Applied Soil Ecology 99 (2016) 1–12

Contents lists available at ScienceDirect

Applied Soil Ecology

journal homepage: www.elsevier.com/locate/apsoil

Effects of organic–inorganic compound fertilizer with reduced

chemical fertilizer application on crop yields, soil biological activity
and bacterial community structure in a rice–wheat cropping system
Jun Zhaoa,b,c,1, Tian Nia,b,c,1, Jing Lia,b,c,1, Qiang Lua,b,c, Zhiying Fanga,b,c, Qiwei Huanga,b,c ,
Ruifu Zhangd , Rong Lia,b,c, Biao Shena,b,c,* , Qirong Shena,b,c
a
Jiangsu Key Lab and Engineering Center for Solid Organic Waste Utilization, Nanjing Agriculture University, Nanjing 210095, China
b
National Engineering Research Center for Organic-based Fertilizers, Nanjing Agricultural University, Nanjing 210095, China
c
Jiangsu Collaborative Innovation Center for Solid Organic Waste Resource Utilization, Nanjing Agricultural University, Nanjing 210095, China
d
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, China

A R T I C L E I N F O A B S T R A C T

Article history: The development of more stable and sustainable agroecosystems for improving food production has
Received 3 July 2015 caused wide public concern in recent years. In the present study, we conducted a field experiment to
Received in revised form 6 November 2015 investigate the effect of pig manure organic–inorganic compound fertilizer with reduced chemical
Accepted 11 November 2015
fertilizer on the crop yields, soil physicochemical properties, biological activities and bacterial
Available online xxx
community structure in a rice–wheat cropping system over two crop seasons (rice and wheat). The
results showed that at all sampling times, this fertilizer regime enhanced the soil nutrient availability,
Keywords:
microbial biomass, enzymatic activities, and soil nitrogen processes and, to some extent, promoted crop
Organic–inorganic compound fertilizer
Rice–wheat cropping system
yields. Across all soil samples, bacterial communities were dominated by Proteobacteria,Acidobacteria,
Temporal dynamics and Chloroflexi at the phylum level. Hierarchical cluster analysis based on the weighted UniFrac distance
Bacterial community structure revealed that the bacterial community structures were strongly separated by the sampling time, and the
Core microbiome treatments in the wheat harvest soils. A Venn diagram of shared OTUs showed a core microbiome across
different treatments and sampling times, in which the relative abundance of each abundant phylum
(class) was stable in the different treatments and at different sampling times. Specifically, the relative
abundance of Alphaproteobacteria, Gammaproteobacteria, Nitrospirae, Bacteroidetes, and Actinobacteria
was largely and particularly enriched under the organic–inorganic compound fertilizer regime,
indicating that soil functions, such as nitrification and the turnover of organic matter, might be
strengthened under this treatment. Collectively, these results indicate that the application of organic–
inorganic compound fertilizer may reduce chemical fertilizer use and improve the long-term
productivity and sustainability of agroecosystems.
ã 2015 Elsevier B.V. All rights reserved.

1. Introduction This has mainly been attributed to the introduction of high-

The rice (Oryza sativa L.)–wheat (Triticum aestivum L.) cropping
system, which covers an area ranging from 9.5 to 13.5 million ha, is
a long-established grain production system in China and is
considered to be of utmost importance for China’s food security
and livelihood (Gupta et al., 2003). Since the early 1980s, the
productivity of rice and wheat in China has increased dramatically.
* Corresponding author at: College of Resources and Environmental Science,
Nanjing Agricultural University, Nanjing 210095, China. Fax: +86 25 84395212.
E-mail address: shenbiao@njau.edu.cn (B. Shen).
1
These authors contributed equally to this paper.
yielding varieties and the increased input of chemical fertilizers
(Ladha et al., 2004; National Bureau of Statistics of China, 2009).
Recently, concerns have been raised regarding the stagnation and
even decline in the productivity and sustainability of this cropping
system (Ladha et al., 2003; Mandal et al., 2003). Therefore, to meet
the ever-increasing food demand in China, the development of
highly productive and sustainable food production practices is
becoming increasingly important.
A rice–wheat cropping system is a nutrient exhaustive system
that requires appropriate fertilizer management to maintain soil
productivity, improve plant nutrition, and increase crop yields.
Organic fertilizers are known to improve soil quality and structure

