Effects of Different Straw Biochar Combined With Microbial Inoculants on Soil Environment of Ginseng

Ginseng is an important cash crop. The long-term continuous cropping of ginseng causes the imbalance of soil environment and the exacerbation of soil-borne diseases, which affects the healthy development of ginseng industry. In this study, ginseng continuous cropping soil was treated with microbial inocula using broad-spectrum biocontrol microbial strain Frankia F1. Wheat straw, rice straw and corn straw were the best carrier materials for microbial inoculum. After treatment with microbial inoculum prepared with corn stalk biochar, the soil pH value, organic matter, total nitrogen, available nitrogen, available phosphorus, and available potassium were increased by 11.18%, 55.43%, 33.07%, 26.70%, 16.40%, and 9.10%, the activities of soil urease, catalase and sucrase increased by 52.73%, 16.80% and 43.80%, respectively. A Metagenomic showed that after the application of microbial inoculum prepared fromwith corn stalk biochar, soil microbial OTUs, Chao1 index, Shannon index, and Simpson index increased by 19.86%, 16.05%, 28.83%, and 3.16%, respectively. Three classes (Alphaproteobacteria, Gammaproteobacteria and Sphingobacteria) were the dominant bacteria in ginseng soil, and their abundance increased by 7.87%, 9.81% and 1.24%, respectively, after treatment with microbial inoculum (corn stalk biochar). Results indicated that the most effective treatment in ginseng soil ould be the combined application of corn stalk biochar and Frankia F1. biochar, peanut shell biochar, wheat straw biochar, and cotton biochar) in combination with antagonistic Frankia F1. We set up four treatments, i.e. CK (control treatment, no fertilizer and biochar), A (wheat straw biochar with microbial inoculum), B (rice straw biochar with microbial inoculum), and C (corn straw biochar with microbial inoculum), to analyze soil physical and chemical properties, fertility, enzyme activity and bacterial richness, evenness, community composition, and structural changes. Extensive studies to identify effective microbial inoculum can accelerate soil ecological recovery, and shorten the interval years during continuous cropping of ginseng; these advances could have far-reaching signicance and potential applications in agriculture.


