Next Article in Journal
ID-APM: Inverse Disparity-Guided Annealing Point Matching Approach for Robust ROI Localization in Blurred Thermal Images of Sika Deer
Previous Article in Journal
Simulation and Experimental Study on the Optimization of Operating Parameters for Coating Pellets of Agropyron Seeds
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Agriculture
Volume 15
Issue 19
10.3390/agriculture15192015
agriculture-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Pesticide Type Distinctly Shapes Soil Resistomes: A Comparative Analysis of Antibiotic and Non-Antibiotic Agro-Chemicals

by
Lilan Lyu
1,
Qinyu Lu
1,
Chanchan Huang
1,
Xiyu Zhang
1,
Jinjie Yao
1,
Huaxian Zhao
2 and
Chengwu Zou
3,*
1
Guangxi Subtropical Crops Research Institute, Laboratory of Quality Risk Assessment for Agro-Products of Ministry of Agriculture and Rural Affairs (Nanning), Key Laboratory of Quality and Safety Control for Subtropical Fruit and Vegetable, Ministry of Agriculture and Rural Affairs, Quality and Testing Center of Subtropical Fruit and Vegetable of Ministry of Agriculture and Rural Affairs, Nanning 530001, China
2
Key Laboratory of Ministry of Education for Environment Change and Resources Use in Beibu Gulf, Guangxi Key Laboratory of Earth Surface Processes and Intelligent Simulation, Nanning Normal University, Nanning 530001, China
3
College of Agriculture, Guangxi University, Nanning 530001, China
*
Author to whom correspondence should be addressed.
Agriculture 2025, 15(19), 2015; https://doi.org/10.3390/agriculture15192015
Submission received: 6 August 2025 / Revised: 10 September 2025 / Accepted: 24 September 2025 / Published: 26 September 2025
(This article belongs to the Section Agricultural Soils)
Browse Figures
Versions Notes

Abstract

Agricultural pesticides are significant drivers of antibiotic resistance in soil. However, the differential impacts of antibiotic versus non-antibiotic pesticides on the soil resistome are poorly characterized. Here, we analyzed sequencing data from soils exposed to either antibiotic or non-antibiotic pesticides to compare differences in antibiotic resistance gene (ARG) burden, diversity, assembly processes, network topology, and host taxonomy. Soils exposed to antibiotic pesticides exhibited a significantly higher ARG burden (0.52% vs. 0.27% of total genes), whereas soils exposed to non-antibiotic pesticides showed significantly higher alpha diversity (p < 0.05). ARG community compositions also differed significantly between antibiotic and non-antibiotic exposures (PERMANOVA, R2 = 0.215, p < 0.001). Assembly analysis using the modified stochasticity ratio indicated that deterministic processes governed ARG community assembly in both groups, with stronger influence observed in non-antibiotic soils. Co-occurrence network analysis revealed contrasting patterns. A compact, highly centralized network emerged in antibiotic-exposed soils, while a larger, more dispersed network characterized non-antibiotic soils. In both networks, aminoglycoside ARGs served as keystone nodes, accompanied by the β-lactam ARG in antibiotic soils and the macrolide ARG in non-antibiotic soils. Pseudomonadota was the predominant ARG host (>60% contribution) across both exposures, though many other phyla exhibited significance (p < 0.05) between group differences in their ARG contributions. Non-pathogenic bacteria comprised the majority of ARG hosts in all samples. When examining ARG contributions from pathogenic hosts, zoonotic and animal-associated pathogens contributed significantly (p < 0.01) more in non-antibiotic soils than in antibiotic soils, whereas the ARG contribution from plant pathogens was comparable between the two pesticide groups. Overall, our study suggests that antibiotic and non-antibiotic pesticides shape distinct ARG network patterns and host–pathogen profiles, posing distinct risks to public health and agricultural ecosystems.

1. Introduction

Antibiotic resistance has rapidly escalated into one of the most urgent global health crises of the 21st century. According to a 2024 analysis published in The Lancet [1], an estimation of 4.71 million (95% UI: 4.23–5.19 million) deaths were associated with bacterial antimicrobial resistance (AMR) in 2021, including 1.14 million (1.00–1.28 million) deaths directly attributable to AMR. By 2050, forecasts indicate that AMR could cause 1.91 million attributable deaths and 8.22 million associated deaths worldwide [1]. At the core of antimicrobial resistance are antibiotic resistance genes (ARGs), which encode the molecular mechanisms that enable bacteria to survive antibiotic pressure [1,2]. These genes are not confined to clinical settings but are widely distributed across diverse environmental compartments, including water, sediments, and especially soils [3,4]. Among them, soils are considered one of the largest and most diverse reservoirs of ARGs, harboring both ancient and emergent resistance elements [4]. Through horizontal gene transfer (HGT), ARGs in the environment can be mobilized into human-associated bacteria, contributing to the emergence of multidrug-resistant pathogens [5]. Therefore, the soil resistome plays a pivotal role in the environmental dissemination of antibiotic resistance.
Agricultural soils, shaped by decades of intensive human use, have become dynamic interfaces between anthropogenic activity and environmental microbiomes [6,7,8]. Among the most pervasive agricultural inputs, pesticides, including but not limited to herbicides, insecticides, fungicides, and other agrochemicals, fundamentally alter soil microbial communities [6,7,8]. While their primary role is to target plant and animal pests, these chemicals inevitably impact non-target soil microbiota, reshaping microbial community structure and function, including ARG composition and transfer [6,7,8,9,10]. Recent studies show that pesticides promote the enrichment of antibiotic resistance genes via multiple mechanisms [8,9,11,12,13,14]. For instance, Liao et al. [9] demonstrated that exposure to three widely used herbicides (glyphosate, glufosinate, and dicamba) significantly increased the prevalence of ARGs in both controlled microcosm experiments and agricultural soils across 11 provinces in China. They further suggest that herbicide exposure may elevate bacterial mutation rates and drive cross-selection, thereby enhancing tolerance to both herbicides and antibiotics [9]. Pesticide exposure has also been shown to promote the horizontal transfer of ARGs [8,9,11]. Zhang et al. [8] reported that the chiral herbicide R-flurtamone significantly enhances ARG conjugation by increasing membrane permeability, inducing excessive reactive oxygen species and SOS responses, and boosting intracellular ATP levels, collectively accelerating ARG propagation.
Additionally, some pesticides targeting bacterial plant pathogens are themselves antibiotics. Worldwide, more than 4000 antimicrobial pesticide products are registered for crop protection [15,16]. Plant production routinely employs three antibiotic classes (aminoglycosides, tetracyclines, and quinolones) that mirror those used in human and veterinary medicine [15,17]. Beyond these, a variety of other antibiotic compounds are applied sporadically or experimentally to manage specific phytopathogens [15]. This direct use of antibiotic-based pesticides creates highly selective environments in agricultural soil, enriching target-specific ARGs and fostering co-selection of linked resistance determinants, ultimately exacerbating the spread of multidrug resistance [18,19,20,21]. For instance, oxytetracycline (a tetracycline class antibiotic commonly used against fire blight) can substantially elevate soil tet gene abundances (tetA, tetL, tetM, tetQ, tetW) by 10- to 1000-fold, even at low exposure levels (2–70 µg/kg), via manure amendment, highlighting its potent selective impact on the soil resistome [20]. This enrichment of ARGs in agricultural soils poses tangible risks, as these genes can be transferred to crops and microorganisms on fresh produce [15,22,23]. A growing number of cross-sectional studies worldwide have detected clinically relevant ARGs and resistant pathogens on fruits and vegetables, while epidemiological evidence links the consumption of raw produce to colonization of the human gut with antimicrobial-resistant bacteria [15,22,23]. Thus, pesticide-influenced soil resistomes in agricultural settings directly threaten ecological and human health, posing a critical challenge to the One Health framework through ARG spread.
Despite clear evidence that both antibiotic pesticides and non-antibiotic pesticides drive ARG proliferation in soils, their relative contributions and distinct mechanisms remain poorly defined. Therefore, a systematic comparison of these two pesticide classes is urgently needed to clarify their effects on ARG and their hosts. Addressing this knowledge gap is critical to inform targeted interventions that mitigate agricultural drivers of AMR within a One Health framework. Hence, in this study, we compiled bacterial sequencing data from agricultural soils exposed to antibiotic and non-antibiotic pesticides to investigate ARG distribution and microbial composition. We then assessed the differential effects of antibiotic versus non-antibiotic pesticides on (1) differences in ARG prevalence, diversity and community assembly processes, (2) pesticide-specific indicator ARGs, and (3) the primary host taxa of these resistance genes.

