Foliar Application of Bacillus thuringiensis Regulates Soil Physicochemical Properties and Microbial Ecological Functions in Tea Plantations Authors: Pumo Cai, Yulin Xiong, Weifeng Wang, Xi Du, Chengcong Lu, Yanyan Li, Qisong Li, Yongcong Hong Published: 2026 ABSTRACT AbstractBacillus thuringiensis (Bt) is widely employed as a biological control agent against pests in tea plantations, yet its impacts on soil health and microbial ecology remain insufficiently understood. This study investigated the effects of two Bt application regimes, namely moderate-frequency conventional Bt application (Bt1, 3 sprays over 21 days) and high-frequency intensive Bt application (Bt2, 6 sprays over 42 days), on soil physicochemical properties, enzyme activities, and microbial community structure and function in tea soils. The moderate-frequency conventional Bt1 significantly improved soil nutrient status by increasing organic matter, available nitrogen, and potassium, and boosted acid protease, sucrase, and cellulase activities, while Bt2 achieved the maximum urease and polyphenol oxidase activities but failed to promote soil available nutrients and organic matter as effectively as Bt1. Although neither Bt treatment induced notable shifts in overall microbial alpha-diversity indices, community composition differed distinctly between the two treatments. Bt1 enriched beneficial taxa related to nitrogen fixation and organic matter degradation including Pseudomonas, Bradyrhizobium, and Sphingomonas, and sharply suppressed pathogenic fungi such as Aspergillus and Curvularia, whereas Bt2 caused an 18.29% reduction in Sphingomonas abundance with limited enrichment of beneficial bacteria. Functional predictions indicated that Bt1 amplified microbial carbon degradation and nitrogen cycling, enriched saprotrophic and symbiotic fungi and reduced plant pathogens by 85.78%, while both Bt treatments elevated saprotrophic fungi yet Bt2 only reduced pathogens by 11.84%. Co-occurrence network analyses revealed enhanced microbial interactions and community stability under Bt1, while excessive high-frequency Bt2 reduced network connectivity and stability compared with CK and Bt1. These results suggest that moderate-frequency conventional Bt application can positively modulate soil microbial communities and ecosystem functions, providing valuable insights for sustainable pest management and soil health maintenance in tea agroecosystems.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1007/s00248-026-02813-3.Keywords: Bt biopesticide, Camellia sinensis, Soil health, Soil microbial community, Physicochemical properties Introduction Tea [Camellia sinensis (L.) Kuntze], a perennial evergreen crop of considerable economic significance, ranks as the second most consumed beverage globally and forms the cornerstone of a multibillion-dollar industry that supports the livelihoods of over 100 million individuals, primarily in Asia, which contributes approximately 80% of global tea production [1]. As the foremost producer of tea globally, China contributes over 40% of the world’s total output, and in addition to its cultural significance as a cherished beverage, tea leaves serve as an important agricultural export commodity that plays an essential role in international trade dynamics [2]. Yet, tea cultivation remains beleaguered by recurring threats from various types of insect pests and pathogens, such as the tea geometrid (Ectropis grisescens) and tea green leafhopper (Empoasca onukii Matsuda). Collectively, these adversities result in annual yield losses ranging from 10% to 30%, while significantly degrading the quality of harvested tea foliage [3]. Current pest management strategies in tea gardens primarily depend on chemical pesticides, which have been effective in the short term but raise critical concerns. For example, repeated application of diamide insecticides, including tetraniliprole, has driven the evolution of pesticide-resistant Spodoptera litura, a major pest in China’s tea industry, with field populations in Nanchang showing over 100-fold resistance [4]. Moreover, chemical residues on tea foliage not only threaten human health but also exceed stringent international residue thresholds. The European Union, for instance, has established a maximum residue limit of 0.01 mg/kg for over 480 pesticides in tea, which restricts market access for exports [5]. Furthermore, the application of insecticides has been shown to disrupt soil microbial communities, adversely affecting nutrient cycling processes and contributing to the long-term degradation of soil quality within tea gardens [6, 7]. These cumulative drawbacks underscore the urgent necessity to develop sustainable and environmentally friendly alternatives to synthetic pesticides. In recent years, Bacillus thuringiensis (Bt), a well-characterized entomopathogenic bacterium, has garnered considerable attention as a promising biopesticide for integrated pest management (IPM) in tea cultivation [8]. Bt is a soil bacterium that produces spores containing crystalline proteins, specifically one or more Cry or Cyt proteins, which target the midgut epithelial cells of certain insect larvae, such as lepidopterans and coleopterans. Importantly, these proteins exhibit minimal impact on non-target organisms, including vertebrates and beneficial insects [9, 10]. Compared to conventional synthetic insecticides, Bt-based formulations are characterized by rapid environmental degradation, which significantly reduces residue accumulation and associated ecological risks. Moreover, their targeted mode of action minimizes disruption to non-target taxa and helps preserve ecological balance within agroecosystems [11]. Recent studies have demonstrated that Bt applications can effectively control tea pests such as hemipteran insects (e.g., E. onukii) [12] and lepidopteran larvae (e.g., Hyposidra talaca Walker, Eterusia aedea L., and E. grisescens) [13]. Field studies in China have demonstrated a reduction in pest population by 60–85% and significant mitigation of yield loss [14, 15]. Despite these promising insecticidal outcomes, the