Specific responses to nitrogen and phosphorusenrichment in cyanobacteria: Factors influencingchanges in species dominance along eutrophicgradients

Virginia Loza, Elvira Perona, Pilar Mateo*

Departamento de Biologıa, Facultad de Ciencias, Universidad Autonoma de Madrid, Madrid, Spain

a r t i c l e i n f o

Article history:

Received 3 May 2013

Received in revised form

10 September 2013

Accepted 6 October 2013

Available online 23 October 2013

Keywords:

Rivers

Streams

Benthic cyanobacteria

Growth

Nutrient regimes

Nitrate

Ammonium

Phosphate

Competition

a b s t r a c t

Anthropogenic eutrophication is aworldwide problem, causing proliferationof cyanobacterial

masses, some of whichmay be toxic. However, little is known about whether the response to

nutrient enrichment differs among cyanobacterial species. To address this issue, distinct

patterns in growth and competitive response of benthic cyanobacteria under N and P nutrient

regimes were studied. Nine cyanobacterial species, collected from Guadarrama river biofilms

at several locations with different nutrient concentrations, were isolated and used for a series

of N and P enrichment bioassays. In competition experiments with a mixture of all nine spe-

cies, a great predominance of certain cyanobacteria over others was noted at high nutrient

conditions,while under lownutrient conditions someothers dominated. On the basis of these

results four selected strains were subjected to a gradient of different concentrations of phos-

phate, nitrate and ammonium, in independent bioassays, both in monocultures and mixed

cultures. Depending on the concentration of N and P, stimulation or inhibition of growth was

observed. Some species grewbetter, dominating at highnutrient concentrations, while higher

yields were recorded for others under low nutrient regimes, dominating in these conditions.

Results from this study clarify previously published field observations, whereby a group of

species occurred mostly in downstream nutrient-rich locations, while other was typical of

upstream oligotrophic conditions. Our findings concerning differential growth in relation to

nutrient concentrationsmay be useful for environmental management, because they help us

predict which cyanobacteria may be expected to occur under certain conditions.

ª 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Human disturbances give rise to a range of alterations in river

systems, in which nitrogen and phosphorus inputs can have

profound effects upon the quality of receiving waters, leading

to eutrophication, which in turn causes proliferation of algal

masses, some of which may be toxic (Dodds and Welch, 2000;

Anderson et al., 2002). Other symptoms of eutrophication

include deep water anoxia in lakes, taste and odour problems,

and changes in the composition of aquatic communities

* Corresponding author. Departamento de Biologıa, Facultad de Ciencias, Universidad Autonoma de Madrid, C/Darwin n� 2, 28049Madrid, Spain. Tel.: þ34 91 497 8184; fax: þ34 91 497 8344.

E-mail address: pilar.mateo@uam.es (P. Mateo).

Available online at www.sciencedirect.com

ScienceDirect

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

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1

0043-1354/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.watres.2013.10.014

(Lewis et al., 2011). Recognition of these problems has

prompted a concerted worldwide effort to develop a rational

predictive framework for preventing and managing fresh-

water eutrophication (European Communities, 2000; USEPA,

2000).

Cyanobacteria are among dominant primary producers in

most freshwater environments, providing the principal en-

ergy base for many aquatic food webs (Scott and Marcarelli,

2012). They play important biogeochemical roles in terms of

nutrient fixation and cycling within the ecosystem (Scott and

Marcarelli, 2012). Nutrient enrichment has been closely linked

to the stimulation of growth of masses of cyanobacteria, but it

did not affect all species equally, although an increase of some

harmful species has been reported (Anderson et al., 2002).

Therefore, an improved understanding of the nature of taxon-

specific responses to N and P influx may help to protect

aquatic ecosystems against future pollution, optimise waste-

water treatment procedures, and resolve the continuing

debate about the respective roles of N and P in regulating

human made eutrophication (Donald et al., 2013).

In previous reports, we described shifts in the structure

and composition of benthic cyanobacterial communities in

rivers with increasing eutrophication in the downstream di-

rection (Perona et al., 1998, 1999; Douterelo et al., 2004; Perona

and Mateo, 2006; Rodriguez et al., 2007). In the Guadarrama

river (central Spain), molecular fingerprint analyses using

temperature gradient gel electrophoresis (TGGE) were

consistent with those obtained from microscopic observa-

tions of field-fixed samples, in which the cyanobacterial

community composition at sampling locations differed be-

tween upstream and downstream locations, whereby some

species predominated at downstream sites with high nutrient

content, while others were scarce or absent (Loza et al., 2013a,

b). However, assessing the causes of the presence or absence

of species is difficult from field observations alone. The causal

factors influencing cyanobacterial species composition are

poorly understood. In trying to detect factors that may affect

species distribution, statistical analyses have been carried out

to detect significant correlations between relative abundances

and water quality variables. However, many environmental

factors are usually interrelated so species distributionsmay be

correlatedwith a broad range of characteristics, although only

one may reflect a causal relationship. Therefore, a better un-

derstanding of the factors controlling the relative dominance

and/or coexistence phenomena in the environment is needed.

We hypothesized that specific characteristics of cyano-

bacterial species, such as differential physiological traits

related to changes in N and P concentration, influence how

species respond to changes in environmental conditions over

space and time, leading to the observed shifts in the structure

of the cyanobacterial communities. Such relationships have

often been proposed, but rarely demonstrated experimentally.

