Environ Sci Pollut Res (2010) 17:1581–1590

DOI 10.1007/s11356-010-0345-8

RESEARCH ARTICLE

Dynamics of the water bloom-forming Microcystis

and its relationship with physicochemical factors in Lake
Xuanwu (China)
Yao Xu & Guoxiang Wang & Wenbin Yang & Renhui Li

Received: 4 December 2009 / Accepted: 17 May 2010 / Published online: 30 May 2010
# Springer-Verlag 2010

Abstract from 3.6×102 cells ml−1 to a peak of 3.8×106 cells ml−1.

Purpose China’s freshwater subtropical shallow lakes are The ratio of mcyB-containing Microcystis subpopulations to
increasingly eutrophic and susceptible to production of the total Microcystis varied, ranging from 0.1% to 12.8%.
heavy growths or water blooms of cyanobacteria. One The abundance of Microcystis containing the mcyB micro-
example was the heavy water bloom that occurred for the cystin gene was shown to be significantly correlated with
first time in Lake Xuanwu, in 2005, an urban lake located concentrations of total phosphorus and phosphate.
in Nanjing city. The aim of the present study was to Conclusions Water temperature was the primary factor
determine dynamics of water bloom dominating Micro- affecting Microcystis abundance in the lake, and phospho-
cystis in this lake. Meanwhile, the relationship between rus loading was shown to be a main factor in governing the
environmental factors and Microcystis populations was also growth of both microcystin-producing genotype and total
analyzed. Microcystis population. Higher Microcystis cell counts at
Materials and methods Molecular detection, using quantita- the bottom of the water column before and after appearance
tive polymerase chain reaction, was used in this lake to study of water blooms in Lake Xuanwu suggested that Micro-
the dynamics of the cyanobacterial community, its Micro- cystis numbers in the water column, especially at the
cystis populations, and the microcystin-producing Micro- bottom of the water column, play an important role in
cystis genotypes from August 2005 to November 2006. forming the next water bloom.
Results It was shown that Microcystis wesenbergii and
Microcystis aeruginosa were the main components of the Keywords Microcystis . mcyB . Quantitative PCR .
cyanobacterial blooms in Lake Xuanwu, and they coexisted Physicochemical factor . Water blooms
with species of the filamentous cyanobacterial genera
Anabaena, Planktothrix, and Anabaenopsis. Microcystis
cells were detected during the entire survey period and in 1 Background, aim, and scope
all sample sites. The cell abundance of Microcystis ranged
In freshwater ecosystems, harmful algal blooms caused by
Responsible editor: Hailong Wang several cyanobacteria genera, including Anabaena, Apha-
nizomenon, Cylindrospermopsis, Microcystis, and Plankto-
Electronic supplementary material The online version of this article
(doi:10.1007/s11356-010-0345-8) contains supplementary material,
thrix (Oscillatoria), have been extensively documented
which is available to authorized users. (Chorus and Bartram 1999). Among these genera, Micro-
Y. Xu : G. Wang : W. Yang
cystis is well known to be the most ecologically damaging
College of Geography Science, Nanjing Normal University, group due to its widespread prevalence in water bodies and
Nanjing 210046, China its toxicity to aquatic and terrestrial organisms (Kuiper-
Goodman et al. 1999; Carmichael et al. 2001). Production
Y. Xu : R. Li (*)
of hepatotoxic microcystin (MC) by Microcystis species has
Institute of Hydrobiology, The Chinese Academy of Sciences,
Wuhan 430072, China led to intensive research on this genus especially the
e-mail: [email protected] underlying causes of such water blooms.
1582 Environ Sci Pollut Res (2010) 17:1581–1590

