and a detection reagent for detecting miRNA located in the first container;

[0030] Wherein, the miRNA is selected from the following group:

[0031] (A) any one marker selected from A1 to A5, or a combination thereof: (A1) hsa-miR-151a-3p; (A2) hsa-miR-22-3p; (A3) hsa-miR-148a-3p; (A4) hsa-miR-122-5p; (A5) hsa-miR-223-3p;

[0032] (B) any one marker selected from B1 to B18, or a combination thereof: (B1) hsa-miR-4433b-5p; (B2) hsa-miR-151a-5p; (B3) hsa-miR-151b; (B4) hsa-miR-548j-5p; (B5) hsa-miR-192-5p; (B6) hsa-miR-146a-5p; (B7) hsa-miR-370-3p; (B8) hsa-miR-744-5p; (B9) hsa-miR -382-5p; (B10)hsa-miR-335-5p; (B11)hsa-miR-1307-3p; (B12)hsa-miR-409-3p; (B13)hsa-let-7d-3p; (B14)h sa-miR-629-5p; (B15) hsa-miR-379-5p; (B16) hsa-miR-432-5p; (B17) hsa-miR-134-5p; (B18) hsa-miR-199b-3p;

[0033] (C) a combination of one or more markers from A1 to A5 and one or more markers from B1 to B18;

[0034] A second container, and a detection reagent for detecting other miRNA located in the second container, wherein the miRNA includes microRNA-122.

[0035] In another preferred embodiment, the kit further comprises a label or instructions.

[0036] In another preferred embodiment, the label or instructions indicate that the kit is used to detect drug-induced liver injury caused by taking anti-tuberculosis drugs.

[0037] In another preferred embodiment, the detecting of drug-induced liver injury caused by taking anti-tuberculosis drugs includes determining the possibility of drug-induced liver injury caused by taking anti-tuberculosis drugs.

[0038] In another preferred example, the judgment includes preliminary judgment.

[0039] The third aspect of the present invention provides a use of a miRNA inhibitor for preparing a drug or pharmaceutical composition for preventing / treating drug-induced liver injury caused by taking anti-tuberculosis drugs;

[0040] wherein the miRNA is selected from the group consisting of: hsa-miR-122-5p; hsa-miR-151a-3p; hsa-miR-192-5p; hsa-miR-148a-3p; hsa-miR-223-3p; hsa-miR-22-3p; hsa-miR-146a-5p; hsa-miR-370-3p; hsa-miR-744-5p; hsa-miR-382-5p; hsa-miR-335-5p; hsa-miR-1307-3p; hsa-miR-409-3p; hsa-let-7d-3p; hsa-miR-629-5p; hsa-miR-379-5p; hsa-miR-432-5p; h sa-miR-134-5p; hsa-miR-199b-3p; hsa-miR-4433b-5p; hsa-miR-151a-5p; hsa-miR-151b; hsa-miR-548j-5p.

[0041] In another preferred embodiment, the inhibitors include: RNA enzymes and DNA expression inhibitors.

[0042] In another preferred embodiment, the RNase comprises an RNase linked to a primer targeting the miRNA.

[0043] In another preferred embodiment, the DNA expression inhibitor is an inhibitor that inhibits the miRNA encoding gene.

[0044] A fourth aspect of the present invention provides a pharmaceutical composition comprising:

[0045] (1) miRNA inhibitors; wherein the miRNA is selected from the group consisting of: hsa-miR-122-5p; hsa-miR-151a-3p; hsa-miR-192-5p; hsa-miR-148a-3p; hsa-miR-223-3p; hsa-miR-22-3p; hsa-miR-146a-5p; hsa-miR-370-3p; hsa-miR-744-5p; hsa-miR-382-5p; hsa-miR-3 35-5p; hsa-miR-1307-3p; hsa-miR-409-3p; hsa-let-7d-3p; hsa-miR-629-5p; hsa-miR-379-5p; hsa-miR-432-5 p; hsa-miR-134-5p; hsa-miR-199b-3p; hsa-miR-4433b-5p; hsa-miR-151a-5p; hsa-miR-151b; hsa-miR-548j-5p;

[0046] (2) anti-tuberculosis drugs; and

[0047] (3) Pharmaceutically acceptable carriers, diluents or excipients.

[0048] In another preferred embodiment, the anti-tuberculosis drugs include: isoniazid, rifampicin, and pyrazinamide.

[0049] In another preferred embodiment, the weight ratio of component (C1) to component (C2) is in the range of 100:1-0.01:1, preferably 10:1-0.1:1, and more preferably 2:1-0.5:1.

[0050] In another preferred embodiment, in the pharmaceutical composition, the content of the component (C1) is 1%-99%, preferably 10%-90%, and more preferably 30%-70%.

[0051] In another preferred embodiment, in the pharmaceutical composition, the content of component (C2) is 1%-99%, preferably 10%-90%, and more preferably 30%-70%.

[0052] In another preferred embodiment, the dosage form of the pharmaceutical composition includes: injection form and oral dosage form.

[0053] In another preferred embodiment, the oral dosage forms include: tablets, capsules, films, and granules.

[0054] In another preferred embodiment, the dosage form of the pharmaceutical composition includes: sustained-release agent and non-sustained-release agent.

[0055] A fifth aspect of the present invention provides a medicine kit, comprising:

[0056] A first container, and a miRNA inhibitor located in the first container; wherein the miRNA is selected from the group consisting of: hsa-miR-122-5p; hsa-miR-151a-3p; hsa-miR-192-5p; hsa-miR-148a-3p; hsa-miR-223-3p; hsa-miR-22-3p; hsa-miR-146a-5p; hsa-miR-370-3p; hsa-miR-744-5p; hsa-miR-382-5p; hsa -miR-335-5p; hsa-miR-1307-3p; hsa-miR-409-3p; hsa-let-7d-3p; hsa-miR-629-5p; hsa-miR-379-5p; hsa-miR-43 2-5p; hsa-miR-134-5p; hsa-miR-199b-3p; hsa-miR-4433b-5p; hsa-miR-151a-5p; hsa-miR-151b; hsa-miR-548j-5p;

[0057] A second container, and an anti-tuberculosis drug located in the second container.

[0058] In another preferred embodiment, the first container and the second container are the same or different containers.

[0059] In another preferred embodiment, the medicine kit further comprises a label or instructions.

[0060] In another preferred embodiment, the label or instructions state that the miRNA inhibitor is used in combination with an anti-tuberculosis drug to prevent / alleviate drug-induced liver injury caused by taking the anti-tuberculosis drug.

[0061] In a sixth aspect, the present invention provides a method for determining whether a patient taking an anti-tuberculosis drug has drug-induced liver injury or is at risk of drug-induced liver injury, or whether a patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking an anti-tuberculosis drug, the method comprising the following steps:

[0062] 1) Detection step: Detect the patient's miRNA;

[0063] 2) Comparison step: comparing the miRNA expression level detected in step 1) with the reference value;

[0064] 3) Judgment step: If the expression level of the miRNA detected in step 1) is significantly changed relative to the reference value, a conclusion is drawn that the patient taking the anti-tuberculosis drug has drug-induced liver injury or is at risk of drug-induced liver injury, or that the patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking the anti-tuberculosis drug;

[0065] Wherein, the miRNA is selected from the following group:

[0066] (A) any one marker selected from A1 to A5, or a combination thereof: (A1) hsa-miR-151a-3p; (A2) hsa-miR-22-3p; (A3) hsa-miR-148a-3p; (A4) hsa-miR-122-5p; (A5) hsa-miR-223-3p;

[0067] (B) any one marker selected from B1 to B18, or a combination thereof: (B1) hsa-miR-4433b-5p; (B2) hsa-miR-151a-5p; (B3) hsa-miR-151b; (B4) hsa-miR-548j-5p; (B5) hsa-miR-192-5p; (B6) hsa-miR-146a-5p; (B7) hsa-miR-370-3p; (B8) hsa-miR-744-5p; (B9) hsa-miR -382-5p; (B10)hsa-miR-335-5p; (B11)hsa-miR-1307-3p; (B12)hsa-miR-409-3p; (B13)hsa-let-7d-3p; (B14)h sa-miR-629-5p; (B15) hsa-miR-379-5p; (B16) hsa-miR-432-5p; (B17) hsa-miR-134-5p; (B18) hsa-miR-199b-3p;

[0068] (C) a combination of one or more markers from A1 to A5 and one or more markers from B1 to B18;

[0069] The reference value is the expression level of the corresponding serum exosomal miRNA in normal people, or the expression level of the corresponding serum exosomal miRNA in patients taking anti-tuberculosis drugs but without drug-induced liver injury.

