SciELO - Scientific Electronic Library Online

 
vol.79 issue4Effect of geotextile and agrotextile covering on productivity and nutritional values in lettuceRepeatability and genotypic stability of agronomic characteristics in Panicum maximum Jacq. author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


Chilean journal of agricultural research

On-line version ISSN 0718-5839

Chil. j. agric. res. vol.79 no.4 Chillán Dec. 2019

http://dx.doi.org/10.4067/S0718-58392019000400531 

RESEARCH

Effect of different grazing intensities on bacterial community composition and diversity in rhizosphere and non-rhizosphere soils in desert steppe of China

Guogang Zhang1 

Jing Huang1 

Meiqing Jia2 

Wei Huang1 

Xin Sui1 

Zhongwu Wang3  4 

Guodong Han3  4 

1Tianjin Normal University, College of Life Sciences, Tianjin 300387, Tianjin, P.R. China.

2Tianjin Normal University, Key laboratory of Water Resource and Environment, Tianjin 300387, Tianjin, P.R. China.

3Inner Mongolia Agricultural University, College of Grassland, Resources and Environment, Hohhot 010019, Inner Mongolian, P.R. China.

4Inner Mongolia Agricultural University, Key Laboratory of the Ministry of Grassland Resources Education, Hohhot 010019, Inner Mongolian, P.R. China.

ABSTRACT

Degradation and desertification are extremely significant environmental problems in arid and semi-arid grassland ecosystems. Long-term overgrazing is the most fundamental cause of grassland degradation. We investigated relationships between grazing intensity and bacterial communities in non-rhizospheric and rhizospheric soils in desert steppe, including 0-10, 10- 20 and 20-30 cm depth soils, as well as Stipa breviflora Griseb., Cleistogenes songorica (Roshev.) Ohwi, Artemisia frigida Willd. and plant community rhizospheric soils. This involved simulating grazing intensities in a long-term localization experiment, using a randomized block design. The effects of grazing on non-rhizospheric soil bacterial abundance were reflected in the 0-10 cm layer, increasing under light grazing and decreasing rapidly under moderate and heavy grazing, mainly related to Bacillus. Bacterial abundance in dominant plant rhizosphere responded differently. In A. frigida Willd. Rhizosphere, it decreased with increasing grazing intensity (a trend repeated in mixed rhizosphere). Bacterial abundance in S. breviflora and C. songorica rhizosphere increased under light and decreased under moderate and heavy grazing. Thus, changes in the dominant plant rhizospheric bacterial community did not significantly affect bacterial abundance in mixed rhizosphere. Changes in the rhizospheric bacterial abundance mainly resulted from levels of the dominant species, Streptomyces and Arthrobacter. There were significantly different results for bacterial community structure. Specifically, grazing had a nonsignificant and significant impact on bacterial community structures in non-rhizospheric (FPERMANOVA = 1.38, p = 0.199) and rhizospheric (FPERMANOVA = 2.03, p = 0.012) soil, respectively, varying significantly among plants (FPERMANOVA = 1.9, p = 0.022). In conclusion, bacterial communities in rhizosphere were mainly affected by plant species and were more sensitive to changing grazing intensity than in non-rhizospheric soil.

Key words: Desert steppe; grazing intensity; non-rhizosphere; rhizosphere; soil bacteria.

INTRODUCTION

The desert steppe ecosystem of Inner Mongolia, which comprises 10.7% of the region’s grassland, is important for both livestock production and preservation of biodiversity (Jia et al., 2017a). Grazing has been one of the most important land-use methods across all Inner Mongolian steppe types for thousands of years. However, overgrazing has caused severe land degradation and desertification in the Inner Mongolian steppe. Compared to other types of grasslands, desert steppe accounts for 39% of total native grassland in Inner Mongolia (Lin et al., 2010a). Therefore, determining a suitable grazing intensity is of great significance in preventing soil degradation and the sustainable development of desert steppe.

Previous studies have considered the impact of grazing on desert steppe ecosystems. The effects of different grazing intensities on soil nutrients (Zhou et al., 2017), soil respiration (Zhao et al., 2017) and gas exchange (Li et al., 2018), plant community structure and productivity (Deng et al., 2014) have been widely studied. A study of the impacts of grazing on C components and total ecosystem organic C in the desert steppe of northern China found that grazing changed the relative distribution of C components in this arid desert grassland (Wang et al., 2017); the same study determined that light to moderate grazing (0.15-0.30 sheep ha-1 mo-1) was beneficial for soil nutrient accumulation in the desert steppe. Grazing intensity has been shown to alter fine-scale processes in desert steppe and cause divergent responses in the spatial distribution of vegetation and soil fertility (Lin et al., 2010b). Grazing also enhances the suppression effect of climatic aridity on seed production in Caragana stenophylla (Xie et al., 2016). Appropriate and efficient grazing exclusion has been found to cause desirable transitions in the plant communities of desert steppe rangelands; this is an available method for counteracting local grassland degradation and promoting rangeland sustainability (Deng et al., 2014).

Soil microorganisms represent the world’s largest reservoir of biological diversity: abundant 16S rRNA gene sequences have been amplified from soil DNA and RNA. They also constitute the most important and sensitive bioactive factor in soil, where they perform an irreplaceable role in the maintenance of healthy desert grassland ecosystems and the restoration of vegetation. Increased soil bacterial diversity not only improves soil ecosystem stability, but also helps to mitigate deterioration of the soil ecological environment (Gao et al., 2017). However, fewer studies have been reported concerning the effects of different grazing intensities on non-rhizospheric and rhizospheric bacterial communities in desert steppes (Eldridge et al., 2017).