Introduction
Ginseng (Panax ginseng Meyer) is a valuable medicinal plant. Its root is used to enhance organ function and prevent various disorders 1,2,3 . This plant is widely cultivated in China, North Korea, South Korea, Japan, and Russia; the Changbai Mountain area is the major producer of ginseng in the world, accounting for 80% of the total production of ginseng in China and 60-70% of global production 4 . However, long-term, continuous, crop rotation has created favorable conditions for the reproduction and spread of pathogenic fungi that affect ginseng production. This practice has allowed diseases to become increasingly prominent, threatening crop yield and quality 5 . Moreover, ginseng pathogenic fungi are extremely resilient, surviving within infected soil and diseased tissue even through winter 6 . More than 70% of ginseng crops may die if the land is infected, making this disease extremely detrimental to agricultural pro ts 7 .
At present, biological control has been reported in many crops 8, 9 . Biocontrol practices are the current focus of disease control research because they are environmentally stable, nontoxic, and e cient. Therefore, the application of biological pesticides and the development of microbial fungicides are the inevitable trend in plant disease control. Studies in China and other countries have shown that various antagonistic antifungal agents such as Chaetomium globosum 10 and Bacillus amyloliquefaciens can inhibit ginseng pathogens 11 .
Biochar is a kind of stable and highly aromatic solid material produced by pyrolysis and carbonization of biomass at high temperature under oxygen limitation. It has a unique structural characteristics and has attracted wide attention. Its physical structure and chemical properties have good bene ts to soil, and it is a potential soil conditioner and adsorbent 12,13 . Not only can biochar reduce nutrient leaching 14,15 and improve soil structure, it can also provide nutrients and living niches for soil microorganisms 16 , reduce the competition among microbes, protect the bene cial soil microorganisms 17 , improve the soil microbial community structure, and enhance soil bacterial diversity 18,19,20 . Biochar may be an effective way to improve the quality of acid soils 21 . For example, Mao et al. 22 pointed out that adding bamboo charcoal and commercial microbial inoculum to pig manure composting caused changes in the relative abundance of Firmicutes and Proteobacteria in the bacterial community during the high-temperature period of composting, thus reducing the production of CH 4 , N 2 O and NH 3 by 69%-80%, 45% and 19%-29%, respectively. Duan et al. 23 found that, compared with the control treatment, the addition of wheat stalk charcoal and microbial inocula (extracted from fresh cow dung) into cow dung compost signi cantly increased the number and abundance of microorganisms in the compost.
Frankia is a kind of actinomycetes that can form nodules in the roots of non-leguminous plants and perform symbiotic nitrogen xation 24 . Plants that can form nodule with Frankia symbiosis are collectively called actinomycetes nodule plants. They have strong symbiotic nitrogen xation ability and are widely distributed. They are important nitrogen suppliers in terrestrial ecosystems, so they have a wide application prospect in agriculture and forestry. Thus, Frankia has the potential to be used in nature as a powerful resource 25 . Other studies have found that Frankia, like other actinomycetes, can produce antimicrobial substances that inhibit the growth of microorganisms, such as Escherichia coli, Staphylococcus aureus, Candida albicans, and Penicillium chrysogenum 26 . Lihua 27 showed that Frankia could improve the nutrient composition, organic matter, total nitrogen, available potassium and available nitrogen contents in Casuarina casuarina forest land, with the increase in available nitrogen being the largest.The biocontrol effect of actinomycetes on plant diseases depends on the stability of their colonization in the environment, which is affected by many factors, such as temperature, humidity, soil type, pH, nutritional status, and microbial community diversity of the soil 28 .
Recently, new generation metagenomics has been applied to comprehensively analyze the structure of the soil microbial community based on updated high-throughput sequencing technology 29 . This method could rapidly provide accurate high-volume production of sequence data and offer an opportunity to achieve a high throughput and deeper insight into the effects of different treatments on the composition of microbial communities 30 . The objective of this study was to evaluate the application of different types of straw-derived biochar (corn straw biochar, rice straw biochar, peanut shell biochar, wheat straw biochar, and cotton biochar) in combination with antagonistic Frankia F1. We set up four treatments, i.e. CK (control treatment, no fertilizer and biochar), A (wheat straw biochar with microbial inoculum), B (rice straw biochar with microbial inoculum), and C (corn straw biochar with microbial inoculum), to analyze soil physical and chemical properties, fertility, enzyme activity and bacterial richness, evenness, community composition, and structural changes. Extensive studies to identify effective microbial inoculum can accelerate soil ecological recovery, and shorten the interval years during continuous cropping of ginseng; these advances could have far-reaching signi cance and potential applications in agriculture.

Soil collection and preparation of related materials
Soil samples were collected in Jilin Province, China (126° 44′ 22′′ E and 42° 39′ 51′′ N). The samples were obtained randomly by dividing a 100 m 2 eld of continuous ginseng cropping into 10 small blocks. In each block, 5 soil samples (>1 kg) were randomly taken from 0 to 20 cm depth. The physicochemical properties of ginseng soil were pH value 4.79, organic matter 14.67 g/kg, total nitrogen 0.81 g/kg, available nitrogen 122.43 mg/kg, available phosphorus 21.93 mg/kg, and available potassium 143.82 mg/kg. The chemicals used in this study were all analytical grade. The Frankia F1 strain was provided by the Department of Environmental Engineering, College of Resources and Environment, Jilin Agricultural University, China. We utilized NCBI-BLAST for the gene sequences with higher homologous in the database, and used the MEGA 7.0 software to construct the phylogenetic tree.
Ginseng pathogens Fusarium solani, Cylindrocarpon destructans, Phytophthora cactorum, Alternaria panax, Rhizoctonia solani, Sclerotinia schinseng, and Botrytis cinerea were provided by Plant Pathology Laboratory of College of Agronomy, Jilin Agricultural University, and stored at 4℃ on agar slants at low temperature.
Veri cation of Frankia F1 against pathogenic fungi of ginseng Filter paper method 53 was used to make bacterial plates of each activated pathogen. Four sterilized lter paper circles with a diameter of 1 cm were placed to the four points 25 mm away from the center of the plate. The lter papers were infused with 20 μL Frankia F1 bacterial suspension and the control treatment with 20 μL sterile water. Cultures were incubated at 28℃ for 7 days. The presence of fungistatic bands was observed, and the diameter of pathogenic bacteria colonies was measured to calculate the fungistatic rate.