Empirical evidence of this nature would be helpful for inter-

preting species distributions in relation to environmental

changes. Therefore, in this study, a series of bioassays were

designed to test individually significant factors that may be

responsible for the observed changes. The effects of different

nutrient regimes (nitrogen and phosphorus) on the growth of

isolated strains from the Guadarrama river were assayed in an

attempt to identify causal relationships.

2. Materials and Methods

2.1. Cyanobacterial cultures

Cyanobacterial strains used for bioassays were isolated from

epilithic biofilms from the riverbed of the Guadarrama river

(central Spain, Madrid), as previously described (Loza et al.,

2013b). Three culture media were used for the isolation in

order to provide a wide range of nutrient concentrations, since

we had previously found different patterns of cyanobacterial

growth to be associated with the particular culture media used

(Berrendero et al., 2008). CHU No. 10modified (CHU10) medium

(Chu, 1942; Gomez et al., 2009) was selected as a low nutrient

content medium (0.99 mg P L�1, 3.5 mg N L�1), BG110 (Rippka

et al., 1979) was selected as an intermediate nutrient level me-

dium (5.5 mg P L�1, 247 mg N L�1), and Allen & Arnon medium

(AAN) (Allen and Arnon, 1955) was chosen as a high nutrient

concentration medium (61.9 mg P L�1, 175 mg N L�1). The nine

isolated cyanobacteria used for the experiments were the

coccoid and non-heterocystous filamentous species Cyanobium

sp., Aphanocapsa muscicola, Pleurocapsa minor, Pseudanabaena

catenata, Leptolyngbya boryana, Leptolyngbya nostocorum, Phormi-

dium sp. and the two heterocyst-forming speciesNostoc carneum

and Tolypothrix tenuis. A detailed description of all isolates used

in this study can be found in Loza et al. (2013b, c).

2.2. Experimental setup

2.2.1. Competition experiments with a mixture of all ninespeciesThedifferential growthofcyanobacterial strainswasanalyzedby

measuring the percentage of the surface of 14-cm diameter agar

plates colonized in the low nutrient CHU10 medium and the

nutrient-rich AAN, described above, simulating the oligo-

mesotrophic and hypertrophic conditions previously found in

the river (Loza et al., 2013a,b). Isolates were inoculated on 2%

agar-platesandwerekept for24h ina16:8h light:darkregimeata

temperature of 18 �C, with an irradiance of 20 mmol photon

m�2 s�1, without shaking, in order to fix inocula on the agar

surface.After that,agarplateswerecoveredby5mLofboth liquid

media (CHU10 and AAN medium) on each, and then put in the

shaker under the same conditions. Media were replaced on day

10 of the experiment. Relative abundance was analyzed using a

dissecting microscope (Leica; Leica Microsystems) and the per-

centageof the area colonizedwas calculated for each isolate. The

experiment was replicated two times, yielding similar results.

Values from a representative experiment are shown.

2.2.2. Nutrient gradient bioassaysOn the basis of the results obtained from the competition exper-

iments with all strains, a series of bioassays under a gradient of

different concentrations of derived nutrients from nitrogen (ni-

trate ðNO�3 Þ and ammonium ðNHþ

4 Þ and phosphorus ðPO3�4 Þwere

carried out. T. tenuis, N. carneum, Phormidium sp., and L. boryana.

were selected for these bioassays. Independent experiments for

each nutrient were performed over 20 days in CHU10medium in

which only the concentration of the ammonium, nitrate or

phosphatewasmodified to final concentrations of 0.2, 3, 6, 10, 50

and 100 mg L�1 (ranges: 0.04e22.58 mg L�1 Ne NO�3 ;

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1 623

0.16e82.35 mg L�1 NeNHþ4 ; 0.06e32.63 mg L�1 PePO3�

4 ). The ex-

periments were carried out with triplicate samples for each

cyanobacterium in monoculture, as well as in competition, in

which two species were placed together. Media at different

nutrient concentrationswere distributed in the different wells of

sterile polystyrene 24-well microtitre plates (IWAKI Microplate).

For each nutrient experiment, isolated cultures were washed

twice in the appropriate medium and 100 mL were transferred to

the sterile plates with 2 mL of each medium at the correct con-

centration. Finally, plates weremaintained for 24 h with a 16:8 h

light:dark period at a temperature of 18 �C, with an irradiance of

20 mmol photonm�2 s�1,without shaking. Theywere thenplaced

on a rotating shaker (200 rpm) to obtain a homogeneous suspen-

sionof theculture.Themediumwaspartially replaced (50%)once

during the 10 day experiment. For each species, the starting

inoculum for all experiments was taken from the same stock

culture, which was grown under low nutrient conditions for one

week.

2.3. Sample counting

Growth of monocultures and mixed cultures was measured as

the totalnumberof cellsmL�1. Sincea thinbiofilmwasdeveloped

at the bottom of the wells at the end of the experiment, due to a

great tendency of epilithic cyanobacteria to attach to the sub-

strata, samples were firmly scraped with a sterile scalpel. Sam-

ples were then preserved adding 10 mL of formaldehyde to each

well (4% final concentration), and kept in dark at 4 �C, taken into

consideration this dilution in further cell counting. Total number

of cells was determinedmicroscopically by counting at least 400

cells in each sample in aNeubauer chamber (BRANDRef.718605).

