Microbial diversity in rhizosphere soil of soybean grass under different cultivation methods in an alpine region

A large number of studies have shown that soybean grass with mixed seeding cultivation can signicantly improve the yield and quality of forage grass compared with clean culture cultivation.This study explores the differences in the characteristics of the composition and diversity of the microbial community in the rhizosphere of soybean grasses between clean culture and mixed seeding methods in an alpine region. We used high-throughput sequencing technology to determine the microbial diversity and analytical methods to determine the physicochemical characteristics of plant rhizosphere soil of Avena sativa L. and Vicia sativa L. the diversity in rhizosphere soil of soybean grass under different The showed that the bacterial microbial diversity in rhizosphere soil of Avena sativa L. was the highest under clean culture, while the bacterial microbial diversity in rhizosphere soil of Vicia L. the lowest under clean culture. The fungi microbial diversity in rhizosphere soil of Vicia L. the highest under clean culture and the lowest under the mixed cultivation of Avena sativaL. There

Forage is a necessity for the survival of animal husbandry and the fundamental driving force for ensuring the sustainable development of animal husbandry. The development of the forage industry aims to promote the development of forage cultivation and pursue high-yield cultivation. A change in cultivation methods is a new technique for the high-yield cultivation of forage. A large number of studies have shown that soybean grass with mixed seeding cultivation can signi cantly improve the yield and quality of forage grass compared with clean culture cultivation, and the development of mixed soybean grassland is a feasible method to increase forage yield and quality [1][2][3][4][5][6][7][8]. Avena sativa L. and Vicia sativa L. are ne forage grasses widely cultivated in alpine areas with strong resistance and of excellent quality [9][10][11][12]. They are widely cultivated in China and abroad to provide high-quality feed for domestic animals in mixed cultivation [13,14].
Grassland soil microorganisms play an important role in the grassland ecosystem [15,16]. There are a large number of microorganisms in the rhizosphere soil of plants. Due to the combined effect of microbial activities and plant metabolism, differences in soil characteristics, microbial community composition, nutrient content, and other aspects exist between the rhizosphere soil and non-rhizosphere soil of plants [17,18]. The diversity of soil microorganisms is an important indicator that can directly re ect the stability and diversity of the plant ecosystem, and it can also visually demonstrate the differences in the function, communities, and species of soil microorganisms in different regions [19]. As an important indicator to describe the characteristics of the soil microbial community, the functional diversity of the soil microbial community can re ect the basic information of the soil [20]. The structure and diversity of the microbial community in rhizosphere soil are usually in uenced by cultivation conditions, the planting environment, plant species, and other factors [21][22][23]. Compared with previously employed methods, such as traditional DNA sequencing, high-throughput sequencing technology can greatly reduce the research cost, save time, and yield a large amount of information with a higher level of accuracy and sensitivity [24][25][26]. At present, there have been many studies on different mixed-planting ratios and grass yield in seeding methods of soybean grass [27][28][29], but there are few studies of the rhizosphere microbial diversity of forage under different cultivation methods.
In this paper, we used the rhizosphere soils of Avena sativa L. and Vicia sativa L. under different cultivation methods in an alpine region as research objects to analyze the changes of microbial diversity using high-throughput sequencing so as to reveal the differences in microbial diversity of the rhizosphere soil of plants under different cultivation methods and to analyze structural and composition changes in the microbial community, thereby providing a basis for the study of microbial ecology of the rhizosphere soil of cultivated forages.

Soilphysicochemical properties soil
The physicochemical properties of the soil samples are shown in Table 1. There was no signi cant difference in pH or in the total nitrogen, total phosphorus, or total potassium content of the soil samples with the three different treatments. Overall, the pH value was 7.50-7.73, the total nitrogen content was 3.16-4.13 g•kg − 1 , the total phosphorus content was low at 1.63-1.91 g•kg − 1 , and the total potassium content was 22.01-22.92 g•kg − 1 . The content of available nitrogen, available phosphorus, available potassium, and organic matter in the soil samples were signi cantly different among the three treatments (P < 0.05). The available nitrogen content was 210.11-275.17 mg•kg − 1 , available phosphorus was 8.41-18.71 mg•kg − 1 , and available potassium content was 67.11-12.18 mg•kg − 1 . The available nitrogen, available potassium, and organic matter contents in the CYCH treatment soil were signi cantly higher than those in the other treatments (P < 0.05), and the contents of available phosphorus in the CYD treatment soil were signi cantly higher than those in the other treatments (P < 0.05).

