ORIGINAL RESEARCH
published: 14 September 2016
doi: 10.3389/fmicb.2016.01446
Edited by:
Graeme W. Nicol,
Université de Lyon, France
Reviewed by:
Gwen-Aelle Grelet,
Landcare Research, New Zealand
Anyi Hu,
Chinese Academy of Sciences, China
*Correspondence:
Davide Francioli
davide.francioli@ufz.de
Specialty section:
This article was submitted to
Terrestrial Microbiology,
a section of the journal
Frontiers in Microbiology
Received: 20 June 2016
Accepted: 30 August 2016
Published: 14 September 2016
Citation:
Francioli D, Schulz E, Lentendu G,
Wubet T, Buscot F and Reitz T
(2016) Mineral vs. Organic
Amendments: Microbial Community
Structure, Activity and Abundance
of Agriculturally Relevant Microbes
Are Driven by Long-Term Fertilization
Strategies. Front. Microbiol. 7:1446.
doi: 10.3389/fmicb.2016.01446
Mineral vs. Organic Amendments:
Microbial Community Structure,
Activity and Abundance of
Agriculturally Relevant Microbes Are
Driven by Long-Term Fertilization
Strategies
Davide Francioli
1
*
, Elke Schulz
1
, Guillaume Lentendu
2
, Tesfaye Wubet
1,3
,
François Buscot
1,3
and Thomas Reitz
1,3
1
Department of Soil Ecology, Helmholtz Centre for Environmental Research – UFZ, Halle, Germany,
2
Department of Ecology,
University of Kaiserslautern, Kaiserslautern, Germany,
3
German Centre for Integrative Biodiversity Research (iDiv),
Halle-Jena-Leipzig, Leipzig, Germany
Soil management is fundamental to all agricultural systems and fertilization practices
have contributed substantially to the impressive increases in food production. Despite
the pivotal role of soil microorganisms in agro-ecosystems, we still have a limited
understanding of the complex response of the soil microbiota to organic and mineral
fertilization in the very long-term. Here, we report the effects of different fertilization
regimes (mineral, organic and combined mineral and organic fertilization), carried out
for more than a century, on the structure and activity of the soil microbiome. Organic
matter content, nutrient concentrations, and microbial biomass carbon were significantly
increased by mineral, and even more strongly by organic fertilization. Pyrosequencing
revealed significant differences between the structures of bacterial and fungal soil
communities associated to each fertilization regime. Organic fertilization increased
bacterial diversity, and stimulated microbial groups (Firmicutes, Proteobacteria, and
Zygomycota) that are known to prefer nutrient-rich environments, and that are involved
in the degradation of complex organic compounds. In contrast, soils not receiving
manure harbored distinct microbial communities enriched in oligotrophic organisms
adapted to nutrient-limited environments, as Acidobacteria. The fertilization regime also
affected the relative abundances of plant beneficial and detrimental microbial taxa,
which may influence productivity and stability of the agroecosystem. As expected,
the activity of microbial exoenzymes involved in carbon, nitrogen, and phosphorous
mineralization were enhanced by both types of fertilization. However, in contrast to
comparable studies, the highest chitinase and phosphatase activities were observed in
the solely mineral fertilized soil. Interestingly, these two enzymes showed also a particular
high biomass-specific activities and a strong negative relation with soil pH. As many
soil parameters are known to change slowly, the particularity of unchanged fertilization
treatments since 1902 allows a profound assessment of linkages between management
and abiotic as well as biotic soil parameters. Our study revealed that pH and TOC were
Frontiers in Microbiology | www.frontiersin.org 1 September 2016 | Volume 7 | Article 1446
Francioli et al. Fertilization Drives Soil Microbiota
the majors, while nitrogen and phosphorous pools were minors, drivers for structure
and activity of the soil microbial community. Due to the long-term treatments studied,
our findings likely represent permanent and stable, rather than transient, responses of
soil microbial communities to fertilization.
