HIV (Human Immunodeficiency Virus) combined non-alcoholic fatty liver diagnosis and prediction marker and application thereof

Abstract

The invention belongs to the technical field of medical diagnosis, and provides an HIV (human immunodeficiency virus) combined non-alcoholic fatty liver diagnosis and prediction marker and application. The marker comprises 39 biomarkers in total, wherein the 39 biomarkers comprise two intestinal microorganisms of enterobacter and klebsiella and 37 plasma metabolites of diacylglycerol, lysophosphatidic acid and the like; by integrating microbiome and metabonomics, the multi-dimensional marker is screened and verified, so that the problems of lack of high specificity and sensitivity, difficulty in early disease screening and incapability of effective risk assessment of an HIV (Human Immunodeficiency Virus) combined NAFLD (Non-Amplified Fragment Leukemia) early diagnosis method caused by unknown pathogenesis in the prior art are solved; the marker combination is applicable to preparation of diagnostic kits, construction of disease prediction models and development of targeted therapy drugs, and a theoretical basis is provided for early discovery, risk early warning and advanced intervention of the HIV combined with the non-alcoholic fatty liver disease.

Description

Technical Field

[0001] This invention belongs to the field of medical diagnostics and relates to biomarkers, specifically to a diagnostic predictive biomarker for HIV-associated non-alcoholic fatty liver disease and its application. Background Technology

[0002] Human immunodeficiency virus (HIV) infection remains a major global public health challenge. With the widespread use of highly effective antiretroviral therapy, the life expectancy of HIV-infected individuals has significantly increased, and their disease spectrum has also shifted from traditional opportunistic infections and tumors to chronic metabolic complications, represented by non-alcoholic fatty liver disease (NAFLD).

[0003] Non-alcoholic fatty liver disease (NAFLD) refers to the presence of steatosis in more than 5% of hepatocytes after excluding excessive alcohol consumption and other known liver damage factors. Notably, the prevalence of NAFLD in HIV-infected individuals is generally higher than in the general population, and they have a higher risk of progression to steatohepatitis, liver fibrosis, and even cirrhosis. It is also frequently accompanied by extrahepatic complications such as cardiovascular disease and type 2 diabetes, becoming one of the leading causes of liver-related morbidity and mortality in this population. However, the pathogenesis of HIV-associated NAFLD is currently poorly understood, and effective interventions are lacking. Therefore, in-depth research into its pathophysiological mechanisms is crucial for early diagnosis and prevention.

[0004] In recent years, the role of the gut microbiome as a "second genome" has become increasingly prominent, and its dysregulation is closely related to various metabolic diseases. Studies have shown that the gut microbiota plays a key role in the development and progression of NAFLD by influencing host metabolic, immune, and inflammatory pathways. For example, meta-analyses have revealed specific changes at the genus level in the gut microbiota of NAFLD patients, such as a decrease in the abundance of beneficial bacteria like Coprococcus and Faecalibacterium, and an increase in the abundance of genus bacteria like Escherichia. Furthermore, research indicates that the gut microbiota structure of HIV-infected patients with NAFLD exhibits unique dysregulation characteristics compared to HIV-infected individuals or HIV-negative fatty liver patients, suggesting that the gut microbiota may be deeply involved in the pathological process of this comorbidity.

[0005] The functional output of the gut microbiota is largely achieved through its metabolites, which are the core mediators of microbiota-host interactions. However, the role of the gut microbiota and its derived circulating metabolites in HIV-associated NAFLD has not yet been systematically elucidated. Therefore, this study integrates gut microbiome and metabolomics to delve into the microbiota-metabolic network behind this comorbidity, aiming to provide new scientific evidence for discovering early diagnostic biomarkers and developing precise intervention strategies. Summary of the Invention

[0006] The purpose of this invention is to provide a diagnostic predictive biomarker for HIV-associated non-alcoholic fatty liver disease and its application. By screening and validating specific combinations of 39 biomarkers containing specific gut microbiota (Enterobacterium and Klebsiella) and 37 plasma metabolites, this invention aims to solve the problem in the prior art of lacking high-specificity and high-sensitivity early diagnosis and prediction methods due to the unclear pathogenesis of HIV-associated non-alcoholic fatty liver disease, as well as the resulting difficulties in early disease screening, inability to conduct effective risk assessment, and clinical intervention.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows: The biomarkers comprise 2 types of intestinal microorganisms and 37 types of plasma metabolites, totaling 39 biomarkers. The intestinal microorganisms are Enterobacter spp. and Klebsiella spp.; the plasma metabolites are diacylglycerol, lysophosphatidic acid, 1-stearoyl-lysophosphatidylethanolamine, phosphatidylcholine, avocadoene-1-acetate, brassinosteroids, and 3β-hydroxy-Δ 5 -cholenic acid, 16,16-dimethyl-prostaglandin A2, 19(S)-hydroxy-eicosatetraenoic acid, carboprost, 15-methyl-15S-prostaglandin E2, 12-hydroxyoctadecanoic acid, N-vinyl-2-pyrrolidone, α-ketoglutarate, glycerol phosphatidylcholine, γ-tocopherol, palmitoyl lysophosphatidylcholine, phosphatidylethanolamine, glycerol phosphatidylethanolamine, glutamyl-glutamine, glutamate, L-3-aminodihydro- 2(3H)-furanolactone, morinone Z, ginkgolic acid, bepadinic acid, litmus acid, nitrile ketone, 5(S),6(R)-11-trans-dihydroxyeicosatetraenoic acid, 15-methylprostaglandin E1, 8,9-epoxy-tetradecanoic-14-enoic acid, glucoside, succinic acid semialdehyde, O-acetyl-L-serine, pipecolic acid, 3-oxo-methyl-L-DOPA, 4-(2-aminoethyl)-5-fluoro-1,2-benzenediol, L-3-cyanalanine.

