Application of Bioorganic Fertilizer on Panax notoginseng Improves Plant Growth by Altering the Rhizosphere Microbiome Structure and Metabolism

Bioorganic fertilizers can alleviate (a) biotic stresses and sustainably increase crop yields. The effect of bioorganic fertilizers on the rhizosphere bacterial community of Panax notoginseng and soil metabolism remains unknown. Here, we tracked the changes in the soil physicochemical properties, bacterial microbiota responses, and soil metabolic functions after the addition of a bioorganic fertilizer in a P. notoginseng field. The application of a bioorganic fertilizer reduced the soil acidification, improved the organic matter, and increased the contents of the total/available soil nutrients. Soil amendment with a bioorganic fertilizer significantly affected the structure of the rhizosphere bacterial community, leading to the enrichment of specific bacterial consortia such as Rhodanobacter, Arthrobacter, Sphingomonas, Devosia, Pseudolabrys, Luteimonas, Lysobacter, Nitrosospira, and Nakamurella. Previously, many of these genera have been associated with nutrient cycling, plant productivity, and disease suppression. Metabolome analysis further highlighted that the bioorganic fertilizer treatment significantly reduced phenolic acids and flavonoids and enhanced organic acids, saccharides and alcohols, and amino acids. This result indicates a high survival of bacterial microbiota in the rhizosphere and an availability of nutrients for P. notoginseng growth. This work showed that the application of bioorganic fertilizers significantly improves soil health status, alters soil metabolic functions, and stimulates a specific subset of rhizosphere microbiota for nutrient cycling and disease protection in P. notoginseng.


Introduction
Soil microorganisms are continuously engaged in several ecosystem services, including decomposition of organic matter, nutrient cycling, plant growth, and disease protection [1][2][3][4]. At a system level, a stable and specific soil microbiome structure predominately determines the agricultural land's positive ecological functions and productivity [5][6][7]. The coevolutionary dynamics between members of soil microbiota can be disrupted by agricultural management strategies such as crop rotation, pesticide use, tillage, and chemical communities. By decrypting the mode of action of the applied bioorganic fertilizer, we sought to determine how bioorganic fertilizers positively affect the Sanqi ginseng growth by modulating the rhizosphere microbiome and metabolites.

Study Site and Sample Collection
The field site is located at Kunming Xundian Undergrowth Planting Base, Yunnan Province, China (22 • 37 N, 20 • 93 E) at an altitude of 1764 m. The region has a subtropical climate with an annual precipitation of 950-1000 mm and an annual temperature of 18-24 • C. In 2017, a bioorganic fertilizer (Yuesheng brand) obtained from Shanghai Luyuan Three Elements Biological Technology Co., Ltd., Kumming, China, was initially applied to the field at a rate of 3000 kg/hectare. Then, P. notoginseng was transplanted into the fertilized soil. The bioorganic fertilizer was used again in 2018 as topdressing on the surface of the planted soil at a rate of 1200 kg/hectare. We ensured that no fertilizer residue was left on the surface of plant leaves. In the same field, a portion without application of the bioorganic fertilizer served as control. The experimental design was a split plot with three replications and two treatments. After two years of P. notoginseng plantation, soil samples were collected. We pooled five P. notoginseng plants from plot together for the collection of the rhizosphere soil samples. A total of three rhizosphere samples were collected for the bioorganic fertilizer treatment (n = 3) and control treatment (n = 3) to analyze bacterial communities and metabolites. Bulked soil samples collected around the P. notoginseng plants were analyzed for physicochemical properties.

Soil Physicochemical Analysis
The physicochemical properties of the soil samples were quantified as previously reported [32]. Soil pH was analyzed using an FE-20 pH meter (Swiss Mettler). The organic matter (OM) contents were analyzed using the dichromate chemical oxygen demand test. Total nitrogen (TN), total phosphorus (TP), and total potassium (TK) were measured after soil being treated by an H 2 SO 4 -H 2 O 2 mixture. An autoAnalyser3 (Bran + Luebbe, Hamburg, Germany) was used to determine TN and TP, while TK was measured by a flame atomic spectrophotometry. Alkali-hydrolyzed nitrogen (AN), available phosphorus (AP), and available potassium (AK) were analyzed by the diffusion method, the Olsen method, and the ammonium acetate extraction flame photometry method, respectively.