Preparation of microbial inocula
First, Frankia F1 was inoculated in 100 mL of Gao's No. uid medium and incubated at 30℃ for 7 days. Corn straw biochar, rice straw biochar, peanut shell biochar, wheat straw biochar, and cotton biochar were selected as candidate materials for the preparation of microbial inocula. The hydroscopicity of the carrier material, the biocompatibility with Frankia F1 and the fungistatic activity of the prepared microbial inoculum against pathogenic fungi were determined, and the surface morphology of the candidate material was visualized using a scanning electron microscope (JSM-7800F, Japan).
Next, each of the 5 carrier materials was sterilized in a 50 mL centrifuge tube with 2 g of the material, and then mixed with 10 mL Frankia F1 bacterial suspension (1×10 9 CFU/mL), put in a sterile bottle, dried in oven at low temperature (30 ℃), and stored at 4℃ and 25 ± 5℃ (room temperature). All ve types of microbial inocula were prepared in the same way.
Finally, at days 1, 7, 14, 21, and 28, the number of living cells in different types of carrier materials was determined by the plate counting method, and the antibacterial activity was determined by the plate-disk incubation method at 7 d.

Remediation of ginseng soil with microbial inocula
The pot experiment was carried out in a solar greenhouse of Jilin Agricultural University College of Resources, Environment and Environment. The whole experiment lasted 28 days. The effects of microbial inoculum on physicochemical properties, fertility, enzyme activities, and bacterial community structure of continuous cropping ginseng soil were studied. Four treatments were set, featuring microbial inocula prepared with different biochars derived from wheat (treatment A), rice (treatment B), corn stalk (treatment C), and unamended control (CK).
The soil (1 kg per pot) was mixed with a treatment inoculum (10 g) and placed into a pot. Each treatment had ve independent repetitions, and a total of 20 planting pots were set up. The soil in each of the planting pot was watered to about 75% water holding capacity by adding distilled water every day by weighing. At the end of the experiment, 3 replicates from each treatment were randomly selected for subsequent analyzes. The samples were equally divided into two parts: one part was frozen at −80℃ for DNA extraction and another part was preserved at 4 ℃ for further analysis. The physicochemical properties and enzyme activities of soil were measured by the published methods 54, 55, 56 .

DNA extraction and PCR ampli cation
Soil samples were collected from four different treatments, and DNA was extracted from the soil using a PowerSoil DNA separation kit (MoBio Laboratories, Carlsbad, CA, USA) in accordance with the protocol recommended in the instructions. The purity and quality of genomic DNA were determined by 1% agarose gel. Target fragments in the v3-v4 regions of the bacterial 16S rRNA hypervariable region were ampli ed with universal primers 338F(5'-ACTCCTACGGGAGGCAGCA-3') and 806R(5'-GGACTACHVGGGTWTCTAAT-3'). A 10-digit barcode sequence was added to the 5 'end of the positive and negative primers of each soil sample (provided by Owesen, Beijing, China). The PCR ampli cation was performed by a Mastercycler (Eppendorf, Germany), and the ampli cation procedure was pre-denaturation at 95℃ for 5 min. There were 32 cycles at 95℃ for 45 s, 55℃ for 50 s and 72℃ for 45 s followed by elongation at 72℃ for 10 min. Each sample was repeated 3 times. After ampli cation, PCR products from the same sample were mixed. The PCR products were detected by electrophoresis, and the target band sizes were ampli ed by 1% agarose gel electrophoresis, and puri ed by an Agencourt AMPure XP nucleic acid puri cation kit and sent to Orvison, Beijing, China, for Illumina Miseq high-throughput sequencing.