http://dx.doi.org/10.1016/j.apsoil.2015.11.006
0929-1393/ ã 2015 Elsevier B.V. All rights reserved.
2 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12
(Bronick and Lal, 2005) as well as stimulate soil microbial biomass
(Masto et al., 2006; Jannoura et al., 2014), enzyme activities (Ge
et al., 2010; Insam et al., 2015; Plaza et al., 2004), and functional
diversity and abundance (Orr et al., 2012) in soil community
structure. However, the effect of organic fertilizers on crop yield is
slow and variable (Khaliq et al., 2006). Thus, farmers prefer to use
inorganic fertilizers rather than organic fertilizers to preserve crop
yield, particularly in intensive agricultural systems under fluctu-
ating environmental conditions (Smith et al., 2007). However, the
long-term oversupply of inorganic fertilizers, especially nitrogen
fertilizer, has resulted in nitrate pollution of groundwater,
eutrophication of surface waters, acid rain and soil acidification,
greenhouse gas emissions, and other forms of air pollution (Ju
et al., 2009). In particular, Guo et al. (2010) reported that soil
acidification had become a major problem in intensive Chinese
agricultural systems. Moreover, the long-term application of
inorganic fertilizers without organic inputs has been identified
as the driving force in the loss of soil organic matter, deterioration
of the soil structure, decrease in biological activities, and thus
reduction in soil fertility (Bronick and Lal, 2005; Ladha et al., 2004;
Zhong and Cai, 2007), which is assumed to be nonsustainable.
Accordingly, the combination of organic and inorganic fertilizers is
a promising approach to develop more sustainable fertilizer
management strategies.
It has been recognized that the addition of organic amendments
and a reduction in synthetic inorganic fertilizer can optimize the
soil microbial-driven internal cycling of nutrients (Germaine et al.,
2010). For instance, reduced inputs of nitrogen fertilizer can
enhance the activity of nitrogen-fixing bacteria and thus promote
biological nitrogen fixation, which supplies nitrogen to agro-
ecosystems (Hsu and Buckley, 2009; Orr et al., 2012). Liu et al.
(2009) reported that the application of organic amendments
coupled with reduced chemical fertilizer enhanced the microbial
biomass, activity, and nutrient availability compared with the use
of chemical fertilizer only. Similarly, combined inorganic/organic
fertilization improves N utilization efficiency and increase rice
productivity through organic carbon accumulation (Pan et al.,
2009). Recently, organic–inorganic compound fertilizer has
become a novel and popular fertilizer in China because of slow
nutrient release, high fertilizer use efficiency, and low environ-
mental pollution (Ni et al., 2010). Thus, it is necessary to
understand the influence of this type of fertilizer on crop
productivity, soil biological activity and microbial assemblages.
Soil microorganisms are considered to play a vital role in
maintaining soil health, productivity, and sustainability (Zhao
et al., 2014a). Therefore, understanding the soil microbial
community and its response to various agricultural management
practices will guide us to select a suitable management strategy for
the establishment of more stable and sustainable agroecosystems
(Li et al., 2012; Zhao et al., 2014b). It is well known that nitrogen
availability often limits plant productivity in terrestrial ecosystems
(LeBauer and Treseder, 2008), and soil nitrogen transformations
are driven directly by a diverse microbial community (Robertson
and Groffman, 2007). For example, biological N2-fixation is the
conversion of dinitrogen (N2) to biologically available ammonium
(NH4+) by free-living, associated and symbiotic diazotrophs from a
wide range of bacterial phyla (Zhang et al., 2006). Therefore, it is
also necessary to understand the dynamics of microbial functional
genes associated with nitrogen cycling in soil applied with
different fertilizer regimes.
We hypothesized that the organic–inorganic compound fertil-
izer with reduced chemical fertilizer would help to establish a
distinct microbial community, facilitate soil microbial-driven
nutrient cycling, enhance enzymatic activities, and thus improve
crop yield. To test this, a field experiment was conducted in a rice–
wheat cropping system to evaluate the effects of this fertilizer
regime on crop yields, soil physiochemical properties, soil enzyme
activities, microbial nitrogen functional genes, and microbial
community structure.
2. Materials and methods
2.1. Field site and experiment description
The experiment was performed in a paddy field under rice–
wheat rotation in Changshu, Jiangsu, China (31
180 N, 120
370 E, 6 m
a.s.l.). This region has a northern subtropical monsoon climate with
an average annual temperature of 15.4
C and a mean annual
precipitation of 1054 mm. The soil type is classified as clay loamy
endogleyic-Fe-leachic-stagnic anthrosol, and the soil properties
before experiment are described by Zhao et al. (2014a).
The fertilization experiments were established in 2010 and
were designed in a randomized complete block with four
replicates, each measuring 40 m2. The treatments were the control
without fertilizer (CK), chemical fertilizers (NPK) (conventional
dosage for this region), 50% chemical fertilizers plus 6000 kg/ha
manure (NPKM) and 30% chemical fertilizers plus 3600 kg/ha pig
manure organic–inorganic compound fertilizer (NPKMOI). The pig
manure organic–inorganic compound fertilizer granules were
comprised of pig manure compost and a suitable amount of multi-
elements with a special coating material. Each plot received the
same levels (except control) of nitrogen, phosphorus, and
potassium from fertilizers applied in the wheat (N: 240 kg/ha,
P2O5: 120 kg/ha and K2O: 100 kg/ha) and rice (N: 300 kg/ha, P2O5:
120 kg/ha and K2O: 100 kg/ha) seasons. The N fertilizer (urea) was
applied as a basal fertilizer and a supplementary fertilizer, while
the P, K, and manure fertilizers were only used as basal fertilizers.
Rice was irrigated routinely to maintain shallow standing water
from the seedling stage to the maturity stage, whereas wheat was
irrigated when necessary for agriculture. The crop grains from the
entire plot were weighed and recorded after air drying.
2.2. Soil sampling and analysis
Soil samples were collected from the plough layer (0–20 cm) of
each replicate plot for each treatment using a 2.5-cm diameter
auger after the harvest of wheat and rice in June and October 2012,
respectively. Each soil sample was a composite of ten soil cores that
were randomly collected, pooled in a sterile plastic bag, and
transported to the laboratory on ice. The soil samples were sieved
(2-mm-mesh sieve) and divided into three subsamples before
being thoroughly homogenized. One portion was stored at 80
C
for DNA extraction, another portion was stored at 4
C for
microbial biomass measurement, and the remainder was air-dried
for soil characteristic and enzyme activity analyses. The soil
edaphic properties of each sample were evaluated in the soil
testing laboratory at the Qiyang Red Soil Experimental Station of
the Chinese Academy of Agricultural Sciences.
2.3. Enzyme activities and microbial biomass analysis
The activities of the soil enzymes were assayed in triplicate with
two non-substrate controls using air-dried soil samples according
to Guan et al. (1986). Alkaline phosphatase (EC 3.1.3.1) activity was
determined using disodium phenyl phosphate as a substrate and
phenol as a product after incubation at 37
C for 24 h. Urease (EC
3.5.1.5) and invertase (EC 3.2.1.26) activities were analyzed by the
released ammonium and glucose equivalent, respectively. Catalase
(EC 1.11.1.6) activity was measured by shaking for 20 min with H2O2
(3.5%) as a substrate, and the suspension was titrated with
0.1 mol L1 KMnO4 solution. Catalase activity was expressed as
0.1 mol L1 KMnO4 ml g1 soil 20 min1, whereas the other enzyme
 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12
activities were expressed as products per gram of dry weight soil
mass per incubation time (24 h).
Microbial biomass carbon (MBC) and nitrogen (MBN) were
determined using the chloroform fumigation–K2SO4 extraction
method (Vance et al., 1987). Fresh soil samples were extracted
using 0.5 mol L1 K2SO4 for 60 min on a rotary shaker after 24 h
fumigation. MBC and MBN were then determined on a LiquiTOC
element analyzer II (Elementar Analysen-system GmbH, Hanau,
Germany) and calculated using correction factors of 0.45 (kEC) and
0.54 (kEN), respectively (Joergensen, 1996; Joergensen and Mueller,
1996).
2.4. DNA extraction
For each soil sample, total microbial genomic DNA was
extracted in triplicate from 0.5 g of frozen soil using a PowerSoil1
DNA Isolation Kit (MoBio Laboratories, Inc., Carlsbad, CA, USA)
according to the manufacturer’s instructions. The triplicate DNA
extracts were pooled to reduce DNA extraction bias. The extracted
DNA was then evaluated on 1% agarose gel, and the quality and
quantity of the extracts were determined using a NanoDrop ND-
2000 spectrophotometer (ThermoScientic, DE, USA). All DNA
samples were diluted to 10 ng ml1 and stored at 20
C until
further use.
2.5. Real-time PCR assay
To generate external standard curves for the real-time PCR
assay, the microbial nitrogen functional genes involved in nitrogen
fixation (nifH) and nitrification (amoA) were PCR-amplified from
soil DNA with each pair of primers listed in Table 1. The PCR
products were gel-purified using the PCR Purification Kit (Axygen
Bio, USA) and ligated into the pMD19-T vector (Takara). The
resulting recombinant plasmids were subsequently transformed
into Escherichia coli Top10 competent cells. After blue–white
screening and re-amplification with the vector-specific primers
RV-M and M13-47, the positive clones were sequenced and BLASTN
was used to identify the insert sequences. Then, the correct insert
clone of each target gene was selected to extract plasmid DNA
using the AxyPrep Plasmid Miniprep Kit (Axygen Bio, USA). The
plasmid DNA concentration was determined on a NanoDrop ND-
2000 spectrophotometer (NanoDrop, Wilmington, DE, USA), and
the copy numbers of each target gene were calculated directly from
the concentration of the extracted plasmid DNA. Ten-fold serial
dilutions of a known copy number of the plasmid DNA were
subjected to real-time PCR assay in triplicate to generate an
external calibration curve. Amplification efficiencies of 94.8–95.2%
were obtained with R2 values of 0.999.
Table 1
Primers and PCR conditions used for the real-time PCR.
Target gene Primer set Sequence (50 –30 )a
Archaeal amoA Arch-amoAF STAATGGTCTGGCTTAGACG
Arch-amoAR GCGGCCATCCATCTGTATGT
Bacterial amoA amoA1F* GGGGHTTYTACTGGTGGT
