Method for assessing the acquisition risk of antibiotic resistance genes by wildlife gut

Abstract

The application provides a method for evaluating the acquisition risk of antibiotic resistance genes in the intestinal tract of wild animals, comprising: collecting leaf samples of forage plants and fecal samples of wild animals from a human interference area in the habitat of the target wild animals, respectively, to obtain plant leaf circle microbial metagenome data and intestinal tract microbial metagenome assembled genomes; determining the phylum of high-risk antibiotic resistance genes in the leaf circle microorganism and the phylum of the intestinal tract microorganism, respectively; if the phylum of the leaf circle microorganism is Proteobacteria or Actinobacteria, and the phylum of the intestinal tract microorganism is Firmicutes or Actinobacteria, it is determined that the acquisition risk is high. The method can accurately evaluate the acquisition risk of antibiotic resistance genes in the intestinal tract microorganism of wild animals to the leaf circle microorganism of forage plants under human interference, and has important theoretical significance and practical value for improving the protection effect of wild animals, improving the management mode of nature reserves, and formulating detection and management strategies for drug-resistant genes.

Description

Technical Field

[0001] This application relates to the field of ecological risk assessment. Specifically, this application relates to a method for assessing the acquired risk of antibiotic resistance genes in the gut of wild animals. Background Technology

[0002] The "One Health" concept emphasizes the close connection between human, animal, and environmental health. While the establishment of nature reserves has yielded significant results in wildlife conservation, spatial overlap between protected areas and human production and living areas remains widespread globally. Agriculture, grazing, and tourism continue to alter natural habitats, promoting the spread of antibiotic resistance genes (ARGs) in the environment. ARGs are now considered emerging environmental pollutants, and their ecological risks depend not only on abundance but also on characteristics such as host pathogenicity, genetic background, and mobility. Wild animals may ingest exogenous ARGs while consuming plants. These ARGs have the potential to be acquired by gut microbiota through horizontal gene transfer, thereby disrupting gut microbiota homeostasis, increasing the proportion of drug-resistant bacteria, and posing a potential threat to animal health.

[0003] However, not all ARGs pose the same ecological risk. An omics-based ARG risk grading framework proposes classifying ARGs into different grades based on abundance, mobile elements (MGEs), and host pathogenicity. Rank I ARGs, possessing high abundance, mobility, and virulence, are defined as high-risk genes with the potential for cross-host transmission and stable integration. Although Rank I ARGs reflect potential danger, their effective acquisition and expression in wild animals are still limited by host specificity and ecological environment. Therefore, relying solely on grading results is insufficient to assess their actual ecological risk. Existing studies have shown that when there is high sequence homology between exogenous ARGs and the gut microbiome, they are more likely to be integrated into the host microbiome through homologous reassortment, thereby affecting its function. Multiple pieces of evidence show that wild animals can acquire gene fragments homologous to exogenous ARGs through the food chain. Therefore, when assessing the ecological risk of Rank I ARGs under human interference, in addition to grading characteristics, their association with gut microbiota should also be considered.

[0004] Therefore, methods for assessing the acquired risk of antibiotic resistance genes in the gut of wild animals still need further research. Summary of the Invention

[0005] This application aims to address, at least to some extent, the technical problems existing in the prior art. To this end, this application proposes a method for assessing the acquired risk of antibiotic resistance genes in the gut microbiota of wild animals. Using this method, the acquired risk of antibiotic resistance genes from phytophyte microorganisms in wild animals under human interference can be accurately assessed. This has significant theoretical and practical value for improving the effectiveness of wildlife conservation, refining the management model of nature reserves, developing strategies for detecting and managing antibiotic resistance genes, and promoting the "One Health" concept in the field of wildlife conservation.

[0006] This application is based on the inventor's discoveries and understanding of the following facts and problems:

[0007] The phyllosphere microbial community is an important "niche" for ARG enrichment due to frequent gene flow between it and environmental microorganisms. Human disturbance is significantly correlated with changes in phyllosphere microbial community structure, ARG abundance, and diversity. When assessing the ecological risk of phyllosphere Rank I ARGs under human disturbance, in addition to hierarchical characteristics, more attention should be paid to their homology and potential collinearity with gut microbiota, as homology can significantly increase the probability of exogenous gene uptake and utilization.

[0008] Significant differences exist in the ecological habits and exposure risks of various wild animals, making non-human primates an ideal group for studying human disturbance and health risks. The Sichuan golden snub-nosed monkey, a typical semi-leavenous primate, exhibits a highly sensitive gut microbiota composition to environmental changes and is considered a "sentinel animal" for monitoring the impact of human disturbance. In the Baihe Nature Reserve, some monkey groups have overlapping ranges with surrounding grazing, foraging, and tourism areas, making them more susceptible to exposure to human-derived microorganisms and resistance factors. Human disturbance significantly alters the gut microbiota composition and function of Sichuan golden snub-nosed monkeys and promotes the accumulation of aeroid glutamate (ARG). Different types of disturbance further affect the structure and ecological function of their gut microbiota. The reserve contains a monkey group affected by human activities and a natural population far removed from human activity, providing an ideal research framework for exploring the transmission risk of foodborne ARG in the gut of wild animals.

[0009] Despite significant progress in the construction and management of protected areas, several key scientific questions remain to be addressed in current conservation practices. For example, under human disturbance, are wild animals at risk due to the enrichment of antibiotic resistance genes in the phytosphere microorganisms of their habitats? If so, through what mechanisms are these resistance genes transferred to and acquired by wild animals?

