Environmental conditions rather than microbial inoculum composition determine the bacterial composition, microbial biomass and enzymatic activity of reconstructed soil microbial communities

Soil Biology & Biochemistry 90 (2015) 10e18

Contents lists available at ScienceDirect

Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio

Environmental conditions rather than microbial inoculum composition determine the bacterial composition, microbial biomass and enzymatic activity of reconstructed soil microbial communities Weibing Xun a, b, Ting Huang c, Jun Zhao a, Wei Ran a, Boren Wang d, Qirong Shen a, Ruifu Zhang a, b, * a

National Engineering Research Center for Organic-based Fertilizers, Jiangsu Key Lab and Engineering Center for Solid Organic Waste Utilization, Nanjing Agricultural University, Nanjing 210095, PR China Key Laboratory of Microbial Resources Collection and Preservation, Ministry of Agriculture, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, PR China c College of Sciences, Nanjing Agricultural University, Nanjing 210095, PR China d Qiyang Red Soil Experimental Station, Chinese Academy of Agricultural Sciences, Qiyang 426182, PR China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 December 2014 Received in revised form 17 July 2015 Accepted 22 July 2015 Available online 4 August 2015

The composition of microbial communities and the level of enzymatic activity in the soil are both important indicators of soil quality, but the mechanisms by which a soil bacterial community is generated and maintained are not yet fully understood. Two soil samples were collected from the same location, but each had been subjected to a different long-term fertilization treatment and was characterized by different microbial diversity, biomass and physicochemical properties. These samples were g-sterilized and swap inoculated. Non-sterilized soil samples along with sterilized and inoculated soil samples were incubated for eight months before their nutrient content, microbial biomass, enzymatic activity and bacterial composition were analyzed. Total phosphorus, and potassium concentrations along with the overall organic matter content of the non-sterilized soil were all equal to those of the same soil that had been sterilized and self/swap inoculated. Additionally, the microbial biomass carbon concentration was not affected by the specific inoculum and varied only by soil type. The activities of catalase, invertase, urease, protease, acid phosphatase and phytase were smaller in the sterilized soils that had been inoculated with organisms from chemical fertilizer amended soil (NPK) when compared to sterilized soil inoculated with organisms from manure and chemical fertilizer amended soil (NPKM) and non-sterilized soil samples. Bacterial 16S rRNA examined by 454-pyrosequencing revealed that the composition of bacterial community reconstructed by immigrant microbial inoculum in the soil was mainly influenced by its physicochemical properties, although the microbial inoculum contained different abundances of bacterial taxa. For example, the pH of the soil was the dominant factor in reconstructing a new bacterial community. Taken together, these results demonstrated that both soil microbial composition and functionality were primarily determined by soil properties rather than the microbial inoculum, which contributed to our understanding of how soil microbial communities are generated and maintained. © 2015 Elsevier Ltd. All rights reserved.

Keywords: g-ray sterilization Soil incubation Bacterial community Microbial biomass Enzymatic activity

1. Introduction Biological activity in the soil is primarily dependent on ecosystem processes that are supported by microbes (Daily, 1997),

* Corresponding author. College of Resources & Environmental Science, Nanjing Agricultural University, Nanjing, 210095, China. Tel.: þ86 25 84396477; fax: þ86 25 84396260. E-mail address: [email protected] (R. Zhang). http://dx.doi.org/10.1016/j.soilbio.2015.07.018 0038-0717/© 2015 Elsevier Ltd. All rights reserved.

which are especially important in maintaining the magnitude and stability of nutrient cycling (Weyhenmeyer et al., 2013). Therefore, identifying the mechanism by which microbial communities are generated and maintained in the soil is essential for understanding its biological functions. In general, organisms that arrive in a new environment either by being artificially transported or by natural dispersal need to cope with two main stresses: environmental changes and novel in
mez et al., 2010). teractions with local biological communities (Go

W. Xun et al. / Soil Biology & Biochemistry 90 (2015) 10e18

Bacteria utilize a range of strategies to adapt to their new environment, including the development of morphological plasticity (Justice et al., 2008), the utilization of an arginine deaminase system in response to an acid environments (Marquis et al., 1987) and the interaction with phagocytic cells in response to oxidative stress (Hassett and Cohen, 1989). A century ago microbial biogeography was described in the following manner: “Everything is everywhere, the environment selects” (Beijerinck, 1913); however, recent thinking has challenged this idea. Organisms become dormant when they faced with unfavorable environmental conditions (Jones and Lennon, 2010). These unfavorable environmental conditions including deviation from optimum pH values, poor soil nutrient, high or low temperatures, toxic substances, etc. are survival challenges for organisms. Understanding these challenges is important for in situ microbial community reconstruction. Dormant microorganisms generate a microbial inoculum (Lennon and Jones, 2011), which is responsible for the diversity and dynamics of communities in the future. Previous study (CruzMartínez et al., 2009) has addressed in situ changes in the composition of microbial communities that have the same original microbial community, but the introduction of an entirely new microbial inoculum into a novel environment, especially a sterilized one, has not yet been well studied. Nannipieri et al. (1983) suggested that each soil has its own ‘biological space’ and therefore maintains a specific microbial biomass value under conditions of equilibrium. Soil type might be the determinant which tend to maintain the composition of bacterial communities (Girvan et al., 2003). Griffiths et al. (2008) established a swapping microbial community of sandy and clayloam soil and reported that the composition of a microbial community depends on soil type rather than the source of the inoculum. This was confirmed by Delmont et al. (2014), who identified the composition of the bacterial communities found in two different sterilized soils that had been inoculated with either a nonsterilized sample from the same soil or with a non-sterilized foreign soil sample. However, these experiments used essentially the different soils with different parent materials and different vegetations. Moreover, the new microbial activity in these reconstructed microbial communities was not examined, and this information may provide insight into the relationship between microbial diversity and microbial functions. We hypothesized that not only the soil bacterial composition but also the soil microbial functionality was primarily determined by soil properties rather than the microbial inoculum. To test this, two soil samples of the same origin but different in fertilization management over the previous 23 years were used. Their microbial communities were swap-inoculated and the composition of the reconstructed bacterial communities was determined by barcode pyrosequencing of the 16S rRNA gene (Borneman and Triplett, 1997). Functional gene abundances were detected using quantitative real-time PCR. Soil enzymatic activity, physicochemical properties and microbial biomass were also assessed. 2. Materials and methods 2.1. Soil sampling Soil samples were collected from the Red Soil Experiential Station (RSES) maintained by the Chinese Academy of Agricultural Sciences, Qiyang (111530 E, 26 450 N), Hunan Province, southern China, which is located at an altitude of 120 m. The soil at this station is known as Ferralic Cambisol which originally developed from Quaternary red clay. Fertilization experiment began in 1990 and included annual rotations of winter wheat (Triticum aestivum

