FEMS Microbiology Ecology, 92, 2016, fiw021
doi: 10.1093/femsec/fiw021
Advance Access Publication Date: 8 February 2016
Research Article
RESEARCH ARTICLE
The role of nitrogen and phosphorus in regulating
Phormidium sp. (cyanobacteria) growth and anatoxin
production
Mark Heath1,∗ , Susie A. Wood2,3 , Roger G. Young2 and Ken G. Ryan4
1
Environmental Science Department, Greater Wellington Regional Council, Shed 39, 2 Fryatt Quay, Pipitea,
Wellington, 6111, New Zealand, 2 Coastal and Freshwater, Cawthron Institute, 98 Halifax St East, Nelson, 7042,
New Zealand, 3 Department of Biological Sciences, University of Waikato, Private bag 3015, Hamilton, 3240,
New Zealand and 4 School of Biological Sciences, Victoria University of Wellington, PO Box 600, Wellington,
6140, New Zealand
∗
Corresponding author: Environmental Science Department, Greater Wellington Regional Council, Shed 39, 2 Fryatt Quay, Pipitea, Wellington 6111,
New Zealand. Tel: +64 4 830 4187; E-mail: [email protected]
One sentence summary: Culture experiment comparing the growth and toxin production of an anatoxin-producing Phormidium sp. strain and a
non-anatoxin-producing Phormidium sp. strain.
Editor: Riks Laanbroek
ABSTRACT
Benthic proliferations of the cyanobacteria Phormidium can cover many kilometres of riverbed. Phormidium can produce
neurotoxic anatoxins and ingestion of benthic mats has resulted in numerous animal poisonings in the last decade.
Despite this, there is a poor understanding of the environmental factors regulating growth and anatoxin production. In
this study, the effects of nitrogen and phosphorus on the growth of two Phormidium strains (anatoxin-producing and
non-anatoxin-producing) were examined in batch monocultures. Cell concentrations were significantly reduced under
reduced nitrogen (ca. <0.100 mM) and phosphorus conditions (ca. <0.003 mM). Cell concentrations and maximum
growth rates were higher for the non-anatoxin-producing strain in all treatments, suggesting there may be an energetic
cost to toxin production. Cellular anatoxin concentrations were lowest (169 fg cell−1 ) under the high-nitrogen and
high-phosphorus treatment. This supports the growth-differentiation balance hypothesis that suggests actively dividing
and expanding cells are less likely to produce secondary-metabolites. Anatoxin quota was highest (>407 fg cell−1 ) in the
reduced phosphorus treatments, possibly suggesting that it is produced as a stress response to growth limiting conditions.
In all treatments there was a 4–5-fold increase in anatoxin quota in the lag growth phase, possibly indicating it may provide
a physiological benefit during initial substrate colonization.
Keywords: nitrogen; phosphorus; anatoxin; homoanatoxin; Phormidium
INTRODUCTION
Proliferations of toxic benthic mat-forming cyanobacteria are
being reported with increasing frequency in freshwater systems
worldwide (Mohamed, El-Sharouny and Ali 2006; Heath, Wood
and Ryan 2011; Quiblier et al. 2013). Concurrently, reports of animal poisonings and fatalities linked to toxin-producing benthic cyanobacteria have increased markedly (Krienitz et al. 2003;
Gugger et al. 2005; Wood et al. 2007, 2010b; Puschner, Hoff and Tor
2008; Faassen et al. 2012). Despite this and the inherent risk to
Received: 27 September 2015; Accepted: 29 January 2016
C FEMS 2016. All rights reserved. For permissions, please e-mail: [email protected]
1
2
FEMS Microbiology Ecology, 2016, Vol. 92, No. 3
human health, research on toxin-producing benthic cyanobacteria remains relatively rudimentary and reactive. There is limited knowledge of the environmental and ecological factors regulating benthic cyanobacterial proliferations. Factors responsible for variations in toxin concentrations observed in the environment are unknown (Wood et al. 2012; Quiblier et al. 2013).
Land-use intensification and urbanization, has led to increased nitrogen and phosphorus concentrations in aquatic systems (Conley et al. 2009). Nitrogen and phosphorus are the major
nutrients favouring the development of planktonic cyanobacteria (Rapala and Sivonen 1998). Consequently, the relationship
between planktonic cyanobacteria and these nutrients has received significant scientific attention (Paerl 1996; Chorus and
Bartrum 1999; Paerl, Hall and Calandrino 2011). In contrast,
there is a limited understanding of the interplay between nitrogen, phosphorus and benthic cyanobacterial proliferations in
lotic systems. In New Zealand rivers, Phormidium spp. blooms
have been associated with elevated nitrogen (ca. >0.1 mg L−1 )
and low-phosphorus concentrations (c. <0.01 mg L−1 ; Wood and
Young 2011, 2012). In contrast, Oscillatoria spp. blooms in the Llobregat River (Spain) have been correlated with high-phosphorus
and low-nitrogen concentrations (Sabater et al. 2003; Vilalta et al.
2003). Additionally, there is limited understanding of the abundance and importance of nitrogen fixing microbes in formation
and maintenance of benthic blooms.
Anatoxin-a (ATX) and its methylene homologue
homoanatoxin-a (HTX) are commonly produced by benthic
cyanobacteria and have been responsible for the majority of
benthic cyanobacteria-related animal poisonings (Quiblier et
al. 2013). ATX and HTX are powerful neuromuscular blocking
agents that act through the nicotinic acetylcholine receptor
(Carmichael 1994). Dihydrogen derivatives, dihydroanatoxin-a
(dhATX) and dihydrohomoanatoxin-a (dhHTX) also bind to
the nicotinic receptor with a reported 10-fold decrease in both
affinity and toxicity compared to ATX (Bates and Rapoport 1979;
Wonnacott et al. 1991; Mann et al. 2012). In affected animals,
these neurotoxins cause convulsions, coma, rigors, cyanosis,
limb twitching, hyper-salivation and/or death.
