Effects of land conversion on soil microbial community structure and diversity

Background : To study the impact of land-use change on soil microbial community structure and diversity in Northeast China, three typical land-use types (plough, grassland, and forest), grassland change to forest land and grassland change to plough, in the Qiqihar region of Heilongjiang Province were taken as research objects. Methods : MiSeq high-throughput sequencing technology based on bacterial 16S rRNA and fungal ITS rRNA was used to study the above community structure of soil bacteria and fungi and to explore the relationship between soil bacteria and soil environmental factors. Results : The results showed that the dominant bacterial phyla changed from Actinobacteria to Acidobacteria , the dominant fungal phyla changed from Ascomycetes to Basidiomycetes , and the ECM functional group increased signicantly after the grassland was completely changed to forest land. After the grassland was changed to plough, the dominant phyla changed from Actinomycetes to Proteobacteria . The functional groups of pathogens and parasites increased signicantly. There was no signicant difference in the diversity of soil bacterial communities, and the diversity of fungal communities increased signicantly. CCA showed that pH, MC, NO 3 - -N, TP and AP of soil were important factors affecting the composition of soil microbial communities, and changes in land-use patterns changed the physical and chemical properties of soils, thereby affecting the structure and diversity of microbial communities. Conclusions : Our research results clarify the impact of changes in land use on the characteristics of soil microbial communities and provide basic data on the healthy use of land. translation, ribosomal and biogenesis; lipid posttranslational protein metabolite biosynthesis, catabolism;


Background
The soil microbial community is the main driving force of ecosystem processes and has the functions of completing the decomposition of soil organic matter and plant litter and mediating the carbon (C) and nitrogen (N) biogeochemical cycles in terrestrial ecosystems [1,2] . However, the composition and diversity of these communities are largely controlled by soil environmental conditions. Therefore, understanding the composition and diversity of soil microbial communities can reveal the interrelationships between soil microorganisms and the local environment and how these communities respond to human disturbance [3] .
Land-use transformation dominated by human activities has a signi cant impact on the composition and structure of soil microbial communities [4,5] . It can fundamentally change soil quality and nutrient cycling, thereby affecting the construction of soil microbial communities, and it can potentially affect soil microbial diversity and ecosystem functions [6,7] . For example, Jangid et al. [8] found that grassland conversion to plough caused signi cant changes in bacterial and fungal abundance and diversity and determined that land-use change was the main determinant of microbial community composition. Wang et al. [9] found that after grassland was transformed into pine forest, the dominant soil bacterial phylum changed from Proteus to Actinomycetes, the dominant fungal phylum changed from Ascomycetes to Basidiomycetes, and grassland afforestation increased ECM fungi but reduced biological nutrition fungus. Mendes et al. [10] found that the contents of acidophilus bacteria and chlamydia in forest soil were higher, the content of actinomycetes in forest logging areas was higher, and the content of nitrifying bacteria and thermophilus bacteria in plough was higher.
Heilongjiang Province, as China's largest commercial grain production base, has fertile soil and a long history of farming. It's soil physical and chemical properties and fertility are crucial to the sustainable development of agriculture [11] . Since the 1950s, with the increase in population, to solve the problems of food and clothing, the area has adopted land reclamation to obtain ploughs. The original grassland was reclaimed into plough, and the natural vegetation disappeared. When the plough land was abandoned, the ground was exposed, which seriously damaged the soil, and it was di cult to restore the original plant community in a short period [12] . At the beginning of the 21st century, the government realized the severity of the ecological and environmental problems and planted some grassland and abandoned land with forest to protect the fragile local ecological environment and promote sustainable and stable economic development [13,14] . In this paper, in the western region of Heilongjiang Province was chosen as the study area, and three land-use types-grass, plough, forest land-were used to determine how longterm land use changed the soil physical and chemical properties and soil bacterial and fungal community structure and diversity; this information is important for the maintenance of soil fertility in the study area, for breeding and for providing a scienti c basis for the protection of soil microbial diversity.