11

L.) and summer maize (Zea mays L.). Different fertilization treatments were implemented with two replicates in a random block design (Chen et al., 2014). Soil samples were collected in November 2011 from two fertilization treatments, i.e. chemical fertilization (nitrogen, phosphate, and potassium fertilizer, NPK) and manure chemical combined fertilization (NPKM). Fresh samples of each treatment were obtained from the two replicated plots (8 random soil cores per plot) and mixed thoroughly for further study. Both soil samples were passed through a 2 mm sieve after having been temporarily preserved in a portable storage box and transported to the lab. Subsamples that were used for measuring enzymatic activity and physicochemical properties were air-dried, subsamples used for DNA extraction were stored at 80  C, and others subsamples were temporarily stored at 4  C for further study. 2.2. Microbial community swapping and incubating After soil samples stored at 4  C were sterilized by gamma ray (60 kGy) irradiation, a 200 g portion sterilized soil was placed into a 500 mL bottle and allowed to stabilize for 8 weeks at room temperature (Ramsay and Bawden, 1983). Next, sterile water was added to maintain a constant moisture level (30% of field capacity), and the samples were pre-incubated at 20  C in the dark for 2 weeks before being tested for sterility. Microbial inocula were prepared from both NPK and NPKM amended soil samples that had been stored at 4  C. First, the soil samples were pre-incubated at 20  C in regularly aerated bags for two weeks. Next, 20 g of soil was placed in 180 mL of sterile water along with 20 g of glass beads (3e4 mm in diameter) and shaken for 20 min. After 20 mL of this soil suspension was mixed with 200 g of g-ray sterilized soil, microbial inocula from NPK amended soil were introduced into sterilized soil samples from both the NPK (self inoculation control, designated as NPKtoNPK) and NPKM (designated as NPKtoNPKM) groups. Conversely, microbial inocula from NPKM amended soil were introduced into sterilized soil samples from both the NPKM (self inoculation control, designated as NPKMtoNPKM) and NPK (designated as NPKMtoNPK) groups. In addition, 200 g of nonsterilized NPK and NPKM soil samples that had been stored at 4  C were placed into a sterilized 500 mL bottle to serve as controls (designated as CKNPK and CKNPKM, respectively). Three replicates of each control, self inoculation control and swap inoculation treatment were randomly blocked and incubated at 20  C at constant moisture (45% of field capacity) for eight months (2012.01e2012.08). 2.3. Soil analysis After eight-month incubation, the soils were passed through a 2 mm sieve, and several measurements were taken. Soil pH was assessed with a PHSe3C mv/pH detector (Shanghai, China) at a soilto-water ratio of 1:5, available P (AP) was extracted with sodium bicarbonate and its concentration was determined by the molybdenum blue method (Olsen et al., 1954), available K (AK) was extracted with ammonium acetate and its concentration was determined by flame photometry (Dahnke, 1988), total N (TN) was determined by Kjeldahl digestion (Bremner and Mulvaney, 1982), total P (TP) and total K (TK) were isolated by HFeHClO4 (Jackson, 1958) and their concentrations were determined by molybdenum-blue colorimetry and flame photometry respectively, soil organic matter (SOM) concentrations were determined by the potassium dichromate volumetric method (Schollenberger, 1931) and microbial biomass carbon (MBC) concentrations were determined by the chloroform fumigation-extraction method (Vance et al., 1987).

12

W. Xun et al. / Soil Biology & Biochemistry 90 (2015) 10e18

The enzymatic activity of six soil enzymes were also measured (Alef and Nannipieri, 1995): (i) catalase, using the permanganimetric method; (ii) urease, using the phenol sodium hypochlorite colorimetric method to determine the level of ammonium produced; (iii) invertase, using the 3, 5-dinitro salicylic acid colorimetric method; (iv) protease, using the Gareth Jiang method; (v) acid phosphatase, using the disodium phenyl phosphate colorimetric method; and (vi) phytase, using the sodium phytate decomposition and stannous chloride colorimetric method. 2.4. Bacterial community composition and functional genes abundances DNA was extracted from a total of 1.0 g soil using the PowerSoil DNA Isolation Kit (Mo Bio Laboratories Inc., Carlsbad, CA, USA). Three successive DNA extractions of each sample were pooled, and the quality of the DNA was assessed based on the 260/280 nm and 260/230 nm absorbance ratios as measured by a NanoDrop ND1000 Spectrophotometer (NanoDrop, ND2000, ThermoScientific, 111 Wilmington, DE). Quantitative real-time PCR was carried out on an ABI 7500 realtime PCR system (Applied Biosystems, America) using SYBR green as a fluorescent dye to determine the relative abundance of functional genes such as ammonia monooxygenase (amoA), nitrate reductase (narG), nitrite reductase (nirK and nirS), nitrous oxide reductase (nosZ) and nitrogenase (nifH) (Table S2) genes (Rotthauwe et al., 1997; Braker et al., 1998; Poly et al., 2001; €ck et al., 2004; Ro €sch and Bothe, Philippot et al., 2002; Throba 2005). All PCR runs started with an initial enzyme activation step performed at 95  C for 5 min followed by 40 cycles of 95  C for 15 s, 60  C for 34 s, and a final temperature increase to 95  C for 15 s. The specificity of the amplified products was also verified. Based on the V1eV3 hypervariable regions of the bacterial 16S rRNA, PCR primers F27: 50 -GCCTTGCCAGCCCGCTCAG-TC-AGAGTTTGATCCTGGCTCAG-30 and R533: 50 -GCCTCCCTCGCGCCATCAG-ACNNNNNNNNNN-TTACCGCGGCTGCTGGCAC-30 were selected as previously described by Dethlefsen et al. (2008) and Huse et al. (2008). Each of these fusion primers included Roche-454 A/B adapters (shown in italics) and a 2-bp linker sequence (shown in bold) followed by a unique, error-correcting barcode sequence (Ns) and finally the 16S rRNA primer. The region amplified by this primer set is well-suited for accurate phylogenetic analysis of bacterial sequences (Liu et al., 2007). The bacterial community in each soil sample was subsequently characterized by sequencing on a 454 GS-FLX Titanium System (Roche, Switzerland) by Majorbio Bio-pharm Technology Co., Ltd (Shanghai, China). Pyrosequencing data were processed using Mothur (version 1.27.1) (Schloss et al., 2009) following the Schloss standard operating procedure (SOP) (http://www.mothur.org/wiki/ 454_SOP) (Schloss et al., 2011). Those sequences that contained a minimum flow length of 360 and a maximum flow length of 450 were de-noised using the Mothur-based re-implementation of the PyroNoise algorithm (Quince et al., 2011) with the default parameters. De-noised sequences that (i) contained more than two mismatches to the forward primer sequence or/and one mismatch to the barcode sequence, (ii) contained non-assigned tags, (iii) contained any ambiguous base calls (Ns), (iv) contained a homopolymer longer than eight nucleotides, or (v) were shorter than 200 bases. The raw sequences were sorted and distinguished by unique 10bp barcodes. The barcode and primer sequences were then trimmed, and then all unique sequences were aligned against the Silva database (Pruesse et al., 2007). Through screening, filtering, preclustering processes, chimera removal and singleton sequences removal, the retained sequences defined as usable reads were used