Anatoxin (often presented as a sum of ATX, HTX, dhATX
and dhHATX) concentrations in Phormidium mats are temporally and spatially variable across New Zealand rivers (Wood
et al. 2010a; Heath, Wood and Ryan 2011; Wood and Young
2011, 2012). Wood et al. (2012) demonstrated that both anatoxinproducing and non-anatoxin-producing genotypes of Phormidium can co-occur in a 1 cm2 section of mat and that total
anatoxin content varies among different anatoxin-producing
genotypes. The total toxin content of a mat can therefore vary
due to the relative abundance of genotypes, and variations in
anatoxin-producing capabilities of genotypes. It is unclear if environmental variables can also influence the amount of anatoxin
produced. Only a limited number of studies have investigated
environmental regulation of anatoxin production and these
have mostly focused on planktonic strains (Rapala et al. 1993; Rapala and Sivonen 1998; Gagnon and Pick 2012). Rapala et al. (1993)
demonstrated that for Dolichospermum (basionym Anabaena) and
Aphanizomenon ATX concentration decreased in high temperatures. Different physicochemical parameters have resulted in
contrasting abundances of microcystin- and non-microcystinproducing genotypes in environmental- (Yoshida et al. 2007;
Briand et al. 2008a) and laboratory-based studies (Vézie et al.
2002; Kardinaal et al. 2007; Briand et al. 2008b). Briand et al.
(2008b) found that microcystin -producing strains of Planktothrix
agardhii were dominant in nitrate limited conditions. To date,
the influence of physio-chemical variables on growth of benthic
anatoxin- and non-anatoxin-producing genotypes has not been
studied.
In this study, two non-axenic, non-heterocytous Phormidium isolates (CAWBG557 = CYN112 and CAWBG48 = VUW14)
were grown in batch monocultures under different concentrations of nitrogen and phosphorus. Strain CAWBG48, isolated
from the Hutt River (North Island, New Zealand), does not produce cyanotoxins (Heath, Wood and Ryan 2010), while strain
CAWBG557, isolated from the Waimakariri River (South Island,
New Zealand), produces ATX, HTX, dhATX and dhHTX (Harland
et al. 2014). The 16S ribosomal RNA sequences (660 bp; JX088082,
GQ451399) of these two strains is similar varying by only 6 bp.
Herein, CAWBG557 is referred to as Phormidium sp. toxic (P-T) and
CAWBG48 as Phormidium sp. non-toxic (P-NT). The growth rates,
cell concentrations and anatoxin quotas (anatoxin content per
cell; P-T only) were assessed. We hypothesized that (i) cell concentrations and growth rates would be reduced in strain P-T, in
comparison to P-NT, because of the greater requirement for nitrogen and phosphorus during anatoxin synthesis and (ii) that
anatoxin concentrations will be lower in the reduced nitrogen
and phosphorus treatments because of the greater requirement
for N and P when synthesizing anatoxin.
MATERIALS AND METHODS
Culture conditions and sampling
Five different variations of MLA growth medium (Bolch and
Blackburn 1996) were used for the experiments; these differed
in their nitrogen (NaNO3 ) and phosphorus (K2 HPO4 .3H2 O) concentrations (Table 1). Concentrations were selected based on
concentrations used by Orr and Jones (1998) and the range of
nutrient conditions observed in a selection of New Zealand
rivers. Potassium levels were maintained by the addition of noninhibitory concentrations of KCl when the K2 HPO4 .3H2 O levels
were reduced. The five MLA nutrient treatments were prepared
in 20 L acid-washed plastic carboys. For each treatment, aliquots
(30 mL) of MLA were pipetted into 33 pre-numbered gamma sterilized polystyrene culture vessels (70 mL, Labserv). Initial inocula
of 6 mg (±0.2 mg) wet weight of either P-NT or P-T were added
to each culture vessel. P-NT and P-T were grown to stationary
phase in MLA medium (full strength) prior to the experiments.
Cultures were incubated at 18◦ C (±1◦ C), under 36 μE m−2 s−1 on
a 12:12 light:dark cycle. The position of culture vessels were randomized, using random number tables following each harvest to
minimize variation in growth due to varying light intensities. After each harvest, gaps created in the culture vessel matrix were
filled with vessels containing water.
Cultures were harvested in triplicate for each treatment at
11 pre-determined time points (days 0, 3, 6, 9, 12, 16, 20, 24,
29, 34 and 39). At harvesting, 15 mL of medium from each vessel was pipetted carefully from the surface to avoid disturbing the filaments. This was filtered (GF/C filters, Whatman) and
Table 1. Nitrate and phosphate concentrations for the five nutrient
treatments.
High nitrate, high phosphate
High nitrate, medium phosphate
High nitrate, low phosphate
Medium nitrate, high phosphate
Low nitrate, high phosphate
Nitrate (mM)
Phosphate (mM)
1.50
1.50
1.50
0.10
0.03
0.1000
0.0025
0.0008
0.1000
0.1000
Heath et al.
the medium from the three replicates combined (to give 45 mL,
which was used) was stored frozen (−20◦ C) for extracellular
toxin analysis. The remaining individual cultures were then homogenized (2 min) using an Ultra-Turrax probe (IKA Laboreknik,
Germany). Homogenization detached filaments from the culture
vessel walls and broke up filaments to assist with cell counts.