Rhizosphere Soil Collection
Rhizosphere soils of P. notoginseng grown in fertilized and non-fertilized fields were collected using the standard protocol. Briefly, P. notoginseng roots were gently shaken to remove bulk soil, and then the roots were transferred to 50 mL Falcon tubes with 25 mL of sterile Silwet L-77 amended PBS buffer. Falcon tubes were continuously rotated on a shaking platform for 20 min at 180 rpm to separate closely adhered soil from the root. The roots were carefully removed after the rhizosphere soil settled down, and the washing buffer was centrifuged at 10,000× g for 20 min. The supernatant was discarded, and the resulted rhizosphere soil samples were used for microbiome and metabolites analysis.

Rhizosphere Microbiome Analysis
Rhizosphere soils DNA extracted using the FastDNA Spin Kit for Soil (MP Biomedicals) by following the manufacturer's guideline. Samples were homogenized in the FastPrep instrument for 40 s at a speed setting of 6.0. The DNA was eluted in 50 µL of elution buffer, and the PCR amplification of the V3-V4 region of 16S rRNA gene carried out using primer pair 338F (5 -ACTCCTACGGGAGGCAGCAG-3 ) and 806R (5 -GGACTACHVGGGTWTCTAAT-3 ) [33]. PCR amplicons were purified using AMPure XT beads (Beckman Coulter Genomics, Danvers, MA, USA) and quantified using Qubit (Invitrogen, Waltham, MA, USA). Finally, the paired-end sequencing of the bacterial amplicons was performed on the Illumina NovaSeq PE250 platform. Raw bacterial reads were first quality-trimmed using Trimmomatic and then assigned to samples based on barcodes. Chimeric sequences were identified using Vsearch software, and sequences characterized as chimeric were removed. Bacterial sequences were binned into operational taxonomic units (OTUs) at ≥97% similarity level through open-reference OTU picking protocol in the UPARSE-pipeline. The most abundant sequences from each OTU were taken as representative sequences for the respective OTU. Taxonomic configuration of OTUs performed using the Silva database.

Metabolites Profiling from Rhizosphere Samples
The soil samples were homogenized in a mixer mill (MM 400, Retsch, Hann, Germany) with a zirconia bead for 1.5 min at 30 Hz. A 100 mg sample from each replicate was weighted and extracted overnight at 4 • C with 1.2 mL 70% aqueous methanol. The extracts were filtrated after centrifugation at 12,000 rpm for 10 min for ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) analysis and analyzed using a UPLC-ESI-MS/MS system.

Statistical Analysis and Data Visualization
We used QIIME to calculate the alpha diversity, including the observed species and Shannon diversity index, and visualized the values in the boxplot using R-3.5.3. The weighted unifrac distance for beta diversity analysis was also calculated using QIIME, and performed the Principal Coordinate Analysis (PCoA) to get principal coordinates and visualize multidimensional data in R package ggplot2. Weighted Unifrac distance data was also used to perform UPGMA clustering of fertilized and non-fertilized soil samples. The relative abundances of the bacteria at different taxonomic levels were calculated based on the classified OTU reads and were subsequently plotted in R with the package ggplot2. Differences between bioorganic fertilizer and non-fertilizer treatments were calculated using a Welch's t-test.
Identified metabolites were annotated using the Kyoto Encyclopedia of Genes and Genomes (KEGG) Compound database (http://www.kegg.jp/kegg/compound/, accessed on 10 August 2020), annotated metabolites then mapped to KEGG Pathway database (http://www.kegg.jp/kegg/pathway.html, accessed on 10 August 2020). Pathways with significantly regulated metabolites mapped to were then fed into metabolite sets enrichment analysis; their significance was determined by hypergeometric test's p-values. The hierarchical cluster analysis (HCA) results of samples and metabolites were presented as heatmaps with dendrograms. In contrast, Pearson correlation coefficients (PCC) between samples were calculated by the cor function in R and presented as only heatmaps. Both HCA and PCC carried out by R package pheatmap. Significantly regulated metabolites between groups were determined by variable importance in projection, VIP ≥ 1 and absolute Log2 FC (fold change) ≥1. VIP values were extracted from the Orthogonal partial least squares discriminant analysis (OPLS-DA) result, which contains score plots and permutation plots, and was generated using R package MetaboAnalystR. The data was log transform (log2) and mean centering before OPLS-DA. To avoid over fitting, a permutation test (200 permutations) was performed. Spearman's correlation index values among the specific microbial taxa and metabolites were calculated using the "psych" package in R 3.4.0, and we removed the correlations with a Spearman's coefficient < 0.7 and p > 0.05.

