Land-use conversion, habitat destruction, climate change, and other intensive anthropogenic activities strongly alter natural ecosystems (Pabst et al., 2013). Over 38 % of all natural landscapes worldwide have been converted to managed systems (Holland et al., 2016). Thus, ecological processes in the conversions should be elucidated to predict the development and sustainability of ecosystem services. Arid and semiarid areas account for 30 % of the total land area worldwide (Ran et al., 2014). These areas are also known for their ecological fragility and resource specificity. The conversion of natural grassland into agricultural fields is a dominant human perturbation commonly occurring in semiarid areas. In the western part of Northeast China, large grassland areas have been converted to upland fields or paddy fields because of high income from farming. This conversion can cause progressive and cumulative soil disturbances and is thus considered a key factor that affects land desertification development, reduces biodiversity, and alters ecological processes in terrestrial ecosystems (Sala et al., 2000). Intensive soil perturbations can also change intrinsic soil properties by promoting soil nutrient cycles and modifying soil physical, chemical, and microbiological properties and the structures of dwelling microbial communities in local ecosystems (Lauber et al., 2008; Lin et al., 2011, 2013; Kuramae et al., 2012). Therefore, land-use or land-cover change is an important factor affecting soil quality. Soil microbial diversity and community structure function as sensitive indicators of soil health and quality and hence rapidly respond to land-use conversions (He et al., 2008).

The effects of land-use conversion on soil physical, chemical, and microbiological properties have been comprehensively investigated. Land-use conversion can elicit significant and long-term effects on the moisture, texture, aeration, pH, nutrient status, microbial biomass, and enzymatic activities of soil largely because of the changes in plant community composition and different management practices across land-use types (Murty et al., 2002; Nishimura et al., 2008; Rahman et al., 2008; Wallenius et al., 2011). However, the effects of land-use conversion on soil biology have not yet been sufficiently assessed. Although land-use conversion has been confirmed to affect the structure of soil microbial communities significantly (Steenwerth et al., 2002; Johnson et al., 2003; Lauber et al., 2008), changes in the abundance of soil-specific taxonomic groups modified during land-use conversion and alterations in plant cover and soil properties remain poorly understood. Soil microbial communities are structured by climate, geology, land use, and soil physical and chemical factors, such as soil organic matter, porosity, and pH (Thomson et al., 2010, 2015; Griffiths et al., 2011; Hartmann et al., 2012). Land-use conversion may lead to unfavorable modification of several environmental variables, which can indirectly or directly affect soil microbial diversity. Therefore, the consequences of changes in soil microbial diversity and community structure during land-use conversion should be investigated to provide relevant information on microbial responses to changes in the soil environment and interactions between soil microbial communities and soil physical and chemical properties.

Culture-independent techniques, such as phospholipid fatty acid biomarkers, polymerase chain reaction–denaturing gradient gel electrophoresis, clone libraries, fluorescence in situ hybridization, enterobacterial repetitive intergenic consensus, and terminal restriction fragment length polymorphism, have been widely employed to examine soil microbial community changes as a consequence of land-cover change, soil pollution, and fertilization (Baath et al., 2003; He et al., 2007; Berg et al., 2012; Suleiman et al., 2013; Thomson et al., 2015). However, these techniques yield unsatisfactory performances in terms of the accurate detection of the changes in ecologically important soil microbial groups under different soil environmental conditions because of resolution limits, fingerprint complexity, and highly abundant indigenous microbes in soil (Bruce et al., 2000; Quince et al., 2009). Conversely, DNA-characterization-based high-throughput methodologies, such as sequencing and subsequent sequence alignment, can directly detect microbial community composition and variations in species with low richness (Shi et al., 2014; Zhao et al., 2014). These methodologies have also been widely adopted to assess the genetic structure and diversity of microbial communities in various environments (Quince et al., 2009; Golebiewski et al., 2014; Liu et al., 2014; Portune et al., 2014; Ren et al., 2014; Singh et al., 2014; Thomson et al., 2015).