amoA2R CCCCTCKGSAAAGCCTTCTTC
nifH PolF TGCGAYCCSAARGCBGACTC
PolR ATSGCCATCATYTCRCCGGA
a
Boldface letters denote degenerate positions. B, C/G/T; H, A/C/T; K, G/T; R, A/G; S, G/C; Y, C/T.
3
Real-time PCR was performed in biological triplicates, and each
involved three technical replicates with two negative controls on
an ABI 7500Cycle (Applied Biosystems, Germany) to enumerate the
abundance of the microbial nitrogen genes described above using
the Premix Ex TaqTM Kit (TaKaRa). The reaction mixture contained
10 ml of Premix Ex TaqTM (2) (Takara), 0.4 ml of ROX Reference Dye
II (50), optimized concentrations of primers (Table 1), and 2 ml of
template DNA, made up to with a final volume of 20 ml. The primer
sets and thermal conditions are listed in Table 1. After the real-time
PCR assay, the specificity of the amplification was verified by
melting-curve analysis and agarose gel electrophoresis. Copy
numbers were log10-transformed to normalize the values prior to
statistical analysis.
2.6. 16S rRNA gene amplification and 454 pyrosequencing
For practical reasons, three biological replicates of each treatment
were randomly selected for pyrosequencing. The V1–
V3 hypervariable region of the bacterial 16S rRNA gene was
amplified using individual bar-coded reverse primers 533R, 50 -
TTACCGCGGCTGCTGGCAC-30 , and the forward primer 27F, 50 -
AGAGTTTGATCCTGGCTCAG-30 . The PCR procedures, including the
reaction mixture and thermal profile, are described in detail by Zhao
et al. (2014b). After PCR amplification, the triplicate PCR products of
each sample were pooled, purified using the PCR Purification Kit
(Axygen Bio, USA), and quantified using a TBS-380 Mini-Fluorometer
(Turner Biosystems, Sunnyvale, CA, USA). The 24 purified amplicons
(3 replicates  4 treatments  2 sampling time points) were then
pooled in equimolar concentrations and subjected to unidirectional
pyrosequencing at Majorbio Bio-pharm Technology Co., Ltd., on a
454 GS-FLX+ instrument (Roche, Switzerland). The sequences have
been deposited in the NCBI Sequence Read Archive (SRA) database
with accession number SRP041099.
2.7. Processing of pyrosequencing data
Raw sequence data generated from pyrosequencing were
processed and analyzed using the Quantitative Insights into
Microbial Ecology (QIIME) software package, version 1.7.0 (Capor-
aso et al., 2010a,b). Briefly, sequences were quality controlled using
the default arguments in the split_libraries.py script with the
exception of increasing the primer mismatch from 0 to 2 and then
assigned to each sample based on unique 10-bp barcodes. After
removing the barcode and primer sequences, the remaining
sequences were clustered into operational taxonomic units (OTUs)
at a level of 97% sequence similarity using a UCLUST-based (Edgar,
2010) two-step subsampled open reference OTU picking workflow
against the Greengenes 13_5 reference database (McDonald et al.,
2012) pre-clustered at 97% identity. The OTUs observed as
Fragment size (bp) Concentration Thermal profile Reference
(nM)
635 200 45 s at 95
C, followed by 40 cycles Francis et al. (2005)
of 5 s at 95
C, 30 s at 52
C and
1 min at 72
C
491 100 30 s at 95
C, followed by 40 cycles Stephen et al. (1998)
of 5 s at 95
C and 34 s at 60
C
361 100 30 s at 95
C, followed by 40 cycles Poly et al. (2001)
of 5 s at 95
C and 34 s at 60
C
4 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12
singletons were discarded from downstream analyses. Represen-
tative sequences (most abundant), which were selected as the
cluster centroid of each OTU, were aligned against the Greengenes
core set using PyNAST (Caporaso et al., 2010a,b). Reads that failed
to align with PyNAST and subsequently identified as chimeras
using Chimera Slayer (Edgar et al., 2011) were eliminated in further
analyses. Taxonomy was assigned to bacterial OTUs by using an
RDP naïve Bayesian rRNA Classifier (Wang et al., 2007) for each
representative sequence against a subset of the Greengenes
13_5 reference database (full database pre-clustered at 97%
sequence identity) with a confidence threshold of 80%. All OTUs
annotated as chloroplasts were removed from the OTU table,
resulting in a set of usable OTUs defined as retained OTUs.
Phylogenetic trees were then built with all representative
sequences of each OTU using the FastTree algorithm (Price et al.,
2010). To equalize the sampling effort, all samples were then
rarefied to 3099 sequences (hereby defined as the rarefied OTUs
table) based on the retained OTUs table. The relative abundance of
a given phylogenetic group was set as the number of sequences
affiliated with that group divided by the total number of sequences
per sample.
Several a-diversity indices were calculated based either on the
rarefied OTU table at a depth of 3099 sequences per sample
(observed OTU richness; the abundance-based coverage estimator,
ACE; Shannon diversity and evenness) or both the rarefied OTU
table and the phylogenetic tree (Faith’s phylogenetic diversity, PD).
To compare between-sample variations in the composition of the
total microbial community, weighted UniFrac distances (Lozupone
and Knight, 2005) were calculated using the QIIME pipeline
(Caporaso et al., 2010a,b).
2.8. Statistical analysis
All statistical analyses were performed using R (version 3.1.1) (R
Development Core Team, 2012) and PASW Statistics 18 (SPSS Inc.).
In all tests, a p value < 0.05 was considered statistically significant.
All data were tested for normality and homogeneity, and the data
were log2(x+c)-transformed when necessary to meet the criteria
for a normal distribution. The constant c was adjusted between
approximately 0.1 and 100 to achieve the best transformation. The
means and standard deviations of the measured variables are
presented. Multiple analyses of variance (MANOVA) using PASW
Statistics 18 (SPSS Inc.) was used to determine the effects of the
fertilizer treatment and sampling time on the dependent variables,
soil physicochemical properties, biological characteristics, a-di-
versity indices, nitrogen functional genes and relative abundance
of abundant taxa. The significant difference between the fertiliza-
tion treatments at each sampling time was tested using one-way
ANOVA followed by Tukey’s post-hoc test. Spearman’s rank
correlations between abundant taxa, a-diversity indices, nitrogen
functional genes, and soil physicochemical properties were
calculated using PASW Statistics 18 (SPSS Inc.).
To compare the microbial community composition across all
samples based on the weighted UniFrac distance matrices,
hierarchical cluster analysis was performed using the “hclust”
function with the average linkage algorithm in the R (R
Development Core Team, 2012) statistical package. To determine
the unique OTUs belonging to each treatment, and shared OTUs
among the different treatments in June, October or both types of
harvest soils, only the OTUs that were present in at least two
biological replicates in each sample were considered technically
reproducible and were retained for further analysis. Therefore,
only the OTUs present in at least four biological replicates were
retained for analyzing the unique OTUs belonging to each fertilizer
treatment in each sampling time, and only eight biological
replicates were retained for analyzing the unique OTUs belonging
to all four fertilizer treatments in each sampling time. For the
shared OTUs in each sampling time, only those present in at least
eight biological replicates were retained.
3. Results
3.1. Crop yields
The two-year (from 2011 to 2012) winter wheat and summer
rice yields obtained in the different treatments are presented in
Table 2. Generally, both wheat and rice yields were significantly
higher in the fertilized treatments than in the unfertilized control,
whereas there were no significant differences between the
fertilized treatments, except for the wheat yield in 2011. The
NPKMOI treatment resulted in the highest yield of rice in 2011
(10.04 t/ha) and wheat in 2012 (5.34 t/ha). Additionally, the wheat
yield in 2011 were highest (5.98 t/ha) in the NPKM treatment
whereas the rice yield in 2012 was highest in the NPK treatment.
3.2. Soil physicochemical characteristics
The selected physicochemical characteristics of the analyzed
soil samples are presented in Table 3. The soil pH, total N (TN),
available N (AN), available P (AP), available K (AK), and NO3-N
contents differed significantly (P < 0.05) between the treatments
in June, whereas only the soil AN and NO3-N contents were
significantly (P < 0.05) different between the treatments in
October. The MANOVA results showed that the soil AN and AP
were significantly (P < 0.01) affected by the different fertilizer
treatments, whereas the organic matter (OM) and TP were
significantly (P < 0.05) affected by sampling time (Table 4). The
soil OM content was higher in the rice harvest soils (21.3 g/kg) than
in June (18.2 g/kg), whereas the soil TP content was higher in
October (1.44 g/kg) than in June (0.92 g/kg) (Table S1). Both
treatments and sampling time had a significant (P < 0.01) effect on
the soil AK. In addition, the treatment, sampling time, and
treatment  time interaction terms significantly (P < 0.001) affect-
ed the soil NO3-N content (Table 4).
Supplementary material related to this article found, in the
online version, at http://dx.doi.org/10.1016/j.apsoil.2015.11.006.
3.3. Microbial biomass and enzymatic activity
The soil MBC varied significantly (P < 0.01) between the
treatments and was the highest in the NPKMOI treatment (June:
635.0 mg/kg; October: 452.6 mg/kg) and the lowest in the CK
treatment (June: 389.3 mg/kg; October: 258.5 mg/kg) in both
sampling times (Tables 3 and 4). Similar to the MBC, the highest
and lowest MBN values were also observed in the NPKMOI
treatment (June: 63.1 mg/kg; October: 42.1 mg/kg) and the CK
treatment (June: 37.6 mg/kg; October: 23.4 mg/kg) in both
Table 2
Two-year wheat and rice yields (t ha1) under different fertilization treatments.
Treatmenta Year: 2011 Year: 2012
Winter wheat Summer rice Winter wheat Summer rice
CK 2.80  0.17 c 6.51  0.24 b 2.99  0.60 b 6.64  0.31 b
NPK 4.83  0.68 b 9.44  0.40 a 4.93  0.56 a 9.83  0.80 a
NPKM 5.98  0.15 a 9.63  0.73 a 5.13  0.40 a 9.27  0.89 a
NPKMOI 5.34  0.56 ab 10.04  0.66 a 5.34  0.37 a 9.00  0.40 a
Values (means  SD, n = 4) within the same column followed by different letters are
significantly different at P < 0.05 according to Tukey’s post-hoc test.
a
Treatment: CK, control without fertilizer; NPK, chemical fertilizers (conven-
tional dosage); NPKM, 50% NPK fertilizer plus 6000 kg/ha pig manure; NPKMOI, 30%
NPK fertilizer plus 3600 kg/ha pig manure organic–inorganic compound fertilizer.
 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12 5

