The effects of pH change and NO− 3 pulse on microbial community

Transcription

RESEARCH ARTICLE
The effects of pH change and NO−3 pulse on microbial
community structure and function: a vernal pool microcosm
study
Sarah R. Carrino-Kyker1,2, Kurt A. Smemo2,3 & David J. Burke1,2
1
Case Western Reserve University, Cleveland, OH, USA; 2The Holden Arboretum, Kirtland, OH, USA; and 3Kent State University, Kent, OH, USA
Correspondence: Sarah R. Carrino-Kyker,
Department of Biology, Case Western
Reserve University, 2080 Adelbert Road,
Cleveland, OH 44106, USA.
Tel.: +1 440 946 4400 x266; fax:
+1 440 602 8005; e-mail: [email protected]
Received 19 December 2011; revised 13 April
2012; accepted 18 April 2012.
Final version published online 14 May 2012.
DOI: 10.1111/j.1574-6941.2012.01397.x
MICROBIOLOGY ECOLOGY
Editor: Tillmann Lueders
Keywords
denitrification; respiration; bacteria; fungi;
denitrifiers.
Abstract
Forest vernal pools experience strong environmental fluctuations, such as
changes in water chemistry, which are often correlated with changes in microbial community structure. However, very little is known about the extent to
which these community changes influence ecosystem processes in vernal pools.
This study utilized experimental vernal pool microcosms to simulate persistent
pH alteration and a pulse input of nitrate (NO
3 ), which are common perturbations to temperate vernal pool ecosystems. pH was manipulated at the onset
and microbial respiration was monitored throughout the study (122 days). On
day 29, NO
3 was added and denitrification rate was measured and bacterial,
fungal, and denitrifier communities were profiled on day 30 and day 31.
Microbial respiration and both bacterial and fungal community structure were
altered by the pH treatment, demonstrating both structural and functional
microbial responses. The NO
3 pulse increased denitrification rate without
associated changes in community structure, suggesting that microbial
communities responded functionally without structural shifts. The functioning
of natural vernal pools, which experience both persistent and short-term environmental change, may thus depend on the type and duration of the change
or disturbance.
Introduction
Vernal pools are seasonally flooded habitats that, in temperate regions, are typically inundated in the spring
owing to snowmelt, high precipitation, and low evapotranspiration. Although they are typically small in size
compared to permanent bodies of water, they are abundant across the landscape (Colburn, 2004) and, therefore,
functionally and ecologically significant. Much of the
research on vernal pool organisms has focused on
amphibians and macroinvertebrates, which take advantage
of the fish-free habitat for breeding (reviewed in Colburn,
2004). Very little is known, however, about the microbial
communities in vernal pools, even though these
communities are responsible for much of the nutrient
cycling and energy flow in wetland habitats (Gutknecht
et al., 2006). In forests, where vernal pools receive high
amounts of organic matter from allochthonous leaf litter
inputs in addition to excrement and debris (e.g.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
exoskeletons) left behind by aquatic organisms (Williams,
2006), microbially mediated energy flow may be especially
important. Consequently, further studies on the microorganisms in vernal pools are necessary to fully understand
the ecological impact of microorganisms in these habitats.
Because of their small size, shallow depth, and short
hydroperiod, vernal pools experience a high degree of
environmental fluctuation, both spatially and temporally
(Brooks, 2000; Carrino-Kyker & Swanson, 2007, 2008).
Over the course of 4 months, Carrino-Kyker & Swanson
(2008) noted changes to microbial community structure
in vernal pools and correlated those changes with temporal fluctuations in water column pH, dissolved oxygen,
and conductivity. Not only are vernal pool microbial
communities temporally dynamic, but they have also
been shown to vary spatially in response to water column
conductivity and dissolved oxygen, to leaf litter quality,
and to the degree of surrounding urbanization (CarrinoKyker et al., 2011). However, whether the observed
FEMS Microbiol Ecol 81 (2012) 660–672
Effects of pH and NO
3 on vernal pool microbial communities
changes to microbial communities in vernal pools also
affect their function is uncertain.
Studies from other systems have shown that environmental conditions can simultaneously affect both
microbial community structure and function (e.g. Balser
& Firestone, 2005; Bell et al., 2009; Smucker & Vis,
2011); however, conflicting studies have found no
association between microbial community structure and
ecosystem processes when one is affected by environmental change (Rich & Myrold, 2004; Menyailo et al., 2010;
Attard et al., 2011), implying functional redundancy in
some communities. It has been proposed that the type
and duration of the environmental change will determine
whether microbial community structure is altered and if
there is a corresponding change in ecosystem function
(Balser et al., 2001). Despite the importance of this topic
and the uncertainty concerning the effects of environmental conditions on microbial community structure and
function, we are unaware of any studies that have
attempted to explore these relationships in vernal pools.
Vernal pools are a good model system for studying the
effects of multiple environmental conditions on microbial
community structure and function because of their high
degree of environmental variability and expected sensitivity to environmental change. Environmental influences on
vernal pools may be persistent and associated with different soil types, underlying geologies, and surrounding
plant communities and land uses (Carrino-Kyker &
Swanson, 2007). However, other environmental changes
may be sporadic, such as nutrient pulses associated with
urban and agricultural runoff following rainfall events
(Moldan & Černý, 1994). Because runoff is a significant
source of water for vernal pool formation (Colburn,
2004), seasonal pulses of pollutants may have a significant
effect on microbial community structure and function,
but this has previously not been investigated.
Here we present the results of an experimental microcosm study that examined the extent to which altered
environmental conditions influenced microbial communities and the processes they mediate. In particular, we were
interested in parameters that are related to human alteration of the environment, such as pH changes that can
result from acid deposition and nutrient enrichment from
pollution. As such, vernal pools were simulated in the
laboratory and subjected to pH manipulation and nitrate
(NO
3 ) enrichment. The pH manipulation was established
at the onset and maintained throughout the 4 months of
the experiment, while the NO
3 was administered only
once part way through the experiment. This simulated
persistent pH changes previously observed in vernal
pools across an urbanization gradient (Carrino-Kyker &
Swanson, 2007) and NO
3 pulses associated with runoff
(Moldan & Černý, 1994). The purpose of this study was
661
to add to the growing literature on the responses of
microbial community structure and ecosystem function
to environmental changes and to specifically address this
question in vernal pools, which are understudied relative
to their ubiquity in temperate regions and ecological
importance.
Materials and methods
Microcosm construction and treatments
Soil from four vernal pools in Northeast Ohio, USA (41°
29′53″N, 81°25′27″W, elevation, 320 m) was randomly
sampled in November 2007. All pools contained standing
water at the time of sampling. The top 2–5 cm layer of
soil was collected with a clean shovel, transported to the
laboratory in opaque plastic bins, immediately transferred
to flat plastic trays, covered with shade cloth, and left to
dry at room temperature. Once dry, the soil was passed
through a 2-mm sieve and stored at 4 °C prior to microcosm construction. The soil was dried and stored at 4 °C
to mimic the winter temperatures and dry conditions that
vernal pools experience in the field in temperate regions
(Colburn, 2004).