[0070] In another preferred embodiment, the miRNA is selected from the group consisting of hsa-miR-151a-3p, hsa-miR-22-3p, hsa-miR-148a-3p, hsa-miR-122-5p, hsa-miR-223-3p, hsa-miR-192-5p, or a combination thereof.

[0071] In another preferred embodiment, the miRNA is selected from the group consisting of hsa-miR-151a-3p, hsa-miR-22-3p, hsa-miR-148a-3p, hsa-miR-122-5p, hsa-miR-223-3p, or a combination thereof.

[0072] In another preferred embodiment, the “detected miRNA expression level is significantly changed relative to the reference value” means that, compared with the reference value:

[0073] The expression level of hsa-miR-4433b-5p increased;

[0074] hsa-miR-151a-5p expression was decreased;

[0075] hsa-miR-151b expression decreased

[0076] The expression level of hsa-miR-548j-5p was decreased;

[0077] hsa-miR-122-5p expression increased;

[0078] The expression of hsa-miR-151a-3p increased;

[0079] hsa-miR-192-5p expression increased;

[0080] The expression of hsa-miR-148a-3p increased;

[0081] hsa-miR-223-3p expression increased;

[0082] hsa-miR-22-3p expression increased;

[0083] The expression level of hsa-miR-146a-5p increased;

[0084] The expression of hsa-miR-370-3p increased;

[0085] The expression level of hsa-miR-744-5p increased;

[0086] hsa-miR-382-5p expression increased;

[0087] The expression level of hsa-miR-335-5p increased;

[0088] The expression level of hsa-miR-1307-3p increased;

[0089] hsa-miR-409-3p expression increased;

[0090] hsa-let-7d-3p expression increased;

[0091] The expression level of hsa-miR-629-5p increased;

[0092] hsa-miR-379-5p expression was decreased;

[0093] hsa-miR-432-5p expression increased;

[0094] The expression of hsa-miR-134-5p increased;

[0095] The expression level of hsa-miR-199b-3p was increased.

[0096] In another preferred embodiment, the miRNA is exosomal miRNA, preferably serum exosomal miRNA.

[0097] In another preferred embodiment, the anti-tuberculosis drugs include: isoniazid and rifampicin.

[0098] In another preferred embodiment, the detection is the detection of an in vitro sample.

[0099] In another preferred embodiment, the in vitro sample includes: a blood sample, a serum sample, or a combination thereof.

[0100] In another preferred embodiment, the miRNA is detected using diagnostic reagents, which include: miRNA-specific binding molecules, miRNA-specific antibodies, miRNA-specific primers, miRNA probes or chips.

[0101] In another preferred embodiment, the diagnostic agent is coupled to or carries a detectable label.

[0102] In another preferred embodiment, the detectable label is selected from the following group: a chromophore, a chemiluminescent group, a fluorophore, an isotope, an enzyme, or a combination thereof.

[0103] In another preferred embodiment, the diagnostic reagents include: antibodies, primers, probes, sequencing libraries, nucleic acid chips or protein chips.

[0104] In another preferred embodiment, the diagnostic reagent further comprises a pharmaceutically acceptable carrier, diluent or excipient.

[0105] In a seventh aspect, the present invention provides a device for determining whether a patient taking anti-tuberculosis drugs has drug-induced liver injury or is at risk of drug-induced liver injury, or whether a patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking anti-tuberculosis drugs, the device comprising the following modules:

[0106] 1) Detection module: a module for detecting the patient's miRNA;

[0107] 2) Comparison module: a module for comparing the expression level of the miRNA detected in step 1) with the reference value;

[0108] 3) Judgment module: If the expression level of the miRNA detected in step 1) is significantly changed relative to the reference value, a module for concluding that the patient taking the anti-tuberculosis drug has drug-induced liver injury or is at risk of drug-induced liver injury, or that the patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking the anti-tuberculosis drug;

[0109] Wherein, the miRNA is selected from the following group:

[0110] (A) any one marker selected from A1 to A5, or a combination thereof: (A1) hsa-miR-151a-3p; (A2) hsa-miR-22-3p; (A3) hsa-miR-148a-3p; (A4) hsa-miR-122-5p; (A5) hsa-miR-223-3p;

[0111] (B) any one marker selected from B1 to B18, or a combination thereof: (B1) hsa-miR-4433b-5p; (B2) hsa-miR-151a-5p; (B3) hsa-miR-151b; (B4) hsa-miR-548j-5p; (B5) hsa-miR-192-5p; (B6) hsa-miR-146a-5p; (B7) hsa-miR-370-3p; (B8) hsa-miR-744-5p; (B9) hsa-miR -382-5p; (B10)hsa-miR-335-5p; (B11)hsa-miR-1307-3p; (B12)hsa-miR-409-3p; (B13)hsa-let-7d-3p; (B14)h sa-miR-629-5p; (B15) hsa-miR-379-5p; (B16) hsa-miR-432-5p; (B17) hsa-miR-134-5p; (B18) hsa-miR-199b-3p;

[0112] (C) a combination of one or more markers from A1 to A5 and one or more markers from B1 to B18;

[0113] The reference value is the expression level of the corresponding serum exosomal miRNA in normal people, or the expression level of the corresponding serum exosomal miRNA in patients taking anti-tuberculosis drugs but without drug-induced liver injury.

[0114] In another preferred embodiment, the miRNA is selected from the group consisting of hsa-miR-151a-3p, hsa-miR-22-3p, hsa-miR-148a-3p, hsa-miR-122-5p, hsa-miR-223-3p, hsa-miR-192-5p, or a combination thereof.

[0115] In another preferred embodiment, the miRNA is selected from the group consisting of hsa-miR-151a-3p, hsa-miR-22-3p, hsa-miR-148a-3p, hsa-miR-122-5p, hsa-miR-223-3p, or a combination thereof.

[0116] In another preferred embodiment, the “detected miRNA expression level is significantly changed relative to the reference value” means that, compared with the reference value:

[0117] The expression level of hsa-miR-4433b-5p increased;

[0118] hsa-miR-151a-5p expression was decreased;

[0119] hsa-miR-151b expression decreased

[0120] The expression level of hsa-miR-548j-5p was decreased;

[0121] hsa-miR-122-5p expression increased;

[0122] The expression of hsa-miR-151a-3p increased;

[0123] hsa-miR-192-5p expression increased;

[0124] The expression of hsa-miR-148a-3p increased;

[0125] hsa-miR-223-3p expression increased;

[0126] hsa-miR-22-3p expression increased;

[0127] The expression level of hsa-miR-146a-5p increased;

[0128] The expression of hsa-miR-370-3p increased;

[0129] The expression level of hsa-miR-744-5p increased;

[0130] hsa-miR-382-5p expression increased;

[0131] The expression level of hsa-miR-335-5p increased;

[0132] The expression level of hsa-miR-1307-3p increased;

[0133] hsa-miR-409-3p expression increased;

[0134] hsa-let-7d-3p expression increased;

[0135] The expression level of hsa-miR-629-5p increased;

[0136] hsa-miR-379-5p expression was decreased;

[0137] hsa-miR-432-5p expression increased;

[0138] The expression of hsa-miR-134-5p increased;

[0139] The expression level of hsa-miR-199b-3p was increased.

[0140] In another preferred embodiment, the device may further include (4) an output module, which outputs the conclusion of the judgment module (3) to a clinician or a patient.

[0141] In another preferred embodiment, the miRNA is exosomal miRNA, preferably serum exosomal miRNA.

[0142] In another preferred embodiment, the anti-tuberculosis drugs include: isoniazid and rifampicin.