Results
Physical and chemical properties of soil in different land-use patterns The physical and chemical properties of the soil of the three land-use patterns are shown in Table 1. The pH of all soils was relatively alkaline, with a signi cant difference between grassland and plough (P<0.05), with the lowest pH value in plough and the highest pH value in grassland. The soil moisture content of the three land-use patterns was signi cantly different (P<0.05), with the highest soil moisture content in forest. The contents of microbial biomass carbon, microbial biomass nitrogen, total phosphorus, available phosphorus and nitrate nitrogen in plough were signi cantly higher than those in forest and grassland (P < 0.05). However, there was no signi cant difference in soil organic matter, total nitrogen, ammonium nitrogen, total potassium or available potassium. Venn diagram of soil microorganisms in different land-use patterns The bacterial Venn diagram of the three land-use patterns is shown in Figure 1A. Effects of land-use patterns on alpha diversity of soil bacteria and fungi There was no signi cant difference between the Shannon index and Simpson index of soil bacteria.
However, the soil bacterial Ace and Chao1 indexes of the three land-use patterns were signi cantly different (P < 0.05). Among them, the Ace index showed plough > forest > grassland, and there were signi cant differences between grassland, forest, and plough (P < 0.05). The Chao1 index result was plough > forest > grassland, and the three land-use types had signi cant differences (P<0.05). The number of soil bacteria OTUs was signi cantly different and showed plough > forest > grassland.
The soil fungal Shannon index, Simpson index, Ace index, Chao1 index, and OTU index were signi cantly different ( Table 2). The OTU index was ranked plough > forest > grassland; the Shannon diversity index was plough > grassland > forest; the Simpson index was forest > grassland> plough; the Ace index was plough > forest > grassland; and the Chao1 index was plough > forest > grassland. Analysis of soil bacterial and fungal community structure in different land-use patterns From the perspective of the overall bacterial community structure, all OTUs belong to 55 bacterial phyla. If the sequence cannot be classi ed to the known phylum level, the phylum can be uniformly classi ed into "others". According to the relative abundance of all phylum levels of the three land-use patterns, the dominant bacteria in the samples were Proteobacteria, Acidobacteria and Actinobacteria ( Figure 3A). The relative abundance of actinomycetes among the dominant bacteria in the original grassland soil was 30.01%, the relative abundance of Acidobacteria was 29.52%, and the relative abundance of Proteobacteria was 17.57% ( Figure 3B). The dominant phylum in plough soil was Proteobacteria, with a relative abundance of 31.22%; additionally, the relative abundance of Actinomycota was 8.73%, and the relative abundance of Acidobacteria was 21.42% ( Figure 3C). The dominant phylum in forest was Acidobacteria, with a relative abundance of 35.7%; additionally, the relative abundance of Proteobacteria was 20.53%, and the relative abundance of Actinomycota was 15.8% ( Figure 3D).
From the perspective of the overall composition of the fungal community structure, all OTUs belong to 35 bacterial phyla, and the sequences that cannot be classi ed to a known phylum level are uniformly classi ed as "others". From the relative abundance of all levels of the three land-use patterns, the dominant phyla in the sample were Ascomycota, Basidiomycota, and Zygomycota ( Figure 4A). The relative abundance of Ascomycota is grassland was 62.74%, making it the dominant soil fungi; additionally, the relative abundance of Basidiomycota was 2.60%, and the relative abundance of Zygomycota was 0.86% ( Figure 4B). After the grassland was converted to plough, the dominant mycoplasma was still Ascomycota, with an abundance of 46.63%, the abundance of Basidiomycota was 11.87%, and the abundance of Zygomycota was 7.28% ( Figure 4C). The dominant phylum of the forest was Basidiomycota, with an abundance of 76.68%; it was followed by Ascomycota, with an abundance of 15.90%, and Zygomycota, with an abundance of 1.18% ( Figure 4D).