Table 3
Physicochemical and biological characteristics of a Chinese paddy soil under different fertilization treatments.

Physicochemical and Treatmenta and sampling time

biological measurements
June October

CK NPK NPKM NPKMOI CK NPK NPKM NPKMOI

pH (H2O) 7.55  0.05 c 7.66  0.03 b 7.64  0.02 bc 7.80  0.05 a 7.60  0.28 a0 7.80  0.02 a0 7.66  0.19 a0 7.73  0.08 a0
Organic matter (g kg1) 17.6  1.6 a 17.9  1.0 a 18.9  2.7 a 18.4  3.0 a 20.1  4.9 a0 22.1  2.0 a0 21.7  2.2 a0 22.8  3.4 a0
Total N (g kg1) 1.78  0.02 b 1.90  0.07 ab 1.96  0.04 a 2.05  0.09 a 1.92  0.29 a0 2.03  0.29 a0 2.10  0.35 a0 2.07  0.29 a0
Total P (g kg1) 1.40  0.06 a 1.45  0.02 a 1.43  0.07 a 1.46  0.07 a 0.88  0.11 a0 0.95  0.07 a0 0.93  0.04 a0 0.92  0.03 a0
Total K (g kg1) 18.4  1.5 a 20.1  3.3 a 19.8  2.3 a 19.7  2.0 a 18.4  4.6 a0 18.6  4.1 a0 19.8  6.1 a0 20.5  4.3 a0
Available N (mg kg1) 119.8  9.4 b 158.2  11.2 a 146.0  19.2 ab 156.6  10.1 a 112.2  7.6 b0 158.1  13.4 a0 145.4  19.2 a0 b0 154.8  20.6 a0
Available P (mg kg1) 15.9  3.3 b 23.9  1.5 a 21.4  1.3 ab 25.3  3.8 a 15.2  4.1 a0 20.8  8.2 a0 21.3  2.9 a0 24.7  2.1 a0
Available K (mg kg1) 104.3  7.6 b 136.3  18.2 ab 124.7  9.1 ab 147.3  15.6 a 66.7  12.0 a0 70.8  8.8 a0 78.2  13.3 a0 81.2  11.2 a0
NO3-N (mg kg1) 6.4  0.5 c 19.5  0.9 a 11.9  0.9 b 21.5  1.1 a 8.4  0.9 b0 10.5  1.1 a0 b0 10.0  1.7 b0 13.4  0.5 a0
Microbial biomass C 389.3  54.8 b 541.2  45.3 ab 561.4  63.9 ab 635.0  56.3 a 258.5  35.7 b0 306.1  41.7 a0 b0 336.0  57.8 a0 b0 452.6  50.8 a0
(mg kg1)
Microbial biomass N 37.6  2.1 b 51.3  4.6 ab 46.2  5.0 ab 63.1  7.1 a 23.4  2.1 b0 35.3  5.4 a0 b0 33.2  4.2 a0 b0 42.1  3.4 a0
(mg kg1)
Invertase activity 49.2  6.6 bc 29.6  7.1 c 61.2  4.8 ab 74.4  1.2 a 58.9  5.1 a0 32.8  4.8 b0 71.8  5.6 a0 77.6  7.7 a0
(mg g1 d1)
Urease activity (mg g1 d1) 3.8  0.3 c 4.4  0.2 bc 5.4  0.3 ab 5.6  0.3 a 4.4  0.2 c0 4.3  0.0 c0 6.3  0.1 b0 6.9  0.0 a0
Alkaline phosphatase 56.3  0.8 b 47.2  2.9 c 64.0  0.3 ab 71.0  2.7 a 55.7  3.2 b0 69.2  11.5 a0 b0 82.8  4.3 a0 68.7  2.0 a0 b0
activity (mg g1 d1)
Catalase activity 6.1  0.1 b 8.0  0.7 ab 9.4  1.2 ab 11.1  1.7 a 8.3  0.4 b0 8.2  0.0 b0 10.6  0.8 a0 b0 12.7  1.7 a0
(mg g1 20 min1)

Values (means  SD, n = 3) within the same row followed by different letters are significantly different at P < 0.05 according to Tukey’s post-hoc test. The “a” and “a0 ” represent
the ANOVA results for the different fertilizer treatments for the wheat and rice season, respectively.
a
Treatment as described in Table 2.

sampling times (Table 3). Both the soil MBC and MBN varied
significantly (P < 0.001) over time and were higher in June
(531.7 mg/kg and 49.6 mg/kg, respectively) than in October
(338.3 mg/kg and 33.5 mg/kg, respectively) (Table S1). In addition,
there were no significant treatment  time interaction effects
(P > 0.05) on the soil MBC and MBN (Table 4).
The soil invertase, urease, alkaline phosphatase and catalase
activities differed considerably (P < 0.05) between the treatments
and exhibited different trends in both sampling times (Table 3).
The highest enzyme activities were measured in the NPKMOI
treatment, with the exception of alkaline phosphatase activity in
October, which was highest in the NPKM treatment. The lowest
invertase activity was observed in the NPK treatment in both
sampling times. The urease and catalase activities were the lowest
in the CK treatment in both sampling times; however, it was
nonsignificant with that in the NPK treatment in October. In
addition, the alkaline phosphatase activity was lowest in the NPK
treatment in June and in the CK treatment in October. Overall, the
NPKM and NPKMOI treatments increased the soil enzyme activity
compared with the CK and NPK treatments. Both fertilizer
treatment and sampling time significantly (P < 0.001) affected
the soil enzyme activities, and there was a significant (P < 0.001)
treatment  time interaction effect on the urease and alkaline
phosphatase activities (Table 4).
3.4. Quantification of nitrogen-cycling genes
The effect of fertilizer treatments on the abundance of bacterial
nifH, archaeal amoA (target AOA groups), and bacterial amoA
(target AOB groups) genes in both sampling times were

Table 4
Overall effects of fertilization treatment and sampling time on the soil physicochemical, biological, and microbial characteristics, and the nitrogen functional gene abundance
analyzed using two-way ANOVA.

Treatment (df = 3) Sampling time (df = 1) Treatment  time (df = 3)

MS F MS F MS F
pH (H2O) 0.042 2.6 0.007 0.431 0.012 0.714
Organic matter 3.4 0.415 56.2 6.9* 3.7 0.449
Total N 0.051 1.0 0.068 1.4 0.005 0.094
Total P 0.004 0.977 1.6 392.6*** 0.001 0.144
Total K 3.4 0.235 0.205 0.014 1.4 0.094
Available N 2250.2 10.5*** 37.7 0.176 18.2 0.085
Available P 95.7 6.1** 7.6 0.483 2.7 0.174
Available K 845.0 5.5** 17469.0 112.8*** 301.6 1.9
NO3-N 118.4 115.4*** 108.5 105.7*** 41.2 40.1***
Microbial biomass C 32666.3 12.3** 149654.9 56.4*** 2264.1 0.853
Microbial biomass N 333.2 16.2*** 1033.6 50.2*** 12.4 0.605
Invertase activity 2249.2 139.3*** 267.3 16.6*** 24.5 1.5
Urease activity 6.9 316.1*** 2.8 129.0*** 0.506 23.4***
Alkaline phosphatase activity 438.2 38.7*** 536.5 47.3*** 240.4 21.2***
Catalase activity 26.2 49.8*** 10.1 19.2*** 0.988 1.9
Archaeal amoA 0.083 14.4*** 0.673 117.0*** 0.002 0.318
Bacterial amoA 0.545 67.0*** 0.479 58.8*** 0.019 2.3
nifH 0.075 13.8*** 0.866 159.1*** 0.013 2.3

df: degrees of freedom; MS: mean square; F: variance ratio.

Values in bold indicate statistically significant, *P < 0.05; **P < 0.01; ***P < 0.001.
6 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12

determined using real-time PCR (Fig. 1). Both fertilizer treatment compared to the CK and NPK treatments; however, in October,
and sampling time significantly (P < 0.001) affected the nifH, AOA, there was no significant difference among the NPK, NPKM, and
and AOB amoA gene copy numbers (Table 4). The nifH gene copy NPKMOI treatments (Fig. 1A). The AOA amoA copy numbers were
numbers were higher in the NPKM and NPKMOI treatments in June highest in the NPKMOI treatment (7.01 107 and 3.36  107 copies

Fig. 1. Abundance of the nifH gene (A), archaeal amoA gene (B), and bacterial amoA gene (C) in a paddy soil under different fertilization treatments in June and October. Error
bars indicate the standard deviations of the means of twelve replicates (three replicates per plot). Different letters at each sampling time indicate significant differences at
P < 0.05 according to Tukey’s post-hoc test. The fertilizer regime abbreviations are defined in Table 2.
 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12 7

Fig. 2. Relative abundance of the dominant bacterial phyla (proteobacterial classes) in a paddy soil under different fertilization treatments in June and October. Error bars
indicate the standard deviations of the means of three replicates. Different letters at each sampling time indicate significant differences at P < 0.05 according to Tukey’s post-
hoc test. The fertilizer regime abbreviations are defined in Table 2.

per gram of dry soil) and lowest in the CK treatment (4.10  107 and
1.70  107 copies per gram of dry soil) in the respective wheat and
the rice harvest soils (Fig. 1B). Similar to AOA, the NPKMOI
treatment had the highest abundance of AOB, and the CK
treatment had the lowest abundance in both sampling times
(Fig. 1C). No significant differences in AOB population size were
observed between the NPK and NPKM treatments in either
sampling time. Additionally, the abundance of the nifH, AOA,
and AOB amoA genes decreased from June to October.
3.5. Bacterial community composition
From the two crop seasons sampled, a total of 95,661 high-
quality 16S rRNA sequences were generated from the 24 soil
samples (2 sampling times  4 fertilizer treatments  3 replicate
plots) in this study (sequences ranged from 3099-5386). After
equalizing the sampling effort, 74,376 sequences were retained for
the analysis. These high-quality sequences were clustered into
11,862 OTUs at 97% sequence similarity, with 1381–1673 OTUs per
sample. The dominant phyla (proteobacterial classes) across all
samples were Alphaproteobacteria, Betaproteobacteria, Deltapro-
teobacteria, Gammaproteobacteria, Chloroflexi, Acidobacteria, Gem-
matimonadetes, Nitrospirae, Bacteroidetes, Actinobacteria,
Planctomycetes, Chlorobi, and WS3 (>1%), accounting for more
than 86% of the bacterial sequences from each soil sample (Fig. 2
and Table S2). In the wheat harvest soils, the phyla (proteobacterial
classes) Chloroflexi, Acidobacteria, Gemmatimonadetes, Nitrospirae,
Bacteroidetes, Actinobacteria, Alphaproteobacteria, and the unclas-
sified bacteria varied significantly (P < 0.05) between the different
treatments (Fig. 2). However, only the phylum Acidobacteria
changed remarkably (P < 0.05) between the different treatments in
the rice harvest soils (Fig. 2). Sampling time had a significant
(P < 0.05) effect on Proteobacteria, Gammaproteobacteria, Deltap-
roteobacteria, Chloroflexi, Chlorobi, and WS3, while Nitrospirae and
Actinobacteria were significantly (P < 0.05) affected by the fertilizer
treatment (Table S3). Moreover, Alphaproteobacteria and Betapro-
teobacteria were considerably (P < 0.05) affected by the fertilizer
treatment and sampling time (Table S3). Acidobacteria and
Gemmatimonadetes were significantly (P < 0.05) influenced by
the fertilizer treatment, sampling time, and their interactions