Fagus grandifolia Ehrh. leaves were sampled from a
mature Beech-Maple forest located at The Holden
Arboretum, Kirtland, OH (41°36′40″N, 81°17′30″W,
elevation, 340 m) on the same date. Only F. grandifolia
leaves were used to control for variability associated
with varying litter quality. Freshly fallen leaves were collected by hand, placed in plastic bins, mixed to eliminate bias by location, covered with shade cloth, and left
to dry at room temperature. Once dry, 1.5-cm-diameter
leaf disks were taken with a punch and stored at 4 °C.
In February of 2008, 72 microcosms were constructed
in 500-mL Mason jars. Each microcosm received an
equal weight (50 g) of premixed vernal pool soil and
fine-grained sterile sand, 800 mg of leaf punches, and
150 mL of autoclaved deionized water. Sand was added
to improve aeration and diffusion because the vernal
pool soil was fine-textured. Each microcosm was
assigned to one of twelve treatments (four pH treatments and three nitrogen (N) treatments in a 4 9 3 full
factorial design), with six replicates per treatment
(n = 72). The initial pH of the microcosms ranged from
4.00 to 4.39. The four pH treatments (pH values of 5,
6, 7, and 8) represented values found in Northeastern
Ohio vernal pools (Carrino-Kyker & Swanson, 2007,
2008) and were manipulated by low-volume addition of
5 M sodium hydroxide (NaOH) based on amounts
determined by titrations in a pilot study. The pH 6, 7,
and 8 microcosms initially received 100, 350, and 500 lL,
respectively, of NaOH. Throughout the remainder of the
S.R. Carrino-Kyker et al.
662
study (122 days), 5 M hydrochloric acid (HCl, as in
Chu et al., 2007) and 5 M NaOH were used to maintain
the pH of each microcosm, which was monitored
biweekly (see Supporting Information, Table S1). While
HCl and NaOH may artificially raise salinity, measurements of sodium ions on day 113 (range of
0.8–304 ppm) and chloride ions on day 114 (range of
60–320 ppm) indicated that their levels in the microcosms were within the range of fresh water salinity
(McArthur, 2006) and we expect the effects on microorganisms to be minimal compared to our experimental
treatments. On day 29, once respiration measurements
were similar in all pH treatments (see Fig. 1), the N
treatments were implemented such that each microcosm
received a 20-mL solution containing 10-mg NO
3 -N in
þ
),
10-mg
NH
-N
in
the
form
the form of KNO3(þNO
3
4
),
or
sterile
DI
water
(N).
of (NH4)2SO4 (þNHþ
4
þ
Therefore, the þNO
3 and þNH4 microcosms received
þ
200 lg per g of dry soil of either NO
3 -N or NH4 -N.
The NO3 added was similar to levels measured in
vernal pools that received wastewater in Pennsylvania
(Laposata & Dunson, 1998). The NHþ
4 addition was
used as a comparison to determine whether any treatment effects were owing to low N (nutrient) availability,
and the DI water treatment was a control. The microcosms were randomly placed in an incubator at
16 ± 1 °C with a 12-hr light/12-hr dark cycle to simulate temperature and natural day/night cycle in Northeastern Ohio in the spring, and rotated within the
incubator to limit bias of location. The jars were
covered with parafilm that included vents to allow air
exchange.
Fig. 1. Community respiration, measured as CO2 efflux from the
microcosms (mean ± standard error), over the course of the study.
Significant P-values (a = 0.05) from a two-way ANOVA, using time as a
repeated factor and pH, are indicated with an *. The data were
natural log transformed prior to statistical analysis.
Microbial processes
Microbial community respiration was determined on days
1, 4, 8, 15, 22, 29, 30, 31, 94, and 122 by measuring CO2
efflux from each microcosm with a portable open-flow
infrared gas analyzer (IRGA; LI-6400; LiCOR® Instruments, Lincoln, NE). Respiration was determined by
scrubbing CO2 from the microcosm headspace and measuring the change in headspace CO2 concentration over
time. We used the average of two separate scrub/flux cycles
and a rate was calculated using the following formula:
Community Respiration = (dc′/dt*system volume)/soil
carbon
where dc′/dt was the rate of change of CO2 over 10 s
determined with the IRGA and corrected for water content, system volume included the volume of microcosm
headspace and IRGA tubing, and soil carbon was the
mass of carbon per gram of dried soil and leaves added
to the microcosms. The final respiration value was
converted to lg CO2-C (g soil C)1 day1.
Potential denitrification rates were determined with a
modified acetylene-block technique (Groffman et al.,
1999). On day 29, when the N treatments were administered, the microcosms were closed with airtight lids that
were fitted with rubber septa and cycled five times between
vacuum and purging with 99.9% pure nitrogen gas (N2).
Jars were then opened, and a syringe and needle were used
to bubble 30 mL of acetylene through the soil. The jars were
then reclosed and the headspace was again flushed with N2
before adding 40 mL of acetylene with a syringe and needle
to the head space via the rubber septa. Between day 29 and
day 31 of the experiment, 20 mL of headspace was sampled
with a syringe and needle at 1, 12, 16, 24, 36, and 40 h following the N additions, without controlling for headspace
pressure, and stored in scintillation vials fitted with airtight
rubber stoppers (GeoMicrobial Technologies, Ochelata,
OK). Following the 16-h headspace sampling, the microcosms were opened for respiration measurements (see
above) and to collect samples for microbial community
analysis (see below). Immediately following this sampling,
additional acetylene was added to the soil and headspace
(as described above) and the jars were again sealed with airtight lids in preparation for the remaining headspace samples. The nitrous oxide (N2O) concentration of each sample
was measured with a gas chromatograph and electron capture detector (GC-2014; Shimadzu Scientific Instruments,
Columbia, MD). Only headspace samples were collected for
N2O determination, which did not include any N2O that
was dissolved in the water. Denitrification rate was calculated following Groffman et al. (1999) by first converting
the N2O measurement in ppm to lL N2O-N/L headspace
and then subtracting this measurement from one time
point by that of an earlier time point and standardizing for
headspace volume and soil dry weight. The proportion of
NO
3 -N that was released as N2O-N during our measurements was calculated for the 16-h and 40-h time points and
summed, as an estimate of the amount of NO
3 that was
denitrified. The 16- and 40-h time points were used because
these were the final time points on day 30 and day 31,
respectively, and the jars were opened between the 2 days of
denitrification measurements.
DNA extraction and molecular analysis
On day 30 (after the 16-h headspace collection) and day
31 (after the 40-h headspace collection), microcosms were
opened to randomly sample one leaf disk and two soil
samples from each jar for microbial community analyses.
Microbial analyses were conducted on these days to
profile the communities at the same time that both respiration and denitrification measurements were made. The
microbial communities in soil and leaf samples were
profiled separately because they have previously been
shown to differ in vernal pool habitats (Carrino-Kyker &
Swanson, 2008). Samples from two microcosms in the
same treatment were randomly pooled prior to DNA
extraction, such that half of a leaf disk and ~125 mg of
soil from each microcosm were used for the molecular
work (i.e. n = 3 soil and leaf samples per treatment group
for the microbial analyses). After cell lysis using a
Precellys 24 Homogenizer (Bertin Technologies, Montigny-le-Bretonneux, France), DNA was extracted separately
from soil and leaf samples using the PowerSoilTM DNA
Isolation Kit (MOBIO Laboratories, Inc., Carlsbad, CA).