broader ecological implications of Bt application, particularly regarding soil health, remain insufficiently explored. Specifically, the impact of Bt on soil physicochemical properties and the composition and function of microbial communities in tea gardens requires comprehensive investigation. Soil microbial communities play a pivotal role in sustaining ecosystem functions within tea plantations, where their diversity and structural stability critically influence nutrient cycling, organic matter decomposition, and the disease resistance capacity of tea plants [16]. These microorganisms facilitate key processes such as litter degradation, nitrogen fixation, phosphorus solubilization, and phytohormone production, collectively maintaining soil health and optimizing plant nutrition. Furthermore, a balanced microbial community suppresses soil-borne pathogens through competitive and antagonistic interactions, thereby mitigating the occurrence of root rot and other soil-borne diseases in tea crops [17, 18]. Given their integral role as indicators of soil health and drivers of ecosystem services, understanding the impacts of Bt applications on soil microbial dynamics and associated functions is essential. Such insight is crucial for assessing the long-term sustainability of Bt-based pest management and ensuring that its deployment does not compromise the ecological integrity underpinning tea production. Against this backdrop, this study aims to assess the effects of different Bt application frequencies on soil physicochemical properties and key enzymatic activities in tea soil. By combining high-throughput sequencing of 16 S/18S rDNA with functional prediction tools such as FAPROTAX and FUNGuild, we intend to characterize the responses of soil microbial community structure, diversity, and functional potential to Bt treatments. The results are expected to deepen our understanding of the interactions between Bt application regimes and soil microbial ecology, thereby providing a scientific basis for the rational optimization of Bt deployment in tea agroecosystems. Materials and Methods Plant Material and Growth Conditions One-year-old tea seedlings of Camellia sinensis ‘Rougui’ with uniform growth were sourced from a local nursery in Wuyishan, Fujian Province, China (27°43’N, 118°01’E). These seedlings were grown in polyethylene pots (18 × 22 × 15 cm) containing 3.0 kg of typical red soil. The soil was collected from the 0–20 cm layer of the research tea garden at Wuyi University in the same region, then air-dried and sieved through a 2 mm mesh prior to use. Baseline soil properties were: pH 4.43, available nitrogen 28.21 mg/kg, available phosphorus 10.12 mg/kg, available potassium 168.26 mg/kg, organic matter 1.46%. Five seedlings were transplanted into each pot, totaling 60 pots. After transplanting, pots received uniform irrigation and were subsequently watered every 7 days. The experiment began following a one-month establishment period. Initially, 150 mL of 0.1% Hoagland nutrient solution was applied per pot, with an additional 30 mL supplied every three days to maintain nutrient availability. Throughout the study, pots were maintained in a greenhouse at 25 ± 2 °C and 70 ± 5% relative humidity to ensure optimal growth conditions. Experimental Design and Bt Treatment The Bacillus thuringiensis subsp. aizawai (Bt) used in this study (32,000 IU/g) was a laboratory-fermented crude extract. To align with practical agricultural management, we varied the application frequency rather than testing different concentrations. Two distinct regimes were established: Bt1 (3 sprays over 21 d), representing the conventional pest control schedule in local tea gardens, and Bt2 (6 sprays over 42 d), simulating an intensive management scenario during severe pest outbreaks. Distilled water was applied as the control (CK) following the same schedule. All treatments were applied using a handheld electric sprayer to ensure consistent pressure. For each application, a fixed volume of the suspension was evenly sprayed onto the foliage until the point of runoff, ensuring uniform coverage across all replicates. Sampling was conducted at the end of the 42-day experimental period. Soil Sampling and Preparation Soil samples from the control (CK), Bt1, and Bt2 treatments were collected following a standardized procedure. In each treatment, tea seedlings were carefully uprooted from 12 randomly selected pots to obtain rhizosphere soil, while the remaining soil from every four pots was combined to form one replicate, resulting in three replicates per treatment. All samples were transferred into sterile plastic bags, labeled, and transported to the laboratory. Upon arrival, plant residues were removed, soils were air-dried, sieved through a 100-mesh sieve, and stored at 4 °C for subsequent physicochemical and enzyme activity analyses. Rhizosphere soil was defined as soil adhering within 1 mm of the root surface [19]. After gently shaking off loosely attached soil, roots were sprayed with sterile phosphate-buffered saline (PBS) to detach tightly adhering particles [20]. The resulting suspension was collected in sterile centrifuge tubes and centrifuged at 12,000 rpm for 15 min at 4 °C. The recovered pellet was immediately frozen at − 80 °C for DNA extraction and microbial diversity analysis. Analysis of Soil Physicochemical Properties and Enzyme Activities Available nitrogen (AN) was quantified through the alkaline diffusion technique, whereas available phosphorus (AP) was extracted using a 0.5 mol·L− 1 sodium bicarbonate solution [21]. Available potassium (AK) was measured via ammonium acetate extraction followed by flame spectrophotometry analysis [22]. Soil organic matter content was assessed employing the potassium dichromate titration method, and pH values were recorded using potentiometric measurement [23]. Soil catalase activity (CAT) was evaluated using the pyrogallol-based colorimetric assay. Urease activity (UE) was determined through a colorimetric procedure involving sodium hypochlorite, while polyphenol oxidase activity (PPO) was also measured using the