High nutrient levels in eutrophic and poorly flushed eutrophic freshwaters worldwide. Kurmayer and Kutzenberger
waters are regarded as the main influencing factors for the (2003) reported that the proportion of MC genotypes with
growth of cyanobacteria water blooms (Philipp et al. 1991; Microcystis was low in Lake Wannsee, Berlin. Vaitomaa et
Paerl et al. 2001). Considerable effort has also been made al. (2003) detected both nontoxic and toxic Anabaena in
to elucidate the relationship between cyanotoxin production Lake Hiidenvesi, Helsinki. The presence of toxic genotypic
and phosphorus (P) and nitrogen (N) concentration (Vézie Microcystis had been found to be positively correlated with
et al. 2002; Downing et al. 2005). Higher P concentrations the nitrate concentrations or total phosphorus (TP) concen-
have been found to cause higher or lower or unchanged trations (Yoshida et al. 2007; Rinta-Kanto et al. 2009).
cyanotoxin levels, respectively, based on investigations However, these studies focused on spatial and temporal
from different laboratories (Utkilen and Gjolme 1995; Oh dynamics of toxic Microcystis during the water bloom
et al. 2000). Field studies have also reached inconsistent seasons and did not pay much attention to the dynamics of
conclusions, with MC production being found to be either Microcystis spp. under unfavorable environmental conditions
positively or negatively correlated with various forms of P before and after the water bloom season. The present study
and N (Wicks and Thiel 1990; Lahti et al. 1997). However, was primarily designed to examine the dynamics of Micro-
little fieldwork has been done to examine how nutrients cystis populations using quantitative PCR method in an
affect the growth of toxic and nontoxic subpopulations urban lake, Lake Xuanwu, from August 2005 to
within Microcystis populations. November 2006. The aim of this study was to (1)
Several studies indicate that Microcystis species rapidly determine temporal and vertical dynamics of water
grew in the water column at the end of spring and then bloom dominating Microcystis in Lake Xuanwu during
dominated during summer period (Preston et al. 1980; the entire year, (2) analyze environmental factors closely
Reynolds et al. 1981). However, Microcystis occurrence at related to Microcystis populations and MC-producing
unfavorable environmental conditions in winter and early Microcystis subpopulations, and (3) provide an insight
spring was also considered to play an important role in the into overwintering of Microcystis and summer recruitment
formation of subsequent water blooms, thus many studies of its numbers in a shallow urban lake.
have paid attention to cyanobacterial recruitment as
determining the contribution of Microcystis in sediments
to subsequent water blooms in the water column (Brunberg 2 Materials and methods
and Blomqvist 2002; Verspagen et al. 2004).
In China, it has recently been shown that many 2.1 Study site
freshwater shallow lakes in subtropical areas have encoun-
tered problems caused by eutrophication and cyanobacterial Lake Xuanwu (32°4′N, 118°47′E), a state-level scenic spot
blooms. Elucidating the mechanism of water bloom in Nanjing city, China, is an urban shallow lake (3.7 km2;
formation and the factors controlling cyanobacterial water mean depth, 1.9 m) located in the highly developed and
blooms have become two major tasks for scientists and lake densely populated Yangtze Delta of Eastern China. Along
managers in China. Therefore, understanding the dynamics with rapid development of the economy during the last
of Microcystis and MC-producing Microcystis in these three decades, increasing external nutrition loads have led
lakes is of great importance. Among these shallow lakes, to hypereutrophication of the lake and a heavy cyanobacte-
the larger ones such as Lake Taihu and Caohu have been rial bloom (Chl-a 671 μg/L) during the summer of 2005.
studied extensively (Chen et al. 2003; Ke et al. 2008). In
addition, not only large shallow lakes but also small urban 2.2 Sampling method and environmental parameters
lakes located at many populated cities in Eastern and
Central China have been found to be eutrophic or Monthly water samples were collected at a depth of 20 cm
hypertrophic and have experienced water blooms (Fan et below the surface at four stations as shown in Fig. 1, from
al. 2005). Different from these larger lakes, urban lakes are August 2005 to November 2006. In addition, monthly
characterized by their small size, subject to resident samples of surface water (at 20 cm), intermediate water (at
activities, less affected by hydrodynamics, and experience 80 cm), and bottom water (at 150 cm) were collected on
less resuspension of sediments. However, no studies have windless sunny days from January to June 2006. One liter
been conducted concerning the dynamics of water bloom- water sample was preserved with Lugol’s solution, for
forming cyanobacteria and MC-producing cyanobacteria in counting of algal cells, and kept in a refrigerator before
these Chinese urban lakes. laboratory analysis, and 500 mL water sample, used for
Recent studies on diversity and/or dynamics of MC- DNA extraction, was filtered through 0.22-μm polycarbon-
producing cyanobacteria using polymerase chain reaction ate membrane filters (Millipore), which were immediately
(PCR)-based molecular approaches have been performed in frozen (−20°C) until processing.
Environ Sci Pollut Res (2010) 17:1581–1590 1583

2.5 Quantitative PCR (qPCR)

The quantitative PCR (qPCR) assays were used to quantita-

tively detect three genetic elements: cyanobacteria-specific
16S rRNA gene (cyano-16S), Microcystis-specific 16S rRNA
gene (mic-16S), and the mcyB gene. Results were presented
as total cyanobacteria, Microcystis, and toxic Microcystis cell
densities, respectively, by converting the resultant gene copy
numbers of these three genes into cell equivalents
corresponding to the standard strain Microcystis aeruginosa
PCC7806. Table 1 shows the primers used in this study in
which DNA sequence specificities were confirmed using a
BLAST search of the DDBJ/EMBL/GenBank databases
(Altschul et al. 1997).
External standards used to determine the cyanobacteria
16S rRNA, Microcystis 16S rRNA, and mcyB gene copy
numbers were prepared using the genomic DNA from M.
aeruginosa strain PCC7806. Copy numbers of the three
genes were estimated on the basis of DNA fragment sizes
Fig. 1 Sampling locations in Lake Xuanwu during 2005 - 2006 (258 bp for the cyano-16S, 212 bp for the mic-16S, and
78 bp for mcyB). A tenfold dilution series of the DNAs


