-
Nitrate amendment reduces biofilm biomass and shiftsmicrobial
communities in remote, oligotrophic ponds
Carmella Vizza1,5, Jennifer L. Pechal2,6, M. Eric Benbow3,7,
Jennifer M. Lang4,8, Dominic T. Chaloner1,9,Stuart E. Jones1,10,
and Gary A. Lamberti1,11
1Department of Biological Sciences, University of Notre Dame,
Notre Dame, Indiana 46556 USA2Department of Entomology, Michigan
State University, East Lansing, Michigan 48824 USA3Department of
Entomology, Department of Osteopathic Medical Specialties, Ecology,
Evolutionary Biology, and Behavior Program,
Michigan State University, East Lansing, Michigan 48824
USA4Department of Biology, University of Dayton, Dayton, Ohio 45469
USA
Abstract: Humans have increased the amount of reactive N
available in the environment by over an order of mag-nitude since
the industrial revolution. Most studies have been conducted in
ecosystems with pervasive anthropo-genic nutrient inputs, so little
is understood about how naïve biofilm communities respond to
elevated nutrients.Our nutrient-diffusing substrate (NDS)
experiments, which were conducted in Alaskan freshwater ponds with
verylittle anthropogenic nutrient inputs, suggest that P limits
biofilm photoautotrophs. However, despite low water-column nutrient
concentrations, overall biofilm biomass was not enhanced by the
addition of N or P. Rather,we observed an ~60% biomass reduction
with NO3
– amendment in 15 oligotrophic ponds across 2 y. This
wide-spread biomass reduction was accompanied by changes in
microbial communities, but these trends were not ob-served with
NH4
1 or P amendment. Nonamended communities (i.e., no nutrient
amendment other than lysogenybroth agar) were characterized by
anaerobic heterotrophs and purple nonsulfur bacteria, whereas
NO3
–-amendedcommunities were characterized by aerobic heterotrophs
and facultatively aerobic heterotrophs (e.g., denitrifiers).These
community patterns suggest that NO3
– can strongly affect microbial interactions during biofilm
formationby altering redox conditions. The effect of NO3
– on microbial biomass may be caused by an NO3– toxicity effect
or
competitive shifts in taxa, both of which may shape biofilm
formation and community assembly. Our results revealpossible
consequences for low-NO3
–, aquatic environments after novel exposure to anthropogenic
NO3– inputs,
suggesting that a legacy of anthropogenic NO3– inputs may have
fundamentally changed microbial community
assembly and biogeochemical cycling in aquatic ecosystems.Key
words: nutrient-diffusing substrate, oligotrophic, nitrate
inhibition, biofilms, microbial community composi-tion,
high-throughput sequencing, redox
From 1860 to the present, humans have increased theamount of
reactive N available by more than 10� (Gallo-way et al. 2004). We
have dramatically transformed theglobal N cycle via fossil fuel
combustion and heavy use ofN in agriculture and industry (Galloway
et al. 2008). Thistransformation can have profound effects on
terrestrialand aquatic food webs (Meunier et al. 2016).
Long-termfertilization studies in terrestrial ecosystems have
shownthat N enrichment can strongly alter soil microbial com-
E-mail addresses: [email protected]; [email protected];
[email protected]; 8PGeffen School of Medicine, University of
California, Los Angeles, California.edu; [email protected]
DOI: 10.1086/697897. Received 10 May 2017; Accepted 28 January
2018; PublFreshwater Science. 2018. 37(2):251–263. © 2018 by The
Society for Freshwate
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
munities and processes (Ramírez et al. 2012, Francioli et
al.2016) and lead to biodiversity loss (Isbell et al. 2013), butthe
long-term effects of anthropogenic N inputs on aquaticecosystems
and their microbial communities are less under-stood.
Primary producers in aquatic ecosystems are equallylikely to be
limited by N or P (Francoeur 2001, Elser et al.2007). However,
atmospheric N deposition (Elser et al.2009) may shift nutrient
limitation from primary N limita-
resent address: Department of Medicine/Division of Cardiology,
David90095 USA, [email protected]; [email protected];
10sjones20@nd
ished online 5 April 2018.r Science. 251
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
-
252 | NO3– amendment reduces biofilm biomass C. Vizza et al.
tion to secondary P limitation. In addition, researchers
havedemonstrated that ambient water-column chemistry andN∶P
stoichiometry (e.g., deviations from the Redfield ra-tio) can be
used to predict nutrient limitation (Keck andLepori 2012, Cooper et
al. 2015). Most studies document-ing relationships between water
chemistry and nutrient-limitation patterns were conducted in
aquatic ecosystemsthat have been receiving regular anthropogenic N
and P in-puts for close to a century. Thus, primary producers
mightbe biologically primed to respond to enhanced nutrient in-puts
(Reisinger et al. 2016).
Not all freshwater ecosystems are limited by N or P.Some are
limited by light availability (Karlsson et al. 2009),whereas some
systems fail to show any limitation (Johnsonet al. 2009). A few
investigators have even found that nu-trient addition can inhibit
biofilm growth (reviewed byFrancoeur 2001). Here, inhibition of
biofilms through Nand P amendment occurred in 1.7 and 3.8% of the
studies,respectively (Francoeur 2001). In general, inhibition
pat-terns are so scarce that they fall within the type 1 errorrange
and, thus, are often ignored (Francoeur 2001). Theseinhibition
patterns are usually documented at a single site(e.g., Biggs et al.
1998) or at particular times (e.g., Bernhardtand Likens 2004) and,
therefore, are difficult to recreate. Forexample, Bernhardt and
Likens (2004) documented N inhi-bition in a heterotrophic stream
outside the growing seasononly. Their hypothesis for this pattern
was that, in environ-ments with ample organic C sources, nutrients
could stim-ulate bacterial heterotrophs, which then inhibit
periphytongrowth by outcompeting them for space or other
resources.If inhibition of periphyton is the result of competitive
shiftsinmicrobial taxa, then examining the effects of nutrients
onmicrobial communities is the key to understanding inhibi-tion
patterns in aquatic ecosystems.
We must study remote ecosystems that receive very lit-tle
anthropogenic N and P to understand whether anthro-pogenic nutrient
input can fundamentally change micro-bial community taxonomic
composition and, as a result,biogeochemical cycling in aquatic
ecosystems. Our studywas conducted in Alaskan ponds in the Copper
River Delta(CRD). The CRD in southcentral Alaska comprises
diversewetland pond habitats, distributed along a gradient of
gla-cial and oceanic influences (Vizza et al. 2017b). The CRD
isconsidered to be a low-nutrient system because of its geo-logical
history and the limited anthropogenic influence inthis remote area
(Bryant 1991). The microorganisms inthese ecosystems may be naïve
to elevated nutrient supplybecause they have not been subjected to
long-term anthro-pogenic nutrient loading. Our basic study
objectives wereto: 1) assess the nutrient-limitation status of
these pondsusing nutrient-diffusing substrate (NDS) experiments
and2) identify howmicrobial biofilm communities were affectedby
nutrient amendment based on targeted high-throughputgene amplicon
sequencing.
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
METHODSStudy area
The Copper River in southcentral Alaska is the 8th-largest river
in the USA (Kammerer 1990). The CopperRiver drains a large region
of the Chugach and WrangellMountains into the Gulf of Alaska, and
the river and the sed-iments it deposits have shaped the largest
contiguous wet-land on the Pacific Coast of North America. The CRD
en-compasses about 283,000 ha of wetland pond habitat andsupports
extraordinary biodiversity (Bryant 1991). Withinthe CRD, different
wetland habitats can be distinguishedalong a gradient of glacial
and oceanic influences (Vizzaet al. 2017b). Ponds were created and
modified by theGreat Alaska earthquake in 1964 that elevated the
Deltaby 124 m (Thilenius 1995). Our study ponds (n 5 15),which we
treated as independent replicates, provided a dis-tinct gradient of
habitats differing in biogeochemistry (Ta-ble 1).
Study designWe conducted 2 separate experiments. We designed
the
1st experiment, conducted in 2013 (n5 9 ponds) and 2014(n 5 6
ponds), to test nutrient limitation using 4 differentNDS treatments
(control, N, P, and N1P). The total num-ber of samples for this
experiment was 600 (15 ponds �4 treatments� 10 replicates). We
conducted the 2nd exper-iment in 2014 in the same 9 ponds sampled
the previousyear to test for the effects of NH4
1 compared toNO3– using
10 replicates of 5 different treatments (control, low NH41,
high NH41, low NO3
–, and high NO3–) for a total of 450
samples (9 ponds � 5 treatments � 10 replicates).