DNA from soil and leaf samples was used to amplify
general bacteria in a PCR using 16S rRNA gene primers
338f and 926r (Muyzer et al., 1993, 1995) following conditions as described in Burke et al. (2006). Fungal PCR
amplified the internal transcribed spacer region of the
rRNA gene using the primer set 58A2F and ITS4 (Martin &
Rygiewicz, 2005) following conditions as described in
Burke et al. (2005) except the extension step was 90 s. To
amplify the community of denitrifying bacteria, we
targeted the nosZ gene using the primers nosZFb and
nosZRb (Rösch & Bothe, 2005) and followed conditions as
described in Rösch et al. (2002). The nosZ gene codes for
N2O reductase, the enzyme that catalyzes the conversion of
N2O to N2 in the denitrification pathway. The nosZ gene is
found in a number of denitrifying bacteria and is considered a good marker for this group of organisms (see
Philippot & Hallin, 2006 for a review); however, the primers used here discriminate against any denitrifiers lacking
the nosZ gene, as well as gram-positive denitrifiers.
Terminal restriction fragment length polymorphism
(TRFLP) profiling was used to examine the diversity and
663
structure of the bacterial, fungal, and denitrifying
communities following protocols previously described
(Burke et al., 2005, 2006, 2008). The endonucleases MspI
and HaeIII were used for bacterial TRFLP and AluI and
HaeIII were used for fungal TRFLP. The denitrifier
TRFLP was conducted using the restriction endonucleases
MboI and TaqI (Rösch & Bothe, 2005). TRFLP profiles
were generated at the Life Sciences Core Laboratories
Center (Cornell University) and analyzed using the
GS600 LIZ size standard and Peak ScannerTM Software
(version 1.0, Applied Biosystems 2006). For our analyses,
only peaks that accounted for > 1% of the relative peak
area were included (i.e. major TRFs; Burke et al., 2008).
Statistical analyses
Because the microcosms were randomly placed in the incubator and rotated within the incubator throughout the
study, they were treated independently for statistical tests.
Differences in respiration were determined over the course
of the 122-day study with a two-way repeated measures
ANOVA using pH treatment and time as factors. For this
ANOVA, respiration measurements made on day 1 were
excluded because they were below detection in several of
the pH 7 and pH 8 microcosms. Respiration differences
were also determined between pH and N addition treatments individually on day 30 and 31 with two-way ANOVAs,
and one-way ANOVAs were used for determining whether
denitrification varied between the pH values. All ANOVAs
were conducted in SIGMASTAT version 3.5 (Systat Software,
Inc., 2006). If needed, the data were natural log
transformed for two-way ANOVAs or one-way ANOVAs were
conducted on ranks to meet homogeneity of variance
requirements. Significance was determined at a < 0.05.
TRFLP profiles were compared between sample types
(i.e. leaf and soil), sample date (i.e. day 30 or 31), and
microcosm treatments (i.e. pH and N addition) with the
nonmetric multidimensional scaling (NMS) ordination
technique and with multi-response permutation procedure
(MRPP) using PC-ORD (Version 5.0; Bruce McCune and
MJM Software, 1999). Because the TRFLP profiles contained relative proportions, they were arcsine-squareroot
transformed prior to running NMS or MRPP to improve
normality (McCune & Grace, 2002). NMS ordinations were
run separately for profiles generated with different TRFLP
enzymes and were conducted on both sample types
together, as well as for soil and leaf communities separately,
for a total of 18 NMS runs. MRPP was also used to determine differences between soil and leaf communities, as well
as to assess treatment differences for soil and leaf communities separately. TRFLP profiles generated with different
enzymes were used in individual MRPP runs for a total of
42 MRPP runs. For our microcosms, differences between
664
groups were considered statistically and ecologically significant if they were greater than by chance (P < 0.05) and the
effect size was large (A > 0.1) (McCune & Grace, 2002).
Pairwise comparisons following MRPP runs revealed the
groups that were significantly different from each other.
Prior to NMS and MRPP, outlier analysis was conducted in
PC-ORD to determine any outlying samples. Samples were
flagged as outliers if their distance measure was more than
two standard deviations from the mean distance (McCune
& Grace, 2002) and were removed from NMS and MRPP
runs if they were shown to be outlying for both TRFLP
enzymes. These criteria only eliminated one sample from
the NMS and MRPP runs of the fungal soil samples.
Operational taxonomic unit (OTU) richness of each
microbial community (bacteria, fungi, and denitrifying
microorganisms) based on TRFLP profiles was determined
in PC-ORD. The relationship between OTU richness for
each microbial community (i.e. bacteria, fungi, and denitrifier) and microbial respiration was assessed with linear
regression using SIGMAPLOT, version 10.0 (Systat Software,
Inc., 2006). Richness was not regressed against denitrification rate because it was undetected in two-thirds of the
microcosms (see below); therefore, reliable regressions
could not be made. Regressions were conducted for the richness values of each microbial community and for each of the
TRFLP enzymes. Separate regressions were made for soil
and leaf samples and for day 30 and day 31 samples for a
total of 24 regressions (three microbial communities 9 two
TRFLP enzymes 9 two sample types 9 two dates). The
data were tested for normality (i.e. Kolmogorov–Smirnov
test) and constant variance in SIGMAPLOT when running the
regressions. If assumptions were violated, regressions were
re-run with CO2 flux (i.e. the dependent variable) values
natural log transformed and/or outliers (visualized on residual plots and prediction/confidence interval plots) removed
to meet normality and equal variance assumptions. Only
five regressions contained outliers that were removed. Significance for regression relationships between richness and
respiration was determined at a < 0.05. Kutner et al. (2005)
was used as a guide for interpretation and diagnostic measures for the regressions.
rates were compared with a two-way ANOVA between pH
and N treatments individually on day 30 and day 31, significant effects of pH treatment (F = 115.36, P < 0.001
and F = 105.91, P < 0.001 for day 30 and 31, respectively), N addition (F = 24.33, P < 0.001 and F = 10.16,
P < 0.001 for day 30 and 31, respectively), and pH 9 N
treatment interactions (F = 4.15, P = 0.002 and F = 3.15,
P = 0.010 for day 30 and 31, respectively) were found on
both dates (Fig. 2). Pairwise multiple comparison tests
(i.e. Tukey tests) within each pH treatment indicated significantly different CO2 efflux between the þNO
3 and
treatments
in
the
pH
6,
7,
and
8
microcosms
on
þNHþ
4
both dates and significantly higher CO2 efflux in the
þNHþ
4 treatment compared to the controls in the pH 6
and 8 microcosms on day 30 only (Table S2).
Denitrification rates were markedly higher in the þNO
3
treatment, compared to the þNHþ
4 and N treatments
where denitrification was not detected on either day 30 or
day 31 (Fig. 3). Within the þNO
3 treatment, there were
no significant effects of pH on denitrification rate on either
date (F = 2.344, P = 0.104 for day 30 and P = 0.060 for
ANOVA on ranks conducted for day 31 data; Fig. 3). At 40 h
following the NO
3 addition, 10.6% of the NO3 -N that was
added to the þNO3 microcosms was recovered as N2O-N.