The Ultra-Turrax probe was rinsed twice with Milli-Q water, and
the contents were added back to the culture, taking the pottle
volume back up to the original 30 mL volume. The homogenized
cultures were then halved; 15 mL was preserved in Lugols iodine
solution (10% (w/v) potassium iodide, 5% (w/v) iodine, 10% (v/v)
acetic acid) and stored in the dark for later enumeration, and the
remainder was frozen (−20◦ C) for later anatoxin analysis.
Cell enumeration
Sub-samples (0.1 to 5 mL, depending on the cell concentration) from the Lugols preserved samples were transferred to
Utermöhl chambers (Utermöhl 1958) and settled in the dark for
at least 1 h. The length of each filament along one to three transects were measured at 400× magnification using an inverted
microscope (Olympus CK40, Japan) to ensure that at least 50 filaments were measured for each sample. Cell length (n > 50), for
each treatment media were measured at 1000× magnification
(Zeiss photomicroscope II, Germany). The mean cell length was
calculated and used to determine the number of cells per millilitre in the original sample for each treatment. Specific growth
rates, μ (day−1 ) were determined using the differential equation:
μ =
100 ln N2
N1
,
T2 − T1
where N1 and N2 are cell concentrations at times T1 and T2,
respectively. Using the specific growth rate, maximum growth
rates (the fastest growth rate observed between two sequential
sampling days) were determined for each treatment.
Anatoxin analysis
The frozen sub-samples were lyophilized (FreeZone6, Labconco,
USA) and suspended in Milli-Q water (5 mL) containing 0.1%
formic acid and sonicated on ice (30 min; Cole Parmer 8890, Biolab, Auckland, NZ). An aliquot (1 mL) was transferred to an
Eppendorf tube (1.7 mL) and centrifuged (4000 × g, 10 min).
The resulting supernatant was transferred to a HPLC vial. The
frozen filtered sub-sample (1 mL) collected for extracellular
toxin was also transferred to an HPLC vial. All samples were
analysed for ATX, dhATX, HTX and dhHTX analysis using liquid chromatography-mass spectrometry as described in Heath,
Wood and Ryan (2011). The instrument was calibrated with dilutions of authentic standards of ATX (A.G. Scientific, CA) in 0.1%
formic acid and has an anatoxin sensitivity of 0.05 ng μL−1 . For
each treatment, anatoxin quota (the anatoxin per cell) was calculated by subtracting the extracellular anatoxin from the total
anatoxin concentration and dividing by the respective cell concentrations for each replicate.
Statistical analysis
To compare differences in growth rates among treatments
and between P-NT and P-T, a two factor analysis of variance (ANOVA) was conducted, with nutrient treatment and
Phormidium strain as the two factors. Where significant differences between species were observed, we performed sepa-
3
rate ANOVAs for P-NT and P-T to identify differences among
treatments.
To compare the effects of nutrient treatments on cell concentration, intracellular anatoxin quotas and total anatoxin, generalized estimating equations (GEE) were used. A Gaussian response function and an autoregressive correlation structure,
where the correlation between observations are modelled as a
function of the distance (time) between observations was used
(Omar et al. 1999). For each response, treatment was the fixed effect. Differences between strains in cell concentration were determined using a GEE model. Effects were only compared for the
last four sampling points (days 24, 29, 34 and 39) to assess the differences at stationary phase when cultures were at carrying capacity. For the GEE model evaluating intracellular anatoxin concentration, day was set as a covariate. This allowed evaluation of
intracellular anatoxin production over time (39 days). When significant treatment effects were observed, pairwise comparisons
were undertaken to determine which treatments differed.
All statistical tests were carried out using R 2.15.1 (R Development Core Team, 2012). GEE models were performed using Package ‘geePack’ (Halekoh, Hojsgaard and Yan 2006). Significance for
all ANOVA tests and GEE models was set at P ≤ 0.05.
RESULTS
Phormidium growth
An initial lag phase was not observed for P-NT with an average
growth rate across all five treatments of 0.35 (cell doubling rate
per day) between days 0 and 3 (Fig. 1a). This was significantly
higher than P-T (0.0 cell doubling rate per day; Fig. 1b) for the corresponding period (F1,20 = 48.50, P < 0.001; ANOVA). Growth rates
for P-NT were highest in the early exponential growth phase
(days 3–9), with the length of the exponential phase depending
on the initial nitrate and phosphate concentration. In the HNHP
treatment, it took 24 days for P-NT to reach stationary phase
(Fig. 1a).
In the high-nitrate treatments P-T, had a typical growth response for all treatments, with an initial lag phase, an exponential growth phase and a stationary phase (Fig. 1b). For these
treatments in the exponential growth phase, maximum growth
rates occurred between days 3 and 9. In the HNHP treatment it
took 34 days for P-T to reach stationary phase, this was 10 more
days than P-NT. The maximum growth rates for P-NT, across all
treatments, were statistically higher than P-T (F1,20 = 28.79, P
< 0.001). There were no differences in maximum growth rates
among treatments for both P-NT (F4,10 = 1.830, P = 0.20) and P-T
(F4,10 = 0.5, P = 0.73).