Keywords: long-term fertilization, soil nutrients, microbial biomass, microbial activity, 454 pyrosequencing, soil
microbial communities
INTRODUCTION
Global demand for agricultural crops is increasing and global
food production is already dependent on intensive agricultural
management. Fertilization is a common farming practice, in
which organic and inorganic fertilizers are used primarily to
improve plant nutrition and hence crop productivity. The type
and quantity of fertilizer amendment affects not only crop
yields but also the physico-chemical properties of the soil,
which, in the long-term, have a significant influence on soil
fertility and productive capacity (
Saha et al., 2008). Fertilization
with organic amendments typically improves soil fertility and
structure by increasing soil nutrient status and organic matter
content (Reganold et al., 1987; Maeder et al., 2002; Liang et al.,
2012
). Consequently organic fertilization enhances soil microbial
biomass (Esperschutz et al., 2007; Gomiero et al., 2011) and
activity (
Ros et al., 2003). Mineral fertilizers, especially nitrogen
(N) inputs, have made a major contribution to the impressive
increases in crop yield achie ved since the 1950s (Robertson
and Vitousek, 2009). Despite the positive effects of inorganic
fertilizers on crop yields, there can be indirect negative effects
on soil quality arising from the complex transformations of N
in the soil. The application of ammonium fertilizers may reduce
soil pH by causing a high rate of proton release to the soil
due to enhanced nitrification processes and ammonium uptake
by the plants. As a consequence, soil acidification can lead
to deficiency of many nutrients, decreases in crop yield and
deterioration in soil fertility (Barak et al., 1997). Concerning
the effects of mineral fertilizers on soil microbial growth and
activity, contrasting results have been found so far. Recently,
Geisseler and Scow (2014) examined 107 datasets of 64 long-
term trials from around the world, reporting that in most of
the studies mineral fertilization led to a significant increase in
the soil microbial biomass, while other field studies based on
short-term application of N amendments found opposite results
(Lupwayi et al., 2011; Roberts et al., 2011; Lazcano et al., 2013).
In addition, positive and negative effects (i.e., bot h increases and
decreases in soil microbial activities) on microbial soil enzyme
activities have been reported in soil receiving inorganic fertilizers
(Gianfreda and Ruggiero, 2006; Guo et al., 2011; Nannipieri
et al., 2012). By comparing the impacts of mineral and organic
fertilization on soil communities, higher soil microbial biomass
and different community structures have already been observed
in agricultural soil with regular organic manure application
(
Marschner et al., 2004; Esperschutz et al., 2007; Lentendu et al.,
2014). Although several studies were addressed to survey the
effects of fertilization on the soil ecosystem, most of them
focused on short-term responses, which are expected to differ
considerably from those in the long-term (
Liang et al., 2015; Eo
and Park, 2016
). Long-term fertilization can have more persistent
impacts on soil characteristics (
Rousk et al., 2011; Körschens
et al., 2013), plant growt h (Clark et al., 2007), and microbial
diversity and activity (Giacometti et al., 2014; Hartmann et al.,
2015). For instance, the effects of inorganic fertilizers can differ
according to its application period, and it may take a long time
for a soil ecosystem to reach an equilibrium state (Fauci and
Dick, 1994
; Moscatelli et al., 2008). Thus, many previous studies
have emphasized the importance of long-term field experiments
to evaluate the effects of different farming systems on soil quality
and productivity (Rasmussen et al., 1998; Marschner et al., 2003;
Chakraborty et al., 2011). The “Static Fertilization Experiment”
(Bad Lauchstädt, Saxony-Anhalt, Germany), established in 1902,
is one of the oldest agricultural field experiments worldwide,
and it aims to provide a comprehensive understanding of the
effects of long-term fertilization on the yields and quality of crops
as well as on soil fertility and ecosystem functions (
Merbach
and Schulz, 2012). In this study, we used a 454-pyrosequencing
approach of the bacterial 16S rDNA and the fungal ITS region to
examine the response of soil microbial communities to 113 years
of different fertilization regimes (mineral, organic, and combined
fertilization) in the “Static Fertilization Experiment.” Moreover,
we analyzed the influence of long-term nutrient addition on
soil properties, microbial biomass and on the activity of soil
hydrolases involved in carbon, nitrogen and phosphorus cycles
in soil.
