Ecotoxicology DOI 10.1007/s10646-015-1476-y

Correlations between cyanobacterial density and bacterial transformation to the viable but nonculturable (VBNC) state in four freshwater water bodies Huirong Chen1 • Ju Shen1 • Gaoshan Pan1 • Jing Liu1 • Jiancheng Li1 Zhangli Hu1

•

Accepted: 28 April 2015  Springer Science+Business Media New York 2015

Abstract Nutrient concentrations, phytoplankton density and community composition, and the viable but nonculturable (VBNC) state of heterotrophic bacteria were investigated in three connected reservoirs and a small isolated lake in South China to study the relationship between biotic and abiotic factors and the VBNC state in bacteria. Nutrient concentrations in the reservoirs increased in the direction of water flow, whereas Wenshan Lake was more eutrophic. Cyanobacterial blooms occurred in all four water bodies, with differing seasonal trends and dominant species. In Xili and Tiegang Reservoirs, the VBNC ratio (percent of VBNC state bacteria over total viable bacteria) was high for most of the year and negatively correlated with cyanobacterial density. Laboratory co-culture experiments were performed with four heterotrophic bacterial species isolated from Wenshan Lake (Escherichia coli, Klebsiella peneumoniae, Bacillus megaterium and Bacillus cereus) and the dominant cyanobacterial species (Microcystis aeruginosa). For the first three bacterial species, the presence of M. aeruginosa induced the VBNC state and the VBNC ratio was positively correlated with M. aeruginosa density. However, B. cereus inhibited M. aeruginosa growth. These results demonstrate that cyanobacteria could potentially regulate the transformation to the VBNC state of waterborne bacteria, and suggest a role for bacteria in cyanobacterial bloom initiation and termination.

Huirong Chen and Ju Shen have contributed equally. & Zhangli Hu [email protected] 1

Shenzhen Key Laboratory of Marine Bioresource and Ecoenvironmental Science, College of Life Science, Shenzhen University, Shenzhen 518060, China

Keywords Cyanobacteria  Culturability  VBNC  Blooms  Bacterial–phytoplankton coupling

Introduction As the eutrophication process accelerates in natural freshwater ecosystems worldwide, blooms occur more frequently, which results in reduced aesthetics, everworsening water quality problems, cyanobacterial toxin pollution, and potential harm to aquatic organisms and human health (Chorus and Bartram 1999; Zurawell et al. 2005). The initiation and termination mechanisms of cyanobacterial blooms are still poorly understood. Abiotic environmental factors have been associated with eutrophication and cyanobacterial overgrowth, but the effect of biotic factors on cyanobacterial blooms has drawn less attention. Some heterotrophic bacteria can degrade cyanobacterial toxins (Jones et al. 1994), or lyse cyanobacterial cells (Shilo 1970; Manage et al. 2001). Thus, interactions between cyanobacteria and heterotrophic bacteria, especially lytic bacteria, could be a key factor influencing cyanobacterial bloom initiation and termination (Rashidan and Bird 2001; Berg et al. 2009). However, cyanobacteria–bacteria interactions have not been fully elucidated, partly because of the complexity of bacterioplankton ecology and the unculturability of most aquatic bacteria. The term viable but nonculturable (VBNC) refers to the stress-induced state of many non-spore forming gramnegative bacteria. Bacteria in a VBNC state were first reported by Xu et al. in 1982. The VBNC state appears to be a general survival strategy in different bacterial species, especially in oligotrophic water bodies (Lemke and Leff 2006; Servais et al. 2009).

