Effects of permafrost thaw-subsidence on soil bacterial communities in the southern Qinghai-Tibetan Plateau

Applied Soil Ecology 128 (2018) 81–88

Contents lists available at ScienceDirect

Applied Soil Ecology journal homepage: www.elsevier.com/locate/apsoil

Effects of permafrost thaw-subsidence on soil bacterial communities in the southern Qinghai-Tibetan Plateau

T

⁎

Xiaodong Wua, Lin Zhaoa, , Guimin Liub, Haiyan Xub, Xiaolan Zhangb, Yongjian Dinga,c,d a Cryosphere Research Station on the Qinghai-Tibetan Plateau, State Key Laboratory of Cryospheric Science, Northwest Institute of Eco-Environment and Resource, Chinese Academy of Sciences, Lanzhou, Gansu 730000, China b School of Environmental and Municipal Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China c Key Laboratory of Ecohydrology of River Basin Sciences, Chinese Academy of Sciences, 320 West Donggang Road, Lanzhou 730000, China d University of Chinese Academy Sciences, 19(A) Yuquan Road, Shijingshan District, Beijing 100049, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Permafrost Bacterial communities Qinghai-Tibetan Plateau Subsidence Organic carbon Illumina dye sequencing

Permafrost thaws cause ground subsidence as the ground ice melts and drains away. Little is known about the effects of this permafrost thaw subsidence on bacterial communities. In this study, using Illumina sequencing methods, we investigated the structure of bacterial communities in the upper 50 cm of the soil in a typical permafrost thaw subsidence area on the southern Qinghai-Tibetan plateau. The micro topographies in the study area were classified as control, collapsing, and subsided types. Results showed that the organic carbon content in the collapsing areas was slightly lower than that in the control areas, while there was a substantial decrease in the subsided areas, with a loss of 23.6 ± 13.2% of organic carbon. The microbial carbon contents showed the highest values in collapsing areas. For all three types of soils, the most abundant microbial groups were Proteobacteria, Acidobacteria, and Bacteroidetes. The Non-metric multidimensional scaling (NMDS) results showed that the bacterial communities were different in the subsided areas than in the control and collapsing areas. In the control and collapsing areas, the soil bacterial communities showed a clear vertical distribution pattern with depth, which was not apparent in the subsided areas. The bacterial communities also correlated with soil variables such as carbon, moisture, nitrogen contents, and the C:N ratio. The ground subsidence can greatly change these variables. The results suggested that permafrost thaw subsidence had important effects on microbial communities via the changes of soil properties.

1. Introduction The permafrost ecosystems have received worldwide attention because climate warming could potentially cause changes of carbon balance in these areas and affect global carbon cycle (Hugelius et al., 2014; Ping et al., 2015; Ping et al., 2008). Microbial decomposition is the basic pathway by which soil organic carbon (SOC) is converted into greenhouse gases, and understanding of soil microbial communities and their relationship to environmental factors is of great importance to evaluate the organic matter decomposition. Under frozen conditions, the microbes are relatively inactive. As permafrost thaws, the microbes will be reactivated. The thawing of permafrost can lead to changes in soil properties, including soil water content, temperature, and pH values, and subsequently, changes in land cover types (Mu et al., 2017; Wu et al., 2017a). These factors have been recognized as important causes of changes in the soil microbes (Brockett et al., 2012; Feng et al., 2014; Kim et al., 2014; Shi et al., 2015). Therefore, soil microbes are

⁎

sensitive to permafrost thaws although they respond differently to permafrost thaws at different depths (Deng et al., 2015). Permafrost degradation not only means a soil temperature increase but also landform changes. Permafrost thaw subsidence is one of the typical landforms of thermokarst terrains due to the ground ice melting and water draining away. The change in landforms due to ground subsidence greatly affects the soil properties, including the SOC content and greenhouse gas fluxes (Abbott and Jones, 2015). It has been suggested that the effects of ground subsidence should be taken into consideration when evaluating permafrost carbon feedback (Schuur et al., 2015). However, there are few reports about the bacterial communities in the different micro topographies in the subsided areas that result from permafrost degradation, and the determinants of the bacterial communities in these areas are unknown. The Qinghai-Tibetan Plateau (QTP) is the largest low-latitude permafrost area in the world. The permafrost area on the QTP accounts for approximately 8% of the global permafrost area and approximately

Corresponding author at: Northwest Institute of Eco-Environment and Resource, Chinese Academy of Sciences, 320 west Donggang Road, Lanzhou, Gansu 730000, China. E-mail address: [email protected] (L. Zhao).

https://doi.org/10.1016/j.apsoil.2018.04.007 Received 13 January 2018; Received in revised form 3 March 2018; Accepted 8 April 2018 Available online 14 April 2018 0929-1393/ © 2018 Elsevier B.V. All rights reserved.

Applied Soil Ecology 128 (2018) 81–88

X. Wu et al.

three quarters of the mountain permafrost area (Zhang, 2012). On the QTP, the ground temperatures are relatively higher than those in circum-arctic regions and just slightly lower than 0 °C (Qin et al., 2016; Zhao et al., 2010). In addition, the rate of warming can be amplified in high-mountain regions (Kang et al., 2010). Thus, the permafrost on the QTP is extremely sensitive to climate change. Similar to the circumarctic regions, the permafrost regions on the QTP store a large amount of SOC, which has been estimated at up to 160 Pg for the upper 25 m soils (Mu et al., 2015). The permafrost degradation can potentially produce emissions of large amounts of greenhouse gases on the QTP (Mu et al., 2016b; Wu et al., 2014). To understand the possible roles of microbial community in organic matter decomposition on the QTP, several studies have been conducted to investigate the microbial communities and their relationship to environmental factors. With the samples collected from several sites with gradients of environmental factors, it was found that pH, SOC, soil moisture content, and even soil depths were determinants of the soil microbial communities (Chu et al., 2016; Zhang et al., 2014). Due to permafrost degradation, thermokarst terrains, including thermokarst lakes, ground subsidence, permafrost collapse, and thaw slumps occur over much of the permafrost areas (Mu et al., 2016a; Niu et al., 2012). The landform changes can also redistribute the SOC and affect the quantities of organic matter (Mu et al., 2016d). Since the microbial community is closely associated with soil variables (Goenster et al., 2017), we hypothesize that 1) thaw-subsidence changed the soil variables and further affected the bacterial community; 2) the vertical distribution of the soil bacterial community in the thaw-subsided area has been changed due to the land form deformation. To test these hypotheses, we investigated the bacterial communities in soils with different microfeatures and their relationships with soil variables within a typical thaw subsidence area on the southern QTP. The results examined bacterial community and its relationship to soil physico-chemical variables in a typical thaw-subsidence area and thus would be beneficial toward understanding the microbial mechanisms of changes of soil biogeochemical cycles in permafrost regions under future global warming scenarios.