The overall objective of the present study was to use a
multidisciplinary approach to carry out an in-depth survey on
the long-term effects of different fertilization strate gies on the
activity and composition of soil microbiome. Assuming that each
fertilization treatment has a different influence on soil organic
matter and nutrient levels, we hypothesized that (i) the soil
edaphic properties, which are strongly affected by long-term
fertilization, in turn affect the structure, quantity and diversity of
soil microbial communities; and (ii) such variations in microbial
communities would reflect changes in microbial activity and
function. Moreover, we hypothesized that (iii) the community
changes caused by long-term fertilization included shifts in
the abundance of various plant-beneficial and detrimental soil
microorganisms, thus influencing agro-ecosystem per formance
and stability.
MATERIALS AND METHODS
Study Site, Experimental Design, and
Soil Sampling Strategy
The Bad Lauchstädt research station is located in the Hercynian
dry region of central Germany (11
◦
53
′
E, 51
◦
24
′
N), which is
Frontiers in Microbiology | www.frontiersin.org 2 September 2016 | Volume 7 | Article 1446
Francioli et al. Fertilization Drives Soil Microbiota
characterized by a mean annual precipitation of 484 mm and
a mean annual temperature of 8.7
◦
C. The “Static Fertilization
Experiment,” situated on a loess-derived loam soil (Haplic
Chernozem; FAO) consisting of 21.0% clay, 67.8% silt and
11.2% sand, was established in 1902 and consists of 18
different fertilization treatments based on a 4-year crop rotation.
A comprehensive description of the experimental site and
treatments was given by
Blair et al. (2006). The crop rotation
consists of sugar beet (B. vulgaris), spring barley (H. vulgare),
potato (S. tuberosum) and winter wheat (T. aestivum). Each year,
all crops are grown in parallel on the four crop strips at the
experimental site. The fertilization treatments chosen for this
study were no fertilization (NF), mineral fertilization (NPK),
farmyard manure fertilization (FYM), and combined farmyard
manure and mineral fertilization (FYM+NPK). Soils fertilized
with FYM received a total of 20 t ha
−1
of farmyard manure (solid,
cattle manure with bedding) in every other year while the soils
fertilized with NPK received an annual dose of mineral fertilizer
(calcium ammonium nitrate + superphosphate + potassium
chloride, Supplementary Table S1). The application of fertilizers is
carried out in November after the crop growing season. Details on
crop yield (for 2012; see Supplementary Table S2) and agricultural
management practices are described elsewhere (
Blair et al., 2006).
Soil samples were collected in October 2012 before fertilizers
had been applied and crops had been sown. Sampling was
performed for the four selected fertilization treatments on
all four crop strips. The size of each plot is 265 m
2
, but
no independent replicates were included when the long-term
fertilization experiment was established. We therefore used
an adapted sampling design in order to obtain five soil
pseudo-replicates representative of each of the four fertilization
treatments (Supplementary Figure S1). Each of the four plots
representing one fertilization treatment was divided into five
subplots. From each subplot, 50 randomly distributed soil cores
(up to a depth of 20 cm and 1.2 cm in diameter) were taken
with a sampling probe and pooled. Afterward we combined
the corresponding soil samples from each strip, obtaining five
composite soil replicates (A–E; Supplementary Figure S1) for
each fertilization treatment. Soil samples were sieved through a
2 mm mesh at the sampling site and kept at 4
◦
C for chemical and
microbial enzyme analyses and at −80
◦
C for molecular analyses.
Analytical Methods
Total Carbon and Nitrogen
Total organic carbon (TOC) and total nitrogen (TN) contents
of the soil samples were determined in triplicate by dry
combustion using a Vario EL III C/H/N analyser (Elementar,
Hanau, Germany). Since the carbonate concentration of the soils
was negligible (<2%), the total C concentration measured was
considered to represent TOC.
Hot Water Extractable C
Hot water extraction from air-dried soil samples was performed
in order to quantify the labile organic C pool, i.e., the potentially
mineralizable and decomposable fraction of the TOC. This was
done by boiling a soil/water suspension (1:5, w/v) for 1 h
under reflux, according to
Schulz (2002). After cooling to room
temperature, 0.1 ml of 1 M MgSO
4
was added to facilitate
soil sedimentation. The sedimented suspensions were then
centrifuged for 10 min at 6700 × g to obtain clear extracts. All
water extracts were filtered (0.45 µm Minisart RC25 single-use
syringe membrane filters, PP-housing, Sartorius AG, Göttingen,
Germany) prior to the determination of hot water extractable
C (HWC) concentrations (mg kg
−1
), which was done using an
elemental analyser for liquid samples (Multi N/C, Analytik Jena,
Germany).