123

123

32.2 (12.5–186) 218.1 (120.1–1002.3) 0.425 (0.35–0.51) 4.25 (3.2–6.13) 0.30 (0.200–0.490) 8.8 (8.3–9.7) 122.5 (107.0–168.0) 8.7 (8.2–10.5) 1.50 Wenshan Lake

0.113

3.20 (1.12–7.51) 114.9 (45.6–25.8) 1.0 (0.50–1.67) 3.2 (2.96–1.79) 0.046 (0.031–0.05) 8.5 (7.5–9.0) 3.0 (2.0–3.9) 7.6 (7.1–9.0) 36.0 Shiyan Reservoir

2.98

3.03 (0.52–9.68) 45.6 (23.3–120.9) 0.136 (0.05–0.36) 1.113 (0.83–1.52) 0.025 (0.020–0.037) 8.6 (6.7–9.0) 2.7 (1.6–3.0) 8.4 (7.6–9.5) 10.0 Tiegang Reservoir

5.20

1.21 (0.13–4.88) 25.0 (4.6–78.1) 0.023 (0.01–0.12) 0.925 (0.615–1.515) 0.038 (0.026–0.048) 7.0 (5.5–9.8) 1.8 (1.2–2.4)

Total phytoplankton (9107 L-1) Chl–a (lg L-1) NH4?–N (mg L-1) TN (mg L-1) TP (mg L-1) DO (mg L-1) COD (mg L-1) pH

7.7 (7.2–8.9) 2.78 31.7

The interconnected Xili, Tiegang and Shiyan Reservoirs are located in Shenzhen, China, with mean depths ranging from 10 to 36 m (Table 1). Water flows from the East River to Xili Reservoir, then to the Tiegang and Shiyan Reservoirs. The outflow from Shiyan serves as the main drinking water supply to the Nanshan District, Shenzhen. The fourth study water body is Wenshan Lake, a highly eutrophic, well-mixed water body on the campus of Shenzhen University, with a mean depth of 1.5 m (Table 1).

Xili Reservoir

Study water bodies and sampling methods

Mean area (km2)

Materials and methods

Mean depth (m)

Many pathogenic bacteria in the VBNC state cannot be detected by routine water quality testing protocols, which rely on culturing techniques, thus they could pose a threat to public health. Currently, the majority of studies of VBNC bacteria have focused on several human pathogens and their activity under specific environmental conditions or with and without a host (Oliver 2010; Pinto et al. 2015). Few studies have considered other members of the ecosystem, such as phytoplankton. Even fewer have paid attention to the VBNC state of bacteria from an ecological viewpoint. Surface biofilms on algal cells could provide a long-term reservoir for VBNC pathogens, and play a role in their distribution (Islam et al. 2004). Despite abundant studies on both cyanobacterial blooms and the bacterial VBNC state, the possible relationship between the two areas has been overlooked. We hypothesize that the occurrence of cyanobacterial blooms could affect the VBNC state of aquatic bacteria, especially algal lytic bacteria, and consequently induce the termination of the bloom. Preliminary studies by the authors on the VBNC state of total bacteria and Aeromonas sobria cultures showed that both pure microcystin (the most ubiquitous cyanobacterial toxin) and a crude extract of cyanobacterial cells induced VBNC bacteria to transform into a culturable state (Han et al. 2005; Pan et al. 2008). It is possible that transformation to the VBNC state could also be induced by cyanobacteria and/or their toxins, and thereby play a role in the initiation and termination of cyanobacterial blooms. In the current study, we investigated four water bodies in Shenzhen, China and measured nutrient concentrations, phytoplankton density and composition, and the culturability of heterotrophic bacteria. In the laboratory, we co-cultured the main bacterial and cyanobacterial species isolated from the field to confirm the VBNC transformation of bacteria and how it was related to cyanobacterial density.