Cell concentrations for P-NT were statistically higher (3–4fold) than the corresponding P-T treatment at stationary phase
(Fig. 1; b = −1.265 ± 0.213; W = 35.2; P < 0.001; GEE). Within
P-NT and P-T, pairwise comparisons of cell concentration for
each treatment differed significantly (Fig. S1, Supporting Information) e.g. the cell concentration for treatment HNHP was
significantly greater than MNHP for P-NT (b = −2.519 ± 0.118;
W = 456.6; P < 0.001; GEE). Cell concentrations in the medium
and low-nitrate treatments were reduced with 0.29 and 0.18 ×
106 cells mL−1 for P-NT and 0.10 and 0.04 × 106 cells mL−1 for
P-T at day 39 (Fig. 1). Treatments HNMP and HNLP had greater
cell concentrations (2.0 and 0.68 × 106 cells mL−1 for P-NT and
0.47 and 0.30 × 106 cells mL−1 for P-T at day 39) than both MNHP
and LNHP. After 39 days, HNHP had the highest cell concentration (P-NT 3.8 × 106 cells mL−1 and P-T 0.91 × 106 cells mL−1 ) in
both experiments.
4
FEMS Microbiology Ecology, 2016, Vol. 92, No. 3
Figure 2. Total anatoxin quota for Phormidium toxic (P-T) under the five different
phosphate and nitrate treatments. HNHP (high nitrate high phosphate), HNMP
(high nitrate medium phosphate), HNLP (high nitrate low phosphate), MNMP
(medium nitrate high phosphate), LNMP (low nitrate high phosphate). Each point
is an average of three replicates.
Figure 1. Cell concentrations for; (a) P-NT (non-anatoxin-producing Phormidium
strain) and (b) P-T (anatoxin-producing Phormidium strain) in five different phosphate and nitrate treatments. HNHP (high nitrate high phosphate), HNMP (high
nitrate medium phosphate), HNLP (high nitrate low phosphate), MNMP (medium
nitrate high phosphate), LNMP (low nitrate high phosphate). Each point is an average of three replicates and error bars represent one standard error.
Anatoxin production
Anatoxin quota (the sum of the intracellular ATX, HTX, dhATX
and dhHTX), ranged from 17 to 1037 fg cell−1 with the highest
concentrations in each treatment corresponding to the initial lag
phase (c. days 3 and 6; Fig. 2). The highest anatoxin quota was
measured in MNHP on day 3 (1037 fg cell−1 ). Overall, the HNMP
treatment had higher intracellular anatoxin concentration than
the high-phosphate treatments HNHP, MNHP and LNHP (P ≥
0.012; GEE: Fig. S2, Supporting Information).
HNLP also had a higher anatoxin quota than HNHP (b = 0.4601
± 0.189; W = 10.05; P = 0.002; GEE). There were no other significant treatment effects. Intracellular anatoxin concentration
fluctuated significantly over time (b = −0.046 ± 0.006; W = 60.340;
P < 0.001; GEE), peaking in the initial lag/early exponential
growth phase before decreasing (Fig. 2). At stationary phase (day
39), the intracellular anatoxin concentration decreased to <120
fg cell−1 for all treatments.
Total anatoxin concentrations (intracellular and extracellular) ranged from 38 (day 39, MNHP) to 2157 μg L−1 (day 29, HNMP)
over the 39 days with concentrations generally increasing with
time (Fig. S3, Supporting Information). All pairwise comparisons
of total anatoxin concentration significantly differed between
treatments (Fig. S4, Supporting Information) except for HNHP
and HNLP, which did not differ (b = −0.086 ± 0.162; W = 0.280;
P = 0.596; GEE).
Figure 3. Total extracellular anatoxin (sum of ATX, HTX, dhATX and dhHTX) in
culture media extract at each sampling point for toxic Phormidium strain (P-T).
HNHP (high nitrate high phosphate), HNMP (high nitrate medium phosphate),
HNLP (high nitrate low phosphate), MNMP (medium nitrate high phosphate),
LNMP (low nitrate high phosphate). Each point is an average of three replicates.
The proportion of extracellular toxin was greatest on day 0
for all treatments; for example 53% of MNHPs total anatoxin
concentration was extracellular. Between days 3 and 39, extracellular toxins accounted for <13% of the total toxin load in all
samples (Fig. 3). With the exception of day 0 concentrations,
the extracellular toxin was greatest in the reduced phosphorus
treatments HNMP and HNLP and was predominately detected
in the late exponential/stationary phase (day 20 onwards). The
majority of the extracellular toxin was composed of dihydrogen
derivatives (data not shown). HTX was only detected on four occasions in the extracellular media and ATX was not detected.
DISCUSSION
Phormidium growth under different nutrient regimes
The HNHP treatment had the highest cell concentrations and
longest exponential growth phase for both P-NT and P-T.
In comparison, the two nitrate-reduced treatments had the
Heath et al.
lowest cell concentrations. In situ observations from New
Zealand rivers have shown that the greatest Phormidium cover
usually occurs when dissolved inorganic nitrogen (DIN) concentrations are over a threshold of ca. 0.2 mg L−1 (Wood and Young
2011, 2012; Heath et al. 2013; Wood et al. 2015). Although some
Phormidium spp. can fix atmospheric nitrogen (Bergman et al.
1997), an initial screen of 13 Phormidium cultures showed they
did not contain the nifD gene involved in nitrogen fixation (Heath
unpublished data). This observation supports the low cell concentrations observed in the reduced nitrate treatments and the
positive relationships between Phormidium proliferations and elevated DIN concentrations in New Zealand rivers (Wood and
Young 2012; Heath et al. 2013; Quiblier et al. 2013). Relationships
between planktonic cyanobacteria that cannot fix atmospheric
nitrogen, e.g. Microcystis, in lentic systems with elevated nitrogen levels are well documented (Vrede et al. 2009).