Mineral N Analysis
NH
4
+
-N and NO
−
3
-N were extracted from 10 g of fresh soil
with 1 M KCl (1:4 w/v) by shaking horizontally for 1.5 h. After
filtration of the suspension (Whatman Schleicher and Schuell 595
1/5 Ø 270 mm), the concentrations of NH
4
+
-N and NO
3
−
-N in
the clear extracts were measured using a flow injection analyzer
(FIAstar 5000, Foss GmbH, Rellingen, Germany).
Plant Available P Analysis
Plant available P was extracted from fresh soil with double
lactate (1:50 w/v, pH 3.6, 1.5 h;
Riehm, 1943). After filtration
of the suspension (Whatman Schleicher and Schuell 595 1/5 Ø
270 mm), the extracted P was quantified colorimetrically using
the molybdenum blue method (Murphy and Riley, 1962).
Microbial Biomass Carbon Using Substrate Induced
Respiration
Microbial biomass carbon (MBC) was estimated with 20 g dry
equivalent of field-moist soil according to
Anderson and Domsch
(1978). All samples were pre-incubated at 22
◦
C for10 days in
polyethylene vessels. The vessels were then incubated in an
automatic respirometer (Respicond V, Nordgren Innovations AB,
Sweden) at a constant temperature of 22
◦
C, and measurements
of CO
2
evolution were taken hourly. After 24 h samples were
amended with 0.8 g of a glucose/talcum mixture (1:1.5 w/w).
From the initial (4–6 h) CO
2
response, we calculated the SIR-
MBC using the regression equation of
Anderson and Domsch
(1978). CO
2
rate from the respirometer readings, given as mg
CO
2
g
−1
h
−1
, was converted into µl CO
2
g
−1
h
−1
.
Soil Enzyme Assays
Determination of the activities of five hydrolytic enzymes
was based on 4 methylumbelliferone (MUB)-coupled
substrates (
German et al., 2011). 4-MUB-β-D-cellobioside,
4-MUB-β-D-glucoside, 4-MUB-β-D-xyloside, 4-MUB-N-
acetyl-β-D-glucosaminide and 4-MUB-phosphate were used
to estimate the activities of enzymes involved in carbon
(β-glucosidase, cellobiohydrolase, xylosidase), nitrogen
(N-acetylglucosaminidase), and phosphorus (phosphatase)
acquisition, respectively (Supplement ary Table S3). The substrate
concentrations in the assays were optimized in a pre-test
to ensure that each enzyme was assayed under saturating
conditions, in order to avoid an underestimation of enzyme
activities (
Nannipieri and Gianfreda, 1998). Soil suspensions
were prepared by adding 0.5 g of fresh soil to 50 ml of 50 mM
Tris (pH 6.4, according to the me an pH of all soil samples)
and subsequent sonication for 5 min. MUB standards (0.16,
Frontiers in Microbiology | www.frontiersin.org 3 September 2016 | Volume 7 | Article 1446
Francioli et al. Fertilization Drives Soil Microbiota
0.625, 1.25, and 2.5 µM) dissolved in buffer and soil suspensions
were used to calculate the emission and quench coefficients
for each sample. Enzyme activities were calculated according
to German et al. (2011). We additionally determined biomass-
specific enzyme activities (µmol g C
−1
h
−1
) by dividing
enzyme activities (nmol g soil
−1
h
−1
) by microbial biomass
(µg C g soil
−1
).
DNA Extraction, Amplicon Library
Preparation, and Pyrosequencing
Total soil DNA was extracted from 0.25 g of soil using a ZR
Soil Microbe DNA MiniPrep kit (Zymo Research, Irvine,
CA, USA), according to the manufacturer’s instructions.
DNA concentrations in the extracts were quantified using
a NanoDrop ND-1000 spectrophotometer (NanoDrop
Technologies, Wilmington, DE, USA). Bacterial and fungal
amplicon libraries were produced using custom fusion primers.