Considering the increasing number of reclamation activities in native meadow grasslands in Northeast China, we applied Illumina MiSeq sequencing to investigate the changes in the diversities, compositions, and relative abundances of the dominant taxa of soil bacterial communities in native meadow grassland and adjacent upland fields, paddy fields, and woodland habitats. We then classified the microbial species and compared their structures among different land-use types. We also analyzed the correlation of specific microbial phyla with soil physical and chemical properties. Thus, we could obtain relevant data to enhance our understanding of the effects of the use change of native grassland on soil microbiota and provide useful information for the sustainable use of ecosystem services in fragile grasslands. We also statistically analyzed our data to evaluate the differences in the structures of soil bacterial communities among different land-use types and to determine their correlations with soil physical and chemical properties. We hypothesized that soil bacterial communities respond to land-use conversion, and these responses are related to variations in soil properties induced by different management measures during changes. This study aimed (1) to determine the changes in the abundances of soil bacteria across land-use types, (2) to identify the alterations in the structure of soil microbial communities in each land-use type, and (3) to observe the responses of specific microbial groups to land-use changes and the correlation of these variations with soil physical and chemical properties.

The purpose of the present study was to evaluate the effects of land-use conversion on the composition and structure of soil bacterial communities and explore the relationship between the physical and chemical properties of the soil and specific phyla in the Horqin region. In this regard, UF, PP, and PF were formerly NMG, which is one of the dominant land-use types in the Horqin Sandy Land. Consistent with studies in other areas (Hartman et al., 2008; Wakelin et al., 2008), our results showed significant differences among the structures of the soil bacterial communities in the land-use types. These differences were related to changes in the physical and chemical properties of the soil (especially in pH and nutrient contents) and thus suggested that the soil properties were significant factors that shape the bacterial community structure. During the conversion from NMG to UF, PF, and PP, the vegetation cover and soil properties were changed according to how the land conversion was managed. Variations in the environment due to land-use conversion have a direct influence on the microbial community inhabiting the soil (Zhang et al., 2014; Mendes et al., 2015). In the Horqin region, the land-use conversion of NMGs usually involves plowing and harrowing the soil and the cultivation of crops or planting trees. This process dramatically altered the soil physical, chemical, and microbiological properties by destroying the vegetation and increasing the rates of organic matter decomposition (Su et al., 2004). Previous studies reported that changes in the soil properties are attributed to cultivation and fertilization of the soil (He et al., 2007; Wang et al., 2012; Shen et al., 2010). Furthermore, the changes in the soils may involve shifts in nutrient contents and soil quality, as the natural grassland or forest plantation areas are expected to accumulate litter deposits, which are relatively more recalcitrant than the crop debris deposited in agricultural areas (Lauber et al., 2008). In an agricultural system, a low input of organic matter, variations in temperature and precipitation, and the management of soils were reported to be related to soil properties and the abundance of organisms (Lauber et al., 2013; Hartmann et al., 2015).

PCoA and CCA showed differences among the structures of the soil microbial communities during the land-use conversion of NMG. CCA was performed on the soil environmental factors and relative abundances of the dominant taxa and indicated that the changes in the environment due to land-use conversion have direct and significant impacts on the community of microorganisms inhabiting the soil. Soil P contents, such as total P, available P, and microbial biomass P, were the most important factors for the microbial community structure in the UF sites. By contrast, the formation of the soil microbial community structures in the PF sites was significantly related to soil moisture, available N, and microbial biomass N (Fig. 5). However, NMG and PP had similar microbial community structures, which were mainly affected by pH, organic matter, total N, and microbial biomass C. Our results demonstrated that pH was an important factor that shaped the structure of the soil microbial community. Some studies showed that soil pH is generally related to the compositions and structures of the soil microbial communities across a geographic scale (Högberg et al., 2007; Jesus et al., 2009; Fierer and Jackson, 2006). Mendes et al. (2015) studied the effects of land-use systems on soil bacterial communities in the southeastern region of the Amazon, and they confirmed that pH is correlated with the structure formation of a bacterial community during land-use alternations. Wakelin et al. (2008) also found that pH influences the structural composition and functional capacity of the soil bacterial communities in Australian agricultural soils. The potential causes of this correlation are the following: (1) integration of pH with other soil physical, chemical, and microbiological properties and (2) a slight variation in the soil pH that could expose microbes to stress, or in other words, the sensitivity of the microbial cell to environmental change (Fierer and Jackson, 2006). Some studies indicated that several parameters other than pH and elements (e.g., Mg, Al, and Cd) are also correlated with specific bacterial groups and structural alternation of microbial communities (Pan et al., 2014; Jesus et al., 2009; Mendes et al., 2015).