NDSsWe used NDSs to assess nutrient limitation in CRD
ponds (Tank et al. 2017). They were constructed from30-mL
plastic cups, which were filled with a 2% lysogenybroth (LB) agar
solution (Novagen; EMD Chemicals Inc.,San Diego, California) and
topped with glass fritted disks.We constructed different treatments
for each of the 2 ex-periments detailed in the study design. For
the nutrient lim-itation experiment, the treatments consisted of
control(CTL; not amended except for LB agar), N (LB 1 0.5 MKNO3), P
(LB 1 0.5 M KH2PO4), and N1P (LB 1 0.5 MKNO3 1 0.5 M KH2PO4). The
2
nd experiment, in whichwe specifically tested for the effects of
N form (NH4
1 orNO3
–) and concentration (high or low), consisted of the fol-lowing
treatments: CTL (LB), low NO3
– (LB 1 0.05 MKNO3), high NO3
– (LB 1 0.5 M KNO3), low NH41 (LB 1
0.05 M NH4Cl), and high NH41 (LB 1 0.5 M NH4Cl).
After a deployment period of 21 to 28 d, we removed sub-strate
disks from ponds, wrapped them in foil, and frozethem until they
could be analyzed for chlorophyll a (Chl a)and ash-free dry mass
(AFDM), or the total amount of
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
-
Table
1.Mean(±SD
)values
forph
ysical
andbiogeochem
ical
variablesat
theCop
perRiver
Delta
(CRD)po
ndssampled
inthesummersof
2013
and2014.W
ater-chemistry
variablesweremeasuredat
thesurfacelayerof
5sitesperpo
nddu
ring
July
andAug
ust(n
510).Allanalytes
werewithindetectionlim
itsexcept
forNO
3–,for
which
allsam-
ples
were<5lgN/L.S
alinitydidno
tvary
withinapo
nd.A
map
ofthepo
ndsalon
gwithdetailedmetho
dsforparameter
measurementwas
publishedby
Vizza
etal.(2017b).
Tem
p5
temperature,S
pC5
specificcond
uctivity,D
O5
dissolvedO
2,D
OC5
dissolvedorganicC,S
RP5
solublereactive
P,T
N5
totalN,T
P5
totalP.
Depth
Daily
Light
SpC
Salin
ity
DO
DOC
NH
41
SRP
TN
TP
Pon
d(m
)temp(7C)
(kilo
lux)
pH(lS/cm
)(PSU
)(m
g/L)
(mg/L)
(lgN/L)
(lgP/L)
(lgN/L)
(lgP/L)
BVN
1.14
±0.02
16.1
±1.7
8.9±18
5.9±0.1
20±1.1
0.01
8.9±0.3
2.7±1.2
1.1±1.0
3.7±1.0
130±62
27±12
BVS
0.86
±0.08
15.5
±1.4
1.8±4.2
6.1±0.2
24±1.1
0.01
6.6±0.3
3.6±0.8
0.9±1.0
3.2±0.9
190±33
34±14
CME
0.83
±0.04
15.4
±1.8
5.4±8.4
5.7±0.1
52±3.1
0.02
2.7±0.7
4.5±0.8
0.8±1.7
2.9±0.8
210±31
18±9
CMW
0.81
±0.04
15.9
±2.3
1.7±3.3
7.7±0.5
47±1.3
0.02
10±0.6
4.5±0.4
0.6±0.6
3.8±1.0
230±32
29±13
EYN
0.52
±0.04
16.6
±2.3
9.3±12
6.3±0.1
9.5±0.3
0.00
7.7±0.5
7.1±0.6
11±2.8
4.8±1.5
270±40
34±8
EYS
0.59
±0.04
17.6
±2.6
14±18
6.6±0.1
8.8±0.2
0.00
8.3±0.4
6.8±0.9
13±5.7
4.8±2.5
280±34
38±13
LIL
0.61
±0.01
15.7
±2.5
6.1±10
7.0±0.1
76±11
0.04
3.7±0.5
4.2±0.5
6.2±0.8
6.9±3.1
130±28
24±7
RHM
0.64
±0.12
13.6
±1.7
7.7±11
7.2±0.1
81±2.4
0.04
3.6±1.0
4.0±0.5
15±5.3
8.6±2.5
120±29
19±8
SCS
0.89
±0.14
16.3
±2.5
11±20
7.4±0.1
53±5.0
0.02
8.2±0.4
2.7±0.4
5.0±1.0
8.7±3.0
110±25
23±7
SME
0.66
±0.02
14.9
±1.2
2.4±4.5
6.3±0.2
44±2.9
0.02
11±0.4
3.8±0.3
0.5±0.5
5.1±3.3
180±32
27±9
SMW
1.04
±0.05
15.6
±1.7
4.7±10
6.7±0.1
70±0.8
0.03
11±0.5
2.9±0.7
0.5±0.9
4.3±2.0
160±44
27±9
STN
0.57
±0.01
19.4
±2.6
13±23
7.6±0.1
42±0.4
0.02
8.5±0.3
8.7±1.6
13±0.9
4.0±0.7
300±98
24±7
STS
0.54
±0.09
19.1
±2.8
13±19
7.6±0.1
63±1.8
0.03
8.1±0.5
5.5±1.7
15±2.6
5.6±2.1
220±89
20±7
TIN
0.60
±0.03
18.3
±2.6
12±21
6.6±0.1
11±0.6
0.00
6.2±1.1
8.0±0.4
14±2.4
5.2±1.0
250±10
31±6
TIS
0.73
±0.04
18.1
±2.7
14±23
6.6±0.1
8.2±0.9
0.00
7.8±0.5
6.4±0.5
11±1.1
4.8±0.9
200±22
27±5
This content downloaded from 129.074.250.206 on June 08, 2018
09:51:28 AMAll use subject to University of Chicago Press Terms and
Conditions (http://www.journals.uchicago.edu/t-and-c).
-
254 | NO3– amendment reduces biofilm biomass C. Vizza et al.
organic matter. Chl a represents the photoautotrophs ofthe
biofilm including both algae and cyanobacteria, whereasAFDM
represents both autotrophs and heterotrophs in thebiofilm.
LB agar promotes colonization of heterotrophic biofilmsbecause
it is supplemented with yeast extract (5 g/L), pep-tone (10 g/L),
and NaCl (10 g/L), whereas the fritted diskpromotes colonization of
autotrophic biofilms (Johnsonet al. 2009). Different substrates,
such as a cellulose sponge,have been used to promote the
heterotrophic community,but our design allowed us to mimic natural
organic sub-strates in these ponds, such as macrophyte stems,
whichcan structurally and chemically support biofilms composedof
microbial autotrophs and heterotrophs (Cattaneo et al.1998, He et
al. 2014). The NDS method used can affectthe nutrient limitation
patterns detected (Capps et al. 2011).For example, Capps et al.
(2011) found slight variations in Nand P colimitation patterns of a
single stream dependingon substrate type and diffusion rates.
Therefore, diffusionrates and ambient chemistry should be reported
when us-ing standardized nutrient limitation methods.
To better understand nutrient release from a controlsubstrate
with LB agar relative to agar-agar, we assessedlaboratory diffusion
rates of dissolved organic C (DOC;Fig. S1A), total N (TN; Fig.
S1B), and total P (TP; Fig. S1C)from these 2 agar types. Our
diffusion rates for LB agar(Appendix S1, Fig. S1A–C) were orders
ofmagnitude lowerthan rates in other studies (Bernhardt and Likens
2004, Ru-genski et al. 2008), but our C, N, and P release rates
tendedto be about anorder ofmagnitude higher for LB agar relativeto
agar-agar. Stoichiometry was similar between the agartypes,
suggesting that biofilms would experience primarilyC limitation
followed by N limitation based on the amend-ments alone (C∶N∶P
after 24 h of LB agar diffusion was35∶11∶1 and that of agar-agar
was 36∶6∶1; Appendix S1,Fig. S1A–C). We also assessed diffusion
rates in high andlow NO3
–-amended LB agar (Appendix S1, Fig. S1A–C) be-cause of the
strong inhibition response exhibited by biofilmson NO3
–-amended substrates.
Chl a and AFDM analysesWithin 60 d of collection, we extracted
Chl a from disks
overnight in 20 mL of 90% buffered acetone. The next day,we used
a fluorometer (TD-700; Turner Designs, San Jose,California; after
Steinman et al. 2017) to measure Chl a in asubsample of the
extract. We estimated total biofilm bio-mass by measuring AFDM
(after Steinman et al. 2017).We air-dried disks and their
respective acetone extracts (in-cluding the subsample used for Chl
a analysis) for a week,and then oven-dried them for at ≥48 h at
607C, weighedthem, and combusted them at 5007C for 4 h. Last, we
re-wetted the disks and dried them at 607C for ≥48 h beforethe
final weighing. We used the difference in mass beforeand after
combustion to estimate AFDM. We report Chl a
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
and AFDM in areal units based on the top surface area ofa
fritted glass disk (3.9 cm2).