We used the primer pair BAC 341F and BAC 907R to amplify
the V3–V5 region of the bacterial 16S rRNA gene (
Muyzer
and Smalla, 1998). We used the primer pair ITS1F and ITS4 to
amplify the fungal internal transcribed spacer (ITS) rRNA region.
Custom primers for unidirectional sequencing were constructed
using 10 bp barcodes, sequencing primers and the BAC 907R and
ITS4 primers (Lentendu et al., 2014); PCR amplifications were
performed in triplicate with a total volume of 50 µL reaction
mix containing 1 µL of soil DNA template, 25 µL GoTaq Green
Master Mix (Promega, Mannheim, Germany) and 1 µL of each
primer (25 µM).
The thermal profile used for preparation of bacterial rDNA
amplicon libraries was as follows: initial denaturation at 98
◦
C
for 1 min, 30 cycles of denaturation at 95
◦
C for 45 s, annealing
at 57
◦
C for 45 s, and extension at 72
◦
C for 90 s, followed
by a final extension period at 72
◦
C for 10 min. The reactions
for preparation of fungal ITS rDNA amplicon libraries were
performed using touchdown PCR conditions with an initial
denaturation for 5 min at 95
◦
C followed by: (1) 10 cycles of
94
◦
C for 30 s, 60–50
◦
C (−1
◦
C per cycle) for 45 s and 72
◦
C for
2 min; and (2) 30 cycles of 94
◦
C for 30 s, 50
◦
C for 45 s and
72
◦
C for 2 min, with a final extension step of 10 min. The PCR
products were analyzed using a 1.5% agarose gel, and amplicons
from the triplicate PCRs were then pooled and purified using
a Qiagen Gel Extraction kit (Qiagen, Hilden, Germany). The
amount of DNA in each of the purified samples was me asured
using the PicoGreen dsDNA assay (Invitrogen, Carlsbad, CA,
USA). Samples were pooled at equimolar concentration and
sequenced unidirectionally using a 454 Titanium amplicon
sequencing kit and a Genome Sequencer FLX 454 System (454
Life S ciences/Roche Applied Biosystems, Mannheim, Germany)
at the Department of Soil E cology, Helmholtz Centre for
Environmental Research (UFZ, Halle, Germany).
Pyrosequencing Data Analysis
Raw reads were first demultiplexed and they were further quality-
trimmed if they c arried the expected barcode and forward primer
sequences with maxima of one and four mismatches, respectively,
using the MOTHUR software package (
Schloss et al., 2009).
For b acterial 16S reads, sequences with a minimum average
quality of a Phred score of 30, a minimum length of 500 nt,
a maximum homopolymer length of eight nucleotides, and no
ambiguous nucleotides, were selected as high quality re ads.
Similarly, fungal ITS reads were retained as high quality reads
if they had a minimum average quality of a Phred score of
25, a minimum length of 450 nt, a maximum homopolymer
length of eight nucleotides and a maximum of eight ambiguous
nucleotides. High quality 16S reads were first aligned against
the aligned version of the reference Silva SSU database (Release
115, non-redundant, clustered at 99% similarity;
Quast et al.,
2013). Chimeras were further removed from the aligned 16S
reads and the high quality IT S reads using UCHIME (Edgar
et al., 2011) as implemented in MOTHUR. Unique sequences
were sorted by decreasing abundance for both the 16S and the
ITS dataset and were clustered into OTUs using CD-HIT-EST
(Fu et al., 2012) at a threshold of 97% pairwise similarity. Low
abundant OTUs wit h 3 or fewer reads were removed as they
potentially originated from sequencing artif acts (
Kunin et al.,
2010). Bacterial 16S OTU representative sequences were classified
against the non-aligned version of t he above mentioned Silva
database using the MOTHUR implement ation of the Wang
et al. (2007) classifier. Representative sequences for fungal ITS
OTUs were classified against the dynamic version of the UNITE
database (version 6, January’ 14; Kõljalg et al., 2013). Those ITS
sequences which could not be assigned further than t he kingdom
Fungi were reclassified against the previous dat abase augmented
with all eukaryotic and non-fungal ITS sequences retrieved from
GenBank (release 202, Benson et al., 2008), in order to remove
non-target organism sequences (those from OTUs not affiliated
to fungi). If a sequence could still not be assigned to a fungal
phylum, it was classified against the full version of the UNITE
database in order to improve its taxonomic annotation.