[0010] Based on this, this application takes the "human-habitat-wildlife" interaction in Baihe Nature Reserve as its starting point and the Sichuan golden snub-nosed monkey as its research subject. Starting from the potential link between the foliar microorganisms of its preferred plants and the gut microbiome, it systematically analyzes ARGs, MGEs, and virulence factors (VFs) in the foliar microbial community, and identifies high-risk Rank I ARGs according to the omics assessment framework. On this basis, homology analysis is used to explore the potential homology between foliar ARG fragments and Sichuan golden snub-nosed monkey gut microbial MAGs. Simultaneously, the impact of human interference on gut microbial function and resistance characteristics is assessed by combining KEGG metabolic pathway annotation and differences in Sichuan golden snub-nosed monkey fecal metabolome data. Thus, by comprehensively using metagenomics, metabolomics, and homology analysis, the potential pathways and ecological effects of the transmission of food plant foliar ARGs to the Sichuan golden snub-nosed monkey gut are elucidated. The results show that significant differences in the composition and structure of foliar microbial ARGs and the relative enrichment of high-risk Rank I ARGs are significantly correlated with human interference. The primary hosts of ARGs in the leaf stalk are Proteobacteria and Actinobacteria, while the primary recipients of ARGs in the gut are Firmicutes and Actinobacteria, exhibiting a clear cross-niche transmission characteristic. Under human disturbance conditions, the gut microbial carbon metabolism pathways and fecal metabolite composition of Sichuan golden monkeys showed significant differences. The findings of this application not only provide a scientific basis for the monitoring and control of emerging pollutants (ARGs) in nature reserves, but also offer a reference for optimizing the management of residents' production and daily life in protected areas and for the protection of wildlife populations. This has significant theoretical and practical value for improving the effectiveness of wildlife protection, refining the management model of nature reserves, developing strategies for detecting and managing drug-resistant genes, and promoting the "One Health" concept in the field of wildlife protection.

[0011] Therefore, this application proposes a method for assessing the acquired risk of antibiotic resistance genes in the gut of wild animals. According to embodiments of this application, the method includes the following steps:

[0012] Leaf samples and fecal samples of the preferred plants of the target wild animals were collected from areas of human disturbance within the target wild animal habitat.

[0013] Metagenomic sequencing was performed on the plant leaf circle microorganisms in the leaf samples of the edible plants to obtain plant leaf circle microbial metagenomic data; metagenomic sequencing was performed on the fecal samples to obtain gut microbial metagenomic data, and the metagenomic genome was reconstructed and assembled based on the gut microbial metagenomic data;

[0014] Based on the metagenomic data of plant chloroplast microorganisms, antibiotic resistance genes with high risk levels were identified according to predetermined risk grading criteria.

[0015] To analyze the cross-niche transmission potential of the high-risk antibiotic resistance genes from the plant leaf zone to the wild animal gut;

[0016] The analysis includes: determining the high-risk antibiotic resistance genes in the phylum phylum phylum phylum phylum and the phylum gut microbiome based on the metagenomic data of the plant phylum phylum phylum and the metagenomic data of the gut microbiome, respectively;

[0017] If the host microbial phylum is Proteobacteria or Actinobacteria, and the recipient microbial phylum is Firmicutes or Actinobacteria, then the target wild animal's gut is determined to have a high risk of acquiring the high-risk antibiotic resistance gene.

[0018] According to the method of this application embodiment, firstly, by simultaneously collecting leaf and fecal samples from plants in areas affected by human disturbance, a complete risk diffusion pathway research system of "environmental exposure source (leaf circle) - animal intestinal receptor" is established; then, by performing metagenomic sequencing on plant leaf circle microorganisms, the exogenous resistance gene spectrum of wild animals through ingestion is directly obtained, and high-risk antibiotic resistance genes are screened out; finally, through cross-niche microbial phylum tracing analysis, a specific high-risk gene transmission pathway of "leaf circle-origin Proteobacteria / Actinomycete host - intestinal Firmicutes / Actinomycete receptor" is discovered. By determining whether leaf circle microorganisms and intestinal microorganisms belong to these two phyla respectively, the acquired risk of antibiotic resistance genes from plant leaf circle microorganisms in wild animals under human disturbance can be accurately assessed. This can not only provide a scientific basis for the monitoring and control of emerging pollutants (ARGs) in nature reserves, but also provide a reference for optimizing the production and life management of residents in protected areas and the protection strategies for wild animal populations. This will help improve the effectiveness of wildlife protection, improve the management model of nature reserves, and promote the "One Belt One Road" initiative. The concept of "One Health" has significant theoretical and practical value in the field of wildlife conservation.

[0019] According to embodiments of this application, the target wild animal is a semi-leaved non-human primate, preferably the Sichuan golden snub-nosed monkey. This type of animal is highly dependent on plant-based foods, which can significantly amplify the risk of exposure to antibiotic resistance genes in the environment and rapidly transmit them to the gut microbiota. Simultaneously, its close evolutionary relationship with humans and similar gut physiological structures make the cross-niche transmission mechanism of antibiotic resistance genes obtained using this animal model a valuable reference for understanding human zoonotic diseases.

[0020] According to embodiments of this application, the analysis further includes: performing homology alignment analysis between the high-risk antibiotic resistance gene sequence and the sequence of the metagenomic assembly genome to screen candidate homologous regions with common origin characteristics at the sequence level; based on this, performing whole-genome collinearity alignment on the genomic fragments containing the candidate homologous regions to identify the consistency of the arrangement of the resistance gene and its upstream and downstream genetic background in the genomes of microorganisms in different ecological niches, and identifying collinear conserved blocks with a length greater than a preset threshold. Homology analysis is used to eliminate non-originating similarity interference caused by the widespread conservatism of resistance genes in nature, and collinearity analysis is used to further determine whether the resistance gene has undergone cross-niche transfer as a complete genetic segment containing multiple functional genes, thereby distinguishing between the background presence and acquired introduction of the resistance gene. Based on the collinear conserved blocks, and combined with the relative abundance of the corresponding metagenomic assembly genome in the gut microbiota, an evaluation index is calculated to quantify the acquired risk of the high-risk antibiotic resistance gene spreading across ecological niches from the plant foliage to the gut of wild animals.