Fig. 1. Sampling sites for the three micro topographies (control, collapsing, and subsided).

AC0_10, AC10_20, AC20_30, AC30_40, and AC40_50. The soil samples from the collapsing and subsided areas were abbreviated similarly, i.e., AE0_10 (0–10 cm soil from the collapsing area), AD0_10 (0–10 cm soil from the subsided area). The soils were collected aseptically using ethanol-disinfected soil augers. We placed the soil samples in sealable, clean plastic bags, and preserved in a car refrigerator at −4 °C. The samples were brought to the laboratory immediately. For soil moisture determination, another 5 soil samples were collected and stored in aluminum boxes and carefully sealed to prevent changes in soil moisture. For the DNA analysis, soils were stored at −80 °C until genomic DNA extraction. 2.2. Laboratory experiments and DNA extraction The soil moisture content was determined using oven-dried method (105 °C for 8 h). The soil pH and conductivity values were measured using soil suspensions (1:5 soil:water ratio). The total soil carbon (TC) and SOC of were measured using a TOC analyzer (Vario TOC cube, Elementra). The total inorganic carbon (TIC) content was calculated as: SIC = TC – SOC. Total nitrogen (TN) was measured using the microKjeldhal procedure. The mass ratios of SOC and TN were calculated as C:N ratios. The light-fraction organic carbon was measured from the SOM that was separated by flotation method (1.8 g cm−3 NaI solution). Microbial carbon (MBC) was determined using the chloroform fumigation–extraction method (Shang et al., 2016). Soil particle distribution was analyzed by a laser diffraction instrument (Mastersizer 2000, Malvern, UK). We extracted total soil DNA from 0.3 g soils using the MoBio PowerSoil DNA Isolation kit (MoBio Laboratories, Carlsbad, CA, USA). For each sample, we performed three replicate extractions, and then pooled the extractions for further analyses. The extracted DNA was measured using a QuBit DNA quantification system (Invitrogen) with QuBit high sensitivity assay reagents. The soil DNA samples were stored frozen at −20 °C pending further analysis.

2. Materials and methods 2.1. Site description and soil sampling An area with ground subsidence in the permafrost regions (91.76°E, 32.00°N, 4744 m) in the southern QTP was studied. From 1980 to 2010, the mean annual precipitation (MAP) is approximately 350 mm (https://data.cma.cn/ or http://www.cma.gov.cn/2011qxfw/ 2011qsjgx/). The sampling area is located in a mountain valley with poorly drainage class. The slope of the sampling area is 1.5°, and the aspect is 90°. The parent materials of the soils are colluvial deposits, and the land cover type is alpine meadow. According to the Soil Taxonomy (ST) (Soil Survey Staff, 2014), the soil is classified as a Glacic Histoturbels (ABAB). From the soil pits, which were excavated in the late September in 2013, the active layer thickness was 2.5 m because the ground ice was present at this depth. The subsidence of the surface ground resulted from melting of the ground ice (Fortier et al., 2010). According to the microfeatures, the subsidence was defined in three stages: control, collapsing and subsided (Fig. 1). The dominant vegetation in the three areas is shown in Table 1. The sampling sites for the control area were collected from the plots approximately 10 m from the boundary of the subsided area (soil samples hereafter abbreviated as AC). The soils from the collapsing areas exposed to the air were also collected (soil samples hereafter abbreviated as AE). The subsided area was abbreviated as AD. For all the three stages, six subsamples of each soil were randomly collected from each area. The surface 50 cm soil layers were collected with increments of 10 cm, and the soil samples from the control areas were abbreviated as

2.3. PCR amplification We conducted the PCR amplification, purification, and sequencing of a region of the 16S rRNA gene (Fierer and Jackson, 2006). The primer set of F515 and R907 was chosen to amplify the V4 and V5 hypervariable regions of the bacterial 16S ribosomal RNA gene (Bates et al., 2011). Template DNA (10 ng) and PCR Pre-Mix (TaKaRa, 25 µL) were mixed with reverse and forward primers (0.3 µM) (Fang et al., 2016). The PCR products were mixed with an equal volume of 1X loading buffer (containing SYBR green) and analyzed by electrophoresis on agarose gels (1.2%). Samples with bright areas between 350 and 82

Applied Soil Ecology 128 (2018) 81–88

X. Wu et al.

Table 1 Soil variables at depths of 0–50 cm for the three micro topographies (n = 6). Micro topography

Depth (cm)

Soil moisture%

pH

Bulk density (g cm−3)

Conductivity (ms/cm)