Fig. 3. Venn diagram of the number of shared and unique OTUs in the different fertilization treatments in June (A), October (B) and both sampling times (C). Only the OTUs
present in two biological replicates of each sample were retained. The fertilizer regime abbreviations are defined in Table 2.
8 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12

(Table S3). Additionally, no significant differences in Bacteroidetes
and Planctomycetes were observed between the different fertilizer
treatments, sampling times, and their interaction terms (Table S3).
Supplementary material related to this article found, in the
online version, at http://dx.doi.org/10.1016/j.apsoil.2015.11.006.
To compare the similarity and dissimilarity in the bacterial
community composition between all samples in June, October,
and both sampling times, Venn diagrams with unique and shared
OTUs were generated (Fig. 3). The number of OTUs unique to CK,
NPK, NPKM, and NPKMOI in the wheat season was 217, 232, 258,
and 302, respectively, accounting for 54.7% of the total observed
OTUs (1845) (Fig. 3A). The majority (63.9–71.2%) of the unique
OTUs in all samples were Alphaproteobacteria, Betaproteobacteria,
Deltaproteobacteria, Chloroflexi, Acidobacteria and Bacteroidetes
(Fig. 4). The relative abundance of Alphaproteobacteria, Gammap-
roteobacteria, Acidobacteria, Gemmatimonadetes, Bacteroidetes, and
Planctomycetes was significantly (P < 0.05) different among the
different treatments; the NPKMOI treatment had the highest
abundance of Gammaproteobacteria, Gemmatimonadetes, and
Bacteroidetes and the lowest abundance of Acidobacteria. In
addition, 253 OTUs were shared among the treatments in June
and accounted for 13.7% of the total observed OTUs. The
majority (83.3–87.6%) of the shared OTUs in all samples
was Alphaproteobacteria, Betaproteobacteria, Deltaproteobacteria,
Gammaproteobacteria, Chloroflexi, Acidobacteria and Gemmatimo-
nadetes, but Chloroflexi and Gemmatimonadetes varied markedly
(P < 0.05) among the treatments (Fig. S1). Similar to the results
from the wheat harvest soils, the percentage of unique and shared
OTUs in the rice harvest soils accounted for 56.1% and 15.3%,
respectively, of the total observed OTUs (Fig. 3B). The dominant
phyla (classes) in the unique and shared OTUs of October were
similar to June (Fig. 4 and Fig. S1). The relative abundance of
Deltaproteobacteria, Gammaproteobacteria, Chloroflexi, Actinobac-
teria, WS3 and others was notably (P < 0.05) changed in the unique
OTUs, whereas only the abundance of Acidobacteria and others
shifted significantly (P < 0.05) in the shared OTUs among the
different treatments (Fig. 4 and Fig. S1). To explore the overall
influences of fertilization practices on the unique and shared OTUs,
we combined the data from both sampling times. Surprisingly, all
of the abundant taxa in the unique OTUs were significantly
(P < 0.05) different among the treatments, indicating that different
fertilization practices had different capacities to alter the soil
microbiomes (Fig. 4). On the contrary, no significant difference
was observed for the abundant taxa in the shared OTUs in different
treatments (Fig. S1), suggesting that a core microbiome sustained
throughout the different treatments.
Supplementary material related to this article found, in the
online version, at http://dx.doi.org/10.1016/j.apsoil.2015.11.006.

Fig. 4. Relative abundance (A) and the summarized statistically significant differences (B) in the dominant bacterial phyla (proteobacterial classes) in the unique OTUs of each
fertilization treatment in June, October, and both sampling times. Error bars indicate the standard deviations of the means of three replicates. Different letters at each sampling
time indicate significant differences at P < 0.05 according to Tukey’s post-hoc test. “ND” in (B) indicates not detected. The fertilizer regime abbreviations are defined in Table 2.
 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12
3.6. Bacterial community structure and a-diversity
Hierarchical cluster analysis of the weighted UniFrac distance
confirmed that the bacterial community structure was distinct
between June and October, indicating that the sampling time is the
major determinant of community structure (Fig. 5). In addition, the
NPKMOI treatment was grouped distinctly from the NPK, NPKM
and CK treatments in June, while the NPK and NPKM treatments
were grouped together, but separated from the CK treatment. In
contrast, there was no distinct separation among different
treatments in October.
The bacterial a-diversity in the individual samples under the
different fertilizer regimes of both sampling times was calculated
based on 3099 sequences (Table S4). Statistically significant
differences (P < 0.05) were only observed for OTU richness and
the ACE in June; the NPKMOI treatment had the highest OTU
richness and ACE. However, all of the a-diversity indices except for
the Shannon evenness varied considerably (P < 0.01) with sam-
pling time (Table S5). The OTU richness, ACE, and Faith's PD were
significantly (P < 0.001) higher in October, indicating that the
samples from this sampling time harbored a richer and more
diverse community than those from June (Table S5). In contrast,
the Shannon evenness was higher in June, indicating that the soil
samples from this sampling time harbored a more equitable
community compared with those from October (Table S4).
Supplementary material related to this article found, in the
online version, at http://dx.doi.org/10.1016/j.apsoil.2015.11.006.
3.7. Correlations between soil characteristics and abundant phyla,
alpha indices and nitrogen functional genes
Spearman’s rank correlation coefficient was used to evaluate
the relationships between the soil properties and the relative
abundance of the phyla (proteobacterial classes), the alpha indices,
and the abundance of the nitrogen functional genes. Most of the
abundant taxa, alpha indices, and the abundance of the functional
genes were positively or negatively correlated with soil TP and AK
(Table 5). The relative abundance of Proteobacteria was positively
correlated with soil OM and negatively correlated with soil TP and
AK. The subgroups of Proteobacteria exhibited different correlation
Fig. 5. Hierarchical cluster dendrogram of the bacterial communities derived from four fertilizer treatments in June (circle) and October (square). Between-sample
similarities were estimated from 3099 sequences per sample using the phylogeny-based UniFrac distance metric. The fertilizer regime abbreviations are defined in Table 2.
9
patterns with the soil properties. Alphaproteobacteria was posi-
tively correlated with soil AN, AP, AK, and NO3-N, whereas
Betaproteobacteria was negatively correlated with soil TP, AP, AK,
and NO3-N. Deltaproteobacteria had a positive correlation with soil
OM and a negative correlation with soil TP, AK, and NO3-N, which
was similar to the observations for the phylum Proteobacteria.
Gammaproteobacteria showed a positive correlation with soil TP
and AK. Additionally, only the phylum Nitrospirae was negatively
correlated with soil pH, and the Planctomycetes phylum was
positively correlated with soil TK. The OTU richness and ACE
exhibited a positive correlation with soil pH and OM, but they were
negatively correlated with soil TP and AK. Faith's PD was negatively
correlated with soil TP and AK, whereas evenness was positively
correlated with soil TP and AK. The abundance of archaeal amoA
and nifH was positively correlated with soil TP, AK, and NO3-N,
while the abundance of nifH was also negatively correlated with
soil OM. There was a positive relationship between the abundance
of bacterial amoA and the soil TP, AN, AP, AK, and NO3-N.
4. Discussion
In the present study, temporal yield variations were observed in
response to the different fertilizer regimes (Table 2), and the
NPKMOI treatment resulted in the highest yield in the 2011 rice
season and the 2012 wheat season. To some extent, the application
of pig manure organic–inorganic compound fertilizer with
reduced chemical fertilizer is capable of enhancing crop yields.
The soil pH changed slightly with different fertilizer regimes in
both sampling times, which is consistent with the results of Zhao
et al. (2014b). Zhong and Cai (2007) and Zhao et al. (2014b)
observed that soil available nutrients (AN, AP, AK, and NO3-N) were
notably altered by fertilization in comparison to CK, which is in
accordance with the present study results. We also observed an
increase, albeit not significant, in soil OM, TN (except for the wheat
season), TP, and TK in response to fertilization in both sampling
times, indicating that these soil properties were gradually, but not
dramatically, changed because of fertilizer application. Generally,
the maximum values of the soil physicochemical properties were
observed in the NPKMOI treatment.
10 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12

Table 5
Spearman’s rank correlation coefficient between soil characteristics and abundant phyla, alpha indices and nitrogen functional genes.