to build a distance matrix with a distance threshold of 0.2. Using the average neighbor algorithm with a cutoff of 97% similarity, all sequences were clustered in operational taxonomic units (OTUs). The remaining sequences were sorted into each sample based on their OTU, and the most abundant sequence in each OTU was selected as a representative sequence. Representative sequences were then taxonomically assigned by the RDP naïve Bayesian rRNA classifier (Wang et al., 2007) with a confidence cutoff of 0.6. If an OTU (with more than one read) only appeared in one sample and was detected in less than two replicates, it was removed. 2.5. Statistical analyses For the pyrosequencing data, the percentage of each taxonomy was designated as the relative abundance. Detrended correspondence analysis (DCA) was employed to determine the transformation and overall functional changes of the bacterial communities (Hill and Gauch, 1980), and redundancy analysis (RDA) was performed to determine the most significant soil variables that shaped their composition and structure (Ramette and Tiedje, 2007). For the RDA, a Mantel test was used to examine the correlation between community structure and each variable. Only variables that were determined to be significant by the Mantel test (P < 0.05) were included in future analyses. The Multiple Response Permutation Procedure (MRPP) was used to compare the community with respect to the dissimilarity of their environmental variables and the Permutational Multivariate Analysis of Variance (adonis) was used to test the differences in communities composition depending on some factors. All statistical analyses were performed with the Vegan package (v.2.0-8) (Dixon, 2003) in R software version 3.0.1. 3. Results 3.1. Varying long-term fertilization had a significant effect on soil properties Despite originating from the same soil, the two soil samples we collected in this study had significantly different characteristics after twenty-three years of different fertilizations. The pH and nutrient content were significantly greater (P < 0.001) in NPKM soil samples than NPK (Chen et al., 2014). This suggests that the current bacterial composition of these two soils is also different even though the bacteria likely originated from the same community many years ago. 3.2. Swapping microbial inocula did not change the soil properties After an eight-month incubation period, both the physicochemical properties (P < 0.001) (Table 1) and the enzyme activities (P < 0.001) of each soil sample (Table 2) were significantly different. While the total phosphorus (TP), total potassium (TK), organic matter (SOM) and microbial biomass carbon (MBC) concentrations did not change significantly in NPKtoNPK or NPKMtoNPK samples as compared to CKNPK, soil pH, available nitrogen (AN), available phosphorus (AP) and available potassium (AK) (Table 1) were significantly different. NPKM soil samples behaved in a similar fashion. 3.3. Microbial inocula had little impact on functional gene abundances and enzymatic activities A real-time PCR-based quantitative analysis of N cycling genes under each treatment condition (Fig. 1) revealed a significant difference between NPK and NPKM soils while less differences was

W. Xun et al. / Soil Biology & Biochemistry 90 (2015) 10e18

13

Table 1 Soil physicochemical characteristics and microbial biomass carbon concentrations of the controls, self inoculation and swap inoculation treatments. Treatmenta NPK soils CKNPK NPKtoNPK NPKMtoNPK NPKM soils CKNPKM NPKMtoNPKM NPKtoNPKM

pHb

AN (mg kg1)

TN (g kg1)b

AP (mg kg1)b

TP (g kg1)b

AK (mg kg1)

TK(g kg1)b

SOM (g kg1)b

MBC (mg kg1)b

4.22 ± 0.01ac 4.34 ± 0.02b 4.45 ± 0.00c

111 ± 0.8a 117 ± 0.1ab 149 ± 0.8c

1.13 ± 0.03a 1.17 ± 0.05a 1.32 ± 0.05b

63.9 ± 0.3a 71.8 ± 0.1b 75.8 ± 0.2c

1.41 ± 0.02a 1.31 ± 0.13a 1.21 ± 0.28a

422 ± 1.0c 391 ± 1.0b 303 ± 1.0a

14.9 ± 0.4a 14.2 ± 0.2a 14.6 ± 1.1ab

14.5 ± 0.1a 12.8 ± 1.2a 14.1 ± 0.6a

176 ± 22.6a 153 ± 28.5a 214 ± 34.1a

6.03 ± 0.02b 5.89 ± 0.03a 5.85 ± 0.02a

117 ± 4.6ab 111 ± 0.8a 120 ± 1.5b

1.61 ± 0.02b 1.61 ± 0.03b 1.47 ± 0.01a

170 ± 0.4a 181 ± 0.2c 174 ± 0.2b

2.64 ± 0.31a 2.09 ± 0.33a 2.70 ± 0.04a

293 ± 0.5a 401 ± 1.5b 406 ± 1.0c

13.7 ± 0.2ab 13.9 ± 0.1ab 13.2 ± 0.1a

21.4 ± 0.7a 20.2 ± 1.1a 20.6 ± 1.0a

448 ± 19.0a 456 ± 27.2a 417 ± 12.3a

AN, soil available nitrogen; TN, soil total nitrogen; AP, soil available phosphorus; TP, soil total phosphorus; AK, soil available potassium; TK, soil total potassium; SOM, soil organic matter; MBC, microbial biomass carbon. a NPK soils, non-sterilized NPK soil (CKNPK) and sterilized NPK soil inoculated with suspensions (NPKtoNPK, sterilized NPK soil inoculated with suspension from NPK amended soil; NPKMtoNPK, sterilized NPK soil inoculated with suspension from NPKM amended soil); NPKM soils, non-sterilized NPKM soil (CKNPKM) and sterilized NPKM soil inoculated with suspensions (NPKMtoNPKM, sterilized NPKM soil inoculated with suspension from NPKM amended soil; NPKtoNPKM, sterilized NPKM soil inoculated with suspension from NPK amended soil). b Values showed significant difference (P < 0.001) between NPK soils and NPKM soils. c Values (mean ± standard deviation) indicate the absolute amount of each characteristic. Different letters in column (shown in Bold) indicate significant differences (P < 0.05) between amendments within same soil according to Duncan's multiple comparison.