TP, total nitrogen (TN), phosphate (PO3
4 ), nitrate (NO3 ), were prepared and amplified with the cyano-16S gene, mic-


ammonium (NHþ 4 ), and nitrite (NO2 ) were analyzed using the 16S gene, and mcyB qPCR assays.
automated wet chemistry analyzer (SAN++, Skalar, Holland). Amplifications and quantifications were performed using
Samples for chlorophyll a measurement were filtered through an ABI Prism 7000 real-time PCR detection (Applied
47-mm glass fiber filter (GF/C; Whatman, Maidstone, Kent, Biosystems, Foster City, CA, USA) equipped with the ABI
UK), and the filters were then grounded using 90% acetone in Prism 7000 SDS fluorescence detection system and
the dark (4°C). The chlorophyll a concentrations were software (version 1.1).
spectrophotometrically determined after extraction in acetone All qPCR reactions were performed in a total
as described by Zhang and Huang (1991). volume of 50 μL containing 25 μL 2× SYBR Green
Water temperature, dissolved oxygen (DO), oxidation- real-time PCR Master Mix (Toyobo, Osaka, Japan),
reduction potential (ORP), and pH were measured via a 0.5 pmol each primer, and 1 μL DNA from a standard
multiparameter meter (YSI 6820, Yellow Spring Instruments, strain or lake water sample and brought to 50 μL with
USA). Water transparency was measured with a 20-cm sterile ultra-pure water. For mcyB, 0.2 pmol of each
diameter black and white Secchi disk. primer was used. Each PCR reaction was run in triplicate
on a 96-well plate (ABI, Foster City, CA, USA) sealed
2.3 Microscopic examination with optical quality sealing tape (ABI, Foster City, CA,
USA). Two negative controls without DNA were included
Fresh and fixed samples of phytoplankton were identified for each PCR run. The PCR program was performed as
and counted under a microscope (BX51, Olympus, Tokyo, follows: The first step was an initial preheating for 1 min
Japan) connected with a digital camera system. at 95°C. For cyano-16S, the initial preheating step was
followed by 40 cycles: 95°C for 20 s, 59°C for 30 s, and
2.4 DNA extraction 72°C for 30 s. For mic-16S, the initial preheating step was
followed by 40 cycles: 95°C for 25 s, 53°C for 30 s, and
The membrane filters with collected phytoplankton cells 70°C for 60 s. In addition, for mcyB, the initial preheating
were cut into small pieces and suspended in lysis buffer step was followed by 40 cycles: 95°C for 15 s, 59°C for
(40 mM EDTA, 400 mM NaCl, 50 mM Tris-hydrochloride, 30 s, and 72°C for 30 s. The melting temperature for the
pH 9.0). Procedures were performed as described by Rinta- qPCR products was determined using the manufacturer’s
Kanto et al. (2005). Purified DNAs were suspended in software. Dilutions of the extracted PCC7806 genomic
sterile 1× TE buffer, pH 8.0. The concentration and purity DNA, for making the standard curve, also run with each
of extracted DNA was determined by a spectrophotometer analysis.
at 260 and 280 nm. The DNA was stored at −20°C prior to The cell abundance was inferred from the standard curve
quantitative PCR analysis. (cell abundance vs Ct) determined in each assay.
1584 Environ Sci Pollut Res (2010) 17:1581–1590

Table 1 Polymerase chain reaction primer sets used in this study

Target gene Primer Primer sequence (5′–3′) Reference

Cyanobacterial 16S rRNA CYA106 F CGGACGGGTGAGTAACGCGTGA Nübel et al. 1997

CYA359 R CCCATTGCGGAAAATTCCCC Nübel et al. 1997
Microcystis 16S rRNA 209F (Mic-16S F) GCCGCRAGGTGAAAMCTAA Neilan et al. 1997
409R (Mic-16S R) AATCCAAAGACCTTCCTCCC Neilan et al. 1997
Microcystis mcyB 30F (mcyB F) CCTACCGAGCGCTTGGG Kurmayer and Kutzenberger 2003
108R (mcyB R) GAAAATCCCCTAAAGATTCCTGAGT Kurmayer and Kutzenberger 2003

F forward primer, R reverse primer

2.6 Statistical analyses to as high as 3.8×106 cells ml−1 on 6 August 2005.

Following this peak, Microcystis cells declined until March
To examine correlations between cell abundances of Micro- 2006 (4.6×102 cells ml−1), and then increased gradually up
cystis, mcyB-containing Microcystis, percentage of mcyB- to another peak in August 2006 (2.8×105 cells ml−1). The
containing Microcystis, and environmental variables, SPSS lowest cell abundance of Microcystis was detected in
statistical software (version 13.0) was used for all analyses. November 2006 (3.6×102 cells ml−1). Similarly, mcyB
Data were available from all of the following environmental genotype, reflecting the presence of MC-producing Micro-
variables: temperature (°C), TP (milligrams per liter), TN cystis, was detected throughout the study period in Lake
(milligrams per liter), PO4–P (milligrams per liter), NH4–N Xuanwu, and mcyB-containing Microcystis cells fluctuated
(milligrams per liter), NO
3
N (milligrams per liter), from 15 to 9.0×103 cells ml−1. As shown in Fig. 4, the
NO
2
N (milligrams per liter), the ratio of TN to TP (TN/ dynamics of the mcyB genotypes in this lake revealed that
TP), pH, chl-a (micrograms per liter), Secchi depth (meter), proportions of mcyB-containing subpopulations in total
ORP (millivolt), and DO (milligrams per liter). Microcystis populations during the survey period were
All the data presented in the study are the combined found to be lower, ranging from 0.1% to 12.8%.
results of the four different sampling sites.
3.3 Correlations between the dynamics of the Microcystis,
mcyB genotype, and environmental factors
3 Results
Variations in environmental factors during the study period
3.1 Microscopic examination were recorded. As shown in Fig. 5a, water temperatures
ranged between 4.4°C and 30.8°C, and the period for water
Phytoplankton and cyanobacterial components were identified temperature below 15°C was observed to be almost
and counted by microscopic examination (Fig. 2). According 5 months (November 2005 to March 2006) during the
to the microscopic counting (see the supplementary Fig. 1), survey seasons. The concentrations of TN and TP were
Microcystis species, mainly Microcystis wesenbergii and M. from 488 to 2,792 μg L−1 and from 51 to 686 μg L−1,
aeruginosa, were observed in Lake Xuanwu on almost all respectively (Fig. 5b), while variations of NO
3
N and
sampling dates except February 2005, May 2005, and PO4–P ranged from 94 to 1,664 μg L−1 and from 9 to
November 2006. Species of filamentous cyanobacteria genera 209 μg L−1, respectively (Fig. 5c). Levels of NH4–N and
Anabaena, Planktothrix, and Anabaenopsis were found to
coexist with Microcystis during the sampling period (Supple- 300
mentary Fig. 1).
Microcystis cell abundances