Microbial community study designWe used automated ribosomal
intergenic spacing anal-
ysis (ARISA) to generate initial microbial community
fin-gerprints from 1 replicate of each treatment per pond in2013
(Lang et al. 2015). ARISA results demonstrated thatmicrobial
communities on disks varied among nutrienttreatments (Appendix S2,
Fig. S2). Therefore, in 2014, weused a more advanced sequencing
platform (MiSeq; Illu-mina, San Diego, California) to obtain more
detailed infor-mation on microbial taxa by analyzing 1 replicate of
eachtreatment per pond. In total we used samples from 4 treat-ments
(CTL, N, P, and N1P) in the 6 ponds sampled in2013/2014 and 5
treatments from 9 other ponds (CTL,low NO3
–, high NO3–, low NH4
1, high NH41) sampled in
2014 to generate a total of 69 samples. In both years,
westerilized 2-mL centrifuge tubes in a boiling water bathand
rinsed them with 95% ethanol because no autoclavewas available near
these remote field sites.We removed bio-film samples from disks
with a flame-ethanol-sterilized ra-zor blade and placed them in the
2-mL centrifuge tubes.Wehad no access to a –807C freezer, so we
covered the tubeswith sterile glass-fiber filters but left them
uncapped to air-dry for ~48 h to prevent mold growth and to
preserve sam-ples. Two weeks later, they were transported to
MichiganState University and frozen at –807C upon arrival.
Studiescomparing preservation methods indicate this approachwas
sufficient to enable us to identify community differencesbased on
environmental factors (Piggott and Taylor 2003,Lauber et al.
2010).
DNA extractionDNA was extracted from biofilm samples according
to
the manufacturer’s protocol using a PowerBiofilm® DNAisolation
kit (Mo Bio, Carlsbad, California). Sufficient readswere obtained
for only 27 of the 69 samples in the first se-quencing run because
of a combination of low quality(probable inhibition) and low
quantity (0.206–67.0 ng/lL)of DNA products. Therefore, we used a
PowerClean ProDNA clean-up kit (Mo Bio) for the remaining 42
samplesand then sequenced these for a 2nd run. These 2 runs
resultedin sequencing all 24 samples from the nutrient
limitationexperiment (CTL, N, P, and N1P), and 35 samples fromthe
N-form experiment. However, only 5 of the 9 ponds(EYN, EYS, SCS,
STN, and TIN) from this experimenthad all treatments (CTL, low
NO3
–, high NO3–, low
NH41, high NH4
1) represented; therefore, we included onlythose 25 samples (5
treatments� 5 ponds) in analyses. Eventhough we conducted 2
different sequencing runs with Illu-minaMiSeq,
-
Volume 37 June 2018 | 255
16S ribosomal RNA (rRNA) gene ampliconhigh-throughput
sequencing
Targeting the 16S rRNA gene allowed us to gather phy-logenetic
information about Bacteria and Archaea. AfterDNA extraction, we
quantified the DNA using a Quant-iT dsDNA HS Assay kit and a Qubit
2.0 (Thermo Fisher,Grand Island, New York) and then stored all
samples at2807C. Illumina MiSeq 16S library construction (2 �
250base pair [bp] paired-end reads) and sequencing was per-formed
in the Michigan State University Genomics CoreFacility with a
modified version of the protocol adaptedfor the Illumina MiSeq
described by Pechal and Benbow(2016). Briefly, V4 regions of the
16S rRNA gene ampliconregion were amplified with region-specific
primers that in-clude Illumina flowcell adapter sequences (515f [50
GTGCCAGCMGCCGCGGTAA] and 806r [50 GGACTACHVGGGTWTCTAAT]) (Caporaso
et al. 2010). All sequenc-ing data were curated using the mothur
software package(version 1.37; https://www.mothur.org/) and the
proceduredetailed at https://www.mothur.org/wiki/MiSeq_SOP(Kozich
et al. 2013). Sequences were classified against theSILVA (version
123) reference taxonomy (Pruesse et al.2007). We assessed the error
rate of our sequences (7.25 �10–5) using the mock community
described by Kozich et al.(2013). We then performed rarefaction to
ensure an evensequence depth of 1000 sequences/sample
subsampled1000�; the range in coverage of these rarefied
sequenceswas 0.879 to 0.996, which indicates sufficient sampling
ofthe microbial communities. Sequence files for all samplesused in
this study are deposited in the Sequence Read Ar-chive at the EMBL
EuropeanNucleotide Archive (ENA; http://www.ebi.ac.uk/ena):
PRJEB19927.
Statistical analysesWe converted the raw Chl a and AFDM data to
average
response ratios (Francoeur 2001) at the site level per
pond,which resulted in 5 response ratios per treatment per
pond.Response ratios were calculated by averaging the 2 repli-cates
per treatment at each deployment site within a pond(if applicable)
and then dividing the average nutrient treat-ments by the average
CTL treatments. For the
nutrient-limitationexperiment,weusedablockedanalysisof
variance(ANOVA) design where either Chl a or AFDM response ra-tio
was the response variable, pond was a blocking variable,and
treatment (N,N1P, P)was the factor of interest.Withinthe ANOVA
design, we tested 3 potential data distributions(normal,
log-normal, and gamma) to determine which dis-tribution had the
lowest Akaike’s Information Criterion valueand, therefore, was the
best distribution to model each re-sponse variable (Burnham and
Anderson 2002). We used alog-normal distribution for the Chl a
response ratios of thenutrient-limitation experiment and a gamma
distributionfor the AFDM response ratios. For the N-form
experiment,we also used a blocked ANOVA design where either Chl a
or
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
AFDM response ratio was the response variable, pond was
ablocking variable, and form of N (NH4
1 or NO3–) and con-
centration (0.05 or 0.5 M) were the factors of interest.
Weincluded an interaction between N form and concentration.We used
a gamma distribution for the Chl a response ratiosfor the N-form
experiment, and a normal distribution for theAFDM response ratios.
We performed post hoc pairwisecomparisons on all 4 ANOVAs (i.e.,
Tukey’s Honestly Sig-nificant Difference [HSD]; Zar 2010). For all
ANOVAs andTukey’s HSD tests, we set a 5 0.05.
To examine the effects of nutrient amendment on mi-crobial
community diversity, we calculated operational tax-onomic unit
(OTU) richness and Shannon entropy, an in-dex that accounts for
both species richness and evenness(Jost 2007). We then performed 2
blocked ANOVAs (1 perexperiment) on OTU richness, which was
log10(x)-trans-formed to ensure normality, with treatment as the
factorof interest and pond as a blocking variable. We used 2
sim-ilar blocked ANOVAs to test Shannon entropy.
To analyze microbial community data, we calculatedpairwise
Bray–Curtis dissimilarity indices for the relativeabundance data
from the 2 experiments separately. To vi-sualize patterns in
microbial community composition inour experiments, we used
principal coordinates analyses(PCoAs) to ordinate microbial
communities. We then eval-uated whether our treatments had
statistically significant ef-fects onmicrobial community
composition of biofilms usingpermutational multivariate analyses of
variance (PERMA-NOVA; Anderson and Walsh 2013). All PERMANOVAmodels
included pond and sequencing run as blocking var-iables, and we set
a 5 0.05.
Last, we used indicator species analysis as a heuristictool to
identify which OTUs were representative of NO3
–
and CTL treatments because these 2 treatments exhibitedthe
greatest difference in biomass and microbial communitycomposition.
Indicator species are those that are stronglyassociated with a
particular habitat (Carignan and Villard2002) or that can be used
to reveal evidence for the effectof environmental changes (McGeoch
1998). To identifywhich OTUs were the best indicator “species” for
eachtreatment, we calculated indicator value indices (Cáceresand
Legendre 2009) for the 1090 OTUs found on the CTLand high-NO3
– treatments from both experiments. We pre-sent only the subset
for which the unadjusted p-value was≤0.01. All statistical analyses
were conducted in the R soft-ware environment using the base,
vegan, and indicspeciespackages (version 3.3.0; R Project for
Statistical Computing,Vienna, Austria).