Table 2 Six enzyme activities of the controls, self inoculation and swap inoculation treatments. Treatmenta NPK soils CKNPK NPKtoNPK NPKMtoNPK NPKM soils CKNPKM NPKMtoNPKM NPKtoNPKM

Catalaseb [mg H2O2 g1 h1]

Invertaseb [mg glucose g1 h1]

Ureaseb [mg NH4 þ  N g1 h1]

Proteaseb [mg NH4 þ  N g1 h1]

Acid phosphataseb [mg PNP g1 h1]

Phytaseb [mg P2O5 g1 h1]

3.49 ± 0.74ac 1.79 ± 0.34a 2.75 ± 0.51a

0.917 ± 0.058c 0.470 ± 0.026a 0.776 ± 0.040b

0.445 ± 0.043b 0.228 ± 0.023a 0.376 ± 0.034b

0.686 ± 0.209b 0.351 ± 0.114a 0.580 ± 0.144b

0.404 ± 0.039b 0.207 ± 0.024a 0.342 ± 0.027b

0.119 ± 0.028b 0.061 ± 0.013a 0.101 ± 0.019b

16.6 ± 3.1b 15.1 ± 3.1ab 12.0 ± 2.9a

2.89 ± 0.12c 2.68 ± 0.12b 2.27 ± 0.05a

1.50 ± 0.10c 1.34 ± 0.10b 1.05 ± 0.05a

1.47 ± 0.06b 1.38 ± 0.06b 1.11 ± 0.07a

0.926 ± 0.080b 0.853 ± 0.080b 0.726 ± 0.047a

0.263 ± 0.058b 0.243 ± 0.057ab 0.190 ± 0.033a

a NPK soils, non-sterilized NPK soil (CKNPK) and sterilized NPK soil inoculated with suspensions (NPKtoNPK, sterilized NPK soil inoculated with suspension from NPK amended soil; NPKMtoNPK, sterilized NPK soil inoculated with suspension from NPKM amended soil); NPKM soils, non-sterilized NPKM soil (CKNPKM) and sterilized NPKM soil inoculated with suspensions (NPKMtoNPKM, sterilized NPKM soil inoculated with suspension from NPKM amended soil; NPKtoNPKM, sterilized NPKM soil inoculated with suspension from NPK amended soil). b Values showed significant difference (P < 0.001) between NPK soils and NPKM soils. c Values (mean ± standard deviation) indicate the absolute activity of each enzyme. Different letters in column (shown in Bold) indicate significant differences (P < 0.05) between amendments within same soil according to Duncan's multiple comparison.

seen between treatments of the same soil. All functional genes involved in N cycling were significantly greater in NPKM soils. All NPKM soil samples displayed greater levels of enzymatic activity when compared to NPK soil samples (Table 2). Among the NPK soil samples, CKNPK had the greatest enzymatic activity,

followed by NPKMtoNPK and NPKtoNPK. Similarly, among NPKM soil samples, CKNPKM had the greatest enzymatic activity, followed by NPKMtoNPKM and NPKtoNPKM. To determine the role of quantity or quality of microbes in the recovery of enzymatic activity, we calculated the ratio of enzyme activity to microbial

Fig. 1. Real-time PCR-based quantitative analysis of nitrogen cycling genes under each treatment condition. Asterisks (*) indicate significance between gene abundances from NPK soils and NPKM soils: *P < 0.05, **P < 0.01.

14

W. Xun et al. / Soil Biology & Biochemistry 90 (2015) 10e18

Table 3 Ratios of enzymatic activity (EA) to microbial biomass carbon (MBC) concentration under the controls, self inoculation and swap inoculation treatments. Treatmenta

Catalase/MBC

CKNPK NPKtoNPK NPKMtoNPK CKNPKM NPKMtoNPKM NPKtoNPKM

0.007 0.004 0.004 0.012 0.011 0.010

± ± ± ± ± ±

0.001bb 0.001a 0.001a 0.002d 0.002cd 0.002c

Invertase/MBC 0.126 0.075 0.088 0.155 0.142 0.131

± ± ± ± ± ±

0.016c 0.013a 0.013b 0.008e 0.009d 0.004c

Urease/MBC 0.061 0.037 0.043 0.081 0.071 0.061

± ± ± ± ± ±

Protease/MBC

0.009c 0.007a 0.007b 0.005e 0.006d 0.003c

0.095 0.056 0.066 0.079 0.073 0.064

± ± ± ± ± ±

0.027c 0.019a 0.017 ab 0.004b 0.005b 0.004ab

Acid phosphatase/MBC 2.32 1.38 1.62 2.07 1.88 1.74

± ± ± ± ± ±

0.33e 0.27a 0.25b 0.17d 0.18cd 0.11bc

Phytase/MBC 0.016 0.010 0.012 0.014 0.013 0.011

± ± ± ± ± ±

0.004d 0.002a 0.003ab 0.003cd 0.003bc 0.002abc

a CKNPK, non-sterilized NPK soil; NPKtoNPK, sterilized NPK soil inoculated with suspension from NPK amended soil; NPKMtoNPK, sterilized NPK soil inoculated with suspension from NPKM amended soil; CKNPKM, non-sterilized NPKM soil; NPKMtoNPKM, sterilized NPKM soil inoculated with suspension from NPKM amended soil; NPKtoNPKM, sterilized NPKM soil inoculated with suspension from NPK amended soil. b Values (mean ± standard deviation) indicate the absolute amount of each characteristic. Different letters in column (shown in Bold) indicate significant differences (P < 0.05) between amendments among both soils according to Duncan's multiple comparison.