Statistical analysis
The raw data were rst screened and the sequences were removed if they were shorter than 200 bp, had a low quality score (≤20), contained ambiguous bases, or did not exactly match the primer sequences and barcode tags. Quali ed reads were separated using the sample-speci c barcode sequences and trimmed with Illumina Analysis Pipeline Version 2.6. Then the dataset was analyzed using vsearch. The sequences were clustered into operational taxonomic units (OTUs) at a similarity level of 97% followed by generating rarefaction curves and calculating the abundance and diversity indices. The Ribosomal Database Project Classi er tool was used to classify all sequences into different taxonomic groups 57 . To examine the similarity between different samples, the clustering analyses and PCA were conducted based on the OTU information from each sample using R 58

Results
Wide-spectrum veri cation of Frankia F1 against pathogenic fungi of ginseng The phylogenetic tree of the F1 strain is shown in Figure 1. It has 100% homology with the Frankia casuarinae strain CCl3, which has been published in GenBank. The sequence accession number is GU296535. The fungistasis spectrum test of Frankia F1 against pathogenic fungi of ginseng ( ve types of carrier material. The pore layered structure of corn stalk biochar, rice straw biochar and wheat straw biochar formed a complex three-dimensional structure, indicating that the three kinds of biochar had a highly porous structure (Figure 2a, b, e). Such structure is expected to be bene cial to the adhesion and reproduction of bacteria, and the diffusion of primary and secondary metabolites supporting the normal metabolism of the introduced biocontrol strain. In contrast, cotton biochar (Figure 2c) had relatively sparse pore structures, and peanut shell biochar (Figure 2d) did not show pore structures suitable for the survival of microorganisms, suggesting bacteria could only attach to the surface of biochar, making for a poor carrier. Based on these results, biochar derived from corn, rice or wheat straw were selected as the most optimal carrier materials for producing microbial inoculum. Effects of microbial inocula on physicochemical properties and enzyme activities of ginseng soil At 28 days, microbial inocula signi cantly altered the characteristics of the soil (Table 3). Compared with the control group, the treatments with wheat straw biochar, rice straw biochar and corn straw biochar increased the pH by 4.43%, 6.55% and 11.18%, and organic matter by 7.43%, 22.10% and 55.43%, respectively. At the same time, there were signi cant differences between four treatments. The contents of total nitrogen, available nitrogen, available phosphorus, and available potassium in the soil in the corn straw biochar treatment were signi cantly increased by 33.07%, 26.70%, 16.40%, and 9.10%, respectively, compared with the control group. Microbial inocula had signi cant effects on the activities of urease, sucrase and catalase in the soil (Table 3), but had no signi cant effect on the activity of phosphatase. Compared with the control group, the contents of urease, catalase and sucrase activities in the soil in the corn straw biochar treatment were increased signi cantly (by 52.73%, 16.80% and 43.80%, respectively).

Composition of soil bacterial community
The four soil treatments showed a total of 421,879 effective bacterial sequences and a total of 7,114 OTUs ( Figure   3). Compared with the unamended control group, total OTUs for wheat straw biochar, rice straw biochar and corn straw biochar treatments increased by 10.24%, 8.76% and 19.86%, respectively (Table 4). Shannon and Simpson indices re ect the diversity of taxa, and these indices were signi cantly increased in comparison to those in the control group (P<0.05). In order to more clearly observe the differences between treatments, we used the Bray-Curtis test to quantify the sample distances ( Figure 4). The results indicated that the differences among different treatments were signi cant. Compared with the control treatment (CK), the distances were large with the treatments wheat straw biochar and corn straw biochar and small with the rice straw biochar treatment.