RESULTSFor the nutrient-limitation experiment, biofilm
photo-
autotrophs (Chl a) and total biomass (AFDM) respondeddifferently
to N and P (Fig. 1A, B, Table 2). PO4
3– (P, N1P)had a positive effect, and approximately doubled the
amount
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
-
256 | NO3– amendment reduces biofilm biomass C. Vizza et al.
of Chl a relative to the CTL substrates, whereas NO3– alone
had relatively little effect (Fig. 1A). In contrast, NO3–
re-
duced biofilm biomass by 60%, but P had relatively little
ef-fect (Fig. 1B). N1P substrates had an intermediate amountof
biomass relative to the N and P treatments (Fig. 1B).
For the N-form experiment, biofilm photoautotrophswere slightly
reduced in the presence of high N concentra-tions, whereas total
biofilm biomass was drastically lowerfor NO3
– amendments at both concentrations (Fig. 2A, B,Table 2). Both
treatments with lower concentrations ofNH4
1 and NO3– had approximately the same amount of
Chl a as CTL substrates, whereas the higher concentrationsof
NH4
1 and NO3– had ~80 and 60% of CTL Chl a, respec-
tively (Fig. 2A). Biofilm biomass exhibited a 40 and 60%
re-duction in the presence of low- and high-NO3
– treatments,respectively. In contrast, NH4
1 treatments had approxi-mately the same AFDM as the CTL
substrates (Fig. 2B).
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
Biofilm a diversity did not respond to N or P. The num-ber of
observed microbial OTUs ranged from 19 to 157(mean ± SD, 60 ± 37)
in the nutrient-limitation experimentand from 16 to 80 in the
N-form experiment (38 ± 17).However, nutrient amendments and pond
did not signifi-cantly affect the number of OTUs present for either
exper-iment (p ≥ 0.17). In addition, species richness/evenness
wassimilar between experiments (Shannon entropy for nutri-ent
limitation: 2.1 ± 0.6; N form: 1.9 ± 0.6). Treatmenthad no effect
on species richness/evenness for either exper-iment (p ≥ 0.16), but
pondwas a significant factor in the nu-trient-limitation experiment
(p 5 0.042).
In contrast to a diversity, biofilm microbial
communitycomposition responded differently to N and P. In
thenutrient-limitation experiment, N-amended communitieswere the
most different from CTL communities, whereasP-amended communities
grouped closer to CTL treatments(Fig. 3A). N1P-amended communities
were clustered be-tween the communities associated with N and P
treatments(Fig. 3A). Only treatment had a significant effect on
biofilmmicrobial communities (PERMANOVA, R2 5 0.57, p 50.001; Fig.
3A). Pond only weakly influenced microbial com-munities (R2 5 0.14,
p5 0.08), and sequencing run was notsignificant (R2 5 0.03, p 5
0.18).
Biofilm microbial community composition also re-sponded
differently to NH4
1 and NO3–. The CTL commu-
nities grouped with the low- and high-NH41-treated com-
munities, and these communities differed from thosegrown at both
concentrations of NO3
– (Fig. 3B). Treatmentand pond had significant effects on the
biofilm microbialcommunities (treatment: R2 5 0.38, p 5 0.001;
pond: R2 50.24, p 5 0.002), whereas sequencing run had a small
andnonsignificant effect (R2 5 0.04, p 5 0.09).
The CTL and high NO3– treatments differed substan-
tially in biomass and microbial community composition,so we
identified OTUs that were associated with these dif-ferences. A
total of 8 and 6 indicator OTUs for the CTLand high NO3
– treatments, respectively, were diagnostic(Table 3). In
general, CTL indicators tended to be anaer-obic chemoorganotrophs
or phototrophs, whereas NO3
–
indicators were aerobic or facultatively aerobic
chemo-organotrophs with an ability to reduce NO3
– (Table 4).
DISCUSSIONOur study suggests that nutrients can have strong
pos-
itive and negative effects on microbial biofilms in low-nutrient
aquatic ecosystems. Pond photoautotrophs (mea-sured as Chl a)
probably were limited by P, but total biofilmbiomass (measured as
AFDM) did not increase in the pre-sence of N or P, but rather
experienced a 60% reductionwith the addition of NO3
–. This reduction in biomass withNO3
– addition was observed across 15 remote, oligotrophicponds
differing in biogeochemistry across 2 y, a result notobserved with
NH4
1 addition. NO3– amendment shifted het-
Figure 1. Mean (±95% CI) response ratios of chlorophyll a(Chl a)
(A) and ash-free dry mass (AFDM) (B) on nutrient-amended substrates
(N, P, and N1P) relative to the control inthe 15 study ponds in
2013 (n 5 9) and 2014 (n 5 6). Responseratios of 1.0 indicate equal
growth relative to the lysogenybroth agar control. Bars with the
same lowercase letters are notsignificantly different (p >
0.05).
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
-
Volume 37 June 2018 | 257
erotrophic microbial composition from predominantly an-aerobic
to aerobic with many of the aerobic taxa capable ofusing both O2
and NO3
– as electron acceptors. These com-munity patterns suggest that
NO3
– probably has a signifi-cant effect on microbial interactions
during biofilm forma-tion, at least in ecosystems that receive
little geologic oranthropogenic inputs of NO3
–.P was the primary nutrient limiting photoautotroph
growth in CRD biofilms as shown by a doubling of Chl a,a widely
used surrogate for algal and cyanobacterial bio-mass (Wetzel and
Likens 2000). P limitation of photoauto-troph growth is common in
both lakes and streams (Elseret al. 1990, Francoeur 2001), and
water nutrient concentra-tions and N∶P stoichiometry are generally
consideredgood predictors of nutrient limitation (Keck and
Lepori2012). In our study, water-column TN∶TP molar ratioswere ~17,
which suggests that the ponds could be on theverge of P limitation
because the ratio is higher than theRedfield N∶P ratio of 16∶1,
which is considered an opti-mal nutrient ratio for oceanic seston
(Redfield 1958). Incontrast, Kahlert (1998) found that periphyton
N∶P > 32indicates P limitation, which suggests that periphyton
as-similating nutrients from the water column in the CRDcould be
limited by N instead of P. Nonetheless, we ob-served widespread P
limitation of primary producers acrossponds with P amendments
significantly increasing Chl a.
In contrast to the photoautotrophs, total biofilmbiomasswas not
enhanced by nutrient addition. Instead, AFDMwasstrongly reduced in
the presence of NO3
–. Results of somealgal studies tend to show similar trends for
Chl a andAFDM (e.g., Wyatt et al. 2010), but Lang et al. (2012)
foundthat N appears to be more limiting for photoautotrophs
thantotal biofilm biomass. In addition, different
nutrient-limitationpatterns for fungi and algae have been
identified when usingwood substrates (Tank andDodds 2003).We also
found dif-
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
ferent response patterns between total biofilm biomass andthe
photoautotrophic components of biofilms on glass frit-ted disks
amended with LB agar. The strong decrease in bio-mass with NO3
– amendment was unexpected given that in-hibition patterns
(i.e., reduction in Chl a or AFDM relativeto the CTL) are rare and
often difficult to reproduce in timeand space (Francoeur 2001, Tank
and Dodds 2003, Bern-hardt and Likens 2004, Sanderson et al. 2009).
We are thefirst investigators to document strong NO3
– inhibition oftotal biofilm biomass across multiple sites and
years.
NH41 did not limit or reduce biofilm biomass. Differen-
tial response to N form usually is expected because NH41 is
energetically less expensive to assimilate than NO3– (Von
Schiller et al. 2007). Nevertheless, some investigators
haveshown that biofilm response does not differ in response
toNO3
– or NH41 (Hoellein et al. 2010), whereas biofilms of
midwestern rivers (USA) responded more to NO3– than
to NH41 (Reisinger et al. 2016). An explanation is that or-
ganisms adapt to the most common N form. For example,Reisinger
et al. (2016) hypothesized that positive biofilm re-sponses to
NO3
– amendment in agriculturally influencedstreams are related to
microbial acclimation to elevatedNO3
– levels from fertilizer runoff. In ecosystems where nu-trient
concentrations are very low, differences in assimi-lation are
unlikely to cause strong reductions in biofilmcaused by NO3
– amendment. However, in the CRD, water-column NO3
– levels were very low (
-
258 | NO3– amendment reduces biofilm biomass C. Vizza et al.
heterotrophs (e.g., Desulfovibrio and Paludibacter),
whichsuggests that biofilms were thick enough to reduce O2.
Forexample, Desulfovibrio is a genus known to use SO4
2– asan alternative electron acceptor (Heidelberg et al. 2004).