[0021] In some embodiments, the preset threshold for similarity is ≥80%, and the preset length of the collinear conservative block is greater than 50bp.

[0022] According to an embodiment of this application, the evaluation index is calculated using the following formula:

[0023]

[0024] Wherein, CAP is the evaluation index;

[0025] M represents the total number of metagenomic assemblages that share collinear conserved blocks with the high-risk antibiotic resistance gene, characterizing the host availability of the antibiotic resistance gene in the gut microbiota; Ci represents the number of collinear conserved blocks in the i-th metagenomic assemblages that form with the specific high-risk antibiotic resistance gene, characterizing the structural integration and genetic retention capacity of the antibiotic resistance gene in that host; Ai represents the total relative abundance of the i-th metagenomic assemblages in the gut microbiota, characterizing the weight of exposure and spread of the antibiotic resistance gene at the community level through host amplification; by weighting Ci and Ai and summing them over M available hosts, the evaluation index simultaneously reflects the cross-host availability, single-host integration capacity, and community spread potential of the antibiotic resistance gene, thereby avoiding misjudgments based solely on a single homology or single transfer event. The CAP value is used to characterize the risk intensity of cross-niche acquisition and spread of a specific high-risk antibiotic resistance gene in the gut microbiota. The larger the CAP value, the wider the host range that the antibiotic resistance gene can be integrated into, the stronger its retention ability in the host, and the higher its amplification potential in the community, thus the greater its risk of spread and colonization.

[0026] According to embodiments of this application, the analysis further includes: based on the homology comparison analysis and whole-genome collinearity comparison analysis results, and combined with the taxonomic annotation information of the plant phylum microbial metagenomic data and the gut microbial metagenomic data, determining the distribution characteristics of phylum microorganisms and gut microorganisms of high-risk antibiotic resistance genes at the phylum level; when the phylum of phylum microorganisms of the high-risk antibiotic resistance gene includes Proteobacteria or Actinobacteria, and the phylum of gut microorganisms of the high-risk antibiotic resistance gene includes Firmicutes or Actinobacteria, it is determined that the target wild animal gut has a high risk of acquiring the high-risk antibiotic resistance gene. Therefore, based on the specific high-risk gene transmission pathway of "phylum-derived Proteobacteria / Actinobacteria host - gut Firmicutes / Actinobacteria receptor," by determining whether phylum microorganisms and gut microorganisms belong to these two phyla respectively, the risk of wild animal gut microorganisms acquiring antibiotic resistance genes from phylum microorganisms of phytophagous plants under human interference can be accurately assessed.

[0027] According to an embodiment of this application, the predetermined risk classification criteria include at least the following conditions: (1) the abundance of the antibiotic resistance gene in the metagenomic data of plant phyllosphere microorganisms reaches a preset threshold; (2) the gene sequence of the antibiotic resistance gene is associated with mobile genetic elements (MGEs) upstream and / or downstream; (3) the plant phyllosphere microorganism carrying the antibiotic resistance gene carries virulence factors; wherein, antibiotic resistance genes that simultaneously meet conditions (1), (2), and (3) are identified as having the highest risk level and are used for the analysis.

[0028] In some embodiments, the preset threshold for abundance is a total number of contigs ≥ 100.

[0029] According to embodiments of this application, the subtypes of antibiotic resistance genes with the highest risk level include at least one of the following: CMH, MfpA, PDC, PmrA, and RphA. These five antibiotic resistance gene subtypes are present in areas affected by human interference but not in areas unaffected by human interference; this characteristic indicates that they can serve as specific biomarkers indicating human activity interference. Therefore, in risk assessment, detecting and analyzing these five specific subtypes can significantly improve the accuracy and specificity of assessing the acquired risks to wild animals.

[0030] According to an embodiment of this application, the method further includes: identifying differentially expressed enzyme genes related to central carbon metabolism in the gut microbiome metagenomic data; identifying differentially abundant metabolites related to central carbon metabolism based on non-targeted metabolomics detection of the fecal sample; mapping the differentially expressed enzyme genes and the differentially abundant metabolites together to the carbon metabolism reference map of the KEGG pathway database, and determining whether high-risk antibiotic resistance genes cause abnormalities in gut microbiome sugar metabolism-related pathways and fecal metabolites.

[0031] By locating "gene expression changes" and "metabolite abundance changes" within the metabolic pathway of "central carbon metabolism," we can accurately determine whether key nodes in microbial energy metabolism (such as glycolysis and the tricarboxylic acid cycle) are interfered with by high-risk antibiotic resistance genes, thereby accurately determining whether the presence of high-risk resistance genes has led to an imbalance in microbial ecosystem function.

[0032] According to embodiments of this application, the differentially expressed enzymes include at least one of the following: phosphoglycerate kinase, isocitrate dehydrogenase, malate kinase, and citrate synthase; the differentially abundant metabolites include at least one of the following: α-ketoglutarate, isocitrate, sedoheptulose-7-phosphate, 6-phosphogluconic acid, and NADH. The aforementioned differentially expressed enzymes and differentially abundant metabolites show significant differences between human-interferenced and non-human-interferenced regions. Based on these enzymes and metabolites, it can be determined whether high-risk antibiotic resistance genes lead to abnormalities in gut microbial glucose metabolism pathways and fecal metabolites.