Functions of soil bacterial and fungal communities in different land-use modes
Using the PICRUSt function prediction software to analyse the soil bacterial community functions in different land-use patterns, it can be seen from Figure 9A that the bacterial community functions are mainly amino acid transport and metabolism; energy production and conversion; signal transduction mechanisms; cell wall/membrane biogenesis; transcription; carbohydrate transport and metabolism; inorganic ion transport and metabolism; translation, ribosomal structure and biogenesis; lipid transport and metabolism; posttranslational modi cation, protein turnover; coenzyme transport and metabolism; secondary metabolite biosynthesis, transport and catabolism; nucleotide transport and metabolism; defence mechanisms; cell cycle control, cell division, chromosome partitioning; RNA processing and modi cation; and chromatin structure and dynamics. It can be seen from Table 3 that except for the three functions of intracellular tra cking, secretion, and vesicular transport, cytoskeleton, and extracellular structures, there were no signi cant differences in the other functions. Note Mean values (means ± SD, n=6) , Significant levels are indicated at the *P < 0.05; **P < 0.01.
Using FUNGuild software to analyse soil fungal community functions under different land-use patterns, it can be seen from Figure 9B that the fungal community functions in the three land patterns are: ectomycorrhizal; animal pathogen; endophyte; dung saprotroph; plant pathogen; arbuscular mycorrhizal; fungal parasite; endomycorrhizal-plant pathogen; bryophyte parasite-ectomycorrhizal; and clavicipitaceous endophyte-plant pathogen. As seen from Table 4, there are signi cant differences in the functions of the ve communities in terms of ectomycorrhizal, animal pathogen, endophyte, dung saprotroph, and fungal parasite. After the grassland was transformed into forest, the ectomycorrhizal functional group increased signi cantly. After the grassland was transformed into plough, the functional groups of animal pathogens, endoparasites, and faecal saprophytic organisms increased signi cantly. Redundancy analysis of soil bacterial and fungal communities and physicochemical properties in different land-use patterns The relationship between soil physical and chemical properties and the community composition of bacteria and fungi at the OTU level was analysed using CCA, and the results are shown in Figure 10.

Discussion
Impact of land use on soil bacterial and fungal community diversity The results of this study show that different land-use patterns signi cantly change the soil bacterial Ace and Chao1 indexes. Compared with grasslands, the soil bacterial richness indexes of plough and forests increased signi cantly, but the Shannon and Simpson indexes did not change ( Table 2), indicating that after the grassland was changed to forest and plough, the soil bacterial community richness increased signi cantly, while its uniformity did not change. Compared with the bacteria, the Shannon, Simpson, Ace, and Chao1 indexes of the three land-use soil fungi were signi cantly different. After the grassland was converted to forest and plough, the diversity of the soil fungal community increased signi cantly ( Table  2). This result may be due to the increase in soil fungal diversity after conversion to forest due to abundant litter. This result is consistent with previous research, where afforestation often stimulates the growth of soil fungal communities [15] , while soil bacteria appear to be less sensitive to land use [16,17] . According to research reports, bacterial community structure, diversity and biomass are more resistant than those of fungi [18] . This difference may be because bacteria can produce a wider range of metabolites to adapt to the new environment. In contrast, fungi depend to a large extent on the presence of their hosts [19] , so the structure and diversity of fungal communities have more dramatic changes based on land use.

Effects of land-use patterns on soil bacterial community composition
At the gate level, the dominant phyla in the three types of soil are Proteobacteria, Acidobacteria, and Actinomyces, which can account for more than 80% of the total bacterial community in each soil sample.
The community structure results are consistent [20,21] . However, this study also found that when the grassland was changed to forest and plough, the abundance of its dominant bacteria changed signi cantly.