Inaddition, purple nonsulfur bacteria (e.g., Rhodoblastus
andRhodocyclus) were also abundant in the CTL communitiesand have
extremely flexible metabolisms; they can be pho-toorganotrophic
(i.e., use light for energy and organic com-pounds as a source for
C and electrons), photolithotrophic(i.e., use light for energy, CO2
as a C source, andH2 or otherinorganic compounds as electron
donors), or chemoorgan-otrophic (i.e., use organic compounds for
sources of energy,C, and electrons) in dark, oxic conditions
(Madigan et al.2014). In contrast, NO3
–-amended communitieswere char-acterized by aerobic heterotrophs
(e.g., Janthinobacterium),probably because the biofilms were not as
thick, but alsoby facultative aerobic heterotrophs (e.g.,
Microvirgula andPseudomonas) that generate energy via fermentation
pro-
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
cesses or by using NO3– as an alternative electron acceptor
when O2 is not present (i.e., denitrification). For
example,Microvirgula aerodenitrificans has the unique ability to
si-multaneously use both O2 and NO3
– as electron acceptorsduring respiration (Patureau et al.
1998). Organisms foundon CTL substrates had different metabolisms
than thoseon NO3
–-amended substrates, so differences in redox con-ditions
probably contributed to these divergent microbialcommunities. These
drastic shifts in microbial communi-ties occurred on an artificial
medium supplemented withorganic C and other nutrients. Therefore,
direct extrapola-tion of these results or those of any NDS
experiment to how
Figure 3. Principal coordinates analysis (PCoA) plots basedon a
Bray–Curtis relative abundance distance measure on 16Sribosomal RNA
gene amplicon sequencing data from the nutrient-limitation
experiment where treatments consisted of control(CTL), N, P, and
N1P (A) and the N-form experiment wheretreatments consisted of CTL,
low NH4
1, high NH41, low NO3
–,and high NO3
– (B). Each dot represents a nutrient-diffusingsubstrate
disk.
Figure 2. Mean (±95% CI) response ratios of chlorophyll a(Chl a)
(A) and ash-free dry mass (AFDM) (B) on nutrient-amended substrates
(low NH4
1, high NH41, low NO3
–, highNO3
–) relative to the control in 9 ponds in 2014. Response ra-tios
of 1.0 indicate equal growth relative to the lysogeny brothagar
control. Bars with the same lowercase letters are not
sig-nificantly different (p > 0.05).
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
-
Volume 37 June 2018 | 259
a natural system would respond would be difficult
withoutconducting an ecosystem-scale manipulation.
NO3– could alter redox conditions in ecosystems with
anaerobic conditions (D’Angelo and Reddy 1999) and rela-tively
low NO3
– inputs from surrounding geology, agricul-
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
tural runoff, and urban wastewater. Most NDS studies havebeen
conducted in well-aerated streams rather than inlakes and wetlands,
which are more likely to experienceanaerobic conditions because of
stagnant water. A biofilmnutrient-limitation survey conducted
across Great Lakes
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
Table 4. Taxonomic and metabolic information for the indicator
operational taxonomic units (OTUs) from indicator species
analysison the control (CTL) and high NO3
– nutrient-diffusing substrate treatments listed in Table 3
including class and the ability to reduceNO3
–. Each taxon is identified to the lowest level possible. UC
stands for unclassified and is used to mark different OTUs
sharingthe same taxonomic group. Taxa that are nonsulfur purple
bacteria, which have flexible metabolisms, are marked with an
asterisk (*).References used to compile this table are listed in
Appendix S4.
Treatment Taxon Class MetabolismReducesNO3
–?
CTL Desulfovibrio Deltaproteobacteria Obligate anaerobic
chemoorganotrophs No
Paludibacter Bacteroidia Obligate anaerobic chemoorganotrophs
No
Phaeospirillum* Alphaproteobacteria Anoxygenic photoorganotrophs
No
Rhodoblastus* Alphaproteobacteria Anoxygenic photoorganotrophs
No
Rhodocyclus* Betaproteobacteria Anoxygenic photoorganotrophs
No
Clostridiaceae Firmicutes Obligate anaerobic chemoorganotrophsor
chemolithotrophs
Some
Veillonellaceae Firmicutes Obligate anaerobic chemoorganotrophs
Some
Bacteroidales Bacteroidia Obligate anaerobic chemoorganotrophs
Some
High NO3– Janthinobacterium Betaproteobacteria Obligate aerobic
chemoorganotrophs Yes
Microvirgula Betaproteobacteria Facultative aerobic
chemoorganotrophs Yes
Paenibacillus Firmicutes Aerotolerant anaerobic
chemoorganotrophs Yes
Pseudomonas Gammaproteobacteria Obligate or facultative aerobic
chemoorganotrophs Some
EnterobacteriaceaeUC1 and 2
Gammaproteobacteria Facultative aerobic chemoorganotrophs
Most
Table 3. Indicator species analysis on the control (CTL) and
high NO3– nutrient-diffusing substrate treatments from 11 ponds
(p ≤ 0.01 for all indicator taxa). Each taxon is identified to
the lowest level possible. UC (unclassified) is used to mark
differentoperational taxonomic units (OTUs) sharing the same
taxonomic group. An indicator value is the product of components A
and B.Component A is the probability that a sample belongs to the
treatment group given that the OTU has been found, whereascomponent
B is the probability of finding the OTU in samples belonging to the
treatment group. Mean (±SD) sequence reads aregiven for each
indicator OTU by treatment.
Treatment Taxon A B Indicator value Reads (CTL) Reads (NO3–)
CTL Clostridiaceae 0.91 1.00 0.96 16 ± 18 1 ± 5
Rhodocyclus 0.99 0.91 0.95 270 ± 220 2 ± 4
Paludibacter 1.00 0.73 0.85 4 ± 6 0 ± 0
Desulfovibrio 1.00 0.73 0.85 54 ± 86 0 ± 0
Phaeospirillum 0.99 0.73 0.85 25 ± 64 0 ± 0
Bacteroidales 0.99 0.73 0.85 37 ± 53 1 ± 1
Rhodoblastus 0.98 0.64 0.79 4 ± 6 0 ± 0
Veillonellaceae 1.00 0.55 0.74 1 ± 2 0 ± 0
High NO3– Enterobacteriaceae UC1 0.98 1.00 0.99 3 ± 5 170 ±
100
Enterobacteriaceae UC2 0.96 1.00 0.98 5 ± 9 120 ± 180
Pseudomonas 0.95 1.00 0.98 8 ± 14 170 ± 160
Microvirgula 0.98 0.82 0.90 4 ± 5 170 ± 130
Paenibacillus 0.98 0.73 0.84 0 ± 1 9 ± 9
Janthinobacterium 0.83 0.82 0.83 2 ± 3 9 ± 8
-
260 | NO3– amendment reduces biofilm biomass C. Vizza et al.
coastal wetlands demonstrated increased biomass in re-sponse to
NO3
– (Cooper et al. 2015), but the communitiesof these wetlands
with their high NO3
– (97 ± 220 lg/L) andTN (920 ± 710 lg/L) concentrations may have
been adaptedto NO3
– inputs. In ecosystems with relatively little anthro-pogenic
NO3
– input like the CRD, NO3– amendment could
provide a novel substrate that alters redox conditions in
an-aerobic biofilms. If thick biofilms experience O2 depletionand
are given a novel electron acceptor, such as NO3
–,denitrifiers could outcompete organisms using other
alter-native electron acceptors because Mn41, Fe31, SO4
2–, andCO2 are less energetically profitable than NO3
– (StummandMorgan 1996). This advantagemay explain why we
ob-served a shift in microbial communities on NO3
–-amendedsubstrates toward organisms with the ability to
reduceNO3
–.The shift in redox conditions can explain the observed
patterns in microbial community composition, but the re-duction
in biofilm biomass caused byNO3
– ismore difficultto interpret. Likens et al. (1970) suggested
that high con-centrations of NO3
– may be toxic to certain bacterial spe-cies, but did not offer
a mechanism or toxicity threshold.NO3
– is used as a preservative to reduce the growth ofClostridium
botulinum and other microorganisms (Roberts1975), and NO2
– can retard lipid oxidation (Gray et al.1981). Reduction of
NO3
– to NO2– in these pond biofilms
could inhibit microbial growth. We did see a larger reduc-tion
in biomass at higher concentrations of NO3
– (>60%at 0.5 M), but still observed a 40% reduction in
biofilmbiomass at the log-lower concentration (0.05 M). The
Ndiffusion rate of the high NO3
– (0.1 mg/h) and low NO3–
(0.01mg/h) substrates early in deployment wasmuch lowerthan for
N substrates used in other studies with comparableconcentrations
(Bernhardt and Likens 2004: 10 mg/h,Rugenski et al. 2008: 0.1–1
mg/h). However, NO2
– stillcould have accumulated in the thick CRD biofilms at
highenough concentrations to approach toxicity.