[0008] The present invention further provides the application of the above-mentioned HIV-associated non-alcoholic fatty liver disease diagnostic predictive biomarkers.

[0009] Preferably, the biomarker is used in the preparation of pharmaceutical compositions for diagnosing or predicting HIV-associated non-alcoholic fatty liver disease.

[0010] Preferably, the pharmaceutical composition comprises an active ingredient that regulates the expression level or concentration of at least one of the 39 biomarkers, the active ingredient comprising one or more of the inhibitors, agonists or antagonists of the 39 biomarkers.

[0011] Preferably, the pharmaceutical composition is used to prepare a formulation for treating HIV-associated non-alcoholic fatty liver disease, the formulation further comprising a pharmaceutically acceptable excipient selected from at least one of pharmaceutically acceptable solvents, solubilizers, cosolvents, emulsifiers, osmotic pressure regulators, stabilizers, suspending agents, anti-adhesives, integrators, permeation enhancers, pH adjusters, buffers, surfactants, absorbents, diluents, filter aids, and sustained-release materials.

[0012] Preferably, the biomarkers are used to prepare a kit for diagnosing or predicting HIV-associated non-alcoholic fatty liver disease, the kit comprising a detection reagent for detecting at least one of the 39 biomarkers.

[0013] Preferably, the detection reagent includes one or more of nucleic acid primers, probes, or antibodies for detecting the intestinal microorganisms, and / or mass spectrometry analysis standards, chromatographic analysis standards, antibodies, or enzyme-linked immunosorbent assay reagents for detecting the plasma metabolites.

[0014] Preferably, the biomarker is used to construct a system for diagnosing or predicting HIV-associated non-alcoholic fatty liver disease.

[0015] Preferably, the diagnostic or predictive method of the system includes the following steps: obtaining a biological sample of the individual to be tested, the biological sample including a fecal sample and / or a plasma sample, then detecting the expression level or concentration of at least one of the 39 biomarkers in the biological sample, and finally comparing the expression level or concentration with a reference value.

[0016] The beneficial effects of this invention are:

[0017] 1. High Specificity and Systemic Relevance of Biomarker Combinations: This invention reveals and validates a multi-dimensional biomarker combination encompassing gut microbiota and plasma metabolites. This combination, from a systems biology perspective of the "gut microbiota-metabolite-host" interaction, can more comprehensively and profoundly reflect the unique pathophysiological state of HIV infection combined with non-alcoholic fatty liver disease. Compared to single microbiome or metabolome biomarkers, this multi-dimensional combination effectively avoids misdiagnosis due to individual differences, significantly improving the specificity and accuracy of diagnosis and prediction.

[0018] 2. This invention provides a core technological foundation for early, non-invasive, or minimally invasive diagnosis: The biomarkers involved in this invention can be detected by collecting fecal and plasma samples from patients. These samples are easily obtained, and the procedure is non-invasive or minimally invasive, offering good clinical applicability and patient compliance. Reagent kits or diagnostic methods developed using this biomarker combination are expected to achieve accurate screening in the early stages of the disease (such as when only hepatic steatosis is present), gaining a valuable time window for clinical intervention and thus preventing the progression of the disease to fatty liver, liver fibrosis, and even liver cancer.

[0019] 3. This invention elucidates the disease mechanism and opens up therapeutic targets: The specific biomarkers disclosed in this invention (such as the enrichment of Enterobacteriaceae and Klebsiella spp., and the dysregulation of a series of lipid, amino acid, and inflammation-related metabolites) provide novel clues and experimental evidence for elucidating the pathogenesis and development of non-alcoholic fatty liver disease in the context of HIV infection. It also provides targets and research directions for developing innovative drugs that target and regulate specific gut microbiota or correct imbalances in key metabolic pathways (such as microbial preparations, metabolic enzyme inhibitors, or agonists).

[0020] 4. It should also have diagnostic and treatment guidance value: The biomarker combination of this invention can be widely applied in multiple fields such as developing diagnostic reagent kits, constructing disease risk prediction models, and screening therapeutic drugs. Furthermore, by dynamically monitoring changes in the levels of these biomarkers in patients, treatment effects can be objectively assessed, disease prognosis predicted, and a basis can be provided for the individualized medical management of HIV-associated non-alcoholic fatty liver disease. Attached Figure Description

[0021] Figure 1 This invention includes the diversity and species composition analysis of the gut microbiota in the HIV-NAFLD group and the HIV group (A is the Alpha diversity comparison analysis, where the Ace index and Chao index assess the abundance differences of bacterial species between groups, and the Shannon index is used to assess the differences in bacterial species diversity between groups; B is the Beta diversity comparison analysis, which constructs a principal coordinate analysis based on weighted UniFrac distance and uses permutation multivariate ANOVA to test the differences between groups; C is the Venn diagram showing the number of ASVs in the two groups; D is the Bar diagram of the top 10 communities at the phylum level; E is the Bar diagram of the top 10 communities at the genus level).