2.4. Statistical analysis

All groups of data were analysed for normality by the

ShapiroeWilk test and for homoscedasticity by the Levene’s

test. Significant differences in growth were identified using

the non-parametric KruskaleWallis one-way analysis of

variance as several of our data did not follow a normal dis-

tribution and/or did not present homogeneity of variances.

Further post hoc comparisons were performed using the

Student-Newman-Keuls method. Statistics were carried out

using SigmaPlot version 9.0 with SigmaStat integration

version 3.1.

3. Results

3.1. Competition experiments with a mixture of all ninespecies

Representation of differential growth trends under nutrient

regimes in cyanobacterial isolates is shown in Fig. 1. The dis-

tribution of the species on day 10 at high nutrient concentra-

tions (AANmedium) showed a dominance of N. carneum (37%)

and Phormidium sp. (27%), followed by A. muscicola, P. catenata

and L. nostocorum (Fig. 1A). However, on day 20 (Fig. 1B) a clear

dominance of Phormidium sp. over the rest of cyanobacteria

was noticed, with almost 59% of the area colonized, and with

N. carneum accounting for only 7.6% of the coverage. T. tenuis

and P. minor were represented by very low percentages of the

colonized surface: 0.4 and 0.7%, respectively (Fig. 1B). On the

other hand, under the low nutrient regime (CHU10medium) T.

tenuiswas dominant on day 10 (Fig. 1A) occupying 41.5% of the

total area colonized, followed by N. carneum (37.5%) and L.

nostocorum (5%). Similar results were obtained after day 20 of

the treatment (Fig. 1B), in which T. tenuis comprised 49% of the

colonized surface, closely followed by N. carneum (40%).

On the basis of these results, a further experiment (Fig. 2)

was carried out in the different culture media with the two

dominant cyanobacteria, Phormidium sp. and T. tenuis, select-

ing N. carneum as the other N-fixing cyanobacteria whose

response varied with the medium and time of culture, and L.

boryana as the non-heterocystous cyanobacteria that grew

similarly in both nutrient regimes (Fig. 1). Similar results were

found concerning the dominant species, whereby for all

exposure durations, Phormidium sp. greatly predominated

under high nutrient conditions, while T. tenuis dominated

under low nutrient conditions. N. carneum also exhibited

greater growth at lower nutrient concentrations, while L.

boryana grewmore under the higher nutrient regime. Based on

these results, these four cyanobacterial species were selected

for subsequent bioassays in order to identify the patterns

among these isolates more clearly.

3.2. Nutrient gradient bioassays

Selected species were subjected to a gradient of different

concentrations of ammonium, nitrate and phosphate. Cya-

nobacterial growthwasmeasured inmonocultures andmixed

cultures. On the basis of previous results, competition ex-

periments were performed mixing T. tenuis with Phormidium

sp., and N. carneum with L. boryana.

3.2.1. Ammonium responsesIn monoculture bioassays, Phormidium sp. and L. boryana grew

significantly more at higher ammonium concentrations

(KruskaleWallis, p < 0.01) (Fig. 3A, D), except for a slight

decrease in L. boryana biomass at 100 mg L�1. At the lowest

concentration (0.2 mg L�1) the number of cells was similar to

that of the initial inoculum, indicating that these conditions

were adverse for both species. Conversely, growth of mono-

cultures of T. tenuis and N. carneum decreased as the concen-

tration of ammonium increased, both with similar growth

patterns (Fig. 3B, E). The same trend was noticed for each

cyanobacterium in competition experiments, leading to a

clear and significant dominance of Phormidium sp. and L. bor-

yana at higher ammonium concentrations, while T. tenuis and

N. carneum dominated significantly at low ammonium con-

centrations (Fig. 3C, F) (KruskaleWallis, p < 0.01).

3.2.2. Nitrate responsesGrowth responses of Phormidium sp. and L. boryana under ni-

trate gradient bioassays also had a significantly higher yield

(KruskaleWallis, p< 0.01) at higher concentrations (Fig. 4A, D).

On the other hand, growth of N. carneum decreased signifi-

cantly (KruskaleWallis, p < 0.05) as the nitrate concentration

in the culture medium increased, whereas in the T. tenuis

bioassay, growth varied little along the nitrate gradient

(Fig. 4B, E). In mixed cultures, again, T. tenuis and N. carneum

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1624

grew more under low nitrate concentrations, while Phormi-

dium sp. and L. boryana dominated significantly at higher

concentrations (KruskaleWallis, p < 0.01) (Fig. 4C, F).

3.2.3. Phosphate responsesIn monocultures, under increased phosphate regimes, Phormi-

dium sp. and L. boryana showed a progressive increase in

biomass, peaking at 10mgL�1, and thendecreasing significantly

atevenhigher concentrations (KruskaleWallis,p< 0.01) (Fig. 5A,

D). T. tenuis and N. carneum showed a significant reduction in

biomass along the gradient (Fig. 5B, E) (KruskaleWallis, p< 0.01,

both species). The trend in competition experiments (Fig. 5C, F)

was the same as that in previous nutrients, whereby T. tenuis

and N. carneum dominated under low phosphate regimes, but

Phormidium sp. and L. boryana dominated at higher phosphate

concentrations (KruskaleWallis, p < 0.01).