2. Materials and Methods

2.1. Data Collection and Compilation

To explore the distribution characteristics and variation patterns of ARGs in soil environments under different pesticides, the following search terms were used: “agricultural soil”, “pesticide”, “antibiotic pesticide”, “16S”, “insecticide”, “fungicide”, “herbicide”, “bacterial community”, “antibiotics”, “bactericide”, “tetracyclines”, “sulfonamides”, “macrolides”, “β-lactams”, and “quinolones”. The use of these broad keywords ensured a more comprehensive retrieval of literature. To ensure the homogeneity and high quality of the included studies, the literature screening and data extraction process of this study was performed according to the following screening criteria: (1) The research must explicitly explore the application of one or more pesticides, including both antibiotic and non-antibiotic types, to agricultural soil. (2) The experiment must be conducted in naturally sourced soil (both in-field and laboratory cultures are acceptable). (3) The core analysis method must be 16S rRNA gene high-throughput sequencing, and its raw sequencing (FASTQ format) must be public and downloadable from a public database (such as NCBI SRA). The included articles are listed in Supplementary Information Table S1.

2.2. Taxonomic and Pathogen Annotation

The raw sequencing data were downloaded from the repository addresses provided in the cited literature (Table S1). The raw paired-end sequencing data were downloaded and analyzed using QIIME 2 [24]. Paired reads were merged and filtered for quality. Amplicon sequence variants (ASVs) were inferred using DADA2 [25]. Taxonomic assignment was then performed with the qiime feature-classifier classify-sklearn method, using a pre-trained SILVA 138 Naive Bayes classifier as the reference taxonomy. We annotated potential pathogens by matching taxa identified in our ASV-based species assignments to the curated species list from the Microbial Pathogen Database (MBPD), which comprises 1986 known animal, plant, and zoonotic bacterial pathogen species [26].

2.3. ARG Annotation and Host Analyses

We predicted functional profiles from 16S rRNA amplicon sequencing data using PICRUSt2, obtaining a comprehensive KEGG Ortholog (KO) abundance table [27]. Based on established methods [28,29,30], we identified ARG-related KOs by matching them to the KO entries designated as resistance genes in the KEGG antimicrobial resistance BRITE hierarchy (KO01504) [26]. This enabled us to separate ARG-associated functions from other metabolic KOs and generate distinct ARG and non-ARG abundance tables. Furthermore, by using the --stratified option in PICRUSt2, we extracted taxon-resolved contributions (pred_metagenome_contribution.tsv.gz) to ARG-associated KOs, allowing attribution of predicted resistance functions to specific microbial taxa, thereby identifying ARG hosts.

2.4. Statistical Analyses

Data normality was assessed using the Shapiro–Wilk test. Normally distributed data were analyzed using Welch’s t-test, whereas non-normal data were analyzed using the Wilcoxon rank-sum test. This study used the R package vegan (v2.7-1) to assess the diversity of ARG communities. Alpha diversity was quantified by the Shannon index [31]. For beta diversity, this study performed Principal Coordinate Analysis (PCoA) based on the Bray–Curtis dissimilarity matrix to visualize the community structure differences between different groups (antibiotic and non-antibiotic) [32]. In addition, this study also used permutational multivariate analysis of variance (PERMANOVA) to test the significance of differences in community structure between different groups.
To reveal the co-occurrence patterns among ARG subtypes, we constructed co-occurrence networks. The networks were constructed based on Spearman’s rank correlation coefficients, retaining only connections that were statistically significant (p < 0.05, after FDR correction) and highly correlated (ρ > 0.7). Topological properties of the co-occurrence network were calculated using the igraph package in R, and Gephi (v0.9.2) was used for visualization. Finally, based on the high betweenness centrality index of nodes in the network, keystone hubs crucial for maintaining the network structure were identified [33].
To elucidate the ecological processes driving the assembly of ARG communities, we employed the Modified Stochasticity Ratio (MST) method based on the NST R package (v3.1.10) to quantify the relative importance of deterministic and stochastic processes [34]. An MST value below 0.5 indicates that deterministic processes are dominant, while a value above 0.5 indicates that stochastic processes are dominant [34].

3. Results

3.1. Pesticide Type Differentially Shapes Soil ARG Composition and Diversity

We detected a total of 99 ARG subtypes across all soil samples, and every subtype was present in both the antibiotic and non-antibiotic groups (Table S2). A significant divergence was observed in the total ARG burden between the two groups. ARGs constituted a minor fraction of the total predicted functional gene pool (Figure S1). Soils exposed to antibiotic pesticides exhibited an ARG burden that was nearly twice as high as that in soils exposed to non-antibiotic pesticides, accounting for 0.52% and 0.27% of the total gene abundance, respectively (Figure 1A).
Specific ARG classes showed differing relative abundances between the two groups. In soils exposed to antibiotic pesticides, the mean abundances of aminoglycoside, β-lactamase, and macrolide ARGs were approximately 4.0-, 3.1-, and 2.9-fold higher, respectively, than in soils exposed to non-antibiotic pesticides (Figure 1A and Table S3). Conversely, the mean abundances of quinolone, tetracycline, and phenicol ARGs were approximately 4.3-, 2.1-, and 1.4-fold higher in the non-antibiotic group compared to the antibiotic group (Figure 1A and Table S3). Wilcoxon test analysis confirmed that the differences in abundance for all major ARG classes between the two groups were statistically significant (p < 0.001) (Figure S2 and Table 1). The ARG compositional structure also differed markedly between the groups (Figure 1B). In the antibiotic group, the resistome was heavily dominated by aminoglycoside ARGs (50.94%) and macrolide ARGs (21.82%), which together constituted over 72% of the total ARG abundance. In contrast, the resistome in the non-antibiotic group was co-dominated by phenicol ARGs (27.30%) and aminoglycoside ARGs (25.34%), with substantial contributions from tetracycline (14.32%) and macrolide (14.56%) ARGs as well.
We used the Shannon index to assess alpha diversity across the groups, which ranged from 6.88 to 7.98 in the antibiotic group and from 6.95 to 8.14 in the non-antibiotic group, with the non-antibiotic group showing significantly greater alpha diversity (Welch’s t-test, p < 0.05) (Figure 2A and Table S4). Principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarity revealed that the first two axes (PC1 and PC2) explained 44.90% and 18.14% of the total variation in ARG composition among samples, respectively (Figure 2B). Samples from the antibiotic and non-antibiotic groups formed distinct clusters, and PERMANOVA confirmed a significant difference in ARG composition associated with pesticide type (R2 = 0.215, p < 0.001) (Figure 2B).