Impact of a Bioorganic Fertilizer Application on Sanqi Ginseng Growth and Soil Physicochemical Properties
After two years of P. notoginseng cultivation, the bioorganic fertilizer treatment produced taller plants than the untreated control ( Figure 1A). Plant fresh weight (FW) and dry weight (DW) were significantly increased in plots that received the bioorganic fertilizer (FW, 8.85 ± 0.12 g; DW, 2.95 ± 0.12 g) compared to non-fertilizer (FW, 6.39 ± 0.28 g; DW, 2.05 ± 0.16 g) treatment ( Figure 1B). Notably, the soil pH was alleviated to 6.04 ± 0.030 from 5.64 ± 0.020, and the soil moisture was reduced to 17.22 ± 0.001 from 26.42 ± 0.002 by the addition of the bioorganic fertilizer ( Figure 1C). The soil total organic matter (TOM) is an essential indicator of soil health. Compared with plots that received no fertilization, the TOM contents significantly increased in soils after addition of the bioorganic fertilizer. Similarly, the TN, AN, TP, AP, TK and AK contents were also improved in soils amended with the bioorganic fertilizer. Overall, the bioorganic fertilizer improved Sanqi ginseng plant growth, alleviated soil acidification and increased essential nutrient contents in the soil.

The Bioorganic Fertilizer Alters P. notoginseng Rhizosphere Bacterial α-and β-Diversity
The impacts of the bioorganic fertilizer on the rhizosphere bacterial community richness and diversity are shown in Figure 2. The boxplots based on the observed species showed that the addition of the bioorganic fertilizer reduced the number of bacterial taxa in the rhizosphere, but the values were not different compared to the mock treatment. In contrast, the Shannon diversity values were significantly decreased for the rhizosphere bacterial community of Sanqi ginseng grown in soils amended with the bioorganic fertilizer relative to the non-fertilizer treatment. This observation points out that the bioorganic fertilizer alters the α-diversity of the rhizosphere microbiome ( Figure 2A). Among the total bacterial OTUs detected from both treatments, 633 OTUs were unique to the bioorganic fertilizer treatment, and 749 OTUs were unique to the non-fertilizer treatment. A total of 2663 OTUs were shared between the two groups ( Figure 2B). Next, we performed a principal coordinate analysis (PCoA) based on the weighted UniFrac distance to observe changes in the rhizosphere bacterial community structure ( Figure 3A). The bacterial community inhabiting the rhizosphere of P. notoginseng grown in the bioorganic fertilizer amended soils was clearly separated from the control treatment. Both treatments separated along the axis 1, and the first coordinate of PCoA 1 explained a maximum variation of 88.37% in the bacterial β-diversity. The observed differences in β-diversity were mainly explained by the change in the proportion of dominant bacterial phyla inhabiting the rhizosphere ( Figure 3B). These results indicate that the application of the bioorganic fertilizer had a significant effect on the rhizosphere community.

Effect of a Bioorganic Fertilizer Application on Rhizosphere Bacterial Community Composition
A total of 386 bacterial genera belonging to 38 phyla were identified in the rhizosphere of P. notoginseng. Among them, only eight phyla and 11 genera have relative abundances greater than 1%. At the phylum level, Proteobacteria (51.3%), Acidobacteria (18.7%), and Actinobacteria (15.6%) dominated the rhizosphere bacterial community. Specifically, the phyla Proteobacteria, Acidobacteria, Actinobacteria, Chloroflexi, and Verrucomicrobia were found to be significantly different in relative abundance between the bioorganic fertilizer and control treatments ( Figure 4).  . Relative abundance of bacterial phyla found to be significantly different between the bioorganic fertilizer and no-fertilizer treatments. The bacterial phyla composition was considered significantly different between treatments when the p-value was <0.05.
The relative abundance of Proteobacteria and Actinobacteria increased, but Acidobacteria, Chloroflexi and Verrucomicrobia decreased in abundance with the bioorganic fertilizer application. At the family level, the relative abundance of Rhodanobacteraceae, Micrococcaceae, Sphingomonadaceae, and Micropepsaceae increased in the bioorganic fertilizer treatment ( Figure S1). A phylogenetic evolutionary tree was constructed to represent the dominant phyla and genera ( Figure 5). Covering the phylum to genus level, the genera that significantly changed in abundance after adding the bioorganic fertilizer were identified using the Welch's t-test. The relative abundance of several bacterial genera including, Rhodanobacter, Arthrobacter, Sphingomonas, Devosia, Pseudolabrys, Luteimonas, Lysobacter, Nitrosospira and Nakamurella increased in the rhizosphere of Sanqi ginseng grown in soils amended with the bioorganic fertilizer. In contrast, the abundance of bacterial genera Bradyrhizobium, Bryobacter, Massilia, Solibacter, and Udaeobacter increased in the non-fertilizer treatment ( Figure 6). Most of the bacterial genera abundant in the bioorganic fertilizer treatment were also positively correlated with the soil physicochemical properties ( Figure S2). Figure 6. Extended error bar plot indicates the differences in bacterial genera relative abundance in the bioorganic fertilizer and no-fertilizer treatments. Only significantly different genera between the two groups were depicted (Welch's t-test, p < 0.05).