An alternative way to explain the reduction in total bio-film
biomass is that NO3
– alters microbial interactions byselecting for taxa that
produce natural antimicrobial com-pounds. Thus, the reduction in
biomass could be caused bycertain bacterial taxa spending their
energy on the produc-tion of toxins instead of growth. The toxins,
in turn, couldreduce the growth of other microbial competitors.
Manyof the indicator OTUs for the NO3
–-amended substrateshave been documented to produce antibiotics.
For exam-ple, Janthinobacterium spp., an indicator of our NO3
–-amended communities, produces a violet pigment with
an-timicrobial properties called violacein (Pantanella et al.2007,
Kim et al. 2012), whereas Janthinobacterium lividumproduces the
antibiotic prodigiosin (Schloss et al. 2010). Astrain of
Paenibacillus, another indicator of our NO3
–-amended communities can produce polymyxin E1, an anti-biotic
active against Gram-negative bacteria, and 2983-Da,
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
an unknown antibiotic active against Gram-positive bacte-ria (He
et al. 2007).We did not find similar evidence of CTLOTUs producing
antimicrobial compounds in the litera-ture. We cannot verify
whether our related OTUs produceantibiotics without culturing these
strains in the laboratoryand testing them under similar conditions
to those in theexperiment, but their potential antimicrobial
propertiescould help explain the reduction in total biofilm
biomass.
ConclusionsWe demonstrated that NO3
– can greatly reduce biofilmbiomass and strongly alter microbial
community composi-tion in low-nutrient environments. The reduced
biomassobserved on NO3
–-amended substrates may be caused byNO3
– toxicity or shifts in competitive advantages amongtaxa, which
affect biofilm formation and community as-sembly. We suggest that
other aquatic ecosystems that re-ceive little geologic or
anthropogenic NO3
– input may ex-hibit strong changes in microbial community
structure andpotentially function. Future researchers should test
whethera legacyofanthropogenicNO3
– inputs fundamentallychangesmicrobial community assembly and
biogeochemical cyclingin aquatic ecosystems.
ACKNOWLEDGEMENTSAuthor contributions: CV, DTC, and GAL designed
the study.
CV performed the fieldwork and water chemistry, Chl a, andAFDM
analyses. JML conducted the ARISA analyses. JLP and MEBhelped with
IlluminaMiSeq analyses. JML, JLP, MEB, and SEJ pro-vided guidance
to CV on the microbial data analyses. CV wrote themanuscript with
significant input on the manuscript’s directionfrom SEJ and all
authors providing editorial advice.
We thank the Cordova Ranger District of the US Departmentof
Agriculture, Forest Service for providing field and logistical
sup-port, particularly Deyna Kuntzsch, Andrew Morin, Sean
Meade,Luca Adelfio, and Ken Hodges, without whom this work on
theCRD would not have been possible. We also thank Gordie Reevesof
the Pacific Northwest Research Station for his leadership
anddirection in the extensive research being conducted on the
CRD.Mike Brueseke, Melanie Runkle, Josephine Chau, and Julia
Hartassisted with analyses of water chemistry, Chl a, and AFDM.
Dayna(Smith) Evans and Julia Hart helped with summer field and
labo-ratory work in 2013 and 2014, respectively. The Center for
Envi-ronmental Science and Technology (CEST) at University of
NotreDame (UND) provided instrumentation and analytical
assistancefor the chemical analyses. Ursula Mahl helped with
chemical anal-yses on the Lachat. Funding was provided by the USDA
ForestService, the Pacific Northwest Research Station, the National
Fishand Wildlife Foundation, UND, and the National Science
Foun-dation Graduate Research Fellowship Program. We also
thankmembers of the Jones laboratory and the Lamberti laboratory
atUND and Jen Tank for their feedback on the manuscript. We
aregrateful for the comments of 2 anonymous referees and
AssociateEditor Antonia Liess, all of whom made suggestions that
improvedthe manuscript.
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
-
Volume 37 June 2018 | 261
LITERATURE CITEDAnderson, M. J., and D. C. I. Walsh. 2013.
PERMANOVA,
ANOSIM, and the Mantel test in the face of
heterogeneousdispersions: what null hypothesis are you testing?
EcologicalMonographs 83:557–574.
Bernhardt, E. S., and G. E. Likens. 2004. Controls on
periphytonbiomass in heterotrophic streams. Freshwater Biology
49:14–27.
Biggs, B. J. F., C. Kilroy, and R. L. Lowe. 1998. Periphyton
devel-opment in three valley segments of a New Zealand
grasslandriver: test of a habitat matrix conceptual model within a
catch-ment. Archiv für Hydrobiologie 143:147–177.
Bryant, M. D. 1991. The Copper River Delta pulse study: an
inter-disciplinary survey of aquatic habitats. General Technical
ReportPNW-GTR-282. PacificNorthwest Research Station, Forest
Ser-vice, US Department of Agriculture, Portland, Oregon.
(Avail-able from:
https://hdl.handle.net/2027/umn.31951d029749523?urlappendp%3Bseqp3)
Burnham, K. P., and D. R. Anderson. 2002. Model selection
andmulti-model inference: a practical information-theoretic
ap-proach. 2nd edition. Springer–Verlag, New York.
Cáceres, M. D., and P. Legendre. 2009. Associations between
spe-cies and groups of sites: indices and statistical inference.
Ecol-ogy 90:3566–3574.
Caporaso, J. G., J. Kuczynski, J. Stombaugh, K. Bittinger, F.
D.Bushman, E. K. Costello, N. Fierer, A. G. Peña, J. K. Goodrich,J.
I. Gordon, G. A. Huttley, S. T. Kelley, D. Knights, J. E. Koenig,R.
E. Ley, C. A. Lozupone, D. McDonald, B. D. Muegge, M.Pirrung, J.
Reeder, J. R. Sevinsky, P. J. Turnbaugh, W. A. Wal-ters, J.
Widmann, T. Yatsunenko, J. Zaneveld, and R. Knight.2010. QIIME
allows analysis of high-throughput communitysequencing data. Nature
Methods 7:335–336.
Capps, K. A., M. T. Booth, S. M. Collins, M. A. Davison, J.
M.Moslemi, R. W. El-Sabaawi, J. L. Simonis, and A. S. Flecker.2011.
Nutrient diffusing substrata: a field comparison of com-monly used
methods to assess nutrient limitation. Journal ofthe North American
Benthological Society 30:522–532.
Carignan, V., and M.-A. Villard. 2002. Selecting indicator
speciesto monitor ecological integrity: a review. Environmental
Mon-itoring and Assessment 78:45–61.
Cattaneo, A., G. Galanti, S. Gentinetta, and A. Susana. 1998.
Epi-phytic algae andmacroinvertebrates on submerged and
floating-leaved macrophytes in an Italian lake. Freshwater Biology
39:725–740.
Cooper, M. J., G. M. Costello, S. N. Francoeur, and G.
A.Lamberti. 2015. Nitrogen limitation of algal biofilms in
coastalwetlands of Lakes Michigan and Huron. Freshwater
Science35:25–40.
D’Angelo, E. M., and K. R. Reddy. 1999. Regulators of
heterotro-phic microbial potentials in wetland soils. Soil Biology
andBiochemistry 31:815–830.
Elser, J. J., T. Andersen, J. S. Baron, A.-K. Bergström, M.
Jansson,M. Kyle, K. R. Nydick, L. Steger, and D. O. Hessen. 2009.
Shiftsin lake N∶P stoichiometry and nutrient limitation driven
byatmospheric nitrogen deposition. Science 326:835–837.
Elser, J. J., M. E. S. Bracken, E. E. Cleland, D. S. Gruner, W.
S.Harpole, H. Hillebrand, J. T. Ngai, E. W. Seabloom, J. B.Shurin,
and J. E. Smith. 2007. Global analysis of nitrogenand phosphorus
limitation of primary producers in freshwa-
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
ter, marine and terrestrial ecosystems. Ecology Letters
10:1135–1142.
Elser, J. J., E. R. Marzolf, and C. R. Goldman. 1990.
Phosphorusand nitrogen limitation of phytoplankton growth in the
fresh-waters of North America: a review and critique of
experimen-tal enrichments. Canadian Journal of Fisheries and
AquaticSciences 47:1468–1477.