4. Discussion

Patterns of growth and competitive response under different

nutrient regimes differed between cyanobacteria. Depending

on the concentration of phosphorus and nitrogen, growthwas

either stimulated or inhibited. Some species grew better at

high nutrient concentrations, while higher yields were

recorded for others under low nutrient regimes, leading to

significant shifts in dominance depending on the N and P,

enrichments, as well as on the redox distinction between ni-

trogen nutrients. These results are consistent with the dif-

ferential distribution of these cyanobacteria in the

Guadarrama river, according to downstream increasing

eutrophication; whereby a group of species occurredmostly in

downstreamnutrient-rich locations, while otherswere typical

of upstream oligotrophic conditions (Loza et al., 2013a,b c).

Therefore, the studied cyanobacteria responded differently in

terms of their nutrient requirements. Those cyanobacteria

which dominated in the nutrient-rich culture medium (AAN)

appeared in great abundance in the river under hypertrophic

conditions, while those with a higher abundance in the sam-

pling site with low nutrient concentrations, in turn, grew

better in the culture medium that simulated oligotrophic

conditions (CHU10). In nutrient gradient bioassays, where

ranges of concentration were representative of downstream

increasing eutrophication (Loza et al., 2013b), Phormidium sp.

Fig. 1 e Representation of the competition experiments with a mixture of nine cyanobacterial species on day 10 (A) and day

20 (B) as percentage of colonized area of agar plates of high nutrient Allen & Arnon (AAN) and low nutrient CHU10 culture

media (see Materials and Methods).

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1 625

and L. boryana grew significantly more and dominated at high

nutrient concentrations, consistent with increasing abun-

dance downstream. However, conversely, dominance of N.

carneum and T. tenuis at low nutrient concentrations was

consistent with increasing abundance upstream. For fresh-

water algae and cyanobacteria, nutrient concentration and

ratios have been invoked as one of the variables controlling

their community structure and biomass, with cyanobacteria

dominating in waters with low N:P ratios (Smith, 1983; Sekar

et al., 2002). In our study increased values of N:P ratios were

found at the river in a downstream direction (Loza et al.,

2013a), as well as in the bioassays with increasing N concen-

trations or decreasing P values (data not shown). However,

depending on the cyanobacteria, a different response for the

same values of N:P was found in the vis-a-vis experiments,

suggesting that the individual nutrient concentrations, either

phosphorus or nitrogen, are more related to the differential

cyanobacterial dominance than to their ratio, as previously

proposed in studies of cyanobacterial dominance in lakes

(Oliver et al., 2012).

There are few studies based on ecophysiological bioassays

of different forms of nutrients in benthic cyanobacteria in

rivers; most of the research focuses on phytoplankton

species in reservoirs and lakes, in which variable responses

to nutrient enrichment have been found within microalgal

communities (Armitage et al., 2006; Donald et al., 2013). Ex-

periments carried out under N-depleted conditions have

revealed a range of behaviours among different cyano

bacteria. For example, Synechococcus PCC 6301 did not grow

under N-limited conditions, especially NeNHþ4 (Ahlgren and

Hyenstrand, 2003), and Oscillatoria simplicissima, isolated

from a South African eutrophic river, showed a reduction in

the length of their filaments in response to N-limiting con-

ditions (Kruskopf and Du Plessis, 2006), indicating growth

inhibition, as described in Anabaena affinis under conditions

of nutrient stress (Smith and Gilbert, 1995). In competition

bioassays, Phormidium tenue was more susceptible to N-

limited conditions than Microcystis aeruginosa, and dominated

at high N concentrations (Fujimoto et al., 1997). In studies of

phytoplankton species in an Australian river, M. aeruginosa

was also dominant when nutrient levels were low, whereas

most non-cyanobacterial phytoplankton (mainly Chlor-

ophyceae) dominated when nutrient levels were high

(Mitrovic et al., 2001). Another study showed that the het-

erocystous cyanobacteria Cylindrospermum stagnale grew bet-

ter under low nitrate concentrations than Leptolyngbya

Fig. 2 e Representation of the competition experiments with a mixture of Phormidium sp., Tolypothrix tenuis, Nostoc carneum,

and Leptolyngbya boryana on day 10 (A) and day 20 (B) as percentage of colonized area of agar plates of high nutrient Allen &

Arnon (AAN) and low nutrient CHU10 culture media.

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1626

foveolarum, and dominated under these conditions while,

conversely, L. foveolarum dominated at high concentrations

(Van der Grinten et al., 2004a).

Phosphorus is widely considered to be the main nutrient

controlling the development of natural populations of cya-

nobacteria in freshwaters, since, depending on its concen-

tration, it can act as a growth-limiting factor or promote

blooms (Schindler, 1977; Anderson et al., 2002; Schindler et al.,

2008). Results from our study showed that each species re-

quires different ranges of P concentrations for optimal

growth, the distinct P regimes being an important factor in

determining the composition and abundance of species.

Sabater et al. (2003) observed that the mass growth of Oscil-

latoriamats in a Spanish river occurred when there was a high

phosphorus content in the water. However, phosphate

differentially affected the growth of two Synechococcus strains

isolated from mesotrophic subalpine lakes, whereby growth

of one of the strains was delayed and inhibited by increasing

phosphate, while the other strain not only needed higher

initial phosphate concentrations to achieve the highest

growth rates, but also tolerated high phosphate concentra-

tions, explaining the differences in their seasonal appearance

in the natural environment (Ernst et al., 2005). Pfeiffer and

Palinska (2002) found that the optimal growth concentration

of six strains of Phormidium was 20 mg phosphate L�1, and

Oscillatoria sp. isolated from a wastewater treatment plant did

not grow at concentrations below 3 mg phosphate L�1 (Marco

and Orus, 1988). In addition, differences were found between

isolates of the same environment, whichmay be explained by

the existence of “microniches” or ecophysiological differ-

ences. For example, two Microcystis strains with different

morphotypes, colonial or unicellular, showed different trends

under distinct P regimes. The colonial form grew well at low P

concentrations, while the growth of the unicellular form was

inhibited under P-limiting conditions (Shen and Song, 2007).