Otu-level Analysis Of Soil Bacteria And Fungi
The results of the OTU statistical analysis are shown in Table 2. The 16S rDNA V3-V4 region of soil samples from all three treatments (CWD, CYD, and CYCH) was sequenced to obtain 48,988-50,302 effective tags. The tags lengths were highly concentrated, averaging 421 bp, 420 bp, and 417 bp for the CWD, CYD, and CYCH treatments, respectively. The sequence length of the 16S rDNA V3-V4 region in each treatment was roughly consistent. Following sequencing analysis, 32,031-42,291 effective tags were obtained from the ITS regions of the DNA obtained from the soil samples of the three treatments Generally, the sequence length was 230-280 bp, and the average length was 243, 230, and 280 bp, for the CWD, CYD, and CYCH treatments, respectively. Rarefaction curves ( Fig. 1a and b) were prepared based on the results. It can be seen from Fig. 1 that the OTU level of the sample increased rapidly with the increase of sequencing fragments, and remained stable after reaching the peak.

Analysis Of Soil Microbial Species Composition
The richness of microbial species composition was > 0.1% higher in the three soil treatment samples ( Fig. 2). We detected many phyla and classes of bacteria and fungi, and most of them contained very few species. For the convenience of observation, only the top ten abundant fungi and bacteria are shown in Fig. 2 and the remaining have been grouped as Others. The Unclassi ed and Unknown groups represent species that were not taxonomically annotated. The ordinate (y-axis) shows the relative abundance. Figure 2a and b, respectively, show the top 10 classes of bacteria and fungi, respectively, according to their relative abundance level.
Alphaproteobacteriawas the most abundant, with a relative abundance of 17.83%, 24.24%, and 18.67%, for the CWD, CYCH, and CYD treatments, respectively. For the CYCH treatment, the abundance of Alphaproteobacteria and Actinobacteria were signi cantly higher than for the CWD and CYD treatment.
For the CYCH treatment, the abundance of Gemmatimonadetes, Nitrospira, and Acidimicrobiia was signi cantly lower than that of the CWD and CYD treatment. With the exception of Actinobacteria, the relative abundance of the main bacterial species in the CWD and CYD soils was relatively close and it is quite different from that in CYCH soil.
As can be seen in Fig. 2b, for all three treatments, the soil fungi were mainly distributed in the following 9 groups: Mortierellomycetes, Sordariomycetes,Leotiomycetes, Dothideomycetes, Tremellomycetes, Agaricomycetes, Archaeorhizomycetes, Eurotiomycetes, Spizellomycetes, or Unclassi ed, of which Mortierellomycetes was the most abundant the CYCH treatment soil (32.66%). Compared with the CWD and CYD treatment soil, the abundance of Sordariomycetes and Leotiomycetes in the CYCH treatment soil was signi cantly decreased, while the distribution of fungal communities was signi cantly different between the soil treatments.

Analysis Of Soil Microbial Diversity
Alpha diversity analysis As can be seen from the alpha diversity index results (Table 3)

Analysis Of Beta Diversity
Heatmaps represent samples through interspecies distance relationships to obtain the sequence relationship between samples, which are drawn by means of matrix language tool software. By analyzing the heatmap, the difference between the two samples can be intuitively re ected by the range of color transformation. Figure 3 shows the heatmap of the three soil treatment samples. The bacterial ora (Fig. 3a) and fungal ora (Fig. 3b) in the CWD and CYD treatment soil had a high level of similarity.
The beta microbial diversity analysis was conducted to obtain a distance matrix. Then, the UPGMA function of R software was used for hierarchical clustering comparison of samples, and the similarity relationship of the species composition in each sample was described. The closer the samples were, the shorter the branch length was, indicating that the species composition of the two samples was more similar. Fig. 4 shows that the OTU level of bacteria and fungi in the CWD treatment soil samples had a relatively high consistency, while the similarity of bacteria and fungi in the CYCH treatment soil sample was relatively low.