Nucleotide Accession Number
The 454 pyrosequencing data generated for this study were
submitted to the European Nucleotide Archive (ENA) under
accession numbers PRJEB9307 and PRJEB9305.
Statistical Analysis
Univariate Analysis of Variance (ANOVA) followed by Tukey’s
honestly significant difference (HSD) post hoc test was used
to test for differences in soil chemical properties, microbial
biomass, and enzyme activities among the four treatments
investigated in this work. All the variables were tested for
normality using Shapiro–Wilk and Jarque–Bera tests and the
equality of group variances was examined using Levene’s test.
A log
10
transformation [log
10
(x + 1)] was applied to all variables
that did not meet the parametric assumptions. Correlation
among the soil parameters and between soil parameters and
enzyme activities were determined using Spearman’s rank
correlation. Diversity of the bacterial and fungal community was
assessed by calculating the Shannon–Wiener index. Differences
in bacterial and fungal OTU richness and diversity were
compared using an ANOVA followed by a Tukey’s HSD post hoc
test. Similarities in the bacterial and fungal community structure
among the four fertilization treatments were investigated using
Frontiers in Microbiology | www.frontiersin.org 4 September 2016 | Volume 7 | Article 1446
Francioli et al. Fertilization Drives Soil Microbiota
an analysis of the similarity (ANOSIM) algorithm. Non-
metric multidimensional scaling (NMDS) based on the Bray–
Curtis dissimilarity index was used to visualize the patterns of
distribution of microbial communities in relation to fertilization
treatments. Permutational multivariate analysis of variances
(PERMANOVA) based on the Bray–Curtis dissimilarity index
was performed to analyze the effect of fertilization treatment on
community composition in each dat a set, using 999 permutations
for each test. We used the option “strata” in the R package
“vegan” to constrain the permutation of samples within each
fertilization group. A model of multivariate analysis of variance
was constructed using distance-based redundancy analysis
(dbRDA) based on the Bray–Curtis distance to determine
the environmental variables that were most influential on the
bacterial and fungal community compositions. Marginal tests
were performed to determine the amounts of variation explained
by the selected variables. Signific ance tests were performed
through nonparametric permutation, which does not rely on the
assumption of multivariate normality (
Taylor et al., 2009). All
the data were analyzed with R version 2.15.3 (R Core Team,
2012
).
RESULTS
Soil Chemical Properties and Microbial
Biomass Carbon
The soil parameters have been strongly influenced by the long-
term application of mineral and organic fertilizers (Table 1). The
unfertilized soil (NF) had a pH of 6.85, while the soil pH had
decreased to a value of 5.74 in the NPK treated soil. Slightly
lower pH values, compared to the NF soil, of 6.56 and 6.42 were
also observed in the organic (FYM) and combined (FYM+NPK)
fertilized soils, respectively. Long-term mineral and organic
nutrient inputs induced a significant increase (p < 0.05) in
TOC, HWC, TN, and available phosphorus [P
(DL)
] in soil and
they increased in the order NF < NPK < FYM < FYM+NPK,
TABLE 1 | Chemical and biological properties of soil samples.