Changes in Soil Metabolites after Addition of the Bioorganic Fertilizer
Soil metabolomics was performed in order to determine the bacterial microbiota responses to the application of the bioorganic fertilizer. Of the 664 metabolites detected by the UPLC-ESI-MS/MS, 175 metabolites were found to be differentially accumulated, and among them, 59 metabolites were enriched in soils amended with the bioorganic fertilizer ( Figure 7A). A heatmap was constructed to show the differences in the abundance of various metabolites assigned to several classes. Metabolites mainly belong to the classes of phenolic acids, lignans and coumarins, amino acids and derivatives, lipids, nucleotides and derivatives, organic acids, flavonoids, alkaloids, and terpenoids ( Figure 7B). Many compounds in the class of phenolic acids, flavonoids, lipids and alkaloids were significantly enriched in the non-fertilizer treatment and decreased in composition after applying the bioorganic fertilizer.  Figure S3). These differentially abundant metabolites were further found to positively and negatively correlate with the soil physicochemical properties in the bioorganic fertilizer and non-fertilizer treatments, respectively ( Figure S4).

Correlations between the Bacterial Community and Soil Metabolism
Soil microorganisms are one of the crucial drivers of the distribution of metabolites. We performed a Spearman's correlation and constructed a co-occurrence network and heatmap to show the relationship between the differential metabolites and significantly impacted bacterial taxa (Figures 8 and S5).

Figure 8.
Heatmap based on Spearman's correlation index values between differential microbial taxa and metabolites responding to bioorganic fertilizer and no fertilizer treatments in the rhizosphere. Correlation were considered significant when the p-value < 0.05 (* p < 0.05, ** p < 0.01).