A gradient of grazing intensity can represent desert steppe ecosystems with different desertification potentials, given the lack of a common method for assessing desertification (Lin et al., 2010a). Therefore, the aim of this study was to investigate the relationship between soil degradation and the bacterial communities of non-rhizospheric and rhizospheric soils in desert steppe. Based on observation of artificial controlled grazing sample plots for 13 consecutive years, the results may contribute to the restoration of soil ecological environment and functioning in degraded grassland, and to the optimization of grazing methods. The findings may also lay a theoretical foundation for the sustainable development of grassland soil ecosystems (Gao et al., 2017).

MATERIALS AND METHODS

Study site

The experiment was conducted in the Inner Mongolia Academy of Agricultural & Animal Husbandry Sciences, located in Shiziwang Banner (41°47’17” N, 111°53’46” E; 1450 m a.s.l.), Inner Mongolia Autonomous Region (IMAR), northern China (Figure 1) (Lin et al., 2010b; Cao et al., 2013). The steppe is characterized by a semiarid continental monsoon climate with a short plant-growing period from May to September (Jia et al., 2017a). The mean annual temperature is 3.4 ℃, and the three highest monthly mean temperatures are 21.5, 24.0, 23.5 °C in June, July and August, respectively (Lin et al., 2010b). The annual precipitation is approximately 280 mm with about more than 80% received from May to September (Cao et al., 2013). Annual average total sunshine time is 3118 h, and the annual mean amount of evaporation is 2300 mm (Cao et al., 2013). Soil in this area was classified as light chestnut soil (Haplic Calcisols, according to the FAO classification). The soil pH was 8.16, soil organic C was 16.29 g kg-1, total N was 1.49 g kg-1, available N was 24.57 mg kg-1, and available P was 3.15 mg kg-1. The plant composition of the study site, Stipa breviflora Griseb. desert steppe community, mainly comprising S. breviflora Griseb., Artemisia frigida Willd., and Cleistogenes songorica (Roshev.) Ohwi, accompanied by Convolvulus ammannii Desr., Aster altaicus Willd., Neopallasia pectinata (Pall.) Poljakov, Kochia prostrata (L.) Schrad., Caragana stenophylla Pojark., and Leymus chinensis (Trin.) Tzvelev. The plant photosynthesis type and composition percentage of the study site is showed in Table 1.

Table 1 Plant composition (%) on different types of grazing treatment in the study site. 

CK: Non-grazed enclosure; LG: light grazing; HG: heavy grazing; MG: moderate grazing.

Figure 1 Study site location, and two-dimensional schematic diagram of experimental treatments. 

Experimental design

The steppe desert where the study site is located has been grazed continuously since 1988, when herdsmen began to settle permanently in the area. A lack of effective management measures prior to the introduction of enclosures led to the steppe desert beginning to degrade (Cao et al., 2013). In May 2004, grazing management experimental plots were established in degraded pasture inside enclosures near the residential area, and in the unenclosed free-grazing area, using a randomized complete block design of four grazing intensities with three replicates (Figure 1). The area of each plot was 4.4 ha. A total of 12 plots were fenced and stocked with sheep at rates of 0, 0.15, 0.30, and 0.45 sheep ha-1 mo-1, defined as non-grazing (CK), light grazing (LG), moderate grazing (MG), and heavy grazing (HG), respectively (Wang et al., 2011a).

The grazing animals were 2-yr-old local Mongolian sheep, which were allowed to graze from 15 May to 15 November in each year. Sheep were not randomly reassigned each year but were retained in their originally assigned plots during the experimental period (May 2004 to date). The sheep were kept in the experiment for 3 yr, and then were replaced by 2-yr-old animals from the same flock (Wang et al., 2017). The daily grazing schedule extended from 06:00 to 18:00 h. During the grazing period, the test sheep were herded to the different sample plots each morning, and then returned to the sheepfold each evening (Cao et al., 2013). Water was provided to the sheep twice per day (early morning and dusk) in their pen, and salt was provided ad libitum over the entire grazing period. During the winter, all experimental sheep were kept in the enclosure and supplied with hay and grains in order to meet their energy maintenance requirements (Wang et al., 2011a; 2017).

Sampling

Five soil cores were randomly selected from each replicated plot and collected using a soil drill (5 cm diameter) in August of each of the years (2013-2016) (this article is based on 2013 data). Each soil core was divided into three layers, namely, non-rhizospheric soil depths of 0-10 (A), 10-20 (B) and 20-30 cm (C). Part of each sample was thoroughly mixed into one composite sample. Three plants representative of the study site (S. breviflora, C. songorica, and A. frigida) and plant community rhizospheric soil were also collected. The method of collecting rhizospheric samples was to uproot the plants, quickly shake off the loose soil around the roots, and then use a blade to gently peel off the soil attached to the roots. When collecting the three dominant plant rhizospheric soils, sampling cores were selected from multiple areas of a single plant species, in order to avoid contamination by other plant species. A 1 m × 1 m square in the study site was randomly selected, and a 10 cm × 10 cm wire box was then used to randomly select a sample square. The plant rhizosphere was then collected by scraping as described above and mixed together as plant community rhizospheric soil. The collected soil was stored in a clean, numbered, sealed plastic pocket, and the soil samples were then quickly transferred to the laboratory and stored at -20 °C for later use.

Isolation and culture of bacteria from soil

Soil cultivable bacteria were cultured with beef protein medium and isolated using the dilution plate method. DNA was extracted using a soil DNA kit (Sangon Biotech Co. Ltd., Shanghai, China) according to the manufacturer’s instructions. A polymerase chain reaction (PCR) was performed on PCR equipment (TC-960F; Blue Marlin, Zurich, Switzerland). The PCR reaction mixtures (50 μL) contained 5 μL 10 × Ex Buffer (plus Mg2+), 4 μL 2.5 mmol L-1 deoxynucleotide (dNTP), 2 μL each of forward and reverse primers (341F: 5'-TACGGGAGGCAGCAG-3'; 534r: 5'-ATTACCGCGGCTGCTGG-3'), 0.5 μL Taq Ex DNA polymerase (5 U μL-1), 10 μL DNA template (10 ng μL-1), and 26.5 μL sterile and DNA-free water.