This study found that the relative abundance of grassland soil actinomycetes was the highest. After conversion to forest and plough, the soil actinomycete content decreased signi cantly. Several studies have shown that actinomycetes are the most widely distributed in the four herbaceous vegetation soils, and their relative abundance is signi cantly higher than that of forests and ploughs; actinomycetes are the dominant mycophytes in grassland soils [22,23] . Actinomyces can degrade cellulose and chitin, which is the main source of the soil nutrient supply. It can decompose more di cult-to-decompose organic carbon by in ltrating its hyphae into large plant tissues, and the spores produced can resist unfavourable external environmental conditions and are considered to be dominant in harsh and stressful soil conditions [24] . The relative abundance of Proteobacteria was lowest in grassland soils. Liu et al. [25] found that the relative abundance of Proteobacteria may be controlled by the difference in soil nutrients. Soil total phosphorus is the main factor affecting the distribution of Proteobacteria, with an interpretation rate as high as 85.3%. Other studies have found that Proteobacteria are relatively abundant in nutrient-rich soils but also relatively abundant in soil that are nutritionally poor [26] . The relative abundance of Planctomycetes was highest in grasslands. Fei et al. [27] found that there was a signi cant positive correlation between oating mould and soil total nitrogen content. The total nitrogen content of grassland was the highest among the three land patterns in this study, making the relative abundance of oating mould the highest in grassland soil. Fu et al. [27] found that the oating fungus phylum occupied a certain proportion in the bacterial community in the green space of the fth ring road in Beijing, re ecting the alkaline and nutrient-poor soils in the study area, and the soil biological activity was low.
The results of this study indicate that the relative abundance of Acidobacteria was the highest after the grassland was transformed into forest land, and it became the dominant bacteria in soil. Acidobacteria can grow on the medium using plant polymer as a substrate, indicating that Acidobacteria play an important role in the degradation of plant residues and degradation of forest litter [29] . Pankratov et al. [30] found that although the Acidobacteria degradation function is not as good as other known cellulosedegrading bacteria, it has strong resistance to stress and can survive in cold northern soils, which plays an important role in cellulose degradation under these conditions. Based on this study, there is less litter content in the grassland and plough patterns, and the forest litter content is signi cantly higher than that in forest land and grassland. Therefore, the content of insoluble matter in litter is also high. As a result, they are more susceptible to litter composition, making Acidobacteria more abundant in forests. Maestre et al. [31] found, in Northeast China, that the abundance of Acidobacteria decreased in the order of the soils of Yanji, Siping, and Tongliao, which may be the result of increased soil drought. In this study, the water content was the highest in forest, which may also be the cause of the increase in soil acid bacilli. Therefore, it can be concluded that the abundance of Acidobacteria is mainly related to the composition and content of litter, which is considered to be related to soil moisture content.
After the grassland was transformed into plough, the soil actinomycete content decreased signi cantly.
Clegg et al. [32] found that the addition of inorganic nitrogen reduced the abundance of actinomycetes compared with the non-fertilized grassland soil. In this study, due to the application of chemical fertilizers throughout the year, the soil NO 3 --N content increased, the soil structure changed, and the relative abundance of soil actinomycetes decreased. Therefore, it can be concluded that the main reason for the decrease in the relative abundance of actinomycetes after grassland conversion to plough may be due to the increase in soil nutrient content in plough. After the grassland was transformed into plough, the relative abundance of Proteobacteria signi cantly increased and became the dominant Mycoplasma in plough. Numerous studies have shown that the relative abundance of Proteobacteria in plough soils makes it the dominant phylum [33,34] . When substrates with high resource availability exist in the soil, Proteobacteria are more abundant in the soil [35] . Li et al. [36] found that Proteobacteria was the main group of saline-alkali soils. The soil in this study was alkaline, and it was also veri ed that Proteobacteria was the main dominant community in alkaline soil. Michael et al. [37] found that a dominant phylum was transformed from an actinomycete to a proteobacterium after conversion from a pasture to a plough.