Dominant vegetation

Control

0–10 10–20 20–30 30–40 40–50 Mean

52.80 ± 7.63 58.47 ± 6.63 66.97 ± 5.34 72.34 ± 5.72 74.29 ± 9.72 65.97 ± 5.16a

7.39 ± 0.51 7.56 ± 0.47 7.70 ± 0.52 7.78 ± 0.56 7.57 ± 0.42 7.60 ± 0.15a

1.05 1.12 1.12 1.04 0.98 1.06

± ± ± ± ± ±

0.23 0.18 0.27 0.19 0.16 0.06

4.56 ± 0.53 2.55 ± 0.43 2.73 ± 0.41 3.71 ± 0.23 4.78 ± 0.36 3.66 ± 1.02a

Kobresia robusta Maxim., K. tibetica,K. willd

Collapsing

0–10 10–20 20–30 30–40 40–50 Mean

36.44 ± 4.14 30.73 ± 4.23 32.98 ± 3.57 33.33 ± 3.16 32.52 ± 2.89 33.20 ± 2.07b

7.55 ± 0.54 7.85 ± 0.55 7.93 ± 0.61 8.00 ± 0.58 7.94 ± 0.57 7.85 ± 0.18b

1.08 0.98 0.88 1.13 1.12 1.04

± ± ± ± ± ±

0.14 0.17 0.16 0.08 0.11 0.11

14.45 ± 2.26 6.04 ± 1.21 7.05 ± 2.09 7.29 ± 1.72 8.61 ± 1.52 8.69 ± 3.35b

Kobresia robusta Maxim., K. tibetica,K. willd

Subsided

0–10 10–20 20–30 30–40 40–50 Mean

50.55 ± 4.38 49.46 ± 4.82 64.45 ± 4.78 58.91 ± 5.88 56.19 ± 5.66 55.91 ± 6.17c

7.79 ± 0.56 7.94 ± 0.52 7.80 ± 0.48 7.83 ± 0.55 7.85 ± 0.60 7.76 ± 0.18b

0.98 0.89 1.03 1.14 0.97 1.03

± ± ± ± ± ±

0.24 0.25 0.19 0.28 0.21 0.09

2.42 ± 0.43 2.29 ± 0.36 3.01 ± 0.35 3.94 ± 0.54 4.07 ± 0.43 5.17 ± 3.35a

Kobresia robusta Maxim., K. humilis

The bold text highlights the mean values from the different depths. The different letters showed the significant differences among the mean values (p < .05, ANOVA followed by LSD test).

correspondence analysis (DCA) showed that the eigenvalue of the bacterial phyla was less than 3.0, and thus a redundancy analysis (RDA) was selected to explore the relationships between the soil variables and bacteria phyla. The mantel test and RDA were performed in Vegan packages in R 3.3.1. All the raw reads have been deposited into the NCBI database (Sequence Read Archive, SRP095474).

3. Results 3.1. Soil properties All the soils were loamy sand. For the soil particle size distribution, the sand accounted for 86.6 ± 4.4%, silt accounted for 11.1 ± 2.9%, and clay for 2.3 ± 1.8%. The proportions of sand, silt, and clay varied considerably, with no obvious patterns among the samples (data not shown). The soil physical and chemical variables from the three microfeatures are summarized in Table 1. The mean soil pH was 7.77. The soils in the collapsing area had the lowest moisture content, but with highest soil conductivity and pH values. The SOC, TN MBC, and LFC content varied at different depths (Fig. 2). For the upper 40 cm layers, the highest SOC contents were recorded in the control soils, whereas the lowest values appeared in the subsided soils. The mean SOC contents in the subsided soils were significantly lower than those in the control (n = 5, p < .01, t-test) and collapsing areas (n = 5, p < .05, t-test). The TN contents varied from 0.24 to 0.4%, and the TN contents of the control areas were significantly higher than those in collapsing areas (n = 5, p < .05, t-test). The LFC and MBC contents showed a clear decreasing trend with the depth in control and subsided soils. The mean LFC contents in control soils were significantly higher than those of the collapsing soils (n = 5, p < .01, t-test) and the subsided soils (n = 5, p < .01, t-test), while the MBC contents in the collapsing soils were significantly higher than those in the control (n = 5, p < .01, t-test) and subsided soils (n = 5, p < .01, t-test). The soil depths had negatively statistically significant relationships with silt, suggesting that the soils were coarser at deeper layers. The soil moisture was significantly negatively correlated with conductivity, C:N ratio, and MBC but positively correlated with TN. The TOC, TIC and TC were significantly positively correlated with each other. There were significant positive associations between the soil carbon and the C:N ratio. The MBC was positively correlated with conductivity and the C:N ratio but negatively correlated with moisture and TN. High proportions

Fig. 2. The soil organic carbon (SOC), total nitrogen (TN), light fraction organic carbon (LFC), and microbial carbon (MBC) contents of different layers of the three micro topographies.

450 bp were chosen for further experiments. We used the QIAquick Gel Extraction Kit (Qiagen, Chatsworth, CA, USA) to purify the PCR products. For sequencing analysis, a single composite sample was prepared by combining approximately equimolar amounts of the PCR products from each sample. Sequencing was carried out on an Illumina Miseq PE300 platform (Majorbio Bio-pharm Technology Co., Ltd., Shanghai, China).

2.4. Processing of sequencing data Sequences were processed and analyzed according to the bioinformatics procedures of (Wang et al., 2016). Paired-end reads from the original DNA fragments were merged by using FLASH software (Mago and Salzberg, 2011). For the quality-filtering using QIIME (version 1.17), three criteria were applied (Wu et al., 2017b): 1) The 300 bp reads were truncated at any site receiving an average quality score < 20 over a 10-bp sliding window. We discarded the truncated reads which were shorter than 50 bp; 2) exact barcode matching; 3) only sequences that overlapped more than 10 bp were assembled. We discarded all the reads that could not be assembled. Using UPARSE (version 7.1), the sequences were assigned to OTUs (at 97% similarity). Meanwhile, the chimeric sequences were identified and removed using UCHIME. The Chao1 metric, Ace, Shannon, and Simpson index were calculated, and the observed OTUs (count of unique OTUs) were also shown. We used mantel test to examine the relationship between the bacterial community structure and each soil variables. A detrended 83

Applied Soil Ecology 128 (2018) 81–88

X. Wu et al.

Table 2 Pearson’s correlation coefficients between environmental factors.

Depth Mois Cond pH TN TC TOC TIC C:N LFC MBC Sand Silt Clay **

Depth

Mois

Cond

pH

TN

TC

TOC

TIC

C:N

LFC

MBC

Sand

Silt

Clay

1 0.23 −0.06 0.42 0.04 −0.23 −0.26 −0.20 −0.20 −0.38 −0.42 0.49 −0.56* −0.29

1 −0.63* −0.41 0.68** −0.27 −0.24 −0.29 −0.66** 0.43 −0.80** −0.08 −0.01 0.21

1 −0.04 −0.41 0.44 0.41 0.45 0.62* −0.32 0.86** 0.35 −0.28 −0.41

1 −0.49 −0.45 −0.51 −0.37 −0.03 −0.75** 0.02 0.29 −.034 −0.15

1 0.05 0.10 −0.00 −0.61* 0.60* −0.53* −0.24 0.14 0.35

1 0.98** 0.98** 0.74** 0.41 0.44 −0.01 0.03 −0.03

1 0.92** 0.71** 0.46 0.41 −0.12 0.14 0.07

1 0.74** 0.35 0.46 0.11 −0.09 −0.13

1 −0.13 0.68** 0.06 0.01 −0.17

1 −0.192 −0.30 0.31 0.24

1 0.06 0.03 −0.20

1 −0.96** −0.88**

1 0.71**

1

*

p < .01; p < .05; n = 15; 2-tailed test. Mois, Moisture, Cond, Conductivity, MBC, microbial organic carbon, LFC, Light fraction carbon. The bold font shows the statistically significant relationships.