The PCR protocol was as follows: 94 °C for 5 min (initial denaturation); 30 cycles of 94 °C, 94 °C for 1 min (denaturation), 55 °C for 45 s (annealing), and 72 °C for 45 s (elongation); and 72 °C for 10 min (elongation). Amplified PCR products were detected using 1.5% agarose gel electrophoresis and sent for Sanger sequencing (BGI, Beijing, China). After sequence alignments with reference sequences in GenBank (http://www.ncbi.nlm.nih.gov), we downloaded high-quality sequences for phylogenetic analysis.

Statistical analysis

One-way ANOVA and least significant difference (LSD) multiple comparison tests were used in post-hoc analysis of significant differences among the factors, using SPSS 22.0 (Qu et al., 2016). Origin 2017 (https://www.originlab.com/2017) was used for data visualization. Two-way ANOVA was used to analyze the effects of grazing or soil depth on bacterial abundance in non-rhizospheric soils, and of grazing or plant species on rhizospheric soil bacterial abundance. Principal coordinate analysis (PCoA) based on the Bray-Curtis matrix was used to analyze changes in the bacterial community structure. Permutational multivariate ANOVA (PERMANOVA) was used to test dissimilarities among different non-rhizospheric soil bacteria at different soil depths or under different grazing treatments. PERMANOVA was also used to test rhizospheric soil bacteria from different plants or different grazing treatments using both the Bray-Curtis and Jaccard distance methods via the vegan package (v. 2.5.2) in R software (v. 3.5.1) (The R Foundation for Statistical Computing, Vienna, Austria). Three non-parametric multivariate statistical methods were used to examine the effects on rhizospheric or non-rhizospheric bacterial communities of different grazing regime, soil depth, or plants, including analysis of similarities (ANOSIM), PERMANOVA, and the multiple response permutation procedure (MRPP) (Liu et al., 2015). The Pearson index was applied to calculate correlation among grazing intensity, plant abundance, litter, and bacterial species abundance using the vegan (v. 2.5.2) and corrplot (v. 0.84) packages in R (v. 3.5.1).

RESULTS

Bacterial abundance in non-rhizospheric and rhizospheric soils

The quantity of cultivable bacteria at 0-30 cm depth in the LG plot was 181.11 × 104 colony-forming units (CFU) g-1; this result was 11.26%, 6.54%, and 15.19% higher than in the CK, MG, and HG plots, respectively. The difference among the four treatments was nonsignificant. This result was more marked in the surface soil. The abundance of soil bacteria at 0-10 cm depth in the LG plot was 338.33 × 104 CFU g-1, which was 40% higher than in the CK plot, and 65.04% and 51.49% higher than in the MG and HG plots, respectively, which showed a significant decrease in soil bacterial abundance (Figure 2, Table 2).

Bacterial abundance was much lower at 10-20 and 20-30 cm soil depths than at 0-10 cm depth, and the results were different when compared with the composite 0-30 cm depth (Figure 2). Therefore, changes in bacterial abundance in the 0-10 cm surface soil determined the trend shown in the 0-30 cm soil depth. The study results also indicated that current grazing intensities are significantly affecting the abundance of bacteria in the surface soil (0-10 cm). Specifically, the abundance of non-rhizospheric soil bacteria in the LG plot was higher than at other grazing intensities or in the CK plot.

The bacterial abundance of S. breviflora (Stibre) and C. songorica (Cleson) rhizosphere (RS, RC) was the highest in the LG plot, at 428.33 and 303.33 × 104 CFU g-1, respectively (Figure 3, Table 3). Abundance then began to decrease with increasing grazing intensity. In particular, bacterial abundance of the Stibre rhizospheric soil in the LG plot was significantly higher than in the CK plot and other grazing intensities. Artemisia frigida (Artfri) rhizospheric soil (RA) bacterial abundance was insensitive to changes in grazing intensity compared with Stibre and Cleson rhizospheric soils. Compared with the CK plot, its bacterial abundance increased in the MG plot but decreased in the LG and HG plots; the difference was however nonsignificant (p > 0.05).

Compared with samples from the three dominant plants, bacterial abundance in the mixed rhizospheric soil (RCom) revealed a significantly different response to changes in grazing intensity. The abundance of RCom bacteria, when compared with the CK plot, gradually decreased with increasing grazing intensity; in the HG plot, this difference was significant (p < 0.05).

CK: Non-grazed enclosure; LG: light grazing; MG: moderate grazing; HG: heavy grazing. The line graph shows the number of bacteria genera in each soil depth. Each value refers to the average of three replicates, and error bars represent the standard error (SE). Different uppercase letters at the same depth indicate significant differences at the p < 0.05 level; different lower-case letters for the same treatment indicate significant differences at the p < 0.05 level, based on a protected least significant difference (LSD) test.

Figure 2 Abundance of cultivable bacteria measured in colony-forming units (CFU) in plots under four treatments at different non-rhizospheric soil depths. 

Table 2 Composition and abundance of cultivable bacteria at different non-rhizospheric soil depths and under different grazing intensities (1 × 104 CFU g-1 dry soil). 

CK: Non-grazed enclosure; LG: light grazing; MG: moderate grazing; HG: heavy grazing; CFU: colony-forming units.

Each value refers to the average of three replicates, and shows the standard error (SE). Different capital letters at the same depth indicate significant differences at the p < 0.05 level; different lower-case letters for the same treatment indicate significant differences at the p < 0.05 level, based on a protected least significant difference (LSD) test.

CK: Non-grazed enclosure; LG: light grazing; MG: moderate grazing; HG: heavy grazing. Stibre: Stipa breviflora; Cleson: Cleistogenes songorica; Artfri: Artemisia frigida; RCom: plant community rhizospheric soil. The line graph shows the number of bacteria genera in each plant rhizosphere. Each value refers to the average of three replicates, and error bars represent the standard error (SE). Different uppercase letters for the same plant indicate significant differences at the p < 0.05 level; different lower-case letters for the same treatment indicate significant differences at the p < 0.05 level, based on a protected least significant difference (LSD) test.