Table 1 Median values of physical–chemical parameters and phytoplankton abundance of four water bodies from 12 monthly samples collected August 2004 to July 2005 (numbers in brackets are minimum and maximum values)

H. Chen et al.

Correlations between cyanobacterial density and bacterial transformation to the viable but…

During 12 monthly sampling visits from August 2004 to July 2005, The pH values and dissolved oxygen (DO) concentrations were measured in the middle of each water body at a depth of 50 cm. Water samples were also collected from 50 cm below the surface and stored in a cooler in 1 L bottles. For each sampling date and water body, two samples were collected: one for water quality parameter analysis and one for phytoplankton identification. The methods used to determine chemical oxygen demand (CODMn), total phosphorus (TP), total nitrogen (TN), ammonium nitrogen (NH4?–N), and chlorophyll a (Chl-a) concentrations can be found in the Chinese Environmental Quality Standard of Surface Water (2002). For phytoplankton enumeration, 15 mL of Lugol’s solution was added to each 1 L water sample. After 48 h, the supernatant was carefully removed to concentrate the sample down to 50 mL. Phytoplankton cells were counted using a hemocytometer under a compound microscope (Olympus-50X, Japan) and identified with the phytoplankton key of Hu et al. (1980). For each sample, five fields in the hemocytometer were counted and the mean value was used to calculate the density, expressed as cell number per liter of water. Phytoplankton, total bacterial and VBNC state bacterial density To closely investigate interactions between phytoplankton (especially cyanobacteria) and total and culturable bacterial density, we carried out a second sampling program in the three reservoirs from March 2005 to February 2006. Water quality parameters, total phytoplankton and cyanobacterial density were determined as above. Samples for bacterial analysis were collected in sterilized sample bottles and transported without cooling. The total number of viable bacteria was measured with direct counting after staining with 40 ,6-diamidino-2-phenylindole (DAPI, Sigma) as described by Porter and Feig (1980). When DAPI binds with nucleic acids in viable bacteria, it emits a blue color under fluorescent light. For each slide, ten visual fields were randomly chosen, and the mean of the middle five counts was used to calculate the total number of viable bacteria per liter of water. The number of culturable bacteria was determined with the direct plate counting method. One part of each water sample was diluted with nine parts bacteria-free water; 100 lL of diluted sample was spread evenly onto Standard Plate Count Agar (APHA, Oxoid Ltd., England), and incubated at 37 C for 24–48 h. Bacterial colonies on each plate were counted and reported as colony-forming units (CFUs) per liter of original water sample. Individual colonies were streaked onto Luria–Bertani (LB) plates (APHA, Oxoid Ltd., England) to grow.

The densities of VBNC bacteria and the VBNC ratio were calculated as follows:
VBNC cell density CFU L1
¼ total viable cell density CFU L1
culturable cell density CFU L1 VBNC ratio ð%Þ ¼ ðVBNC cell density=total viable cell densityÞ  100

Cyanobacteria and bacteria co-culture experiments Based on the results from our preliminary water quality investigation, Wenshan Lake has the highest nutrient concentrations and is dominated by cyanobacteria yearround. Water from Wenshan Lake was therefore used to isolate aquatic bacteria to co-culture with cyanobacteria. Surface water samples were collected during cyanobacterial blooms, inoculated onto Standard Plate Count Agar, and incubated at 37 C for 24 h. Single colonies were streaked to LB plates and identified by the Microbial Detection and Analysis Center of Guangdong Province. Four main bacterial species were identified from the water samples: Escherichia coli, Klebsiella peneumoniae, Bacillus megaterium, and Bacillus cereus. Given that Microcystis aeruginosa was the main cyanobacterial species in Wenshan Lake, strain M. aeruginosa 905 from the Institute of Hydrobiology, Chinese Academy of Sciences, was used for the co-culture experiment. Axenic M aeruginosa 905 cultures were grown in BG11 medium (ATCC 616) at 25 C under continuous light (20 lmol m-2 s-1) until they reached the log phase (approximately 106 cells mL-1). Bacterial cultures were grown overnight, centrifuged to remove the LB medium, and resuspended in BG11 medium to a density of 107 CFU mL-1. The bacterial solution was then inoculated into the cyanobacterial culture above at a ratio of 1:100 (0.3 mL bacterial culture to 30 mL cyanobacterial culture) and mixed well. The co-cultures were grown in a 25 C incubator with 20 lmol m-2 s-1 continuous light. The control group comprised 0.3 mL bacterial culture added to 30 mL BG11 medium without cyanobacteria. Cultures were collected daily to monitor the cell density (cell number per mL) of M. aeruginosa using a hemocytometer under a microscope, and the density of total and culturable bacterial cells using the methods mentioned above. Statistical analyses were conducted with SPSS 13.0 for Windows. For field data, a single sample was collected for each site. For laboratory co-culture experiments, means and standard deviations of triplicates are presented. A oneway ANOVA was used to compare parameters among reservoirs (alpha 0.05).