Acidobacteria (31.8%), Bacteroidetes (13.9%), Chloroflexi (4.7%), Actinobacteria (3.2%), Planctomycetes (2.9%), and Nitrospirae (2.9%). The other phyla in different samples varied with abundances of lower than 4.7% across all the samples (Fig. 3). At the class level, the dominant classes were Actinobacteria (31.8%), Betaproteobacteria (17.3%), Alphaproteobacteria (9.0%), Sphingobacteria (5.1%), Gammaproteobacteria (4.7%), Flavobacteria (3.8%), and Deltaproteobacteria (3.6%). There were five classes of the phylum Proteobacteria. In all the samples, the Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria, and Deltaproteobacteria were abundant, while Epsilonproteobacteria were only found in the sample of AD 40_50 (Supplementary Material 1). At the genus level, the microbial communities were mainly composed of RB41_norank (10.7%), Subgroup_6_norank (10.1%), 1124_norank (6.6%), Comamonadaceae_unclassified (5.3%), Flavobacterium (3.9%), Nitrospira (2.8%), Anaerolineaceae_uncultured (2.8%), and Nitrosomonadaceae_uncultured (2.3%). Other genera were present at less than 2% (Supplementary Material 2). The dominant bacterial phyla showed different vertical patterns among the soils. The relative abundance of Proteobacteria decreased with depth in both control and collapsing areas but showed a moderate increasing trend in subsided areas. The Acidobacteria abundances in control and collapsing soils were largely higher than 40% and increased with depth. In the subsided areas, the Acidobacteria abundances were much lower and showed a trend opposite to those of the control and collapsing areas. The Bacteroidetes varied considerably among different samples and had a lower mean value than the control in collapsing (p > .05, t-test, n = 5) and subsided (p > .05, t-test, n = 5) areas (Fig. 3). The Chloroflexi abundances in control and collapsing areas were much lower than those in subsided areas. The Actinobacteria abundances showed decreasing trends with depths in control and collapsing areas, whereas they fluctuated in soils from subsided areas (Fig. 3). The mean Planctomycete abundances in collapsing areas were higher than in the control (p > .05, t-test, n = 5) and subsided (p < .05, t-test, n = 5) areas. The nonmetric multidimensional scaling (NMDS) plots clearly indicated that the bacterial communities showed differences among the soils (Fig. 4). The samples from soils from control and collapsing areas were largely distributed in the left part of the chart. The chart also reflected the effects of depth on the bacterial communities, i.e., the positions of the soils from lower depths were higher than those from shallower depths. In contrast, the samples from soils from subsided areas were distributed in the right part of the chart without a vertical distribution of soil samples.

of sand were significantly associated with low proportions of silt and clay (Table 2). Bulk densities had no statistically significant relationships with any of the variables (data not shown). 3.2. Sequencing data For each sample, there were 27,681 reads, and a total of 20,385 OTUs (at the 3% evolutionary distance) for all the samples identified. The Good’s coverage estimator of the OTUs ranged from 98.1% to 98.7% in the samples. The rarefaction curves demonstrated that the sequencing data sufficiently covered the bacterial community diversity (data not shown). The highest OTU value of 1579 occurred in the sample of AC 10_20, and the lowest value of 1156 in AE40_50 (Table 3). The ACE, Chao1, and the OTU numbers showed similar patterns in the samples (Table 3). The Shannon-Wiener index was significantly positively correlated with the OTU (p < .01, Pearson, two-tailed) but negatively correlated with the Simpson index (p < .01, Pearson, two-tailed). The Chao, Ace, and OTUs significantly correlated with each other (p < .01, Pearson, twotailed). The Shannon-Wiener indices ranged from 4.30 to 5.72. 3.3. Bacterial diversity and community structure The OTUs belonged to 47 phyla, 107 classes, 240 orders, 425 families and 725 genera. Bacteria phyla varied greatly among different samples, and the dominant phyla were Proteobacteria (34.6%), Table 3 Estimated OTU richness, diversity indices, and sample coverage. Samples

OTUs

Ace

Chao1

Coverage

Shannon index

Simpson index

AC0_10 AC10_20 AC20_30 AC30_40 AC40_50 AE0_10 AE10_20 AE20_30 AE30_40 AE40_50 AD0_10 AD10_20 AD20_30 AD30_40 AD40_50

1448 1579 1485 1426 1224 1310 1454 1406 1308 1156 1313 1418 1266 1224 1368

1817 2082 2080 1868 1790 1665 1912 1937 1877 1673 1691 1953 1764 1655 1860

1820 2060 2086 1843 1817 1650 1928 1928 1881 1652 1721 1969 1744 1658 1844

0.985 0.982 0.981 0.984 0.983 0.987 0.984 0.983 0.983 0.984 0.986 0.983 0.984 0.986 0.983

5.52 5.72 5.20 5.29 4.72 5.15 5.48 5.00 4.81 4.30 5.43 5.28 4.56 4.86 4.87

0.013 0.009 0.023 0.016 0.026 0.037 0.012 0.027 0.040 0.057 0.014 0.018 0.063 0.032 0.033

OTUs, Operate taxonomic units, Ace, the Ace metric (Abundance-based coverage estimator), Chao1, the Chao1 richness estimator. 84

Applied Soil Ecology 128 (2018) 81–88

X. Wu et al.

Fig. 3. The relative abundance of bacterial phyla (%) in the soils from the three micro topographies. AC, control areas, AE, collapsing areas, AD, subsided areas.