Pascault et al. [38] found that Proteobacteria had the fastest decomposition rate of beans in plough, indicating that Proteobacteria had a good effect on the degradation of crop residues. Plough soil Proteobacteria were the most abundant in this study, which shows that it plays an important role in the crop decomposition process. After the grassland was converted to plough, Gemmatimonadetes increased. Several studies have shown that Bacillus has a relatively high abundance in plough soils [39] . Gemmatimonadetes is an alkalophilic microorganism and can produce spores, which can resist dehydration and adapt to drought and extreme environmental conditions. Some Gemmatimonadetes species have strong nitrogen-xing effects and play an important role in the biological control of the production and release of plant hormones and soil-derived plant pathogens (such as fungi) [40,41] .
Mahoney et al. [42] found that winter wheat soil bacterial communities were rich in Gemmatimonadetes.
Monreal et al. [43] found rich Gemmatimonadetes communities in rapeseed agricultural soils in Ottawa, Canada. This result shows that Gemmatimonadetes has a higher abundance in plough soil. Plough is an arti cial ecosystem with a monoculture. Due to the effect of external nitrogen application, the available nitrogen content in the soil is high. Because Gemmatimonadetes has a strong nitrogen-xing capacity, its content is highest in ploughs. After the grassland was converted to plough, Bacteroidetes increased. Bacteroidetes are mainly anaerobic or facultative anaerobic bacteria and can be found in a variety of habitats, including soil, sediment and seawater. Li et al. [44] studied the black soil plough in Northeast China and found that Bacteroidetes were the dominant bacteria in the soil. These ora were found to be the most common ora in plough and forest soil. Turner et al. [45] and Donn et al. [46] also found that the abundance of Bacteroidetes was higher in the eld soils of wheat and pea. Gkarmiri et al. [47] and Xiao et al. [48] found a large number of Bacteroidetes in soils in rapeseed elds, alfalfa and other plants.
Bergkemper et al. [49] found that the relative abundance of Bacteroidetes was positively related to available phosphorus, and available phosphorus may be one of the important factors affecting the bacterial community. In this study, due to the application of chemical fertilizers to the plough land, the available phosphorus content was the highest, and the relative abundance of Bacteroidetes increased in the plough soil.

Effects of land-use patterns on the composition of soil fungal communities
Among the three land-use patterns, the soil fungal groups were mainly Ascomycota, Basidiomycota, and Zygomycota. Ascomycota was the dominant phylum in grassland soil. Cao Hongyu et al. [50] also found that the grassland soil fungus Ascomycota accounted for the highest abundance, which was mainly due to the faster evolution rate of Ascomycota, drought resistance and radiation resistance, suitability for bare sand with a low vegetation canopy and harsh living environments such as land and grassland [51] .
Zhang et al. [52] found that Ascomycota was the dominant bacteria in the most primitive grassland, and its dominant orders are mainly Hypocreales and SoCCAriales. Its abundance in forestland decreases signi cantly with increasing age. Most SoCCAriales are saprophytic, usually found on faeces or rotten plants. In our study area, animal dung was found in grazing grasslands, and animal and human dung in plough lands are common fertilizers. Therefore, Ascomycetes become the dominant bacteria in grasslands and ploughs.
After the grassland was transformed into forestland, Basidiomycota increased signi cantly in the forestland and became the dominant phylum in the soil. This result is consistent with previous studies.