[0033] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0034] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0035] Figure 1 This study presents a framework for assessing the genetic risk of antibiotic resistance in wildlife health caused by human habitat disturbance, including (a) a risk assessment framework for antibiotic resistance genes (ARGs) based on multi-omics data; and (b) key processes by which ARGs are transferred from plant phyllosphere microbes to wildlife gut microbes through the food chain.

[0036] Figure 2 This diagram illustrates the composition and high-risk characteristics of antibiotic resistance genes in phyllosphere microorganisms remodeled by human interference; (a) the class composition, quantity, and relative abundance of all antibiotic resistance genes (ARGs) in phyllosphere samples; (b) the transmission trend and risk level distribution of ARGs in samples from different sources; (c) the differences in subclass composition and quantity of Rank I ARGs in the PCK and PHD groups, as well as the ARG subclasses specific to the PHD group; and (d) the differences in the composition of predicted host bacteria of Rank I ARGs at the phylum level (p<0.05).

[0037] Figure 3 The study showed the availability of ARGs from gut microbiota to phylum microorganisms; (a) the difference in collinear acquisition ability between the CK and HD groups at different phyla; (b) the compositional differences of collinear genes at the KEGG Level 2 functional annotation level (comparison between CK and HD); (c, d) the compositional differences of collinear ARGs in the CK and HD groups at the antibiotic resistance type category level, respectively, and reflected their differential characteristics from the perspective of the host microbiota structure of gut microbiota ARGs.

[0038] Figure 4The effects of HD on gut microbiota function and gut metabolism in wild animals were demonstrated.

[0039] Figure 5 A structural equation model of plant input-driven nutrient regulation and its cascade effects on bacterial community structure and resistance-related genetic factors is presented; (a) a correlation heatmap shows the Pearson correlation coefficients (r) among all variables; (b) the model demonstrates the hypothetical structural relationships between plant inputs, nutrient status, bacterial community diversity, mobile genetic elements (MGEs), antibiotic resistance genes (ARGs), and CAP. In the figure, A–H represent: reducing sugars, amino acids, water, bacterial community diversity, mobile genetic elements (MGEs), antibiotic resistance genes (ARGs), cephalexin (CAP), and relative plant abundance, respectively. Detailed Implementation

[0040] The following will explain the solution of this application with reference to embodiments. Those skilled in the art will understand that the following embodiments are for illustrative purposes only and should not be considered as limiting the scope of this application. Where specific techniques or conditions are not specified in the embodiments, they are performed according to the techniques or conditions described in the literature in the art or according to the product instructions. Reagents or instruments whose manufacturers are not specified are all conventional products that can be obtained commercially.

[0041] Example 1

[0042] 1 Experimental Methods

[0043] 1.1 Research Subjects and Sample Collection

[0044] Baihe National Nature Reserve (104°01'–104°12′ E, 33°10′–33°22′ N) is one of the important core habitats of the Sichuan golden snub-nosed monkey (Rhinopithecus roxellana), and also an area with high population size and density. To assess the impact of human disturbance on the Sichuan golden snub-nosed monkey and its food source microorganisms, two monkey groups with significantly different ecological conditions were selected: a group inhabiting the Xiapingdi area, which was more heavily affected by human activities, was selected as the human disturbance group (MHD), and a group inhabiting a remote mountainous area inaccessible to humans was selected as the wild control group (MCK). Fecal and food plant samples were collected simultaneously in the respective habitats of the two monkey groups. Figure 1 ).

[0045] Fifteen fresh fecal samples were randomly collected from each of the MHD and MCK groups at 9:00 and 17:00 daily, immediately after defecation, to minimize the impact of environmental exposure. All fecal samples were transported in dry ice for subsequent 16S rRNA gene sequencing and metagenomic sequencing.

[0046] Plant samples were collected during the main feeding period of Sichuan golden monkeys (8:00–10:00) by tracking their feeding behavior. Young leaves and branches of the corresponding plants were selected, and species identification was performed (Table 1). Three–5 leaves were collected from each sampling point and aggregated into a single biological replicate. Samples from the human interference area were defined as the plant-human interference group (PHD), and samples from the wild area were defined as the plant-wild control group (PCK). Plant samples were transported in dry ice after collection for metagenomic sequencing.

[0047] Table 1. Types of plant samples

[0048]

[0049] 1.2 Binning Analysis of Golden Monkeys' Preference for Plants

[0050] Plant samples were transported to Shanghai Meiji Biotechnology Co., Ltd. on dry ice for metagenomic sequencing. The metagenomic sequencing method was the same as in our previous studies. The data has been stored in the NCBI Sequence Reading Archive (SRA) database, accession number PRJNA947945. During metagenomic analysis, we used open reading frames (ORFs) to predict non-redundant gene contigs and compared the ORFs with the Comprehensive Antibiotic Research Database (CARD, http: / / arpcard.mcmaster.ca; blastp, e-value: 1e-5) to obtain ARG information. Virulence factors (VFs) were identified using the Virulence Factor Database (VFDB, setB: http: / / www.mgc.ac.cn / VFs / ; blastp, e-value: 1e-5). The MGEs database includes plasmids (https: / / ftp.ncbi.nlm.nih.gov / refseq / release / plasmid), ICEberg (https: / / bioinfo-mml.sjtu.edu.cn / ICEberg2 / index.php), Integrall (http: / / integrall.bio.ua.pt / ), and ISfinder (https: / / www-is.biotoul.fr) for identifying MGEs (blastp, e-value: 1e-5). The NR database (https: / / www.ncbi.nlm.nih.gov; blastp, e-value: 1e-5) is used to obtain microbial species. The NT database (https: / / www.ncbi.nlm.nih.gov; blastp, e-value: 1e-5) is used to obtain plant information.