pH OM TN TP TK AN AP AK NO3-N
Proteobacteria 0.27 0.56** 0.03 0.77** 0.15 0.10 0.16 0.59** 0.16
Alphaproteobacteria 0.07 0.10 0.03 0.39 0.08 0.51* 0.49* 0.61** 0.77**
Betaproteobacteria 0.10 0.26 0.04 0.60** 0.12 0.15 0.43* 0.67** 0.41*
Deltaproteobacteria 0.23 0.60** 0.01 0.69** 0.15 0.15 0.29 0.69** 0.46*
Gammaproteobacteria 0.02 0.19 0.08 0.52* 0.13 0.32 0.25 0.60** 0.40
Chloroflexi 0.16 0.02 0.31 0.42* 0.39 0.13 0.18 0.42* 0.03
Acidobacteria 0.37 0.46* 0.24 0.70** 0.01 0.14 0.22 0.61** 0.07
Gemmatimonadetes 0.16 0.44* 0.18 0.80** 0.11 0.17 0.19 0.75** 0.38
Nitrospirae 0.49* 0.41* 0.13 0.10 0.06 0.15 0.07 0.13 0.04
Bacteroidetes 0.07 0.28 0.21 0.20 0.12 0.08 0.10 0.21 0.12
Actinobacteria 0.32 0.19 0.04 0.00 0.08 0.32 0.43* 0.02 0.51*
Planctomycetes 0.08 0.10 0.14 0.23 0.41* 0.08 0.23 0.23 0.29
Chlorobi 0.17 0.34 0.10 0.64** 0.32 0.21 0.40 0.68** 0.48*
WS3 0.14 0.50* 0.25 0.42* 0.35 0.03 0.19 0.56** 0.27
OTU richness 0.45* 0.51* 0.24 0.63** 0.20 0.04 0.13 0.62** 0.06
ACE 0.45* 0.50* 0.24 0.68** 0.24 0.04 0.04 0.69** 0.09
Shannon 0.31 0.31 0.13 0.01 0.18 0.14 0.29 0.02 0.13
Faith’s PD 0.38 0.39 0.21 0.60** 0.09 0.09 0.07 0.66** 0.16
Shannon evenness 0.22 0.37 0.10 0.81** 0.00 0.07 0.19 0.70** 0.28
Archaeal amoA 0.08 0.29 0.02 0.81** 0.39 0.25 0.32 0.90** 0.51*
Bacterial amoA 0.25 0.05 0.24 0.50* 0.20 0.61** 0.66** 0.67** 0.87**
nifH 0.09 0.44* 0.08 0.78** 0.24 0.21 0.37 0.86** 0.53**

Values in bold indicate statistically significant, *P < 0.05, **P < 0.01.