pyrogallol colorimetric assay. Acid protease (ACPT) activity was assessed by a ninhydrin colorimetric method. Additionally, soil cellulase activity (CE) was quantified via the anthrone colorimetric assay, and sucrase activity (IE) was determined using the 3,5-dinitrosalicylic acid technique [24]. Soil Total DNA Extraction Total DNA from soil samples was isolated using the BioFast Soil Genomic DNA Extraction Kit (BioFlux, Hangzhou, China). The integrity and purity of the extracted DNA were evaluated via 1% agarose gel electrophoresis. DNA concentration was then measured using a NanoDrop 2000 C Spectrophotometer (Thermo Scientific, Waltham, MA, USA). Only DNA samples meeting the quality criteria were selected for subsequent high-throughput sequencing analysis of the microbial community. Soil Microbial Diversity Analysis via 16 S and ITS rDNA Sequencing The 16 S rDNA and ITS rDNA regions of soil samples were amplified using specific primers and thermal cycling conditions detailed in Table S1. PCR amplification was performed on an ABI GeneAmp® 9700 system (Perkin Elmer, Waltham, MA, USA). Subsequently, microbial community sequencing was conducted using the Illumina HiSeq platform. Raw sequencing reads were processed with the QIIME pipeline (http://qiime.org/scripts/assign_taxonomy.html) to filter out low-quality sequences, defined as those with an average quality score below 20 (Q < 20) or shorter than 100 base pairs [25]. Operational taxonomic units (OTUs) were then clustered at a 97% sequence similarity threshold using Uparse across all effective tags obtained from the samples [26]. Taxonomic annotation of representative OTU sequences was performed using the BLAST algorithm within QIIME against the UNITE database (v7.2) to determine microbial abundances at various taxonomic levels. Data Analysis Significant analysis was performed using IBM SPSS version 26. One-way analysis of variance (ANOVA) was conducted to assess the effects of different treatment conditions on soil physicochemical properties and enzyme activities. Significant differences among treatment means were determined using Duncan’s multiple range test at a significance level of P < 0.05. Data visualization, including boxplots and functional profiles, was generated using the “ggplot2” package in R software (version 4.3.1). Soil microbial community data were subjected to alpha diversity analysis, including Chao1 and Simpson indices, and beta diversity analysis based on non-metric multidimensional scaling (NMDS). Genus-level co-occurrence networks were constructed using the “psych” package in R (version 4.3), followed by visualization in Gephi (v0.9.7) and Cytoscape (v3.9.1) [27]. Random forest (RF) classification was performed to identify key microbial biomarkers that discriminate among treatments. In the model, treatment groups served as the response variable, while microbial genus abundances were used as predictors. Functional predictions of soil bacterial and fungal communities were performed using the FAPROTAX database (version 1.2.1) for bacteria and the FUNGuild database (http://www.funguild.org/) for fungi, respectively. Results Effects of Bt Application on Tea Soil Physicochemical Properties and Enzyme Activities As shown in Fig. 1, Bt application significantly improved the nutrient status of tea soils. Compared with the control group (CK), the Bt-treated groups (Bt1 and Bt2) exhibited increased soil pH values, indicating that Bt application alleviated soil acidification to some extent. Soil organic matter content increased with Bt application frequency, with the highest level observed in the Bt1 group, reaching 2.26%, which was significantly different from the CK (P < 0.05). Moreover, the levels of available soil nutrients peaked under Bt1 treatment, notably, available nitrogen and available potassium were significantly higher in Bt1 than in CK and Bt2 (P < 0.05), while available phosphorus in Bt1 was significantly higher than in Bt2 but did not differ significantly from CK. Soil physicochemical content and enzyme activities under different Bt treatments. CK: distilled water control; Bt1: 3 sprays, 21 d; Bt2: 6 sprays, 42 d. Asterisks indicate significant differences (*, P < 0.05, **, P < 0.01, ***, P < 0.001) As illustrated in Fig. 1, Bt application significantly influenced soil enzyme activities. Compared to CK, activities of acid protease (ACPT), urease (UE), sucrase (IE), polyphenol oxidase (PPO), and cellulase (CE) were significantly enhanced under Bt treatments (P < 0.05). Acid protease, sucrase, and cellulase activities reached their maximum levels in the Bt1 group, measured at 47.25 µg/g, 395.80 µg/g, and 310.4 µg/g, respectively. Conversely, urease and polyphenol oxidase activities peaked in the Bt2 group, with respective values of 232.57 µg/g and 35.25 µg/g. No significant changes were observed in catalase activity (CAT) across different Bt treatments. These results indicate that moderate-frequency conventional Bt application markedly improves soil physicochemical properties and enhances the content of available nutrients and multiple enzyme activities, suggesting a beneficial effect of Bt on improving soil quality. Effects of Bt Application on Tea Soil Microbial OTUs Venn diagram analysis was conducted to compare bacterial and fungal operational taxonomic units (OTUs) among different treatments. As illustrated in Fig. 2A, a total of 10,794 bacterial OTUs were detected across the three treatments, with 5,742 OTUs shared among all groups. The number of unique bacterial OTUs was 1,201, 1,096, and 1,213 for the CK, Bt1, and Bt2 treatments, respectively. Compared to CK, the total bacterial OTU number under Bt2 treatment showed no significant change, whereas Bt1 treatment resulted in a decrease in bacterial OTU richness. The rank order of bacterial OTU abundance across treatments was Bt2 > CK > Bt1. Venn diagram of soil microbial OTUs under different intensities of Bt application. CK: distilled water control; Bt1: 3 sprays, 21 d; Bt2: 6 sprays, 42 d As shown in Fig. 2B, a total of 1,922 fungal OTUs were identified, with 1,057 OTUs shared among all treatments. The