Functional gene prediction
PICRUSt software (http://picrust.github.io/picrust/) was used to compare the species composition information obtained from the 16S sequencing tags to infer the functional gene composition of the samples and then analyze the functional differences between the different soil treatments. According to the of Kyoto Encyclopedia of Genes and Genomes (KEGG) metabolism histogram (Fig. 5), the functional genes of the soil microbial community for each treatment were basically similar in their metabolic pathways.

Discussion
Previous studies by many scholars have con rmed that the abundance and ora composition of soil microbial populations are closely related to pH, organic matter content, water content, the content of various nutrients, soil particle size, and other physical and chemical properties of soil [37][38][39][40][41][42]. In this study, there were no signi cant differences in pH, total nitrogen, total phosphorus, and total potassium in the three plant rhizosphere soil treatment samples of Avena sativa L. and Vicia sativa L. under clean culture and mixed seeding cultivation, but there were signi cant differences in available nitrogen, available phosphorus, available potassium, and organic matter content (P < 0.05). The content of available nitrogen, available potassium, and organic matter of soil of Avena sativa L. and Vicia sativa L. in the mixed seeding treatment were signi cantly higher than those in the other treatments (P < 0.05), and the content of available phosphorus of Avena sativa L. in the clean culture treatment was signi cantly higher than in the other treatments (P < 0.05). The relationship between plant rhizosphere microorganisms and soil nutrient content requires further study.
Rhizosphere microorganisms play an important role in connecting the soil and plant microecosystems. Plant root secretions can provide energy for the life activities of microorganisms, and microorganisms can affect plants by decomposing organic matter and transforming nutrient substances or indirectly affect plant growth and the dynamics of non-rhizosphere microorganisms via material cycling and energy ow [43,44]. Yu et al. [45] studied the diversity of rhizosphere and rhizome microorganisms cultivated by Impatiensbalsamina in Hainan, China, and the seasonal variation of its community at different altitudes. The results showed that the number of rhizospheric fungi and bacterial OTUs, Shannon-Weiner index, and richness index varied greatly with altitude and with the wet and dry seasons.
Lijun et al. [46] studied the microbial diversity in rhizosphere soil of Davidiainvolucrata in different regions. They found that the microbial communities in the rhizosphere soil of Davidiainvolucrata in three regions all had high biodiversity, but there were differences in species composition and distribution. Yating et al. [47] studied the rhizosphere microbial community diversity of maize in Yunnan, China, and they concluded that altitude and variety type were important factors in uencing the rhizosphere microbial community diversity of maize. In our study, the microbial diversity in rhizosphere soil of soybean grass was studied under different cultivation methods. The results showed that the bacterial microorganism diversity in rhizosphere soil was the highest when Avena sativa L. was under clean culture cultivation, while the bacterial microorganism diversity in rhizosphere soil of Vicia sativa L. under clean culture cultivation was the lowest. The fungal microorganism diversity of rhizosphere soil of Vicia sativa L. under clean culture cultivation was the highest, and the fungal microorganism diversity of rhizosphere soil under mixed seeding of Avena sativa L. and Vicia sativa L. was the lowest. The diversity of rhizosphere microorganisms was signi cantly different under different cultivation methods, which was consistent with the results of other researchers. In similar environments, the structure of microbial communities is determined by the plant types, and the structure composition of soil microbial communities varies between different vegetation types [48]. The change in soil microbial diversity depends on soil water content, pH, aeration, nitrogen content, and organic carbon affected by vegetation. Based on the characteristic analysis of rhizosphere exudates and plant litters, the main factors affecting soil microbial diversity may be vegetation type and quantity [49]. In this study, the species structure of the plant rhizosphere microbial community under the three treatments was at class level, and the abundance of Alphaproteobacteria and Actinobacteria was signi cantly higher in the mixed sowing treatment of Avena sativa L. and Vicia sativa L. than that in clean culture treatment of both Vicia sativa L. and Avena sativa L. The abundance of Gemmatimonadetes, Nitrospira, and Acidimicrobiia was signi cantly lower than that of soybean grass under clean culture. The Mortierellomycetes of fungi were most abundant under mixed seeding treatment (32.66%), while the abundance of Sordariomycetes and Leotiomycetes was signi cantly lower than that of soybean grass under clean culture, while the species distribution of bacteria and fungi communities in soil was signi cantly different under the different various treatments.
The KEGG metabolic analysis showed that the functional genes of soil microbial communities were similar in their metabolic pathways. The structure and diversity of the microbial community in the rhizosphere of plants changes greatly under different cultivation modes, and the in uence of plants, soil, and the environment required further study.