Figure 3 Abundance of cultivable bacteria measured in colony-forming units (CFU) in plots under four treatments in the rhizospheric soil of different plants. 

Structure and composition of bacterial communities in desert steppe

Tables 2 and 3 show the bacterial genera identified in the present study and their abundance in non-rhizospheric and rhizospheric soil. A total of 16 bacterial genera were identified, seven belonging to Firmicutes and Actinobacteria in non- rhizospheric soil, and nine belonging to Proteobacteria, Firmicutes, and Actinobacteria in rhizospheric soil. The genera common to both rhizospheric and non-rhizospheric soils were Bacillus, Arthrobacter, and Streptomyces (Tables 2 and 3).

Bacillus had an overall relative abundance of over 10% in non-rhizospheric soil treatments and could therefore be regarded as a dominant genus. The quantity of bacteria in non-rhizospheric soils was the largest in the 0-10 cm soil layer relative abundance of Bacillus was 80.56%, 67.57%, 76.09%, and 82.83% in the CK, LG, MG, and HG plots, respectively (Table 2). The change in Bacillus abundance was consistent with that in the non-rhizospheric surface soil (Figure 4; R = 0.82, p < 0.01), indicating that the change in bacterial abundance in the surface soil was mainly caused by the dominant genus, Bacillus. In addition, compared with CK, new genera, Streptomyces and Rhodococcus, emerged in LG and MG but disappeared in HG.

Table 3 Composition and abundance of cultivable bacteria in different plant rhizospheric soil under different grazing intensities (1 × 104 CFU g-1 dry soil). 

CK: Non-grazed enclosure; LG: light grazing; MG: moderate grazing; HG: heavy grazing; CFU: colony-forming units.

Each value refers to the average of three replicates, and shows the standard error (SE). Different uppercase letters for the same plant indicate significant differences at the p < 0.05 level; different lower-case letters for the same treatment indicate significant differences at the p < 0.05 level, based on a protected least significant difference (LSD) test.

According to the same standard analysis methodology (Table 3), the dominant genera in the rhizospheric soil were Streptomyces and Arthrobacter. For example, the relative abundance of Streptomyces in the CK, LG, MG, and HG plots of the plant community rhizospheric soil was 40.17%, 32.68%, 35.66%, and 44.13%, respectively. The relative abundance of Arthrobacter was 34.4%, 30.85%, 25.68%, and 16.83%, respectively (Table 3). These changes in Streptomyces or Arthrobacter with different grazing intensities were significantly correlated with changes in bacterial abundance in different rhizospheric soils (Figure 5; p < 0.05), indicating that the change in bacterial abundance in rhizospheric soil was also mainly caused by changes in the abundance of dominant genera.

*, **, *** Significant at the 0.05, 0.01 and 0.001 probability level, respectively. Tot.bio: Aboveground plants; Total: bacterial abundance; Arth: Arthrobacter; Bac: Bacillus; Jeo: Jeotgalicoccus; Koc: Kocuria; Oce: Oceanobacillus; Rho: Rhodococcus; Stre: Streptomyces; UN: Unknown genus.

Figure 4 Correlation analysis of grazing intensity, plant abundance, and bacterial abundance in non-rhizospheric soil. 

*, **, *** Significant at the 0.05, 0.01 and 0.001 probability level, respectively. Tot.bio: Aboveground plants; Total: bacterial abundance; Artfri: Artemisia frigida; Cleson: Cleistogenes songorica; Stibre: Stipa breviflora; Aqu: Aquincola; Arth: Arthrobacter; Bac: Bacillus; Jan: Janthinobacterium; Mas: Massilia; Pae: Paenibacillus; Rhi: Rhizobium; Sph: Sphingopyxis; Stre: Streptomyces; UN: Unknown genus.

Figure 5 Correlation analysis of grazing intensity, plant abundance, and bacterial abundance in rhizospheric soil.  

β-Diversity-based statistical tools were employed to explore structural variances in non-rhizospheric and rhizospheric soil bacterial communities. According to the PCoA, there were significant differences in the bacterial community structure of rhizospheric soil compared with non-rhizospheric bacterial communities at different soil depths and when mixed (Figure 6; PERMANOVA, p = 0.001). Further dissimilarity tests revealed significant differences in bacterial communities between non-rhizospheric and rhizospheric soil. PERMANOVA tests, based on both Jaccard and Bray-Curtis distances, were used to confirm the significance results (Table 4). This method showed that differences among grazing treatments and plant species were significant and non-significant in rhizospheric and non-rhizospheric bacterial communities, respectively, which indicated that the disturbance effect of grazing on the former was greater.

On the other hand, dissimilarity tests for each treatment using MRPP, ANOSIM, and PERMANOVA methods based on Bray-Curtis distance, conducted to compare differences among soil depths or plant species under the same grazing treatment, showed a significant difference between non-rhizospheric and rhizospheric bacterial communities, in that the former decreased from CK (FPERMANOVA = 11.36, p = 0.003) to MG (FPERMANOVA = 1.12, p = 0.389), but the latter decreased from CK (FPERMANOVA = 3.12, p = 0.005) to HG (FPERMANOVA = 0.68, p = 0.858) (Table 5). As Table 5 shows, the bacterial community structure in the rhizospheric soils of different plants differed among different grazing intensities. The results of the three test methods showed that a change in grazing intensity had a nonsignificant and significant effect on changes in the bacterial community structure Cleson and Artfri rhizospheric soils, respectively. The MRPP and ANOSIM results revealed a significant difference among different grazing intensities in the bacterial community structure of the Stibre and Com rhizospheric soils (Table 5).