Table 4 Statistical and significant test results for the effects of the controls, self inoculation and swap inoculation treatments on the overall microbial community structure assessed by two different statistical approaches. Taxa

CKNPK-NPKtoNPKe CKNPK-NPKMtoNPKe NPKMtoNPK-NPKtoNPKe CKNPKM-NPKMtoNPKMe CKNPKM-NPKtoNPKMe NPKtoNPKM-NPKMtoNPKMe NPK soils-NPKM soilsf

OTUb c

d

MRPP [d (P)]

Adonis [F (P)]

MRPP [d (P)]

Adonis [F (P)]

0.263 0.206 0.235 0.403 0.346 0.353 0.411

(0.092) (0.106) (0.085) (0.123) (0.096) (0.125) (0.001)

0.642 0.513 0.449 0.458 0.504 0.551 0.801

0.164 0.169 0.145 0.319 0.339 0.328 0.309

0.592 0.462 0.437 0.408 0.408 0.503 0.781

(0.106) (0.074) (0.023) (0.001) (0.001) (0.001) (0.001)

(0.112) (0.095) (0.115) (0.112) (0.104) (0.104) (0.001)

(0.079) (0.051) (0.015) (0.001) (0.001) (0.001) (0.001)

Numbers in brackets (shown in Bold) indicate significances (P). a Microbial community structures compared on taxonomic level. b Microbial community structures compared on phylogenetic level. c Multi-response permutation procedure. Statistic is the overall weighted mean of within-group means of the pairwise dissimilarities among sampling units. d Permutational multivariate analysis of variance. Significance tests were carried out using F-tests based on sequential sums of squares from permutations of the raw data. e Comparison between each pair of treatments amended with the same soil. f Comparison between different soils.

biomass carbon (EA/MBC) (Table 3), which was found to be significantly greater in CK conditions when compared to the other two treatments for each soil sample. 3.4. Bacterial community composition was influenced by the soil environment The composition of the bacterial community in each sample was determined by using barcoded pyrosequencing of the 16S rRNA gene (Borneman and Triplett, 1997). Sequences were identified at a value of 90 and then binned into operational taxonomic units (OTUs) at 97% similarity (Roesch et al., 2007). Community diversity and detected OTUs were calculated by using phylogenetic measures, whereas sample similarity was calculated by using both phylogenetic and taxonomic measures. We observed that bacterial communities from NPK soil samples were significantly different from those of NPKM soil samples on both taxonomic and OTU level (Table 4). Multiple Response Permutation Procedure (MRPP) and Permutational Multivariate Analysis of Variance (adonis) analyses revealed a significant difference between the bacterial compositions of NPK and NPKM soils. Adonis analysis also revealed significant differences in composition of NPKM but not in the NPK soil samples. At taxonomic level (Fig. 2), the largest proportional decrease in the composition of NPK soil samples was Bacteroidetes, followed by the phyla Gemmatimonadetes, Verrucomicrobia and Nitrospirae. In contrast, the largest proportional increase was Chloroflexi, followed by Firmicutes and TM7. At the OTU level (Table S1), we observed that richness indexes (Chao and ACE), diversity indexes (Shannon and Simpson) and DOtuN were all significantly greater in NPKM soil samples. In addition, detrended correspondence analysis (DCA)

(Fig. 3) could separate communities of the two soils by DCA1 (37.3%) and could separate communities of NPKM soils by DCA2 (7.7%), which was not possible for the communities of NPK soils. A redundancy analysis (RDA) (Fig. 4), performed by fitting ten soil variables, demonstrated that microbial community composition

Fig. 2. Distribution of 16S rRNA sequences across bacterial phyla under all treatment conditions.

W. Xun et al. / Soil Biology & Biochemistry 90 (2015) 10e18

Fig. 3. Detrended correspondence analysis (DCA) of bacterial communities based on OTUs at a distance of 3% for individual samples. The first two components are 37.3% and 7.7%.

was significantly (r ¼ 0.756, P ¼ 0.001) shaped by several key physical and chemical variables in the soil. Bacterial communities of NPKM and NPK soil samples could be well separated by RDA1 (36.8%) according to some major variables such as pH, SOM, microbial biomass, etc. To determine the relative contribution of soil pH, other soil characteristics such as nutrient contents and inoculum sources to the composition of the bacterial community, a variance partitioning analysis (VPA), which seeks to partition the total variance of the dependent variable into various portions, was performed. To assess the effect of soil nutrient content, a group of characteristic parameters (SOM, MBC, TN and TP) were selected by the BioEnv procedure. These variables explained 51.6% of the observed variation, leaving 48.4% of the variation unexplained (Fig. 5). Soil pH

15

Fig. 5. Variation partitioning analysis (VPA) was used to determine the effects of soil pH (H), inocula sources (I), soil characteristics (C) and interactions between these parameters on the structure of the bacterial community. Circles on the edges of the triangle show the percentage of variation explained by each set of factors alone. The percentage of variation explained by interactions between two or three of these factors is designated by rectangles on the sides or a circle in the middle of the triangle. The unexplained variation is depicted as a rectangle on the bottom.

explained the largest portion (23.7%, P ¼ 0.002) of the observed variation, followed by other soil characteristics (12.9%, P ¼ 0.021), and the sources of the inoculum only explained a very small amount of the variation (4.2%, P ¼ 0.075). The interactions between these variables also accounted for part of the observed variation, e.g., interactions between soil pH and other soil characteristics accounted for 6.2% (P ¼ 0.031) and interactions with inoculum explained other 1.8 þ 2.3 þ 0.5 ¼ 4.6% of variation. Thus, soil pH, soil characteristics and the interaction between the two are important factors in explaining changes in the composition of the bacterial community in soil. 4. Discussion

Fig. 4. Redundancy analysis (RDA) identified nine selected environmental variables among bacterial communities.