Conclusions
In this study, there were no signi cant differences in pH, total nitrogen, total phosphorus, and total potassium in the three plant rhizosphere soil treatment samples of Avena sativa L. and Vicia sativa L. under clean culture and mixed seeding cultivation, but there were signi cant differences in available nitrogen, available phosphorus, available potassium, and organic matter content (P < 0.05). The contents of available nitrogen, available potassium, and organic matter of soil of Avena sativa L. and Vicia sativa L. under mixed seeding treatment were signi cantly higher than those under the other treatments (P < 0.05), and the available phosphorus content in the soil of Avena sativa L. under clean culture treatment was signi cantly higher than those of other treatments (P < 0.05).
We investigated the microbial diversity in rhizosphere soil of soybean grass under different cultivation methods. The results showed that the bacterial microbial diversity in rhizosphere soil of Avena sativa L. was the highest under clean culture, while the bacterial microbial diversity in rhizosphere soil of Vicia sativa L. was the lowest under clean culture. The fungi microbial diversity in rhizosphere soil of Vicia sativa L. was the highest under clean culture and the lowest under the mixed cultivation of Avena sativaL. and Vicia sativa L. There were signi cant differences in rhizosphere microbial diversity under different cultivation methods.
In this study, the species structure of the rhizosphere microbial community was at the class level under the three treatments. The abundance of Alphaproteobacteria and Actinobacteria under the mixed cultivation of Avena sativa L. and Vicia sativa L. was signi cantly higher than that of Avena sativa L. and Vicia sativa L. under clean culture cultivation, and the abundance of Gemmatimonadetes, Nitrospira, and Acidimicrobiia was signi cantly lower than that of soybean grass under clean culture. The Mortierellomycetes class of fungi was the most abundant under mixed seeding treatment (32.66%), while the abundance of Sordariomycetesand Leotiomycetes was signi cantly lower than that of soybean grass under clean culture. The species distribution of the bacteria and fungi community in soil microorganisms was signi cantly different between the soil treatments. KEGG metabolic analysis showed that the functional genes of soil microbial communities were similar in their metabolic pathways.

Methods
Overview of the experimental area The experiment was conducted at Sanjiangyuan Field Ecological Observation Station of Qinghai University (33°24'30" N, 97°18'00" E), which is located in Zhenqin Town, Chengduo County, Yushu Prefecture, Qinghai province, China. The experimental area is situated at an altitude is 4,270 m and has a typical plateau continental climate, with an annual average temperature of − 5.6 °C to 3.8 °C and an annual average precipitation of 562.2 mm, mainly distributed from June to September. Its alpine meadow soil has a rich humus content but poor soil fertility due to poor decomposition. The soil pH is 7.32, organic matter content is 58 g •kg − 1 , and available potassium, phosphorus, and nitrogen is 76.5, 7.0, and 14.0 mg •kg − 1 , respectively, of which ammonia nitrogen is 5.1 mg •kg − 1 and nitrate nitrogen is 8.9 mg •kg − 1 . The area is irrigated during droughts.

Experimental Materials And Design
This experiment was designed as a single-factor randomized block study. The experimental materials were Avena sativa L. and Vicia sativa L., the seeds were provided by the Key Laboratory of Superior Forage Germplasm in the Qinghai-Tibetan Plateau. We set up three treatments: Vicia sativa L. with clean culture cultivation (CWD), Avena sativa L. with clean culture cultivation (CYD), and a mixture of Avena sativa L. and Vicia sativa L. in equal proportions (CYCH). There were three plots of 4 × 5 m 2 each. The plots were drilled, and the rows were spaced at intervals of 30 cm. Diammonium phosphate (30 kg/mu) was applied prior to sowing. The amount of seed used in the plot was calculated according to the plot area. The amount of Avena sativa L. under clean culture cultivation was 15 kg/mu, that of Vicia sativa L. under clean culture cultivation was 5 kg/mu, and the mixed sowing amount is was 7.5 kg/mu of Avena sativa L. plus 2.5 kg/mu of Vicia sativa L.