[0143] In another preferred embodiment, the detection is the detection of an in vitro sample.

[0144] In another preferred embodiment, the in vitro sample includes: a blood sample, a serum sample, or a combination thereof.

[0145] In another preferred embodiment, the miRNA is detected using diagnostic reagents, which include: miRNA-specific binding molecules, miRNA-specific antibodies, miRNA-specific primers, miRNA probes or chips.

[0146] In another preferred embodiment, the detection reagent is coupled with or carries a detectable label.

[0147] In another preferred embodiment, the detectable label is selected from the following group: a chromophore, a chemiluminescent group, a fluorophore, an isotope, an enzyme, or a combination thereof.

[0148] In another preferred embodiment, the diagnostic reagents include: antibodies, primers, probes, sequencing libraries, nucleic acid chips or protein chips.

[0149] In another preferred embodiment, the diagnostic reagent further comprises a pharmaceutically acceptable carrier, diluent or excipient.

[0150] In an eighth aspect, the present invention provides a method for constructing a diagnostic model for tuberculosis and / or drug-induced liver injury caused by anti-tuberculosis drugs, the method comprising the steps of:

[0151] (1) Providing samples containing miRNA from tuberculosis patients with normal liver function and tuberculosis patients with drug-induced liver injury caused by anti-tuberculosis drugs;

[0152] (2) Quantifying the miRNAs in the two samples to obtain miRNA expression data; comparing the expression data of the same miRNA in the two samples to obtain differentially expressed miRNAs;

[0153] (3) screening the differentially expressed miRNAs to obtain candidate miRNAs;

[0154] (4) The miRNA expression data and candidate miRNAs are used to train the support vector machine (SVM) model to obtain a diagnostic model.

[0155] In another preferred embodiment, the sample is a serum sample.

[0156] In another preferred embodiment, the miRNA is exosomal miRNA, preferably serum exosomal miRNA.

[0157] In another preferred embodiment, the anti-tuberculosis drugs include: isoniazid, rifampicin. In another preferred embodiment, in step (1), it includes: isolating the miRNA from the sample.

[0158] In another preferred embodiment, the miRNA is separated from the sample by size exclusion chromatography.

[0159] In another preferred embodiment, before step (2), the method further includes: identifying the exosomal miRNA in the two samples.

[0160] In another preferred embodiment, the identification includes: morphological identification; particle size identification; marker identification; or a combination thereof.

[0161] In another preferred embodiment, the morphological identification is performed using a transmission electron microscope.

[0162] In another preferred embodiment, the particle size identification is performed using nanoflow cytometry.

[0163] In another preferred embodiment, a miRNA detection reagent is used for marker identification.

[0164] In another preferred embodiment, the detection reagent is coupled with or carries a detectable label.

[0165] In another preferred embodiment, the detectable label is selected from the following group: a chromophore, a chemiluminescent group, a fluorophore, an isotope, an enzyme, or a combination thereof.

[0166] In another preferred embodiment, the detection reagent is an antibody.

[0167] In another preferred embodiment, in step (3), the screening is performed through literature mining and database.

[0168] In another preferred embodiment, in step (3), the candidate miRNA is related to tuberculosis and / or liver damage.

[0169] In another preferred embodiment, the candidate miRNA is selected from the following group or a combination thereof: hsa-miR-4433b-5p; hsa-miR-151a-5p; hsa-miR-151b; hsa-miR-548j-5p; hsa-miR-122-5p; hsa-miR-151a-3p; hsa-miR-192-5p; hsa-miR-148a-3p; hsa-miR-223-3p; hsa-miR-22-3p; hsa-miR-1 46a-5p; hsa-miR-370-3p; hsa-miR-744-5p; hsa-miR-382-5p; hsa-miR-335-5p; hsa-miR-1307-3p; hsa-miR-409 -3p; hsa-let-7d-3p; hsa-miR-629-5p; hsa-miR-379-5p; hsa-miR-432-5p; hsa-miR-134-5p; hsa-miR-199b-3p.

[0170] In another preferred embodiment, in step (4), the method further includes the step of using a recursive feature elimination (RFE) algorithm to obtain preferred miRNAs, thereby obtaining a diagnostic model.

[0171] In another preferred embodiment, step (4) includes the following steps:

[0172] (4.1) Using miRNA expression data and candidate miRNAs for SVM model training to obtain the weights of candidate miRNAs;

[0173] (4.2) Using the recursive feature elimination (RFE) algorithm, the miRNAs with the smallest weight or the lowest weight ranking are removed to obtain the preferred miRNAs;

[0174] (4.3) Repeat steps (4.1) and (4.2) to perform model training. When the model performance no longer improves, terminate the training to obtain a diagnostic model.

[0175] In another preferred embodiment, the preferred miRNAs include hsa-miR-22-3p, hsa-miR-151a-3p, hsa-miR-148a-3p, hsa-miR-122-5p and hsa-miR-223-3p.

[0176] A ninth aspect of the present invention provides a system for diagnosing tuberculosis and / or drug-induced liver injury caused by anti-tuberculosis drugs, comprising:

[0177] An input module, wherein the input module is configured to input data, wherein the input data includes: miRNA expression data and miRNA type of the object to be tested;

[0178] a diagnostic module configured to execute a diagnostic model for tuberculosis and / or drug-induced liver injury caused by anti-tuberculosis drugs, thereby obtaining a diagnostic result for the subject; the diagnostic model being constructed using the method described in the eighth aspect of the present invention;

[0179] An output module is configured to output a diagnosis result of the diagnosis module.

[0180] It should be understood that within the scope of the present invention, the above-mentioned technical features of the present invention and the technical features described in detail below (such as in the embodiments) can be combined with each other to form new or preferred technical solutions. Due to space limitations, they will not be listed here one by one. BRIEF DESCRIPTION OF THE DRAWINGS

[0181] Figure 1Identification of serum exosomes is shown. (A) Under electron microscopy, serum exosomes appear as cup- or saucer-shaped, membrane-enclosed microstructures with well-defined edges. These structures are round or nearly round, with diameters ranging from 30 to 150 nanometers (nm). (B) Serum exosomes express exosomal positive protein markers such as Tsg101 / Alix / CD9 / HSP70, but lack the exosomal negative protein marker Calnexin (right column). (C) Detection results show that the exosome size distribution is concentrated between 30 and 150 nm, which is the typical size range for exosomes.

[0182] Figure 2 The results of differentially expressed genes (DEGs) analysis are shown. Left: Venn diagram of DEGs. A circle represents a gene set, and the area where different circles overlap represents the intersection of these gene sets. Unoverlapped areas are unique to this gene set, and the numbers on the diagram indicate the number of genes in the corresponding area. Right: Volcano plot of DEGs. Each dot represents a miRNA. The horizontal axis represents the logarithm of the fold difference in expression between the two samples for a particular miRNA; the vertical axis represents the negative logarithm of the false discovery rate (FDR). A larger absolute value on the horizontal axis indicates a greater fold difference in expression between the two samples; a larger value on the vertical axis indicates a more significant differential expression and a more reliable differentially expressed miRNA. Black dots represent miRNAs with no differential expression, red dots represent upregulation, and blue dots represent downregulation.

[0183] Figure 3 The cluster diagram of miRNA differential expression is shown. The columns represent different samples, and the rows represent different miRNAs. 10 The (TPM+1) values ​​were clustered, with red indicating highly expressed miRNAs and blue indicating low expression.

[0184] Figure 4 This figure shows the GO annotation classification statistics of differentially expressed miRNA target genes. The horizontal axis represents the GO classification, while the vertical axis represents the percentage of genes on the left and the number of genes on the right. This figure shows the gene enrichment of each GO secondary function in the context of differentially expressed miRNA target genes and the context of all genes.

[0185] Figure 5 The KEGG classification diagram of differentially expressed miRNA target genes is shown. The vertical axis represents the name of the KEGG metabolic pathway, the horizontal axis and the numbers on the bar graph represent the number of differentially expressed miRNA target genes annotated to that pathway, and the percentage on the bar graph represents the ratio of the number of differentially expressed miRNA target genes annotated to that pathway to the total number of differentially expressed miRNA target genes annotated to that pathway.