In this study, soil samples that were taken from quaternary red clay with the same vegetation and climate differed in soil pH, nutrient content, microbial biomass, enzymatic activity and microbial diversity after years of exposure to different fertilizers. After eight months of incubation, the microbial biomass of sterilized and inoculated soil matched that of the control soil, which suggests that the properties of the soil rather than the microbial inoculum were more important in determining the quantity of soil microbiota. This observation seems to support the hypothesis by Nannipieri et al. (1983) that each soil has a specific biological content under equilibrium conditions. In other words, the overall content is due to the soil microenvironment providing the right set of conditions for soil microbiota to grow. At the microniche levels, nutrient availability, pH, soil moisture, organic carbon content and vegetation type may affect microbial composition (Buckley and Schmidt, 2002). The most crucial requirement when bacteria migrate to a new habitat may be the ability to adapt before surviving and establishing a new
mez et al., 2010). community (Go Dick et al. (1988) reported that some enzyme activities have positive correlation and some have negative correlation with soil pH, soil organic C and total N content. Experiments in both

16

W. Xun et al. / Soil Biology & Biochemistry 90 (2015) 10e18

sterilized and inoculated soil samples demonstrated that local soil properties were more important than the microbial inoculum in affecting the overall enzymatic activity because sterilized and nonsterilized soils showed identical levels except the sterilized soil that received the NPK inoculum. These results suggest that low pH and/ or inhibitors of enzymatic activity/synthesis that were present in the NPK inoculum may have slowed down the recovery of enzymatic activity in these samples. An incubation time longer than eight months would be required to test this postulation. g-ray sterilized soil still exhibits enzymatic activity when suitable substrates are present (McLaren et al., 1962). Various enzymes have been shown to resist irradiation, and their location has been suggested to be factor in this resistance (Lensi et al., 1991). In general, different enzymes have different sensitivity to gamma rays (Shih and Souza, 1978); therefore, the activity of each enzyme would vary to a different degree in soil irradiated with 60 kGy gamma irradiation. NPKM soil samples exhibited significantly greater enzymatic activity when compared to NPK samples. However, only three EA/MBC ratios were significantly greater in NPKM soils, indicating that the activity of several enzymes (protease, acid phosphatase and phytase) aligned with microbial biomass, while other soil enzymes (catalase, invertase and urease) gain more beneficial effects when exposed to relatively good soil conditions. These results suggested that the quantity of microbes contributed more to the recovery of the activities of protease, acid phosphatase and phytase, whereas the quality of microbes contributed more to the recovery of the activities of catalase, invertase and urease. With regards to NPK and NPKM soils respectively, CK samples always displayed the greatest enzymatic activity, followed by NPKMto samples and NPKto samples. Interestingly, EA/MBC ratios were also greatest in CK samples and smallest in NPKto samples for NPK and NPKM soils respectively, which suggests that: (I) Enzymatic activity in the soil was at least partially destroyed with 60 kGy gamma irradiation; (II) This activity gradually returned to normal levels during the reconstruction of microbial communities; and (III) The degree of recovery was affected by the source of the inocula. Bacterial richness and diversity are greater in soils with a neutral pH and smaller in acidic soils (Rousk et al., 2010). We observed that richness indexes (Chao and ACE), diversity indexes (Shannon and Simpson) and DOtuN were all significantly greater in NPKM soil samples (Table S1). Differences in bacterial diversity and richness between these soils could primarily be explained by soil pH and several other characteristics (Fig. 5). In addition, microbial biomass depended only on incubating conditions. Thus, bacteria richness, diversity and microbial biomass in NPKM soil samples were significantly greater because of the acid pH and greater nutrient content of that soil. Dormant bacteria may account for up to 40% of taxonomic richness in nutrient-poor systems, though they are relatively scarce in productive ecosystems (Jones and Lennon, 2010). Here, Acidobacteria and Proteobacteria were both dominate in NPK and NPKM soils, whereas Bacteroidetes, Gemmatimonadetes, Verrucomicrobia and Nitrospirae were proportional decreased and Chloroflexi, Firmicutes and TM7 were proportional increased in NPK soils. These results suggested that inocula from NPK amended soil that was subsequently incubated in NPKM soil established a community similar to the community from CKNPKM soil, whereas inocula from NPKM amended soil that was incubated in NPK soil established a community similar to the community from CKNPK soil. Therefore, both inocula appeared to contain bacterial communities with nearly the same taxonomic composition (but different abundances) because they evolved from the same community despite different fertilization conditions. Interestingly, bacterial communities from NPK soil samples showed no significant differences in bacterial composition and relative abundance according to MRPP, adonis and

DCA analysis, whereas bacterial communities from NPKM soil samples differed significantly according to both adonis and DCA (but not MRPP) analysis. The similarity of these soil bacterial communities was largely related to environmental characteristics as opposed to geographical proximity (Fierer and Jackson, 2006), which suggests that the more closely related the environmental characteristics these bacterial communities encounter are, the more similar bacterial communities will be reconstructed. In general, functional gene expression is correlated with bacterial community composition; therefore, more functional gene copies were detected in NPKM soil samples. Consider two sterilized soil samples with different environmental characteristics such as soil pH, nutrient contents, moisture, available carbon, etc. When inocula are introduced to one of these soils as bacterial inocula, they initially develop randomly after a short adaptation period because of the abundant space and nutrients. However, as bacteria proliferate, environmental stresses that act at the single cell level will enhance until bacterial activity and environmental stress finally achieve a dynamic equilibrium. Based on these limitations imposed by environmental characteristics and using nonmetric multidimensional scaling of bacterial communities, we constructed a model to describe the final state of bacterial community development under varying environmental conditions. First, bacterial communities develop in soils with the same pH (Fig. 6a). Bacterial inoculum (point A) is inoculated into a soil sample where they find a potential development space (space B), which allows several types of bacterial communities to be constructed. The potential direction of development (DD1, DD2, DD3, DD4…DDi) is randomly decided. However, the radii of

Fig. 6. A model used to describe the potential development of bacterial communities under different environmental conditions. (a) Bacterial community development in soils with the same pH. Point A represents “bacterial inoculum” and space B represents “soil environment” for bacterial community development. DD1, DD2, DD3, DD4…DDi are potential directions for this development. R is the radius of development for each soil sample. (b) Influence of soil pH (from acidic to alkaline). Cross sections are the same conditions previously described in (a).