123

H. Chen et al.

Results Nutrient levels and phytoplankton composition All the water bodies were alkaline (pH [ 7.1), and surface waters were well-oxygenated (DO concentrations 5.5–9.8 mg L-1; Table 1). With the exception of TP, nutrient concentrations increased along the direction of flow from Xili to Shiyan Reservoir. The COD, DO, TP, TN, NH4?–N, and Chl-a concentrations and total phytoplankton density were higher in Shiyan Reservior than Xili and Tiegang Reservoirs (P \ 0.05). However, compared with Wenshan Lake, the nutrient concentrations in Shiyan Reservoir were much lower (Table 1). Of note, the COD, TP concentrations and phytoplankton density were ten times higher in Wenshan Lake than in the three reservoirs. In the first survey, the phytoplankton density of the four water bodies was between 1.50 9 106 and 1.86 9 109 cells L-1, with cyanobacterial cells representing an average of 50, 63, 65 and 70 % of total phytoplankton in Xili, Tiegang, Shiyan and Wenshan Lake, respectively (Table 1; Fig. 1). A total of 113 strains were identified, belonging to 42 genera from six phyla (Table 2). Dominant cyanobacterial species were Oscillatoria spp. in the three reservoirs, and M. aeruginosa and Merismopedia sp. in Wenshan Lake (Table 3). A seasonal trend was evident in the three reservoirs, whereby cyanobacteria dominated (over 80 %) during summer (June to August), while green algae and diatoms were high in density during winter (December to February). However, Wenshan Lake was dominated by cyanobacteria year-round, even in the winter (Fig. 1). For Fig. 1 Phytoplankton composition (% of cell density) in the four water bodies from August 2004 to July 2005

123

all four water bodies, no correlations were found between phytoplankton density and nutrient concentrations, but cyanobacterial density was significantly correlated with TN concentrations (r = 0.777, P \ 0.001). Relationship between cyanobacterial and bacterial dynamics In the second survey, Xili Reservoir had the highest bacterial density among the reservoirs. Bacterial density peaked in May and January and was higher than the phytoplankton density in all months except in September and October, when the maximum phytoplankton density was observed (Fig. 2). In Tiegang and Shiyan Reservoirs, bacterial density was more variable, with higher phytoplankton density than in Xili reservoir (Fig. 2). The mean cyanobacterial percentages in Xili and Tiegang Reservoirs were 52, 71, 66 %, respectively, similar to the first survey. The peak cyanobacterial percentage occurred in September and October in all three reservoirs (Fig. 2). In Xili and Tiegang Reservoirs, the VBNC ratio remained at or near 100 %, except in September when it fell to approximately 65 % (Fig. 2). This indicated that some VBNC bacteria resuscitated during cyanobacterial blooms (Fig. 2). There was a negative relationship between VBNC ratio and cyanobacterial density in both Xili (r = –0.92, P \ 0.001) and Tiegang Reservoirs (r = –0.67, P = 0.018). Shiyan Reservoir showed more fluctuation in VBNC ratio (Fig. 2) and the VBNC ratio was not correlated with cyanobacterial or phytoplankton density (P = 0.108). The lowest VBNC ratio (39.6 %) in Shiyan Reservoir occurred in October, just