[0051] 1.3 Metagenomics Technology for the Gut Microbiome of Golden Monkeys

[0052] Metagenomic analysis was performed on fecal samples from Sichuan golden snub-nosed monkeys. Fecal samples from the MHD and MCK groups were frozen on dry ice and sent to Shanghai Meiji Biotechnology Co., Ltd. for metagenomic sequencing. The obtained sequences were assembled, and contiguous groups were subjected to binning analysis to reconstruct the bacterial metagenomic assembled genomes (MAGs). The raw sequencing data were stored in the NCBI Sequence Reading Archive (SRA) database, accession number PRJNA1331459. We selected contiguous groups corresponding to high-quality MAGs (integrity ≥80% and contamination ≤10%) for reassembly and gene prediction. The predicted open reading frames (ORFs) were aligned with the Comprehensive Antibiotic Resistance Database (CARD; http: / / arpcard.mcmaster.ca; blastp, e-value:1e-5) to annotate antibiotic resistance genes (ARGs). Subsequently, pathway annotation was performed using the KEGG database (https: / / www.genome.jp / kegg; blastp, e-value:1e-5) for subsequent analysis.

[0053] 1.4 Risk assessment of the spread of phyllosphere microorganisms ARGs

[0054] A two-step assessment framework was employed to evaluate the ecological risk of antibiotic resistance genes (ARGs) from phyllosphere microbes spreading into the gut of Sichuan golden snub-nosed monkeys. Figure 2(a) The first step is risk level identification: ARGs in the leaf microbiome are classified according to three criteria, namely (1) high abundance (total contigs ≥ 100); (2) carrying mobile genetic elements; and (3) their host microorganisms carry virulence factors. Accordingly, ARGs are divided into four levels: Rank I (highest risk), which meets all three criteria; Rank II (high risk), which meets criteria (1) and (2); Rank III (medium risk), which meets only criterion (1); and Rank IV (low risk), which does not meet criterion (1). The second step is host availability assessment: In order to assess the potential of Rank I ARGs to be acquired by the gut microbiome, regions in gut MAGs that have high similarity (identity ≥ 80%) to leaf microbiome Rank I ARGs are first screened by homology-based sequence alignment (BLASTn, e-value ≤ 1e-5). Building upon this, Nucmer (v4.0) was used to perform genome-wide collinearity alignment of homologous regions, identifying and screening high-confidence collinearity events longer than 50 bp located in contiguous blocks. These collinear blocks indicate high structural compatibility between the relevant ARGs and intestinal MAGs, suggesting their potential for cross-host migration. Therefore, if an intestinal MAG exhibits the aforementioned collinearity events with a specific Rank I ARG, the MAG is considered to have the potential to acquire that ARG. To quantify this potential, a Colinear Acquisition Potential (CAP) index was defined, calculated as follows:

[0055]

[0056] Where M represents the total number of MAG species detected to be collinear with the Rank I ARG, Ci represents the number of ARGs in the i-th MAG that are collinear with the ARG of that species, and Ai represents the total relative abundance of the i-th MAG in the gut microbiota. A larger CAP value indicates a wider host range for antibiotic resistance genes to be integrated into, a stronger ability to be maintained in the host, and a higher potential for amplification in the community, thus increasing the risk of their spread and colonization.

[0057] 1.5 Non-targeted metabolomics analysis of golden monkey feces by LC-MS

[0058] Solid samples (100 mg) were added to extraction buffer (methanol / water = 4:1, including internal standard), followed by grinding (-10℃, 6 min) and low-temperature sonication (5℃, 30 min) to extract metabolites. Samples were incubated at -20℃ and centrifuged, and the supernatant was used for LC-MS / MS analysis. Raw data underwent peak extraction, alignment, and metabolite annotation using Progenesis QI (HMDB, Metlin, and a self-built library). Data preprocessing was then performed on the Meiji Cloud platform, including 80% rule-based removal of missing values, minimum value imputation, total normalization, removal of variables with RSD > 30% in QC, and log10 transformation. Preprocessed data were analyzed using the R package ropls for PCA and OPLS-DA, with seven cross-validations used to assess model stability. Metabolites with significant differences were screened based on VIP > 1 and p < 0.05. Finally, pathway annotation was performed using KEGG, and pathway enrichment analysis was completed using Python scipy.stats based on Fisher's exact test.

[0059] 1.6 Effects of phyllosphere microbial ARGs on gut ARGs under HD environment

[0060] To assess the impact of high-risk antibiotic resistance genes (Rank I ARGs) from the leaf circle under human disturbance on the gut microbiota function of Sichuan golden snub-nosed monkeys, we screened relevant gut MAGs based on homology analysis for multi-level functional analysis. First, we performed KEGG homology annotation on the MAG-encoding genes, constructing a non-redundant gene-function matrix. At the Level 2 level, Fisher's exact test and log2 Fold Change were used to calculate functional differences, and significant functional entries were selected after FDR correction. Second, we conducted cross-omics integration analysis, co-mapping differentially expressed enzyme genes related to carbon metabolism with differentially expressed metabolites identified from non-targeted metabolomics. Fecal metabolites were detected by LC-MS / MS, and OPLS-DA modeling was performed. The model stability was assessed through 7 cross-validations and 200 permutation tests. The selection criteria for differentially expressed metabolites were VIP>1 and p<0.05. Finally, we jointly mapped differentially expressed enzymes and metabolites to KEGG pathways, constructing a "microbial functional gene-host metabolite" association network. Furthermore, we calculated the Pearson correlation coefficient (r) to quantify the linear relationship between environmental factors, microbial characteristics, and resistance group indicators. Figure 1 Figure b illustrates the key process by which ARGs are transferred from plant phytosphere microbes to wild animal gut microbes through the food chain, and also demonstrates the digestive and absorption mechanism of food through the synergistic interaction between the gut microbiota and the host gut. This figure integrates and showcases the proposed genetic risk assessment system for drug resistance and the multi-omics joint analysis approach from multiple dimensions.