Furthermore,Microcystis toxic strains exhibited higher growth

at high P concentrations than non-toxic strains, although both

also appear to have an elevated requirement for N (Vezie et al.,

2002). In competition experiments, L. foveolarum grew better

and dominated under high P concentrations with respect to C.

Fig. 3 e Growth of monocultures and mixed cultures along a gradient of different concentrations of ammonium, on day 20.

(A) monoculture of Phormidium sp., (B) monoculture of Tolypothrix tenuis, (C) mixed culture of Phormidium sp. and T. tenuis, (D)

monoculture of Leptolyngbya boryana, (E) monoculture of Nostoc carneum, (F) mixed culture of L. boryana and N. carneum.

Error bars indicate standard errors (n [ 3).

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1 627

stagnale (Van der Grinten et al., 2004b). Planktothrix agardii

(formerly Oscillatoria agardii), which is highly abundant in

many eutrophic lakes in the Netherlands, was not competitive

under P-limiting conditions, wherein it is displaced by the

prochlorophyta Prochlorothrix hollandice in competition exper-

iments (Ducobu et al., 1998).

There is renewed interest in understanding patterns in the

distribution not only of taxa but also of their traits. Trait-based

approaches to microbial biodiversity and biogeography offer

the possibility of advancing ecological theory and predicting

responses to environmental change (Green et al., 2008).

Climate change scenarios predict that rivers and other fresh-

water ecosystems will experience higher temperatures and

altered nutrient loading (Carey et al., 2012; Elliot, 2012). Elliot

(2012) indicated that most of the modelling studies that

included a change innutrients showedanenhancementunder

thehigher nutrient scenarios of the cyanobacterial response to

the climate drivers. Paerl and Paul (2012) pointed out that the

expansion of harmful bloom-forming cyanobacteria, linked to

climate change, has serious consequences for humandrinking

and irrigation water supplies, fisheries and recreational re-

sources, and this has ramifications for water management. In

addition, temperature, light, and nutrients may have interac-

tive effects on cyanobacterial growth and toxin production,

with consequent effects on aquatic food webs (Paerl and Paul,

2012; Elliot, 2012). These environmental drivers will have

substantial effects on species composition, and it has been

predicted that spatial variation in climate change will interact

with physiological variation in cyanobacteria to alter the

dominant cyanobacterial taxa from region to region (Carey

et al., 2012). The results of our research show that ecological

rangesof individual cyanobacterial populations candifferwith

nutrient concentration, and the persistence of several species

under low or high nutrient concentrations in the river can be

explained on the basis of their different ecophysiological

properties. Increased levels of ammonium and/or nitrate-

nitrogen favour the development of the non-heterocystous

cyanobacteria Phormidium sp. and L. boryana, whereas com-

bined nitrogen scarcity facilitates the development of the

heterocystous cyanobacteria N. carneum and T. tenuis. The

Fig. 4 e Growth of monocultures and mixed cultures along a gradient of different concentrations of nitrate, on day 20. (A)

monoculture of Phormidium sp., (B) monoculture of Tolypothrix tenuis, (C) mixed culture of Phormidium sp. and T. tenuis, (D)

monoculture of Leptolyngbya boryana, (E) monoculture of Nostoc carneum, (F) mixed culture of L. boryana and N. carneum.

Error bars indicate standard errors (n [ 3).

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1628

dominance ofheterocystous cyanobacteria in these conditions

may be due to some advantageous ecophysiological trait such

as their ability to carry out nitrogenase activity, fixing atmo-

spheric N2when the levels of combinedN are low. Increases in

the proportion of N2-fixing cyanobacteria has been reported as

a result of N-deficiency in benthic communities (Sekar et al.,

2002; Scott and Marcarelli, 2012, and references therein), and

our previous studies showed that generally heterocystous

cyanobacteria were dominant in different rivers at sites with

low levels of combined N (Perona et al., 1998; Douterelo et al.,

2004; Rodrıguez et al., 2007).

Regarding phosphorus, cyanobacterial responses to P limi-

tation or P enrichment were also species-specific, whereby

Phormidium sp. and L. boryana dominated at high P concentra-

tions, while T. tenuis and N. carneum dominated at low P re-

gimes. Several adaptive ecophysiological traits of dominant

species in oligotrophic environments have been suggested by

various authors. Cyanobacteria respond in different ways to P

starvationandsubsequently in themanner inwhich theyadapt

to environments with low or fluctuating P concentrations.