unique fungal OTU counts for CK, Bt1, and Bt2 were 166, 100, and 195, respectively. Relative to CK, Bt2 treatment increased fungal OTU richness, while Bt1 treatment slightly reduced it. The abundance of fungal OTUs followed the order Bt2 > CK > Bt1. Effects of Bt Application on Tea Soil Microbial Diversity Alpha diversity indices are commonly used to evaluate the richness and diversity of microbial communities. Boxplots of bacterial richness (Chao1) and diversity (Simpson) indices (Fig. 3A) showed no significant differences in Chao1 between CK and both Bt1 and Bt2 treatments, whereas a significant difference was observed between Bt1 and Bt2. Notably, the Simpson index was significantly increased in the Bt2 treatment group (P < 0.05). Meanwhile, as shown in Fig. 3B, Bt application at both frequencies significantly increased the Simpson index of the fungal community (P < 0.05), while the fungal Chao1 index significantly decreased after Bt2 treatment (P < 0.05). Soil microbial community diversity and composition under different treatments. (A) and (B) show Chao1 and Simpson diversity indices for soil bacteria and fungi, respectively. (C) and (D) present NMDS analyses of bacterial and fungal community structures. (E) and (F) display the phylum-level composition of soil bacterial and fungal communities. CK: distilled water control; Bt1: 3 sprays, 21 d; Bt2: 6 sprays, 42 d. Asterisks indicate significant differences (* P < 0.05, ** P < 0.01, *** P < 0.001) Non-metric multidimensional scaling (NMDS) was further employed to assess the effects of different Bt application frequencies on soil microbial community structure at the OTU level (Fig. 3C and D). Clear separation between CK and Bt-treated groups was observed for both bacterial and fungal communities, indicating that Bt application significantly altered the composition of tea soil microbial communities. In summary, Bt treatment had a limited impact on microbial alpha diversity but significantly influenced the structure of bacterial and fungal communities. Taxonomic analysis at the phylum level (relative abundance > 1%) revealed that Proteobacteria, Chloroflexi, and Actinobacteria dominated the bacterial communities across CK, Bt1, and Bt2 groups (Figs. 3E). For the fungal community, Ascomycota and Basidiomycota were the predominant phyla, with Ascomycota showing a significant dominance in the CK group, but exhibiting a marked decline under Bt1 and Bt2 treatments (Fig. 3F). Symbiotic networks of soil bacteria (A) and fungi (B) under different intensities of Bt application treatments. Node size represents abundance, and edge thickness indicates correlation strength (Spearman’s r > 0.70, P < 0.01). Topological characteristics of bacterial (C) and fungal (D) symbiotic networks under varying Bt treatment intensities. CK: distilled water control; Bt1: 3 sprays, 21 d; Bt2: 6 sprays, 42 d Effects of Bt Application on Tea Soil Microbial Co-occurrence Networks To investigate whether Bt application alters the interactions within soil microbial communities, co-occurrence network analyses and topological feature calculations were performed on soil microbial communities under different Bt application frequencies, aiming to assess changes in microbial community complexity and connectivity. The co-occurrence network results showed that for bacteria (Fig. 4A), CK had 289 nodes, 2816 edges, and an average degree of 9.744. In the Bt1 treatment, the number of nodes remained stable at 289, while edges increased to 3024 and the average degree rose to 10.464. However, in the Bt2 treatment, nodes slightly decreased to 286, edges dropped to 2623, and the average degree declined to 9.171. For fungi (Fig. 4B), CK exhibited 297 nodes, 2716 edges, and an average degree of 9.145, in Bt1, nodes slightly decreased to 292, edges increased to 2980, and average degree increased to 9.959, whereas Bt2 showed a further reduction to 256 nodes, 2108 edges, and an average degree of 8.234. Topological feature analysis (Fig. 4C and D) revealed that the Bt1 treatment significantly increased betweenness centrality, closeness centrality, and node degree compared to CK, indicating greater complexity and connectivity in the microbial co-occurrence network under conventional Bt application. Conversely, the Bt2 treatment exhibited declines in these parameters, reflecting reduced network connectivity and stability. In summary, Bt application altered the structure of tea soil microbial co-occurrence networks, with conventional application (Bt1) promoting microbial community stability and complexity, while excessive application (Bt2) diminished connectivity and overall network stability. Effects of Bt Application on Tea Soil Microbial Community Structure A random forest classification model was employed to identify the key microbial taxa that most significantly contribute to the discrimination between treatments. The model identified 15 bacterial and 15 fungal genera as primary biomarkers based on their Mean Decrease Accuracy scores (Fig. 5A and C). Key microbial genera differing under various Bt application intensities identified by random forest regression analysis. (A) and (B) show the 15 key bacterial genera and their relative abundances, respectively. (C) and (D) present the 15 key fungal genera and their relative abundances, respectively. CK: distilled water control; Bt1: 3 sprays, 21 d; Bt2: 6 sprays, 42 d. Asterisks indicate significant differences (* P < 0.05, ** P < 0.01, *** P < 0.001) For the bacterial community, the most influential discriminators predominantly belonged to the phyla Proteobacteria and Nitrospirae. Compared with CK, the conventional Bt1 treatment significantly enriched several functional biomarkers, including of Pseudomonas (112.91%), Novosphingobium (51.86%), Bradyrhizobium (47.09%), and the nitrifying genus Nitrospira (26.66%) (Fig. 5B). In contrast, the intensive Bt2 treatment led to a distinct microbial signature, notably characterized by a 18.29% decrease in the relative abundance of Sphingomonas. Similarly, the RF model for fungal communities identified key taxa involved in treatment differentiation (Fig. 5C). Bt application consistently suppressed several potentially pathogenic biomarkers identified by the model. Specifically, in the Bt1 treatment, the relative abundances of Aspergillus, Thanatephorus, and Curvularia decreased by 69.06%, 86.26%, and 47.29%, respectively (Fig. 5D). These results demonstrate that while Bt application restructures the tea rhizosphere microbiome, the recruitment of beneficial bacteria and the suppression of pathogens are highly sensitive to the frequency of application. Effects of Bt Application on Tea Soil Bacterial Functional Metabolism and Fungal Trophic Modes Changes in microbial community structure often result in shifts in ecosystem functions. In this study, bacterial functional annotation under varying Bt application frequencies was performed using the FAPROTAX database. A total of 10,794 operational taxonomic units (OTUs) were identified, among which 5,742 OTUs (53.19%) were successfully annotated, covering 30 functional categories mainly related to carbon and nitrogen cycling as well as various degradation processes. Significant alterations in bacterial functional profiles were detected across different Bt treatments (P < 0.05), indicating statistically meaningful functional predictions. Key bacterial ecological functions included aromatic compound degradation, cellulose degradation, nitrogen fixation, nitrate reduction, nitrite reduction, and nitrogen oxide reduction. Results showed that Bt application markedly enhanced bacterial degradation-related functions (Fig. 6A), especially cellulose hydrolysis and aromatic compound degradation, with all Bt-treated groups exhibiting significant increases, and Bt1 achieving the highest proportions. Predicted functions of microbial communities under different treatments. (A) Bacterial community functions predicted by FAPROTAX. (B) Fungal trophic modes predicted by FUNGuild. CK: distilled water control; Bt1: 3 sprays, 21 d; Bt2: 6 sprays, 42 d. Different letters indicate significant differences (Duncan’s test, P < 0.05). Circle size and color represent relative abundance Moreover, Bt1 treatment significantly elevated the relative abundances of functions related to urea hydrolysis, nitrate respiration, nitrogen fixation, denitrification, nitrite denitrification, and nitrate denitrification (P < 0.05), reaching 79.67%, 52.15%, 78.09%, 2.05%, 99.31%, and 99.36%, respectively. Conversely, the Bt2 group exhibited decreases in nitrite respiration, nitrate reduction, and nitrification functions by 9.24%, 7.15%, and 3.87%, respectively, while other functions were slightly higher than CK but lower than Bt1. Fungal functional guild prediction based on the FUNGuild database (Fig. 6B) revealed that fungal trophic modes mainly comprised symbiotrophs, saprotrophs, and pathotrophs. The proportion of symbiotrophs in Bt1 was significantly higher than in CK and Bt2 (P < 0.05), whereas saprotrophs were enriched in both Bt1 and Bt2 relative to CK (P < 0.05). The abundance of pathotrophs was lower in Bt treatments than in CK. Further breakdown indicated that Bt1 significantly increased the proportions of ectomycorrhizal fungi, plant saprotrophs, dung saprotrophs, and wood saprotrophs. Notably, compared to CK, the relative abundances of plant pathogens were significantly reduced by 85.78% and 11.84% in Bt1 and Bt2 treatments, respectively. In summary, Bt application significantly influences key microbial metabolic functions in soil. Conventional Bt application notably enhances bacterial organic matter degradation functions, simultaneously, Bt1 treatment promotes the abundance of saprotrophic and ectomycorrhizal fungi while reducing plant pathogenic fungi, suggesting a beneficial restructuring of microbial functional profiles conducive to soil ecological health. Effects of Bt Application on Soil Environmental Variables and Microbial Correlations Redundancy analysis (RDA) was employed to elucidate the influence of soil environmental factors on bacterial and fungal communities (Fig. 7A and B). The first two RDA axes (RDA1 and RDA2) jointly explained 83.35% of the variance in bacterial community composition and 90.09% in fungal community composition. Except for catalase (CAT), all other soil variables significantly affected soil microbial communities. Soil pH, cellulase activity (CE), sucrase activity (IE), acid protease activity (ACPT), available phosphorus (AP), and urease activity (UE) exhibited significant positive correlations with each other and exerted a stronger influence on bacterial and fungal communities under Bt1 treatment. In contrast, fungal and bacterial communities under Bt2 treatment were more strongly associated with polyphenol oxidase (PPO) activity. Redundancy analysis (RDA) and correlation relationships between microbial communities and soil environmental factors. (A) and (B) show RDA of key bacterial and fungal communities with soil environmental factors, respectively. (C) and (D) present correlation analyses among key bacterial and fungal genera, microbial functions, and soil factors. CK: distilled water control; Bt1: 3 sprays, 21 d; Bt2: 6 sprays, 42 d. Red lines indicate positive correlations, while blue lines indicate negative correlations Building on these findings, correlations among microbial functional groups, key microbial genera, and physicochemical properties in tea soils were further analyzed (Fig. 7C and D). Among key bacterial genera, Pseudomonas was primarily involved in nitrate respiration and denitrification processes and showed a significant positive correlation with urease activity (UE). Bacillus, known for its organic matter degradation capacity, was positively correlated with soil organic matter (SOM) content, potentially accelerating C and N cycling. This suggests that the accumulation of organic matter and nitrogen transformation are driven by specific functional bacterial groups. Bradyrhizobium and Nitrosospira, which dominate nitrogen fixation and nitrification respectively, were positively correlated with increased available nitrogen (AN) content. Sphingomonas showed a positive correlation with aromatic compound