Increasing phosphate concentration in this culture-based
study increased cell biomass. In contrast, Phormidium blooms are
generally observed in New Zealand rivers with low-phosphate
(<0.01 mg L−1 ; Heath, Wood and Ryan 2011; Wood and Young
2012; Wood et al. 2015). Many cyanobacteria are able to store
phosphates as polyphosphates in polyphosphate bodies (also
called volutine granules; Kromkamp 1987) which enables them
to perform two to four cell divisions, comparable to a 4—32-fold
increase in biomass (Mur, Skulberg and Kilen 1999). This may
provide this species with a competitive advantage over other
cyanobacteria/algae especially during the early growth phase.
As Phormidium mats become established they can create internal geochemical conditions (i.e. elevated pH, low dissolved oxygen) that facilitate phosphorus release from incorporated sediments (Wood, Wagenhoff and Young 2014). Under these conditions, iron, aluminium and magnesium reduction leads to the
release of bound phosphates (Ruban and Demare 1998; Khoshmanesh et al. 2002). In a recent study of New Zealand Phormidium mats, Wood et al. (2015) found within Phormidium mat water had 320-fold higher DRP concentrations than the overlying
river water, and in concert with elevated concentrations of elements including iron and aluminium indicating phosphorus release from entrapped sediment. In this study, we provided an
easily accessible phosphorus source without any competition;
hence, the positive relationship observed between cell biomass
and phosphorus concentration.
In contrast to New Zealand Phormidium proliferations,
Sabater et al. (2003) observed Oscillatoriales blooms in the Llobregat river occur when concentrations of water column phosphorus are high and nitrogen low. Sabater et al. (2003) demonstrated that the Oscillatoriales mats had high aminopeptidase
enzymatic activity in these conditions indicating inorganic nitrogen was being obtained from organic sources. Loza, Perona
and Mateo (2014) in a study on factors influencing changes in
species dominance found benthic cyanobacteria have adapted
and specialized to survive in a wide range of nitrogen and phosphorus conditions in both laboratory and field studies. Collectively these data demonstrate benthic Oscillatoriales have a
highly adaptable physiology allowing them to thrive in a wide
range of nitrogen and phosphorus conditions.
Anatoxin production
Results from the few studies undertaken on anatoxin production on planktonic species indicate that growth phase and environmental factors can affect the amount of cellular anatoxin
produced (Rapala et al. 1993; Gagnon and Pick 2012). For example, ATX quota in two Dolichospermum species was higher in
5
nitrogen-depleted than in nitrogen-rich treatments (Rapala et al.
1993). Gagnon and Pick (2012) demonstrated that ATX quota in
Cuspidothrix (basionym Aphanizomenon) issatschenkoi was significantly higher in ‘medium’ (75 mg L−1 NaNO3 ) nitrogen concentrations than low and high concentrations (15 and 1500 mg L−1 ).
These findings are similar to the current study where ATX quotas were lowest in the HNHP treatment, despite having greater
nutrients, higher cell concentrations and a faster growth rates
over a sustained period. These results support the growthdifferentiation balance hypothesis which suggests that actively
dividing and expanding cells are less likely to invest in producing secondary metabolites, while those cells exposed to intermediate levels of resources will have the highest concentrations of metabolites (Herms and Mattson 1992). The reduced
phosphate treatments (HNMP and HNLP) had the highest anatoxin quotas. This is of concern because low water column DRP
and elevated DIN concentrations have been linked to Phormidium proliferations in New Zealand rivers (Wood and Young 2012;
Heath et al. 2013).
The anatoxin quota of P-T for all five treatments peaked in
the lag phase (c. day 3), with the exception of the HNMP treatment that peaked in the early exponential growth phase (c. day
6). Harland et al. (2013) also showed that ATX quota peaked in
the lag/early exponential growth phase coinciding with substrate acquisition. They suggested that ATX may be alleopathic
to other Phormidium cells and/or bacteria. In this study, the peak
in ATX quota occurred during the spread of filaments across the
substrate, suggesting that anatoxins may play an important role
in initial colonization processes. Contrasting growth/toxin relationships have been observed in other benthic species, for example, in cylindrospermopsin-producing benthic Oscillatoria sp.
PCC 6505 toxin peaked in the exponential and stationary growth
phases under varying light levels (Bormans et al. 2013). In planktonic cyanobacteria, ATX quota peaks at different times during
growth phases. For example, in Rapala et al. (1993), a 5–10-fold increase in ATX quota in two Dolichospermum species was observed
in the first two weeks of incubation in the lag/early exponential growth phase. In contrast, Gagnon and Pick (2012) found a
3–5-fold increase in ATX quota in the late exponential phase in
C. issatschenkoi and Gupta, Bhaskar and Rao (2002) recorded the
highest ATX quotas in weeks 5 and 6 of growth in Dolichospermum flos-aquae. Research on microcystin production has generally demonstrated that synthesis increases during the exponential phase reaching a maximum late in this phase (e.g. Orr and
Jones 1998; Lyck 2004; Downing et al. 2005; Yepremian et al. 2007).
This has led some researchers to hypothesize environmental parameters have an indirect effect on microcystin production, as
a result of their direct impact on cell growth (Briand et al. 2008b,
2012). In this study, there was no correlation between growth
and anatoxin production. However, reductions in anatoxin production during the exponential growth phase coincided with decreases in the specific growth rate. The link between specific
growth rate and anatoxin quota is an interesting avenue for future research and may help improve understanding of the ecological role of this toxin.