Correlations between cyanobacterial density and bacterial transformation to the viable but… Table 2 Phytoplankton genera identified from the four water bodies sampled from August 2004 to July 2005 (numbers in brackets indicate number of species identified for each genus) Chlorophyta

Bacillariophyta

Cyanophyta

Pandorina sp.(1)

Microcystis spp. (2)

Melosira sp. (1)

Acanthosphaera sp. (1)

Aphanocapsa spp. (4)

Stephanodiscus spp. (2)

Characium spp. (3)

Chroococccus spp. (3)

Synedra spp. (2)

Golenkinia radiate Chodat

Merismopedia spp. (4)

Schroederia spp. (3)

Dactylococcopsis spp. (2)

Chlorella sp. (1)

Raphidiopsis spp. (2)

Chodatella spp. (3)

Anabaena spp. (2)

Tetrae¨dron spp. (9)

Spirulina spp. (2)

Gymnodinium spp. (2)

Kirchnecriella spp. (2)

Oscillatoria spp. (5)

Ceratium hirundinella (Mu¨ll.) Schr.

Selenastrum spp. (3) Treubaria spp. (2)

Lyngbya spp. (3) Phormidium spp. (3)

Nitzschia spp. (3) Cryptophyta Cryptomonas sp. (1) Pyrrophyta

Euglenophyta Euglena sp. (1) Trachelomonas spp. (3)

Echinosphaerilla limnetica G. M. Smith. Dictyosphaerium sp. (1) Actinastrum sp. (1) Pediastrum spp. (7) Scenedesmus spp. (12) Tetrastrum spp. (2) Crucigenia quadrata Morrer Coelastrum spp. (2) Staurastrum spp. (6) Cosmarium spp. (2) Westella sp. (1)

one month later than in the Xili and Tiegang Reservoirs (Fig. 2). Effect of M. aeruginosa on VBNC ratio of bacterial isolates from Wenshan Lake The growth of axenic cultures of M. aeruginosa was inhibited when co-cultured with B. cereus isolated from Wenshan Lake (Fig. 3). By comparison, the cell density of M. aeruginosa co-cultured with the other three bacterial cultures increased over time, which indicated that these bacteria had little or no inhibitory effect on M. aeruginosa growth (Fig. 3). After 5 days of incubation, the culturable proportion of the total density of E. coli was low (\ 10 %) in both control and co-culture groups, indicating that more than 90 % of E. coli lost their culturability (Fig. 4a). A positive correlation was detected between M. aeruginosa cell density and the E. coli VBNC ratio (r = 0.813, P = 0.004). The growth of M. aeruginosa inhibited the culturability of E. coli, and encouraged them to transform into the VBNC state. This inhibitory effect became evident after 3 days of incubation. The strong inhibitory effect of M. aeruginosa on E. coli culturability was not observed in K. peneumoniae (Fig. 4b), suggesting that other factors in the culture

medium supported the transformation of VBNC cells into viable cells, but the presence of M. aeruginosa in the coculture suppressed this transformation. The M. aeruginosa cell density was positively correlated with the VBNC ratio of K. peneumoniae (r = 0.947, P = 0.05). The viable density of B. megaterium in co-culture was lower than the control group beginning at day 2, suggesting an inhibitory effect of M. aeruginosa. A strong positive correlation between M. aeruginosa density and B. megaterium VBNC ratio (r = 0.967, P = 0.03) suggests that M. aeruginosa supported the transformation of B. megaterium into the VBNC state. The viable density of B. cereus was lower in the control than the co-culture during the first 3 days, suggesting an inhibitory effect of BG-11medium on the growth of bacteria (Fig. 4d). The VBNC ratio.of B. cereus remained stable throughout the experiment and no correlation was found between M. aeruginosa cell density and the VBNC ratio of B. cereus (r = –0.125, P [ 0.05).