Microbial community structure
MRPP revealed significant differences (A > 0.1 and
P < 0.05) between soil and leaf communities for bacteria,
fungi, and denitrifiers (Table 1). NMS ordinations also
showed differences between soil and leaf communities
Results
Microbial processes
Respiration rate differed significantly between the pH
treatments (F = 60.12, P < 0.001) and over time
(F = 172.49, P < 0.001), and the pH 9 time interaction
was also significant (F = 13.52, P < 0.001; Fig. 1). The
pH 7 and pH 8 microcosms showed little CO2 efflux on
day 1 and ended with a greater rate than the pH 5 and
pH 6 microcosms on day 122 (Fig. 1). When respiration
Fig. 2. Community respiration, measured as CO2 efflux from the
microcosms, for each pH and N treatment on day 30 and day 31
(mean with standard error bars). Significant P-values (a = 0.05) from
a two-way ANOVA, using pH and N treatments as factors, are indicated
with an *. The data were natural log transformed prior to statistical
analysis.
665
(Fig. S1). Because of these differences between soil and leaf
communities, they were analyzed separately to determine
the effects of microcosm treatments (pH and N addition)
and date (day 30 or 31). The soil and leaf ordination diagrams shown here used the TRFLP enzymes HaeIII for bacteria, HaeIII for Fungi, and TaqI for denitrifiers. The other
TRFLP enzymes produced similar results (Fig. S2). For the
NMS graphs shown in Fig. 4, the leaf bacterial and soil
denitrifier NMS ordinations resulted in two-dimensional
solutions, while the soil bacterial, leaf denitrifier, and
Fig. 3. Denitrification rate (mean with standard error bars) for each
treatment group. Rates for the þNHþ
4 and N treatments were below
the level of detection of our method (indicated by ‘ND’). The P-values
are from a one-way ANOVA conducted within each þNO
3 treatment.
fungal ordinations resulted in three-dimensional solutions.
For NMS ordinations that resulted in three-dimensional
solutions, the graphs shown here include the two dimensions that explained the most variation (i.e. higher R2 values). NMS and MRPP showed community differences
between pH treatments for both bacteria and fungi
(Table 1; Fig. 4a–d). For bacteria, this was apparent in both
soil and leaf samples (Table 1; Fig. 4a and b), but fungi
were only affected in leaf samples and the effect size (i.e. A
statistic) was lower than for bacteria (Table 1; Fig. 4c and
d). TRFLP profiles using the nosZ gene showed no effect of
pH on soil denitrifier communities (Table 1; Fig. 4e). For
the leaf denitrifier communities, the TaqI enzyme showed
no significant effect of pH on denitrifier community structure (Table 1; Fig. 4f), but this result was not consistent
across enzymes as the MboI enzyme indicated a significant
pH effect (Table 1); however, this difference was only
between the pH 8 microcosms and the remaining three pH
treatments (Fig. S2). Thus, these results are inconclusive as
to whether pH influenced the denitrifier community structure in leaf samples. Contrary to the pH treatment, the
þ
addition of NO
3 or NH4 had little effect on microbial
community structure, even for denitrifiers, which use NO
3
as an electron acceptor (Table 1; Fig. 4). Similarly, there
was no effect of sampling date on microbial community
structure (Table 1; Fig. 4). The differences in community
structure between pH treatments and substrate type were
owing to both the presence/absence of and changes in the
relative abundance of individual terminal restriction fragments (TRFs; Tables S3–S5). In general, TRFs were consistent between the N treatments (Tables S6–S8).
Table 1. MRPP results for each microbial community TRFLP profile
Bacteria
Fungi
MspI
Grouping
factor
Leaf vs. soil
Leaf samples
pH
treatment
N addition
Date
Soil samples
pH
treatment
N addition
Date
HaeIII
Denitrifiers
AluI
HaeIII
MboI
A
P
A
P
A
P
0.55
< 0.001
0.14
< 0.001
0.16
< 0.001
0.15
< 0.001
0.14
< 0.001
0.15
< 0.001
0.26
< 0.001
0.36
< 0.001
0.13
< 0.001
0.15
< 0.001
0.18
< 0.001
0.077
< 0.001
0.0020
0.0051
0.33
0.93
0.045
< 0.001
0.093
0.0077
< 0.001
0.077
0.014
0.0018
0.23
0.028
0.017
0.041
0.27
< 0.001
0.0054
0.013
0.013
0.00075
0.085
0.32
P
TaqI
A
P
A
0.0099
0.0016
0.050
0.26
0.0090
0.0029
0.056
0.17
0.025
0.0043
0.016
0.0087
0.057
0.051
< 0.001
0.11
0.018
0.0020
< 0.001
0.73
0.15
< 0.001
0.012
0.061
0.053
< 0.001
< 0.001
0.027
0.0034
0.023
0.0047
P
A
0.0032
0.90
< 0.001
0.019
0.16
þ
Community differences between pH treatment (5, 6, 7, and 8), N addition (þNO
3 , þNH4 , and N), and sample date (day 30 or day 31) were
determined independently on soil and leaf samples because the communities were significantly different between these two sample types. TRFLP
enzymes were also analyzed separately. Differences between groups were considered significant if they were greater than by chance (P < 0.05)
and had a large effect size (A > 0.1) (McCune & Grace, 2002). Significant differences are shown in bold and were confirmed by visualizing nonmetric multidimensional scaling (NMS) graphs (see Fig. 4, Figs S1 and S2).
666
(a)
(b)
(c)
(d)
(e)
(f)
Fig. 4. Nonmetric multidimensional scaling ordination graphs showing the bacterial (a, b), fungal (c, d), and denitrifier (e, f) community
structures as determined by TRFLP. TRFLP enzymes shown here are HaeIII for the bacteria (a, b), HaeIII for the fungi (c, d), and TaqI for the
þ
denitrifiers (e, f). The pH treatments are indicated in the legend, while N treatments are represented by squares (þNO
3 ), diamonds (þNH4 ), and
circles (N). Ellipses show the general change in community structure between pH treatments (where significant, see Table 1). The two sample
dates (day 30 and day 31) are not shown on the graphs, but no differences in community structure were seen between dates (see Table 1).
Regressions of richness vs. respiration
CO2 flux was positively correlated to bacterial richness,
but negatively related to the richness of nosZ-containing
denitrifiers (Table 2; Fig. 5). Although there were some
discrepancies between TRFLP enzymes, these relationships
were significant for the bacterial community in three of
the four soil samples and were unanimously significant
for the denitrifier community in leaf samples (Table 2).
No significant regressions were found using fungal richness as the independent variable (Table 2).
Discussion
The effect of pH
In our microcosms, pH was found to influence the community structure of soil and leaf bacteria and leaf fungi
within 30 days following the pH manipulation. Such
community changes with pH have also been shown in
vernal pools (Carrino-Kyker & Swanson, 2008; CarrinoKyker et al., 2011) and upland soils (Fierer & Jackson,
2006; Rousk et al., 2010). These compositional changes
could be owing to different microbial taxa having unique
physiological constraints, such as the maintenance of proton motive force, which allows survival within narrow pH
ranges (i.e. pH has a direct effect on the community
structure; Booth, 1985). Another possibility is that pH
influenced microbial community structure indirectly
because pH influences a number of other environmental
characteristics in aquatic systems, such as nutrient availability and redox potential (Stumm & Morgan, 1996). See
Lauber et al. (2009) for further discussion on why environmental pH change could affect microbial community
structure directly or indirectly in soil. The bacterial community was affected in both soil and leaf samples, but
changes to the fungal community were only observed in
leaf samples, suggesting that fungi were more abundant
and more active in the leaf samples than in soil. This is
typical for wetlands where fungi often have a substantial
role in decomposing leaf litter, but can be in low abundance or absent from soils owing to anoxic conditions
(van der Valk, 2006). Further, in the leaf samples, the
effect of pH on fungal community structure was weaker
than for bacterial community structure. Rousk et al.