[0186] Figure 6This figure shows a GO enrichment plot of differentially expressed genes. The horizontal axis represents the GeneRatio (the ratio of genes of interest annotated in that entry to the total number of differentially expressed genes), and the vertical axis represents each BP entry. The size of the dot represents the number of differentially expressed miRNA target genes annotated in that pathway, and the color represents the enrichment significance value (Q value or P value, see the legend for details). Redder colors indicate lower significance values.

[0187] Figure 7 A bubble plot showing KEGG pathway enrichment for differentially expressed miRNA target genes is shown. The horizontal axis represents GeneRatio (the ratio of genes of interest annotated in that entry to all differentially expressed genes), while the vertical axis represents each KEGG pathway entry. The dot size represents the number of differentially expressed miRNA target genes annotated in that pathway, and the dot color represents the adjusted p-value of the hypergeometric test.

[0188] Figure 8 The evaluation results of SVM-RFE are shown. (A) SVM-RFE feature selection performance curve. The model performance changes when the SVM-RFE algorithm screens different numbers of miRNA features. The horizontal axis represents the number of retained feature genes, and the vertical axis represents the AUC value (Area Under Curve) based on ROC analysis. The larger the value, the better the model classification performance. The peak of the curve shows that the AUC is the highest (0.97) when the number of features is 5, indicating that the combination of hsa-miR-22-3p, hsa-miR-151a-3p, hsa-miR-148a-3p, hsa-miR-122-5p, and hsa-miR-223-3p is the optimal feature subset. (B) SVM model confusion matrix. Prediction results of sample grouping based on the SVM model of the five miRNA combinations. The matrix shows: In the tuberculosis group (TB, n = 7), 6 cases were correctly predicted (true negative, TN), and 1 case was misclassified as TB-DILI (false positive, FP); in the tuberculosis with liver injury group (TB-DILI, n = 5), all cases were correctly predicted (true positive, TP = 5), and no non-liver injury cases were misclassified (false negative, FN = 0). Based on this calculation, sensitivity (Sensitivity) = TP / (TP + FN) = 5 / 5 = 100%; specificity (Specificity) = TN / (TN + FP) = 5 / 5 = 85.7%. (C) ROC curve comparison between the SVM model and a single miRNA. ROC curve analysis of the SVM model for the five-miRNA combination and a single miRNA. The horizontal axis represents the false positive rate (1-specificity), and the vertical axis represents the true positive rate (sensitivity). DETAILED DESCRIPTION

[0189] Through extensive and in-depth experiments, the inventors identified 23 key miRNAs for the first time in patients with liver damage caused by taking anti-tuberculosis drugs. Compared with patients who do not suffer liver damage from taking anti-tuberculosis drugs, these miRNAs are significantly expressed in patients with liver damage caused by taking anti-tuberculosis drugs. Among them, hsa-miR-122-5p, hsa-miR-151a-3p, hsa-miR-192-5p, hsa-miR-148a-3p, hsa-miR-223-3p, and hsa-miR-22-3p are closely related to liver cell damage pathways and drug metabolism pathways, and can be used as markers for early diagnosis of liver damage caused by anti-tuberculosis drugs. In addition, based on the key differential miRNAs discovered, the present invention further constructed a diagnostic model for drug-induced liver damage caused by tuberculosis and anti-tuberculosis drugs. When the model includes five miRNAs—hsa-miR-22-3p, hsa-miR-151a-3p, hsa-miR-148a-3p, hsa-miR-122-5p, and hsa-miR-223-3p—it achieves optimal sample differentiation. This model can accurately diagnose both tuberculosis and drug-induced liver injury caused by anti-tuberculosis drugs. This is the basis for the present invention.

[0190] It should be understood that the specific methods and experimental conditions of the present invention are described below in various levels of detail to provide a more detailed understanding of the present invention. The following provides definitions of certain terms used in this specification. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present invention belongs.

[0191] the term

[0192] Where a numerical range is provided, it is understood that every intermediate integer of that value, every tenth of each intermediate integer of that value, between the upper and lower limits of that range, and any other intermediate values ​​in the specified range are encompassed within the present invention, unless the context clearly indicates otherwise. The upper and lower limits of these smaller ranges may independently be included in the smaller ranges and are also encompassed within the present invention, subject to any express exclusions in the specified range. For example, "1 to 50" includes "2 to 25," "5 to 20," "25 to 50," "1 to 10," etc.

[0193] As used herein, the terms "comprising" or "including" may be open, semi-closed, or closed. In other words, the terms also include "consisting essentially of" or "consisting of."

[0194] As used herein, the term "and / or" refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0195] As used herein, the term "significant" means that in a hypothesis test, the observed effect (e.g., the difference between the experimental and control groups) is unlikely to be caused by random error alone. A hypothesis test consists of: the null hypothesis (H0), which assumes that the observed effect does not exist (e.g., there is no difference between the experimental and control groups); the p-value, which is the probability of observing the current or more extreme effect if H0 is true; and the significance threshold (α). The significance threshold is usually used to determine whether the hypothesis test is significant. Generally, the significance threshold is 0.05. If the p-value ≤ α, H0 is rejected, meaning that the observed effect exists, and the result is called "significant."

[0196] As used herein, the terms "microRNA" and "miRNA" are used interchangeably. They are a class of evolutionarily conserved, small, non-coding RNA molecules, typically 21-23 nucleotides in length, that regulate gene expression at the translational level. MicroRNAs exist in various forms in bodily fluids, including serum and urine, with exosomes being the primary mode of expression.

[0197] As used herein, the terms "exosomal microRNA" and "exosome-derived microRNA" refer to microRNAs present in exosomes. Exosomes are 30-150 nm extracellular vesicles that protect the microRNAs within them from degradation by enzymes in the body, thereby enhancing their stability in body fluids. Exosome secretion increases significantly during oxidative stress, hypoxia, and cellular damage, and the amount of microRNA contained in exosomes also increases accordingly.

[0198] As used herein, the term "SVM-RFE" is support vector machine-recursive feature elimination, which is a feature selection method that combines support vector machine and recursive feature elimination. First, the SVM is trained and the feature weights are obtained. Among them, the SVM calculates the weight of each feature, and the larger the absolute value of the weight, the more important the feature; then, the features are sorted from large to small by the absolute value of the weight, and the features with the smallest weight are removed, or the features with the lowest weight ranking are removed. Repeat training and elimination until a specified number of features remain. Training and elimination are terminated until the preset number of features is reached or the model performance does not improve. In general, cross-validation is used to evaluate the performance of the model.

[0199] miRNA of the present invention and its application

[0200] The inventors have identified 23 key miRNAs in patients with liver damage caused by anti-tuberculosis drugs. These miRNAs can serve as markers for the early diagnosis of drug-induced liver damage, and can therefore be used as markers for detecting drug-induced liver damage caused by anti-tuberculosis drugs, as well as for the preparation of diagnostic reagents or kits for detecting drug-induced liver damage caused by anti-tuberculosis drugs.

[0201] Based on the miRNA of the present invention, the present invention also provides a method for determining whether a patient taking anti-tuberculosis drugs has drug-induced liver injury or is at risk of drug-induced liver injury, or whether the patient will have drug-induced liver injury or will be at risk of drug-induced liver injury after taking anti-tuberculosis drugs, including the steps of detecting the patient's miRNA, comparing the expression level of the detected miRNA with a reference value, and, based on the change in the expression level of the detected miRNA relative to the reference value, making a conclusion that the patient taking anti-tuberculosis drugs has drug-induced liver injury or is at risk of drug-induced liver injury, or the patient will have drug-induced liver injury or will be at risk of drug-induced liver injury after taking anti-tuberculosis drugs.

[0202] Corresponding to the above-mentioned method for determining whether a patient taking anti-tuberculosis drugs has drug-induced liver injury or is at risk of drug-induced liver injury, or whether a patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking anti-tuberculosis drugs, the present invention also provides a device for implementing the method, the device comprising a detection module, a comparison module, and a judgment module. In a preferred embodiment, the device may further include an output module for outputting the conclusion of the judgment module to a clinician or patient.