The soil MBC and MBN were higher in June than in October,
which is in line with the results reported by Zhao et al. (2014b).
This result might be attributable to differences in soil temperature,
soil moisture, and the oxygen gradient between June and October
(Díaz-Raviña et al., 1995; Gu et al., 2009; Noll et al., 2005). Organic
manure contains vast quantities of readily utilizable energy
sources, and inorganic fertilizer is rich in available nutrients,
which stimulates the growth of soil microorganisms and results in
higher soil microbial biomass (Kandeler et al., 1999). In the present
study, the NPKMOI treatment had the greatest capacity to increase
microbial biomass. Soil enzymatic activity is widely used as an
indicator of soil quality and relates to soil nutrient transformation
(Dick, 1994; Yang et al., 2008). The enzymes selected in the current
study, i.e., invertase, urease, alkaline phosphatase, and catalase, are
related to soil C, N, and P transformations (Dick, 1994). Moreover,
Acosta-Martinez and Harmel (2006) reported that the soil
microbial biomass was positively correlated with soil enzymatic
activities. Higher enzymatic activities were observed in the
treatments with the organic manure, i.e., NPKM and NPKMOI,
which is coherent with the findings reported by Ge et al. (2010),
indicating that these treatments can strengthen the soil nutrient
(e.g., C, N, and P) cycling.
The detection of microbial functional genes involved in nitrogen
cycling using real-time PCR provides useful information on soil
nitrogen transformation (Hayden et al., 2010). In the present study,
two key nitrogen cycling genes, nifH and amoA, were quantified to
determine the potential nitrogen fixation and ammonia oxidation
in response to different fertilizer regimes. Considerable spatial and
temporal shifts in the abundance of the nifH, AOA, and AOB amoA
genes were observed in the current study, which is similar to the
results reported by He et al. (2007) and Orr et al. (2012). The
abundance of the nifH genes was significantly (P < 0.05) higher in
the NPKM and NPKMOI treatments in June, whereas it was
comparable among the NPK, NPKM, and NPKMOI treatments in
October. Orr et al. (2012) also observed that the abundance of the
nifH gene was higher under organic compared with conventional
management. He et al. (2007) found that the population sizes of
AOA and AOB were higher in the NPK+OM treatment in both winter
and summer. In our study, we also found that the NPKMOI
treatment had a higher abundance of AOA and AOB compared with
the CK and NPK treatments in both sampling times, whereas the
population sizes of AOA and AOB were comparable in the NPKM
and NPK treatments in both sampling times (Fig. 1). Previous
studies showed that the abundance of AOA and AOB was positively
and significantly correlated with soil pH in acidic soils (He et al.,
2007), and the abundance of AOB had a significant negative
relationship with soil pH in alkaline soils (Shen et al., 2008). No
significant correlation was observed between the soil pH and the
population size of AOA and AOB, which might be due to the
different soil types and the mild pH fluctuations (a range of 7.27 to
7.84 for all samples) in our study. However, both the AOA and AOB
amoA copy numbers were positively and significantly (P < 0.05)
correlated with soil NO3-N, indicating that an increase in soil NO3-
N might result in an increase in AOA and AOB abundance.
Moreover, the AOB population size also had a significant positive
correlation with soil AN, suggesting that AOB played a major role in
ammonia oxidation in the current study. The NPKMOI treatment
might activate soil microbes related to nitrogen cycling and lead to
a high potential for nitrogen fixation and ammonia oxidation.
The pyrosequencing analyses revealed that Proteobacteria,
Chloroflexi, and Acidobacteria were the most abundant phyla in
all soil samples in this study. This result is in agreement with
several previous studies, demonstrating that Proteobacteria,
Chloroflexi, and Acidobacteria are the dominant soil bacterial taxa
based on 16S rRNA gene clones or pyrosequencing (Janssen, 2006;
Zhao et al., 2014a). Significant differences in the abundance of
phyla (classes) were observed for Chloroflexi, Acidobacteria,
Gemmatimonadetes, Nitrospirae, Bacteroidetes, Actinobacteria,
Alphaproteobacteria, and unclassified bacteria among all treat-
ments in June, whereas only Acidobacteria differed notably among
different treatments in October, suggesting that the practice of
periodic flooding resulted in an anaerobic environment and thus
normalized the soil microbiome in October. However, significant
differences were also observed between different sampling times,
and there were significant fertilizer  time interaction effects. In
particular, the relative abundance of Betaproteobacteria and
Deltaproteobacteria was higher in October while that of the
Alphaproteobacteria, Gammaproteobacteria, Acidobacteria and Gem-
matimonadetes was higher in June. Many studies have shown that
the soil pH is the strongest factor determining the soil microbial
community structure (Lauber et al., 2009; Rousk et al., 2010; Xiong
et al., 2012). In our study, only the phylum Nitrospirae was
 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12
negatively correlated with soil pH in our study, suggesting that the
mild pH range in the field scale does not provide insight into the
relationship between soil pH and soil bacterial communities. This
result was similar to the findings reported by Zhao et al. (2014b)
who demonstrated that the soil OM, AK, TN, and TP could drive
bacterial communities. In the present study, we found that the soil
TP and AK were correlated with the most abundant phyla, which
also might be a good explanation for the difference in the relative
abundance between June and October.
Furthermore, we analyzed the unique and shared OTUs present
in CK, NPK, NPKM, and NPKMOI treatments in June, October, and
both sampling times. Of the total observed OTUs, 13.7%, 14.5%, and
22.7% were shared in the June, October, and both sampling times,
respectively (Fig. 3). This observation indicated that there might be
a core microbiome that is consistent and stable across different
treatments and sampling times, which was verified by analyzing
the relative abundance of the abundant taxa in the shared OTUs
(Fig. S1). Not surprisingly, the relative abundance of the abundant
taxa in the shared OTUs remained constant over the two sampling
times, predominated by Alphaproteobacteria, Betaproteobacteria,
Deltaproteobacteria, Chloroflexi, Acidobacteria, and Gemmatimona-
detes. In contrast, the relative abundance of the abundant taxa in
the unique OTUs belonging to each treatment in June, October, and
both sampling times showed a different pattern, i.e., the abundant
taxa differed significantly between the different treatments in the
two sampling times. Specifically, the relative abundance of
Alphaproteobacteria, Gammaproteobacteria, Nitrospirae, Bacteroi-
detes, and Actinobacteria was higher in the NPKM and NPKMOI
treatments in the two sampling times, while the relative
abundance of Acidobacteria and Gemmatimonadetes was increased
in the NPK treatment (Fig. 4). Members classified as Alphaproteo-
bacteria, Gammaproteobacteria, and Nitrospirae are known to
participate in nitrification (Baker et al., 2013), which corresponds
with the qPCR results of the AOA and AOB in the current study.
Actinobacteria are a group of bacteria involved in organic matter
turnover and which produce a vast array of antimicrobial and
xenobiotic compounds (Schrijver and Mot, 1999; Basilio et al.,
2003). Jones et al. (2009) reported that the abundance of
Acidobacteria was negatively correlated with soil pH; in our study
the enrichment of this phylum did not coincide with a reduction in
soil pH, suggesting that this taxa might be a predictor for the soil
pH variation.
Soil biodiversity is considered to be critical for the integrity,
function and sustainability of soil ecosystems (Kennedy and Smith,
1995). Zhao et al. (2014b) found that the microbial community
diversity and richness, to a certain extent, were higher in the
NPKMOI treatment compared with other fertilizer regimes,
although these responses were not significant. These findings
are consistent with the results of our study, i.e., the community
diversity and richness were higher in the NPKMOI treatment
relative to other treatments. This treatment might lead to a more
stable agroecosystem and contribute to sustainable crop produc-
tion. In addition, the richness (OTU richness and ACE) and diversity
(Shannon and Faith’s PD) were positively correlated with the soil
pH and OM although the correlations between the diversity, pH
and OM were not significant. These results demonstrate that
organic fertilizers can enhance soil bacterial richness and diversity
by increasing soil pH and OM contents in the soils of this study.
In the present study, temporal factors predominantly influenced
the soil biomass, enzymatic activities, nitrogen functional genes and
bacterial community (Table. 4 and Fig. 5). These results are consistent
with several previous studies (Hannula et al., 2012; Kennedy et al.,
2006; Zhao et al., 2014b). In addition, the NPKMOI treatment in June
was grouped separately from the other treatments, indicating that
this treatment was the utmost factor in shaping the bacterial
community structure in the soils of wheat harvest.
11
5. Conclusion
To the best of our knowledge, this is the first study to
comprehensively evaluate the effects of pig manure organic–
inorganic compound fertilizer with reduced chemical fertilizer on
crop yields, soil edaphic properties, biological attributes (microbial
biomass and enzyme activity), biochemical traits (microbial
nitrogen functional genes), and the bacterial community structure
in a rice–wheat cropping system under the field conditions. The
NPKMOI treatment was able to increase crop yields, soil nutrient
availability, microbial biomass, enzymatic activity, nitrogen
processes, community richness and diversity. Therefore, the
NPKMOI treatment might be a sound fertilization practice for
the sustainable food production in the future.
Acknowledgments
This research was financially supported by the National Key
Basic Research Program of China (2015CB150502 and
2011CB100503), the Priority Academic Program Development
(PAPD) of Jiangsu Higher Education Institutions and the 111 Project
(B12009). We would like to thank the graduate students from
Nanjing Agricultural University, for their help in managing the field
experiment and collecting field data. We also would like to thank
the anonymous referees for their constructive comments, which
significantly improve the manuscript.
References
Acosta-Martinez, V., Harmel, R.D., 2006. Soil microbial communities and enzyme
activities under various poultry litter application rates. J. Environ. Qual. 35,
1309–1318.
Baker, B.J., Sheik, C.S., Taylor, C.A., Jain, S., Bhasi, A., Cavalcoli, J.D., Dick, G.J., 2013.
Community transcriptomic assembly reveals microbes that contribute to deep-
sea carbon and nitrogen cycling. ISME J. 7, 1962–1973.
Basilio, A., Gonzalez, I., Vicente, M.F., Gorrochategui, J., Cabello, A., Gonzalez, A.,
Genilloud, O., 2003. Patterns of antimicrobial activities from soil actinomycetes
isolated under different conditions of pH and salinity. J. Appl. Microbiol. 95,
814–823.
Bronick, C.J., Lal, R., 2005. Soil structure and management: a review. Geoderma 124,
3–22.
Caporaso, J.G., Bittinger, K., Bushman, F.D., DeSantis, T.Z., Andersen, G.L., Knight, R.,
2010a. PyNAST: a flexible tool for aligning sequences to a template alignment.
Bioinformatics 26, 266–267.
Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K.,
Fierer, N., Pena, A.G., Goodrich, J.K., Gordon, J.I., 2010b. QIIME allows analysis of
