Antibiotic resistance genes (ARGs), a new type of environmental pollutant, are widely enriched in farmland soils, promoting the spread of drug resistance and seriously affecting soil safety (Huijbers et al., 2015; Rosenkrantz et al., 2019). The soil–plant system is a significant source of exposure to environmental resistance genes, which can spread from the environment to humans (Wang et al., 2022a,b). The food chain is the most direct and main pathway for resistant bacteria and genes to enter the human body in soil–plant systems (Blau et al., 2018). With the rapid development of the global economy, resistance genes in agricultural products can spread rapidly worldwide through food processing, preservation, and global transportation, triggering public safe risks and affecting global health (Nowakowska et al., 2019; Zhu et al., 2019). Therefore, to remove and control ARGs in farmland soils, it is particularly important to clarify the influence of soil management practices on ARGs and their relationship with microbial communities.

Cropping system soil, one of the habitats with the highest microbial diversity on Earth, is a huge reservoir of ARGs (Forsberg et al., 2012; Sanderson et al., 2016; Van Goethem et al., 2018). Internal resistance and exogenous input are the main sources of ARGs in agricultural soil (Riber et al., 2014; Nõlvak et al., 2016). Internal resistance refers to the inherent genome of microorganisms, that can be inherited by subsequent generations. Exogenous input involves the horizontal transfer and enrichment of ARGs directly into the soil through the addition of external sources, such as manure, sludge, and reclaimed water. The composition and abundance of ARGs are closely linked to the microbial community and can be affected by changes in microbial community diversity and composition (Forsberg et al., 2014; Liu et al., 2019; Zhang et al., 2021). Studies have shown that Actinobacteria produce the most natural antibiotics, and Firmicutes and Actinobacteria have been identified as the main groups carrying ARGs (Zhou et al., 2015; Xun et al., 2016). Chen et al. (2016) found that the abundance of tetracyclines ARGs increased as the abundance of Actinomycete increased and the abundance of Chloroflexi decreased. A study have also shown that the increased abundance of Proteobacteria was accompanied by increasing β-lactam-resistance genes (Wang et al., 2016). In addition, Xiang et al. (2018) and Urra et al. (2019) showed that the decrease of soil microbial α-diversity led to an increase in MLSB, aminoglycosides, β-lactams, amide alcohols, sulfonamides ARGs and transposase gene tnpA. In summary, previous studies mainly focuses on manure application, sewage-irrigated, and typically contaminated soils (Achiba et al., 2010; Xie et al., 2018), however, there are no reports on the influence mechanism of tillage management measures on agricultural soil ARGs.

The introduction and enrichment of ARGs in farmland soils are commonly attributed to agricultural practices (Wu et al., 2022). For example, manure application can considerably enrich resistant bacteria and ARGs in agricultural soils (Munk et al., 2018; Xie et al., 2018). Fertilizer application can significantly increase the abundance of important resistance genes, including intI1, tetW, tetO, sul1, and sul2 (Xie et al., 2018). Zhao et al. (2019) found that the abundance of tetracycline- and ampicillin-resistant bacteria in greenhouse soils was substantially higher than that in field soils. Besides the change of soil physical and chemical properties (i.e., pH, soil organic carbon (SOC), total nitrogen (TN), NH₄⁺, NO₃⁻, soil texture, etc) may create different niches for bacterial populations and alter the microbial communities, thereby affecting the composition and abundance of ARGs (Dumbrell et al., 2010). In addition, Zhao et al. (2017) found that the concentration of some heavy metals (such as Cu, Zn, Mn, Fe) in soil may also drive the distribution and spread of ARGs in the environment. Conservation tillage, that is, reduced or no tillage with straw return, can decrease soil disturbance and improve its physical and chemical properties, thereby leading to changes in microbial habitats and communities, consequently affecting likely the abundance of ARGs in agricultural soil (Wang et al., 2017, 2021). This practice has been widely used in dry farmlands in northern China (Wang et al., 2020). However, the impact of long-term no-tillage on ARGs in agricultural soil remains unknown, hampering a complete understanding of ARGs in this context.

Therefore, based on a long-term (> 11 yr) conservation tillage experimental field, we investigated the effect on the abundance of ARGs and microbial community composition using shotgun metagenomic sequencing under two different tillage systems, no tillage and plough tillage, in a winter wheat–summer maize rotation system. We hypothesized that long-term no-till application could reduce the abundance and accumulation of ARGs by improving soil properties and microbial community structure. Therefore, three main objectives were established: (1) to analyze the effects of different tillage treatments on the relative abundance of ARGs; (2) to study the contribution of the microbial community to ARGs under long-term no tillage vs. plough tillage; and (3) to determine the correlations between soil variables and changes in ARGs under different tillage treatments.