Some cyanobacteria respond to P-limiting conditions by

increasing surface phosphatase activities and becoming

completely suppressed under high ambient phosphate con-

centrations; these enzymes hydrolyze a variety of P-containing

organicmolecules and release P, making it available for uptake

(Whitton et al., 2005; Whitton and Mateo, 2012). Increased

phosphate uptake rates have also been found in phosphate-

limited cyanobacteria (Ritchie et al., 2001; Fu et al., 2005). In

addition to the ability of carrying out phosphatase activity

when P concentrations are low, a distinct phosphorus-storage

capacity is considered a fundamental strategy in organisms

characteristic of oligotrophic waters (Whitton et al., 2005; Shen

and Song, 2007). Other strategies exploit the presence of

mucilage, which helps these organisms survive in nutrient-

poor media, where they play an important role in nutrient

uptake and assimilation (Reynolds, 2007), or the possibility of

increasing bioavailabilty of P at the thin layer of fine sediment,

commonlyobservedunderwell developed cyanobacterialmats

(Quiblier et al., 2013). Our previous results showed increasing in

situ phosphatase activity in populations of Nostoc and Tolypo-

thrixatupstream locationsofaMediterraneanstreamwithvery

low concentrations of P (Mateo et al., 2010) and that P luxury

Fig. 5 e Growth of monocultures andmixed cultures along a gradient of different concentrations of phosphate, on day 20. (A)

monoculture of Phormidium sp., (B) monoculture of Tolypothrix tenuis, (C) mixed culture of Phormidium sp. and T. tenuis, (D)

monoculture of Leptolyngbya boryana, (E) monoculture of Nostoc carneum, (F) mixed culture of L. boryana and N. carneum.

Error bars indicate standard errors (n [ 3).

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1 629

consumption, by which cells can be taken up in excess of im-

mediate requirement and stored in form the polyphosphate

granules, could assist to overcome P deficient conditions

(Mateo et al., 2006;Munoz-Martın et al., 2013). Furthermore, the

phosphatase activity of Nostoc and Tolypothrix was induced

under P deficiency, and there was also a negative correlation

between this activity and external inorganic P concentration

(Mateo et al., 2006; Munoz-Martın et al., 2013). Therefore, in an

environment with variable and intermittent P inputs, some

cyanobacteria can overcome changes in nutrient loading by

regulatingabsorptionmechanismsandPutilization for optimal

growth. In addition, different cyanobacterial taxa can also vary

in their relative capacity and efficiency in relation to these

processes, which can also explain the differential response

found between the studied cyanobacteria.

5. Conclusions

- Taxa-specific variability in cyanobacterial growth responses

to nutrient enrichment demonstrates that these primary

producer components of fluvial systems do not respond

uniformly to changes in nutrient input, leading to shifts in

the structure and composition of communities.

- Although it is generally assumed that nutrient-rich condi-

tions favour cyanobacterial growth, results from this study

indicate that cyanobacteria should be treated individually

rather than as a homogeneous group. Some species prefer

low nutrient conditions, while others prefer eutrophic to

hypertrophic conditions.

- Specific adaptive ecophysiological traits offer the possibility

for cyanobacteria to occupy distinct niches, which would

explain their environmental divergence, and therefore their

differential distribution.

- Our study provides further information about species-

specific autecologies and interactions between species,

and in turn, highlights the importance of understanding the

dominance of each one, according to their different

ecological preferences, which may provide more informa-

tion about the trophic state in which they occur.

- Information about species-specific autecologies and in-

teractions between speciesmay be useful for environmental

management, because it helps to predict which cyanobac-

teria may be expected to occur under certain conditions.

Acknowledgements

This work was supported by grants from the Ministerio de

Educacion y Ciencia, Spain (CGL2008-02397/BOS), and from the

Comunidad Autonoma de Madrid, Spain (S2009/AMB-1511).

r e f e r e n c e s

Ahlgren, G., Hyenstrand, P., 2003. Nitrogen limitation effects ofdifferent nitrogen sources on nutritional quality of two

freshwater organisms, Scenedesmus quadricauda(Chlorophyceae) and Synechococcus sp. (Cyanophyceae). J.Phycology 39 (5), 906e917.

Allen, M.B., Arnon, D.I., 1955. Studies on nitrogen-fixing blue-green algae. I. Growth and nitrogen fixation by Anabaenacylindrica Lemm. Plant Physiol. 30 (4), 366e372.

Anderson, D.M., Glibert, P.M., Burkholder, J.M., 2002. Harmfulalgal blooms and eutrophication: nutrient sources,composition, and consequences. Estuaries 25 (4b), 704e726.

Armitage, A.R., Frankovich, T.A., Fourqurean, J.W., 2006. Variableresponseswithin epiphytic andbenthicmicroalgal communitiesto nutrient enrichment. Hydrobiologia 569, 423e435.

Berrendero, E., Perona, E., Mateo, P., 2008. Genetic andmorphological characterization of Rivularia and Calothrix(Nostocales, Cyanobacteria) from running water. Int. J. Syst.Evol. Microbiol. 58, 447e460.

Carey, C.C., Ibelings, B.W., Hoffmann, E.P., Hamilton, D.P.,Brookes, J.D., 2012. Eco-physiological adaptations that favourfreshwater cyanobacteria in a changing climate. Water Res. 46(5), 1394e1407.

Chu, S.P., 1942. The influence of the mineral composition of themedium on the growth of planktonic algae. Part 1. Methodsand culture media. J. Ecol. 30, 284e325.

Dodds, W.K., Welch, E.B., 2000. Establishing nutrient criteria instreams. J. North Am. Benthological Soc. 19 (1), 186e196.

Donald, D.B., Bogard, M.J., Finlay, K., Bunting, L., Leavitt, P.R.,2013. Phytoplankton-specific response to enrichment ofphosphorus-rich surface waters with ammonium, nitrate, andurea. PLoS ONE 8 (1), e53277. http://dx.doi.org/10.1371/journal.pone.0053277.