Francioli, D., E. Schulz, G. Lentendu, T. Wubet, F. Buscot, and
T.Reitz. 2016. Mineral vs. organic amendments: microbial com-munity
structure, activity and abundance of agriculturally rel-evant
microbes are driven by long-term fertilization strategies.Frontiers
in Microbiology 7. (Available from:
https://doi.org/10.3389/fmicb.2016.01446)
Francoeur, S. N. 2001. Meta-analysis of lotic nutrient
amendmentexperiments: detecting and quantifying subtle responses.
Jour-nal of the North American Benthological Society
20:358–368.
Galloway, J. N., F. J. Dentener, D. G. Capone, E. W. Boyer, R.
W.Howarth, S. P. Seitzinger, G. P. Asner, C. C. Cleveland, P.
A.Green, E. A. Holland, D. M. Karl, A. F. Michaels, J. H. Porter,A.
R. Townsend, and C. J. Vörösmarty. 2004. Nitrogen cycles:past,
present, and future. Biogeochemistry 70:153–226.
Galloway, J. N., A. R. Townsend, J.W. Erisman,M. Bekunda, Z.
Cai,J. R. Freney, L. A. Martinelli, S. P. Seitzinger, and M. A.
Sutton.2008. Transformation of the nitrogen cycle: recent trends,
ques-tions, and potential solutions. Science 320:889–892.
Gray, J. I., B. MacDonald, A. M. Pearson, and I. D. Morton.
1981.Role of nitrite in cured meat flavor: a review. Journal of
FoodProtection 44:302–312.
He, D., L. Ren, and Q. L. Wu. 2014. Contrasting diversity of
epi-biotic bacteria and surrounding bacterioplankton of a com-mon
submerged macrophyte, Potamogeton crispus, in fresh-water lakes.
FEMS Microbiology Ecology 90:551–562.
He, Z., D. Kisla, L. Zhang, C. Yuan, K. B. Green-Church, and A.
E.Yousef. 2007. Isolation and identification of a
Paenibacilluspolymyxa strain that coproduces a novel lantibiotic
and poly-myxin. Applied and Environmental Microbiology
73:168–178.
Heidelberg, J. F., R. Seshadri, S. A. Haveman, C. L. Hemme, I.
T.Paulsen, J. F. Kolonay, J. A. Eisen, N. Ward, B. Methe, L.
M.Brinkac, S. C. Daugherty, R. T. Deboy, R. J. Dodson, A. S.Durkin,
R. Madupu, W. C. Nelson, S. A. Sullivan, D. Fouts,D. H. Haft, J.
Selengut, J. D. Peterson, T.M. Davidsen, N. Zafar,L. Zhou, D.
Radune, G. Dimitrov, M. Hance, K. Tran, H.Khouri, J. Gill, T. R.
Utterback, T. V. Feldblyum, J. D. Wall,G. Voordouw, and C. M.
Fraser. 2004. The genome sequenceof the anaerobic, sulfate-reducing
bacteriumDesulfovibrio vul-garis Hildenborough. Nature
Biotechnology 22:554–559.
Hoellein, T. J., J. L. Tank, J. J. Kelly, and E. J.
Rosi-Marshall. 2010.Seasonal variation in nutrient limitation of
microbial biofilmscolonizing organic and inorganic substrata in
streams. Hydro-biologia 649:331–345.
Isbell, F., P. B. Reich, D. Tilman, S. E. Hobbie, S. Polasky,
and S.Binder. 2013. Nutrient enrichment, biodiversity loss, and
con-sequent declines in ecosystem productivity. Proceedings ofthe
National Academy of Sciences of the United States ofAmerica
110:11911–11916.
Johnson, L. T., J. L. Tank, and W. K. Dodds. 2009. The
influenceof land use on stream biofilm nutrient limitation across
eightNorth American ecoregions. Canadian Journal of Fisheriesand
Aquatic Sciences 66:1081–1094.
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=19892979&crossref=10.1126%2Fscience.1176199&citationId=p_15https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1890%2F12-2010.1&citationId=p_1https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1890%2F12-2010.1&citationId=p_1https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.2307%2F1468034&citationId=p_19https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.2307%2F1468034&citationId=p_19https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.4315%2F0362-028X-44.4.302&citationId=p_23https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.4315%2F0362-028X-44.4.302&citationId=p_23https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1046%2Fj.1365-2427.1998.00325.x&citationId=p_12https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1007%2Fs10750-010-0276-x&citationId=p_27https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1007%2Fs10750-010-0276-x&citationId=p_27https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=17922835&crossref=10.1111%2Fj.1461-0248.2007.01113.x&citationId=p_16https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1007%2Fs10533-004-0370-0&citationId=p_20https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1046%2Fj.1365-2426.2003.01161.x&citationId=p_2https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2F1574-6941.12414&citationId=p_24https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1890%2F08-1823.1&citationId=p_6https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1890%2F08-1823.1&citationId=p_6https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1086%2F684646&citationId=p_13https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=23818582&crossref=10.1073%2Fpnas.1310880110&citationId=p_28https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=23818582&crossref=10.1073%2Fpnas.1310880110&citationId=p_28https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=23818582&crossref=10.1073%2Fpnas.1310880110&citationId=p_28https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1139%2Ff90-165&citationId=p_17https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1139%2Ff90-165&citationId=p_17https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1127%2Farchiv-hydrobiol%2F143%2F1998%2F147&citationId=p_3https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1899%2F10-146.1&citationId=p_10https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1899%2F10-146.1&citationId=p_10https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1128%2FAEM.02023-06&citationId=p_25https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1016%2FS0038-0717%2898%2900181-3&citationId=p_14https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1016%2FS0038-0717%2898%2900181-3&citationId=p_14https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1139%2FF09-065&citationId=p_29https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1139%2FF09-065&citationId=p_29https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=18487183&crossref=10.1126%2Fscience.1136674&citationId=p_22https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1023%2FA%3A1016136723584&citationId=p_11https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1023%2FA%3A1016136723584&citationId=p_11https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1038%2Fnbt959&citationId=p_26https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=20383131&crossref=10.1038%2Fnmeth.f.303&citationId=p_8
-
262 | NO3– amendment reduces biofilm biomass C. Vizza et al.
Jost, L. 2007. Partitioning diversity into independent alpha
andbeta components. Ecology 88:2427–2439.
Kahlert, M. 1998. C: N: P ratios of freshwater benthic
algae.Ergebnisse der Limnologie 51:105–114.
Kammerer, J. C. 1990. Largest rivers in the United States.
Open-File Report 87–242. US Geological Survey, Water
ResourcesDivision, Reston, Virginia. (Available from:
https://pubs.usgs.gov/of/1987/ofr87-242/pdf/ofr87242.pdf )
Karlsson, J., J. Ask, L. Persson, M. Jansson, P. Byström, and P.
Ask.2009. Light limitation of nutrient-poor lake ecosystems.
Na-ture 460:506–510.
Keck, F., and F. Lepori. 2012. Can we predict nutrient
limitationin streams and rivers? Freshwater Biology
57:1410–1421.
Kim, S. J., S. C. Shin, S. G. Hong, Y. M. Lee, H. Lee, J. Lee,
I.-G.Choi, and H. Park. 2012. Genome sequence of Janthino-bacterium
sp. strain PAMC 25724, isolated from alpine glaciercryoconite.
Journal of Bacteriology 194:2096.
Kozich, J. J., S. L. Westcott, N. T. Baxter, S. K. Highlander,
andP. D. Schloss. 2013. Development of a dual-index
sequencingstrategy and curation pipeline for analyzing amplicon
se-quence data on the MiSeq Illumina sequencing platform. Ap-plied
and Environmental Microbiology 79:5112–5120.
Lang, D. A., R. S. King, and J. T. Scott. 2012. Divergent
responsesof biomass and enzyme activities suggest differential
nutrientlimitation in stream periphyton. Freshwater Science
31:1096–1104.
Lang, J. M., R. W. McEwan, and M. E. Benbow. 2015. Abiotic
au-tumnal organic matter deposition and grazing disturbance
ef-fects on epilithic biofilm succession. FEMS MicrobiologyEcology
91. doi:10.1093/femsec/fiv060
Lauber, C. L., N. Zhou, J. I. Gordon, R. Knight, and N.
Fierer.2010. Effect of storage conditions on the assessment of
bacte-rial community structure in soil and human-associated
sam-ples. FEMS Microbiology Letters 307:80–86.
Likens, G. E., F. H. Bormann, N. M. Johnson, D. W. Fisher, andR.
S. Pierce. 1970. Effects of forest cutting and herbicide treat-ment
on nutrient budgets in the Hubbard Brook watershed-ecosystem.
Ecological Monographs 40:23–47.