[0061] 1.7 Statistical Analysis Methods

[0062] The Wilcoxon signed-rank test was used to analyze the differences between the CK and HD groups, with a significance level of p < 0.05. R 4.4.1 was used for Wilcoxon signed-rank test and ANOVA. Basic data processing was performed using Excel 2024, and OriginPro 2024 was used to create bar charts, bubble charts, Sankey plots, and heatmaps. The VennDiagram package in R 4.4.1 was used to create Venn plots, and the tidyverse package was used to process the data and create relative abundance and quantity facets of ARGs.

[0063] 2 Results

[0064] 2.1 HD increases the ecological risk of ARGs in food to golden snub-nosed monkeys.

[0065] Metagenomic sequencing data from plant samples showed that HD, overall, increased the number and relative abundance of chlorophyll microbial resistance genes. Figure 2 (a). All ARGs can be divided into 21 major categories. Except for Diaminopyrimidine and Triclosan, the abundance and number of the other ARG categories in the PHD group were significantly higher than those in the PCK group. Figure 2 (b)

[0066] In addition, the omics-based risk assessment framework revealed differences in the four risk levels of ARGs between the PCK and PHD groups at the class level. Focusing on high-risk Rank I ARGs, it was found that although there were no significant differences in the relative abundance of ARG subtypes, the PHD group introduced five ARG subtypes not present in the PCK group, including CMH, MfpA, PDC, PmrA, and RphA. Figure 2 (c). Subsequently, the analysis of host bacterial species in the NR database of Rank I ARGs showed that Acidobacteria, Actinobacteria, and Firmicutes were significantly higher in the PHD group than in the PCK group, while there were no significant differences in the other categories. Figure 2 (d).

[0067] Version 2.2 HD increased the availability of Rank I ARGs in food for Sichuan golden snub-nosed monkeys.

[0068] Overall, the structural features and homology analysis based on MAGs showed that the number of collinear events and the collinear acquisition potential index detected in the HD environment were significantly higher than those in the control group, indicating an enhanced potential structural association between plant-derived ARGs and gut microbiota. Furthermore, host classification results based on the NR database showed that host microorganisms capable of acquiring ARGs were mainly distributed across nine phyla. Among these, the comparison of phylum-level collinear acquisition potential showed that Acidobacteria, Actinobacteria, Firmicutes, and Other / Unclassified were significantly higher in the MHD group than in the MCK group (p<0.05). During the collinear flow of ARGs, the main source phyla of ARGs in plant samples were Proteobacteria and Actinobacteria, while the main accepting phyla of ARGs in gut samples were Firmicutes and Actinobacteria. This distribution characteristic remained consistent in both the CK and HD groups. Figure 3 (c and d).

[0069] At the KEGGLevel2 functional annotation level, the overall pathway abundance in the MHD group was higher than that in the MCK group. Figure 3 (b). Further comparison of 22 differentially expressed pathways revealed that the abundance of pathways such as Replication and Repair, Translation, Signal Transduction, Metabolism of Cofactors and Vitamins, and Membrane Transport was significantly higher in the MHD group than in the MCK group. Figure 3 (b) Collinearity analysis of ARGs detected 15 ARGs collinear in the gut microbiota, of which Elfamycin was detected only in the MHD group, while the remaining 14 ARGs were detected in both groups. Figure 3 (c and d).

[0070] 2.3 HD leads to changes in the metabolic structure of gut microbiota

[0071] Figure 4 This study demonstrates the effects of hemolytic hyperplasia (HD) on gut microbiome function and metabolism in wild animals. Carbon metabolism-related enzymes were mapped onto a KEGG pathway map to illustrate the changes in gut microbiota in two groups under HD interference. Circular markers indicate changes in gut metabolites, while square markers indicate changes in gut microbiome enzymes. The central carbon metabolism pathway comprises glycolysis, the tricarboxylic acid cycle, the pentose phosphate pathway, and oxidative phosphorylation.

[0072] KEGG enzyme pathway annotation based on homologous ARGs revealed that HD interference affected the carbon metabolism-related functions of the gut microbiota in Sichuan golden snub-nosed monkeys. Among the five key enzymes involved, four—Phosphoglycerate kinase, Isocitrated dehydrogenase, Malic enzyme, and Citrate synthase—were upregulated in the MHD group. Metabolomics data analysis showed that, compared to the MCK group, the MHD group exhibited decreased levels of α-Ketoglutarate, Isocitrate, and Sedoheptulose-7-phosphate, while increasing levels of 6-Phosphogluconate and NADH. The trends in α-Ketoglutarate and NADH were consistent with the changes in their corresponding enzyme, Isocitrate dehydrogenase.