3.2. Deterministic Processes Dominate ARG Assembly Across Different Pesticide-Use Soils

In this study, we applied the modified stochasticity ratio (MST) to evaluate the influence of stochastic processes on ARG composition and to elucidate the assembly mechanisms of ARG communities (Figure 3). The results showed that the MST values of ARGs in antibiotic pesticide soils averaged 0.32 ± 0.30, while those in non-antibiotic pesticide soils averaged 0.26 ± 0.19 (Figure 3). MST values were predominantly distributed below the 0.5 threshold, indicating that ARG community assembly was largely governed by deterministic processes in both groups (Figure 3). Moreover, the MST values in the non-antibiotic group were significantly (Wilcoxon rank-sum test, p < 0.05) lower than those in the antibiotic group (Figure 3), suggesting a relatively stronger influence of deterministic processes in the non-antibiotic pesticide soils.

3.3. ARG Co-Occurrence Patterns Vary with Pesticide Type

To investigate the co-occurrence patterns of ARGs in soils with different pesticide exposures, co-occurrence networks were constructed based on significant Spearman correlations (ρ > 0.7, p < 0.05) (Figure 4). The two networks presented contrasting topological properties. The antibiotic network (12 nodes, 24 edges) (Figure 4A) was substantially smaller than the non-antibiotic network (44 nodes, 166 edges) (Figure 4B). Network topology metrics (Table S5) revealed that the antibiotic network was more densely connected and exhibited greater degree centralization, indicating a more compact and centralized co-occurrence structure. In contrast, the non-antibiotic network was generally sparser and displayed a higher average degree, reflecting more connections per node.
Keystone nodes were identified as the three ARGs with highest betweenness centrality (Table S6). In the antibiotic network, the keystone nodes were TEM-5, aac(3)-Ib, and aac(3)-Iva (Table S5). Among these, TEM-5 is a β-lactam resistance gene, while both aac(3)-Ib and aac(3)-Iva are aminoglycoside resistance genes. In the non-antibiotic network, the keystone nodes were aac(2′)-Ia, aac(2′)-Id, and erm(B) (Table S5). The genes aac(2′)-Ia and aac(2′)-Id are aminoglycoside resistance genes, whereas erm(B) is a macrolide resistance gene.

3.4. Distinct ARG Host Community Under Antibiotic and Non-Antibiotic Pesticides

Our analysis of ARG-hosting taxa showed that taxa differed markedly in their contributions to ARG abundance in the soil resistome. The host community was dominated by five core phyla: Pseudomonadota, Actinomycetota, Bacteroidota, Bacillota, and Acidobacteriota, which collectively accounted for over 90% of the total ARG-host contribution (Figure S3). These core phyla showed distinct contribution patterns in the antibiotic versus non-antibiotic groups. Pseudomonadota was the predominant contributor, accounting for 64.5% in the antibiotic group versus 61.5% in the non-antibiotic group (Figure S3). Similarly, contributions of Actinomycetota and Acidobacteriota were higher in the antibiotic group than in the non-antibiotic group (Figure S3). In contrast, Bacillota and Bacteroidota displayed higher contributions in the non-antibiotic group than in the antibiotic group (Figure S3). Between-group comparisons of mean contributions per sample showed that all core phyla differed significantly (p < 0.05) except Pseudomonadota (Figure S4).
We further analyzed the primary hosts of different ARG types across samples with different pesticide backgrounds. For certain resistance types, the dominant host phylum was consistent between the two groups (Figure 5). For instance, Pseudomonadota was the major contributor to β-lactam ARGs in both the antibiotic (100%) and non-antibiotic (99.8%) groups (Figure 5). In contrast, many ARG types had different main contributors across the two groups. For quinolone resistance, Actinomycetota was the main contributor in the antibiotic group (94.5%), whereas Pseudomonadota contributed most in the non-antibiotic group (94.6%) (Figure 5). The contribution of Actinomycetota to phenicol resistance was markedly higher in the antibiotic group (45.39%) than in the non-antibiotic group (7.41%) (Figure 5).
We assessed the contribution of potential pathogenic hosts of ARGs using the MBPD and found that non-pathogenic bacteria were the main contributors (Figure S5). Within pathogenic ARG hosts, zoonotic pathogens contributed the most, followed by animal-associated pathogens, while plant pathogens made the lowest contribution (Figure S6 and Table S7). Statistical comparison between groups showed that the contributions of zoonotic and animal-associated pathogens were significantly (p < 0.01) higher in the non-antibiotic group than in the antibiotic group (Figure S7 and Table 2). In contrast to the other pathogen groups, the contribution from plant pathogens did not differ significantly between the two groups (Figure S7 and Table 2).