Anatoxin concentrations in Phormidium mats across New
Zealand rivers are temporally and spatially variable (Wood et al.
2010a; Heath, Wood and Ryan 2011; Wood and Young 2011; Heath
et al. 2013). The total toxin content of a mat can vary due to the
relative abundance of anatoxin and non-anatoxin-producing
genotypes, and variations in anatoxin-producing capabilities of
genotypes (Wood et al. 2010a, 2012; Heath, Wood and Ryan 2011).
This study demonstrates that anatoxin concentration can also
vary due to the stage of growth and nutrient concentrations.
6
FEMS Microbiology Ecology, 2016, Vol. 92, No. 3
The initial inoculums were prepared by breaking up Phormidium mats and taking sub-samples (using tweezers) that were
then placed into culture vessels containing media. The day
0 samples were collected within 5 min of the initial inoculation. This process would have disrupted cells and likely caused
release of anatoxins into the medium explaining the high
extra-cellular ATX content on day 0. Anatoxin degrades rapidly
(Wonnacott et al. 1991), and this may explain the low concentrations throughout the remainder of the experiments.
Comparison of growth and anatoxin production
between non-toxic (P-NT) and toxic (P-T) strains
Limited studies have explored the effect of physicochemical
factors on the selection of toxin-producing and non-toxinproducing genotypes and these have focused on planktonic
microcystin-producing cyanobacteria (Hadas 2005; Kardinaal
et al. 2007; Briand et al. 2008b). In New Zealand, Phormidium
mats can contain anatoxin and non-anatoxin-producing genotypes (Heath, Wood and Ryan 2010; Wood et al. 2012). The degree to which physicochemical influences regulate the relative
proportions of these genotypes is unknown. In this monoculture study, despite equivalent starting inoculums P-NT cell concentrations were approximately 3–4-fold higher than P-T for
each treatment. In addition, the maximum growth rates were
significantly higher for P-NT treatments than for P-T treatments. During periods of extensive Phormidium coverage in
New Zealand rivers, dramatic shifts from toxic to non-toxic
samples (and vice versa) can be observed on weekly timescales (Heath, Wood and Ryan 2011; Wood and Young 2012).
These data suggest that under our culture conditions, the nonanatoxin-producing strain, P-NT, is more likely to outcompete over P-T. However, a competitive advantage cannot be
assumed solely from monoculture experiments (Schatz et al.
2005; Oberhaus et al. 2007; Briand et al. 2008b). For example, when different strains of microcystin-producing and nonmicrocystin-producing strains of P. agardhii were grown together,
the fitness of the non-microcystin-producing strains was greater
than microcystin-producing strains in non-limiting conditions,
while the microcystin-producing strains had a greater fitness in
physiologically limiting conditions (Briand et al. 2008b). Further
experiments using a mixture of toxin and non-toxic Phormidium
are planned.
CONCLUSIONS
Increases in nitrate and phosphate concentration resulted in
greater Phormidium biomass. However, Phormidium spp. proliferations generally occur in New Zealand when water column DRP
concentrations are low. This suggests Phormidium may have a
competitive advantage over other algal species under these conditions, but is outcompeted when DRP concentrations increase.
Anatoxin quota increased when media phosphate concentrations were reduced. This is of considerable concern with aquatic
restoration projects worldwide aiming to lower and limit phosphorus loads. Additionally, in all treatments anatoxin quota was
highest in the lag growth phase, indicating it may provide a
physiological benefit during initial substrate colonization. The
non-toxic strain grew faster than the toxic strain, potentially
suggesting an energetic cost of toxin production. This study
has demonstrated that anatoxin concentration can vary due to
growth stage and nitrogen and phosphorus concentration. Future research aimed at understanding spatial and temporal variability in toxin production and developing predictive models for
forecasting anatoxin concentrations should include these variables.
SUPPLEMENTARY DATA
Supplementary data are available at FEMSEC online.
ACKNOWLEDGEMENTS
Rafael Barbieri (Victoria University) is thanked for assistance
with statistical analysis.
FUNDING
This study was funded by the New Zealand Ministry for Business, Innovation and Employment Cumulative Effects programme (CO1×0803) and by VUW grant 80837. Mark Heath
thanks the Bay of Plenty Regional Council, Hawke’s Bay Regional
Council, Horizons Regional Council, Greater Wellington Regional
Council, Environment Canterbury and Environment Southland
for financial support.
Conflict of interest. None declared.
REFERENCES
Bates HA, Rapoport H. Synthesis of anatoxin a via intramolecular
cyclization of iminium salts. J Am Chem Soc 1979;101:1259–65.
Bergman B, Gallon JR, Rai AN et al. N2 fixation by nonheterocystous
cyanobacteria.
FEMS
Microbiol
Rev
1997;19:139–85.
Bolch C, Blackburn S. Isolation and purification of Australian isolates of the toxic cyanobacterium Microcystis aeruginosa Kütz.
J Appl Phycol 1996;8:5–13.
Bormans M, Lengronne M, Brient L et al. Cylindrospermopsin accumulation and release by the benthic cyanobacterium Oscillatoria sp. PCC 6506 under different light conditions and
growth phases. B Environ Contam Tox 2013;91:1–5.
Briand E, Bormans M, Quiblier C et al. Evidence of the cost of
the production of microcystins by Microcystis aeruginosa under differing light and nitrate environmental conditions. Plos
One 2012;7:e29981.
Briand E, Gugger M, Francois JC et al. Temporal variations in the
dynamics of potentially microcystin-producing strains in a
bloom-forming Planktothrix agardhii (cyanobacterium) population. Appl Environ Microb 2008a;74:3839–48.