W. Xun et al. / Soil Biology & Biochemistry 90 (2015) 10e18

different soil samples are not identical, e.g., acidified (or alkalized) soils will restrict bacterial community development. VPA analysis has revealed that soil characteristics are good predictors of bacterial populations. Previous studies have also demonstrated that soil moisture (Horn et al., 2014), soil organic carbon (Nakatsu et al., 2005) and soil C: N ratio (Shen et al., 2013) may influence bacterial communities. Second, when soil pH is taken into consideration (Fig. 6b), the model looks like a rugby. In the middle are neutral soils, where bacterial inocula have a larger space for development and are therefore able to construct more varieties of stable communities among different soil compositions. 5. Conclusions This study demonstrated that the bacterial composition, microbial biomass and enzymatic activity of reconstructed soil microbial community were primarily determined by soil properties but not on the microbial inoculum. Among the soil properties studied here, pH was the dominant factor in reconstructing a new bacterial community. This results support the findings of Griffiths et al. (2008) and Delmont et al. (2014). The activities of catalase, invertase, urease, protease, acid phosphatase and phytase were greater in non sterilized soil when compared to sterilized soil and were greater in NPKMto sample when compared to NPKto sample for NPK and NPKM soils respectively. It is likely that biological recovery was not complete when NPK amended soil was used as an inoculum. Moreover, NPKM soil samples contain more functional gene copies. Thus, we concluded that the soil microbial functionality was primarily determined by soil properties, as is the case for bacterial community composition. Acknowledgments The authors thank the staff at the Qiyang Red Soil Experimental Station for managing the field experiments and helping with the collection of soil samples. This research was financially supported by the Chinese Ministry of Science and Technology (2015CB150500), the Fundamental Research Funds for the Central Universities (KYTZ201404) and the China Postdoctoral Science Foundation (2014M560430). RZ and QS were also supported by the Priority Academic Program Development (PAPD) of Jiangsu Higher Education Institutions and the 111 Project (B12009). The authors are grateful for Prof. Paolo Nannipieri of University of Florence for his helpful discussion on this manuscript. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.soilbio.2015.07.018. References Alef, K., Nannipieri, P., 1995. Methods in Applied Soil Microbiology and Biochemistry. Acad. Press. €n. Jaarb. Van K. Beijerinck, M.W., 1913. De infusies en de ontdekking der backterie Akad. Van Wet., pp. 1e28. Borneman, J., Triplett, E.W., 1997. Molecular microbial diversity in soils from eastern Amazonia: evidence for unusual microorganisms and microbial population shifts associated with deforestation. Applied and Environmental Microbiology 63, 2647e2653. Braker, G., Fesefeldt, A., Witzel, K.P., 1998. Development of PCR primer systems for amplification of nitrite reductase genes (nirK and nirS) to detect denitrifying bacteria in environmental samples. Applied and Environmental Microbiology 64, 3769e3775. Bremner, J.M., Mulvaney, C.S., 1982. Methods of Soil Analysis. Part 2. Chemical and Microbiological Properties. Methods Soil Anal., pp. 595e624. Buckley, D.H., Schmidt, T.M., 2002. Exploring the biodiversity of soilda microbial rain forest. In: Biodivers. Microb. Life. Wiley-Liss Inc.

17

Chen, L., Xun, W., Sun, L., Zhang, N., Shen, Q., Zhang, R., 2014. Effect of different longterm fertilization regimes on the viral community in an agricultural soil of Southern China. European Journal of Soil Biology 62, 121e126. Cruz-Martínez, K., Suttle, K.B., Brodie, E.L., Power, M.E., Andersen, G.L., Banfield, J.F., 2009. Despite strong seasonal responses, soil microbial consortia are more resilient to long-term changes in rainfall than overlying grassland. ISME Journal 3, 738e744. Dahnke, W.C., 1988. Recommended Chemical Soil Test Procedures for the North Central Region. Bulletin 499. Daily, G., 1997. Nature's Services: Societal Dependence on Natural Ecosystems. Island Press. Delmont, T.O., Francioli, D., Jacquesson, S., Laoudi, S., Mathieu, A., Nesme, J., Ceccherini, M.T., Nannipieri, P., Simonet, P., Vogel, T.M., 2014. Microbial community development and unseen diversity recovery in inoculated sterile soil. Biology and Fertility of Soils 50, 1069e1076. Dethlefsen, L., Huse, S., Sogin, M.L., Relman, D.A., 2008. The pervasive effects of an antibiotic on the human gut microbiota, as revealed by deep 16S rRNA sequencing. PLoS Biology 6, e280. Dick, R.P., Rasmussen, P.E., Kerle, E.A., 1988. Influence of long-term residue management on soil enzyme activities in relation to soil chemical properties of a wheat-fallow system. Biology and Fertility of Soils 6, 159e164. Dixon, P., 2003. VEGAN, a package of R functions for community ecology. Journal of Vegetation Science 14, 927e930. Fierer, N., Jackson, R.B., 2006. The diversity and biogeography of soil bacterial communities. Proceedings of the National Academy of Sciences 103, 626e631. Girvan, M.S., Bullimore, J., Pretty, J.N., Osborn, A.M., Ball, A.S., 2003. Soil type is the primary determinant of the composition of the total and active bacterial communities in arable soils. Applied and Environmental Microbiology 69, 1800e1809.
mez, J.P., Bravo, G.A., Brumfield, R.T., Tello, J.G., Cadena, C.D., 2010. A phylogenetic Go approach to disentangling the role of competition and habitat filtering in community assembly of Neotropical forest birds. Journal of Animal Ecology 79, 1181e1192. Griffiths, B.S., Hallett, P.D., Kuan, H.L., Gregory, A.S., Watts, C.W., Whitmore, A.P., 2008. Functional resilience of soil microbial communities depends on both soil structure and microbial community composition. Biology and Fertility of Soils 44, 745e754. Hassett, D.J., Cohen, M.S., 1989. Bacterial adaptation to oxidative stress: implications for pathogenesis and interaction with phagocytic cells. FASEB Journal: Official Publication of the Federation of American Societies for Experimental Biology 3, 2574e2582. Hill, M.O., Gauch Jr, H.G., 1980. Detrended correspondence analysis: an improved ordination technique. Vegetatio 42, 47e58. Horn, D.J.V., Okie, J.G., Buelow, H.N., Gooseff, M.N., Barrett, J.E., Takacs-Vesbach, C.D., 2014. Soil microbial responses to increased moisture and organic resources along a salinity gradient in a polar desert. Applied and Environmental Microbiology 80, 3034e3043. Huse, S.M., Dethlefsen, L., Huber, J.A., Welch, D.M., Relman, D.A., Sogin, M.L., 2008. Exploring microbial diversity and taxonomy using SSU rRNA hypervariable tag sequencing. PLoS Genetics 4, e1000255. Jackson, M.L., 1958. Soil Chemical Analysis. Prentice Hall, Englewood Cliffs, NJ. Jones, S.E., Lennon, J.T., 2010. Dormancy contributes to the maintenance of microbial diversity. Proceedings of the National Academy of Sciences 107, 5881e5886. Justice, S.S., Hunstad, D.A., Cegelski, L., Hultgren, S.J., 2008. Morphological plasticity as a bacterial survival strategy. Nature Reviews Microbiology 6, 162e168. Lennon, J.T., Jones, S.E., 2011. Microbial seed banks: the ecological and evolutionary implications of dormancy. Nature Reviews Microbiology 9, 119e130. Lensi, R., Lescure, C., Steinberg, C., Savoie, J.-M., Faurie, G., 1991. Dynamics of residual enzyme activities, denitrification potential, and physico-chemical properties in a g-sterilized soil. Soil Biology and Biochemistry 23, 367e373. Liu, Z., Lozupone, C., Hamady, M., Bushman, F.D., Knight, R., 2007. Short pyrosequencing reads suffice for accurate microbial community analysis. Nucleic Acids Research 35 e120ee120. Marquis, R.E., Bender, G.R., Murray, D.R., Wong, A., 1987. Arginine deiminase system and bacterial adaptation to acid environments. Applied and Environmental Microbiology 53, 198e200. McLaren, A.D., Luse, R.A., Skujins, J.J., 1962. Sterilization of soil by irradiation and some further observations on soil enzyme activity 1. Soil Science Society of America Journal 26, 371. Nakatsu, C.H., Carmosini, N., Baldwin, B., Beasley, F., Kourtev, P., Konopka, A., 2005. Soil microbial community responses to additions of organic carbon substrates and heavy metals (Pb and Cr). Applied and Environmental Microbiology 71, 7679e7689. Nannipieri, P., Muccini, L., Ciardi, C., 1983. Microbial biomass and enzyme activities: production and persistence. Soil Biology and Biochemistry 15, 679e685. Olsen, S.R., Cole, C.V., Watanabe, F.S., Dean, L.A., 1954. Estimation of Available Phosphorus in Soils by Extraction with Sodium Bicarbonate. US Department of Agriculture, Washington, DC. Philippot, L., Piutti, S., Martin-Laurent, F., Hallet, S., Germon, J.C., 2002. Molecular analysis of the nitrate-reducing community from unplanted and maize-planted soils. Applied and Environmental Microbiology 68, 6121e6128. re, F., Monrozier, L.J., 2001. Comparison of Poly, F., Ranjard, L., Nazaret, S., Gourbie nifH gene pools in soils and soil microenvironments with contrasting properties. Applied and Environmental Microbiology 67, 2255e2262.