After 29 years of pine planting in the wasteland, the relative abundance of Basidiomycota increased from 10.9% to 68.7% [24] . During the fungal succession of the Damma glacier fore eld in central Switzerland, it was found that the community dominated by Ascomycota became a community dominated by Basidiomycota [53] . Basidiomycota are the dominant ectomycorrhizal species, which are more abundant in oak and oak forests. The genera Mycelium and Lactobacillus in Basidiomycota are common mycorrhizal fungi in forest soils, which can be symbiotic with Pinus sylvestris var. Mongolica and thus account for a large proportion of soil fungi [54] . Other Basidiomycota ora, especially white rot fungi, can breakdown litter with high lignin and aromatic substrates. However, only a small group of fungal groups have the ability to secrete enzymes that catalyse the degradation of complex macromolecules such as lignin [55] , and they are largely con ned to the Agaricus species in Basidiomycota [56] . In this study, litterfall increased signi cantly after grassland afforestation, requiring more decomposing bacteria, which is also the reason for the increase in soil Basidiomycota.
After the grassland was transformed into plough land, the dominant fungal phylum was still Ascomycetes, but Zygomycota signi cantly increased. Part of Zygomycota is a saprophytic fungus that mainly decomposes plant litter and changes soil chemical properties. Angela et al. [56] found that the majority of Zygomycota in the genus Zygomycota were predominant in Colombia, which is consistent with this study. Qian et al. [58] found that the relative abundance of soil fungal Zygomycota increased after grass growing in apple orchards, indicating that grass would affect the relative abundance of soil Zygophyta. It will convert matter to humus and provide a carbon source to increase soil organic carbon.
Zygomycota are mostly saprophytic, which can make good use of the saprophytic environment. Zygomycota are also pathogenic bacteria, which can be parasitic when plants are weak, easily causing postpartum diseases. The study found that the relative abundance of Zygomycota had a signi cant positive correlation with the soil nitrate nitrogen content and with the increase in the soil nitrate nitrogen content. The highest nitrate nitrogen content in ploughs in this study may cause an increase in Zygomycota. Li et al. [59] found that the relative abundance of Zygomycota in apple eld and corn eld was greater than that in an intercropped eld, and the relative abundance of soil fungi in different categories and subgenera was also different, indicating that, due to the differences in crop roots, residues, secretions, and crop management and maintenance, the method affects the physical and chemical properties of the soil, and then changes the microbial species composition and its structure.
Although the species composition of soil bacterial communities is similar between different land-use patterns, the relative abundance of soil bacterial phyla and genera may be different because of different plant patterns and differences in the form and content of nutrients provided to the soil [60,61 ] .

Effects of land-use patterns on functional changes in soil bacterial and fungal communities
After land-use change, the function of the inherent bacterial community changed due to differences in the aboveground vegetation community, surface litter composition, decomposition rate, and degree of interference from human activities. In this study, it was found that except for a small number of three functions, including intracellular tra cking, secretion, and vesicular transport, cytoskeleton, and extracellular structures, there were signi cant changes in the functions of other bacterial communities (Table 3). This result shows that the change in land-use patterns has a signi cant effect on the function of surface soil bacterial communities. Zhang et al. [62] found that during the transition from secondary forest to larch plantation, due to soil acidi cation and a reduction in effective nutrient content, land-use patterns had a greater impact on soil bacterial communities. We found that after a long-term change in the original grassland, the bacterial function in the 0-20 cm soil changed, and the bacterial function of the plough decreased compared with that of the grassland and forest ( Table 3). The nutrient conversion and return and the litter quality and quantity of forest and grassland were higher than those of plough, which was consistent with the research results of many scholars [63,64] .
After land-use changes, the functional groups of fungal communities changed signi cantly. The grassland was transformed into forest Inocybe. The abundance of Inocybe signi cantly increased. The species belonged to Basidiomycota and was an ECM fungi. ECM fungi are reported to be most widely distributed in trees in northern temperate regions [65] . Because ECM fungi are strongly affected by the host, their richness is positively related to the proportion of ECM plants and species richness. In northern temperate deciduous forests, ECM fungi accounted for 34.1% of all taxonomic units, while in grasslands, they accounted for only 11.9%, which re ected the lack of host plants in grassland ecosystems [66] . After the grassland was converted to plough, the relative abundance of Mortierella, Chaetomium and Microdochium increased signi cantly, and they were common saprophytic fungi in soil. Liang et al. [67] studied a vineyard and found that the most abundant fungal genera included Mortierella, Chaetomium and Microdochium, which may be considered to play a key role in planting soil. Li [68] found that the inoculation of corn with Mortierella signi cantly increased soil nutrient transformation, increased the content of indole acetic acid and abscisic acid in corn roots, and increased the biomass of corn seedlings. In addition, it had the ability to decompose cellulose, hemicellulose and lignin, increase carbon nutrients, increase soil organic matter and nutrient content, and dissolve phosphorus in the soil.