NR: Non-rhizosphere; R: rhizosphere; CK: non-grazed enclosure; LG: light grazing; MG: moderate grazing; HG: heavy grazing. Plots a-d respectively show the differences in bacterial community structure between non-rhizospheric (a: 0-10 cm; b: 10-20 cm; c: 20-30 cm; d: three layers combined) and rhizospheric (Com: compound plant community rhizosphere) soils.

Figure 6 Principal coordinate analysis (PCoA) ordination of bacterial communities based on Bray-Curtis distances.  

DISCUSSION

The purpose of this study was to determine the effects of different grazing intensities on bacterial community abundance and structure in different plant rhizospheric soils and in different depths of non-rhizospheric soil. Our results showed that bacterial abundance and community structure were significantly different among different depths, but that grazing intensity had no significant impact. On the other hand, bacterial abundance and community structure were significantly different among the plant rhizospheric soils, and showed a significant response to grazing intensity. These results partially supported previous reports that changes in ecological processes caused by large herbivore grazing (Qu et al., 2016; Zhao et al., 2017) and changes in the soil microbial community appear to be more dependent on soil attributes or specific plant species (Aldezabal et al., 2015).

Table 4 Dissimilarity tests of soil bacteria communities using permutational multivariate analysis of variance (PERMANOVA) based on Jaccard and Bray-Curtis distances. 

*, **, *** Significant at the 0.05, 0.01 and 0.001 probability level, respectively.

NR: Non-rhizosphere; R: rhizosphere; A: 0-10 cm; B: 10-20 cm; C: 20-30 cm; CK: non-grazed enclosure; LG: light grazing; MG: moderate grazing; HG: heavy grazing; S: Stipa breviflora; C: Cleistogenes songorica; A: Artemisia frigida; Com: compound plant community rhizosphere.

Table 5 Dissimilarity tests for each treatment using multiple response permutation procedure (MRPP), analysis of similarities (ANOSIM) and permutational multivariate analysis of variance (PERMANOVA) methods based on the Bray-Curtis distance. 

*, **, *** Significant at the 0.05, 0.01 and 0.001 probability level, respectively.

NR: Non-rhizosphere; R: rhizosphere; CK: non-grazed enclosure; LG: light grazing; MG: moderate grazing; HG: heavy grazing; Stibre: Stipa breviflora; Cleson: Cleistogenes songorica; Artfri: Artemisia frigida; Com: compound plant community rhizosphere.

Variation in bacterial community in relation to grazing intensity

Grazing is one of the most extensive grassland management strategies. It has the potential to change the availability of the soil matrix utilized by bacteria, and so has an important impact on soil bacterial communities (Stark et al., 2015; Zhao et al., 2017). Grazing livestock and plants can affect soil bacterial ecology in desert steppe regions through a range of specific factors, including plant community composition and biomass, feces and urinary sedimentation, rhizosphere exudation, soil texture, and soil physical and chemical properties (Liu et al., 2015; Abdalla et al., 2018). These factors may have positive, neutral, or negative effects on bacterial community structure (Wang et al., 2011b; Hu et al., 2015; Liu et al., 2015; Mueller et al., 2017). Numerous studies have shown that properly managed grazing has a positive impact on soil bacterial diversity by altering soil physical and chemical properties in grassland ecosystems. However, when grazing intensity exceeds grassland limits, there are negative impacts on grassland aboveground vegetation or underground organisms (Xie et al., 2014; Qu et al., 2016).

The present study showed that bacterial abundance in non-rhizospheric soils at different depths did not vary consistently with increasing grazing intensity. Bacterial abundance at a depth of 0-10 cm responded more obviously to changes in grazing intensity. Compared with the CK plot, the LG treatment tended to increase bacterial abundance, while the MG and HG treatments reduced soil bacterial abundance (Figure 2). This indicated that the impact of grazing on the soil bacterial community mainly depends on grazing intensity (Zhao et al., 2017). In addition, Bacillus was clearly dominant in the non- rhizospheric soil bacterial composition, and was the main reason for the changes in soil bacterial abundance. The bacterial abundance of non-rhizospheric soil at the 0-10 cm depth increased in LG, probably because grazing had a significant effect on topsoil quality (Wang et al., 2018). We found that differences in grazing intensity had nonsignificant effect on the bacterial community structure of non-rhizospheric soil. Moreover, the response of non-rhizospheric soil bacterial communities to changes in grazing intensity was complex, and was rarely significantly correlated with aboveground plant responses (Figure 4). Our analysis indicated that non-rhizospheric soil bacterial communities were mainly affected by the properties of vertically distributed soils (Liu et al., 2015; Stark et al., 2015; Zhong et al., 2016).

Variation in the rhizospheric bacterial community in relation to grazing intensity and different plants

Most soil bacteria are heterotrophic, using plant exudates or decomposing plant material for growth (Bais et al., 2006; Bulgarelli et al., 2013; Edwards et al., 2015; Beckers et al., 2017). Therefore, plant species can affect bacterial community structures in the rhizosphere. The abundance and composition of rhizospheric microorganisms can vary among different plant species, resulting in rhizospheric bacterial communities with their own unique features (Rosenzweig et al., 2013; Murphy et al., 2016). Our study showed that plant species, grazing intensity, and the interaction between plants and grazing had significant impacts on the size and structure of rhizospheric bacterial communities (Tables 4 and 6). There were significant negative correlations between grazing intensity and the abundance of aboveground plants, the abundance of Artfri, and the bacterial abundance of both. Moreover, the abundance of Stibre and Cleson, their rhizospheric bacterial abundance, and the abundance of any genus in the rhizospheric soil was significantly related to changes in grazing intensity (Figure 5). In summary, the results showed that grazing may inhibit the abundance of aboveground plants, the abundance of Artfri, and the rhizospheric bacterial abundance of both in desert steppe; this would not be conducive to the development of desert steppe plants and rhizospheric bacterial diversity.