Discussion
Excessive uses of chemical fertilizers and pesticides have caused a deterioration of soil health and microbiome communities. Hence, replacing the inorganic with bioorganic fertilizers is a prerequisite for sustainable agriculture [13]. Previous studies have shown that applying organic fertilizers improves soil and plant health and contributes to high crop yield [12]. In this study, we investigated the effect of a bioorganic fertilizer on the soil physicochemical characteristics, soil metabolites, Sanqi ginseng growth, and rhizosphere microbiome. Compared to the no-fertilizer treatment, the bioorganic fertilizer application improved Sanqi ginseng plant fresh weight, dry weight, shoot length, and root length. Soil organic matter, total and alkali-hydrolyzednitrogen, and total and available phosphorus were also higher in soil amended with bioorganic fertilizer. Generally, the increased nutrient level in soils has been linked with improved plant performance. High amounts of available nitrogen, phosphorus and potassium in soil improve crop quality and yield [34]. It has been previously observed that the amount of organic matter in soil is directly proportional to the yield production [35]. Moreover, the addition of inorganic fertilizers does not affect total nitrogen content, but the organic fertilizers are confirmed to improve soil nutrients for maintaining stable yields [36][37][38]. Amendment of excessive inorganic nitrogen fertilizers is also hypothesized to cause soil acidification because of soil nitrification [39][40][41]. In our study, the application of a bioorganic fertilizer altered soil pH from 5.64 to 6.04, thus reduced the soil acidification. Previously, Zhang et al. [42] demonstrated that applying a high proportion of organic fertilizers relative to a low proportion alleviates soil acidification. These results highlight that the soil amendment of organic fertilizers improves soil nutrient and organic matter and reduces soil acidification associated with high plant performance.
Microorganisms are an important component of soil ecosystems that is directly associated with plant health [43]. Organic and inorganic amendments significantly affect the soil microbiome structure and functions. In our study, the addition of a bioorganic fertilizer affected the bacterial diversity. Shannon diversity decreased, and bacterial community composition shifted in the bioorganic fertilizer treatment compared to non-fertilizer treatment. However, the species richness was not significantly different between both treatments. A decrease in bacterial Shannon diversity might be due to plant response to changes in soil environmental conditions caused by bioorganic fertilization leading to the enrichment of a specific subset of functional microbiota in the rhizosphere. Application of the bioorganic fertilizer enriched bacterial phyla Proteobacteria and Actinobacteria and decreased the abundance of phylum Acidobacteria. Some of the members of Proteobacteria and Actinobacteria are copiotrophic bacteria [44], while some members of Acidobacteria are fastidious oligotrophic bacteria [44]. The increased relative abundance of some of the members of Proteobacteria in the rhizosphere has been positively correlated with the increased soil nutrient level [45]. Several bacterial species within phylum Proteobacteria dominate the nutrient-rich environment and play a key role in C and N cycling [46]. Similarly, increased availability of C and N is also known to induce the abundance of some of the members of Actinobacteria in the soil [47]. For Acidobacteria, the low soil pH increased its relative abundance in the soil as some bacterial taxa in this phylum are often negatively correlated with the soil pH [48,49]. This phenomenon highlights the general life-history strategies that addition of nutrients favoring the fast growing and copiotrophic bacteria [50]. Overall, these results suggest that the addition of a bioorganic fertilizer is likely to favor the growth of some copiotrophic bacteria (e.g., some bacteria within phylum Proteobacteria and Actinobacteria) over some oligotrophic bacteria (e.g., some bacteria within phylum Acidobacteria) because of their ability to live in nutrient-sufficient or nutrient-limited environments, respectively.
Soils from the bioorganic fertilization treatment are enriched in several bacterial genera, including Rhodanobacter, Arthrobacter, Sphingomonas, Devosia, Pseudolabrys, Luteimonas, Lysobacter, Nitrosospira and Nakamurella. Bacterial species in the genus Rhodanobacter are Gram-negative, rod-shaped and aerobic, mainly catalase-and oxidase-positive [51]. Rhodanobacter was previously positively correlated with the soil pH and negatively with the soil NO 3 − N [52]. In this study, Rhodanobacter was also positively correlated with the soil pH and other soil chemical properties. Interestingly, Rhodanobacter and Lysobacter were also found as antagonistic to fungal pathogens in previous reports [53,54]. Lysobacter was not significantly correlated with soil properties in the present study and, therefore, might be involved in functions for the plants other than nutrient acquisition. The abundance of some Arthrobacter species was increased with the addition of the bioorganic fertilizer [55] and it has been found that they promote plant growth [56] and also restrain pathogenic bacteria and fungi [57]. Sphingomonas is ubiquitous in natural habitats and has been reported to be involved in disease suppression [58][59][60]. Devosia sp. was previously found to form a unique nitrogen-fixing foot-nodule symbiosis with the aquatic legume Neptunia natans (Lf) Druce [61] and has been correlated with nitrogen in the soil [62]. With reference to these studies, we deduce that Devosia sp. increase available nitrogen contents in soil. The genus Pseudolabrys was previously reported to increase in relative abundance after NK treatment [63]. Luteimonas species are known for their catalytic activities related to oxidase, catalase, alkaline phosphatase, esterase, and esterase lipase and are involved in the biodegradation metabolism of the organic matter [64][65][66]. The enrichment of Luteimonas in soil is a valuable indicator of soil amelioration [67]. The relative abundance of Nitrosospira was positively correlated with nitrification activity following long-term inorganic and organic fertilization [68]. Most of the bacterial genera enriched in the rhizosphere of P. notoginseng due to the addition of the bioorganic fertilizer were positively correlated with soil properties such as pH, TOM, TN, AN, TP, AP, TK, and AK. In contrast, P. notoginseng grown in soil without fertilizer was enriched for bacterial genera Bradyrhizobium, Bryobacter, Massilia, Solibacter, Udaeobacter, and Acidibacter. Bradyrhizobium spp. are agriculturally important because they can form root nodules and be involved in nitrogen fixation.
Soil metabolites mainly source from microbial metabolites, soil organic matter, plant decomposition, and root exudates [69]. In rhizosphere metabolomics, differentiating the metabolites from plants and microorganisms is still a big challenge [70]. Soil microorganisms mainly control most of the reactions in the N cycle [71][72][73]; therefore, an intense interaction may occur between bacterial microbiota after the addition of bioorganic fertilizers. In this study, the addition of a bioorganic fertilizer altered the metabolomic profile of rhizosphere soils and decreased the composition of many compounds belonging to the classes of phenolic acids, flavonoids, lipids, and alkaloids. We observed that the p-Coumaric acid was the predominantly enriched metabolite in the non-fertilizer treatment, and its abundance decreased after adding the bioorganic fertilizer. Previous studies have demonstrated that accumulation of phenolic acids in the soil inhibits seedling growth [74], especially p-Coumaric acid, which may inhibit plant growth [75]. Furthermore, we found a significant and positive correlation between the genus Massilia and 5-O-p-Coumaroylquinic acid, 3-O-p-Coumaroylquinic acid, and caffeic acid in the non-fertilizer treatment. Previous studies found that the presence of pathogens increased the release of caffeic acid and that the caffeic acid and infected plant exudates have reasonably similar effects on microbial community composition [76]. The enrichment of caffeic acid indicates that the Sanqi ginseng plants in the non-fertilizer treatment were probably under attack by the soil-borne pathogens. Flavonoids contents are found to be related to available nitrogen [77]. Plants accumulate more flavonoids under limiting nitrogen availability compared to those that are well supplied [78]. The metabolites correlated with enriched genera for the bioorganic fertilizer effect included those from the class of organic acids, saccharides and alcohols, and amino acids and derivatives. As organic acids, saccharides and alcohols, and amino acids were enriched in the bioorganic fertilizer treatment; therefore, we focused on deciphering their ecological role in plant-microbe interactions. Organic acids act as chemoattractant signals to microbes and are important nutrients and chelators of poorly soluble mineral nutrients [79][80][81]. Similarly, sugars and sugar alcohols have been shown to act as chemotaxis substances for a range of microbiota in the rhizosphere [79,82]. Zhu et al. [83] reported an increased secretion of sugars and sugar alcohols in the rhizosphere when nitrogen was sufficiently available to the plants. The high composition of amino acids in soil amended with the bioorganic fertilizer indicates that the Sanqi ginseng plants had sufficient nitrogen for growth. It has been documented that plants release the low amount of amino acids when subjected to nitrogen deficiency [84,85]. The above results pointed out the role of bioorganic fertilizers in nutrient availability, disease protection, soil microbial ecology, and crop productivity.