Therefore, it has been recognized as a bene cial soil microorganism by the genus Sporella. Most microsporum fungi are saprophytic, some species are parasitic or symbiotic, and most are phytopathogenic [69] . Therefore, the fungal functions of plough soils are mainly saprophytic, parasitic, animal pathogens, and mycorrhizal.

Effects of soil physical and chemical properties on soil microbial community composition
Land-use and management patterns will change the type of vegetation on the ground and then affect the physical and chemical properties of the soil [70,71] . Changes in soil physical and chemical properties will affect the structure and composition of soil microbial communities. Consistent with most other studies, pH is an important factor affecting soil microbial community structure. Barka et al. [72] found that there was a signi cant positive correlation between Actinomycetes and soil pH. Actinomycetes grew healthily in soils with a neutral pH and grew fastest between pH 6 and 9. Rousk et al. [73] found that both bacterial and fungal communities were affected by soil pH, but bacterial communities were more affected by pH than were fungal communities, which may be due to the relatively narrow optimal pH range for bacterial growth, while the pH range for fungal growth is very wide. Although soil pH has a direct impact on microbial community structure, soil pH can also indirectly change microbial communities through other variables, such as nutrient utilization and organic carbon content.
As an indispensable source of energy and nutrients for microorganisms, SOC plays an important role in shaping the microbial community and signi cantly changes the proportion of bacteria and fungi in the soil [74] . However, the SOC content in this study may not have caused changes in soil microbial communities. In this study, NO 3 --N was the most important factor affecting the soil bacterial and fungal communities ( Figure 10). Nitrogen restrictions are common in most terrestrial ecosystems and often lead to erce competition between microorganisms and plants [75] . With the increase in nitrogen availability, the taxonomic and functional characteristics of soil microbial communities change, including the decrease in relative abundance of mycorrhizal fungi and the slow growth of bacterial groups. Due to the low soil nitrogen content and low litter mass in forestland, fungi appear to be the main decomposers of complex litter and soil organic matter and have largely affected related bacterial communities and their activities [76] . Soil moisture is also an important limiting factor that strongly affects soil microbial communities [77] . In this study, MC plays a key role in soil fungal diversity. Not only can it protect soil organic matter from decomposition and leaching by combining with aggregates, it can also provide a larger surface area for the growth of soil microorganisms [78] .
In this study, the TP and AP contents of plough and forestland were signi cantly higher than those of grassland. The reason may be that the interception of rainwater by the forest canopy makes the surface runoff smaller, the soil surface organic matter and mineral nutrients are retained, and the loss is less. The arti cial fertilization in the plough compensates for the nutrients in the soil. Other studies have shown that under eutrophic conditions, the limiting effect of phosphorus on the original microbial community has been greatly reduced, and the metabolic activity of microorganisms has changed, which may change the species composition of microorganisms [79] . However, the grassland is not supplemented with external nutrients, and the growth of vegetation has absorbed phosphorus in the soil, which ultimately results in lower total phosphorus and available phosphorus in the soil. He et al. [80] found that P is the most critical contributor to differences in fungal communities, and phosphorus in forestland is usually less than that in managed ecosystems due to fertilization. Therefore, although the exact mechanism is not yet clear, P may be an important driving force for the construction of soil fungal communities across land-use types. DNA collection and high-throughput sequencing Genomic DNA was isolated from 0.5 g of each pooled soil sample from each sample plot (n = 18) with the PowerSoil DNA Isolation Kit per the manufacturer's instructions. The extracts of three technical repeats were mixed into a single DNA sample. Extracted genomic DNA was detected by 1% agarose gel electrophoresis. PCR was carried out on a GeneAmp 9700 PCR system. Based on previous reports, the primers 338F (5'-ACTCCTACGGGAGGCAGCA-3')-806R (5'-GGACTACHVGGGTWTCTAAT-3') were used for the 16S rRNA genes. Ampli ed products were detected by 2% agarose gel electrophoresis and recovered from the gel using the AxyPrep DNA gel extraction kit, washed with Tris-HCl, and veri ed by 2% agarose gel electrophoresis. PCR products were quanti ed using the QuantiFluorTM-ST uorometer, and the samples were adjusted as needed for sequencing. Sequencing was conducted by Shanghai Majorbio Biopharm Technology (Shanghai, China) using an Illumina MiSeq platform.