[0203] Method for constructing the diagnostic model of the present invention

[0204] Methods for isolating exosomal miRNA are well known to those skilled in the art, and any method capable of isolating exosomal miRNA from a sample (such as a serum sample) falls within the scope of protection of the present invention. In the present invention, size exclusion chromatography is used for separation.

[0205] Methods for quantifying miRNA are well known to those skilled in the art. Generally speaking, the method for quantifying RNA includes the steps of: performing RNA sequencing on the obtained sample; aligning the sequencing data (reads) to the reference genome; and counting the number of successfully aligned reads using a suitable method. Before performing the alignment, the obtained sequencing data can be quality controlled to remove data that may affect subsequent analysis. In some cases, it may be necessary to identify and annotate the reads to confirm whether they are new transcription products.

[0206] The analysis method of differential expression is also well known to those skilled in the art. Generally, the method of differential expression analysis includes the steps of providing expression data of a gene / RNA / protein under two treatment backgrounds, standardizing the expression data, comparing the two groups of expression data using a suitable method, and performing a statistical test to determine whether the two groups of expression data are significantly different. The genes / RNA / proteins with significant differential expression are used as data for subsequent analysis. Preferably, the genes / RNA / proteins with significant differential expression obtained are screened for fold change (FC). Preferably, the miRNAs with significant differential expression of FC≥1.5 are used as data for subsequent analysis. Before performing differential analysis, the data to be processed can be corrected.

[0207] In some embodiments, the differential expression analysis further comprises performing functional analysis on the significantly differentially expressed miRNAs.

[0208] The present invention also predicts and functionally analyzes target genes of significantly differentially expressed miRNAs.

[0209] The main advantages of the present invention include:

[0210] (1) The present invention differentially analyzed the miRNA expression profiles of patients who did not develop liver damage from taking anti-tuberculosis drugs and those who developed liver damage from taking anti-tuberculosis drugs. A total of 128 significantly differentially expressed miRNAs were identified. In patients with liver damage from anti-tuberculosis drugs, 83 miRNAs were upregulated and 45 were downregulated.

[0211] (2) Based on the screening of expression levels, the present invention further identified 23 key differential miRNAs, among which hsa-miR-122-5p, hsa-miR-151a-3p, hsa-miR-192-5p, hsa-miR-148a-3p, hsa-miR-223-3p, and hsa-miR-22-3p are closely related to the hepatocellular damage pathway and drug metabolism pathway, and can be used as markers for the early diagnosis of liver damage caused by anti-tuberculosis drugs.

[0212] (3) Based on the key differential miRNAs discovered, the present invention further constructed a diagnostic model for tuberculosis and drug-induced liver injury caused by anti-tuberculosis drugs. When the model includes five miRNAs: hsa-miR-22-3p, hsa-miR-151a-3p, hsa-miR-148a-3p, hsa-miR-122-5p, and hsa-miR-223-3p, the ability to distinguish samples is optimal.

[0213] (4) The model can accurately diagnose tuberculosis with a specificity of ≥85.7% and can also accurately diagnose drug-induced liver injury caused by anti-tuberculosis drugs with a specificity of 100%.

[0214] The present invention will be further described below in conjunction with specific examples. It should be understood that these examples are intended to illustrate the present invention only and are not intended to limit the scope of the invention. The experimental methods in the following examples, for which specific conditions are not specified, are generally based on conventional conditions, such as those described in Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989), or according to the conditions recommended by the manufacturer. Unless otherwise stated, percentages and parts are weight percentages and parts by weight.

[0215] Materials and methods

[0216] 1. Collection of patient serum:

[0217] This study aimed to analyze the differences in serum exosomal miRNAs in tuberculosis-infected patients receiving anti-tuberculosis drug therapy (an isoniazid-rifampicin-based regimen) with and without drug-induced liver injury (DILI). Twelve tuberculosis-infected patients receiving anti-tuberculosis treatment at the Shanghai Public Health Clinical Center between December 2024 were included. Seven patients had normal liver function, five had DILI (diagnosed as confirmed by the RUCAM score), and six healthy controls served as a control group. No statistically significant differences in gender or age were observed between the groups (P>0.05).

[0218] 2. Serum exosome isolation:

[0219] Serum exosomes were separated by size exclusion chromatography: 1 mL of thawed serum was pre-filtered through a 0.8 μm filter membrane, rinsed with PBS, and loaded onto an Enze Kangtai purification column (kit ES911). After equilibration with 10 mL of 0.1 M PBS, 2 mL of the eluate was collected and transferred to a 100 kDa ultrafiltration tube (Millipore UFC810096). The column was concentrated to 200 μL by centrifugation at 4000 × g, and the volume was made up with PBS before aliquoting for storage.

[0220] 3. Serum exosome identification:

[0221] Morphological identification by transmission electron microscopy: 10 μL of exosome sample was dripped onto a plasma-cleaned copper grid, incubated at room temperature for 10 min, washed with sterile water, and negatively stained with 2% uranyl acetate for 1 min. The grid was dried under an incandescent lamp for 2 min and imaged using a Hitachi H-7650 transmission electron microscope at an accelerating voltage of 80 kV.

[0222] BCA protein quantification: The total protein concentration was determined using the BCA method: bovine serum albumin standard was serially diluted (0–2000 ng / μL), and the sample was mixed with BCA working solution (solution A:solution B = 50:1) at a volume ratio of 1:8. The samples were incubated at 37°C for 30 min. The absorbance at 562 nm was measured using a microplate reader (TECAN infiniteF50), and the concentration was calculated based on the standard curve.

[0223] Western Blot Protein Marker Detection: Based on BCA quantification results, 10–30 μg of protein was denatured at 95°C for 5 min, separated by 10% SDS-PAGE electrophoresis, and transferred to a PVDF membrane at 200 mA for 1 h. After blocking with 5% skim milk powder for 2 h, the membrane was incubated with primary antibodies (Tsg101 ABclonal A5789; Alix abcam ab186429; CD9 abcam ab263019; HSP70 abcam ab181606; Calnexin Proteintech 10427-2-AP) overnight at 4°C and HRP secondary antibodies for 1 h at room temperature. The membrane was then developed using ECL chemiluminescence reagent (Solarbio PE0010).

[0224] NanoFCM nanoflow cytometric particle size analysis: Detection was performed using a NanoFCM N30E nanoflow cytometer: the sample was diluted with PBS to an appropriate concentration, the flow rate was calibrated with 250 nm silica fluorescent microspheres (NanoFCM QS2502), and the particle size distribution and particle concentration were analyzed based on a scattered light-particle size standard curve established using 68 / 91 / 113 / 155 nm silica standard spheres (NanoFCM S16M-Exo).

[0225] 4. Data Analysis:

[0226] Sequencing data quality control: Raw sequencing data (Raw Reads) were obtained using the Illumina NovaSeq platform. A multi-stage filtering process was used to generate high-quality sequences (Clean Reads): (1) sequences with a Q30 (base misidentification rate ≤ 0.1%) ratio of less than 97.7% were removed; (2) reads with an unknown base N content ≥ 10% were removed; and (3) sequences without 3' adapters and lengths < 15 nt or > 35 nt were removed.

[0227] sRNA annotation and genome alignment: Clean reads were aligned to noncoding RNA databases (Silva / rRNA, GtRNAdb / tRNA, Rfam / snRNA / snoRNA, Repbase repeat sequences) using Bowtie v1.0.0 (parameter: -v0). Unannotated reads were further aligned to the GRCh38 human reference genome (Ensembl release-110) to obtain mapped reads.

[0228] miRNA identification and novel miRNA prediction: miRNA identification was performed using miRDeep2 v2.0.5 (animal parameters: -g -1 -b0). Mapped reads were aligned with the miRBase v22 database to identify known miRNAs, and novel miRNAs were screened by precursor hairpin structure prediction (RNAfold v2.1.7) and random folding energy calculation (randfold v2.0, -s 99).