Conclusions
This study demonstrates that applying bioorganic fertilizers improves the Sanqi ginseng plant growth. We highlight the importance of the rhizosphere bacterial microbiota and their combined functions for nutrient availability and disease protection. The application of the bioorganic fertilizer significantly altered the rhizosphere microbiome and stimulated specific plant-beneficial bacterial consortia, which are involved in nutrient cycling and pathogen suppression. Moreover, the soil bioorganic fertilizer's addition regulated the correlation between rhizosphere bacterial microbiota and soil metabolism, affecting plant rhizosphere microecology. We recommend using bioorganic fertilizers in Sanqi ginseng production, considering the initial soil nutrient level to ensure a sufficient optimum supply of essential nutrients to plants and stimulate indigenous soil microbiota for natural disease suppression.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms10020275/s1, Figure S1: Relative abundance of bacterial families found significantly different between bioorganic fertilizer and no-fertilizer treatments. The bacterial family composition was considered statistically significant differences when p value was less than 0.05; Figure S2: Pearson correlation analysis between significantly enriched bacterial genera (abundant in bioorganic and non-fertilizer treatment) and soil physicochemical properties. The non-significant correlations is shown with cross mark; Figure S3: Differentially abundant top fold change metabolic compounds in bioorganic (red bar) and no-fertilizer (green bar) treatments; Figure S4: Pearson correlation analysis between differentially abundant top fold change metabolic compounds (abundant in bioorganic and no-fertilizer treatments) and soil physicochemical properties. The non-significant correlations is shown with cross mark; Figure S5: Co-occurrence network between differentially abundant bacterial genera and differential metabolites. The red line segment represents positive correlation. The green line segment represents negative correlation. The size of circle represents relative abundance or metabolite expression of bacterial genera.

Data Availability Statement:
The raw sequencing data has been submitted to NCBI SRA PR-JNA752500. All remaining data used in this manuscript are available in the text and additional files.

Conflicts of Interest:
The authors declare no conflict of interest.