[0022] Figure 2 The differences in gut microbiota between the HIV-NAFLD group and the HIV group at three taxonomic levels (phylum, genus, and species) in this invention are: A is the phylum level; B is the genus level; C is the species level.

[0023] Figure 3 This is a histogram showing the distribution of LDA values ​​(LAD score ≥ 3) of the two groups of gut microbiota at the phylum to species level in this invention.

[0024] Figure 4 This invention includes fecal differential metabolite profile analysis and significance analysis between the HIV-NAFLD group and the HIV group.

[0025] Figure 5 This is the pathway by which the fecal metabolite KEGG is differentially enriched in the HIV-NAFLD group and the HIV group in this invention.

[0026] Figure 6 This invention includes plasma differential metabolite profile analysis and significance analysis between the HIV-NAFLD group and the HIV group.

[0027] Figure 7 This invention relates to the differential enrichment pathway of KEGG, a differentially metabolized metabolite in plasma of the HIV-NAFLD group and the HIV group.

[0028] Figure 8 This is a differential enrichment pathway score map of KEGG, a differential metabolite in plasma of the HIV-NAFLD group and the HIV group in this invention.

[0029] Figure 9 This invention presents correlation analyses between key gut differential metabolites and plasma differential metabolites and differential bacterial genera in the HIV-NAFLD and HIV groups (A is the correlation analysis between gut differential metabolites and differential bacterial genera; B is the correlation analysis between plasma differential metabolites and differential bacterial genera; red represents positive correlation, blue represents negative correlation, * indicates significant correlation between metabolites and microorganisms: * p<0.05, ** p<0.01, *** p<0.001).

[0030] Figure 10 This invention presents the feature importance and ROC curve analysis of the classification model based on microbiome and plasma metabolites (A is the classification contribution model; B is the prediction model built with 2 gut microbes; C is the prediction model built with 37 plasma metabolites; D is the prediction model built with 2 gut microbes + 37 plasma metabolites; E is a comparison chart of the 3 models). Detailed Implementation

[0031] Unless otherwise specified, the experimental methods used in the following examples are conventional methods.

[0032] Unless otherwise specified, all materials and reagents used in the following examples are commercially available.

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0034] Example 1

[0035] The study included 60 HIV-infected patients, of whom 36 had NAFLD (HIV-NAFLD group) and the other 24 had HIV infection alone (HIV group).

[0036] 1. Method

[0037] 1.1 Data and Sample Collection

[0038] Demographic and clinical data of enrolled patients were recorded, and stool and blood samples were collected, processed, and preserved. Demographic data included age, sex, body mass index (BMI) = weight (kg) / height (m), waist circumference, hip circumference, HIV diagnosis time, ART initiation time, and specific protocol. Clinical data mainly included platelet count, liver and kidney function tests, CD4+ T and CD8+ T lymphocyte counts, lipid metabolism indicators such as total cholesterol, triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, fasting blood glucose assessment, and past medical history related to metabolism, such as hypertension, diabetes, and cardiovascular disease. Blood and stool samples were collected on the day of enrollment under standardized conditions. Participants fasted for at least 8 hours before blood was collected, and plasma was separated by centrifugation and stored at -80°C until analysis. Fresh stool samples were aliquoted into sterile tubes and transferred to a -80°C freezer for storage within 2 hours.

[0039] 1.2 Assessment of hepatic steatosis

[0040] Patients who met the inclusion criteria but not the exclusion criteria underwent transient elastography (FibroScan®, Echosens, Paris, France) in a fasting state to assess the degree of hepatic steatosis. At least 10 valid measurements were obtained for each subject. The final result was the median of the controlled attenuation parameter (CAP) and reported in decibels per meter (dB / m). A cap ≥ 238 dB / m was defined as fatty liver (238~259 dB / m was mild fatty liver, 260~291 dB / m was moderate fatty liver, and ≥ 292 dB / m was severe fatty liver).

[0041] 1.3 Statistical Analysis

[0042] Clinical characteristic data were statistically analyzed using SPSS version 26 (SPSS Inc., Chicago, Illinois, USA). Continuous variables were expressed as mean (standard deviation) or median (interquartile range), and categorical variables were expressed as frequency (percentage). Normality was tested for each parameter using the Shapiro-Wilk test. Continuous variables were analyzed using the Mann-Whitney U test or Student's t-test based on normality, and categorical variables were analyzed using the chi-square test.