(2010) also found a weaker influence of pH change on
fungal communities than on bacterial communities in
upland soil and suggested that this result may be owing
to the wide pH optimum for fungal growth (often 5–9
pH units for many fungal species). Improved competitive
abilities of bacteria in response to pH increases could also
limit fungal response to pH changes through direct or
indirect competitive interactions (Rousk et al., 2010).
Regardless of the reason for the community pH response,
it was accompanied by significant differences in microbial
respiration.
The microcosms with more basic pH had significantly
higher CO2 efflux throughout most of the study. We
believe the initial low CO2 efflux values in the pH 7 and
pH 8 microcosms resulted from the microcosm having
the greatest departure from the initial pH of 4.00–4.39
667
Table 2. Summary of simple linear regressions investigating the
relationship between richness and microbial respiration
Sample type
and day
Bacterial richness
MspI
Soil day 30*,†
Soil day 31
Leaf day 30
Leaf day 31*,†
HaeIII
Soil day 30
Soil day 31
Leaf day 30
Leaf day 31
Fungal richness
AluI
Soil day 30
Soil day 31*,†
Leaf day 30
Leaf day 31*,†
HaeIII
Soil day 30
Soil day 31*,†
Leaf day 30*
Leaf day 31
Denitrifier richness
MboI
Soil day 30
Soil day 31*
Leaf day 30
Leaf day 31*
TaqI
Soil day 30
Soil day 31
Leaf day 30*,†
Leaf day 31
Regression equation
R2
P-value
y = 0.112x + 1.624
y = 0.108x + 1.718
y = 0.311x + 30.068
y = 0.0266x + 3.118
0.713
0.513
0.0070
0.0468
< 0.001
< 0.001
0.63
0.205
y = 5.195x 51.038
y = 1.201x + 17.791
y = 1.1913x + 6.531
y = 2.794x 5.300
0.633
0.0418
0.122
0.266
< 0.001
0.232
0.036
0.001
y
y
y
y
=
=
=
=
0.447x + 43.686
0.00517x + 3.439
0.348x + 30.777
0.0110x + 3.398
0.0168
0.00185
0.0422
0.0456
0.451
0.806
0.229
0.211
y
y
y
y
=
=
=
=
0.462x + 25.701
0.00268x + 3.588
0.00952x + 3.390
0.485x + 29.075
0.0180
0.000858
0.0375
0.0859
0.435
0.867
0.258
0.083
y = 0.986x + 48.513
y = 0.00955x + 3.669
y = 1.712x + 51.866
y = 0.0393x + 3.870
0.132
0.0254
0.423
0.326
0.030
0.353
< 0.001
< 0.001
y = 0.726x + 41.256
y = 1.869x + 53.895
y = 0.0571x + 3.990
y = 1.564x + 48.690
0.302
0.270
0.702
0.489
0.311
0.001
< 0.001
< 0.001
Richness was determined from TRFLP profiles and was used as the
independent (x) variable. Microbial respiration was measured as CO2
flux and was used as the dependent (y) variable. The richness determined from different sample types (leaf or soil) and on different dates
(day 30 or day 31) was regressed independently against microbial respiration values. Different TRFLP enzymes were also analyzed separately. Significant regressions (a = 0.05) are shown in bold.
*Indicates a regression where the dependent variable (CO2 flux) was
natural log transformed to meet normality and homogeneity of variances.
†
Indicates a regression where an outlier(s) was removed.
and any CO2 released from microbial respiration and dissolved in the water would be used for buffering via the
carbonate system (Wetzel, 2001). Between day 1 and day
29, CO2 efflux in each pH treatment increased and
decreased in a pattern similar to that of microbial growth
in culture (Madigan et al., 2003). We believe the populations in each microcosm initially experienced exponential
growth, and by day 29, had declined to a size that could
668
(a)
(b)
Fig. 5. Regressions of operational taxonomic unit (OTU) richness and
community respiration for the bacterial community (a) and the
denitrifier community (b) on day 30. Mean respiration values (±
standard error) for two pooled microcosms from the same treatment
(see Materials and methods for details) are shown. The TRFLP
enzymes used for determining richness were MspI for bacteria (a) and
TaqI for denitrifiers (b). The summary of these regressions can be
found in Table 2.
be maintained throughout the remainder of the experiment. By day 29, the CO2 efflux was similar for all the
pH treatments, suggesting that each microcosm had
reached CO2 equilibrium and the microorganisms had
adapted to the new pH. Following the N additions on
day 29, when the soil in the microcosms was disturbed by
our acetylene additions, CO2 efflux was higher in the pH
7 and pH 8 treatments across N addition treatments. This
result was consistent throughout the remainder of the
study and, we believe, reflects the respiration rate of the
adapted microbial community in a buffered aquatic system, which is more reflective of a natural environment.
In natural aquatic habitats, a number of studies have
noted higher microbial respiration at higher pH, perhaps
because of inhibition of microbial activity accompanying
the high levels of aluminum common under acidic conditions (Rao et al., 1984; Dangles et al., 2004; Dorea &
Clarke, 2008). Others have suggested that, in soils, fungal
respiratory activity is highest at low pH, while that of
bacteria is highest under more basic conditions
(Blagodatskaya & Anderson, 1998). Our measurement of
CO2 efflux in the current study encompasses respiration
from a wide range of microorganisms, both prokaryotic
and eukaryotic, thus limiting conclusions about the activities of specific microbial groups. However, significant
regressions were found between CO2 efflux and the richness of certain microbial groups; specifically, both leaf
and soil bacterial richness had a positive relationship with
respiration. Similar to the current study, McGrady-Steed
et al. (1997) and Bell et al. (2005) demonstrated that
increased bacterial richness is correlated with higher CO2
flux in aquatic microcosms, suggesting that increases in
bacterial species traits can have an impact on carbon
cycling in aquatic systems. However, a review by Nielsen
et al. (2011) found that positive relationships between
richness and respiration were more likely to be found in
microcosm studies where 10 species or less were manipulated, but this relationship is less clear in studies including more diverse communities. The relationship between
richness and respiration observed in the current study
might not accurately reflect natural settings, where bacterial richness is typically higher than in a microcosm
where dominant, fast-growing organisms might be
selected for (Nielsen et al., 2011). Interestingly, we found
a negative relationship between denitrifier richness (in
both leaf and soil samples) and CO2 efflux, which, to our
knowledge, is the first documentation of such a relationship. It may be that our observed increases in denitrifier
richness signify increasingly anoxic conditions in our
microcosms and, therefore, a general reduction in CO2
production, which could explain the negative relationship
between denitrifier richness and CO2 efflux. Examining
the mechanism behind this relationship will require additional study, but the significant regressions between bacterial richness and CO2 efflux in our study, coupled with
our finding that the bacterial and fungal community
composition was different between microcosms with different CO2 flux rates, suggest that the difference in
microbial respiration between treatments could be the
result of microbial community structural changes.