4. Discussion

Although agricultural pesticides are widely recognized for their capacity to alter soil ARG profiles, the differential impacts of antibiotic versus non-antibiotic pesticides remain poorly characterized. Our study clearly revealed how these two pesticide classes distinctly shape soil resistomes, identifying significant differences in ARG burden, diversity, community composition, assembly mechanisms, co-occurrence networks, and host taxonomic profiles. We recognize that predicting ARGs based on 16S rRNA data may have inherent limitations in precision. However, our study aims to identify the systematic differences in soil resistome profiles shaped by different pesticide types through a comparative analysis, a research strategy that is well-supported in the literature [28,29,30]. Notably, Chen et al. [28]. found that PICRUSt2 results showed good consistency with those from qPCR. Therefore, the methodology used in our study can provide important insights into the ecological impact of pesticides on the soil resistome.
Our findings reveal that the direct application of antibiotic pesticides imposes a more substantial burden on the soil resistome than non-antibiotic pesticides, although a shared set of 99 ARGs was detected in both soil types. Soils exposed to antibiotic pesticides harbored nearly double the total ARG abundance compared to those exposed to non-antibiotic pesticides (0.52% vs. 0.27%) (Figure 1A). This result is likely a direct consequence of strong, targeted selective pressure: antibiotic compounds eliminate susceptible bacteria, thereby allowing intrinsically resistant microorganisms and those carrying corresponding ARGs to proliferate and dominate the ecological niche [18,19,20,21]. This was particularly evident in the 4.0-fold higher abundance of aminoglycoside resistance genes in the antibiotic group, a finding consistent with the widespread use of aminoglycosides (e.g., streptomycin) in agriculture and previous reports of their enrichment in soils amended with antibiotic-contaminated manure [35,36].
In contrast, while non-antibiotic pesticides resulted in a lower overall ARG burden, they appeared to foster a more diverse resistome, exhibiting significantly higher alpha diversity (Figure 2A). This presents an intriguing paradox. Although antibiotic-exposed soils have a higher ARG load, they exhibit lower diversity. This highlights their different mechanisms of action. Instead of the direct “sledgehammer” selection of antibiotic pesticides, non-antibiotic pesticides likely promote ARG proliferation and diversification through indirect, non-lethal mechanisms. These can include inducing bacterial SOS responses, increasing cell membrane permeability to facilitate horizontal gene transfer, and triggering cross-selection due to overlapping tolerance mechanisms, as reported in studies on herbicides like glyphosate and R-flurtamone [8,9,11,12,13,14]. This explains our observation that certain ARG classes, such as those resistant to quinolones and tetracyclines, were significantly more abundant in the non-antibiotic group (Figure 1A), suggesting a different spectrum of co-selection pressures.
The mechanisms underlying ARG maintenance play a crucial role in shaping their composition and influencing their potential for dissemination [37,38]. According to niche and neutral theories, the structure of ARG communities may be governed by a combination of deterministic processes (such as selective pressures) and stochastic processes (including random dispersal and ecological drift) [37,38,39]. It is generally believed that under stronger environmental stress, deterministic forces are more likely to dominate the assembly of ARG communities [40,41]. Our results demonstrate that ARG community assembly in soils exposed to both antibiotic-based and non-antibiotic pesticides is governed predominantly by deterministic processes (Figure 3), indicating that each class of pesticide imposes strong environmental selection on the soil resistome. Studies of ARG assembly in soils under pesticide exposure or in the presence of pesticide residues remain scarce. Liu et al. [37] investigated soils from forest, grassland and cropland across temperate monsoon, temperate continental, subtropical monsoon and plateau mountain climates and similarly found that deterministic factors, chiefly pH, dominate resistome assembly. In contrast, Yu et al. [38] reported that stochastic processes predominated ARG assembly in paddy soils from large and small farms in Jiangsu Province, China. These divergent findings suggest that the balance between deterministic and stochastic control of soil ARGs may vary with environmental context and spatial scale. Moreover, we found that deterministic assembly of ARG communities was stronger in soils associated with non-antibiotic pesticide use than in those with antibiotic pesticide exposure (Figure 3). It suggests that non-antibiotic agrochemicals exert a more complex and multifaceted selective pressure on the entire microbial community structure, which in turn deterministically shapes the resistome [8,9,11,12,13,14].
To explore the complex relationships among ARGs under different pesticide backgrounds, we constructed co-occurrence networks based on significant correlations. These networks were used to identify key ARG interactions and compare how pesticide class shapes the overall architecture of the soil resistome. The contrasting topologies of the antibiotic and non-antibiotic pesticide networks reveal fundamentally different co-occurrence patterns of ARGs (Figure 4 and Table S5). The antibiotic network is compact, with fewer nodes and edges but a higher density of connections (Figure 4 and Table S5), suggesting a tightly knit cluster of ARGs with strong mutual dependencies [33,42]. This configuration is closely tied to keystone nodes (e.g., TEM-5, aac(3)-Ib, and aac(3)-Iva), which exert significant control over the network (Figure 4 and Table S6). Such cohesiveness implies that the loss of these critical genes may destabilize the network and compromise its structural integrity [33,42]. In contrast, the non-antibiotic network is larger, with more nodes and edges and a higher average degree per ARG (Figure 4 and Table S5). This indicates that non-antibiotic pesticides foster a more diverse and dispersed resistome, where ARGs are less reliant on specific keystone nodes [33,42]. In other words, the loss of keystone genes in the non-antibiotic network would have a milder impact on overall network stability than in the antibiotic network [33,42]. Indeed, we found that keystone genes in the antibiotic network exhibit higher betweenness centrality compared to those in the non-antibiotic network (Table S6), positioning them as reliable indicators of the presence of co-occurring ARGs [33,42,43]. In both networks, aminoglycoside resistance genes dominate the keystone set (two of three in each) (Figure 4 and Table S6), suggesting a foundational role in maintaining ARG co-occurrence across pesticide types. This is consistent with their high relative abundance in both groups (Figure 1B). However, the third keystone differs markedly: a β-lactam resistance gene (TEM-5) in the antibiotic network versus a macrolide resistance gene (erm(B)) in the non-antibiotic network (Figure 4 and Table S6). Though aminoglycoside resistance genes were universally central, the specific third keystone node uniquely differentiated the networks, highlighting how pesticide class shapes which additional ARG types attain central importance. In summary, our findings reveal that antibiotic and non-antibiotic pesticides shape distinct ARG network patterns. Given the antibiotic network’s greater reliance on keystone nodes, mitigating ARG dissemination in such soils could involve altering environmental conditions that influence these pivotal ARGs.
Quantifying the contributions of different host taxa provided further insight into the biological drivers of ARG dissemination in soils associated with antibiotic- and non-antibiotic pesticide exposure. We found that Pseudomonadota (formerly Proteobacteria) serves as the predominant host of ARGs in both antibiotic and non-antibiotic pesticide soils, contributing over 60% to ARGs in each (Figure S1). This observation aligns with findings from various environmental studies [44,45,46,47]. For instance, Wang et al., using high-throughput qPCR, found that Proteobacteria (the former name for Pseudomonadota) were the main potential hosts of ARGs in agricultural soils undergoing solarization and manure amendment [45]. Similarly, a metagenomic investigation of soils under long-term fertilization revealed that bacterial genera significantly associated with ARG subtypes primarily belonged to Proteobacteria [46]. Furthermore, Proteobacteria have also been identified as the dominant ARG hosts in pig farm wastewater treatment systems and in soils fertilized with pig manure [47]. These findings establish Pseudomonadota as a key reservoir for ARGs in various ecosystems. Our study further reveals that the contribution of Pseudomonadota remained remarkably stable across both pesticide types, with no significant difference observed (p > 0.05) (Figure S4), suggesting that the type of pesticide applied has a limited impact on this principal contributor.
In contrast to the stability of Pseudomonadota, the contribution of many phyla varied marked. For example, Actinomycetota’s contribution was significantly increased in samples associated with antibiotic pesticide use compared to those with non-antibiotic pesticides (Figures S3 and S4). As the producers of over two-thirds of nature’s antibiotics, Actinomycetota have evolved innate self-protection mechanisms, often encoded by genes located within their antibiotic synthesis clusters, which are themselves functional ARGs [48,49,50]. Their pronounced contribution in the antibiotic pesticide-influenced group is likely a direct consequence of this pre-adaptation, which allows them to withstand the lethal effects of antibiotic pesticides and thrive where other bacteria perish. This selective pressure also provides a compelling explanation for the observed “shifts” in the primary hosts of specific ARG types. For example, Actinomycetota’s contribution to quinolone resistance was markedly higher in the group associated with antibiotic pesticide use (Figure 5). This is likely because the antibiotic pesticide strongly selected for the “resistance specialists”—Actinomycetota. As this phylum proliferated, the quinolone resistance genes it harbors were consequently amplified, making it the apparent dominant host under this condition. The hosts of certain ARGs, such as β-lactam resistance genes, remained consistently dominated by Pseudomonadota across both pesticide groups (Figure 5). This observation further emphasizes the fundamental role of Pseudomonadota as a key ARG reservoir and highlights its position as the primary and steadfast host of β-lactam resistance genes in soils exposed to different pesticide types.
Finally, we assessed ARG-host pathogenicity to reveal distinct health-risk profiles for each pesticide type. Reassuringly, non-pathogenic bacteria constituted the vast majority (>84%) of ARG hosts in all samples, indicating a low risk profile for the bulk of the host community (Figure S5). To provide a more detailed assessment of the potential health risks, we further analyzed the pathogenic fraction of the ARG hosts (Figure S6). This examination revealed divergent trends between the two groups (Figure S7 and Table 2). In the non-antibiotic group, the contributions of zoonotic and animal-associated pathogens as ARG hosts were significantly higher (p < 0.05) than in the antibiotic group (Figure S7 and Table 2). This suggests that non-antibiotic pesticides, while not directly selecting for resistance, may pose a greater public health risk by disrupting the soil’s microbial equilibrium, thereby creating niches for opportunistic pathogens to proliferate and acquire ARGs. In contrast, we found no significant difference in the ARG contribution from plant pathogens (Figure S7 and Table 2). This finding suggests that antibiotic and non-antibiotic pesticides exert selective pressures of a comparable magnitude on this host group.