[0043] 2. Results

[0044] Statistical analysis of baseline characteristics of the two groups of patients revealed significant differences between the HIV group and the HIV-NAFLD group in body mass index, CAP, CD8+ T cell count, CD4 / CD8 ratio, triglycerides, total cholesterol, LDL cholesterol, AST, ALT, GGT, and uric acid. The median CAP in the HIV-NAFLD group was 301 (range 277-317), classifying it as severe fatty liver. The ALT levels in the HIV-NAFLD group were significantly higher than those in the HIV-NAFLD group (HIV-NAFLD: 38.50 (24.00-60.75) vs. HIV: 20.50 (16.25-34.00), P < 0.00. The levels of aspartate aminotransferase (AST) [HIV-NAFLD: 27.00 (22.00-39.25) vs HIV: 21.00 (17.00-29.25), P=0.004] were significantly higher in the HIV group than in the HIV group. The levels of triglycerides (HIV-NAFLD: 2.65 (2.34) vs HIV: 1.39 (0.64), P=0.004), total cholesterol (HIV-NAFLD: 4.63 (1.07) vs HIV: 3.84 (1.03), P=0.006), and low-density lipoprotein cholesterol (LDL-C) (HIV-NAFLD: 2.77 (0.71) vs HIV: 2.22 (0.82), P=0.007) were also significantly higher in the HIV group than in the HIV group. There were no statistically significant differences between the two groups in terms of age, sex, history of metabolic diseases, liver stiffness value, time of HIV infection, duration of ART, and medication regimen. The overall data are shown in Table 1.

[0045] Table 1 General characteristics of the research subjects

[0046]

[0047] Example 2: Amplicon Sequencing Study of Gut Microbiota

[0048] 1. Method

[0049] 1.1 Sample Processing

[0050] First, the collected fecal samples need to be subjected to DNA extraction, polymerase chain reaction (PCR) amplification, and sequencing library construction. According to the FastPure Stool DNA Isolation Kit (MJYH, Shanghai, China) instructions, total genomic DNA of the microbial community is extracted. The integrity of the extracted genomic DNA is detected by 1% agarose gel electrophoresis, and the DNA concentration and purity are determined by NanoDrop2000 (ThermoFisher Scientific, USA).

[0051] Using the extracted DNA as a template, the full-length 16S rRNA gene was amplified by PCR using barcode-containing primers. The PCR reaction system was as follows: 4 μL 5×FastPfu buffer, 2 μL 2.5 mM dNTPs, 0.8 μL upstream primer (5 μM), 0.8 μL downstream primer (5 μM), 0.4 μL FastPfu polymerase, 0.2 μL BSA, 10 ng template DNA, and a final volume of 20 μL. Each sample was replicated in triplicate. The amplification program was as follows: 95℃ pre-denaturation for 3 min, 27 cycles (95℃ denaturation for 30 s, 60℃ annealing for 30 s, 72℃ extension for 30 s), followed by a stable extension at 72℃ for 10 min, and finally storage at 4℃ (PCR instrument: T100 Thermal Cycler PCR, USA). After 2% agarose gel electrophoresis, the samples were purified using magnetic beads, and the purified products were quantified using a Qubit 4.0 (Thermo Fisher Scientific, USA). The samples were then mixed in appropriate proportions according to the sequencing volume requirements for each sample.

[0052] Sequencing libraries were constructed using the SMRTbell prep kit 3.0: (1) DNA damage repair; (2) end repair; and (3) adapter ligation. Sequencing was performed using the PacBio Sequel IIe System. HiFi reads were generated from the sequenced subreads using the CCS mode of SMRT-Link v11.0 for subsequent data analysis.

[0053] 1.2 Analysis of Gut Microbiota Sequencing Data

[0054] Data from each sample was differentiated based on the barcode sequence, and length filtering and orientation correction were performed, retaining sequences of 1000-1800 bp (bacteria) / 300-900 bp (fungi). Using default parameters, the DADA2 plugin in the Qiime2 workflow was used to denoise the optimized sequences after quality control assembly. Sequences denoised by DADA2 are typically referred to as ASVs (amplicon sequence variants). All non-compliant sequences annotated in all samples were removed. To minimize the impact of sequencing depth on subsequent Alpha and Beta diversity data analysis, the sequence count for all samples was flattened to 6000; even after flattening, the average sequence coverage for each sample remained at 99.09%. Based on the Sliva 16S rRNA gene database, the Naive Bayes classifier in Qiime2 was used for species taxonomy analysis of the ASVs.

[0055] 1.3 Statistical Analysis of Gut Microbiota Sequencing Data

[0056] The Ace, Chao, and Shannon indices of Alpha diversity were calculated using Mothur software (http: / / www.mothur.org / wiki / Calculators), and inter-group differences in Alpha diversity were analyzed using the Wilxocon rank-sum test. Principal coordinates analysis (PCoA) based on the Bray-Curtis distance algorithm was used to test the similarity of microbial community structure between samples, and permutation multivariate ANOVA was used to test whether the differences in microbial community structure between the two groups were significant. The Wilxon rank-sum test was used to compare the different species in the two groups, and LEfSe analysis (http: / / huttenhower.sph.harvard.edu / LEfSe) was used to identify bacterial taxa with significant differences in abundance from the phylum to the species level between different groups, with the criteria being P < 0.05 and a linear discriminant analysis (LDA) score ≥ 3.