Unlike the bacterial and fungal communities, the structure of the nosZ-containing denitrifiers, at least in soil,
was unchanged between pH treatments; and unlike respiration rate, denitrification rate was unchanged between
pH treatments. There is evidence to suggest that denitrifiers perform optimally at the pH of their native environment (Parkin et al., 1985; Cavigelli & Robertson, 2000;
Šimek et al., 2002). Specifically, Šimek et al. (2002)
showed that, in soils where the pH was manipulated, the
activity of denitrifiers was likely to be highest at the natural pH. However, we saw no significant change in denitrification rate between the pH treatments of our
microcosms. We suggest that vernal pool denitrifiers are
metabolically flexible because under natural conditions
they commonly experience seasonal pH fluctuations
between 5 and 8 (Colburn, 2004; Carrino-Kyker & Swanson, 2007). Therefore, denitrifiers in vernal pools are likely
adapted to this fluctuating pH and may maintain high
denitrification rates throughout this range. Consequently,
pH fluctuations may not influence denitrification rate in
vernal pools in ways expected for upland soil denitrifiers.
The effect of NO−3 pulse
Denitrification rate was high in microcosms that
received the NO
3 pulse, but not detected in the absence
of NO
3 , as expected, as denitrification is a facultative
respiratory process. Respiration was also significantly
affected by the N treatment and was higher in the
þNHþ
4 microcosms, compared to the þNO3 microcosms in the higher pH treatments. This suggests a possible nutrient limitation that was overcome by adding
(NH4)2SO4. Unlike microbial processes, the addition of
þ
NO
3 or NH4 did not lead to immediate microbial
community structural changes for any of the groups
examined (bacteria, fungi, and denitrifiers). It is likely
that the lack of community responses to N is the result
of the short time that they were exposed to NO
3 or
NHþ
4 , because evidence from studies on chronic N
inputs has shown effects on both soil bacterial and fungal communities (Demoling et al., 2008; Hassett et al.,
2009; Edwards et al., 2011). However, studies on denitrifier communities have shown no community composition changes in response to NO
3 , both after 14 days of
incubation in the laboratory (Deiglmayr et al., 2006)
and after over 50 years of field fertilization (Enwall
et al., 2005). It is possible we saw no changes in denitrifier community structure because a significant portion
of denitrifiers lacks the nosZ gene (Jones et al., 2008)
and the PCR primers used to amplify nosZ are known
to discriminate against gram-positive bacteria (Throbäck
et al., 2004). In addition, acetylene, which was added to
measure denitrification of our microcosms, actively
inhibits the reduction of N2O to N2 (Yoshinari &
Knowles, 1976) and we profiled the denitrifier community with the gene that codes for N2O reductase, which
could limit our conclusions. However, in the studies by
Deiglmayr et al. (2006) and Enwall et al. (2005)
described above, the narG gene (codes for NO
3 reductase) was targeted, suggesting that the lack of response
to NO
3 might be more biological than methodological;
669
however, additional studies are necessary to fully
understand how NO
3 addition impacts vernal pool
denitrifier communities.
Although we profiled the community shortly after
NO
3 addition, our data suggest that substantial increases
in denitrification rate were not necessarily preceeded by
changes to the community structure of nosZ-containing
denitrifiers in our microcosms. There is conflicting evidence in the literature on whether changes in denitrification are correlated with changes to denitrifier community
structure or abundance. A number of studies have found
no link between function (including measurements of
potential denitrification rate and denitrifying enzyme
activity) and denitrifier community structure or the copy
number of denitrification genes (Rich & Myrold, 2004;
Deiglmayr et al., 2006; Kandeler et al., 2006, 2009; Ma
et al., 2008; Hallin et al., 2009; Attard et al., 2011; Dandie
et al., 2011), while others have documented changes to
activity in response to changes in denitrifier community
composition or gene abundance (Cavigelli & Robertson,
2000; Rich et al., 2003; Dong et al., 2009; Morales et al.,
2010). It has been suggested that the amount of conflicting evidence indicates that denitrifier activity may in
some cases, such as under ideal levels of oxygen and carbon, be influenced by the denitrifying community, while
at other times is predominantly affected by environmental
factors (Wallenstein et al., 2006). Further, the conflicting
reports may be because absolute denitrification rates are
more influenced by the availability of NO
3 in the environment, while overall denitrification efficiency is controlled by the abundance of denitrification genes (e.g.
Hallin et al., 2009; Graham et al., 2010). Therefore, further experimentation is needed to fully understand the
relationship between denitrification and the community
structure and population size of the organisms performing this process. However, it is well known that denitrifiers comprise a taxonomically diverse group of organisms
that include more than 50 genera and are known to use
other respiratory pathways, including aerobic respiration
and sulfate reduction (reviewed in Knowles, 1982;
Philippot & Hallin, 2006). It is plausible that denitrifiers
in our microcosms simply used other electron acceptors
for respiration in the absence of NO
3.
Conclusion
Overall, our results indicate that pH could have an effect
on both microbial community structure and ecosystem
function in vernal pools, especially in leaf samples where
fungi might be more active. Pulses of NO
3 , however, may
have a substantial and immediate impact on function that
is not dependent on a prior change in microbial community structure. Thus, short-term increases in NO
3
670
availability, such as after a rainfall event, might have a different effect on microbial communities in vernal pools
than longer-term environmental changes, such as pH. Our
results lend some support for the idea that different
microbial groups respond differently (in terms of their
structure and function) to different types of environmental change (Balser et al., 2001).
Acknowledgements
We thank Dr Wendy Mahaney, Dr Juan C. López-Gutiérrez,
and Charlotte Hewins for their help with the design and
implementation of this experiment and Dr Darren Bade
at Kent State University for GC access.
References
Attard E, Recous S, Chabb A, De Berranger C, Guillaumaud
N, Labreuches J, Philippot L, Schmid B & Le Roux X
(2011) Soil environmental conditions rather than
denitrifier abundance and diversity drive potential
denitrification after changes in land uses. Glob Change Biol
17: 1975–1989.
Balser TC & Firestone MK (2005) Linking microbial
community composition and soil processes in a California
annual grassland and mixed-conifer forest. Biogeochemistry
73: 395–415.
Balser TC, Kinzig AP & Firestone MK (2001) Linking soil
microbial communities and ecosystem functioning. The
Functional Consequences of Biodiversity (Kinzig AP, Pacala
SW & Tilman D, eds), pp. 265–293. Princeton University
Press, Princeton, NJ.
Bell T, Newman JA, Silverman BW, Turner SL & Lilley AK
(2005) The contribution of species richness and
composition to bacterial services. Nat Lett 436: 1157–1160.
Bell CW, Acosta-Martinez V, McIntyre NE, Cox S, Tissue DT
& Zak JC (2009) Linking microbial community structure
and function to seasonal differences in soil moisture and
temperature in a chihuahuan desert grassland. Microb Ecol
58: 827–842.