Our study showed that there was nonsignificant difference in rhizospheric bacterial abundance among the three dominant plants (p > 0.05). On the other hand, bacterial abundance in the Stibre and Cleson rhizosphere and in the plant community rhizosphere showed significant differences in the CK plot (p < 0.05). In addition, LG promoted bacterial abundance in the Stibre and Cleson rhizosphere. Particularly, rhizospheric bacterial abundance was significantly increased in Stibre, the most important species in this study site, primarily because of the obvious increase of Streptomyces in the LG plot (Table 4). The abundance of Arthrobacter and Streptomyces, dominant genera in the rhizospheric soil, significantly increased at 0-10 cm depth in non-rhizospheric soil in the LG plot (Table 2), which may be one of the reasons for the observed increase in bacterial abundance at this depth in the LG plot (Figures 2 and 3). In summary, these results indicated that the non-rhizospheric and rhizospheric bacterial communities interacted with each other through common genera that were dominant in their various communities.

Table 6 Two-way analysis of variance of the effects of grazing intensity and plant composition, grazing intensity, and soil depth on soil bacterial abundance. 

*, **Significance at the 0.05 and 0.01 probability levels, respectively. R: Rhizospheric soil; NR: non-rhizospheric soil.

The decrease in bacterial abundance in Artfri rhizospheric soil after grazing was consistent with the results of a study by Zang et al. (2017). In addition, there was a gradual decreasing of RA and RCom bacterial abundance with grazing intensity; this may be affected by grazing, the suppression of aboveground plants, and Artfri quantity (Figure 5). The rhizospheric bacterial community structure of Artfri was significantly different from that of Cleson and Com (Table 7). The effects of grazing on rhizospheric bacterial community structure varied with plant species. For example, although RA, RS, and RCom bacterial community structures varied significantly with grazing intensity, bacterial community structure changes in RC were limited, without significant changes (Table 5). The effect observed in this study seemed to indicate that the roots of different desert steppe plants may affect their rhizospheric bacterial communities in different ways, and there may also be differences in their response to external environmental changes and the corresponding results. Similar studies were conducted by Cleary et al. (2012), who observed that the variation in rhizospheric bacterial communities with depth in a salt marsh historically contaminated by mercury was mainly affected by plant species.

Table 7 Comparison of similarity index between each layer or plant across four grazing intensities based on the Bray- Curtis and Jaccard indices. 

*, **, *** Significant at the 0.05, 0.01 and 0.001 probability level, respectively.

A: 0-10 cm; B: 10-20 cm; C: 20-30 cm; NR: non-rhizosphere; R: rhizosphere; CK: non-grazed enclosure; LG: light grazing; MG: moderate grazing; HG: heavy grazing; Stibre: Stipa breviflora; Cleson: Cleistogenes songorica; Artfri: Artemisia frigida; Com: compound plant community rhizosphere.

Different impacts of grazing intensity on soil bacterial communities of non-rhizospheric or rhizospheric soil

In this study, PCoA showed significant differences in the bacterial community structure in rhizospheric and non-rhizospheric soil along grazing intensities (Figure 6). A total of 13 genera were isolated from the soil of the study site, belonging to three major groups: Proteobacteria, Firmicutes and Actinobacteria (Tables 2 and 3). This finding was consistent with the study of Jia et al. (2017b). The bacterial community species composition of rhizospheric soil was more abundant than that of non-rhizospheric soil. A possible reason for this was that high plant diversity could improve soil bacterial activity (Lange et al., 2015). Therefore, a reduced plant diversity at the study site due to grazing explained why RCom bacterial abundance gradually decreased with enhanced grazing intensity (Figure 3).

The PERMANOVA testing based on different beta diversity distance matrices (Bray-Curtis, Jaccard) was used to examine differences in bacterial communities at different soil depths and grazing intensities (Abdalla et al., 2018). The smaller the Bray-Curtis dissimilarity coefficient, the more similar was the relationship between the two bacterial community structures. The effect of grazing on soil bacterial community structure was similar to that for bacterial abundance. The bacterial community structures in non-rhizospheric soils were significantly different among soil depths, but grazing intensity did not significantly contribute to differences in bacterial community structure. On the other hand, grazing intensity, plant species, grazing intensity, and the combination all three had significant impacts on bacterial community structures in the rhizospheric soil. Previous studies and the present study revealed that variations in soil bacterial community structure are closely related to changes in the soil environment, plant species, and grazing intensity. Additionally, grazing affects soil bacterial abundance and community structure by affecting the major bacterial groups (Figures 4 and 5) (Eldridge et al., 2017).

CONCLUSIONS

Our study showed that the effect of grazing on bacterial abundance in non-rhizospheric soil was mainly reflected in the surface soil. Bacterial abundance increased in the light grazing (LG) plot, and decreased rapidly in the moderate (MG) and heavy grazing (HG) plots, mainly related to the abundance of Bacillus. The differences in how rhizospheric bacteria responded to changes in grazing intensity were mainly reflected in the decrease in bacterial abundance in the mixed rhizospheric soil with increasing grazing intensity. This was consistent with variations in bacterial abundance in the Artemisia frigida rhizosphere; bacterial abundance in the Stipa breviflora and Cleistogenes songorica rhizosphere increased in the LG plot and began to decrease in the MG and HG plots. Thus, the response of bacterial abundance to grazing intensity in the mixed rhizospheric soil was not significantly affected by changes in the rhizospheric bacterial community of the dominant plants. Variations in bacterial abundance in the rhizospheric soil with grazing intensity were mainly caused by changes in the abundance of the dominant Streptomyces and Arthrobacter species. In addition, the effects of grazing intensity on bacterial community structures in the non-rhizospheric and rhizospheric soils were significantly different: grazing had nonsignificant impact on bacterial community structures in non-rhizospheric soil, but had a significant effect on rhizospheric soil. There were also significant variations in bacterial community structures among plant species. In conclusion, bacterial communities in rhizospheric soil were mainly affected by different plant species, and were also more susceptible to changes in grazing intensity than those in non-rhizospheric soil.