[0057] 2. Results

[0058] 2.1 Differences in gut microbiota composition between HIV-NAFLD group and HIV group

[0059] The results showed no statistically significant differences in the three diversity indices: Ace index (P=0.316), Chao index (P=0.298), and Shannon index (P=0.850), suggesting that the two groups of patients were similar in terms of bacterial abundance and diversity. Figure 1(A). Further construct a principal coordinate analysis score map based on weighted UniFrac distance ( Figure 1 The Beta diversity of the HIV-NAFLD group and the HIV group was compared using a Venn diagram (P=0.020). To assess the impact of HIV-NAFLD on gut microbiota composition, ASV of the HIV and HIV-NAFLD groups was compared and analyzed. Figure 1 (C) shows that the number of ASVs shared between the two groups is relatively small, accounting for only 10.76% (324) of the total ASVs. Notably, the HIV-NAFLD group has a large number of unique ASVs, as high as 1554, accounting for 51.63% of the total ASVs in this group. In contrast, the number of unique ASVs in the HIV group is 1132, accounting for 37.61% of the total ASVs in this group. Further investigation into the specific changes in the microbial composition behind these differences, and in-depth analysis of the microbial structure of the two groups at the phylum and genus levels, revealed that the common dominant bacteria at the phylum level in both groups are mainly Firmicutes (Bacillota), Pseudomonadota, Bacteroidota, and Actinomycetota, etc. Figure 1 In the middle D), the common dominant bacteria at the genus level are mainly Blautia, Streptococcus, Escherichia, and Ruminococcus, etc. Figure 1 (E).

[0060] The Wilcoxon rank-sum test was further used to compare the top 10 most abundant gut microbiota at the phylum, genus, and species levels between the HIV-NAFLD group and the HIV group. Figure 2 Proteobacteria were significantly enriched in the HIV-NAFLD group (P < 0.050). Figure 2 (A). Among the top 10 most abundant differentially abundant bacterial genera, the abundance of Gram-negative bacteria *Klebsiella*, *Dialister*, *Citrobacter*, and *Enterobacter* was significantly increased in the HIV-NAFLD group compared to the HIV group. Beneficial bacteria genera *Faecalibacterium* and *Anaerobutyricum*... Figure 2 (B) and beneficial bacteria Faecalibacterium_prausnitzii and Anaerobutyricum_hallii ( Figure 2 The abundance of C was significantly reduced in the HIV-NAFLD group.

[0061] LEfSe analysis shows ( Figure 3 The differences in gut microbiota between the two groups of patients at different taxonomic levels from phylum to species include enriched bacteria such as Gammaproteobacteria, Pseudomonadota, Enterobacter ales, and Enterobacteriaceae. These are all opportunistic pathogens that are enriched in the HIV-NAFLD group and contribute significantly to the differences in gut microbiota composition, which may be related to the inflammatory response associated with fatty liver.

[0062] Example 3: Non-targeted metabolomics study of feces and plasma

[0063] 1. Method

[0064] 1.1 Sample Processing

[0065] fecal samples

[0066] ① Take 50mg±5mg of fecal sample into a 2mL centrifuge tube and add a 6mm diameter grinding bead;

[0067] ② Metabolite extraction was performed using 400 μL of extraction buffer (methanol:water = 4:1 (v:v)) containing 0.02 mg / mL of internal standard (L-2-chlorophenylalanine);

[0068] ③ The sample solution was ground in a cryo-tissue homogenizer for 6 min (-10℃, 50 Hz);

[0069] ④ Low-temperature ultrasonic extraction for 30 min (5℃, 40 kHz);

[0070] ⑤ Place the sample at -20℃ for 30 min, centrifuge for 15 min (4℃, 13000 g), and transfer the supernatant into a vial with an inner tube for analysis.

[0071] ⑥ Take 20 μL of supernatant from each sample, mix them, and use them as quality control samples;

[0072] plasma samples

[0073] ① Accurately transfer 100 μL of sample into a 1.5 mL centrifuge tube;

[0074] ② Add 100 μL of extraction buffer (methanol:acetonitrile = 1:1 (v:v)), containing four internal standards (L-2-chlorophenylalanine (0.02 mg / mL), etc.);

[0075] ③ After vortexing for 30 seconds, extract using low-temperature ultrasonication for 30 minutes (5℃, 40KHz).

[0076] ④ Let the sample stand at -20℃ for 30 minutes;

[0077] ⑤ Centrifuge for 15 min (13000 g, 4℃), transfer the supernatant, and dry it with nitrogen gas;

[0078] ⑥ Add 100 μL of reconstitution solution (acetonitrile:water = 1:1) to reconstitute;

[0079] ⑦ Vortex mix for 30 seconds, then perform low-temperature ultrasonic extraction for 5 minutes (5℃, 40KHz).

[0080] ⑧ Centrifuge for 10 min (13000 g, 4℃), transfer the supernatant to a vial with an inner tube for analysis;

[0081] ⑨ Take 20 μL of supernatant from each sample, mix them, and use them as quality control samples;

[0082] 1.2 LC-MS Analysis

[0083] The ultra-high performance liquid chromatography-tandem Fourier transform mass spectrometry (UHPLC-Exploris240) system was used.