5. Conclusions

This study revealed that antibiotic and non-antibiotic pesticides distinctly shape agricultural soil resistomes. We reveal that these two pesticide classes shape ARG profiles through distinct mechanisms, affecting their abundance, diversity, co-occurrence networks, and host compositions. Notably, our findings demonstrate a critical trade-off: antibiotic pesticides correlate with a higher ARG burden, likely driven by direct selective pressure, while non-antibiotic pesticides correlate with greater ARG diversity, suggesting that they promote resistance through indirect, non-lethal mechanisms. Deterministic processes predominantly govern ARG community assembly under both conditions, with the selective pressure appearing stronger and more complex in non-antibiotic soils. Our network analysis uncovered contrasting ARG co-occurrence patterns: a compact, centralized network in antibiotic-influenced soils versus a more dispersed, resilient network in non-antibiotic soils. Across these different structures, aminoglycoside resistance genes consistently held keystone positions, highlighting their foundational role in agricultural resistomes. Pseudomonadota emerged as the stable, predominant ARG host regardless of pesticide type. Our analysis of host pathogenicity identified divergent and critical health-risk profiles: non-antibiotic pesticides amplified the role of zoonotic and animal-associated pathogens as ARG hosts, posing a greater potential risk to public and animal health. Conversely, the comparable contribution of plant pathogens to ARGs under both pesticide regimes indicates that non-antibiotic pesticides may pose risks to crop health similar to those of antibiotic pesticides. Overall, these findings provide a new perspective on how different pesticide classes contribute to antimicrobial resistance, highlighting the need for tailored management strategies. Future research employing a range of analytical methods, including qPCR, shotgun metagenomics, and third-generation sequencing, can be used to further validate the ecological patterns revealed in this study.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/agriculture15192015/s1, Figure S1: Relative abundance of genes within the total predicted functional gene pool; Figure S2: Boxplots of the relative abundance of major ARG classes between soils associated with antibiotic and non-antibiotic pesticide use; Figure S3: Relative contribution of dominant bacterial phyla to the total ARG pool; Figure S4: Boxplots of the ARG contribution from dominant host phyla in two pesticide groups; Figure S5: Contribution of pathogenic versus non-pathogenic bacteria to the ARG pool; Figure S6: Contribution of different pathogenic categories to the ARG pool; Figure S7: Boxplots of the contribution of different pathogen categories to ARG pool in soils associated with antibiotic and non-antibiotic pesticide use; Table S1: Details of the literature sources and datasets included in this study; Table S2: Annotation and relative abundance of the 99 ARG subtypes detected in soils associated with antibiotic and non-antibiotic pesticide use; Table S3: Statistical comparison of the relative abundance (%) of major ARG classes between soils associated with antibiotic and non-antibiotic; Table S4: Shannon and Simpson indexes for ARG communities in each soil sample; Table S5: Topological properties of the ARG co-occurrence networks for the antibiotic and non-antibiotic pesticide groups; Table S6: Keystone ARGs and their Betweenness Centrality values in each co-occurrence network; Table S7: Statistical comparison of the relative contribution (%) of different pathogen categories to ARG pool in soils associated with antibiotic.

Author Contributions

Conceptualization, L.L. and C.Z.; methodology, L.L. and C.Z.; software, Q.L.; validation, C.H. and X.Z.; formal analysis, J.Y.; investigation, L.L. and C.Z.; resources, L.L.; data curation, L.L., H.Z. and C.H.; writing—original draft preparation, L.L.; writing—review and editing, L.L.; visualization, L.L., Q.L. and C.Z.; supervision, C.Z.; project administration, X.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Guangxi Key Research and Development Plan (Guike AB24010031) and the Guangxi Academy of Agricultural Sciences Basic Scientific Research Project (Guinongke 2021YT148).

Institutional Review Board Statement

Not applicable.

Data Availability Statement

The data used to support the findings of this study are available from the corresponding author upon request.