The vegetation cover might affect the diversity and structure of the soil microbial community because the differences in plant community compositions can contribute to changes in litter quality and quantity, which then alter the content and cycling processes of soil nutrients (Miki et al., 2010). In this study, the four sites differed in terms of vegetation cover. Therefore, differences among the soil properties and structures of the microbial communities in different land-use types can be expected. In the agricultural sites of our study (UF and PF), the cultivation of rice and corn was performed through conventional management, while in NMG and PP, the cultivation was performed through no-tillage management. The long-term cultivation might affect the whole soil community. Souza et al. (2013) showed that the major differences among the soil biodiversity and microbial community structures were associated with tillage management. Differences among the soil structures induced by different tillage systems and land-use conversion influence the structures of microbial communities (Sessitsch et al., 2001; Peixoto et al., 2006). Considering that the four land-use types have the same soil type and the sampling sites were adjacent to each other, we can assume that land-use conversion is a principal driving force that alters the soil properties and microbial community structures. Jangid et al. (2011) investigated the rRNA gene diversity of soil bacteria at two successional gradients with different vegetation, and they found that bacterial diversity remained unchanged in both gradients; their results confirmed that land-use history was the main determinant of the species composition in soil microbial communities rather than soil properties and vegetation. Our results showed that NMG, UF, PF, and PP have nearly the same dominant phyla, genera, and species, although the relative abundances of the taxa were different from one another, which also confirms that the species composition of soil microbial communities was not distinctly affected by land-use change and duration. Similarly, Suleiman et al. (2013) reported that land-use history might affect the present-day structure of the soil microbial community because many highly resilient and resistant microbial populations unrelated to soil type, soil properties, or land use are widely present in the microbial community. Our results showed that the land-use types have nearly the same dominant phyla, genera, and species, although the relative abundances of the taxa were different from one another. For this reason, the influences of land-use conversion were suggested to have more quantitative effects than qualitative effects. In other words, the structure of the soil microbial community may be affected by changes in the relative abundances of specific groups. The structure may change in response to land-use conversion, probably because the abundance of specific bacterial taxa sensitive to soil environment variation are extremely low and thus difficult to detect.

The observed phyla are typically present in most soils. Their relative abundances varied among the four land-use types. Hence, land use possibly affected their abundances. Proteobacteria and Acidobacteria were the most abundant phyla in all of the samples. Proteobacteria are commonly described as abundant free-living bacteria in many habitats (Zhou et al., 2004; Drees et al., 2006; Yang et al., 2015). The Acidobacterial community is correlated with soil properties (e.g., pH, P, K, Ca, Mg, and Al contents), and its abundance can be altered by soil type and land-use change (Navarrete et al., 2013; Mendes et al., 2015). Our study showed a decrease in the relative abundance of the Acidobacteria group during the conversion of NMG to UF, PF, and PP. At the genus level, the relative abundance of the dominant genera (Acidobacteria bacterium, Blastocatella, and Sphingomonas) also decreased during this process. This effect indicated that bacterial community structures were altered during land-use conversion.

Our results confirm that land-use conversion directly alters soil properties and microbial community structures in soil. These changes may produce potential effects on the ecological functions of meadow grassland ecosystems. However, the effect of land-use conversion on the functional profile of soil microbes remains unknown. Therefore, further studies on microbial functional profiles in Horqin soils should be performed to understand the function of soil microbes and to evaluate the sustainability of land-use systems.

In this study, land-use conversion in the NMG of the Horqin Sandy Land elicits a distinct effect on the physical and chemical parameters of soil and the structure of soil bacterial communities. Differences among the structures of soil microbial communities are caused by land-use change and variability in land management practices, which directly influence soil physical, chemical, and microbiological properties.