[0084] Chromatographic conditions:

[0085] The chromatographic column was an ACQUITY UPLC HSS T3 (100 mm × 2.1 mm id, 1.8 µm; Waters, Milford, USA); mobile phase A was 95% water + 5% acetonitrile (containing 0.1% formic acid), and mobile phase B was 47.5% acetonitrile + 47.5% isopropanol + 5% water (containing 0.1% formic acid). The injection volume was 3 μL, and the column temperature was 40 °C.

[0086] Mass spectrometry conditions:

[0087] The sample was ionized by electrospray ionization, and mass spectrometry signals were acquired using both positive and negative ion scanning modes. Specific parameters are shown in Table 2.

[0088] Table 2 Mass Spectrometry Parameter Conditions

[0089]

[0090] Quality control:

[0091] Quality control (QC) samples are prepared by mixing equal volumes of extracts from all samples. Each QC sample has the same volume as the sample and is processed and tested using the same methods as the analytical samples. During instrumental analysis, a QC sample is inserted every 5-15 analytical samples to examine the stability of the entire testing process.

[0092] Substance identification and analysis:

[0093] After the initial setup, the raw LC-MS data was imported into the metabolomics processing software Progenesis QI v3.0 (Waters Corporation, Milford, USA) for processing, including baseline correction, peak identification, integration, retention time correction, and peak alignment. This resulted in a data matrix containing retention time, mass-to-charge ratio (m / z), and peak intensity information. Subsequently, the software was used to identify metabolites from characteristic peaks. The MS and MS / MS data were matched against the public metabolite databases HMDB (http: / / www.hmdb.ca / ) and Metlin (https: / / metlin.scripps.edu / ), with the MS quality error tolerance set to <10 ppm. The metabolite identification results were also referenced against the MS / MS matching scores. To ensure data quality, the data matrix after database searching underwent preprocessing. First, the 80% rule was used for missing value filtering, retaining only variables whose non-zero value proportion was greater than or equal to 80% in at least one sample set. The remaining missing values ​​were imputed using the minimum value of the original matrix. To reduce errors caused by sample preparation and instrument fluctuations, a summation normalization method was used to normalize the peak intensities of the sample mass spectrometry. Furthermore, variables with a relative standard deviation greater than 30% in the QC samples were removed, and the data were transformed using log10 to obtain the final data matrix for subsequent statistical analysis.

[0094] 1.3 Statistical Analysis of Metabolomics Data

[0095] The ropls package (version 1.6.2) in R was used to perform partial least squares discriminant analysis (PLS-DA) and orthogonal partial least squares discriminant analysis (OPLS-DA) on the preprocessed data matrix, and the stability of the model was evaluated using a 7-cycle cross-validation. The selection of significantly differentially expressed metabolites was determined based on the variable projection importance and p-value obtained from the OPLS-DA model; metabolites with VIP > 1 and p < 0.05 were considered significantly differentially expressed metabolites. Subsequently, metabolic pathway annotation of the differentially expressed metabolites was performed using the Kyoto Encyclopedia of Genetics and Genomes (KEGG) database to identify the pathways involved. Pathway enrichment analysis was performed using the scipy.stats package in Python, and Fisher's exact test was used to identify the biological pathways most relevant to the experimental treatment.

[0096] 2. Results

[0097] After preprocessing, the relative concentrations of each metabolite in the raw sequencing data were standardized using unit variance. Metabolites identified in both anion and cation combined modes were analyzed using PLS-DA (…). Figure 4 (A) and OPLS-DA ( Figure 4 Analysis was conducted in group B to comprehensively assess the overall differences in fecal and plasma metabolite levels between the HIV-NAFLD group and the HIV group. The fecal metabolomics analysis showed that the PLS-DA model had good predictive ability. Figure 4 (C) No overfitting was observed. According to the PLS-DA and OPLS-DA analysis results shown in the figure, there were significant differences in the intestinal flora metabolite profiles between the two groups of patients.

[0098] Based on the variable projection importance (VIP) obtained from the OPLS-DA model, biologically significant differentially expressed metabolites were identified using VIP>1 and P<0.05 for comparison of relative metabolite concentrations between the two groups as statistical significance criteria. A total of 551 differentially expressed metabolites were detected between the two groups. The HIV-NAFLD group showed upregulation of 187 metabolites and downregulation of 364 metabolites compared to the HIV group. Among these, hepatoprotective substances such as N-lactylglycine, 5-phenylvaleric acid, dodecanoic acid, and kynurenic acid were significantly downregulated in the HIV-NAFLD group compared to the HIV group, while hepatotoxic substances such as cholesterol and γ-glutamyl-L-putrescine were significantly upregulated. Figure 4 (D).

[0099] Further KEGG functional pathway enrichment analysis of differentially metabolites revealed the top 20 pathways with the highest significance and enrichment rate. Figure 5 The differential pathways that were significantly enriched in the feces of the two groups mainly involved lipid, amino acid and energy metabolism, including primary bile acid biosynthesis, arginine and proline metabolism, tryptophan metabolism, β-alanine metabolism, riboflavin metabolism and degradation of chloroalkanes and chloroolefins.