Blagodatskaya EV & Anderson TH (1998) Interactive effects of
pH and substrate quality on the fungal-to-bacterial ratio
and QCO2 of microbial communities in forest soils. Soil
Biol Biochem 30: 1269–1274.
Booth IR (1985) Regulation of cytoplasmic pH in bacteria.
Microbiol Rev 49: 359–378.
Brooks RT (2000) Annual and seasonal variation and the
effects of hydroperiod on benthic macroinvertebrates of
seasonal forest (“vernal”) ponds in central Massachusetts,
USA. Wetlands 20: 707–715.
Burke DJ, Martin KJ, Rygiewicz PT & Topa MA (2005)
Ectomycorrhizal fungi identification in single and pooled
root samples: terminal restriction fragment length
polymorphism (TRFLP) and morphotyping compared. Soil
Burke DJ, Kretzer AM, Rygiewicz PT & Topa MA (2006) Soil
bacterial diversity in a loblolly pine plantation: influence of
ectomycorrhizas and fertilization. FEMS Microbiol Ecol 57:
409–419.
Burke DJ, Dunham SM & Kretzer AM (2008) Molecular
analysis of bacterial communities associated with the roots
of Douglas fir (Pseudotsuga menziesii) colonized by different
ectomycorrhizal fungi. FEMS Microbiol Ecol 65: 299–309.
Carrino-Kyker SR & Swanson AK (2007) Seasonal
physicochemical characteristics of thirty Northern Ohio
temporary pools along gradients of GIS-delineated human
land-use. Wetlands 27: 749–760.
Carrino-Kyker SR & Swanson AK (2008) Temporal and
spatial patterns of eukaryotic and bacterial communities
found in Ohio vernal pools as determined by denaturing
gradient gel electrophoresis. Appl Environ Microbiol 74:
2554–2557.
Carrino-Kyker SR, Swanson AK & Burke DJ (2011) Changes
in eukaryotic microbial communities of vernal pools along
an urban-rural land use gradient. Aquat Microb Ecol 62:
13–24.
Cavigelli MA & Robertson GP (2000) The functional
significance of denitrifier community composition in a
terrestrial ecosystem. Ecology 81: 1402–1414.
Chu Z, Jin X, Iwami N & Inamori Y (2007) The effect of
temperature on growth characteristics and competitions of
Microcystis aeruginosa and Oscillatoria mougeotii in a
shallow, eutrophic lake simulator system. Hydrobiologia
581: 217–223.
Colburn EA (2004) Vernal Pools: Natural History and
Conservation. The McDonald and Woodward Publishing
Company, Blacksburg, VA.
Dandie CE, Wertz S, Leclair CL, Goyer C, Burton DL, Patten
CL, Zebarth BJ & Trevors JT (2011) Abundance, diversity
and functional gene expression of denitrifier communities in
adjacent riparian agricultural zones. FEMS Microbiol Ecol
72: 69–82.
Dangles O, Gessner MO, Geurold F & Chauvet E (2004)
Impacts of stream acidification on litter breakdown:
implications for assessing ecosystem functioning. J Appl Ecol
41: 365–378.
Deiglmayr K, Philippot L & Kandeler E (2006) Functional
stability of the nitrate-reducing community in grassland
soils towards high nitrate supply. Soil Biol Biochem 38:
2980–2984.
Demoling F, Nilsson LO & Bååth E (2008) Bacteria and fungal
response to nitrogen fertilization in three coniferous forest
soils. Soil Biol Biochem 40: 370–379.
Dong L, Smith CJ, Papaspyrou S, Stott A, Osborn AM &
Nedwell DB (2009) Changes in benthic denitrification,
nitrate ammonification, and annamox process rates and
nitrate and nitrite reductase gene abundances along an
estuarine gradient (the Colne estuary, United Kingdom).
Appl Environ Microbiol 75: 3171–3179.
Dorea CC & Clarke BA (2008) Effect of aluminium on
microbial respiration. Water Air Soil Pollut 89: 353–358.
Edwards IP, Zak DR, Kellner H, Eisenlord SD & Pregitzer KS
(2011) Simulated atmospheric N deposition alters fungal
community composition and suppresses ligninolytic gene
expression in a northern hardwood forest. PLoS ONE 6:
e20421.
Enwall K, Philippot L & Hallin S (2005) Activity and
composition of the denitrifying bacterial community
respond differently to long-term fertilization. Appl Environ
Microbiol 71: 8335–8343.
Fierer N & Jackson RB (2006) The diversity and biogeography
of soil bacterial communities. P Natl Acad Sci USA 103:
626–631.
Graham DW, Trippett C, Dodds WK, O’Brien JM, Banner
EBK, Head IM, Smith MS, Yang RK & Knapp CW (2010)
Correlations between in situ denitrification activity and
nir-gene abundances in pristine and impacted prairie
streams. Environ Pollut 158: 3225–3229.
Groffman PM, Holland EA, Myrold DD, Robertson GP & Zou
X (1999) Denitrification. Standard Soil Methods for LongTerm Ecological Research (Robertson GP, Coleman DC,
Bledsoe CS & Sollins P, eds), pp. 272–288. Oxford
University Press, New York, NY.
Gutknecht JLM, Goodman RM & Balser TC (2006) Linking
soil process and microbial ecology in freshwater wetland
ecosystems. Plant Soil 289: 17–34.
Hallin S, Jones CM, Schloter M & Philippot L (2009)
Relationship between N-cycling communities and ecosystem
functioning in a 50-year-old fertilization experiment. ISME J
3: 597–605.
Hassett JE, Zak DR, Blackwood CS & Pregitzer KS (2009) Are
basidiomycete laccase gene abundance and composition
related to reduced lignolytic activity under elevated
atmospheric NO3- deposition in a northern hardwood
forest? Microb Ecol 57: 728–739.
Jones CM, Stres B, Rosenquist M & Hallin S (2008)
Phylogenetic analysis of nitrite, nitric oxide, and nitrous
oxide respiratory enzymes reveal a complex evolutionary
history for denitrification. Mol Biol Evol 25: 1955–1966.
Kandeler E, Deiglmayr K, Tscherko D, Bru D & Philippot L
(2006) Abundance of narG, nirS, nirK, and nosZ genes of
denitrifying bacteria during primary successions of a glacier
foreland. Appl Environ Microbiol 72: 5957–5962.
Kandeler E, Brune T, Enowashu E, Dörr N, Guggenberger G,
Lamersdorf N & Philippot L (2009) Response of total and
nitrate-dissimilating bacteria to reduced N deposition in a
spruce forest soil profile. FEMS Microbiol Ecol 67: 444–
454.
Knowles R (1982) Denitrification. Microbiol Rev 46: 46–70.
Kutner MH, Nachtsheim CJ, Neter J & Li W (2005) Applied
Linear Statistical Models. The McGraw-Hill Companies, New
York, NY.
Laposata MM & Dunson WA (1998) Effects of boron and
nitrate on hatching success of amphibian eggs. Arch Environ
Contam Toxicol 35: 615–619.
Lauber CL, Hamady M, Knight R & Fierer N (2009)
Pyrosequencing-based assessment of soil pH as a predictor
671
of soil bacterial community structure at the continental
scale. Appl Environ Microbiol 75: 5111–5120.