degradation ability and was significantly associated with PPO activity. Among key fungal genera, Aspergillus exhibited negative correlations with wood saprotrophs, PPO activity, and available phosphorus (AP). Metarhizium was negatively correlated with plant pathogens, suggesting that it may enhance tea plant disease resistance through antagonistic effects. Additionally, soil organic matter content negatively correlated with Fusarium, indicating that organic matter improvement may suppress pathogenic fungi. Collectively, Bt application influenced soil microbial community structure by modulating soil physicochemical factors, with key microbial functions and metabolic activities closely correlated with environmental variables, revealing potential mechanisms by which Bt treatments optimize soil ecosystem functions. Discussion Bt application Enhances Soil Environment and Promotes Microbial Functional Strengthening The microbial pesticide Bacillus thuringiensis (Bt) is a cornerstone of current and future sustainable agriculture [9, 10]. Its widespread use has prompted extensive research into the environmental behavior of Bt insecticidal proteins and their complex interactions with soil microbial ecology [28]. Therefore, investigating the effects of different Bt application regimes on soil physicochemical properties and microbial diversity in tea plantation is essential for understanding its broader ecological footprint and guiding green farming practices. Our results indicated that Bt application influenced soil environmental conditions, notably by mitigating acidification. Compared with CK, Bt treatments (particularly Bt1) led to a significant increase in soil pH (Fig. 1). This neutralizing effect may be attributed to Bt stimulating tea plant roots to exude alkaline substances or enhancing the activity of alkali-producing nitrifying microorganisms [29, 30]. This is further supported by the enrichment of the nitrifying genus Nitrospira (Fig. 5B), which may contribute to the observed shifts in soil alkalinity. Interestingly, the pH increase was not intensity-dependent, as the more frequent Bt2 treatment did not yield a further rise in pH. This suggests that while Bt can alleviate acidification, the response may be constrained by the experimental duration or the limited buffering capacity of young tea plant root systems [24, 30]. Furthermore, the persistence of Bt cells and toxins introduced via sprays can vary significantly depending on soil characteristics, potentially leading to a threshold effect in ecological responses [31]. These findings highlight the complexity of soil-microbe-plant interactions under different Bt application intensities. Concurrently, the Bt1 treatment substantially increased the contents of available nitrogen and available potassium, and showed a slight but not necessarily significant increase in soil organic matter within the short experimental period. This suggests that moderate-frequency Bt application may promote nutrient mineralization and release by enhancing microbial carbon and nitrogen metabolism, such as cellulose degradation and urease activity [32, 33]. This inference is supported by significantly increased acid protease, sucrose, and cellulase activities, indicating that Bt application may modulate soil microbial community structure to activate functional decomposer groups, thereby improving soil nutrient availability and fertility [34]. Given the absence of organic matter input during the experiment, the observed short-term changes likely reflect enhanced nutrient cycling rather than a substantial accumulation of soil organic matter. Notably, although the Bt2 treatment exhibited the highest urease and polyphenol oxidase activities, its associated increase in available nutrients was lower than that of Bt1, implying that excessive Bt application may shift microbial metabolism toward defense mechanisms, consequently reducing nutrient utilization efficiency. The rapid shifts in soil chemical properties observed within the 42-day period can be attributed to the specific conditions of our pot experiment. Unlike open-field systems, the confined soil volume in pots tends to amplify the impact of external inputs and plant-soil feedback loops. Furthermore, our parallel research has demonstrated that foliar Bt application significantly enhances tea plant vigor by modulating the phyllosphere microbiome and activating the plant’s internal defense mechanisms [35]. It is well-established that healthier plants can rapidly alter soil nutrient cycling through modified rhizodeposition and root-microbe interactions. Therefore, the significant changes in organic matter and available nutrients likely reflect a synergistic effect of direct nutrient supplementation from the Bt crude extract and indirect biological feedback from the improved health of the tea plants. Bt application Restructures Soil Microbial Communities to Facilitate Nutrient Cycling in the Tea Rhizosphere As vital components of soil ecosystems, microbes decompose insoluble organic substrates into soluble forms, facilitating soil organic matter cycling and transformation, while the soil provides their habitat and supports complex symbiotic and competitive microbial interactions [30, 36]. Previous study has shown that increased bacterial diversity enhances microbial interspecies connectivity, thereby reinforcing soil ecosystem stability [37]. Our findings suggest that the soil microbial response to Bt is sensitive to the frequency of application. Although both treatments altered community structure, the significant loss of fungal richness specifically under Bt2 indicates that high-frequency application may impose a selective pressure that exceeds the tolerance of certain fungal taxa. In contrast, the bacterial community appeared more resilient to increased Bt inputs. This divergence suggests that while routine Bt application (Bt1) supports microbial diversity, intensive regimes (Bt2) may shift the community toward a state dominated by a few resilient taxa, potentially filtering out sensitive species [38, 39]. Despite the lack of significant changes in