(r = 0.22, p = .03), and TIC (r = 0.31, p = .01). The r value between moisture and bacterial community was 0.18, with a p value of 0.06. Other factors such as pH, MBC, LFC, and the C:N ratio did not have statistically significant relationships with the bacterial communities. The statistically significant relationships between phyla and

3.4. Relationships between bacterial community structure and soil properties A Mantel test demonstrated that the bacterial community statistically significantly correlated with TC (r = 0.27, p = .02), TOC

Fig. 4. The nonmetric multidimensional scaling (NMDS) plots for the bacterial community structures in the soils from the three micro topographies. AC, control areas, AE, collapsing areas, AD, subsided areas. 85

Applied Soil Ecology 128 (2018) 81–88

X. Wu et al.

Table 4 Pearson’s correlation coefficients between environmental factors and relative abundance of bacterial phyla. Proteobac Depth Mois Cond TC TOC TIC C:N MBC Sand Silt Clay

−0.26 0.40 −0.46 −0.42 −0.31 −0.53* −0.53* −0.48 −0.37 0.33 0.36

Acidobac

Bacteroidetes

Chloroflexi

Actinobac

Planctomycetes

0.30 −0.36 0.50 0.52* 0.40 0.61* 0.56* 0.47 0.38 −0.35 −0.36

−0.22 0.29 −0.48 −0.39 −0.27 −0.51 −0.42 −0.31 −0.61* 0.57* 0.57*

−0.15 0.14 −0.45 −0.66** −0.63* −0.67** −0.46 −0.34 −0.06 0.09 0.00

−0.75 −0.27 0.35 0.15 0.19 0.11 0.29 0.45 −0.14 0.25 −0.06

−0.48 −0.45 0.60* 0.67** 0.59* 0.74*** 0.71** 0.66** 0.17 −0.12 −0.22

**

Nitrospirae *

0.62 0.44 −0.21 −0.04 −0.12 0.04 −0.15 −0.47 0.61* −0.69** −0.36

Chlorobi

Gemmatimonadetes

0.14 0.29 −0.43 −0.58* −0.55* −0.59* −0.56* −0.52* 0.00 −0.08 0.13

0.21 −0.31 0.56* 0.60* 0.51 0.68** 0.58* 0.48 0.32 −0.28 −0.33

** p < .01; *p < .05; n = 15; 2-tailed test. Mois, Moisture, Cond, Conductivity, MBC, microbial organic carbon, LFC, Light fraction carbon. Proteobac, Proteobacteria, Acidobac, Acidobacteria, Actinobac, Actinobacteria. The bold font shows the statistically significant relationships. The pH, soil moisture, total nitrogen, light fraction carbon and bulk density did not have a statistically significant relationship with any phyla and these variables were not shown. The Firmicutes and Bacteria_unclassified were not statistically significantly correlated the variables and were not shown.

Gemmatimonadetes showed significant, positive relationships with TC, TIC, and C:N ratio, while Chlorobi was largely negatively correlated with these factors. The Nitrospirae correlated with soil particle distribution (Table 4).

environmental factors are shown in Table 4. The depth was significantly negatively correlated with Proteobacteria, but positively correlated with Acidobacteria, Nitrospirae, and Gemmatimonadetes. The Bacteroidetes were significantly negatively correlated with active layer thickness but positively correlated with pH and mean annual precipitation. The Firmicutes were positively correlated with TOC and TN. The bulk densities, proportion of sand, gravel content, and C:N ratios also had statistically significant relationships with bacterial phyla. The moisture, pH, sand, depth, conductivity, and bulk density were selected to investigate the effects of environmental factors on the soil bacterial communities since other factors were strongly correlated with these factors (r > 0.7, p < .01, Table 3). The first and second axes of RDA explained 52.3% and 1.8% of the total phyla variances. The RDA result showed that the first horizontal axis was mainly correlated with moisture, TN, depth, conductivity, and TOC, whereas pH, sand, and bulk density mainly correlated with the vertical axis, which explained much lower phyla variances (Fig. 5). There are statistically significant relationships between specific phyla and environmental variables. For example, the Acidobacteria and

4. Discussion The collapse of soil mass on the boundaries of subsided areas and non-subsided areas can expose soil profiles, and these soils generally experience structural deformation. The soil moisture content was much lower in the collapsing areas in this study, which could be explained by the exposure of natural soil profiles and the deformation of the soil structure leading to higher evaporation and improved drainage conditions (Jensen et al., 2014). The higher pH and conductivity in the collapsing soils showed the effects of evaporation on the soluble salinity in this area, consistent with the significant negative relationship between moisture and conductivity. There were no significant differences among the bulk densities of the soils in the three areas, which suggested that the soils were homogeneous before the ground subsidence occurred. The results indicated that ground subsidence greatly changed the soil physiochemical characteristics. The SIC pools were primarily present in arid and semi-arid regions (Schlesinger, 2002). The study area belongs to the semiarid regions since the mean annual precipitation is approximately 350 mm. Our findings are in agreement with previous studies showing that TIC contents were comparable to the TOC in permafrost regions on the QTP (Mu et al., 2016c). The SOC contents in the control area were higher than those in the collapsing area, which revealed that the ground subsidence caused by permafrost thaw decreased the organic carbon content. It has been shown that the organic carbon in the surface 10 cm layer of the collapsing area could be 30% lower than that of control area (Mu et al., 2016d). This could be explained if the lower amount of soil moisture in the collapsing area accelerated the microbial decomposition of SOC (Curiel Yuste et al., 2007), which would be consistent with significantly higher contents of MBC in the soils of collapsing areas. In addition, the MBC was significantly negatively correlated with the C:N ratio, which showed that more organic matter was decomposed where there was a higher MBC content. The high content of MBC in collapsing areas could be explained by low soil moisture favoring the growth of microbes (Jansson and Tas, 2014). The low C:N ratio resulted from the rate of production of organic matter being higher than the decomposition rate (Post et al., 1985). The lower LFC content of the soils in collapsing areas suggested that the organic matter in these soils was more decomposed because LFC is a labile fraction of organic matter and can be easily utilized by microbes. Previous studies demonstrated that particulate organic matter and dissolved organic matter can be transferred to lower terrains in permafrost regions (Paré and Bedard-

Fig. 5. The redundancy analysis (RDA) of the bacterial community structures in the soils from the three micro topographies. AC, control areas, AE, collapsing areas, AD, subsided areas. 86

Applied Soil Ecology 128 (2018) 81–88

X. Wu et al.

moisture and TN played opposite roles in shaping the bacterial community structure. Moisture has been shown to be an important factor affecting soil microbial communities (Zhang et al., 2014). In addition, the effects of moisture and TN on the microbial community could reflect the effects of C:N ratios (Rousk and Baath, 2007). In this study, the C:N ratios were negatively correlated with the moisture and TN contents. This could be explained by the fact that the TN contents showed small changes in comparison with soil organic carbon, which may have experienced decomposition (Mu et al., 2016d). The landform changes caused by the permafrost thaw can greatly affect the stock and chemical characteristics of SOC. Our study indicated that the permafrost thaw subsidence can also greatly change the microbial community structure due to the effects of loss of organic carbon, moisture, and the mixing of soils during the subsidence. Further studies are required to understand the effects of the microbial community structure changes on the emissions of greenhouse gases in permafrost regions.