250
( 104 cells•ml-1)

3.2 Dynamics in the cell abundances of Microcystis 200

150
Microcystis cells were detected at all sample sites in Lake
100
Xuanwu during the whole survey period. Based on qPCR
assays, patterns of temporal dynamics among the cyano- 50

bacterial community, Microcystis population, and the mcyB 0

genotype subpopulation, were found to be similar except Aug-05 Oct-05 Dec-05 Feb-06 Apr-06 Jun-06 Aug-06 Oct-06

during November 2005 and November 2006 (Fig. 3). Cell Fig. 2 Microcystis cell numbers by microscope in Lake Xuanwu during
abundance of Microcystis ranged from 3.6×102 cells ml−1 the study period. The error bars indicate standard deviations (n=4)
Environ Sci Pollut Res (2010) 17:1581–1590 1585

Fig. 3 Changes of cells abundan-

ces for total cyanobacteria,
Microcystis, and mcyB-containing
Microcystis based on real-time
polymerase chain reaction detec-
tions using primers of cyno-16S,
mic-16S, and mcyB, respectively.
The error bars indicate standard
deviations (n=4)

−1
NO
2
N were shown to be 40 to 551 μg L and 6 to Microcystis, Y2 is the cell number of mcy genotype, and Y3 is
−1
74 μg L , respectively (Fig. 5d). Chlorophyll a concen- the cell number of cyanobacteria.
trations ranged from a minimum of 3.0 μg L−1 in March
2006 to a maximum of 697.3 μg L−1 in August 2005. 3.4 Dynamics of Microcystis in the water column
As shown in Table 2, amounts of each gene examined at different depths during winter and spring
were positively correlated with chlorophyll a, TP, PO4–P,
and pH and negatively correlated with TN:TP ratio. Water Using the qPCR method, Microcystis cells were detected in
temperature was positively correlated with the number of the whole water column during the 2006 survey and ranged
Microcystis and cyanobacterial cells, rather than mcyB from 4.6×102 to 2.0×105 cells ml−1 (Fig. 6). It was shown
genotype levels. The percentage of mcyB genotype within in 2006 that Microcystis cells declined from January to
the Microcystis population was shown to be positively April then sharply increased in June.
correlated with TN, NO
3
N, and TN:TP ratio and Variation in Microcystis cells among different depths in
negatively correlated with PO4–P. In addition, no significant the water column was observed from January to June 2006,
correlation was found between N concentration and cell and vertical profiles of Microcystis cell density (Fig. 6)
numbers of Microcystis and mcyB gene-containing Micro- revealed that Microcystis abundance at the bottom of the
cystis. Major environmental factors were taken to estimate water column were higher than that at the surface and middle
the contributions to the growths of phytoplankton by of the water column in January, May, and June 2006.
stepwise regressions analysis. Equations were given below:

lnðX 1Þþ1:002 lnðX 2Þ
Y 1 ¼ 1014:689þ1:081 ;
4 Discussion
8:363þ0:698 lnðX 1Þþ1:034 lnðX 2Þ
Y 2 ¼ 10 ;

lnðX 1Þþ0:691 lnðX 2Þ
1:086 lnðX 3Þ 4.1 Dynamics of cyanobacteria and Microcystis
Y 3 ¼ 1015:212þ0:567 ;

where X1 is the concentration of PO4, X2 is the concentration This outbreak of a heavy water bloom in Lake Xuanwu has
of TP, X3 is the concentration of NO3, Y1 is cell number of led to several studies for investigating the compositions of

Fig. 4 Percentage of Microcys-

tis cells in total cyanobacterial
cells and mcyB-containing
Microcystis cells for total
numbers of Microcystis cells in
Lake Xuanwu from August
2005 to November 2006, using
real-time polymerase chain re-
action. The error bars indicate
standard deviations (n=4)
1586 Environ Sci Pollut Res (2010) 17:1581–1590

Fig. 5 Phytoplankton and environmental parameters for the lake from nitrogen (TN) and TP concentrations. c Nitrate (NO
3 ) and phosphate
þ

August 2005 to November 2006. a Changes in the density of (PO3
4 ) concentrations. d Ammonium (NH4 ) and nitrite (NO2 )
chlorophyll a (chl-a) concentration and water temperatures. b Total concentrations. The error bars indicate standard deviations (n=4)