Madigan, M. T., K. S. Bender, D. H. Buckley, J. M. Martinko,
andD. A. Stahl. 2014. Brock biology of microorganisms. 14 edi-tion.
Pearson, London, UK.
McGeoch,M. A. 1998. The selection, testing and application of
ter-restrial insects as bioindicators. Biological Reviews
73:181–201.
Meunier, C. L., M. J. Gundale, I. S. Sánchez, and A. Liess.
2016.Impact of nitrogen deposition on forest and lake food websin
nitrogen-limited environments. Global Change Biology
22:164–179.
Pantanella, F., F. Berlutti, C. Passariello, S. Sarli, C. Morea,
and S.Schippa. 2007. Violacein and biofilm production in
Janthino-bacterium lividum. Journal of Applied Microbiology
102:992–999.
Patureau, D., J.-J. Godon, P. Dabert, T. Bouchez, N. Bernet, J.
P.Delgenes, and R. Moletta. 1998. Microvirgula
aerodenitrificansgen. nov., sp. nov., a new Gram-negative bacterium
exhibitingco-respiration of oxygen and nitrogen oxides up to
oxygen-saturated conditions. International Journal of Systematic
andEvolutionary Microbiology 48:775–782.
Pechal, J. L., and M. E. Benbow. 2016. Microbial ecology of
thesalmon necrobiome: evidence salmon carrion decomposition
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
influences aquatic and terrestrial insect microbiomes.
Envi-ronmental Microbiology 18:1511–1522.
Piggott, M. P., and A. C. Taylor. 2003. Extensive evaluation of
fae-cal preservation and DNA extraction methods in Australiannative
and introduced species. Australian Journal of
Zoology51:341–355.
Pruesse, E., C. Quast, K. Knittel, B. M. Fuchs, W. Ludwig,
J.Peplies, and F. O. Glöckner. 2007. SILVA: a comprehensiveonline
resource for quality checked and aligned ribosomalRNA sequence data
compatible with ARB. Nucleic Acids Re-search 35:7188–7196.
Ramírez, K. S., J. M. Craine, and N. Fierer. 2012. Consistent
ef-fects of nitrogen amendments on soil microbial communitiesand
processes across biomes. Global Change Biology 18:1918–1927.
Redfield, A. C. 1958. The biological control of chemical factors
inthe environment. American Scientist 46:205–221.
Reisinger, A. J., J. L. Tank, andM.M. Dee. 2016. Regional and
sea-sonal variation in nutrient limitation of river biofilms.
Fresh-water Science 35:474–489.
Roberts, T. A. 1975. The microbiological role of nitrite and
ni-trate. Journal of the Science of Food and Agriculture
26:1755–1760.
Rugenski, A. T., A. M. Marcarelli, H. A. Bechtold, and R.
S.Inouye. 2008. Effects of temperature and concentration onnutrient
release rates from nutrient diffusing substrates. Jour-nal of the
North American Benthological Society 27:52–57.
Sanderson, B. L., H. J. Coe, C. D. Tran, K. H. Macneale, D.
L.Harstad, and A. B. Goodwin. 2009. Nutrient limitation of
pe-riphyton in Idaho streams: results from nutrient diffusing
sub-strate experiments. Journal of the North American
Bentho-logical Society 28:832–845.
Schloss, P. D., H. K. Allen, A. K. Klimowicz, C. Mlot, J. A.
Gross, S.Savengsuksa, J. McEllin, J. Clardy, R. W. Ruess, and J.
Han-delsman. 2010. Psychrotrophic strain of
Janthinobacteriumlividum from a cold Alaskan soil produces
prodigiosin. DNAand Cell Biology 29:533–541.
Steinman, A. D., G. A. Lamberti, and P. R. Leavitt. 2017.
Biomassand pigments of benthic algae. Pages 223–241 in F. R.
HauerandG. A. Lamberti (editors).Methods in stream ecology. 3rd
edi-tion. Elsevier, Amsterdam, The Netherlands.
Stumm, W., and J. J. Morgan. 1996. Aquatic chemistry:
chemicalequilibria and rates in natural waters. 3 edition. John
Wileyand Sons, New York.
Tank, J. L., and W. K. Dodds. 2003. Nutrient limitation
ofepilithic and epixylic biofilms in ten North American
streams.Freshwater Biology 48:1031–1049.
Tank, J. L., A. J. Reisinger, and E. J. Rosi. 2017. Nutrient
limitationand uptake. Pages 147–171 in F. R. Hauer and G. A.
Lamberti(editors). Methods in stream ecology. 3rd edition.
Elsevier,Amsterdam, The Netherlands.
Thilenius, J. F. 1995. Phytosociology and succession on
earth-quake-uplifted coastal wetlands, Copper River Delta,
Alaska.General Technical Report PNW-GTR-346. Pacific
NorthwestResearch Station, Forest Service, US Department of
Agricul-ture, Portland, Oregon. (Available from:
https://www.fs.fed.us/pnw/pubs/pnw_gtr346.pdf )
Vizza, C., W. E. West, S. E. Jones, J. A. Hart, and G. A.
Lamberti.2017a. Regulators of coastal wetland methane production
and
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=18027744&crossref=10.1890%2F06-1736.1&citationId=p_30https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1899%2F09-072.1&citationId=p_56https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1899%2F09-072.1&citationId=p_56https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1046%2Fj.1365-2427.2003.01067.x&citationId=p_60https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2Fgcb.12967&citationId=p_45https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1038%2Fnature08179&citationId=p_34https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1038%2Fnature08179&citationId=p_34https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1071%2FZO03012&citationId=p_49https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1086%2F685829&citationId=p_53https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1086%2F685829&citationId=p_53https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1899%2F12-031.1&citationId=p_38https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.2307%2F1942440&citationId=p_42https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1089%2Fdna.2010.1020&citationId=p_57https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1089%2Fdna.2010.1020&citationId=p_57https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=17947321&crossref=10.1093%2Fnar%2Fgkm864&citationId=p_50https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=17947321&crossref=10.1093%2Fnar%2Fgkm864&citationId=p_50https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2Fj.1365-2427.2012.02802.x&citationId=p_35https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1002%2Fjsfa.2740261118&citationId=p_54https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2Fj.1365-2486.2012.02639.x&citationId=p_51https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1128%2FJB.00096-12&citationId=p_36https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1899%2F07-046R1.1&citationId=p_55https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&system=10.1899%2F07-046R1.1&citationId=p_55https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1017%2FS000632319700515X&citationId=p_44https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2F1462-2920.13187&citationId=p_48https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2F1462-2920.13187&citationId=p_48https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=23793624&crossref=10.1128%2FAEM.01043-13&citationId=p_37https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&pmid=23793624&crossref=10.1128%2FAEM.01043-13&citationId=p_37https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2Fj.1574-6968.2010.01965.x&citationId=p_41
-
Volume 37 June 2018 | 263
responses to simulated global change. Biogeosciences
14:431–446.
Vizza, C., J. A. Zwart, S. E. Jones, S. D. Tiegs, and G. A.
Lamberti.2017b. Landscape patterns shape wetland pond
ecosystemfunction from glacial headwaters to ocean. Limnology
andOceanography 62:S207–S221.
Von Schiller, D., E. Martí, J. L. Riera, and F. Sabater. 2007.
Effectsof nutrients and light on periphyton biomass and nitrogen
up-take in Mediterranean streams with contrasting land
uses.Freshwater Biology 52:891–906.
This content downloaded from 129.0All use subject to University
of Chicago Press Terms
Wetzel, R. G., and G. E. Likens. 2000. Composition and biomassof
phytoplankton. Pages 147–174 in R. G. Wetzel and G. E.Likens
(editors). Limnological analyses. Springer, New York.
Wyatt, K. H., R. J. Stevenson, and M. R. Turetsky. 2010. The
im-portance of nutrient co-limitation in regulating algal
commu-nity composition, productivity and algal-derived DOC in
anoligotrophic marsh in interior Alaska. Freshwater
Biology55:1845–1860.
Zar, J. H. 2010. Biostatistical analysis. 5th edition. Pearson
Prentice–Hall, Upper Saddle River, New Jersey.
74.250.206 on June 08, 2018 09:51:28 AMand Conditions
(http://www.journals.uchicago.edu/t-and-c).
https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.5194%2Fbg-14-431-2017&citationId=p_64https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2Fj.1365-2427.2010.02419.x&citationId=p_69https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1002%2Flno.10575&citationId=p_66https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1002%2Flno.10575&citationId=p_66https://www.journals.uchicago.edu/action/showLinks?doi=10.1086%2F697897&crossref=10.1111%2Fj.1365-2427.2007.01742.x&citationId=p_67