microbial diversity, the Bt1 treatment significantly increased the number of edges, average degree, and node centrality within bacterial co-occurrence networks (Fig. 3C and D), indicating strengthened bacterial interspecific associations and community stability. This phenomenon may result from adaptive microbial adjustments in symbiotic relationships to cope with shared survival demands and dispersal constraints [40, 41]. Additionally, Bt application may induce the release of specific metabolites from plant roots, ameliorating soil nutrient conditions and enhancing soil-plant interactions, which indirectly fosters microbial community symbioses [25]. Further analyses showed that the Bt1 treatment significantly enriched the relative abundances of Proteobacteria, Chloroflexi, and Acidobacteria within bacterial communities, and Basidiomycota within fungal communities (Fig. 3E and F). Previous research confirmed that Proteobacteria and Basidiomycota are key decomposers of soil organic matter, lignin, and cellulose, playing pivotal roles in soil nutrient cycling and plant growth [42–44]. Random forest modeling identified key bacterial genera that were specifically enriched under the Bt1 treatment (Fig. 5A and B). Among these, Sphingomonas is recognized for its high ecological adaptability and its specialized role in the degradation of aromatic compounds [45]. Furthermore, Pseudomonas and Novosphingobium were significantly promoted. These two genera often exhibit synergistic relationships where they efficiently decompose glucose and cellulose respectively, thereby enhancing carbon utilization through complementary metabolic pathways [46, 47]. In contrast, these beneficial functional enrichments were significantly attenuated under the intensive Bt2 treatment. For instance, the relative abundance of Sphingomonas decreased by 18.29% under Bt2, while the nitrifying genus Nitrospira showed no significant response. These results indicate that conventional Bt application (Bt1) effectively recruits metabolically diverse taxa, whereas excessive frequency (Bt2) may impose a physiological stress that inhibits these key functional groups. In the fungal community, Bt1 treatment significantly increased the abundances of Chaetosphaeria and Gibelulopsis (Fig. 5C and D). The symbiotic association between Chaetosphaeria and Trichoderma, both of which are capable of organic matter degradation and promoting plant growth, is well documented [48]. The enrichment of these key taxa implies that conventional Bt application fosters organic matter decomposition within soil carbon cycles, enhancing carbon sequestration and supplying richer nutrients that support tea plant growth. Functional prediction analyses revealed that Bt1 treatment significantly enhanced bacterial community functions related to the degradation of cellulose and aromatic organic compounds (Fig. 6A). Research has suggested that an increase in saprophytic fungi can indirectly boost fungal degradation of organic matter and autotoxic compounds in soil [49, 50]. Fungal community shifts mirrored this trend, with significant increases in relative abundance of saprophytic groups, including both endomycorrhizal and ectomycorrhizal fungi, under Bt1 treatment (Fig. 6B). Collectively, these data suggest that Bt application enhances organic matter degradation processes in soil microbes, thereby promoting soil carbon cycling. Moreover, Burkholderia and Bradyrhizobium were significantly enriched under Bt1 treatment. Burkholderia is recognized as a plant growth-promoting bacterium, commonly harboring the nifH gene essential for biological nitrogen fixation [51, 52]. Bradyrhizobium is well known for its nitrogen-fixing capabilities and close symbiotic associations with plant roots, containing relevant nitrogen fixation gene clusters [53]. Regarding nitrogen-cycling microbes, Nitrospira, a key nitrifying genus that facilitates nitrogen transformation, showed increased abundance under Bt1 treatment [54]. The enrichment of these functional keystone taxa suggests that conventional Bt application promotes beneficial microbial populations involved in nutrient cycling, thereby enhancing soil nutrient turnover. Redundancy analysis (RDA) further identified soil pH, urease activity, and organic matter as principal drivers of bacterial community restructuring under Bt1, while polyphenol oxidase activity and available phosphorus predominantly influenced fungal community variability (Fig. 7C and D), consistent with findings reported by Kunito et al. [55]. In bacterial communities, Bradyrhizobium, Nitrospira, Pseudomonas, and Sphingomonas exhibited strong positive correlations (r > 0.8) with urease activity, available nitrogen, and organic matter, indicating that the enrichment of these beneficial taxa likely mediates improvements in soil nutrient content. Bt application Suppresses Soil-Borne Fungal Pathogens Under varying Bt application frequencies, significant alterations in the abundance of certain soil-borne fungal pathogens were observed. Pathogens such as Curvularia, Aspergillus, and Thanatephorus were markedly reduced in Bt-treated soils (Fig. 5C and D). Curvularia species are known to infect diverse plants, causing root rot, flower rot, and leaf blight diseases [56]. Aspergillus, a ubiquitous fungal genus, can induce spoilage and harm plants [57]. Meanwhile, Metarhizium, a commonly distributed entomopathogenic fungus, effectively suppresses insect infestations [58]. The enrichment of Burkholderia may contribute to pathogen suppression through antibiotic secretion or induction of systemic resistance, which aligns with previous findings showing that Burkholderia colonization in the tea plant rhizosphere, especially under infestation by Ectropis grisescens, enhances beneficial microbial associations and modulates tea seedling metabolic pathways related to defense. Inoculation with Burkholderia strains has been demonstrated to promote soil nitrogen cycling functions and stimulate the production of defensive secondary metabolites in tea leaves, highlighting its promising role in integrated pest and disease management within