[0073] 2.4 Plant Input-Driven Nutrition-Microbe-Resistance Cascade

[0074] To characterize the potential causal relationships among plant inputs, nutrient status, bacterial community structure, and genetic factors, a structural equation model (SEM) was constructed that includes plant-related inputs, microbial community structure, MGEs, ARGs, and homology relationships. Figure 5 A structural equation model (SEM) of plant input-driven nutrient regulation and its cascade effects on bacterial community structure and resistance-related genetic factors was presented. The model characterizes the potential causal relationships among plant input, nutrient status, bacterial community diversity, mobile genetic elements (MGEs), antibiotic resistance genes (ARGs), and collinear acquisition potential (CAP). Path coefficients are standardized estimates (β), with solid lines representing statistically significant paths (p < 0.05) and dashed lines representing non-statistical paths. Correlation heatmaps show the Pearson correlation coefficients (r) between variables. The overall correlation structure is highly consistent with SEM paths, supporting the ecological process framework revealed by the model. The overall model fit is moderate (χ²). 2 =21.80, df=14, p=0.083; CFI=0.701; RMSEA=0.136), but the core causal structure was stable and reliable. The results showed that plant-related factors had a significant positive impact on nutrient status (β=0.648, p=0.007), and nutrient status further significantly promoted bacterial community diversity (β=0.680, p=0.019), constituting the key transmission pathway of the model ( Figure 5 (a)

[0075] Correlation analysis showed a high degree of consistency with SEM results. Plant richness was significantly positively correlated with RS (r=0.307), AA (r=0.506), and moisture (r=0.448), with AA showing the strongest correlation with moisture (r=0.509). Moisture was significantly positively correlated with bacterial community diversity (r=0.507), while the correlation between plant richness and bacterial community diversity was weaker (r=0.211), supporting the key mediating role of nutrient status between plant input and microbial structure. At the genetic level, plant correlation indices were significantly positively correlated with collinear acquisition potential (CAP) (r=0.422), and RS was significantly positively correlated with ARGs (r=0.360). Furthermore, the correlation directions of bacterial community diversity, MGEs, ARGs, and collinear acquisition potential (CAP) were consistent with the cascade relationship set by SEM, further validating the model's reasonable characterization of ecological processes. Figure 5 (b)

[0076] Human interference promotes a significant enrichment of antibiotic resistance genes (ARGs) in phytosphere microbes. These ARGs can enter the gut of wild animals via the food chain and further integrate into the gut microbiome through horizontal gene transfer. This process alters the functional structure of the wild animal gut microbiome, thereby affecting gut metabolic function at the overall level, with particularly pronounced damage to central carbon metabolism. Previous studies, consistent with our findings, have shown that phytotoxicity (HD) significantly increases the overall ARG load in the phytosphere microbiota of wild animals' preferred plants, thus harming the health of individual wild animals and their populations.

[0077] The above results indicate that HD does not merely alter the abundance of environmental resistance genes, but rather exerts a systemic impact on wildlife health risks through the "environment-host-microbe-metabolism" pathway. HD may have a comprehensive effect on the wildlife-related resistance gene pool by altering the composition and gene structure characteristics of environmental microorganisms. Unlike previous studies that mainly focused on changes in the abundance and diversity of ARGs, this application further observed the differential characteristics of ARG risk structures. Although the overall abundance and diversity of ARGs in the PHD group were significantly higher than those in the PCK group, the relative abundance of common subtypes of Rank I ARGs did not show a significant difference, suggesting that the impact of HD on the resistance gene pool may be more reflected in structural adjustments rather than simple quantitative accumulation. This structural change is mainly manifested in the introduction of new subtypes and their enhanced potential for transmission.

[0078] Homology analysis further revealed that under high-disturbance (HD) conditions, some ARGs may exhibit higher potential mobility due to changes in host distribution or enhanced association with mobile genetic elements (MGEs). Although core ARG subtypes remained relatively stable in abundance, reflecting their niche conservation, their genetic background and dispersal potential may differ under different disturbance scenarios. This characteristic is consistent with reports in other ecosystems of a small number of high-risk ARGs having strong dispersal capabilities, suggesting that when assessing the potential ecological risks of ARGs, both their host spectrum and genetic background characteristics should be considered. The difference in ARG host composition between the leaf sphere and gut microbiota further reflects the complex pattern of their cross-niche dispersal: although the two niches share some ARG types, the main host phyla are significantly different, with Proteobacteria and Actinobacteria dominating in the leaf sphere, while Firmicutes and Actinobacteria are dominant in the gut. This host replacement phenomenon is consistent with the view that ARGs mainly rely on MGEs for movement between microbial communities. Under HD conditions, the relative abundance of Proteobacteria in the leaf sac increases, which may reflect the influence of changes in exogenous input or selection pressure on the ARG diffusion process, indicating that different ecological niches may play different functional roles in the ARGs dissemination network.

[0079] In the gut ecosystem, changes in ARGs are significantly associated with enhanced functions of multiple carbon metabolism pathways, involving core metabolic processes such as glycolysis (EMP), the pentose phosphate pathway (PPP), and the tricarboxylic acid cycle (TCA). Our findings are consistent with animal model studies of antibiotic exposure-induced disturbances in the gut metabolic network, suggesting that ARG colonization may be related to adjustments in the functional structure of the gut microbiota. However, whether this process directly confers an ecological competitive advantage to drug-resistant bacteria requires further experimental verification. Furthermore, the observed incomplete synchronization between the metabolome and microbiome indicates that host metabolic regulation may play a buffering role in responding to changes in microbial function, suggesting that host metabolic regulation may be crucial in maintaining gut homeostasis.

[0080] Structural equation modeling (SEM) and correlation analysis revealed that plant inputs indirectly correlated with the distribution patterns of resistance groups (ARGs) by influencing nutrient status and bacterial community structure, suggesting that changes in resistance groups may be embedded in a plant-nutrient-microbe coupling network. However, the relationships reflected in this model are still at the statistical correlation level, and their specific mechanisms of action require further verification.