ACKNOWLEDGEMENTS

Jing Huang contributed equally to this work and should be regarded as co-first authors. The study was supported by grants from National Natural Science Foundation of China (31770499, 31100330, 31500365, 31270502, 41403082, 31300386), Tianjin Science and Technology Support Program (15ZCZDSF00410) and Tianjin Natural Science Foundation (12JCYBJC19700, 18JCYBJC96500). We gratefully thank Guodong Han, Guogang Zhang, Meiqing Jia and many others for setting up the experiment. We also thank Wei Huang and Shuai Xu for DNA extraction and soil microbe culture.

REFERENCES

Abdalla, M., Hastings, A., Chadwick, D.R., Jones, D.L., Evans, C.D., Jones, M.B., et al. 2018. Critical review of the impacts of grazing intensity on soil organic carbon storage and other soil quality indicators in extensively managed grasslands. Agriculture Ecosystems & Environment 253:62-81. doi:10.1016/j.agee.2017.10.023. [ Links ]

Aldezabal, A., Moragues, L., Odriozola, I., and Mijangos, I. 2015. Impact of grazing abandonment on plant and soil microbial communities in an Atlantic mountain grassland. Applied Soil Ecology 96:251-260. doi:10.1016/j.apsoil.2015.08.013. [ Links ]

Bais, H.P., Weir, T.L., Perry, L.G., Gilroy, S., and Vivanco, J.M. 2006. The role of root exudates in rhizosphere interactions with plants and other organisms. Annual Review of Plant Biology 57:233-266. doi:10.1146/annurev.arplant.57.032905.105159. [ Links ]

Beckers, B., De Beeck, M.O., Weyens, N., Boerjan, W., and Vangronsveld, J. 2017. Structural variability and niche differentiation in the rhizosphere and endosphere bacterial microbiome of field-grown poplar trees. Microbiome 5:17. doi:10.1186/s40168-017-0241-2. [ Links ]

Bulgarelli, D., Schlaeppi, K., Spaepen, S., van Themaat, E.V.L., and Schulze-Lefert, P. 2013. Structure and functions of the bacterial microbiota of plants. Annual Review of Plant Biology 64:807-838. doi:10.1146/annurev-arplant-050312-120106. [ Links ]

Cao, J., Wang, X., Sun, X., Zhang, L., and Tian, Y. 2013. Effects of grazing intensity on soil labile organic carbon fractions in a desert steppe area in Inner Mongolia. Springerplus 2:S1. doi:10.1186/2193-1801-2-s1-s1. [ Links ]

Cleary, D.F.R., Oliveira, V., Gomes, N.C.M., Pereira, A., Henriques, I., Marques, B., et al. 2012. Impact of sampling depth and plant species on local environmental conditions, microbiological parameters and bacterial composition in a mercury contaminated salt marsh. Marine Pollution Bulletin 64:263-271. doi:10.1016/j.marpolbul.2011.11.020. [ Links ]

Deng, L., Sweeney, S., and Shangguan, Z.P. 2014. Grassland responses to grazing disturbance: plant diversity changes with grazing intensity in a desert steppe. Grass and Forage Science 69:524-533. doi:10.1111/gfs.12065. [ Links ]

Edwards, J., Johnson, C., Santos-Medellin, C., Lurie, E., Podishetty, N.K., Bhatnagar, S., et al. 2015. Structure, variation, and assembly of the root-associated microbiomes of rice. Proceedings of the National Academy of Sciences of the United States of America 112:E911-E920. doi:10.1073/pnas.1414592112. [ Links ]

Eldridge, D.J., Delgado-Baquerizo, M., Travers, S.K., Val, J., Oliver, I., Hamonts, K., et al. 2017. Competition drives the response of soil microbial diversity to increased grazing by vertebrate herbivores. Ecology 98:1922-1931. doi:10.1002/ecy.1879. [ Links ]

Gao, X., Han, G., and Zhang, G. 2017. Soil microbial community structure and composition of Stipa breviflora on the desert steppe. Acta Ecologica Sinica 37:5129-5136. doi:10.5846/stxb201605100902. [ Links ]

Hu, J., Hou, X.-Y., Wang, Z., Ding, Y., Li, X.-L., Li, P., et al. 2015. Effects of mowing and grazing on soil nutrients and soil microbes in rhizosphere and bulk soil of Stipa grandis in a typical steppe. Chinese Journal of Applied Ecology 26:3482-3488. doi:10.13287/j.1001-9332.20150820.005. [ Links ]

Jia, M., Huang, J., Meng, Y., Han, G., Zhang, G., and Yang, Y. 2017b. Composition and spatial distribution of cultivable soil bacteria in the Stipa breviflora desert steppe of Inner Mongolia. Journal of Tianjin Normal University (Natural Science Edition) 37:54-58+65. doi:10.3969/j.issn.1671-1114.2017.04.011. [ Links ]

Jia, M., Liu, C., Li, Y., Xu, S., Han, G., Huang, J., et al. 2017a. Response of fungal composition and diversity to simulated nitrogen deposition and manipulation of precipitation in soils of an Inner Mongolia desert steppe of northern China. Canadian Journal of Soil Science 97:613-625. doi:10.1139/cjss-2016-0135. [ Links ]

Lange, M., Eisenhauer, N., Sierra, C.A., Bessler, H., Engels, C., Griffiths, R.I., et al. 2015. Plant diversity increases soil microbial activity and soil carbon storage. Nature Communications 6:6707. doi:10.1038/ncomms7707. [ Links ]

Li, X., He, H., Yuan, W., Li, L., Xu, W., Liu, W., et al. 2018. Response of soil methane uptake to simulated nitrogen deposition and grazing management across three types of steppe in Inner Mongolia, China. Science of the Total Environment 612:799- 808. doi:10.1016/j.scitotenv.2017.08.236. [ Links ]