Table 2 Correlations matrix between cyanobacteria and physicochemical variables

Variable Mic-16S Cyano-16S mcyB %Ba

r P n r P n r P n r P n

Water temperature 0.64 ** 16 0.66 ** 16 0.31 ns 16 −0.74 ** 16

DO 0.14 ns 16 −0.04 ns 16 0.35 * 16
pH 0.46 ** 16 0.56 ** 16 0.39 * 16 −0.44 ** 16
ORP −0.60 ** 16 −0.53 * 16 −0.76 ** 16 −0.03 ns 16
Secchi depth −0.85 ** 16 −0.92 ** 16 −0.69 ** 16
TN −0.05 ns 16 −0.18 ns 16 0.25 ns 16 0.50 ** 16
TP 0.63 ** 16 0.78 ** 16 0.69 ** 16 −0.24 ns 16
TN:TP −0.66 ** 16 −0.87 ** 16 −0.57 ** 16 0.43 ** 16
NO
3 −0.32 ns 16 −0.41 * 16 −0.02 ns 16 0.59 ** 16
NHþ4 0.15 ns 16 0.13 ns 16 0.16 ns 16 0.06 ns 16
NO
2 0.02 ns 16 0.05 ns 16 0.17 ns 16 0.11 ns 16
PO3

4 0.72 ** 16 0.79 ** 16 0.72 ** 16 −0.33 * 16
Chlorophyll a concentrations 0.86 ** 16 0.95 ** 16 0.66 ** 16 −0.60 ** 16
a
%B is percentages of mcyB-containing Microcystis cells in total Microcystis cells
ns not significant
*P<0.05
**P<0.01
Environ Sci Pollut Res (2010) 17:1581–1590 1587

Fig. 6 Variations of Microcystis

in the water column with
depth for Lake Xuanwu from
January to June of 2006. The
error bars indicate standard
deviations (n=4)

bacteria and cyanobacteria in the lake (Wang et al. 2007; Microcystis community during the water bloom season,
Zheng et al. 2008). However, no studies have been much higher than those in waters of European and
performed to examine seasonal dynamics of cyanobacteria American countries. Xu et al. (2008) concluded that M.
and Microcystis using a molecular approach. wesenbergii from Chinese waters represent a non-MC
During the study period, two peaks of Microcystis producer, which explains that the lower mcyB-containing
abundance were found, one in August 2005 and one in Microcystis ratios detected in Lake Xuanwu were caused
August 2006, while the lowest number occurred in by the relative abundance of M. wesenbergii.
February 2006. This reflected the water temperature as the Cell numbers of Microcystis detected by the qPCR
primary factor affecting Microcystis abundance in the lake. analysis were found to be 1–30 times higher than those by
The results from both microscopic examination and qPCR microscopic examination, which is consistent with results
indicated that the cyanobacterial community was mainly of several previous studies (Vaitomaa et al. 2003; Hotto et
composed of Microcystis (over 95%) during the blooms, al. 2007; Yoshida et al. 2008). Lower values of cell
followed by Anabaena flos-aquae and Planktothrix agard- abundance examined by the microscopic counting method,
hii. Larger numbers of Anabaena and Planktothrix species especially for Microcystis and cyanobacteria that occur in
in the cyanobacterial community after the Microcystis- late winter and early spring having low numbers, may result
dominated water blooms led to a pattern as found in this from the low detection limit, colonial decomposition, and
study, where higher total cyanobacterial cells but lower cell loss during the longer fixation procedure by the
Microcystis cells were found in November of both 2005 and microscopic counting approach (Hotto et al. 2007).
2006, indicating that Lake Xuanwu exhibited a significant
population succession in the cyanobacterial community. 4.2 Relationship between Microcystis and environmental
It is generally known that cyanobacterial blooms, dominated factors
by either Microcystis or Planktothrix, consist of both toxic and
nontoxic populations (Kurmayer et al. 2002, 2004; Yoshida et The present study revealed no significant correlation between
al. 2008). Several recent studies using qPCR or competitive water temperature and relative abundance of the mcyB
PCR to amplify fragments of the mcy gene involved in MC genotypes in Lake Xuanwu. This finding was consistent
production have revealed that proportions of mcy-containing with the field studies by Kurmayer and Kutzenberger (2003)
Microcystis were relatively low, i.e., 1–38% for mcyB and Yoshida et al. (2007), in which the relative abundance of
subpopulations in Lake Wannsee, Germany, 0.5–35% for mcyB genotypes did not depend on temperature. In Lake
mcyA subpopulations in Lake Mikata, Japan, 0–37% for Xuanwu, pH showed a significant positive correlation with
mcyD subpopulations in Lake Oneida, USA, and 0–48% abundance of Microcystis, cyanobacteria, and mcyB geno-
for mcyD subpopulations in Lake Erie, respectively type. This is consistent with previous reports in which
(Kurmayer and Kutzenberger 2003; Yoshida et al. 2007; cyanobacteria could outcompete other phytoplankton species
Hotto et al. 2008; Rinta-Kanto et al. 2009). However, in our under high pH conditions using their carbon concentration
present study, the proportions of mcyB genotypic subpopu- mechanisms (Wick and Thiel 1990; Kotak et al. 2000;
lations in Lake Xuanwu were shown to be even lower than Rantala et al. 2006).
those of the above studies, i.e., 0.1–12.8%. M. wesenbergii It is striking that the abundance of Microcystis and mcyB
was found to account for an average of 56% of the total genotypes were significantly positively correlated with TP
1588 Environ Sci Pollut Res (2010) 17:1581–1590