[0229] miRNA quantification and differential analysis: Based on the UMI (Unique Molecular Identifier) ​​of the QsRNA-seq library, PCR amplification deviation was corrected and the expression level was normalized using the TPM algorithm. Count data were processed using limma v3.40.2 for differential analysis. The screening criteria were: |log2FC| ≥ 0.585 (FC ≥ 1.5) and the corrected p value ≤ 0.05, and miRNAs expressed in at least 2 / 3 of the samples in the group were retained. Four groups of comparisons were set: Control_vs_TB, Control_vs_TB-DILI, Control_vs_TB_TB-DILI, and TB-DILI_vs_TB. The difference results were analyzed using volcano plots and hierarchical clustering (log 10 (TPM+1)) visualization.

[0230] Target gene prediction and functional annotation: Animal target gene prediction was performed using a combination of miRanda v3.3a (parameters: -sc 150.0 -en-30 -scale 4.0) and RNAhybrid v2.1.1 (parameters: -d 1.9, 0.28 -b 1-e-30). BLAST v2.2.26 (parameter: -e 1e-5) was used to align the target genes against nine major databases, including NR, Swiss-Prot, GO, and KEGG. Functional annotation was obtained for 19,326 genes (99.66%). Differential miRNA target genes were enriched for GO functions (topGO v2.18.0, node size = 6) and KEGG pathways (clusterProfiler). Hypergeometric-corrected p-values ​​< 0.05 were considered significant.

[0231] Disease relevance screening: Disease associations of differentially annotated miRNAs were analyzed using the Human Disease miRNA Database (HMDD v3.0). Focus was placed on screening miRNAs associated with liver injury and tuberculosis (e.g., hsa-miR-122-5p), and their potential as biomarkers was assessed based on literature support.

[0232] 5. Feature selection and diagnostic model construction based on SVM-RFE:

[0233] Key miRNAs were screened using the support vector machine-recursive feature elimination (SVM-RFE) algorithm. Based on the maximum margin principle of the SVM, this algorithm iteratively trains the model, calculates feature importance scores, removes the lowest-scoring features, and repeats this process to ultimately determine the optimal feature combination. Model performance was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), with higher AUCs indicating better diagnostic performance. Concurrently, sensitivity and specificity were calculated using a confusion matrix. Sensitivity reflects the model's ability to identify positive samples (TP / (TP + FN)), while specificity reflects its ability to identify negative samples (TN / (TN + FP)).

[0234] Example 1: Identification of serum exosomes

[0235] Under an electron microscope, serum exosomes appear as cup-shaped or saucer-shaped membrane-enclosed microstructures with clear edges. The structures are round or nearly round. The diameter ranges from 30 to 150 nanometers (nm). The membrane structure usually has a smooth or slightly textured appearance, with small protrusions or depressions visible on the membrane ( Figure 1 A).

[0236] Western Blot protein marker detection showed that serum exosome samples expressed positive protein markers Tsg101 / Alix / CD9 / HSP70, but no Calnexin expression ( Figure 1 B). NanoFCM nano-flow cytometry analysis shows the particle size distribution of serum exosomes, including the average particle size, main peak particle size, and particle size distribution width parameters. It can also obtain particle concentration information and particle size-number distribution. The figure shows that the particle size distribution of exosomes is concentrated in the range of 30-150nm ( Figure 1 C).

[0237] Example 2: Identification and Characterization of miRNA

[0238] This example involves identifying miRNAs and analyzing their characteristics through quality control and comparison.

[0239] First, sequencing data quality control and sRNA annotation were performed. Small RNA sequencing was performed on 18 plasma exosome samples using the Illumina NovaSeq platform, yielding a total of 403.06M raw reads (mean, 22.39M / sample). After rigorous quality control (excluding low-quality sequences, sequences with N > 10%, and sequences <15nt or > 35nt in length), the total number of clean reads reached 249.02M (mean, 13.84M), with an average Q30 base quality of 97.83%. sRNA classification and annotation revealed that unannotated reads accounted for 70.82%-88.78%, of which 41.19%-74.08% were mapped to the GRCh38 genome, with a positive-strand localization rate of 87.09%-90.79%, consistent with the characteristics of exosomal miRNAs.

[0240] Next, miRNAs were identified and their signatures analyzed. A total of 701 miRNAs, including 682 known miRNAs and 19 novel miRNAs, were identified using miRDeep2 v2.0.5 combined with hairpin structure prediction (RNAfold). The dominant length peak for known miRNAs was 22 nt (accounting for 68.3%), while the length distribution of novel miRNAs showed a bimodal distribution (20 nt and 24 nt). The secondary structure free energy (MFE ≤ -20 kcal / mol) of the novel miRNA precursors was analyzed. Among them, novel_miR_12 was stably expressed in 15 samples (TPM ≥ 5), and its hairpin structure was verified by randfold (p < 0.05).

[0241] Example 3: Differential miRNA analysis

[0242] This embodiment involves differential expression analysis of miRNAs.

[0243] Based on UMI-corrected TPM quantitative analysis (limma algorithm), a total of 128 differential miRNAs (FC ≥ 1.5, p ≤ 0.05) were identified between the TB-DILI group and the TB group, of which 83 were upregulated and 45 were downregulated in the TB-DILI group.

[0244] Through gene expression level analysis, the volcano plot of significantly differentially expressed genes was statistically analyzed and drawn. The volcano plot shows two key indicators: fold change and corrected p-value. The T-test analysis showed that the significantly differentially expressed genes between the two samples were obtained, with log2 (fold change) as the horizontal axis and -log 10 (padj) is the vertical axis. It can be seen that the expression levels of miRNAs in the two (groups) of samples are different, TB-DILI_vs_TB: 83 genes are upregulated and 45 genes are downregulated ( Figure 2 right).

[0245] According to cluster analysis, miRNAs with the same or similar expression patterns can be grouped into one category. MiRNAs with similar expression patterns may have similar functions, participate in the same metabolic process or exist in the same cellular pathway. Cluster analysis can be used to infer the functions of unknown genes or new functions of known genes ( Figure 3 ).

[0246] Then, differential miRNAs were filtered by expression. Differential genes were selected using FC>=1.5 and PValue<=0.05, and the average UMI expression of all samples was ≥50. The VENN diagram shows the differential expression of miRNAs between the groups. It can be seen that there are 38 differentially expressed miRNAs in the serum exosomes of patients with TB-DILI (anti-tuberculosis drug-induced liver injury) and TB (no drug injury) groups ( Figure 2 left).

[0247] Example 3: Prediction and functional analysis of miRNA target genes

[0248] This embodiment involves predicting and analyzing the function of miRNA target genes, wherein the functional analysis includes biological process, molecular function, cellular component, and KEGG pathway analysis.

[0249] Based on the gene sequence information of known miRNAs and newly predicted miRNAs and the corresponding species, miRanda and RNAhybrid were used to predict target genes, and a total of 19,392 target genes were predicted.

[0250] GO analysis was performed on the differentially expressed miRNA target genes. The R package clusterProfiler was used to perform enrichment analysis of the differentially expressed miRNA target genes in terms of biological process, molecular function, and cellular component. The enrichment analysis used the hypergeometric test method to find GO terms that were significantly enriched compared with the whole genome background ( Figure 4 ).

[0251] The significant enrichment of these functions indicates that the differentially expressed miRNA target genes have specific regulatory effects in biological processes, cellular components and molecular functions. GTPase activity regulation may be involved in a variety of intracellular signal transduction processes, while protein autophosphorylation is closely related to the regulation of key enzyme activities. Cell migration regulation and regulation of BMP signaling pathways may play an important role in tissue repair or pathological changes. In terms of cellular components, the enrichment of cytoplasm and cell junctions suggests that these genes may be related to intercellular communication and structural maintenance. The significance of nuclear chromatin and synaptic membranes further indicates that epigenetic regulation and neural signaling may be affected. In addition, the prominent performance of functions such as calcium ion binding and chromatin binding suggests the potential role of these genes in the stability of the intracellular environment and the regulation of gene expression. In addition, functions such as GTPase, plasma membrane, and synapse show consistency in different branches, which may suggest their synergistic effects in related biological processes. These functions can be further explored in the future to reveal their potential mechanisms in the process of liver damage caused by anti-tuberculosis drugs ( Figure 6 ).