Haughn (2013)), and thus, the SOC contents in lower terrains can be higher than those in collapsing areas (Mu et al., 2016d). In our study area, the SOC content (3%) in the control areas was much lower than in the organic soils of permafrost regions. Therefore, the lower terrains may contain less particulate and dissolved organic matter. This could be an explanation for the lowest SOC contents being in the subsided areas. Another possible explanation is that the SOC in the subsided areas experiences higher decomposition rates than production rates. The vegetation has been greatly changed by the subsidence (Fig. 2); thus, it is reasonable to infer that the net carbon ecosystem has also been changed. Proteobacteria, Acidobacteria, Chloroflexi, Bacteroidetes, and Actinobacteria are the dominant phyla in most soils under different climatic conditions (Feng et al., 2014; Rousk et al., 2010; Wallenstein et al., 2007). The abundances of these phyla, with the exception of Actinobacteria, were comparable to those in most studies of different soils. In this study, the Actinobacteria accounted for 3.2% of the total bacteria, which was lower than the values of approximately 10% in the 29 soils in the Arctic (Chu et al., 2010). In the alkaline soils on the QTP, it was found that the relative abundance of Actinobacteria in the upper 10 cm layers was approximately 15% (Zhang et al., 2014). The highest content of Actinobacteria in the present study was 7.9%, which was recorded in the 0–10 cm layer of the collapsing areas. This phylum was found to be positively related to pH value (Liu et al., 2015). Therefore, the relatively lower abundance of Actinobacteria was possibly due to the lower pH values in the soils (usually lower than 8). It has been shown that pH was the most important factor controlling soil bacterial communities on the regional scale (Chu et al., 2016; Fierer and Jackson, 2006; Tripathi et al., 2012). In this study, the pH was not significantly correlated with bacterial community structure. This could be explained by the fact that our study was conducted in the same area with micro topographies, and thus, the pH gradients were not sufficient to impose substantial effects on the bacterial community. In addition to pH, other soil variables, including carbon availability, total nitrogen, soil moisture (Drenovsky et al., 2004), and C:N ratios (Wan et al., 2014), are associated with soil bacterial communities. The specific phyla of bacteria, including Planctomycetes and Gemmatimonadetes, were significantly associated with soil carbon, C:N ratios and MBC. However, the bacterial community structure had a significant relationship with SOC, but it had no statistically significant relationships with C:N ratios, TN, or moisture. These results showed that the changes in bacterial community structure among micro topographies may not be ascribed to the factors that greatly affected bacteria on a regional scale. The soil bacterial communities changed with depth because soil properties, including pH, oxygen availability, and nutrients, varied greatly among different soil layers (Douterelo et al., 2010; Fierer et al., 2003). In some studies, the difference in bacteria between different small areas (the environments surrounding a plant) could be greater than between large areas (across a field) (Ramette and Tiedje, 2007). In our study, the NMDS result clearly showed that soil bacterial community structure had similar vertical patterns in both control and collapsing areas, whereas this pattern was mixed in the subsided areas. In subarctic areas, it was also found that soil depth was one of the most influential soil properties determining the bacterial community (Kim et al., 2014). The vertical pattern of soil bacterial communities in subsided areas suggested that the vertical distribution of soil bacteria was greatly changed in the subsided area, which could be ascribed to the soil mixture resulting from ground surface deformation during the subsidence. The RDA results showed that the SOC, depth, conductivity, TN, and moisture mainly affected the first axis. On the QTP, the SOC content was largely positively associated with moisture and TN on a regional scale, while it decreased with depth (Wu et al., 2016). In this study, these relationships were not found, which could be explained by the fact that the soils were collected from micro topographies, and there were no significant environmental gradients. In addition, the soil

5. Conclusions The microbial community structure and its relationship to environmental factors in the soils of different micro topographies in a typical permafrost thaw subsidence area were investigated on the south QTP. The permafrost thaw ground subsidence considerably changed the soil properties, especially the organic carbon content, moisture, and microbial carbon content. Consequently, the soil bacterial community structure in the collapsing area was different from that in the control area with a similar vertical pattern. In comparison with the control and collapsing areas, the soil bacterial community structure of the subsided areas showed no vertical pattern, suggesting that the bacterial community was mixed due to the deformation of the soils during permafrost thaw subsidence. The bacterial community structure was mainly associated with soil organic carbon, while pH values had a significant relationship with the bacterial community. In addition, the soil bacterial community correlated with moisture and conductivity, which also reflected a relationship with the C:N ratio. Our results showed that permafrost thaw subsidence has significant effects on soil properties and that the mixing of soils and loss of carbon are the main factors in determining the soil bacterial community in subsided areas. Acknowledgements This work was supported by the State Key Laboratory of Cryospheric Science (SKLCS-ZZ-2018), the National Natural Science Foundation of China (41721091, 41661013, 41730751), and the Key Research Program of Frontier Sciences, CAS (Grant No. QYZDY-SSW-DQC021). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.apsoil.2018.04.007. References Abbott, B.W., Jones, J.B., 2015. Permafrost collapse alters soil carbon stocks, respiration, CH4, and N2O in upland tundra. Glob. Change Biol. 21, 4570–4587. Bates, S.T., Berglyons, D., Caporaso, J.G., Walters, W.A., Knight, R., Fierer, N., 2011. Examining the global distribution of dominant archaeal populations in soil. ISME J. 5, 908–917. Brockett, B.F.T., Prescott, C.E., Grayston, S.J., 2012. Soil moisture is the major factor influencing microbial community structure and enzyme activities across seven biogeoclimatic zones in western Canada. Soil Biol. Biochem. 44, 9–20. Chu, H., Fierer, N., Lauber, C.L., Caporaso, J.G., Knight, R., Grogan, P., 2010. Soil bacterial diversity in the Arctic is not fundamentally different from that found in other biomes. Environ. Microbiol. 12, 2998–3006. Chu, H., Sun, H., Tripathi, B.M., Adams, J.M., Huang, R., Zhang, Y., Shi, Y., 2016. Bacterial community dissimilarity between the surface and subsurface soils equals horizontal differences over several kilometers in the western Tibetan Plateau. Environ. Microbiol. 18, 1523–1533. Curiel Yuste, J., Baldocchi, D., Gershenson, A., Goldstein, A., Misson, L., Wong, S., 2007.