tea plantations [59, 60]. Functional prediction further indicated a significant reduction in plant pathogen-related fungal taxa following Bt1 application (Fig. 6B). Correlation analyses revealed significant negative relationships between key pathogenic fungi such as Aspergillus, Fusarium, and Curvularia and soil enzyme activities. Previous studies have recognized that soil ecological imbalance and pathogen accumulation are primary factors driving soil-borne disease outbreaks [61, 62]. Bt, as an insecticidal bacterium, has been widely reported to enhance tea orchard ecosystem stability [8, 63]. Our findings that conventional Bt application reduces pathogen abundance suggest an inhibitory effect on soil-borne pathogens, thereby potentially enhancing tea plant disease resistance. Furthermore, the restructuring of the soil microbial community observed in this study suggests an integrated response of the tea plantation ecosystem to Bt application. Our previous findings showed that foliar Bt application promotes the enrichment of beneficial microbes on tea leaves and boosts leaf flavonoid biosynthesis [35]. This improved aboveground physiological status potentially serves as a driver for belowground changes. By enhancing plant defense and nutrient assimilation, Bt application likely stimulates the roots to recruit beneficial soil taxa, such as Pseudomonas and Bradyrhizobium, which we found to be enriched in the rhizosphere. Consequently, the central mechanism for Bt’s positive impact goes beyond simple pest control; it involves a sophisticated coordination between the phyllosphere and rhizosphere, ultimately promoting soil ecological functions and plant-soil health. Limitations and Future Perspectives While this study offers valuable insights, several limitations warrant consideration. First, regarding the potential “nutrient effect”, although Bt was applied via foliar spraying, the crude extract contains residual fermentation media and organic components [64]. These substances could indirectly influence the soil through plant-mediated pathways, such as altered rhizodeposition or phyllosphere-rhizosphere signaling. Therefore, the observed shifts in soil nutrients likely reflect a combination of Bt’s specific biological activity and an indirect physiological response of the tea plant to the organic components in the extract [65]. Second, in terms of experimental monitoring, we did not quantify the persistence of B. thuringiensis cells or their toxins on the leaf surface or their eventual movement into the soil system. The longevity of these components is critical for interpreting long-term ecological impacts, as the persistence of introduced Bt can vary significantly depending on the environment [28, 31]. Without data on the degradation rates of Bt toxins, it remains challenging to fully decouple the specific biological effects of Bt from the plant’s general systemic response to foliar treatments. Finally, the experimental design for Bt1 and Bt2 was intended to align with practical management, which coupled application frequency with total duration. Future research should utilize inactivated crude extracts as foliar controls to effectively isolate the biological impacts of Bt from the indirect effects of the fermentation carrier on plant-soil feedback loops. Conclusion Bt application at different frequencies effectively improves tea soil conditions by alleviating soil acidification. The moderate-frequency conventional Bt application (Bt1) significantly increased soil pH, organic matter, available nitrogen and available potassium, as well as activated soil enzyme activities. Although Bt2 showed the highest urease and polyphenol oxidase activities, nutrient accumulation was lower than in Bt1. Bt application significantly altered soil microbial community structure and co-occurrence networks, enriching key functional taxa that enhance organic matter degradation and nutrient cycling, thereby improving soil ecology. Bt1 application also increased saprotrophic fungi and ectomycorrhizal abundance while suppressing pathogenic fungi, enhancing soil disease resistance. Overall, appropriate Bt use optimizes soil microbial structure and function, promoting soil stability and tea plant growth. Future studies should explore the long-term ecological effects of Bt and its interactions with plant roots, apply multi-omics to reveal metabolic networks of key functional microbes, and evaluate Bt’s efficacy across diverse soils and crops to support its sustainable application. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 (XLSX 9.77 KB) Author Contributions Y.X. and H.L.: Conceptualization, Visualization, Methodology, Writing—original draft, Formal analysis, Writing—review and editing. D.L. and W.X.: Formal analysis, Writing—review and editing. Z.W., J.W., W.C. and X.F.: Methodology, Investigation, Writing—original draft. C.N., X.D., C.Y., and Y.L.: Methodology, Investigation. P.C. and Y.H. : Conceptualization, Visualization, Methodology, Writing—original draft, Formal analysis, Writing—review & editing, Funding acquisition. All authors have read and agreed to the published version of the manuscript. Funding This research was funded by the Fujian Province Young and Middle-Aged Research and Education Projects (JZ240071, JAT241155), the Project of Fujian Provincial Department of Science and Technology (2025N0013, 2025FJSYCXM03), the Nanping Academy of Resource Industrialization Chemistry Project (N2023Z005, N2023Z007, N2024Z009, N2025K019), and Wuyi University Talent Introduction Project (YJ202307, YJ202432, YJ202433). Data Availability The data presented in this study are available upon request from the corresponding author. Declarations Competing Interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Pumo Cai and Yulin Xiong contributed equally to this work. Contributor Information Qisong Li, Email: liqisong@wuyiu.edu.cn. Yongcong Hong, Email: wyxyhyc@wuyiu.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1 (XLSX 9.77 KB) Data Availability Statement The data presented in this study are available upon request from the corresponding author.