[0252] KEGG analysis was performed on differentially expressed miRNA target genes. KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analysis was performed on each gene using the R package clusterProfiler. The hypergeometric test method was used to identify KEGG pathways that were significantly enriched compared to the entire genome background. Figure 5 ).

[0253] This study found that the PI3K-Akt pathway was abnormally activated, which may be involved in regulating cell survival and apoptosis, thereby affecting the functional state of liver cells; changes in the endocytic pathway may be related to exosome-mediated substance transport and signal transduction; and disturbances in the calcium signaling pathway may further trigger an imbalance in intracellular calcium homeostasis, leading to cell dysfunction. In addition, abnormalities in the Wnt signaling pathway may involve disorders in cell proliferation and differentiation processes, while changes in the cholinergic synaptic pathway may have an important impact on neurotransmitter-related signal transduction. These results suggest that differentially expressed miRNA target genes play an important role in liver damage caused by anti-tuberculosis drugs through the combined action of multiple key pathways ( Figure 7 ).

[0254] Example 4: Differential miRNA functional annotation and pathway analysis

[0255] This embodiment involves further screening candidate miRNA markers based on existing information and exploring the biological pathways in which they participate or mediate.

[0256] Through literature mining and the HMDD database, a total of 23 key differentially expressed miRNAs were identified in the comparison between the TB-DILI (anti-tuberculosis drug-induced liver injury) group and the TB (drug-uninjured) group. Among them, 10 miRNAs are known to be clearly associated with drug metabolism, hepatocellular injury, and inflammatory immune pathways: hsa-miR-122-5p, hsa-miR-151a-3p, hsa-miR-192-5p, hsa-miR-148a-3p, hsa-miR-223-3p, hsa-miR-22-3p, hsa-miR-146a-5p, hsa-miR-370-3p, hsa-miR-744-5p, and hsa-miR-382-5p. Among them, miR-146a / 370 / 382 were also highly expressed in the control group and were therefore excluded; miR-744 was reported in previous studies to be negatively correlated with the degree of liver damage, which was inconsistent with this study; therefore, this study finally screened out 6 miRNAs that were significantly upregulated in the serum of TB-DILI patients as subsequent verification targets: hsa-miR-122-5p, hsa-miR-151a-3p, hsa-miR-192-5p, hsa-miR-148a-3p, hsa-miR-223-3p, and hsa-miR-22-3p. Among them, hsa-miR-122-5p, a sensitive biomarker of liver injury, regulates the hepatocellular injury pathway and the PI3K / MAPK pathway; hsa-miR-151a-3p, a key marker of acetaminophen hepatotoxicity, mediates the drug metabolism pathway; hsa-miR-192-5p, involved in bile secretion regulation, affects the hepatocellular apoptosis pathway; hsa-miR-148a-3p, which promotes hepatocellular CYP protein expression, regulates the drug metabolism pathway; hsa-miR-223-3p, which alleviates liver injury by inhibiting the NLRP3 inflammasome, participates in the hepatocellular apoptosis pathway; and hsa-miR-22-3p, an early liver injury marker that inhibits liver cancer cell proliferation, targets the MAPK / ERK pathway. This biomarker combination system covers the three major pathological links of drug metabolism imbalance, hepatocellular injury, and inflammatory immune imbalance, providing a multi-target validation basis for the TB-DILI diagnostic system (Table 1).

[0257] Table 1. Functional annotation and pathway analysis of differentially expressed miRNAs

[0258]

[0259]

[0260] Example 5: Optimization and screening of miRNA combinations and validation of model diagnostic efficacy

[0261] This example involves screening the miRNAs obtained above and constructing a model to evaluate the efficacy of the model in diagnosing tuberculosis and liver damage induced by anti-tuberculosis drugs.

[0262] SVM-RFE feature selection showed that when the following five miRNAs (hsa-miR-22-3p, hsa-miR-151a-3p, hsa-miR-148a-3p, hsa-miR-122-5p, and hsa-miR-223-3p) were retained, the model AUC reached the highest value (0.97), indicating that this combination had the best ability to distinguish the groups of samples ( Figure 8 A).

[0263] The confusion matrix validation results showed that for 7 tuberculosis (TB) samples, the model correctly predicted 6 cases (specificity 85.7%); for 5 tuberculosis combined with drug-induced liver injury (TB-DILI) samples, the model correctly predicted all of them (sensitivity 100%) ( Figure 8 B).

[0264] ROC curve analysis further confirmed that the area under the curve of the SVM model (AUC = 0.97) was significantly higher than that of a single miRNA (such as hsa-miR-122-5p), indicating that the combination strategy can improve the ability to distinguish between TB and TB-DILI, and has better diagnostic performance ( Figure 8 C).

[0265] All documents mentioned in this application are incorporated herein by reference, just as if each document were incorporated herein by reference individually. It should also be understood that after reading the above teachings of the present invention, those skilled in the art may make various changes or modifications to the present invention, and that such equivalents also fall within the scope of the claims appended hereto.

Claims

1. A use of miRNA or a detection reagent thereof, characterized in that: (i) for use as a marker for detecting drug-induced liver injury caused by taking anti-tuberculosis drugs; and / or (ii) for use in preparing a diagnostic reagent or kit for detecting drug-induced liver injury caused by taking anti-tuberculosis drugs; Wherein, the miRNA is selected from the following group: (A) any one marker selected from A1 to A5, or a combination thereof: (A1) hsa-miR-151a-3p; (A2) hsa-miR-22-3p; (A3) hsa-miR-148a-3p; (A4) hsa-miR-122-5p; (A5) hsa-miR-223-3p; (B) any one marker selected from B1 to B18, or a combination thereof: (B1) hsa-miR-4433b-5p; (B2) hsa-miR-151a-5p; (B3) hsa-miR-151b; (B4) hsa-miR-548j-5p; (B5) hsa-miR-192-5p; (B6) hsa-miR-146a-5p; (B7) hsa-miR-370-3p; (B8) hsa-miR-744-5p; (B9) hsa-miR -382-5p; (B10)hsa-miR-335-5p; (B11)hsa-miR-1307-3p; (B12)hsa-miR-409-3p; (B13)hsa-let-7d-3p; (B14)h sa-miR-629-5p; (B15) hsa-miR-379-5p; (B16) hsa-miR-432-5p; (B17) hsa-miR-134-5p; (B18) hsa-miR-199b-3p; (C) A combination of one or more markers from A1 to A5 and one or more markers from B1 to B18.

2. A kit for detecting drug-induced liver injury caused by taking anti-tuberculosis drugs, characterized in that: The kit contains: a first container, and a detection reagent for detecting miRNA located in the first container; Wherein, the miRNA is selected from the following group: (A) any one marker selected from A1 to A5, or a combination thereof: (A1) hsa-miR-151a-3p; (A2) hsa-miR-22-3p; (A3) hsa-miR-148a-3p; (A4) hsa-miR-122-5p; (A5) hsa-miR-223-3p; (B) any one marker selected from B1 to B18, or a combination thereof: (B1) hsa-miR-4433b-5p; (B2) hsa-miR-151a-5p; (B3) hsa-miR-151b; (B4) hsa-miR-548j-5p; (B5) hsa-miR-192-5p; (B6) hsa-miR-146a-5p; (B7) hsa-miR-370-3p; (B8) hsa-miR-744-5p; (B9) hsa-miR -382-5p; (B10)hsa-miR-335-5p; (B11)hsa-miR-1307-3p; (B12)hsa-miR-409-3p; (B13)hsa-let-7d-3p; (B14)h sa-miR-629-5p; (B15) hsa-miR-379-5p; (B16) hsa-miR-432-5p; (B17) hsa-miR-134-5p; (B18) hsa-miR-199b-3p; (C) a combination of one or more markers from A1 to A5 and one or more markers from B1 to B18; A second container, and a detection reagent for detecting other miRNA located in the second container, wherein the miRNA includes microRNA-122.