87

Applied Soil Ecology 128 (2018) 81–88

X. Wu et al.

slump in the Qinghai-Tibet plateau. Cold Reg. Sci. Technol. 83, 131–138. Paré, M.C., Bedard-Haughn, A., 2013. Soil organic matter quality influences mineralization and GHG emissions in cryosols: a field-based study of sub-to high Arctic. Glob. Change Biol. 19, 1126–1140. Ping, C., Jastrow, J., Jorgenson, M., Michaelson, G., Shur, Y., 2015. Permafrost soils and carbon cycling. Soil 1, 147–171. Ping, C.L., Michaelson, G.J., Jorgenson, M.T., Kimble, J.M., Epstein, H., Romanovsky, V.E., Walker, D.A., 2008. High stocks of soil organic carbon in the North American Arctic region. Nat. Geosci. 1, 615–619. Post, W.M., Pastor, J., Zinke, P.J., Stangenberger, A.G., 1985. Global patterns of soil nitrogen storage. Nature 317, 613–616. Qin, Y., Wu, T., Ren, L., Yu, W., Wang, T., Zhu, X., Wang, W., Hu, G., Tian, L., 2016. Using ERA-Interim reanalysis dataset to assess the changes of ground surface freezing and thawing condition on the Qinghai-Tibet Plateau. Environ. Earth Sci. 75, 826. Ramette, A., Tiedje, J.M., 2007. Biogeography: An emerging cornerstone for understanding prokaryotic diversity, ecology, and evolution. Microb. Ecol. 53, 197–207. Rousk, J., Baath, E., 2007. Fungal and bacterial growth in soil with plant materials of different C/N ratios. FEMS Microbiol. Ecol. 62, 258–267. Rousk, J., Baath, E., Brookes, P.C., Lauber, C.L., Lozupone, C., Caporaso, J.G., Knight, R., Fierer, N., 2010. Soil bacterial and fungal communities across a pH gradient in an arable soil. ISME J. 4, 1340–1351. Schlesinger, W., 2002. Inorganic carbon and the global carbon cycle. Encyclopedia of soil science. Marcel Dekker, New York, 706-708. Schuur, E., McGuire, A., Schädel, C., Grosse, G., Harden, J., Hayes, D., Hugelius, G., Koven, C., Kuhry, P., Lawrence, D., 2015. Climate change and the permafrost carbon feedback. Nature 520, 171–179. Shang, W., Wu, X., Zhao, L., Yue, G., Zhao, Y., Qiao, Y., Li, Y., 2016. Seasonal variations in labile soil organic matter fractions in permafrost soils with different vegetation types in the central Qinghai-Tibet Plateau. Catena 137, 670–678. Shi, Y., Xiang, X., Shen, C., Chu, H., Neufeld, J.D., Walker, V.K., Grogan, P., 2015. Vegetation-associated impacts on arctic tundra bacterial and microeukaryotic communities. Appl. Environ. Microbiol. 81, 492–501. Soil Survey Staff, 2014. Keys to Soil Taxonomy, Twelfth Edition. USDA-Natural Resources Conservation Service, Washington, DC. Tripathi, B.M., Kim, M., Singh, D., Leecruz, L., Laihoe, A., Ainuddin, A.N., Go, R., Rahim, R.A., Husni, M.H.A., Chun, J., 2012. Tropical soil bacterial communities in Malaysia: pH dominates in the equatorial tropics too. Microb. Ecol. 64, 474–484. Wallenstein, M.D., Mcmahon, S., Schimel, J.P., 2007. Bacterial and fungal community structure in Arctic tundra tussock and shrub soils. FEMS Microbiol. Ecol. 59, 428–435. Wan, X., Huang, Z., He, Z., Yu, Z., Wang, M., Davis, M.R., Yang, Y., 2014. Soil C: N ratio is the major determinant of soil microbial community structure in subtropical coniferous and broadleaf forest plantations. Plant Soil 387, 103–116. Wang, N.F., Zhang, T., Yang, X., Wang, S., Yu, Y., Dong, L.L., Guo, Y.D., Ma, Y.X., Zang, J.Y., 2016. Diversity and composition of bacterial community in soils and lake sediments from an arctic lake area. Front. Microbiol. 7. Wu, X., Fang, H., Zhao, L., Wu, T., Li, R., Ren, Z., Pang, Q., Ding, Y., 2014. Mineralisation and changes in the fractions of soil organic matter in soils of the permafrost Region, Qinghai-Tibet Plateau, China. Permafrost Periglac. Process. 25, 35–44. Wu, X., Fang, H., Zhao, Y., Smoak, J.M., Li, W., Shi, W., Sheng, Y., Zhao, L., Ding, Y., 2017a. A conceptual model of the controlling factors of soil organic carbon and nitrogen densities in a permafrost-affected region on the eastern Qinghai-Tibetan Plateau. J. Geophys. Res. Biogeosci. 122, 1705–1717. Wu, X., Zhao, L., Fang, H., Zhao, Y., Smoak, J.M., Pang, Q., Ding, Y., 2016. Environmental controls on soil organic carbon and nitrogen stocks in the high-altitude-arid western Qinghai-Tibetan Plateau permafrost region. J. Geophys. Res. Biogeosci. 121, 176–187. Wu, X.D., Xu, H.Y., Liu, G.M., Ma, X.L., Mu, C.C., Zhao, L., 2017b. Bacterial communities in the upper soil layers in the permafrost regions on the Qinghai-Tibetan plateau. Appl. Soil Ecol. 120, 81–88. Zhang, T., 2012. Progress in global permafrost and climate change studies. Quat. Sci. 32, 27–38. Zhang, X., Xu, S., Li, C., Zhao, L., Feng, H., Yue, G., Ren, Z., Cheng, G., 2014. The soil carbon/nitrogen ratio and moisture affect microbial community structures in alkaline permafrost-affected soils with different vegetation types on the Tibetan plateau. Res. Microbiol. 165, 128–139. Zhao, L., Wu, Q., Marchenko, S., Sharkhuu, N., 2010. Thermal state of permafrost and active layer in Central Asia during the International Polar Year. Permafrost Periglac. Process. 21, 198–207.