NF NPK FYM FYM+NPK
pH 6.85 (0.04)
a
5.74 (0.05)
c
6.56 (0.03)
b
6.42 (0.05)
b
TOC (g/kg) 160 (1.2)
d
188 (4.1)
c
210 (3.7)
b
229 (2.2)
a
TN (g/kg) 11 (0.2)
d
14 (0.4)
c
17 (0.8)
b
20 (0.2)
a
HWC (mg/kg) 284 (0.62)
d
360 (14.3)
c
451 (5.9)
b
518 (5.8)
a
NH
4
+
-N (mg/kg) 0.76 (0.03)
b
1.30 (0.04)
a
0.72 (0.05)
b
0.92 (0.04)
b
NO
3
−
-N (mg/kg) 10.1 (0.1)
d
17.9 (0.2)
b
13.71 (0.3)
c
28.1 (0.3)
a
P
(DL)
(mg/kg) 44.6 (2.2)
d
83.0 (1.4)
c
124.7 (1.4)
b
171.6 (2.3)
a
C/N (ratio) 14.1 (0.05)
a
13.9 (0.3)
a
12.4 (0.4)
b
11.3 (0.05)
b
MBC (µg/g) 131 (7.9)
c
146 (7.7)
c
197 (10.9)
b
239 (13.7)
a
Values are means and letters denote significant differences among the treatments
(p < 0.05). Standard error of the means are indicated in parentheses. NF, no
fertilization; NPK, mineral fertilization; FYM, organic fertilization; FYM+NPK, organic
and mineral fertilization; TOC, total organic carbon; TN, total nitrogen; HWC,
hot water extractable carbon; P
(DL)
, plant available phosphorous; MBC, microbial
biomass carbon.
whereas an opposite trend was observed for C/N ratio. TOC
was positively correlated with TN (r = 0.97, p < 0.01),
HWC (r = 0.88, p < 0.01) and P
(DL)
(r = 0.90, p < 0.01).
Furthermore, HWC was positively correlated with TN (r = 0.94,
p < 0.01) and P
(DL)
(r = 0.95, p < 0.01) and the latter
was also strongly correlated with TN (r = 0.93, p < 0.01).
No significant differences in soil ammonium concentration
were observed among the soils studied, with the exception
of t he one fertilized with NPK, which showed a higher
ammonium content. Nitrate concentrations were significantly
higher (p < 0.05) in all fertilized soils and increased in the order
NF < FYM < NPK < FYM+NPK.
Farmyard manure fertilized soils showed a strong increase
of MBC, while long-term application of mineral fertilizer
induced only a slight increase in MBC relative to the
unfertilized treatments (Table 1). MBC increased by 10,
53, and 79% in the NPK, FYM, and FYM+NPK fertilized
soils, respectively, compared to NF soil. A significant positive
correlation (p < 0.01) was found between MBC and the amount
of TOC, TN, and HWC with r values of 0.71, 0.73, and 0.65,
respectively.
Enzyme Activities
Long-term fertilization significantly increased (p < 0.05) the
activities of β-glucosidase, cellobiohydrolase, N-acetylgluco-
saminidase, phosphatases, and xylosidase compared to their
activities in the NF soil (Figure 1). β-glucosidase and xylosidase
activities were approximately two times higher in all fertilized
treatments than in the NF soil. Cellobiohydrolase activity was
about four times higher in the NPK and FYM+NPK soils,
and three times higher in the FYM soil, than in the NF soil.
N-acetylglucosaminidase showed the highest activity in the NPK
soil, followed by FYM+NPK and FYM soils, while in the NF
soil its activity was low. The phosphatase activity in the NPK
soil was nearly five times higher than that in the control, three
times higher than that in the FYM soil and almost two times
higher than that in the FYM+NPK treatment. The hydrolase
activities were significantly correlated among each other and
β-glucosidase, cellobiohydrolase and xylosidase were further
positively correlated with TOC (r = 0.59; r = 0.61; r = 0.54;
respectively, p < 0.05) and with TN (r = 0.53; r = 0.59;
r = 0.47; respectively, p < 0.05). N-acetylglucosaminidase and
phosphatase activities were negatively related to pH (r = −0.83;
r = −0.91, respectively, p < 0.001) and positively correlated
with ammonium (r = 0.57; r = 0.60, respectively, p < 0.01)
and nitrate (r = 0.65; r = 0.69, respectively, p < 0.01)
concentrations.
Bacterial and Fungal Richness and
Diversity
A total of 200,477 16S bacterial and 76,267 ITS fungal raw
reads were obtained from the 20 soil samples (4 treatments × 5
replicates). After quality filtering steps and exclusion of singletons
doubletons and tripletons, 135,761 bacterial and 54,560 fungal
high quality reads were recovered. Sequences were clustered into
3,122 b acterial and 610 fungal OTUs.
Frontiers in Microbiology | www.frontiersin.org 5 September 2016 | Volume 7 | Article 1446