Discussion The four water bodies located in a subtropical area are eutrophic or hypereutrophic, with phytoplankton communities dominated by cyanobacteria. However, during the two

123

H. Chen et al. Table 3 Mean cell density (9106 cells mL-1) of the main phytoplankton species in the four studied water bodies from August 2004 to July 2005 Xili

Tiegang

Shiyan

Wenshan

Oscillatoria

5.13

21.8

16.3

7.50

Raphidiopsis Aphanocapsa

1.54 1.47

2.22 2.10

2.20 2.73

1.54 58.7

Merismopedia

0.517

0.821

1.90

1367

Microcystis

0

0

0.844

1760

Cyanophyta

Chlorophyta

Fig. 2 Total phytoplankton and bacterial density cyanobacteria and VBNC ratio in the three reservoirs from March 2005 to February 2006

Fig. 3 Growth curve of the cyanobacterium M. aeruginosa cocultured with four bacterial species

survey periods, only Wenshan Lake experienced summer cyanobacterial blooms, probably due to its shallow depth and long water retention time. By comparison, the three reservoirs are deeper, and had lower nutrient concentrations, which might have prevented the occurrence of surface blooms with even higher cyanobacterial density. The dominant cyanobacterial species were different between the

123

Scenedesmus

0.310

0.721

0.938

9.75

Coelastrum

0

0

0

522

three reservoirs and the lake. Strains of Oscillatoria Vauch were dominant in the three reservoirs. By comparison, Wenshan Lake was dominated by Microcystis sp., which was not observed in Xili or Tiegang Reservoirs. The increase in nutrient concentrations with flow direction in the three reservoirs indicated that nutrient inputs from the drainage area played an important role in eutrophication. Phytoplankton, cyanobacterial, and bacterial densities in the four water bodies varied according to the eutrophication level. Both cyanobacterial and culturable bacterial densities were positively correlated with TN concentrations, whereas total bacterial density was correlated with TP concentrations. The important role of nitrogen on microbial community composition has also been found in other freshwater and marine environments (e.g., Ling et al. 2012). The positive correlation between bacterial culturability and cyanobacterial density in Xili and Tiegang Reservoirs could be the manifestation of several mechanisms. For example, cyanobacteria could provide dissolved organic matter to support bacterial growth (Eiler and Bertilsson 2007) or provide shelter for attached active cells (Islam et al. 2004). High cyanobacterial density could also resuscitate bacterial cells in the VBNC state. However, since we only used plate count agar to identify culturable bacteria, increased culturability could either result from an increase in bacterial activity, or could be the result of more bacteria that prefer the medium been isolated. The observation that cyanobacterial density and bacterial culturability were not correlated in Shiyan Reservoir could be explained by the higher nutrient concentrations, and/or how the cyanobacterial species present affected the composition and seasonal variation of attached bacteria (Turner et al. 2009). In addition, toxins produced by some cyanobacteria could also affect the culturability of symbiotic bacteria as observed by Ashton et al. (2003) for the toxic benthic dinoflagellate Ostreopsis lenticularis. Our

Correlations between cyanobacterial density and bacterial transformation to the viable but… Fig. 4 Change of total and culturable cell density of four bacterial species co-cultured with M. aeruginosa