Microbial soil respiration and its dependency on carbon inputs, soil temperature and moisture. Glob. Change Biol. 13, 2018–2035. Deng, J., Gu, Y., Zhang, J., Xue, K., Qin, Y., Yuan, M., Yin, H., He, Z., Wu, L., Schuur, E.A.G., Tiedje, J.M., Zhou, J., 2015. Shifts of tundra bacterial and archaeal communities along a permafrost thaw gradient in Alaska. Mol. Ecol. 24, 222–234. Douterelo, I., Goulder, R., Lillie, M., 2010. Soil microbial community response to landmanagement and depth, related to the degradation of organic matter in English wetlands: Implications for the in situ preservation of archaeological remains. Appl. Soil Ecol. 44, 219–227. Drenovsky, R.E., Vo, D.D., Graham, K.J., Scow, K.M., 2004. Soil water content and organic carbon availability are major determinants of soil microbial community composition. Microb. Ecol. 48, 424–430. Fang, R.-L., Chen, L.-X., Shu, W.-S., Yao, S.-Z., Wang, S.-W., Chen, Y.-Q., 2016. Barcoded sequencing reveals diverse intrauterine microbiomes in patients suffering with endometrial polyps. Am. J. Transl. Res. 8, 1581–1592. Feng, Y., Grogan, P., Caporaso, J.G., Zhang, H., Lin, X., Knight, R., Chu, H., 2014. pH is a good predictor of the distribution of anoxygenic purple phototrophic bacteria in Arctic soils. Soil Biol. Biochem. 74, 193–200. Fierer, N., Jackson, R.B., 2006. The diversity and biogeography of soil bacterial communities. PNAS 103, 626–631. Fierer, N., Schimel, J.P., Holden, P.A., 2003. Variations in microbial community composition through two soil depth profiles. Soil Biol. Biochem. 35, 167–176. Fortier, D., Godin, E., Perreault, N., Levesque, E., 2010. Periglacial landscape stabilization following rapid permafrost degradation by thermo-erosion, Bylot Island, Nunavut, Canadian Arctic Archipelago, AGU Fall Meeting, pp. 0497. Goenster, S., Gründler, C., Buerkert, A., Joergensen, R.G., 2017. Soil microbial indicators across land use types in the river oasis Bulgan sum center, Western Mongolia. Ecol. Ind. 76, 111–118. Hugelius, G., Strauss, J., Zubrzycki, S., Harden, J.W., Schuur, E., Ping, C.-L., Schirrmeister, L., Grosse, G., Michaelson, G.J., Koven, C.D., 2014. Estimated stocks of circumpolar permafrost carbon with quantified uncertainty ranges and identified data gaps. Biogeosciences 11, 6573–6593. Jansson, J.K., Tas, N., 2014. The microbial ecology of permafrost. Nat. Rev. Microbiol. 12, 414–425. Jensen, A., Lohse, K., Crosby, B., Mora, C., 2014. Variations in soil carbon dioxide efflux across a thaw slump chronosequence in northwestern Alaska. Environ. Res. Lett. 9, 025001. Kang, S., Xu, Y., You, Q., Flügel, W.-A., Pepin, N., Yao, T., 2010. Review of climate and cryospheric change in the Tibetan Plateau. Environ. Res. Lett. 5, 015101. Kim, H.M., Jung, J.Y., Yergeau, E., Hwang, C.Y., Hinzman, L., Nam, S., Hong, S.G., Kim, O.S., Chun, J., Lee, Y.K., 2014. Bacterial community structure and soil properties of a subarctic tundra soil in Council, Alaska. FEMS Microbiol. Ecol. 89, 465–475. Liu, J., Sui, Y., Yu, Z., Shi, Y., Chu, H., Jin, J., Liu, X., Wang, G., 2015. Soil carbon content drives the biogeographical distribution of fungal communities in the black soil zone of northeast China. Soil Biol. Biochem. 83, 29–39. Mago, T., Salzberg, S.L., 2011. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics 27, 2957. Mu, C., Zhang, T., Wu, Q., Peng, X., Cao, B., Zhang, X., Cheng, G., 2015. Organic carbon pools in permafrost regions on the Qinghai-Xizang (Tibetan) Plateau. Cryosphere 9, 479–486. Mu, C., Zhang, T., Wu, Q., Peng, X., Zhang, P., Yang, Y., Hou, Y.W., Zhang, X., Cheng, G., 2016a. Dissolved organic carbon, CO2, and CH4 concentrations and their stable isotope ratios in thermokarst lakes on the Qinghai-Tibetan Plateau. J. Limnol. 75, 313–319. Mu, C., Zhang, T., Zhang, X., Cao, B., Peng, X., 2016b. Sensitivity of soil organic matter decomposition to temperature at different depths in permafrost regions on the northern Qinghai-Tibet Plateau. Eur. J. Soil Sci. 67, 773–781. Mu, C., Zhang, T., Zhang, X., Cao, B., Peng, X., Cao, L., Su, H., 2016c. Pedogenesis and physicochemical parameters influencing soil carbon and nitrogen of alpine meadows in permafrost regions in the northeastern Qinghai-Tibetan Plateau. Catena 141, 85–91. Mu, C., Zhang, T., Zhang, X., Li, L., Guo, H., Zhao, Q., Cao, L., Wu, Q., Cheng, G., 2016d. Carbon loss and chemical changes from permafrost collapse in the northern Tibetan Plateau. J. Geophys. Res.Biogeosciences 121, 1781–1791. Mu, C.C., Abbott, B.W., Zhao, Q., Su, H., Wang, S.F., Wu, Q.B., Zhang, T.J., Wu, X.D., 2017. Permafrost collapse shifts alpine tundra to a carbon source but reduces N2O and CH4 release on the northern Qinghai-Tibetan Plateau. Geophys. Res. Lett. 44, 8945–8952. Niu, F., Luo, J., Lin, Z., Ma, W., Lu, J., 2012. Development and thermal regime of a thaw

88