Lin, Y., Han, G., Zhao, M., and Chang, S.X. 2010a. Spatial vegetation patterns as early signs of desertification: a case study of a desert steppe in Inner Mongolia, China. Landscape Ecology 25:1519-1527. doi:10.1007/s10980-010-9520-z. [ Links ]

Lin, Y., Hong, M., Han, G., Zhao, M., Bai, Y., and Chang, S. 2010b. Grazing intensity affected spatial patterns of vegetation and soil fertility in a desert steppe. Agriculture Ecosystems & Environment 138:282-292. doi:10.1016/j.agee.2010.05.013. [ Links ]

Liu, N., Kan, H.M., Yang, G.W., and Zhang, Y.J. 2015. Changes in plant, soil, and microbes in a typical steppe from simulated grazing: explaining potential change in soil C. Ecological Monographs 85:269-286. doi:10.1890/14-1368.1. [ Links ]

Mueller, P., Granse, D., Nolte, S., Hai Thi, D., Weingartner, M., Hoth, S., et al. 2017. Top-down control of carbon sequestration: grazing affects microbial structure and function in salt marsh soils. Ecological Applications 27:1435-1450. doi:10.1002/eap.1534/full. [ Links ]

Murphy, C., Foster, B., and Gao, C. 2016. Temporal dynamics in rhizosphere bacterial communities of three perennial grassland species. Agronomy 6:17. doi:10.3390/agronomy6010017. [ Links ]

Qu, T.B., Du, W.C., Xia, Y., Yang, Z.M., Liu, D.B., Wang, D.L., et al. 2016. Impacts of grazing intensity and plant community composition on soil bacterial community diversity in a steppe grassland. PLOS ONE 11:e0159680. doi:10.1371/journal.pone.0159680. [ Links ]

Rosenzweig, N., Bradeen, J.M., Tu, Z.J., McKay, S.J., and Kinkel, L.L. 2013. Rhizosphere bacterial communities associated with long-lived perennial prairie plants vary in diversity, composition, and structure. Canadian Journal of Microbiology 59:494-502. doi:10.1139/cjm-2012-0661. [ Links ]

Stark, S., Männistö, M.K., Ganzert, L., Tiirola, M., and Häggblom, M.M. 2015. Grazing intensity in subarctic tundra affects the temperature adaptation of soil microbial communities. Soil Biology and Biochemistry 84:147-157. doi:10.1016/j.soilbio.2015.02.023. [ Links ]

Wang, H., Dong, Z., Guo, J., Li, H., Li, J., Han, G., et al. 2017. Effects of grazing intensity on organic carbon stock characteristics in Stipa breviflora desert steppe vegetation soil systems. Rangeland Journal 39:169-177. doi:10.1071/rj16007. [ Links ]

Wang, Z., Jiao, S., Han, G., Zhao, M., Willms, W.D., Hao, X., et al. 2011a. Impact of stocking rate and rainfall on sheep performance in a desert steppe. Rangeland Ecology & Management 64:249-256. doi:10.2111/rem-d-09-00033.1. [ Links ]

Wang, X., Song, N.P., Yang, X.G., Wang, L., and Chen, L. 2018. Grazing exclusion-induced shifts, the relative importance of environmental filtering, biotic interactions and dispersal limitation in shaping desert steppe communities, northern China. Journal of Arid Land 10:402-415. doi:10.1007/s40333-018-0411-5. [ Links ]

Wang, B., Sun, G., Luo, P., Wang, Z., Zhang, Y., Wu, N., et al. 2011b. Microbial communities of Alpine meadow soil in the eastern Qinghai-Tibetan Plateau subjected to experimental warming and grazing. Chinese Journal of Applied Environmental Biology 17:151-157. doi:10.3724/sp.J.1145.2011.00151. [ Links ]

Xie, L., Chen, W., Gabler, C.A., Han, L., Guo, H., Chen, Q., et al. 2016. Effects of grazing intensity on seed production of Caragana stenophylla along a climatic aridity gradient in the Inner Mongolia Steppe, China. Journal of Arid Land 8:890-898. doi:10.1007/s40333-016-0050-7. [ Links ]

Xie, Z., Le Roux, X., Wang, C.P., Gu, Z.K., An, M., Nan, H.Y., et al. 2014. Identifying response groups of soil nitrifiers and denitrifiers to grazing and associated soil environmental drivers in Tibetan alpine meadows. Soil Biology and Biochemistry 77:89-99. doi:10.1016/j.soilbio.2014.06.024. [ Links ]

Zang, X., Zhang, H., Wang, X., Yuandan, M.A., Baoyintaogetao, G., Gao, Y., et al. 2017. Effect on functional diversity of Artemisia frigida rhizosphere soil microbial community with grazing. Journal of Zhejiang A&F University 34:86-95. doi:10.11833/j.issn.2095-0756.2017.01.013. [ Links ]

Zhao, F., Ren, C., Shelton, S., Wang, Z., Pang, G., Chen, J., et al. 2017. Grazing intensity influence soil microbial communities and their implications for soil respiration. Agriculture, Ecosystems & Environment 249:50-56. doi:10.1016/j.agee.2017.08.007. [ Links ]

Zhong, L., Hoogendoorn, C.J., Bowatte, S., Li, F.Y., Wang, Y., and Luo, D. 2016. Slope class and grazing intensity effects on microorganisms and nitrogen transformation processes responsible for nitrous oxide emissions from hill pastures. Agriculture, Ecosystems & Environment 217:70-78. doi:10.1016/j.agee.2015.11.009. [ Links ]

Zhou, G., Zhou, X., He, Y., Shao, J., Hu, Z., Liu, R., et al. 2017. Grazing intensity significantly affects belowground carbon and nitrogen cycling in grassland ecosystems: a meta-analysis. Global Change Biology 23:1167-1179. doi:10.1111/gcb.13431. [ Links ]

Received: February 01, 2019; Accepted: June 15, 2019

*Corresponding author (jiameiqing@tjnu.edu.cn).

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License