Acknowledgments

The authors would like to thank their college and the laboratory, as well as the reviewers who provided helpful and gratefully appreciated suggestions for this manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Naghavi, M.; Vollset, S.E.; Ikuta, K.S.; Swetschinski, L.R.; Gray, A.P.; Wool, E.E.; Robles Aguilar, G.; Mestrovic, T.; Smith, G.; Han, C.; et al. Global Burden of Bacterial Antimicrobial Resistance 1990–2021: A Systematic Analysis with Forecasts to 2050. Lancet 2024, 404, 1199–1226. [Google Scholar] [CrossRef]
  2. Tacconelli, E.; Carrara, E.; Savoldi, A.; Harbarth, S.; Mendelson, M.; Monnet, D.L.; Pulcini, C.; Kahlmeter, G.; Kluytmans, J.; Carmeli, Y.; et al. Discovery, Research, and Development of New Antibiotics: The WHO Priority List of Antibiotic-Resistant Bacteria and Tuberculosis. Lancet Infect. Dis. 2018, 18, 318–327. [Google Scholar] [CrossRef] [PubMed]
  3. Zhuang, M.; Achmon, Y.; Cao, Y.; Liang, X.; Chen, L.; Wang, H.; Siame, B.A.; Leung, K.Y. Distribution of Antibiotic Resistance Genes in the Environment. Environ. Pollut. 2021, 285, 117402. [Google Scholar] [CrossRef]
  4. Delgado-Baquerizo, M.; Hu, H.-W.; Maestre, F.T.; Guerra, C.A.; Eisenhauer, N.; Eldridge, D.J.; Zhu, Y.-G.; Chen, Q.-L.; Trivedi, P.; Du, S.; et al. The Global Distribution and Environmental Drivers of the Soil Antibiotic Resistome. Microbiome 2022, 10, 219. [Google Scholar] [CrossRef] [PubMed]
  5. Liao, H.; Li, H.; Duan, C.-S.; Zhou, X.-Y.; An, X.-L.; Zhu, Y.-G.; Su, J.-Q. Metagenomic and Viromic Analysis Reveal the Anthropogenic Impacts on the Plasmid and Phage Borne Transferable Resistome in Soil. Environ. Int. 2022, 170, 107595. [Google Scholar] [CrossRef]
  6. Meena, R.S.; Kumar, S.; Datta, R.; Lal, R.; Vijayakumar, V.; Brtnicky, M.; Sharma, M.P.; Yadav, G.S.; Jhariya, M.K.; Jangir, C.K.; et al. Impact of Agrochemicals on Soil Microbiota and Management: A Review. Land 2020, 9, 34. [Google Scholar] [CrossRef]
  7. Li, J.; Yang, Z.; Zhu, Q.; Zhong, G.; Liu, J. Biodegradation of Soil Agrochemical Contamination Mitigates the Direct Horizontal Transfer Risk of Antibiotic Resistance Genes to Crops. Sci. Total Environ. 2023, 901, 166454. [Google Scholar] [CrossRef] [PubMed]
  8. Zhang, Y.-Q.; Cheng, L.-C.; Zhao, F.-J.; Chen, M.-M.; Wang, P. Chiral Pesticides Selectively Influence the Dissemination of Antibiotic Resistance Genes: An Overlooked Environmental Risk. Environ. Sci. Technol. 2025, 59, 13374–13384. [Google Scholar] [CrossRef]
  9. Liao, H.; Li, X.; Yang, Q.; Bai, Y.; Cui, P.; Wen, C.; Liu, C.; Chen, Z.; Tang, J.; Che, J.; et al. Herbicide Selection Promotes Antibiotic Resistance in Soil Microbiomes. Mol. Biol. Evol. 2021, 38, 2337–2350. [Google Scholar] [CrossRef]
  10. Zeng, Y.; Feng, R.; Huang, C.; Liu, J.; Yang, F. Antibiotic Resistance Genes in Agricultural Soils: A Comprehensive Review of the Hidden Crisis and Exploring Control Strategies. Toxics 2025, 13, 239. [Google Scholar] [CrossRef]
  11. Huang, C.; Cui, M.; Li, T.; Zheng, C.; Qiu, M.; Shan, M.; Li, B.; Zhang, L.; Yu, Y.; Fang, H. Migration of Fungicides, Antibiotics and Resistome in the Soil-Lettuce System. J. Hazard. Mater. 2025, 484, 136725. [Google Scholar] [CrossRef]
  12. Qiu, D.; Ke, M.; Zhang, Q.; Zhang, F.; Lu, T.; Sun, L.; Qian, H. Response of Microbial Antibiotic Resistance to Pesticides: An Emerging Health Threat. Sci. Total Environ. 2022, 850, 158057. [Google Scholar] [CrossRef]
  13. Zhang, H.; Chen, S.; Zhang, Q.; Long, Z.; Yu, Y.; Fang, H. Fungicides Enhanced the Abundance of Antibiotic Resistance Genes in Greenhouse Soil. Environ. Pollut. 2020, 259, 113877. [Google Scholar] [CrossRef] [PubMed]
  14. Rangasamy, K.; Athiappan, M.; Devarajan, N.; Parray, J.A. Emergence of Multi Drug Resistance among Soil Bacteria Exposing to Insecticides. Microb. Pathog. 2017, 105, 153–165. [Google Scholar] [CrossRef] [PubMed]
  15. Miller, S.A.; Ferreira, J.P.; LeJeune, J.T. Antimicrobial Use and Resistance in Plant Agriculture: A One Health Perspective. Agriculture 2022, 12, 289. [Google Scholar] [CrossRef]
  16. USEPA What Are Antimicrobial Pesticides? Available online: https://www.epa.gov/pesticide-registration/what-are-antimicrobial-pesticides (accessed on 18 April 2025).
  17. WHO Critically Important Antimicrobials for Human Medicine: 6th Revision. Available online: https://www.who.int/publications/i/item/9789241515528 (accessed on 18 April 2025).
  18. Sundin, G.W.; Monks, D.E.; Bender, C.L. Distribution of the Streptomycin-Resistance Transposon Tn 5393 among Phylloplane and Soil Bacteria from Managed Agricultural Habitats. Can. J. Microbiol. 1995, 41, 792–799. [Google Scholar] [CrossRef]
  19. Verhaegen, M.; Bergot, T.; Liebana, E.; Stancanelli, G.; Streissl, F.; Mingeot-Leclercq, M.-P.; Mahillon, J.; Bragard, C. On the Use of Antibiotics to Control Plant Pathogenic Bacteria: A Genetic and Genomic Perspective. Front. Microbiol. 2023, 14, 1221478. [Google Scholar] [CrossRef]
  20. Shentu, J.; Zhang, K.; Shen, D.; Wang, M.; Feng, H. Effect from Low-Level Exposure of Oxytetracycline on Abundance of Tetracycline Resistance Genes in Arable Soils. Environ. Sci. Pollut. Res. 2015, 22, 13102–13110. [Google Scholar] [CrossRef] [PubMed]
  21. Herbert, A.; Hancock, C.N.; Cox, B.; Schnabel, G.; Moreno, D.; Carvalho, R.; Jones, J.; Paret, M.; Geng, X.; Wang, H. Oxytetracycline and Streptomycin Resistance Genes in Xanthomonas Arboricola Pv. Pruni, the Causal Agent of Bacterial Spot in Peach. Front. Microbiol. 2022, 13, 821808. [Google Scholar] [CrossRef]
  22. Cerqueira, F.; Matamoros, V.; Bayona, J.M.; Berendonk, T.U.; Elsinga, G.; Hornstra, L.M.; Piña, B. Antibiotic Resistance Gene Distribution in Agricultural Fields and Crops. A Soil-to-Food Analysis. Environ. Res. 2019, 177, 108608. [Google Scholar] [CrossRef]
  23. Guo, Y.; Qiu, T.; Gao, M.; Sun, Y.; Cheng, S.; Gao, H.; Wang, X. Diversity and Abundance of Antibiotic Resistance Genes in Rhizosphere Soil and Endophytes of Leafy Vegetables: Focusing on the Effect of the Vegetable Species. J. Hazard. Mater. 2021, 415, 125595. [Google Scholar] [CrossRef]
  24. Caporaso, J.G.; Kuczynski, J.; Stombaugh, J.; Bittinger, K.; Bushman, F.D.; Costello, E.K.; Fierer, N.; Peña, A.G.; Goodrich, J.K.; Gordon, J.I.; et al. QIIME Allows Analysis of High-Throughput Community Sequencing Data. Nat. Methods 2010, 7, 335–336. [Google Scholar] [CrossRef] [PubMed]
  25. Callahan, B.J.; McMurdie, P.J.; Rosen, M.J.; Han, A.W.; Johnson, A.J.A.; Holmes, S.P. DADA2: High-Resolution Sample Inference from Illumina Amplicon Data. Nat. Methods 2016, 13, 581–583. [Google Scholar] [CrossRef] [PubMed]
  26. Yang, X.; Jiang, G.; Zhang, Y.; Wang, N.; Zhang, Y.; Wang, X.; Zhao, F.; Xu, Y.; Shen, Q.; Wei, Z. MBPD: A Multiple Bacterial Pathogen Detection Pipeline for One Health Practices. iMeta 2023, 2, e82. [Google Scholar] [CrossRef]