Briand E, Yéprémian C, Humbert J-F et al. Competition between microcystin- and non-microcystin-producing Planktothrix agardhii (cyanobacteria) strains under different
environmental conditions. Environ Microbiol 2008b;10:
3337–48.
Carmichael WW. The toxins of cyanobacteria. Sci Am 1994;13:
64–72.
Chorus I, Bartrum J. Toxic Cyanobacteria in Water: A Guide to Their
Public Health Consequences, Monitoring and Management. London and New York: E & FN Spon, 1999.
Conley DJ, Paerl HW, Howarth RW et al. Controlling eutrophication: nitrogen and phosphorus. Science 2009;323:1014–5.
Downing TG, Sember CS, Gehringer MM et al. Medium N:P ratios and specific growth rate co-modulate microcystin and
protein content in Microcystis aeruginosa PCC7806 and Maeruginosa UV027. Microbial Ecol 2005;49:468–73.
Faassen EJ, Harkema L, Begeman L et al. First report
of (homo)anatoxin-a and dog neurotoxicosis after
Heath et al.
ingestion of benthic cyanobacteria in the Netherlands.
Toxicon 2012;60:378–84.
Gagnon A, Pick FR. Effect of nitrogen on cellular production
and release of the neurotoxin anatoxin-a in a nitrogen-fixing
cyanobacterium. Front Microbiol 2012;3:211.
Gugger M, Lenoir S, Berger C et al. First report in a river in France
of the benthic cyanobacterium Phormidium favosum producing anatoxin-a associated with dog neurotoxicosis. Toxicon
2005;45:919–28.
Gupta N, Bhaskar ASB, Rao PVL. Growth characteristics and toxin
production in batch cultures of Anabaena flos-aquae: effects of
culture media and duration. World J Microb Biot 2002;18:29–35.
Hadas O. Ecological implications of the emergence of non-toxic
subcultures from toxic Microcystis strains. Environ Microbiol
2005;7:798–805.
Halekoh U, Hojsgaard S, Yan J. The R package geepack for generalized estimating equations. J Stat Softw 2006;15:1–11.
Harland F, Wood S, Moltchanova E et al. Phormidium autumnale
growth and anatoxin-a production under iron and copper
stress. Toxins 2013;5:2504–21.
Harland F, Wood SA, Broady P et al. Polyphasic studies of benthic
freshwater cyanobacteria in Canterbury, New Zealand. New
Zeal J Bot 2014;52:116–35.
Heath MW, Wood SA, Brasell KA et al. Development of habitat
suitability criteria and in-stream habitat assessment for benthic cyanobacteria Phormidium. River Res Appl 2013;31:98–108.
Heath MW, Wood SA, Ryan KG. Polyphasic assessment of freshwater benthic mat forming cyanobacteria in New Zealand.
FEMS Microbiol Ecol 2010;73:95–109.
Heath MW, Wood SA, Ryan KG. Spatial and temporal variability in Phormidium and associated anatoxin production in two
New Zealand rivers. Aquat Microb Ecol 2011:64:69–79.
Herms DA, Mattson WJ. The dilemma of plants: to grow or defend. Q Rev Biol 1992;67:283–35.
Kardinaal WEA, Tonk L, Janse I et al. Competition for light between toxic and nontoxic strains of the harmful cyanobacterium Microcystis. Appl Environ Microb 2007;73:2939–46.
Khoshmanesh A, Hart BT, Duncan A et al. Luxury uptake of phosphorus by sediment bacteria. Water Res 2002:36:774–8.
Krienitz L, Ballot A, Kotut K et al. Contribution of hot spring
cyanobacteria to the mysterious deaths of lesser flamingos
at Lake Bogoria, Kenya. FEMS Microbiol Ecol 2003;43:141–8.
Kromkamp J. Formation and functional significance of storage
products in cyanobacteria. New Zeal J Mar Fresh 1987;21:
457–65.
Loza V, Perona E, Mateo P. Specific responses to nitrogen and
phosphorus enrichment in cyanobacteria: factors influencing changes in species dominance along eutrophic gradients.
Water Res 2014;48:622–31.
Lyck S. Simultaneous changes in cell quotas of microcystin,
chlorophyll a, protein and carbohydrate during different
growth phases of a batch culture experiment with Microcystis
aeruginosa. J Plankton Res 2004;26:727–36.
Mann S, Cohen M, Chapuis-Hugon F et al. Synthesis, configuration assignment, and simultaneous quantification by liquid chromatography coupled to tandem mass spectrometry,
of dihydroanatoxin-a and dihydrohomoanatoxin-a together
with the parent toxins, in axenic cyanobacterial strains and
in environmental samples. Toxicon 2012;60:1404–14.
Mohamed ZA, El-Sharouny HM, Ali WSM. Microcystin production in benthic mats of cyanobacteria in the Nile River and
irrigation canals, Egypt. Toxicon 2006;47:584–90.
Mur L, Skulberg O, Kilen U. Cyanobacteria in the environment.
In: Chorus I, Bartram J (eds). Toxic Cyanobacteria in Water: A
7
Guide to Their Public Health Consequences, Monitoring and Management. London: E & P Spon, 1999, 41–111.
Oberhaus L, Briand JF, Leboulanger C et al. Comparative effects
of the quality and quantity of light and temperature on
the growth of Planktothrix agardhii and P. rubescens. J Phycol
2007;43:1191–9.
Omar RZ, Wright EM, Turner RM et al. Analysing repeated measurements data: a practical comparison of methods. Stat Med
1999;18:1587–603.