group also found that addition of the cyanobacterial toxin, microcystin-LR, induced an increase in culturable bacterial cell density, a decrease in the VBNC ratio in water samples (Han et al. 2005) and caused transformation to the VBNC state in bacterial isolates (Chen et al. 2012). In future studies, the seasonal variation in main bacterial species and the toxin concentrations in water samples should be monitored. The co-culture experiments showed that the effect of the cyanobacterium M. aeruginosa on bacteria was speciesspecific. Three bacterial species showed higher VBNC ratios when co-cultured with M. aeruginosa, indicating inhibition of bacterial cell division by the cyanobacterium. However, B. cereus suppressed the growth of M. aeruginosa. Similar results were reported by Mouget et al. (1995): antibiotics produced by the golden algae Prymnesium parvum and the diatom Asterionella notata combined with nutrient deficiency to produce a synergistic inhibitory effect on bacteria Staphyloccocus aureus, fecal Streptococcus faecalis, and E. coli. In our study, the presence of M. aeruginosa induced the VBNC state in E. coli, K. peneumoniae, B. cereus, and B. megaterium, though the VBNC ratio varied among the bacterial species. Few researchers have worked on the phytoplankton– bacteria interactions during bloom formation. Islam et al. (2002) found that mucinase plays a key role in the interaction between the cyanobacterium Anabaena sp. and symbiotic Vibrio cholerae O1 during VBNC transformation. Our group added crude microcystin to a culture of Aeromonas sobria, and found the extract induced culturable A. sobria cell to transform to the VBNC state (Pan et al. 2008). However, the exact mechanisms driving phytoplankton growth inhibition or enhancement by

heterotrophic bacteria are unknown. Because many algaelysing bacteria are also human pathogens in aquatic ecosystems, the change in their activity status (VBNC state) could not only relate with the cyanobacterial bloom initiation and termination, but also the spread of human disease. In conclusion, our data from four eutrophic to hypereutrophic water bodies showed a variation in cyanobacteria community and bacterial VBNC state dynamics. Cyanobacterial density and/or cyanobacterial toxins could regulate the transformation of bacteria to and from the VBNC state, which might responsible for the bloom occurrence and termination, and also cause public health problems. Acknowledgments This study was financially supported by National Natural Science Fund (Nos. 41176106, 31200092, 31470431) (http://www.nsfc.gov.cn/), Guangdong Natural Science Foundation (2014A030308017) and Shenzhen Grant Plan for Science and Technology to Zhangli Hu and Huirong Chen. The funders had no interest conflicts, and played no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors gratefully acknowledge support provided by the Shenzhen Water Authority (http://www.waterchina.com/) for sampling in Xili, Shiyan, and Tiegang Reservoirs. Conflict of interest The authors declare that they have no conflict of interest associated with this publication.

References Ashton M, Rosado W, Govind NS, Tosteson TR (2003) Culturable and nonculturable bacterial symbionts in the toxic benthic dinoflagellate Ostreopsis lenticularis. Toxicon 42:419–424. doi:10.1016/S0041-0101(03)00174-0

123

H. Chen et al. Berg KA, Lyra C, Sivonen K, Paulin L, Suomalainen S, Tuomi P, Rapala J (2009) High diversity of cultivable heterotrophic bacteria in association with cyanobacterial water blooms. ISME J 3:314–325. doi:10.1038/ismej.2008.110 Chen H, Fu L, Luo L, Lu J, White WL, Hu Z (2012) Induction and resuscitation of the viable but nonculturable state in a cyanobacteria-lysing bacterium isolated from cyanobacterial bloom. Microb Ecol 63:64–73. doi:10.1007/s00248-011-9928-2 Chorus I, Bartram J (eds) (1999) Toxic Cyanobacteria in water: a guide to their public health consequences, monitoring and management. E & FN Spon, London. doi:10.4324/9780203478073 Eiler A, Bertilsson S (2007) Flavobacteria blooms in four eutrophic lakes: linking population dynamics of freshwater bacterioplankton to resource availability. Appl Environ Microbiol 73: 3511–3518. doi:10.1128/AEM.02534-06 Han Q, Hu Z, Lei A, Song L, Liu Y (2005) Studies on the relationship between microcystin-LR and viviform state of bacterium in natural bodies of water. Acta Hydrobiologia Sinica 29:199–202 Hu HJ, Li YY, Wei YX, Zhu HZ, Chen JY, Shi ZX (1980) Freshwater algae in China. Shanghai Science and Technology Press, Shanghai Islam MS, Goldar MM, Morshed MG, Khan MNH, Islam MR, Sack RB (2002) Involvement of the hap gene (mucinase) in the survival of Vibrio cholerae O1 in association with the blue-green alga, Anabaena sp. Can J Microbiol 48:793–800. doi:10.1139/ w02-073 Islam MS, Mahmuda S, Morshed MG, Bakht HBM, Khan MNH, Sack RB, Sack DA (2004) Role of cyanobacteria in the persistence of Vibrio cholerae O139 in saline microcosms. Can J Microbiol 50:127–131. doi:10.1139/w03-114 Jones GJ, Bourne DG, Blakeley RL, Doelle H (1994) Degradation of the cyanobacterial hepatotoxin microcystin by aquatic bacteria. Nat Toxins 2:228–235. doi:10.1002/nt.2620020412 Lemke MJ, Leff LG (2006) Culturability of stream bacteria assessed at the assemblage and population levels. Microb Ecol 51:365–374. doi:10.1007/s00248-006-9026-z Ling J, Dong JD, Wang YS, Zhang YY, Deng C, Lin L, Wu ML, Sun FL (2012) Spatial variation of bacterial community structure of the Northern South China Sea in relation to water chemistry. Ecotoxicology 21:1669–1679. doi:10.1007/s10646-012-0941-0