and PO4 and negatively correlated with TN:TP, suggesting tion. The present study showed that Microcystis spp. were
that P loading may be a factor in governing the growth of present in all of the samples and were able to be detected
both MC-producing genotypes and Microcystis population. even at very low cell abundances during low temperature
This finding agrees with Rinta-Kanto et al. (2009) who periods using the qPCR method, indicating that Microcystis
obtained similar results during a 3-year survey in Lake Erie. can overwinter in the water column even during low
Davis et al. (2009) also found P loading yielded signifi- temperatures of 4–5°C for lakes of the middle and lower
cantly high growth rates for both toxic and nontoxic reaches of the Yangtze River.
populations of Microcystis. Rapala and Sivonen (1998) Chen et al. (2003) and Ke et al. (2008) found that
found that hepatotoxic water blooms dominated by Micro- temperature played an important role in changes within the
cystis spp. were associated with high PO4–P concentration, phytoplankton community during the spring–summer succes-
whereas nontoxic blooms occurred over a wide range of sion period in Lake Taihu. Based on their laboratory and field
nutrient concentrations. However, the importance of nutri- studies, Robarts and Zohary (1987) showed that the temper-
ent factors may vary with different environments. Yoshida ature optima for the growth rate of Microcystis was 25°C or
et al. (2007) reported high NO
3
N loading as a significant greater, and growth of Microcystis will be severely limited by
factor enhancing the growth of MC-producing genotypes temperatures below 15°C. In our present study, Microcystis
within the Microcystis community. abundance was found to drop to its lowest level along with a
Interestingly, the percentage of mcyB within the Micro- decrease in water temperature below 15°C in late November,
cystis population was positively correlated with TN and and to increase sharply in June when the water temperature
NO
3
N and negatively correlated with PO4–P. This rose to 27°C. This result is similar to the finding by Zhang et
implies that higher N and lower P would favor the growth al. (2005) that the cyanobacterial biomass sharply increased in
of non-MC-producing strains instead of MC-producing Lake Taihu during the month of May.
ones in interspecies competition within the growth of a It is generally acknowledged that overwintering Microcystis
Microcystis population under complete nutritional mainte- populations might become inocula in the water column
nance. Vézie et al. (2002) demonstrated that higher levels during the spring and promote the development of dense
of N and P may favor MC-producing strains outcompeting surface blooms of Microcystis during summer. Visser et al.
non-MC-producing strains. While Davis et al. (2009) (1995) demonstrated that Microcystis sank during the
found, based on experiments under different concentrations autumn, and the colony number was lowest during spring.
of N and P, that P loading yielded significantly higher This was due to formation of a cellular carbohydrate ballast at
growth rates for both toxic and nontoxic populations of reduced temperatures. We observed that Microcystis abun-
Microcystis. These findings reflect the different response dance at the bottom of the water column was 20 times as
strategies that Microcystis populations have under different much as that at the surface in January 2006, and then
environmental conditions. Microcystis cells in the whole water column sharply
decreased during February, March, and April 2006. During
4.3 Microcystis populations in the water column May 2006, Microcystis abundance at the bottom of water
during the low temperature season column increased greatly and was much higher than that of
the surface water (Fig. 6). This made it apparent that
Overwintering of Microcystis cells (colonies) has been Microcystis colonies (cells) were distributed differently at
regarded to play a key role in formation of water blooms depths in the water column, and this difference has attributed
the following spring. However, previous studies performed to the higher detecting capacity of the qPCR method. Such a
in European deep or shallow lakes led to the argument of difference has not been documented for shallow lakes so far.
overwintering locations, whether in sediments or in the Despite the fact that Microcystis cells were detected in the
water column (Boström et al. 1989; Brunberg and Blomqvist water column at depths for only 6 months in this study, the
2002). In subtropical shallow waters of China, detecting tendency, for higher Microcystis numbers at the bottom of
overwintering Microcystis using the traditional microscopic the water column before and after water blooms in Lake
counting and pigment measurement methods has been Xuanwu, was evident. Therefore, it is suggested that Micro-
performed in different scales. Li et al. (2004) found that cystis in the water column, especially at the bottom of the
Microcystis cells would not be detected in the water column water column, play an important role in forming next seasons
when the water temperature went below 10°C in a fish pond water bloom This conclusion supports those by Verspagen et
located in Wuhan, China. Song et al. (2007) also found that al. (2004) and Cao et al. (2008). More importantly, these
Microcystis cells can widely distribute in the water column results suggest that lake management should pay more
of Lake Taihu, except in February 2005 and January 2006, attention to Microcystis abundances at the bottom of the
during their 2004–2006 survey using microscopic examina- water column in order to effectively control water blooms.
Environ Sci Pollut Res (2010) 17:1581–1590 1589