Clustering analysis of bacterial community at the class level
In all the soil samples, we detected 39 phyla, 97 classes, 153 orders, 225 families, and 306 genera. The twelve soil samples from four different treatments were divided into two categories ( Figure 5). The soil samples from the unamended control were clustered into one branch, whereas the soil samples from the wheat straw biochar, rice straw biochar and corn straw biochar treatments were clustered into another branch. The relative abundance of bacterial community composition was analyzed at the class level, and there were ve dominant phyla (abundance of > 2%) whose relative abundance in the soil was signi cantly different among the treatments. Compared with the control, the relative abundance of Alphaproteobacteria, Gammaproteobacteria and Sphingobacteria was signi cantly higher, and the relative abundance of Actinobacteria and Thermoleophilia was signi cantly lower. In three of the treatments, the relative abundance of class Actinobacteria accounted for 18.29% in the wheat straw biochar treatment, almost 16.53% in the rice straw biochar treatment and 15.62% in the corn straw biochar treatment. After the corn straw biochar treatment, the abundance of Alphaproteobacteria, Gammaproteobacteria and Sphingobacteria increased by 7.87%, 9.81% and 1.24%, respectively. In summary, the effects of the treatments wheat straw biochar and rice straw biochar on the improvement of bacterial community were similar, and both of them mainly enhanced the relative abundance of Alphaproteobacteria and Gammaproteobacteria.
The soil samples from different treatments were clustered by taxa, or similarity of abundance among samples, and the clustered data were used to construct a hierarchical clustering heatmap ( Figure 6). The soil bacterial communities treated by wheat straw biochar, rice straw biochar and corn straw biochar were grouped into two groups, one of which was the community with high relative abundance of Thermoleophilia, Sphingobacteria, Actinobacteria, Alphaproteobacteria, and Gammaproteobacteria.
Signi cant differences in taxa among the treatment groups Linear discriminant analysis (LDA), which allows comparisons among the treatment groups, also performs subgroup comparisons within the group comparisons to nd taxa with signi cant differences in abundance among the groups. When LDA>4.0, there were 26 taxa groups with signi cant differences among wheat straw biochar, rice straw biochar, corn straw biochar and CK at each classi cation level (Figure 7a), and CK had the most signi cant differences with 14 taxa. Among the other three treatments, ve taxa were signi cantly different in the wheat straw biochar treatment, and the rice straw biochar treatment had 1. There were many signi cantly different taxa in the corn straw biochar treatment; at LDA>4.0 it had six and at LDA>5.0 it had one Saccharibacteria. These results indicated that Saccharibacteria contributed greatly to the signi cance of differences and was the most important taxa that caused the differences among the four treatments.
LEfSe analysis showed (Figure 7b) that there were 24 signi cantly different taxa among the four treatments, including three in the wheat straw biochar treatment, whereas rice straw biochar had zero, corn straw biochar had 3, and CK had 18. The number and abundance of different taxa were the highest in the CK treatment. Different taxa treated in the wheat straw biochar treatment were mainly Xanthomonadaceae, Xanthomonadales and Gammaproteobacteria, and in the corn straw biochar treatment were mainly Sphingomonadaceae, Sphingomonadales and Alphaproteobacteria. After treatments wheat straw biochar, rice straw biochar, corn straw biochar and CK, the bacterial taxa with the largest contribution were Proteobacteria, Actinobacteria, Chloro exi, and Saccharibacteria, respectively.
Redundancy Analysis (RDA) of Bacterial Community Structure and Soil Environmental Factors Soil pH, organic matter (OM), total nitrogen (TN), and dominant taxa in the bacterial community were analyzed by RDA ( Figure 8). Organic matter and total nitrogen were the main factors in uencing bacterial community composition in ginseng soil after different treatments. In the RDA analysis, the distance between the corn straw biochar treatment and CK was large, and the two treatments occupied separate quadrants, indicating that the soil bacterial community structure of the corn straw biochar and CK treatments was signi cantly different. Treatments wheat straw biochar and rice straw biochar occupied a common quadrant, indicating the similar bacterial community structure. The correlation analysis showed that Alphaproteobacteria was signi cantly positively correlated with organic matter and total nitrogen (P<0.05). Acidobacteria, Thermoleophilia, Actinobacteria, and Spartobacteria were negatively correlated with organic matter and total nitrogen (P < 0.05).