Agricultural management practices, such as fertilization types, agricultural planting patterns, and planting crops, are assumed to introduce or enrich a large number of ARGs and exert selection pressure on indigenous soil microbial communities (Forsberg et al., 2014; Wang et al., 2017; Li et al., 2022). However, there is a dearth of knowledge regarding how long-term conservation tillage influences the patterns of a wide spectrum of ARGs in dry agricultural soils. In this study, non-metric multidimensional scaling (NMDS) revealed that the soil ARGs clustered into two groups based on the two tillage treatments (Figure 2). These findings indicate that the community structure and abundance of ARGs in the soil were influenced by long-term ZT and PT applications. The total abundance of microbial antibiotic resistance, sensitive, and target genes was observed to be lower after long-term ZT application than after PT (Figure 1). This suggests that the abundance of soil ARGs may decrease with reduced soil disturbance intensity.

Among the general mechanisms of antibiotic resistance, multidrug efflux pumps excluding drugs serve as critical one in multidrug resistance (Hernando-Amado et al., 2016). Our findings indicate that long-term ZT significantly reduced the abundance of soil efflux pump conferring ARGs, with a significant difference of 23.46% compared to PT soil. Simultaneously, a large number and wide distribution of broad-spectrum efflux pumps, such as oqxB, mexK, efpA, adeH, acrD, lrfA vgaALC, amrB, mexT, acrE, lmrB, and mdtL in no-tillage soils than in plough tillage soils (Supplementary Table S1). These results suggest that long-term ZT can reduce the accumulation of antibiotic-resistant microorganisms by weakening the generalized role of broad-spectrum efflux pumps in protecting cells from multiple antibiotics encountered in the agricultural environment (Zhang et al., 2021). In addition, compared to PT soil, the abundance of fluoroquinolone, aminoglycosides, isoniazid, trimethoprim, multi-drug resistance, polymyxins, fosfomycin, MLSB, β-lactams, chloramphenicol, sulfonamides, and ethambutol resistance genes under long ZT soil were also significantly decreased. This may be because long-term ZT creates a soil environment suitable for microbial proliferation, which significantly increases the abundance of beneficial microorganisms, thereby reducing the abundance of antibiotic-resistant microbial hosts (Dunivin and Shade, 2018). These results indicate that long-term ZT application for 11 consecutive years can inhibit the extensive enrichment of ARG abundance in agricultural soil and reduce the spread of resistance (Gullberg et al., 2014; Blanco et al., 2016). This can greatly contribute to the control of soil resistance gene pollution and achieve sustainable agricultural production.

We used mantel test to identify the key factors affecting the accumulation of ARGs, including physical and chemical soil properties, microbial diversity, and microbial metabolic activity (Supplementary Table S2). The conservation tillage has important effects on the accumulation of ARGs in soil by regulating soil microhabitats. Specifically, the abundance of antibiotic biosynthesis, resistance, sensitivity, and target genes was significantly regulated by soil pH (Figure 5). This is consistent with previous studies showing that pH is the most important factor affecting the microbial community and accumulation of antibiotic resistance genes (Rousk et al., 2010; Li et al., 2020). Changes in soil pH led to significant changes in the soil physicochemical properties and microbial communities, resulting in changes in the accumulation of ARGs in the soil (Liu et al., 2017). In addition, mental analysis showed that the SOC content had a significant effect on AGR accumulation (Figure 5). The ARGs in the no-tillage soil were significantly lower than those in the plough tillage soil, mainly because long-term no-tillage formed a carbon-rich microenvironment that was conducive to the balanced proliferation of various microbial populations, thereby inhibiting the niche of the ARGs host population and reducing the accumulation of ARGs in the soil (Forsberg et al., 2014). Besides, our result also found that the concentration of soil available Zn had a significant effect on the Ars family genes, and the concentration of available Fe had a significant effect on the ASs family genes (Figure 5). This was mainly due to long-term no-tillage significantly changed the concentration of available Zn and Fe, which had an effect on soil bacteria and bacteriophages, and thus inhibited the accumulation of ARGs in no-tillage soil (Dickinson et al., 2019).