Conclusions
Processing of sequencing data The raw sequence les were analysed and quality-ltered using QIIME (version 1.9.1) with the following criteria: (i) the 250-bp reads were truncated at any site receiving an average quality score of b20 over a 50-bp sliding window; (ii) the exact barcode matching two nucleotide mismatches in primer matching reads containing ambiguous characters were removed; and (iii) only sequences with N10 bp overlap were assembled according to their overlap sequence. Reads that could not be assembled were discarded. The chimeric sequences were identi ed and removed using UCHIME software. The operational taxonomic units (OTUs) with a 97% similarity cut-off were clustered using UPARSE software. The representative sequence of each OTU was taxonomically classi ed by the Ribosomal Database Project (RDP) classi er against the SILVA (SSU123) database for 16S rRNA and the UNITE database for ITS rRNA using a con dence threshold of 70%. The sequencing depth of the soil bacteria and fungi in all samples was N98%, indicating that they were reliable sequencing results.

Statistical analyses
Mothur software was used to calculate the community richness parameters (Chao1, Ace index) and community diversity parameters (Simpson, Shannon index) as part of the alpha diversity analysis. PCoA and h-cluster analysis are based on the Bray-Curtis matrix and were implemented using R software. The bioenv method was used to test the soil environmental factors, and the environmental factors with signi cant differences were selected for CCA. One-way analysis of variance (ANOVA) was used to analyse the differences in the diversity of both soil bacterial and fungal communities among the plough, grassland, and forest sites. Tukey's HSD (honestly signi cant difference) test was used for multiple comparisons when the homogeneity of variance test was successful, and signi cance was observed at P = 0.05. Stepwise regressions were performed to identify the best independent soil factors affecting soil bacterial and fungal diversity. One-way ANOVA, Tukey's HSD test and stepwise regressions were conducted using SPSS 16.0. The functions of bacteria and fungi were analysed using PICRUSt and FUNGuild function prediction software, respectively.

Declarations
Ethics approval and consent to participate Sampling permission has been obtained for qiqihaer in Heilongjiang Province, Northeastern China and eld studies were conducted in accordance with local legislation.

Consent
Availability of data and materials All data generated or analyzed during this study are included in this published article and its supplementary information les. The raw data are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests. Note: in the gure above, different colors represent different groups, the Numbers of overlapping parts represent the number of species that are common to multiple groups, and the Numbers of nonoverlapping parts represent the number of species that are unique to each group.In the gure below, the abscissa is the number of common or unique groups, and the length of the horizontal column above represents the corresponding number of species Differences in the level abundance of bacterial phylums in different land use patterns Note: The Y axis represents the name of a species at a certain taxonomic level, the X axis represents the average relative abundance of different groups of species, and the columns of different colors represent different groups; the far right is the P value, * 0.01 <P ≤ 0.05, ** 0.001 <P ≤ 0.01, *** P ≤ 0.001. The same below.