Douterelo, I., Perona, E., Mateo, P., 2004. Use of cyanobacteria toassess water quality in running waters. Environ. Pollut. 127 (3),377e384.

Ducobu, H., Huisman, J., Jonker, R.R., Mur, L.R., 1998. Competitionbetween a prochlorophyte and a cyanobacterium undervarious phosphorus regimes: comparison with the droopmodel. J. Phycology 34 (3), 467e476.

Elliot, J.A., 2012. Is the future blue-green? A review of the currentmodel predictions of how climate change could affect pelagicfreshwater cyanobacteria. Water Res. 46, 1364e1371.

Ernst, A., Deicher, M., Herman, P.M.J., Wollenzien, U.I.A., 2005.Nitrate and phosphate affect cultivability of cyanobacteriafrom environments with low nutrient levels. Appl. Environ.Microbiol. 71 (6), 3379e3383.

European Communities, 22/12/2000. Directive 2000/60/EC of theEuropean Parliament and of the Council of the 23 October 2000establishing a framework for community action in the field ofwater policy. Official J. Eur. Communities L 327.

Fu, F.X., Zhang, Y., Bell, P.R.F., Hutchins, D.A., 2005. Phosphateuptake and growth kinetics of Trichodesmium (Cyanobacteria)isolates from the North Atlantic Ocean and the Great BarrierReef, Australia. J. Phycology 41 (1), 62e73.

Fujimoto, N., Sudo, R., Sugiura, N., Inamori, Y., 1997. Nutrient-limited growth of Microcystis aeruginosa and Phormidium tenueand competition under various N: P supply ratios andtemperatures. Limnol. Oceanogr. 42 (2), 250e256.

Gomez, N., Donato, J.C., Giorgi, A., Guasch, H., Mateo, P.,Sabater, S., 2009. La biota de los rıos: los microorganismosautotrofos. In: Elosegi, A., Sabater, S. (Eds.), Conceptos ytecnicas en ecologıa fluvial. Fundacion BBVA., Bilbao, Spain,pp. 219e242.

Green, J.L., Bohannan, B.J.M., Whitaker, R.J., 2008. Microbialbiogeography: from taxonomy to traits. Science 320,1039e1043.

Kruskopf, M., Du Plessis, S., 2006. Growth and filament length ofthe bloom forming Oscillatoria simplicissima (Oscillatoriales,Cyanophyta) in varying N and P concentrations. Hydrobiologia556, 357e362.

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1630

Lewis Jr., W.M., Wurtsbaugh, W.A., Paerl, H.W., 2011. Rationale forcontrol of anthropogenic nitrogen and phosphorus to reduceeutrophication of inland waters. Environ. Sci. Technol. 45 (24),10300e10305.

Loza, V., Perona, E., Mateo, P., 2013a. Molecular fingerprinting ofcyanobacteria from river biofilms as a water qualitymonitoring tool. Appl. Environ. Microbiol. 79 (5), 1459e1472.

Loza, V., Berrendero, E., Perona, E., Mateo, P., 2013b. Polyphasiccharacterization of benthic cyanobacterial diversity frombiofilms of the Guadarrama (Spain): morphological, molecularand ecological approaches. J. Phycology 49 (2), 282e297.

Loza, V., Perona, E., Carmona, J., Mateo, P., 2013c. Phenotypic andgenotypic characteristics of Phormidium-like cyanobacteriainhabiting microbial mats are correlated with the trophicstatus in running waters. Eur. J. Phycology 48 (2), 235e252.

Marco, E., Orus, M.I., 1988. Variation in growth and metabolismwith phosphorus nutrition in two cyanobacteria. J. PlantPhysiol. 132 (3), 339e344.

Mateo, P., Douterelo, I., Berrendero, E., Perona, E., 2006.Physiological differences between two species ofcyanobacteria in relation to phosphorus limitation. J.Phycology 42 (1), 61e66.

Mateo, P., Berrendero, E., Perona, E., Loza, V., Whitton, B.A., 2010.Phosphatase activities of cyanobacteria as indicators ofnutrient status in a Pyrenees river. Hydrobiologia 652 (1),255e268.

Mitrovic, S.M., Bowling, L.C., Buckney, R.T., 2001. Responses ofphytoplankton to in-situ nutrient enrichment; potentialinfluences on species dominance in a river. Int. Rev.Hydrobiology 86 (3), 285e298.

Munoz-Martın, M.A., Martınez-Rosell, A., Perona, E., Fernandez-Pinas, F., Mateo, P., 2013. Monitoring bioavailable phosphorusin lotic systems: a polyphasic approach based oncyanobacteria. Sci. Total Environ.. http://dx.doi.org/10.1016/j.scitotenv.2013.06.076.

Oliver, R.L., Hamilton, D.P., Brookes, J.D., Ganf, G.G., 2012.Physiology, blooms and prediction of planktoniccyanobacteria. In: Whitton, B.A. (Ed.), Ecology ofCyanobacteria II. Their Diversity in Space and Time. Springer,Dordrecht, pp. 155e194.

Paerl, H.W., Paul, V.J., 2012. Climate change: links to globalexpansion of harmful Cyanobacteria. Water Res. 46,1349e1363.

Perona, E., Bonilla, I., Mateo, P., 1998. Epilithic cyanobacteriacommunities and water quality: an alternative tool formonitoring eutrophication in the Alberche River (Spain). J.Appl. Phycology 10 (2), 183e191.