5 Conclusions tion in phytoplankton of eutrophic lakes. Can J Fish Aquat Sci

57:1584–1593
Kuiper-Goodman T, Falconner I, Fitzgerald J (1999) Human health
Water temperature was the primary factor affecting Micro- aspects. In: Chorus I, Bartram J (eds) Toxic cyanobacteria in
cystis abundance in the lake, and P loading was shown to water: a guide to their public health consequences, monitoring
be a main factor in governing the growth of both MC- and management. E and FN Spon, London, pp 113–153
Kurmayer R, Kutzenberger T (2003) Application of real-time PCR for
producing genotype and total Microcystis population.
quantification of microcystin genotypes in a population of the toxic
Higher Microcystis cell counts at the bottom of the water cyanobacterium Microcystis sp. Appl Environ Microb 69:6723–
column before and after appearance of water blooms in 6730
Lake Xuanwu suggested that Microcystis numbers in the Kurmayer R, Dittmann E, Fastner J, Chorus I (2002) Diversity of
microcystin genes within a population of the toxic cyanobacte-
water column, especially at the bottom of the water column, rium Microcystis spp. in Lake Wannsee (Berlin, Germany).
play an important role in forming the next water bloom. Microb Ecol 43:107–118
Kurmayer R, Christiansen G, Fastner J, Börner T (2004) Abundance
of active and inactive microcystin genotypes in populations of
the toxic cyanobacterium Planktothrix spp. Environ Microbiol
References
6:831–841
Lahti K, Rapala J, Färdig M, Niemelä M, Sivonen K (1997) Persistence
Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, of cyanobacterial hepatotoxin, microcystin-LR in particulate mate-
Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new rial and dissolved in lake water. Water Res 31:1005–1012
generation of protein database search programs. Nucleic Acids Li K, Song L, Wang N (2004) Studies on recruitment and growth
Res 25:3389–3402 characteristic of Microcystis in sediment. Acta Hydrobiologica
Boström B, Pettersson AK, Ahlgren I (1989) Seasonal dynamics of a Sin 28:113–118 (in Chinese)
cyanobacteria-dominated microbial community in surface sedi- Neilan BA, Jacobs D, Del Dot T, Blackall LL, Hawkins PR, Cox PT,
ments of a shallow eutrophic lake. Aquat Sci 51:153–178 Goodman AE (1997) rRNA sequences and evolutionary relation-
Brunberg AK, Blomqvist P (2002) Benthic overwintering of Micro- ships among toxic and nontoxic cyanobacteria of the genus
cystis colonies under different environmental conditions. J Microcystis. Int J Syst Bacteriol 47:693–697
Plankton Res 24:1247–1252 Nübel U, Garcia-Pichel F, Muyzer G (1997) PCR primers to amplify
Cao HS, Tao Y, Kong FX, Yang Z (2008) Relationship between 16S rRNA genes from cyanobacteria. Appl Environ Microbiol
temperature and cyanobacterial recruitment from sediments in 63:3327–3332
laboratory and field studies. J Freshwater Ecol 23:405–412 Oh H, Lee SJ, Jang M, Yoon B (2000) Microcystin production by
Carmichael WW, Azevedo MFO, An JS, Molica RJR, Jochimsen EM, Microcystis aeruginosa in a phosphorus limited chemostat. Appl
Lau S, Rinehart KL, Shaw GR, Eagelsham GK (2001) Human Environ Microbiol 66:176–179
fatalities from cyanobacteria: chemical and biological evidence Paerl HW, Fulton RS, Moisander PH, Dyble J (2001) Harmful
for cyanotoxins. Environ Health Persp 109:663–668 freshwater algal blooms, with an emphasis on cyanobacterial.
Chen YW, Qin BQ, Teubner K, Dokulil MT (2003) Long-term dynamics Scientific World Journal 1:76–113
of phytoplankton assemblages: Microcystis-domination in Lake Philipp R, Rowland MGM, Baxter PJ, McKenzie C, Bell RH (1991)
Taihu, a large shallow lake in China. J Plankton Res 25:445–453 Health risks from exposure to algae. CDR (London: England
Chorus I, Bartram J (1999) Toxic cyanobacteria in water. A guide to Rev) 1:R67–R68
their public health consequences, monitoring and management. E Preston T, Stewart WDP, Reynolds CS (1980) Bloom-forming
and FN Spon, London cyanobacterium Microcystis aeruginosa overwinters on sediment
Davis TW, Berry DL, Boyer GL, Gobler CJ (2009) The effects of surface. Nature 288:365–367
temperature and nutrients on the growth and dynamics of toxic Rantala A, Rajaniemi-Wacklin P, Lyra C, Lepisto L, Rintala J,
and non-toxic strain of Microcystis during cyanobacteria blooms. Mankiewiez-Boczek J, Sivonen K (2006) Detection of
Harmful Algae 8:715–725 microcystin-producing cyanobacteria in Finnish lakes with
Downing TG, Meyer C, Gehringer MM, Venter MVD (2005) genus-specific microcystin synthetase gene E (mcyE) PCR and
Microcystin content of Microcystis aeruginosa is modulated by associations with environmental factors. Appl Environ Microbiol
nitrogen uptake rate relative to specific growth rate or carbon 72:6101–6110
fixation rate. Environ Toxicol 20:257–262 Rapala J, Sivonen K (1998) Assessment of environmental conditions
Fan C, Yang X, Shi L, Xu D, Zhang Q, Wu C (2005) Characteristics that favor hepatotoxic and neurotoxie Anabaena spp. strains
and cause of lake eutrophication in Jiangsu province with cultured under light limitation at different temperature. Microb
suggestions on its control measures. Resour Environ Yangtze Ecol 36:181–192
Basin 14:218–223, in Chinese Reynolds CS, Jaworski GHM, Cmiech HA, Leedale GF (1981) On the
Hotto AM, Satchwell MF, Boyer GL (2007) Molecular characterization of annual cycle of the blue-green algae Microcystis aeruginosa
potential microcystin-producing cyanobacteria in lake Ontario embay- Kütz. Emend. Elenkin. Philos T Roy Soc B 293:419–477
ments and nearshore waters. Appl Environ Microb 73:4570–4578 Rinta-Kanto JM, Ouellette AJA, Boyer GL, Twiss MR, Bridgeman
Hotto AM, Satchwell MF, Berry DL, Gobler CJ, Boyer GL (2008) Spatial TB, Wilhelm SW (2005) Quantification of toxic Microcystis spp.
and temporal diversity of microcystins and microcystin-producing during the 2003 and 2004 blooms in western Lake Erie using
genotypes in Oneida Lake, NY. Harmful Algae 7:671–681 quantitative real-time PCR. Environ Sci Technol 39:4198–4205
Ke Z, Xie P, Guo L (2008) Controlling factors of spring–summer Rinta-Kanto JM, Konopko EA, Debruyn JM, Bourbonniere RA, Boyer
phytoplankton succession in Lake Taihu (Meiliang Bay, China). GL, Wilhelm SW (2009) Lake Erie Microcystis: relationship
Hydrobiologia 607:41–49 between microcystin production, dynamics of genotypes and
Kotak BG, Lam AKY, Prepas EE, Hrudey SE (2000) Role of chemical environmental parameters in a large lake. Harmful Algae 8:665–
and physical variables in regulating microcystin-LR concentra- 673
1590 Environ Sci Pollut Res (2010) 17:1581–1590