123

Manage PM, Kawabata Z, Nakano S (2001) Dynamics of cyanophagelike particles and algicidal bacteria causing Microcystis aeruginosa mortality. Limnology 2:73–78. doi:10.1007/s102010170002 Ministry of Environmental Protection of P.R China (2002) Environmental quality standards for surface water (GB3838-2002) Mouget JL, Dakhama A, Lavoie MC, de la Nou¨e J (1995) Algal growth enhancement by bacteria: is consumption of photosynthetic oxygen involved? FEMS Microbiol Ecol 18:35–43. doi:10. 1111/j.1574-6941.1995.tb00159.x Oliver JD (2010) Recent findings on the viable but nonculturable state in pathogenic bacteria. FEMS Microb Rev 34:415–425. doi:10. 1111/j.1574-6976.2009.00200.x Pan G, Hu Z, Lei A, Li S (2008) Effect of crude microcystin on the viable but non-culturable state of Aeromonas sobria in aquatic environment. J Lake Sci 20:105–109 Pinto D, Santos MA, Chambel L (2015) Thirty years of viable but nonculturable state research: unsolved molecular mechanism. Crit Rev Microbiol 41:61–76. doi:10.3109/1040841X.2013.794127 Porter KG, Feig YS (1980) The use of DAPI for identifying and counting aquatic microflora. Limnol Oceanogr 25:943–948. doi:10.4319/lo.1980.25.5.0943 Rashidan KK, Bird DF (2001) Role of predatory bacteria in the termination of a cyanobacterial bloom. Microb Ecol 41:97–105. doi:10.1007/s002480000074 Servais P, Prats J, Passerat J, Garcia-Armisen T (2009) Abundance of culturable versus viable Escherichia coli in freshwater. Can J Microbiol 55:905–909. doi:10.1139/W09-043 Shilo M (1970) Lysis of blue–green algae by Myxobacter. J Bacteriol 104:453–461 Turner JW, Good B, Cole D, Lipp EK (2009) Plankton composition and environmental factors contribute to Vibrio seasonality. ISME J 3:1082–1092. doi:10.1038/ismej.2009.50 Xu HS, Roberts N, Singleton FL, Attwell RW, Grimes DJ, Colwell RR (1982) Survival and viability of nonculturable Escherichia coli and Vibrio cholerae in the estuarine and marine environment. Microb Ecol 8:313–323. doi:10.1007/BF02010671 Zurawell RW, Chen H, Burke JM, Prepas EE (2005) Hepatotoxic cyanobacteria: a review of the biological importance of microcystins in freshwater environments. J Toxicol Environ Health B 8:1–37. doi:10.1080/10937400590889412