Ma WK, Bedard-Haughn A, Siciliano SD & Farrell RE (2008)
Relationship between nitrifier and denitrifier community
composition and abundance in predicting nitrous oxide
emissions from ephemeral wetland soils. Soil Biol Biochem
40: 1114–1123.
Madigan MT, Martinko JM & Parker J, eds (2003) Brock
Biology of Microorganisms. Pearson Education Inc., Upper
Saddle River, NJ.
Martin KJ & Rygiewicz PT (2005) Fungal specific primers
developed for analysis of the ITS region of environmental
DNA extracts. BMC Microbiol 5: 28.
McArthur JV (2006) Microbial Ecology: An Evolutionary
Approach. Elsevier Inc., Burlington, MA.
McCune B & Grace JB (2002) Analysis of Ecological
Eommunities. MjM Software Design, Gleneden Beach, OR.
McGrady-Steed J, Harris PM & Morin PJ (1997) Biodiversity
regulates ecosystem predictability. Nature 390: 162–165.
Menyailo OV, Abraham WR & Conrad R (2010) Tree species
affect atmospheric CH4 oxidation without altering
community composition of soil methanotrophs. Soil Biol
Biochem 42: 101–107.
Moldan B & Černý J, eds (1994) Biogeochemistry of Small
Catchments. John Wiley & Sons Ltd, Chichester.
Morales SE, Cosart T & Holben WE (2010) Bacterial gene
abundances as indicators of greenhouse gas emissions in
soils. ISME J 4: 799–808.
Muyzer G, de Waal EC & Uitterlinden AG (1993) Profiling of
complex microbial populations by denaturing gradient gel
electrophoresis analysis of polymerase chain reactionamplified genes coding for 16S rRNA. Appl Environ
Microbiol 59: 695–700.
Muyzer G, Teske A & Wirsen CP (1995) Phylogenetic
relationships of Thiomicrospira species and their
identification in deep-sea hydrothermal vent samples by
denaturing gradient gel electrophoresis of 16S rDNA
fragments. Arch Microbiol 164: 165–172.
Nielsen UN, Ayres E, Wall DH & Bardgett RD (2011) Soil
biodiversity and carbon cycling: a review and synthesis of
studies examining diversity-function relationships. Eur J Soil
Sci 62: 105–116.
Parkin TB, Sexstone AJ & Tiedje JM (1985) Adaptation of
denitrifying populations to low soil pH. Appl Environ
Microbiol 49: 1053–1056.
Philippot L & Hallin S (2006) Molecular analyses of soil
denitrifying bacteria. Molecular Techniques for Soil,
Rhizosphere and Plant Microorganisms (Cooper JE & Rao JR,
eds), pp. 146–165. CAB International Pub, Cambridge, MA.
Rao SS, Paolini D & Leppard GG (1984) Effects of low-pH
stress on morphology and activity of bacteria from lakes
receiving acid precipitation. Hydrobiologia 114: 115–121.
Rich JJ & Myrold DD (2004) Community composition and
activities of denitrifying bacteria from adjacent agricultural
soil, riparian soil, and creek sediment in Oregon, USA. Soil
672
Rich JJ, Heichen RS, Bottomley PJ, Cromack K & Myrold DD
(2003) Community composition and functioning of
denitrifying bacteria from adjacent meadow and forest soils.
Appl Environ Microbiol 69: 5974–5982.
Rösch C & Bothe H (2005) Improved assessment of
denitrifying, N2-fixing, and total-community bacteria by
terminal restriction fragment length polymorphism analysis
using multiple restriction enzymes. Appl Environ Microbiol
71: 2026–2035.
Rösch C, Mergel A & Bothe H (2002) Biodiversity of
denitrifying and dinitrogen-fixing bacteria in an acid forest
soil. Appl Environ Microbiol 68: 3818–3829.
Rousk J, Bååth E, Brookes PC, Lauber CL, Lozupone C,
Caporaso JG, Knight R & Fierer N (2010) Soil bacterial and
fungal communities across a pH gradient in an arable soil.
ISME J 4: 1340–1351.
Šimek M, Jı́ŝová L & Hopkins DW (2002) What is the socalled optimum pH for denitrification in soil? Soil Biol
Biochem 34: 1227–1234.
Smucker NJ & Vis ML (2011) Acid mine drainage affects the
development and function of epilithic biofilms in streams.
J N Am Benthol Soc 30: 728–738.
Stumm W & Morgan JJ (1996) Aquatic Chemistry, Chemical
Equilibria and Rates in Natural Waters, 3rd edn. John Wiley
& Sons, Inc., New York, NY.
Throbäck IN, Enwall K, Jarvis Å & Hallin S (2004) Reassessing
PCR primers targeting nirS, nirK, and nosZ genes for
community surveys of denitrifying bacteria with DGGE.
FEMS Microbiol Ecol 49: 401–417.
van der Valk AG (2006) The Biology of Freshwater Wetlands.
Oxford University Press, Oxford, UK.
Wallenstein MD, Myrold DD, Firestone M & Voytek M (2006)
Environmental controls on denitrifying communities and
denitrification rates: insights from molecular methods. Ecol
Appl 16: 2143–2152.
Wetzel RG (2001) Limnology: Lake and River Ecosystems, 3rd
edn. Academic Press, San Diego, CA.
Williams DD (2006) The Biology of Temporary Waters. Oxford
University Press, Oxford, UK.
Yoshinari T & Knowles R (1976) Acetylene inhibition of
nitrous oxide reduction by denitrifying bacteria. Biochem
Biophys Res Commun 69: 705–710.
Fig. S1. Nonmetric multidimensional scaling (NMS) ordination graphs showing the bacterial (a,b), fungal (c,d),
and denitrifier (e,f) community structures as determined
by terminal restriction fragment length polymorphism
(TRFLP).
Fig. S2. Nonmetric multidimensional scaling ordination
graphs showing the bacterial (a,b), fungal (c,d), and denitrifier (e,f) community structures as determined by terminal restriction fragment length polymorphism (TRFLP).
Table S1. Measured pH values (mean ± standard deviation) of microcosms at various time points throughout
the study.
Table S2. Multiple comparison tests of CO2 efflux measured on day 30 and day 31 within each pH and N treatment.
Table S3. Average relative abundance in each pH treatment (n = 18 microcosms) for the bacterial terminal
restriction fragments.
Table S4. Average relative abundance in each pH treatment (n = 18 microcosms) for the fungal terminal
restriction fragments.
Table S5. Average relative abundance in each pH treatment (n = 18 microcosms) for the denitrifier terminal
restriction fragments (determined by profiling the nosZ
gene).
Table S6. Average relative abundance in each N treatment
(n = 24 microcosms) for the bacterial terminal restriction
fragments.
(n = 24 microcosms) for the fungal terminal restriction
fragments.
(n = 24 microcosms) for the denitrifier terminal restriction fragments (determined by profiling the nosZ gene).
Please note: Wiley-Blackwell is not responsible for the
content or functionality of any supporting materials supplied by the authors. Any queries (other than missing
material) should be directed to the corresponding author
for the article.
Supporting Information
Additional Supporting Information may be found in the
online version of this article:

The effects of pH change and NO− 3 pulse on microbial community

Transcription

Similar documents