Robarts RD, Zohary T (1987) Temperature effects on photosynthetic Wicks RJ, Thiel PG (1990) Environmental factors affecting the
capacity, respiration, and growth rates of bloom-forming cyano- production of peptide toxins in floating scums of the cyanobac-
bacteria. New Zeal J Mar Fresh 21:391–399 terium Microcystis aeruginosa in a hypertrophic African reser-
Song L, Chen W, Peng L, Wan N, Gan N, Zhang X (2007) voir. Environ Sci Technol 24:1413–1418
Distribution and bioaccumulation of microcystins in water Xu Y, Wu Z, Yu B, Peng X, Yu G, Wei Z, Wang G, Li R (2008)
columns: a systematic investigation into the environmental fate Non-microcystin producing Microcystis wesenbergii (Komárek)
and the risks associated with microcystins in Meiliang Bay, Lake Komárek (Cyanobacteria) representing a main waterbloom
Taihu. Water Res 41:2853–2864 forming species in Chinese waters. Environ Pollut 156:162–
Utkilen H, Gjolme N (1995) Iron-stimulated toxin production in 167
Microcystis aeruginosa. Appl Environ Microbiol 61:797–800 Yoshida M, Yoshida T, Takashima Y, Hosoda N, Hiroishi S (2007)
Vaitomaa J, Rantala A, Halinen K, Rouhiainen L, Tallberg P, Mokelke Dynamics of microcystin-producing and non-microcystin-
L, Sivonen K (2003) Quantitative real-time PCR for determina- producing Microcystis populations is correlated with nitrate
tion of microcystin synthetase E copy numbers for Microcystis concentration in a Japanese lake. FEMS Microbiol Lett 266:49–
and Anabaena in lakes. Appl Environ Microbiol 69:7289–7297 53
Verspagen JMH, Snelder EOFM, Visser PM, Huisman J, Mur LR, Yoshida M, Yoshida T, Kashima A, Takashima Y, Hosoda N,
Ibelings BW (2004) Recruitment of benthic Microcystis (Cyano- Nagasaki K, Hiroishi S (2008) Ecological dynamics of the toxic
phyceae) to the water column: internal buoyancy changes or bloom-forming cyanobacterium Microcystis aeruginosa and its
resuspension? J Phycol 40:260–270 cyanophages in Freshwater. Appl Environ Microbiol 74:3269–
Vézie C, Rapala J, Vaitomaa J, Seitsonen J, Sivonen K (2002) Effect 3273
of nitrogen and phosphorus on growth of toxic and nontoxic Zhang ZS, Huang XF (1991) A manual on methods for plankton
Microcystis strains and on intracellular microcystin concentra- research in freshwater. Science Press, Beijing, pp 333–356
tions. Microb Ecol 43:443–454 Zhang X, Kong F, Cao H, Tan J, Tao Y, Wang M (2005)
Visser PM, Ibelings BW, Mur LR (1995) Autumnal sedimentation of Recruitment dynamics of bloom-forming cyanobacteria in
Microcystis spp. as result of an increase in carbohydrate ballast at Meiliang Bay of Taihu Lake. Chinese J Appl Ecol 16:1346–
reduced temperature. J Plankton Res 17:919–933 1350 (in Chinese)
Wang Y, Li J, Wu M, Wang Y, Weng Y (2007) Composition of Zheng X, Xiao L, Ren J, Yang L (2008) Variation of bacterial
Microcystis species of the cyanobacterial bloom in xuanwu lake community in the outbreak and decline of Microcystis spp. bloom
of Nanjing. Environ Sci 28:2187–2191 (in Chinese) in Lake Xuanwu. Environ Sci 29:2956–2962 (in Chinese)