Orr PT, Jones GJ. Relationship between microcystin production
and cell division rates in nitrogen-limited Microcystis aeruginosa cultures. Limnol Oceanogr 1998;43:1604–14.
Paerl HW. A comparison of cyanobacterial bloom dynamics in
freshwater, estuarine and marine environments. Phycologia
1996;35:25–35.
Paerl HW, Hall NS, Calandrino ES. Controlling harmful cyanobacterial blooms in a world experiencing anthropogenic and
climatic-induced change. Sci Total Environ 2011;409:1739–45.
Puschner B, Hoff B, Tor ER. Diagnosis of anatoxin-a poisoning in
dogs from North America. J Vet Diagn Invest 2008:20:89–92.
Quiblier C, Wood SA, Echenique-Subiabre I et al. A review of current knowledge on toxic benthic freshwater cyanobacteria—
ecology, toxin production and risk management. Water Res
2013;47:5464–79.
R Development Core Team. R: a language and environment for
statistical computing. 2.15.1 edn. Vienna, Austria: R Foundation for Statistical Computing, 2012.
Rapala J, Sivonen K. Assessment of environmental conditions
that favor hepatotoxic and neurotoxic Anabaena spp. strains
cultured under light limitation at different temperatures. Microbial Ecol 1998;36:181–92.
Rapala J, Sivonen K, Luukkainen R et al. Anatoxin-a concentration in Anabaena and Aphanizomenon under different environmental conditions and comparison of growth by toxic and
non toxic Anabaena-strains—a laboratory study. J Appl Phycol
1993;5:581–91.
Ruban V, Demare D. Sediment phosphorus and internal phosphate flux in the hydroelectric reservoir of Bort-les-Orgues,
France. In: Amiard JC, Rouzic B, Berthet B, Bertru G (eds).
Oceans, Rivers and Lakes: Energy and Substance Transfers at Interfaces, Vol. 131. Netherlands: Springer, 1998, 349–59.
Sabater S, Vilalta E, Gaudes A et al. Ecological implications of
mass growth of benthic cyanobacteria in rivers. Aquat Microb
Ecol 2003;32:175–84.
Schatz D, Keren Y, Hadas O et al. Ecological implications of the
emergence of non-toxic subcultures from toxic Microcystis
strains. Environ Microbiol 2005;7:798–805.
Utermöhl H. Zur Vervollkommung der quantitativen phytoplankton methodik. Verhand Internat Verein Theor Angew Limnol 1958;9:1–38.
Vézie C, Rapala J, Vaitomaa J et al. Effect of nitrogen and phosphorus on growth of toxic and nontoxic Microcystis strains
and on intracellular microcystin concentrations. Microbial
Ecol 2002;43:443–54.
Vilalta E, Guasch H, Muñoz I et al. Ecological factors that cooccur with geosmin production by benthic cyanobacteria.
The case of the Llobregat River. Arch Hydrobiol Algol Studies
2003;109:579–92.
Vrede T, Ballantyne A, Mille-Lindblom C et al. Effects of N:P loading ratios on phytoplankton community composition, primary production and N fixation in a eutrophic lake. Freshwater Biol 2009;54:331–44.
Wonnacott S, Jackman S, Swanson KL et al. Nicotinic pharmacology of anatoxin analogs. II. Side chain structure-activity
8
FEMS Microbiology Ecology, 2016, Vol. 92, No. 3
relationships at neuronal nicotinic ligand binding sites. J
Pharmacol Exp Ther 1991;259:387–91.
Wood S, Depree C, Brown L et al. Entrapped sediments as a source
of phosphorus in epilithic cyanobacterial proliferations in
low nutrient rivers. PloS One 2015;10:e0141063.
Wood SA, Heath MW, Kuhajek J et al. Fine-scale spatial variability in anatoxin-a and homoanatoxin-a concentrations in
benthic cyanobacterial mats: implication for monitoring and
management. J Appl Microbiol 2010a;109:2011–18.
Wood SA, Heath MW, McGregor G et al. Identification of a benthic
microcystin producing Planktothrix sp. and an associated dog
poisoning in New Zealand. Toxicon 2010b;55:897–903.
Wood SA, Selwood AI, Rueckert A et al. First report of
homoanatoxin-a and associated dog neurotoxicosis in New
Zealand. Toxicon 2007;50:292–301.
Wood SA, Smith FMJ, Heath MW et al. Within-mat variability in
anatoxin-a and homoanatoxin-a production among benthic
Phormidium (cyanobacteria) strains. Toxins 2012;4:900–12.
Wood SA, Wagenhoff A, Young R. The effect of flow and nutrients on Phormidium abundance and toxin production in rivers
in the Manawatu-Whanganui region. Prepared for Horizons
Regional Council. Cawthron Report No. 2575. Nelson, New
Zealand, 2014.
Wood SA, Young RG. Benthic cyanobacteria and toxin production
in the Manawatu -Wanganui region. Cawthron Report 1959.
Nelson, New Zealand, 2011.
Wood SA, Young RG. Review of benthic cyanobacteria monitoring programme. Cawthron Report 2217. Nelson, New Zealand,
2012.
Yepremian C, Gugger MF, Briand E et al. Microcystin ecotypes
in a perennial Planktothrix agardhii bloom. Water Res 2007;41:
4446–56.
Yoshida M, Yoshida T, Takashima Y et al. Dynamics of
microcystin-producing and non-microcystin-producing Microcystis populations is correlated with nitrate concentration
in a Japanese lake. FEMS Microbiol Lett 2007:266:49–53.