3. A use of a miRNA inhibitor, characterized in that: For preparing a drug or pharmaceutical composition for preventing / treating drug-induced liver injury caused by taking anti-tuberculosis drugs; wherein the miRNA is selected from the group consisting of: hsa-miR-122-5p; hsa-miR-151a-3p; hsa-miR-192-5p; hsa-miR-148a-3p; hsa-miR-223-3p; hsa-miR-22-3p; hsa-miR-146a-5p; hsa-miR-370-3p; hsa-miR-744-5p; hsa-miR-382-5p; hsa-miR-335-5p; hsa-miR-1307-3p; hsa-miR-409-3p; hsa-let-7d-3p; hsa-miR-629-5p; hsa-miR-379-5p; hsa-miR-432-5p; h sa-miR-134-5p; hsa-miR-199b-3p; hsa-miR-4433b-5p; hsa-miR-151a-5p; hsa-miR-151b; hsa-miR-548j-5p.

4. A pharmaceutical composition, characterized in that The pharmaceutical composition comprises: (1) miRNA inhibitors; wherein the miRNA is selected from the group consisting of: hsa-miR-122-5p; hsa-miR-151a-3p; hsa-miR-192-5p; hsa-miR-148a-3p; hsa-miR-223-3p; hsa-miR-22-3p; hsa-miR-146a-5p; hsa-miR-370-3p; hsa-miR-744-5p; hsa-miR-382-5p; hsa-miR-3 35-5p; hsa-miR-1307-3p; hsa-miR-409-3p; hsa-let-7d-3p; hsa-miR-629-5p; hsa-miR-379-5p; hsa-miR-432-5 p; hsa-miR-134-5p; hsa-miR-199b-3p; hsa-miR-4433b-5p; hsa-miR-151a-5p; hsa-miR-151b; hsa-miR-548j-5p; (2) anti-tuberculosis drugs; and (3) Pharmaceutically acceptable carriers, diluents or excipients.

5. A medicine box, characterized in that: The medicine box comprises: A first container, and a miRNA inhibitor located in the first container; wherein the miRNA is selected from the group consisting of: hsa-miR-122-5p; hsa-miR-151a-3p; hsa-miR-192-5p; hsa-miR-148a-3p; hsa-miR-223-3p; hsa-miR-22-3p; hsa-miR-146a-5p; hsa-miR-370-3p; hsa-miR-744-5p; hsa-miR-382-5p; hsa -miR-335-5p; hsa-miR-1307-3p; hsa-miR-409-3p; hsa-let-7d-3p; hsa-miR-629-5p; hsa-miR-379-5p; hsa-miR-43 2-5p; hsa-miR-134-5p; hsa-miR-199b-3p; hsa-miR-4433b-5p; hsa-miR-151a-5p; hsa-miR-151b; hsa-miR-548j-5p; A second container, and an anti-tuberculosis drug located in the second container.

6. A method for determining whether a patient taking an anti-tuberculosis drug has drug-induced liver injury or is at risk of drug-induced liver injury, or whether a patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking an anti-tuberculosis drug, characterized in that: The method comprises the following steps: 1) Detection step: Detect the patient's miRNA; 2) Comparison step: comparing the miRNA expression level detected in step 1) with the reference value; 3) Judgment step: If the expression level of the miRNA detected in step 1) is significantly changed relative to the reference value, a conclusion is drawn that the patient taking the anti-tuberculosis drug has drug-induced liver injury or is at risk of drug-induced liver injury, or that the patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking the anti-tuberculosis drug; Wherein, the miRNA is selected from the following group: (A) any one marker selected from A1 to A5, or a combination thereof: (A1) hsa-miR-151a-3p; (A2) hsa-miR-22-3p; (A3) hsa-miR-148a-3p; (A4) hsa-miR-122-5p; (A5) hsa-miR-223-3p; (B) any one marker selected from B1 to B18, or a combination thereof: (B1) hsa-miR-4433b-5p; (B2) hsa-miR-151a-5p; (B3) hsa-miR-151b; (B4) hsa-miR-548j-5p; (B5) hsa-miR-192-5p; (B6) hsa-miR-146a-5p; (B7) hsa-miR-370-3p; (B8) hsa-miR-744-5p; (B9) hsa-miR -382-5p; (B10)hsa-miR-335-5p; (B11)hsa-miR-1307-3p; (B12)hsa-miR-409-3p; (B13)hsa-let-7d-3p; (B14)h sa-miR-629-5p; (B15) hsa-miR-379-5p; (B16) hsa-miR-432-5p; (B17) hsa-miR-134-5p; (B18) hsa-miR-199b-3p; (C) a combination of one or more markers from A1 to A5 and one or more markers from B1 to B18; The reference value is the expression level of the corresponding serum exosomal miRNA in normal people, or the expression level of the corresponding serum exosomal miRNA in patients taking anti-tuberculosis drugs but without drug-induced liver injury.

7. A device for determining whether a patient taking anti-tuberculosis drugs has drug-induced liver injury or is at risk of drug-induced liver injury, or whether a patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking anti-tuberculosis drugs, characterized in that: The device includes the following modules: 1) Detection module: a module for detecting the patient's miRNA; 2) Comparison module: a module for comparing the expression level of the miRNA detected in step 1) with the reference value; 3) Judgment module: If the expression level of the miRNA detected in step 1) is significantly changed relative to the reference value, a module for concluding that the patient taking the anti-tuberculosis drug has drug-induced liver injury or is at risk of drug-induced liver injury, or that the patient will have drug-induced liver injury or be at risk of drug-induced liver injury after taking the anti-tuberculosis drug; Wherein, the miRNA is selected from the following group: (A) any one marker selected from A1 to A5, or a combination thereof: (A1) hsa-miR-151a-3p; (A2) hsa-miR-22-3p; (A3) hsa-miR-148a-3p; (A4) hsa-miR-122-5p; (A5) hsa-miR-223-3p; (B) any one marker selected from B1 to B18, or a combination thereof: (B1) hsa-miR-4433b-5p; (B2) hsa-miR-151a-5p; (B3) hsa-miR-151b; (B4) hsa-miR-548j-5p; (B5) hsa-miR-192-5p; (B6) hsa-miR-146a-5p; (B7) hsa-miR-370-3p; (B8) hsa-miR-744-5p; (B9) hsa-miR -382-5p; (B10)hsa-miR-335-5p; (B11)hsa-miR-1307-3p; (B12)hsa-miR-409-3p; (B13)hsa-let-7d-3p; (B14)h sa-miR-629-5p; (B15) hsa-miR-379-5p; (B16) hsa-miR-432-5p; (B17) hsa-miR-134-5p; (B18) hsa-miR-199b-3p; (C) a combination of one or more markers from A1 to A5 and one or more markers from B1 to B18; The reference value is the expression level of the corresponding serum exosomal miRNA in normal people, or the expression level of the corresponding serum exosomal miRNA in patients taking anti-tuberculosis drugs but without drug-induced liver injury.

8. A method for constructing a diagnostic model for drug-induced liver injury caused by tuberculosis and / or anti-tuberculosis drugs, characterized in that: The method comprises the steps of: (1) Providing samples containing miRNA from tuberculosis patients with normal liver function and tuberculosis patients with drug-induced liver injury caused by anti-tuberculosis drugs; (2) quantifying miRNAs in the two samples and performing differential expression analysis to obtain miRNA expression data and differentially expressed miRNAs; (3) screening the differentially expressed miRNAs to obtain candidate miRNAs; (4) The miRNA expression data and candidate miRNAs are used to train the support vector machine (SVM) model to obtain a diagnostic model.

9. The method according to claim 8, wherein In step (4), the method further includes the step of using a recursive feature elimination (RFE) algorithm to obtain the preferred miRNA, thereby obtaining a diagnostic model.

10. A system for diagnosing drug-induced liver injury caused by tuberculosis and / or anti-tuberculosis drugs, characterized in that: include: An input module, wherein the input module is configured to input data, wherein the input data includes: miRNA expression data and miRNA type of the object to be tested; a diagnostic module configured to execute a diagnostic model for tuberculosis and / or drug-induced liver injury caused by anti-tuberculosis drugs, thereby obtaining a diagnostic result for the subject to be tested; the diagnostic model is constructed using the method of claim 8; An output module is configured to output a diagnosis result of the diagnosis module.

Images