  27. Douglas, G.M.; Maffei, V.J.; Zaneveld, J.R.; Yurgel, S.N.; Brown, J.R.; Taylor, C.M.; Huttenhower, C.; Langille, M.G.I. PICRUSt2 for Prediction of Metagenome Functions. Nat. Biotechnol. 2020, 38, 685–688. [Google Scholar] [CrossRef]
  28. Chen, M.; Yan, X.; Tang, Q.; Liu, M.; Yang, M.; Chai, Y.; Wei, Y.; Shen, P.; Zhang, J. Particle Size Transfer of Antibiotic Resistance Genes in Typical Processes of Municipal Wastewater Treatment Plant. Bioresour. Technol. 2025, 424, 132288. [Google Scholar] [CrossRef] [PubMed]
  29. Dželalija, M.; Kvesić-Ivanković, M.; Jozić, S.; Ordulj, M.; Kalinić, H.; Pavlinović, A.; Šamanić, I.; Maravić, A. Marine Resistome of a Temperate Zone: Distribution, Diversity, and Driving Factors across the Trophic Gradient. Water Res. 2023, 246, 120688. [Google Scholar] [CrossRef]
  30. Gao, Y.-X.; Li, X.; Zhao, J.-R.; Zhang, Z.-X.; Fan, X.-Y. Response of Microbial Communities Based on Full-Scale Classification and Antibiotic Resistance Genes to Azithromycin and Copper Combined Pollution in Activated Sludge Nitrification Laboratory Mesocosms at Low Temperature. Bioresour. Technol. 2021, 341, 125859. [Google Scholar] [CrossRef]
  31. Kemp, P.F.; Aller, J.Y. Bacterial Diversity in Aquatic and Other Environments: What 16S RDNA Libraries Can Tell Us. FEMS Microbiol. Ecol. 2004, 47, 161–177. [Google Scholar] [CrossRef]
  32. Stefanowicz, A.M.; Kapusta, P.; Stanek, M.; Rola, K.; Zubek, S. Herbaceous Plant Species Support Soil Microbial Performance in Deciduous Temperate Forests. Sci. Total Environ. 2022, 810, 151313. [Google Scholar] [CrossRef]
  33. Schmidt, J.E.; Kent, A.D.; Brisson, V.L.; Gaudin, A.C.M.M. Agricultural Management and Plant Selection Interactively Affect Rhizosphere Microbial Community Structure and Nitrogen Cycling. Microbiome 2019, 7, 146. [Google Scholar] [CrossRef]
  34. Ning, D.; Deng, Y.; Tiedje, J.M.; Zhou, J. A General Framework for Quantitatively Assessing Ecological Stochasticity. Proc. Natl. Acad. Sci. USA 2019, 116, 16892–16898. [Google Scholar] [CrossRef]
  35. Wang, F.; Han, W.; Chen, S.; Dong, W.; Qiao, M.; Hu, C.; Liu, B. Fifteen-Year Application of Manure and Chemical Fertilizers Differently Impacts Soil ARGs and Microbial Community Structure. Front. Microbiol. 2020, 11, 62. [Google Scholar] [CrossRef]
  36. Jauregi, L.; Epelde, L.; Alkorta, I.; Garbisu, C. Antibiotic Resistance in Agricultural Soil and Crops Associated to the Application of Cow Manure-Derived Amendments From Conventional and Organic Livestock Farms. Front. Vet. Sci. 2021, 8, 633858. [Google Scholar] [CrossRef]
  37. Liu, Z.; Zhao, Y.; Zhang, B.; Wang, J.; Zhu, L.; Hu, B. Deterministic Effect of PH on Shaping Soil Resistome Revealed by Metagenomic Analysis. Environ. Sci. Technol. 2023, 57, 985–996. [Google Scholar] [CrossRef]
  38. Yu, F.; Zhang, Y.; Wen, Y.; Peng, Z.; Xu, D.; Xiao, J. Divergent Risk Gene Profiles in Smallholder and Large-Scale Paddy Farms. Environ. Res. 2025, 285, 122406. [Google Scholar] [CrossRef] [PubMed]
  39. Zhou, J.; Ning, D. Stochastic Community Assembly: Does It Matter in Microbial Ecology? Microbiol. Mol. Biol. Rev. 2017, 81, e00002-17. [Google Scholar] [CrossRef] [PubMed]
  40. Peng, F.; Guo, Y.; Isabwe, A.; Chen, H.; Wang, Y.; Zhang, Y.; Zhu, Z.; Yang, J. Urbanization Drives Riverine Bacterial Antibiotic Resistome More than Taxonomic Community at Watershed Scale. Environ. Int. 2020, 137, 105524. [Google Scholar] [CrossRef]
  41. Vanwonterghem, I.; Jensen, P.D.; Dennis, P.G.; Hugenholtz, P.; Rabaey, K.; Tyson, G.W. Deterministic Processes Guide Long-Term Synchronised Population Dynamics in Replicate Anaerobic Digesters. ISME J. 2014, 8, 2015–2028. [Google Scholar] [CrossRef]
  42. Zhao, H.; Brearley, F.Q.; Huang, L.; Tang, J.; Xu, Q.; Li, X.; Huang, Y.; Zou, S.; Chen, X.; Hou, W.; et al. Abundant and Rare Taxa of Planktonic Fungal Community Exhibit Distinct Assembly Patterns Along Coastal Eutrophication Gradient. Microb. Ecol. 2023, 85, 495–507. [Google Scholar] [CrossRef] [PubMed]
  43. Li, B.; Yang, Y.; Ma, L.; Ju, F.; Guo, F.; Tiedje, J.M.; Zhang, T. Metagenomic and Network Analysis Reveal Wide Distribution and Co-Occurrence of Environmental Antibiotic Resistance Genes. ISME J. 2015, 9, 2490–2502. [Google Scholar] [CrossRef]
  44. Zheng, D.; Yin, G.; Liu, M.; Hou, L.; Yang, Y.; Van Boeckel, T.P.; Zheng, Y.; Li, Y. Global Biogeography and Projection of Soil Antibiotic Resistance Genes. Sci. Adv. 2022, 8, eabq8015. [Google Scholar] [CrossRef]
  45. Wang, X.; Zhang, L.; Gu, J.; Feng, Y.; He, K.; Jiang, H. Effects of Soil Solarization Combined with Manure-Amended on Soil ARGs and Microbial Communities during Summer Fallow. Environ. Pollut. 2023, 333, 121950. [Google Scholar] [CrossRef] [PubMed]
  46. Zhang, Y.; Cheng, D.; Zhang, Y.; Xie, J.; Xiong, H.; Wan, Y.; Zhang, Y.; Chen, X.; Shi, X. Soil Type Shapes the Antibiotic Resistome Profiles of Long-Term Manured Soil. Sci. Total Environ. 2021, 786, 147361. [Google Scholar] [CrossRef] [PubMed]
  47. Zhang, R.-M.; Liu, X.; Wang, S.-L.; Fang, L.-X.; Sun, J.; Liu, Y.-H.; Liao, X.-P. Distribution Patterns of Antibiotic Resistance Genes and Their Bacterial Hosts in Pig Farm Wastewater Treatment Systems and Soil Fertilized with Pig Manure. Sci. Total Environ. 2021, 758, 143654. [Google Scholar] [CrossRef] [PubMed]
  48. Takahashi, Y.; Nakashima, T. Actinomycetes, an Inexhaustible Source of Naturally Occurring Antibiotics. Antibiotics 2018, 7, 45. [Google Scholar] [CrossRef]
  49. Barka, E.A.; Vatsa, P.; Sanchez, L.; Gaveau-Vaillant, N.; Jacquard, C.; Klenk, H.-P.; Clément, C.; Ouhdouch, Y.; van Wezel, G.P. Taxonomy, Physiology, and Natural Products of Actinobacteria. Microbiol. Mol. Biol. Rev. 2016, 80, 1–43. [Google Scholar] [CrossRef]
  50. Fatahi-Bafghi, M. Antibiotic Resistance Genes in the Actinobacteria Phylum. Eur. J. Clin. Microbiol. Infect. Dis. 2019, 38, 1599–1624. [Google Scholar] [CrossRef]
Agriculture 15 02015 g001
Figure 1. Differential abundance and composition of ARG classes in soils exposed to antibiotic versus non-antibiotic pesticides. (A) Comparison of total ARG abundance, expressed as percentage of total genes, between soils exposed to non-antibiotic versus antibiotic pesticides. (B) Compositional structure of ARG classes in the two exposure groups.
Figure 1. Differential abundance and composition of ARG classes in soils exposed to antibiotic versus non-antibiotic pesticides. (A) Comparison of total ARG abundance, expressed as percentage of total genes, between soils exposed to non-antibiotic versus antibiotic pesticides. (B) Compositional structure of ARG classes in the two exposure groups.
Agriculture 15 02015 g001