[0081] This application systematically analyzes the changes in antibiotic resistance genes, microbial community structure, and metabolic characteristics in the Sichuan golden snub-nosed monkey ecosystem under human disturbance. The results show that human disturbance is significantly associated with changes in the composition and structure of ARGs (antibiotic resistance genes) in the leaf sphere and gut microbiota, host distribution characteristics, and the function of multiple core carbon metabolism pathways. Differences in ARG host composition between the leaf sphere and gut further support the view that ARGs mainly spread between different ecological niches through gene flow rather than overall bacterial migration. Simultaneously, partial decoupling between the metabolome and microbiome suggests that host metabolic regulation may, to some extent, buffer the ecological impacts of altered microbial function. Furthermore, plant input regulates bacterial community structure through nutrient status as a key mediator, and is further correlated with the distribution of resistance genetic factors, reflecting a potentially stable synergistic relationship between environmental input, microbial structure, and resistance evolution. Overall, this application provides a multi-omics evidence base for understanding the ecological processes of resistance genes in wild animals under human disturbance conditions and offers a scientific reference for conducting comprehensive assessments of resistance risks within the "One Health" framework.

[0082] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for assessing the acquired risk of antibiotic resistance genes in the gut of wild animals, characterized in that, Includes the following steps: Leaf samples and fecal samples of the preferred plants of the target wild animals were collected from areas of human disturbance within the target wild animal habitat. Metagenomic sequencing was performed on the plant leaf circle microorganisms in the leaf samples of the edible plants to obtain plant leaf circle microbial metagenomic data; metagenomic sequencing was performed on the fecal samples to obtain gut microbial metagenomic data, and the metagenomic genome was reconstructed and assembled based on the gut microbial metagenomic data; Based on the metagenomic data of plant chloroplast microorganisms, antibiotic resistance genes with high risk levels were identified according to predetermined risk grading criteria. To analyze the cross-niche transmission potential of the high-risk antibiotic resistance genes from the plant leaf zone to the wild animal gut; The analysis includes: determining the high-risk antibiotic resistance genes in the phylum phylum phylum phylum phylum and the phylum gut microbiome based on the metagenomic data of the plant phylum phylum phylum and the metagenomic data of the gut microbiome, respectively; If the host microbial phylum is Proteobacteria or Actinobacteria, and the recipient microbial phylum is Firmicutes or Actinobacteria, then the target wild animal's gut is determined to have a high risk of acquiring the high-risk antibiotic resistance gene.

2. The method according to claim 1, characterized in that, The target wild animal is a semi-leaf-eating non-human primate.

3. The method according to claim 1, characterized in that, The target wild animal is the Sichuan golden snub-nosed monkey.

4. The method according to claim 1, characterized in that, The analysis further includes: Homology comparison analysis was performed between the high-risk antibiotic resistance gene sequence and the sequence of the metagenomic assembled genome to screen out homologous regions with sequence similarity reaching a preset threshold. Whole-genome collinearity alignment was performed on the homologous regions to identify collinear conserved blocks with a length greater than a preset value; Based on the collinear conserved blocks, and combined with the relative abundance of the corresponding metagenomic assembled genome in the gut microbiota, an evaluation index is calculated to determine the cross-niche transmission potential of the high-risk antibiotic resistance genes from the plant foliage to the wild animal gut.

5. The method according to claim 4, characterized in that, The evaluation index is calculated using the following formula: Wherein, CAP is the evaluation index; M represents the total number of metagenomic assembly genomes that share collinear conserved blocks with the high-risk antibiotic resistance genes. Ci represents the number of collinear conserved blocks in the i-th metagenomically assembled genome that form a collinearity with the specific high-risk antibiotic resistance gene; Ai represents the total relative abundance of the i-th metagenomic assembly genome in the gut microbiota.

6. The method according to claim 4, characterized in that, Further includes: Based on the homology comparison analysis and whole-genome collinearity comparison analysis results, and combined with the taxonomic annotation information of the plant phylum microbial metagenomic data and the gut microbial metagenomic data, the distribution characteristics of phylum microorganisms and gut microorganisms with high-risk antibiotic resistance genes were determined at the phylum level. If the phylum of the high-risk antibiotic resistance gene is Proteobacteria or Actinobacteria, and the phylum of the gut microbiota of the high-risk antibiotic resistance gene is Firmicutes or Actinobacteria, then the gut microbiota of the target wild animal is deemed to have a high risk of acquiring the high-risk antibiotic resistance gene.

7. The method according to claim 1, characterized in that, The predetermined risk classification criteria include at least the following conditions: (1) The abundance of the antibiotic resistance gene in the metagenomic data of plant phyllosphere microorganisms reaches a preset threshold; (2) The gene sequence of the antibiotic resistance gene is associated with mobile genetic elements upstream and / or downstream; (3) Plant phyllosphere microorganisms carrying the antibiotic resistance gene carry virulence factors; Among them, antibiotic resistance genes that simultaneously meet conditions (1), (2), and (3) are identified as having the highest risk level and are used for the analysis.

8. The method according to claim 7, characterized in that, The subtypes of the antibiotic resistance genes with the highest risk level include at least one of the following: CMH, MfpA, PDC, PmrA, and RphA.

9. The method according to claim 1, characterized in that, Further includes: Identify differentially expressed enzyme genes related to central carbon metabolism in the gut microbiome metagenomic data; Based on non-targeted metabolomics detection of the fecal samples, differentially abundant metabolites related to central carbon metabolism were identified. The differentially expressed enzyme genes and the differentially abundant metabolites are jointly mapped to the carbon metabolism reference map of the KEGG pathway database to determine whether the high-risk antibiotic resistance genes cause abnormalities in gut microbial sugar metabolism-related pathways and fecal metabolites.

10. The method according to claim 9, characterized in that, The differentially expressed enzymes include at least one of the following: phosphoglycerate kinase, isocitrate dehydrogenase, malic acid kinase, and citrate synthase; The differential abundance metabolites include at least one of the following: α-ketoglutarate, isocitrate, sedoheptulose-7-phosphate, 6-phosphogluconic acid, and NADH.

Images