18

W. Xun et al. / Soil Biology & Biochemistry 90 (2015) 10e18

€ckner, F.O., Pruesse, E., Quast, C., Knittel, K., Fuchs, B.M., Ludwig, W., Peplies, J., Glo 2007. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Research 35, 7188e7196. Quince, C., Lanzen, A., Davenport, R.J., Turnbaugh, P.J., 2011. Removing noise from pyrosequenced amplicons. BMC Bioinformatics 12, 38. Ramette, A., Tiedje, J.M., 2007. Multiscale responses of microbial life to spatial distance and environmental heterogeneity in a patchy ecosystem. Proceedings of the National Academy of Sciences 104, 2761e2766. Ramsay, A.J., Bawden, A.D., 1983. Effects of sterilization and storage on respiration, nitrogen status and direct counts of soil bacteria using acridine orange. Soil Biology and Biochemistry 15, 263e268. Roesch, L.F.W., Fulthorpe, R.R., Riva, A., Casella, G., Hadwin, A.K.M., Kent, A.D., Daroub, S.H., Camargo, F.A.O., Farmerie, W.G., Triplett, E.W., 2007. Pyrosequencing enumerates and contrasts soil microbial diversity. ISME Journal 1, 283e290. € sch, C., Bothe, H., 2005. Improved assessment of denitrifying, N2-fixing, and totalRo community bacteria by terminal restriction fragment length polymorphism analysis using multiple restriction enzymes. Applied and Environmental Microbiology 71, 2026e2035. Rotthauwe, J.H., Witzel, K.P., Liesack, W., 1997. The ammonia monooxygenase structural gene amoA as a functional marker: molecular fine-scale analysis of natural ammonia-oxidizing populations. Applied and Environmental Microbiology 63, 4704e4712. Rousk, J., Bååth, 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 Journal 4, 1340e1351.

Schloss, P.D., Gevers, D., Westcott, S.L., 2011. Reducing the effects of PCR amplification and sequencing artifacts on 16S rRNA-based studies. PLoS One 6, e27310. Schloss, P.D., Westcott, S.L., Ryabin, T., Hall, J.R., Hartmann, M., Hollister, E.B., Lesniewski, R.A., Oakley, B.B., Parks, D.H., Robinson, C.J., Sahl, J.W., Stres, B., Thallinger, G.G., Horn, D.J.V., Weber, C.F., 2009. Introducing mothur: opensource, platform-independent, community-supported software for describing and comparing microbial communities. Applied and Environmental Microbiology 75, 7537e7541. Schollenberger, C.J., 1931. Determination of soil organic matter. Soil Science 31, 483e486. Shen, C., Xiong, J., Zhang, H., Feng, Y., Lin, X., Li, X., Liang, W., Chu, H., 2013. Soil pH drives the spatial distribution of bacterial communities along elevation on Changbai mountain. Soil Biology and Biochemistry 57, 204e211. Shih, K.L., Souza, K.A., 1978. Degradation of biochemical activity in soil sterilized by dry heat and gamma radiation. Origin of Life 9, 51e63. Throb€ ack, I.N., Enwall, K., Jarvis, A., Hallin, S., 2004. Reassessing PCR primers targeting nirS, nirK and nosZ genes for community surveys of denitrifying bacteria with DGGE. FEMS Microbiology Ecology 49, 401e417. Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring soil microbial biomass C. Soil Biology and Biochemistry 19, 703e707. Wang, Q., Garrity, G.M., Tiedje, J.M., Cole, J.R., 2007. Naïve bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Applied and Environmental Microbiology 73, 5261e5267.
n, E., 2013. Shifts in phytoplankton species Weyhenmeyer, G.A., Peter, H., Wille richness and biomass along a latitudinal gradient e consequences for relationships between biodiversity and ecosystem functioning. Freshwater Biology 58, 612e623.