Sample Collection
The experiment initiated in May 2019 and the experimenter visited the plot in August 2019. Using a vepoint sampling method, root samples ≤ 20 cm long were collected in triplicate from Avena sativa L. andVicia sativa L.. The large soil particles were shaken off using the shake-down method [30], and then a brush was used to brush off excess soil surrounding the roots, leaving a layer of 0-5 mm of soil, which was considered the rhizosphere soil. The two species were evenly mixed, placed into sterile Ziplock bags, stored at a low temperature, and brought back to the laboratory as quickly as possible. The collected soil samples for each treatment were divided into two parts, one to determine the nutrient content and the other to determine the microbial diversity index of the soil.

Determination Of Soil Physicochemical Properties
The indices of the physicochemical properties of the soil samples were determined by conventional analysis methods [31]. Soil pH was determined using a pH potentiometer method. Organic matter content was determined by high-temperature potassium dichromate oxidation capacity, available nitrogen content by a potassium chloride extraction-spectrophotometric method, available phosphorus content by sodium bicarbonate extraction and molybdenum-antimony colorimetry, and available potassium content by an ammonium acetate extraction-ame photometer method. Total nitrogen content was determined by Nessler colorimetry, total phosphorus content by molybdenum-antimony colorimetry, and total potassium content by a ame photometer method.

Total Dna Extraction From Soil Microorganisms
The total genomic DNA was extracted from the soil microorganisms using an OMEGA soil DNA extraction kit. Each sample was mixed before processing in triplicate to reduce the chances of missing DNA from organisms that were sparsely distributed. Following extraction, polymerase chain reaction (PCR) ampli cation of the V3-V4 region of the bacterial 16S rRNA gene and the fungal internal transcribed spacer (ITS)1 region was performed using universal primers (bacterial V3-V4 region: 341F, 5 -CCTACGGGNGGCWGCAG-3 and 805R, 5 -GACTACHVGGGTATCTAATCC-3 and fungal ITS1 region: ITS1F, 5 -CTTGGTCATTTAGAGGAAGTAA-3 and ITS2R, GCTGCGTTCTTCATCGATGC-3 ). The PCR ampli cation conditions were performed with reference to the method used by Zhao et al. [32]. The ampli ed products were puri ed and quanti ed, and a sequencing library was constructed, which was sequenced by Sangon Biotech (Shanghai) Co., Ltd. (Shanghai, China) on an Illumina HiSeq PE250 highthroughput sequencing platform.

Bioinformatics Analysis
After cutting the barcode and primer sequences from raw data, the reads of each sample were spliced to obtain the raw tags. The raw tags were inspected and the chimeras were removed to obtain effective tags. We used Qiime software (http://qiime.org/) to perform operational taxonomic unit (OTU) clustering when the sequence similarity was > 97%. We used mothur software (https://www.mothur.org/) [33], the Silva ribosomal RNA sequence database (https://www.arb-silva.de/), and the UNITE fungal DNA database (https://unite.ut.ee/) [34]to annotate species (threshold value, 0.8-1.0) to determine the classi cation information of the sequence and obtain taxa at all levels of phylum, class, order, family, and genus. The PyNAST tool (https://omictools.com/pynast-tool) [35] was used for multi-sequence comparison with data information in the Greengenes database (https://greengenes.secondgenome.com/) [36]. Finally, the tags were standardized, and Qiime l.7.0 software was used to calculate the alpha diversity index and perform the beta diversity analysis. We plotted the dilution curve using the vegan package of R 2.15.3 software (https://www.r-project.org/). The R language tool was used to draw the heatmap, and the unweighted pair-group method with arithmetic mean (UPGMA) was used to draw the phylogenetic tree by hierarchical clustering.

Data analysis
All data were processed and analyzed using SPSS 21.0 software (IBM Corp. in Armonk, NY). One-way analysis of variance and Duncan's new multiple range test were used to analyze the signi cant differences.A probability value (P) of < 0.05 was considered statistically signi cant. Anailability of data and materials All data generated of analysed of this study are described in this paper.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Competing interests
The authors declare that they have no con ict of interest.  Histogram of KEGG metabolism of soil samples