In addition to the influence of soil properties on ARGs, the diversity and structural composition of the microbial community also significantly affect ARGs (Dunivin and Shade, 2018). Our previous study found that long-term ZT significantly increased soil bacterial and fungal alpha diversity (Wang et al., 2021), which may be another key biological factor in reducing ARGs accumulation in the soil. A higher soil microbial diversity can significantly inhibit the spread of resistance genes (Chen et al., 2019). Previous studies have confirmed that the decrease of α-diversity of soil microbial communities can lead to the increase the abundance of MLSB, aminoglycosides, β-lactams, amide alcohols, sulfonamides and tnpA resistance genes (Han et al., 2018; Urra et al., 2019). These results are consistent with those of this study (Table 3), which may be because long-term ZT increased the content of organic matter and mineral nutrients in soil, thus increasing the niche breadth of microorganisms in the ZT soil (Forsberg et al., 2014), alleviating the selective pressure on microorganisms (Chase and Myers, 2011), and compressing the living space of antibiotic-carrying host microorganisms (Dunivin and Shade, 2018). Furthermore, Forsberg et al. (2014) and Li B. et al., (2015) found robust correlations between bacterial phyla and ARGs, suggesting that the microbial structure composition could be one of the factors affecting the soil resistome. Our previous studies have also shown that long-term ZT application has significant effects on soil bacterial and fungal community structures (Wang et al., 2017, 2021), suggesting that ZT may lead to changes in the abundance and composition of host populations carrying antibiotic resistance genes. For examples, compared with PT, the relative abundance of Firmicutes in ZT soils increased significantly, whereas the relative abundance of Alphaproteobacteria, Betaproteobacteria and γ-proteobacteria decreased significantly (Figure 4). The average number of ARGs in the genome of Proteobacteria is the largest, and their proliferation can promote the growth and creation of drug-resistant bacteria in the soil (Zhou et al., 2015; Xun et al., 2016). Udikovic-Kolic et al. (2014) found that the alteration of the soil resistome can mainly be attributed to an increase in certain bacterial populations, such as Pseudomonas spp., Janthinobacterium sp. and Psychrobacter pulmonis, which harbor ARGs in farmland soil.

Microbial metabolic activity also had a significant effect on the accumulation of antibiotic resistance genes (Figure 5). The results showed that the long-term ZT soil microbial metabolic activity was significantly higher than that of the PT soil, which may be another key factor in reducing the accumulation of ARGs in long-term ZT soils. This is mainly because the higher the metabolic activity of microorganisms, the higher their versatility and health, thereby occupying a niche and crowding out the living space of antibiotic-carrying host microorganisms (Yang et al., 2022). Su et al. (2015) reported similar findings, demonstrating that soil with stronger microbial metabolic activity and a higher relative balance of community structure composition can inhibit the abundance and activity of pathogenic microorganisms and antibiotic gene hosts more effectively.

The long-term application of ZT significantly reduced the accumulation of fluoroquinolone, aminoglycosides, isoniazid, trimethoprim, multi-drug resistance, polymyxins, fosfomycin, MLSB, β-lactams, chloramphenicol, sulfonamides, and ethambutol resistance genes under dryland conditions, compared to PT soil. Such antibiotic resistance gene reduction is closely associated with soil properties, microbial community composition, and metabolic activity. Our findings suggest that long-term ZT practices can reduce the accumulation of antibiotic resistance genes through the following mechanisms: decreased pH and increased SOC content are key physicochemical factors that inhibit ARG accumulation in no-tillage soil, while decreased abundance of Proteobacteria, Actinobacteria, and Acidobacteria communities, and increased abundance of Gemmatimonadetes and Chloroflexi communities, as well as microbial metabolic activity, are key biological factors that inhibit the reproduction of antibiotic-resistant host microbes. This implies that long-term no-tillage practices inhibit the accumulation of antibiotic resistance genes in farmland soil, which is a promising agricultural management measure to reduce the accumulation risk of soil ARGs. Therefore, combining conservation tillage with other high-yield and efficient agricultural practices will greatly improve the sustainability of agricultural production and make a significant contribution to human health in future agricultural production. Future research should identify the risks of active antibiotic resistance gene host microbial populations in the agricultural environment at the transcriptome level, based on RNA sequencing, to ensure that ZT is used sustainably.

Raw reads were uploaded to the NCBI Sequence Read Archive database (accession no. PRJNA641749).

WW: conceptualization, investigation, data curation, writing—original draft, review, editing, and funding acquisition. PS and ZL: partial data investigation. FM: conceptualization, writing—reviewing and editing. YL: conceptualization and funding acquisition. XW: conceptualization, funding acquisition, supervision, writing—reviewing and editing. All authors contributed to the article and approved the submitted version.

This study was financially supported by the Science and Technology Plan Project of Shaanxi Province (Key Industry Innovation Chain Project) (grant number 2023-ZDLNY-03), Starting Research Fund from the Northwest A&F University (grant number 2452021112), and National Natural Science Foundation of China (grant number 31971860).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.