[0100] The PLS-DA model of plasma metabolites suggests that there are overall differences in plasma metabolites between the HIV-NAFLD and HIV groups. Figure 6In plasma, 18 upregulated differential metabolites were detected and identified in the combined cation and anion mode of plasma samples, including glutamate, 2-ketoglutaric acid, and succinate semialdehyde, while 105 downregulated differential metabolites were identified, such as isolithocholic acid, arachidonic acid, 7-ketolithocholic acid, N-stearoyl tyrosine, 13-L-hydroperoxylinoleic acid, and 20-carboxyarachidonic acid.

[0101] Further KEGG pathway analysis was performed on the differentially metabolites. Figure 7 The differentially enriched pathways in the HIV-NAFLD group and the HIV group include antioxidant response-related signaling pathways such as the hypoxia-inducible factor-1 (HIF-1) signaling pathway, as well as pathways related to butyrate metabolism, arginine biosynthesis and alanine, aspartic acid and glutamate metabolism.

[0102] Difference abundance score ( Figure 8 The results showed that pathways related to antioxidation, such as the HIF-1 signaling pathway, differed significantly between the two groups, and their expression trend was downregulated in the HIV-NAFLD group.

[0103] In summary, the commonalities between gut microbiota and plasma metabolites are that most differential metabolites belong to amino acids, lipids, and bile acids and are mainly enriched in metabolic pathways related to these substances. Among them, the amino acid metabolic pathway is the main differential enrichment pathway for gut microbiota metabolites, while the antioxidant-related signaling pathway (hypoxia-inducible factor-1 signaling pathway) shows significant differences in the enrichment pathways of plasma metabolites between the HIV-NAFLD group and the HIV group.

[0104] Example 4: Association Analysis of Gut Microbiota with Fecal and Plasma Metabolomics

[0105] 1. Method

[0106] Spearman correlation coefficient was used to identify differentially metabolites influenced by differential microorganisms. The Spearman correlation coefficient (cor) measures the strength and direction of the monotonic relationship between two variables, ranging from -1 to +1. cor ≥ 0.6 indicates a strong correlation, 0.4 ≤ cor < 0.6 indicates a moderate correlation, 0.2 ≤ cor < 0.4 indicates a weak correlation, and cor < 0.2 indicates a very weak correlation. A two-tailed test was used to calculate the p-value, and the Benjamini-Hochberg method was used to correct the p-value to control for false detection. A corrected p-value < 0.05 was considered statistically significant. Statistical analysis was performed using the scipy.stats package in Phython (version 2.7.10). Correlation results are presented as a heatmap, with significant correlations marked with an asterisk (* P < 0.05, ** P < 0.01, *** P < 0.001).

[0107] 2. Results

[0108] A total of 23 key products from 7 classes (such as amino acids, bile acids, and lipids) of differentially metabolized gut microbiota were screened, which may be involved in the pathogenesis of fatty liver. An association was found between differentially metabolized gut microbiota and their metabolites. Figure 9 In the HIV-NAFLD group, the significantly decreased abundance of the genera *Faecalibacterium* and *Anaerobutyricum* in the gut microbiota was positively correlated with the downregulation of hepatoprotective gut microbiota metabolites, including dodecanedioic acid (P < 0.001), kynurenic acid (P < 0.001), and 5-Phenylvaleric acid. Conversely, the significantly increased abundance of *Klebsiella*, *Enterobacter*, *Citrobacter*, and *Enterococcus* was negatively correlated with hepatoprotective substances. Furthermore, Faecalibacterium and Anaerobutyricum were negatively correlated with the significantly upregulated hepatotoxic substances Cholesterol and Gamma-Glutamyl-L-Putrescine, while Klebsiella, Enterobacter, Citrobacter, and Enterococcus were positively correlated with both substances.

[0109] Furthermore, a correlation analysis was also performed on differentially expressed bacterial species and plasma metabolites. A total of 15 key metabolites from 9 classes (such as amino acids, bile acids, and unsaturated fatty acids) were screened from the differentially expressed plasma metabolites, which may be involved in the pathogenesis of fatty liver. The results showed that... Figure 9 In group B, beneficial bacteria genera *Anaerobutyricum* and *Faecalibacterium* were positively correlated with hepatoprotective plasma metabolites. In the HIV-NAFLD group, the abundance of *Anaerobutyricum* and *Faecalibacterium* decreased, and 20-Carboxy Arachidonic acid, a hepatoprotective substance, was significantly downregulated. Potentially pathogenic bacteria genera were negatively correlated with hepatoprotective plasma metabolites. In the HIV-NAFLD group, the abundance of *Klebsiella*, *Citrobacter*, and *Enterobacter* increased, while N-Stearoyl Tyrosine, Gamma-Tocopherol, and 20-Carboxy Arachidonic acid, substances with hepatoprotective effects, were significantly downregulated.

[0110] The above results reveal the synergistic effect of gut microbiota dysbiosis and metabolic disorders in HIV-NAFLD patients. In particular, the gut microbiota may participate in the disease progression by regulating metabolism, increasing substances that cause liver damage, and decreasing substances that protect the liver.

[0111] Example 5: Random Forest Machine Algorithm

[0112] 1. Method

[0113] To identify biomarkers in gut microbiota and plasma metabolites that could differentiate between the HIV-NAFLD group and the HIV group, a random forest model based on gut microbiota and plasma metabolites was constructed using the R package "Random Forest" (version 4.7-1.1) to identify important features. The area under the receiver operating characteristic curve (ROC) was plotted using the R package "pROC" (version 1.18.0) to evaluate model performance.