high-throughput community sequencing data. Nat. Methods 7, 335–336.
Díaz-Raviña, M., Acea, M.J., Carballas, T., 1995. Seasonal changes in microbial
biomass and nutrient flush in forest soils. Biol. Fertil. Soils 19, 220–226.
Dick, R.P., 1994. Soil enzyme activities as indicators of soil quality. In: Doran, J.W.,
Coleman, D.C., Bezdicek, D.F., Stewart, B.A. (Eds.), Defining Soil Quality for a
Sustainable Environment. Soil Science Society of America.
Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST.
Bioinformatics 26, 2460–2461.
Edgar, R.C., Haas, B.J., Clemente, J.C., Quince, C., Knight, R., 2011. UCHIME improves
sensitivity and speed of chimera detection. Bioinformatics 27, 2194–2200.
Francis, C.A., Roberts, K.J., Beman, J.M., Santoro, A.E., Oakley, B.B., 2005. Ubiquity and
diversity of ammonia-oxidizing archaea in water columns and sediments of the
ocean. Proc. Natl. Acad. Sci. 102, 14683–14688.
Ge, G., Li, Z., Fan, F., Chu, G., Hou, Z., Liang, Y., 2010. Soil biological activity and their
seasonal variations in response to long-term application of organic and
inorganic fertilizers. Plant Soil 326, 31–44.
Germaine, K.J., Chhabra, S., Song, B., Brazil, D., Dowling, D.N., 2010. Microbes and
sustainable production of biofuel crops: a nitrogen perspective. Biofuels 1,
877–888.
Gu, Y., Zhang, X., Tu, S., Lindström, K., 2009. Soil microbial biomass, crop yields, and
bacterial community structure as affected by long-term fertilizer treatments
under wheat–rice cropping. Eur. J. Soil Biol. 45, 239–246.
Guan, S., Zhang, D., Zhang, Z., 1986. Soil Enzymes and Its Methodology. Agriculture
Press, Beijing (in Chinese).
Guo, J., Liu, X., Zhang, Y., Shen, J., Han, W., Zhang, W., Christie, P., Goulding, K.W.T.,
Vitousek, P.M., Zhang, F., 2010. Significant acidification in major Chinese
croplands. Science 327, 1008–1010.
Gupta, R.K., Naresh, R.K., Hobbs, P.R., Zheng, J., Ladha, J.K., et al., 2003. Sustainability
of post-green revolution agriculture: the rice–wheat cropping systems of the
Indo-Gangetic Plains and China. In: Ladha, J.K. (Ed.), Improving the Productivity
12 J. Zhao et al. / Applied Soil Ecology 99 (2016) 1–12
and Sustainability of Rice–Wheat Systems: Issues and Impacts. ASA, Madison,
WI, pp. 1–25.
Hannula, S.E., De, B.W., Van, V.J., 2012. A 3-year study reveals that plant growth
stage, season and field site affect soil fungal communities while cultivar and
GM-trait have minor effects. PLoS One 7, e33819.
Hayden, H.L., Drake, J., Imhof, M., Oxley, A.P.A., Norng, S., Mele, P.M., 2010. The
abundance of nitrogen cycle genes amoA and nifH depends on land-uses and
soil types in South-Eastern Australia. Soil Biol. Biochem. 42, 1774–1783.
He, J., Shen, J., Zhang, L., Zhu, Y., Zheng, Y., Xu, M., Di, H., 2007. Quantitative analyses
of the abundance and composition of ammonia-oxidizing bacteria and
ammonia-oxidizing archaea of a Chinese upland red soil under long-term
fertilization practices. Environ. Microbiol. 9, 2364–2374.
Hsu, S.F., Buckley, D.H., 2009. Evidence for the functional significance of diazotroph
community structure in soil. ISME J. 3, 124–136.
Insam, H., Gómez-Brandón, M., Ascher, J., 2015. Manure-based biogas fermentation
residues—friend or foe of soil fertility? Soil Biol. Biochem. 84, 1–14.
Jannoura, R., Joergensen, R.G., Bruns, C., 2014. Organic fertilizer effects on growth, crop
yield, and soil microbial biomass indices in sole and intercropped peas and oats
under organic farming conditions. Eur. J. Agron. Part B 52, 259–270.
Janssen, P.H., 2006. Identifying the dominant soil bacterial taxa in libraries of 16S
rRNA and 16S rRNA genes. Appl. Environ. Microbiol. 72, 1719–1728.
Joergensen, R.G., 1996. The fumigation–extraction method to estimate soil
microbial biomass: calibration of the kEC value. Soil Biol. Biochem. 28, 25–31.
Joergensen, R.G., Mueller, T., 1996. The fumigation–extraction method to estimate
soil microbial biomass: calibration of the kEN value. Soil Biol. Biochem. 28, 33–
37.
Jones, R.T., Robeson, M.S., Lauber, C.L., Hamady, M., Knight, R., Fierer, N., 2009. A
comprehensive survey of soil acidobacterial diversity using pyrosequencing and
clone library analyses. ISME J. 3, 442–453.
Ju, X., Xing, G., Chen, X., Zhang, S., Zhang, L., Liu, X., Cui, Z., Yin, B., Christie, P., Zhu, Z.,
Zhang, F., 2009. Reducing environmental risk by improving N management in
intensive Chinese agricultural systems. Proc. Natl. Acad. Sci. 106, 3041–3046.
Kandeler, E., Stemmer, M., Klimanek, E.M., 1999. Response of soil microbial biomass,
urease and xylanase within particle size fractions to long-term soil
management. Soil Biol. Biochem. 31, 261–273.
Kennedy, A.C., Smith, K.L., 1995. Soil microbial diversity and the sustainability of
agricultural soils. Plant Soil 170, 75–86.
Kennedy, N., Brodie, E., Connolly, J., Clipson, N., 2006. Seasonal influences on fungal
community structure in unimproved and improved upland grassland soils. Can.
J. Microbiol. 52, 689–694.
Khaliq, A., Abbasi, M.K., Hussain, T., 2006. Effects of integrated use of organic and
inorganic nutrient sources with effective microorganisms (EM) on seed cotton
yield in Pakistan. Bioresour. Technol. 97, 967–972.
Ladha, J.K., Pathak, H., Tirol-Padre, A., Dawe, D., Gupta, R.K., et al., 2003. Productivity
trends in intensive rice–wheat cropping systems in Asia. In: Ladha, J.K. (Ed.),
Improving the Productivity and Sustainability of Rice–Wheat Systems: Issues
and Impacts. ASA, CSSA, and SSSA, Madison, WI, pp. 45–76.
Ladha, J.K., Khind, C.S., Gupta, R.K., Meelu, O.P., Pasuquin, E., 2004. Long-term effects
of organic inputs on yield and soil fertility in the rice–wheat rotation. Soil Sci.
Soc. Am. J. 68, 845–853.
Lauber, C.L., Hamady, M., Knight, R., Fierer, N., 2009. Pyrosequencing-based
assessment of soil pH as a predictor of soil bacterial community structure at the
continental scale. Appl. Environ. Microbiol. 75, 5111–5120.
LeBauer, D.S., Treseder, K.K., 2008. Nitrogen limitation of net primary productivity in
terrestrial ecosystems is globally distributed. Ecology 89, 371–379.
Li, R., Khafipour, E., Krause, D.O., Entz, M.H., de Kievit, T.R., Fernando, W.G.D., 2012.
Pyrosequencing reveals the influence of organic and conventional farming
systems on bacterial communities. PLoS One 7, e51897.
Liu, M., Hu, F., Chen, X., Huang, Q., Jiao, J., Zhang, B., Li, H., 2009. Organic
amendments with reduced chemical fertilizer promote soil microbial
development and nutrient availability in a subtropical paddy field: the
influence of quantity, type and application time of organic amendments. Appl.
Soil Ecol. 42, 166–175.
Lozupone, C., Knight, R., 2005. UniFrac: a new phylogenetic method for comparing
microbial communities. Appl. Environ. Microbiol. 71, 8228–8235.
Mandal, U.K., Singh, G., Victor, U.S., Sharma, K.L., 2003. Green manuring: its effect on
soil properties and crop growth under rice–wheat cropping system. Eur. J.
Agron. 19, 225–237.
Masto, R.E., Chhonkar, P.K., Singh, D., Patra, A.K., 2006. Changes in soil biological and
biochemical characteristics in a long-term field trial on a sub-tropical inceptisol.
Soil Biol. Biochem. 38, 1577–1582.
McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A.,
Andersen, G.L., Knight, R., Hugenholtz, P., 2012. An improved Greengenes
taxonomy with explicit ranks for ecological and evolutionary analyses of
bacteria and archaea. ISME J. 6, 610–618.
National Bureau of Statistics of China, 2009. Irrigated area, consumption of chemical
fertilizers and rural hydropower stations and electricity consumption in rural
areasChina Statistical Yearbook 2009. China Stat. Press, Beijing. . (Chapter 12.6)
http://www.stats.gov.cn/tjsj/ndsj/2009/indexeh.htm.
Ni, B., Liu, M., Lu, S., Xie, L., Wang, Y., 2010. Multifunctional slow-release organic–
inorganic compound fertilizer. J. Agric. Food Chem. 58, 12373–12378.
Noll, M., Matthies, D., Frenzel, P., Derakshani, M., Liesack, W., 2005. Succession of
bacterial community structure and diversity in a paddy soil oxygen gradient.
Environ. Microbiol. 7, 382–395.
Orr, C.H., Leifert, C., Cummings, S.P., Cooper, J.M., 2012. Impacts of organic and
conventional crop management on diversity and activity of free-living nitrogen
fixing bacteria and total bacteria are subsidiary to temporal effects. PLoS One 7,
e52891.
Pan, G., Zhou, P., Li, Z., Smith, P., Li, L., Qiu, D., Zhang, X., Xu, X., Shen, S., Chen, X.,
2009. Combined inorganic/organic fertilization enhances N efficiency and
increases rice productivity through organic carbon accumulation in a rice paddy
from the Tai Lake region, China. Agric. Ecosyst. Environ. 131, 274–280.
Plaza, C., Hernandez, D., Garcia-Gil, J.C., Polo, A., 2004. Microbial activity in pig slurry-
amended soils under semiarid conditions. Soil Biol. Biochem. 36, 1577–1585.
Poly, F., Ranjard, L., Nazaret, S., Gourbière, F., Monrozier, L.J., 2001. Comparison of
nifH gene pools in soils and soil microenvironments with contrasting
properties. Appl. Environ. Microbiol. 67, 2255–2262.
Price, M.N., Dehal, P.S., Arkin, A.P., 2010. FastTree 2—approximately maximum-
likelihood trees for large alignments. PLoS One 5, e9490.
R Development Core Team, 2012. R: A Language and Environment for Statistical
Computing. R Foundation for Statistical Computing, Vienna, Austria ISBN 3-
900051-07-0.
Robertson, G.P., Groffman, P.M., 2007. Nitrogen transformations. In: Paul, E.A. (Ed.),
Soil Microbiology, Ecology, and Biochemistry. Springer, New York, pp. 341–364.
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 J. 4, 1340–1351.
Schrijver, A.D., Mot, R.D., 1999. Degradation of pesticides by actinomycetes. Crit. Rev.
Microbiol. 25, 85–119.
Shen, J., Zhang, L., Zhu, Y., Zhang, J., He, J., 2008. Abundance and composition of
ammonia-oxidizing bacteria and ammonia-oxidizing archaea communities of
an alkaline sandy loam. Environ. Microbiol. 10, 1601–1611.
Smith, R.G., Menalled, F.D., Robertson, G.P., 2007. Temporal yield variability under
conventional and alternative management systems. Agron. J. 1629–1634.
Stephen, J.R., Kowalchuk, G.A., Bruns, M.-A.V., McCaig, A.E., Phillips, C.J., Embley, T.
M., Prosser, J.I., 1998. Analysis of b-subgroup proteobacterial ammonia oxidizer
populations in soil by denaturing gradient gel electrophoresis analysis and
hierarchical phylogenetic probing. Appl. Environ. Microbiol. 64, 2958–2965.
Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring
soil microbial biomass C. Soil Biol. Biochem. 19, 703–707.
Wang, Q., Garrity, G.M., Tiedje, J.M., Cole, J.R., 2007. Naive Bayesian classifier for
rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl.
Environ. Microbiol. 73, 5261–5267.
Xiong, J., Liu, Y., Lin, X., Zhang, H., Zeng, J., Hou, J., Yang, Y., Yao, T., Knight, R., Chu, H.,
2012. Geographic distance and pH drive bacterial distribution in alkaline lake
sediments across Tibetan Plateau. Environ. Microbiol. 14, 2457–2466.
Yang, L., Li, T., Li, F., Lemcoff, J.H., Cohen, S., 2008. Fertilization regulates soil
enzymatic activity and fertility dynamics in a cucumber field. Sci. Hortic. 116,
21–26.
Zhang, Y., Li, D., Wang, H., Xiao, Q., Liu, X., 2006. Molecular diversity of nitrogen-fixing
bacteria from the Tibetan Plateau, China. FEMS Microbiol. Lett. 260, 134–142.
Zhao, J., Ni, T., Li, Y., Xiong, W., Ran, W., Shen, B., Shen, Q., Zhang, R., 2014a. Responses
of bacterial communities in arable soils in a rice–wheat cropping system to
different fertilizer regimes and sampling times. PLoS One 9, e85301.
Zhao, J., Zhang, R., Xue, C., Xun, W., Sun, L., Xu, Y., Shen, Q., 2014b. Pyrosequencing
reveals contrasting soil bacterial diversity and community structure of two
main winter wheat cropping systems in China. Microb. Ecol. 67, 443–453.
Zhong, W., Cai, Z., 2007. Long-term effects of inorganic fertilizers on microbial
biomass and community functional diversity in a paddy soil derived from
quaternary red clay. Appl. Soil Ecol. 36, 84–91.