Discussion
Reasons for the problems associated with continuous cropping are extremely complex, involving soil, plants, microorganisms, and many other aspects. Some studies have shown that the change of soil physical and chemical properties and the lack of soil fertility are the main reasons for problems with continuous cropping 31 33,34 . In this study, the biochar prepared with corn straw , wheat straw and rice straw were applied to the soil; they improved soil pH value, organic matter, fertility, and enzyme activity. Yulan et al. 35 found that, compared with organic fertilizer application alone, the mixed application of organic fertilizer and microbial inoculum could greatly improve the content of organic matter and available phosphorus in the soil of an abandoned greenhouse. Therefore, the application of microbial inocula can signi cantly improve the soil nutrient status and increase crop yield. This may be because the application of microbial inocula can signi cantly enhance soil enzyme activity and promote the transformation of nutrients into available forms 36 .
To more deeply probe the microbial diversity, we conducted 16S rRNA analysis on the soils. In the ginseng soil tested in our study, the top ve classes of bacteria included Actinobacteria (26.54%), Alphaproteobacteria (14.88%), Gammaproteobacteria (6.84%), Thermoleophilia (12.38%), Sphingobacteria (2.44%). Biochar application with different microbial materials had different effects on various microbial taxa. After the corn straw biochar treatment, the abundance of Alphaproteobacteria, Gammaproteobacteria and Sphingobacteria increased by 7.87%, 9.81% and 1.24%, respectively. Alphaproteobacteria and Gammaproteobacteria belong to Proteobacteria, and they include nitrogen-xing bacteria that are symbiotic with plants 37 . Gammaproteobacteria often utilize nutrients such as ammonia and methane generated by the decomposition of organic substances 38,39 . Sphingobacteria are conducive to the degradation of cellulose in litter. Actinobacteria can promote the rapid decay of litter and is an important participant in the decomposition of chitin and lignin 40 .
Four soil treatments generated 421,879 effective bacterial sequences and 7,114 OTUs ( Figure 3). The relative abundance Chao1 index and the diversity index of bacterial community increased after wheat straw biochar, rice straw biochar and corn straw biochar treatments in soil (Table 4). This is consistent with the research results of Guangming 41 , with the addition of biochar improving the population structure diversity of soil microorganisms in ash desert soil. The promotion effect of biochar on soil bacterial community may be attributed to the following reasons: biochar provides more carbon sources for soil bacteria and improves the environmental conditions for soil bacteria; the complex pore structure of biochar provides a good habitat for the growth of soil bacteria and protects them from being preyed upon by other organisms 42 . Frankia F1, which has antifungal effect, was selected as a biocontrol microorganism. It is an actinomycete that can become a symbiont with non-legumes to form root nodules and x atmospheric nitrogen. It forms vesicles on the stalk at the top of the mycelium. It has a very high nitrogen xation e ciency and a wide range of host plants across different families. Studies on the in uence of biochar on soil bacteria mostly focus on nitrogen bacteria. Many scholars have found that nitrogen bacteria can promote nitrogen xation and inhibit denitri cation 43 .
In ecology, redundancy analysis is a method to identify the relationship between environment, taxa and vegetation.
The bacterial taxa in soils were closely related to the pH value, the total carbon, and the total nitrogen in the soils 44 .
Using the redundancy analysis, Jian et al. 45 concluded that soil organic carbon and total nitrogen were soil indicators in the process of rocky deserti cation in mountainous karst areas. In this study, redundancy analysis was conducted on bacterial community structure and the soil environmental factors. The RDA results showed that pH value, organic matter and total nitrogen were the main environmental factors affecting the community composition of ginseng soil bacteria. Studies have shown that soil pH, organic carbon and nitrogen are important environmental factors that affect the composition of bacterial communities 46 , and organic matter can increase the activity of soil enzymes and improve soil biological activity 47 by regulating the capacity of soil microorganisms to utilize carbon sources 48 . Fertilization can not only change soil physical and chemical properties, but also regulate soil enzyme activities and promote soil microorganisms coordinating nutrient cycling in soil.
Biochar is relatively stable, which signi cantly affects the activities of soil microorganisms. The pore structure of biochar and its adsorption of water and fertilizer can provide a good habitat for microorganisms 49 . Jin et al. 50 found that the addition of biochar could signi cantly increase soil microbial biomass nitrogen proportionally with an increase in the amount of biochar. The application of biochar can cause changes in the bacterial community structure in the soil, and the bacterial community structure is signi cantly correlated with the soil pH value and nutrients 51,52 . It is concluded that the proportion of biocontrol bacteria and bene cial bacteria in soil can be controlled by adjusting soil pH value, organic matter and total nitrogen in production practice. In summary, we conclude that the changes in the bacterial compositions in our study are caused by the application of microbial inoculum prepared with microorganisms and biochar. Declarations