Perona, E., Bonilla, I., Mateo, P., 1999. Spatial and temporalchanges in water quality in a Spanish river. Sci. Total Environ.241 (1e3), 75e90.

Perona, E., Mateo, P., 2006. Benthic cyanobacterial assemblages asindicators of nutrient enrichment regimes in a Spanish river.Acta Hydrochimica et Hydrobiologica 34 (1e2), 67e72.

Pfeiffer, C., Palinska, K.A., 2002. Characterization of marinePhormidium isolates-conformity between molecular andecophysiological results. J. Plant Physiol. 159 (6), 591e598.

Quiblier, C., Wood, S., Echenique-Subiabre, I., Heath, M.,Villeneuve, A., Humbert, J.F., 2013. A review of currentknowledge on toxic benthic freshwater cyanobacteria e

ecology, toxin production and risk management. Water Res..http://dx.doi.org/10.1016/j.watres.2013.06.042.

Reynolds, C.S., 2007. Variability in the provision and function ofmucilage in phytoplankton: facultative responses to theenvironment. Hydrobiologia 578 (1), 37e45.

Rippka, R., Deruelles, J., Waterbury, J.B., Herdman, M.,Stanier, R.Y., 1979. Generic assignments, strain histories, andproperties of pure cultures of cyanobacteria. J. Gen. Microbiol.111, 1e61.

Ritchie, R.J., Trautman, D.A., Larkum, A.W.D., 2001. Phosphatelimited cultures of the cyanobacterium Synechococcus arecapable of very rapid, opportunistic uptake of phosphate. NewPhytol. 152 (2), 189e201.

Rodrıguez, V., Aguirre de Carcer, D., Loza, V., Perona, E., Mateo, P.,2007. A molecular fingerprint technique to detect pollution-related changes in river cyanobacterial diversity. J. Environ.Qual. 36 (2), 464e468.

Sabater, S., Vilalta, E., Gaudes, A., Guasch, H., Munoz, I.,Romanı, A., 2003. Ecological implications of mass growth ofbenthic cyanobacteria in rivers. Aquat. Microb. Ecol. 32 (2),175e184.

Schindler, D.W., 1977. Evolution of phosphorus limitation inlakes. Science 195, 260e262.

Schindler, D.W., Hecky, R.E., Findlay, D.L., Stainton, M.P.,Parker, B.R., Paterson, M.J., Beaty, K.G., Lyng, M.,Kasian, S.E.M., 2008. Eutrophication of lakes cannot becontrolled by reducing nitrogen input: results of a 37-yearwhole-ecosystem experiment. Proc. Natl. Acad. Sci. UnitedStates America 105 (32), 11254e11258.

Scott, J.T., Marcarelli, A.M., 2012. Cyanobacteria in freshwaterbenthic environments. In: Whitton, B.A. (Ed.), Ecology ofCyanobacteria II. Their Diversity in Space and Time. Springer,Dordrecht, pp. 271e289.

Sekar, R., Nair, K.V.K., Rao, V.N.R., Venugopalan, V.P., 2002.Nutrient dynamics and successional changes in a lenticfreshwater biofilm. Freshw. Biol. 47, 1893e1907.

Shen, H., Song, L.R., 2007. Comparative studies on physiologicalresponses to phosphorus in two phenotypes of bloom-formingMicrocystis. Hydrobiologia 592, 475e486.

Smith, V.H., 1983. Low nitrogen to phosphorus ratios favordominance by blue-green-algae in lake phytoplankton.Science 221, 669e671.

Smith, A.D., Gilbert, J.J., 1995. Spatial and temporal variability infilament length of a toxic cyanobacterium (Anabaena affinis).Freshw. Biol. 33 (1), 1e11.

USEPA (United States Environmental Protection Agency), 2000.Nutrient Criteria. Technical Guidance Manual. Rivers andStreams. EPA-822-B-00-002. United States EnvironmentalProtection Agency, Washington, pp. 17e28.

Van der Grinten, E., Simis, S.G.H., Barranguet, C., Admiraal, W.,2004a. Dominance of diatoms over cyanobacterial species innitrogen-limited biofilms. Archiv fur Hydrobiologie 161 (1),99e112.

Van der Grinten, E., Janssen, M., Simis, S.G.H., Barranguet, C.,Admiraal, W., 2004b. Phosphate regime structures speciescomposition in cultured phototrophic biofilms. Freshw. Biol.49 (4), 369e381.

Vezie, C., Rapala, J., Vaitomaa, J., Seitsonen, J., Sivonen, K., 2002.Effect of nitrogen and phosphorous on growth of toxic andnontoxic Microcystis strains on intracelular microcystinconcentrations. Microb. Ecol. 43 (4), 443e454.

Whitton, B.A., Mateo, P., 2012. Rivulariaceae. In: Whitton, B.A.(Ed.), Ecology of Cyanobacteria II. Their Diversity in Space andTime. Springer, Dordrecht, pp. 561e592.

Whitton, B.A., Al-Shehri, A.M., Ellwood, N.T.W., Turner, B.L., 2005.Ecological aspects of phosphatase activity in cyanobacteria,eukaryotic algae and bryophytes. In: Turner, B.L., Frossard, E.,Baldwin, D.S. (Eds.), Organic Phosphatase in the Environment.CABI, Wallingford, pp. 205e242.

wat e r r e s e a r c h 4 8 ( 2 0 1 4 ) 6 2 2e6 3 1 631