[0114] 2. Results

[0115] A total of 39 biomarkers were identified, including 2 gut microbes and 37 plasma metabolites. The results are shown in Table 3.

[0116] Table 3 39 markers

[0117]

[0118] The model achieved the lowest classification error when selecting 2 gut microbes and 37 plasma metabolites. The contributions to classification accuracy were ranked, showing that Enterobacter and glutamate contributed the most to the model's predictions, followed by Stearoyl Lysophosphatidylethanolamine and 12-Hydroxyoctadecanoic acid. Figure 10 (A). Gut microbiota has weak discriminative ability when used alone as a biomarker (AUC=0.625). Figure 10 (B), while metabolomics alone provides better predictive performance (AUC=0.913). Figure 10 In the middle C), the synergistic effect of gut microbiota and plasma metabolites provides the best predictive performance (AUC=0.925). Figure 10 (D). When gut microbiota and plasma metabolomics are used in combination, the predictive power of the combined model is significantly better than that of the single-analysis model (AUC: 0.925 vs. 0.913 / 0.625). Figure 10 The study (E) validated the importance of multi-omics integration in improving the accuracy of disease classification.

[0119] The above-described embodiments are merely preferred embodiments of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A diagnostic predictive biomarker for HIV-associated non-alcoholic fatty liver disease, characterized in that: The biomarkers comprise 2 gut microbiota and 37 plasma metabolites, totaling 39 biomarkers. The gut microbiota are Enterobacteriaceae and Klebsiella; the plasma metabolites are diacylglycerol, lysophosphatidic acid, 1-stearoyl-lysophosphatidylethanolamine, phosphatidylcholine, avocadoene-1-acetate, brassinosteroids, and 3β-hydroxy-Δ 5 -cholenic acid, 16,16-dimethyl-prostaglandin A2, 19(S)-hydroxy-eicosatetraenoic acid, carboprost, 15-methyl-15S-prostaglandin E2, 12-hydroxyoctadecanoic acid, N-vinyl-2-pyrrolidone, α-ketoglutarate, glycerol phosphatidylcholine, γ-tocopherol, palmitoyl lysophosphatidylcholine, phosphatidylethanolamine, glycerol phosphatidylethanolamine, glutamyl-glutamine, glutamate, L-3-aminodihydro- 2(3H)-furanolactone, morinone Z, ginkgolic acid, bepadinic acid, litmus acid, nitrile ketone, 5(S),6(R)-11-trans-dihydroxyeicosatetraenoic acid, 15-methylprostaglandin E1, 8,9-epoxy-tetradecanoic-14-enoic acid, glucoside, succinic acid semialdehyde, O-acetyl-L-serine, pipecolic acid, 3-oxo-methyl-L-DOPA, 4-(2-aminoethyl)-5-fluoro-1,2-benzenediol, L-3-cyanalanine.

2. The application of the HIV-associated non-alcoholic fatty liver disease diagnostic predictive biomarker according to claim 1, characterized in that: The biomarker is used to prepare pharmaceutical compositions for the diagnosis or prediction of HIV-associated non-alcoholic fatty liver disease.

3. The application according to claim 2, characterized in that: The pharmaceutical composition includes an active ingredient that regulates the expression level or concentration of at least one of the 39 biomarkers, the active ingredient including one or more of the inhibitors, agonists or antagonists of the 39 biomarkers.

4. The application according to claim 2, characterized in that: The pharmaceutical composition is used to prepare a formulation for treating HIV-associated non-alcoholic fatty liver disease. The formulation further includes pharmaceutically acceptable excipients selected from at least one of pharmaceutically acceptable solvents, solubilizers, cosolvents, emulsifiers, osmotic pressure regulators, stabilizers, suspending agents, anti-adhesives, integrators, penetration enhancers, pH adjusters, buffers, surfactants, absorbents, diluents, filter aids, and sustained-release materials.

5. The application of the HIV-associated non-alcoholic fatty liver disease diagnostic predictive biomarker according to claim 1, characterized in that: The biomarkers are used to prepare a kit for diagnosing or predicting HIV-associated non-alcoholic fatty liver disease, the kit comprising a detection reagent for detecting at least one of the 39 biomarkers.

6. The application according to claim 5, characterized in that: The detection reagents include one or more of nucleic acid primers, probes, or antibodies for detecting the intestinal microorganisms, and / or mass spectrometry standards, chromatographic standards, antibodies, or enzyme-linked immunosorbent assay reagents for detecting the plasma metabolites.

7. The application of the HIV-associated non-alcoholic fatty liver disease diagnostic predictive biomarker according to claim 1, characterized in that: The biomarkers are used to construct a system for diagnosing or predicting HIV-associated non-alcoholic fatty liver disease.

8. The application according to claim 8, characterized in that: The system's diagnosis or prediction includes the following steps: obtaining a biological sample from the individual to be tested, the biological sample including a fecal sample and / or a plasma sample, then detecting the expression level or concentration of at least one of the 39 biomarkers in the biological sample, and finally comparing the expression level or concentration with a reference value.

Images