Agriculture 15 02015 g002
Figure 2. ARG diversity (alpha and beta) under antibiotic vs. non-antibiotic pesticide exposure (A) Alpha diversity of ARG communities, measured by the Shannon index, in soils exposed to antibiotic and non-antibiotic pesticides. Different letters (a, b) above the boxplots indicate a statistically significant difference between the groups (Welch’s t-test, p < 0.05). (B) PCoA of ARG community composition based on Bray–Curtis dissimilarity. Each point represents a soil sample, colored by pesticide type. The percentage of variation explained by each principal coordinate is shown on the axes.
Figure 2. ARG diversity (alpha and beta) under antibiotic vs. non-antibiotic pesticide exposure (A) Alpha diversity of ARG communities, measured by the Shannon index, in soils exposed to antibiotic and non-antibiotic pesticides. Different letters (a, b) above the boxplots indicate a statistically significant difference between the groups (Welch’s t-test, p < 0.05). (B) PCoA of ARG community composition based on Bray–Curtis dissimilarity. Each point represents a soil sample, colored by pesticide type. The percentage of variation explained by each principal coordinate is shown on the axes.
Agriculture 15 02015 g002
Agriculture 15 02015 g003
Figure 3. Ecological processes governing ARG community assembly. Boxplots showing the MST values for ARG communities in soils exposed to antibiotic and non-antibiotic pesticides. MST values below the dashed line (0.5) indicate that community assembly is dominated by deterministic processes, while values above indicate dominance by stochastic processes. Different letters (a, b) indicate a statistically significant difference in MST values between the two pesticide groups (Wilcoxon rank-sum test, p < 0.05).
Figure 3. Ecological processes governing ARG community assembly. Boxplots showing the MST values for ARG communities in soils exposed to antibiotic and non-antibiotic pesticides. MST values below the dashed line (0.5) indicate that community assembly is dominated by deterministic processes, while values above indicate dominance by stochastic processes. Different letters (a, b) indicate a statistically significant difference in MST values between the two pesticide groups (Wilcoxon rank-sum test, p < 0.05).
Agriculture 15 02015 g003
Agriculture 15 02015 g004
Figure 4. Co-occurrence networks of ARGs in soils under different pesticide exposures. Networks visualizing significant positive correlations (Spearman’s ρ > 0.7, p < 0.05) among ARG subtypes in (A) antibiotic-exposed soils and (B) non-antibiotic-exposed soils. Each node represents an ARG subtype, with node size proportional to its betweenness centrality. Nodes are colored by the ARG class they belong to. Keystone ARGs, identified by the highest betweenness centrality, are indicated in bold text.
Figure 4. Co-occurrence networks of ARGs in soils under different pesticide exposures. Networks visualizing significant positive correlations (Spearman’s ρ > 0.7, p < 0.05) among ARG subtypes in (A) antibiotic-exposed soils and (B) non-antibiotic-exposed soils. Each node represents an ARG subtype, with node size proportional to its betweenness centrality. Nodes are colored by the ARG class they belong to. Keystone ARGs, identified by the highest betweenness centrality, are indicated in bold text.
Agriculture 15 02015 g004
Agriculture 15 02015 g005
Figure 5. Contribution of bacterial phyla to different classes of ARG burden under distinct pesticide exposures. Chord diagrams illustrating the contribution of bacterial phyla to major ARG classes in (A) antibiotic-exposed soils and (B) non-antibiotic-exposed soils.
Figure 5. Contribution of bacterial phyla to different classes of ARG burden under distinct pesticide exposures. Chord diagrams illustrating the contribution of bacterial phyla to major ARG classes in (A) antibiotic-exposed soils and (B) non-antibiotic-exposed soils.
Agriculture 15 02015 g005
Table 1. Statistical comparison of the relative abundance (%) of major ARG classes between soils associated with antibiotic and non-antibiotic pesticide use.
Table 1. Statistical comparison of the relative abundance (%) of major ARG classes between soils associated with antibiotic and non-antibiotic pesticide use.
Arg TypeAntibioticNon-Antibioticp-Value
Aminoglycoside0.302250.05103p < 0.001
β-lactamase0.037970.00617p < 0.001
Fosfomycin0.002830.00660p < 0.001
Macrolide0.135940.03074p < 0.001
Phenicol0.059360.06717p < 0.001
Quinolone0.000860.01135p < 0.001
Rifampicin0.001290.00030p < 0.001
Sulfonamide0.001050.00113p < 0.001
Tetracycline0.015990.03075p < 0.001
Trimethoprim0.000790.00093p < 0.001
Total ARGs0.652740.20158p < 0.001
Values are presented as median relative abundance (%). The statistical significance of differences between the two groups was assessed using the Wilcoxon rank-sum test.
Table 2. Statistical comparison of the relative contribution of different pathogen categories to ARG pool in soils associated with antibiotic and non-antibiotic pesticide use.
Table 2. Statistical comparison of the relative contribution of different pathogen categories to ARG pool in soils associated with antibiotic and non-antibiotic pesticide use.
Pathogenic Bacteria TypeAntibioticNon-Antibioticp-Value
Animal0.2001.760p < 0.001
Plant0.0150.030p > 0.05
Zoonotic0.1809.380p < 0.001
Values are presented as median relative abundance (%). The statistical significance of differences between the two groups was assessed using the Wilcoxon rank-sum test.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lyu, L.; Lu, Q.; Huang, C.; Zhang, X.; Yao, J.; Zhao, H.; Zou, C. Pesticide Type Distinctly Shapes Soil Resistomes: A Comparative Analysis of Antibiotic and Non-Antibiotic Agro-Chemicals. Agriculture 2025, 15, 2015. https://doi.org/10.3390/agriculture15192015

AMA Style

Lyu L, Lu Q, Huang C, Zhang X, Yao J, Zhao H, Zou C. Pesticide Type Distinctly Shapes Soil Resistomes: A Comparative Analysis of Antibiotic and Non-Antibiotic Agro-Chemicals. Agriculture. 2025; 15(19):2015. https://doi.org/10.3390/agriculture15192015

Chicago/Turabian Style

Lyu, Lilan, Qinyu Lu, Chanchan Huang, Xiyu Zhang, Jinjie Yao, Huaxian Zhao, and Chengwu Zou. 2025. "Pesticide Type Distinctly Shapes Soil Resistomes: A Comparative Analysis of Antibiotic and Non-Antibiotic Agro-Chemicals" Agriculture 15, no. 19: 2015. https://doi.org/10.3390/agriculture15192015

APA Style

Lyu, L., Lu, Q., Huang, C., Zhang, X., Yao, J., Zhao, H., & Zou, C. (2025). Pesticide Type Distinctly Shapes Soil Resistomes: A Comparative Analysis of Antibiotic and Non-Antibiotic Agro-Chemicals. Agriculture, 15(19), 2015. https://doi.org/10.3390/agriculture15192015

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop