This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Introduction
Noxious cyanobacterial blooms have increased in freshwaters due to eutrophication caused by high nutrient loading from wastewater discharge and large-scale aquaculture (Paerl & Huisman 2008). Cyanobacteria, particularly the Microcystis genus, are widely distributed in lakes, rivers, ponds, brackish waters and even in harsh environments such as hot springs (Zurawell et al. 2005; Belykh et al. 2013). These species can produce a potent class of hepatotoxins called microcystins that may have serious impact on the functioning and health of aquatic ecosystems, as well as on wildlife and human health (Carmichael 1996). As they co-exist in the same environment, Microcystis sp. and zooplankton interact and affect each other. Microcystis sp. and herbivorous grazers interact in complex ways (Sarnelle et al. 2005) that may be attributed to genetic and/or physiological variation within and among prey and consumer populations (Wilson, Hay 2007). Cyanobacteria have defense mechanisms that reduce or inhibit herbivory, including aggregation into large growth forms (e.g. colonies and filaments) and/or production of secondary metabolites and allelochemicals, including toxins (Watanabe et al. 1996; Żak, Kosakowska 2016). As for the defense mechanism through production of toxins, the effect of these toxins on predators is not clear (Wilson et al. 2006). Some studies have demonstrated their negative effects on zooplankton (Zurawell et al. 2005), while others have shown no anti-herbivore effect of cyanotoxins (Wilken et al. 2010). Despite the defense of cyanobacteria against zooplankton grazing, some copepods show high ingestion rates on large forms of toxic cyanobacteria (Koski et al. 2002; Kozlowsky-Suzuki et al. 2003). This may be due to ability of these zooplankton species to capture such large forms (DeMott 1990) and to develop a physiological adaptation and resistance to cyanotoxins (Fulton, Paerl 1987). Nevertheless, the ingestion rates of Microcystis sp. by zooplankton differ significantly among copepod species (Ger et al. 2010), as copepods respond differently to anti-grazing defense of cyanobacteria (Engstrom et al. 2000). Such differences in cyanobacteria ingestion rates among copepods could affect the composition of the zooplankton community, and may be useful for controlling harmful cyanobacterial blooms in lakes (Wang et al. 2010). Therefore, ingestion rates for Microcystis sp. and other toxic cyanobacteria should be addressed for many copepod species to show how the latter tolerate Microcystis sp. and their toxins more than others (Lurling 2003; Ger et al. 2010). Limited studies, however, were performed on copepods grazing on Microcystis sp. (Work & Havens 2003; Wilson et al. 2006; Ger et al. 2010). One of the most important copepods is C. vicinus, which is regularly found in the zooplankton communities of many lakes and reservoirs, and is often referred to as a predator of many zooplankton and phytoplankton species (Devetter, Seda 2006). In Egyptian fishponds, the copepod Cyclops genus is abundant in freshwaters and has been observed to frequently co-occur with Microcystis blooms (Zakaria A. Mohamed, personal communication). Therefore, the objective of the present study was to investigate the grazing effects of Cyclops on toxic M. aeruginosa as a potential bioagent in the biocontrol of harmful cyanobacterial blooms in freshwater sources.
Materials and methods
Organisms
The cyanobacterium M. aeruginosa was obtained from the microalgal culture collection of the Botany & Microbiology Department, Faculty of Science, Sohag University, Egypt. This strain was originally isolated from fish ponds in the Sohag region, Egypt, and reported as a microcystin producer (Bakr 2016). The organisms were grown in 250-ml Erlenmeyer flasks for 21 days in a climatic chamber under controlled light (50 μmol photons m-2 s-1), photoperiod (16:8 hours light-dark) and temperature (30 ± 0.5°C) in a BG-11 liquid medium. The green alga Ankistrodesmus falcatus was isolated from fish ponds and cultured under the same conditions as M. aeruginosa.
Cyclops vicinus individuals were collected from Egyptian fish ponds, where blooms of Microcystis are not known to occur. Cyclops vicinus were transferred individually using a pipette with a large tip opening to glass aquariums filled with GF/C Whatman (pore size = 1.2 μm) filtered water from the same fishpond. Before being used for culturing, the filtered fishpond water was sterilized in an autoclave at 1.05 kg cm-2 (15 psi) and 121°C. The species was verified under a binocular microscope (Olympus BH-2, Japan) at ×100 total magnification, using the identification key of freshwater crustacean zooplankton of Europe (Bledzki & Rybak 2016). Individual clones were not separated in order to study grazing at a population level. Prior to grazing experiments, Cyclops vicinus cultures were grown for one week at the same temperature and light cycle as Ankistrodesmus falcatus cultures, while feeding with A. falcatus every 2 days at an approximate concentration ranging from 5000 to 10 000 cells ml-1.
Grazing experiments
Short-term grazing experiments were conducted to determine if Cyclops are able to feed on M. aeruginosa in the presence of a known edible prey (A. falcatus). These experiments were carried out in 6 one-liter glass bottles containing 500 ml filtered fishpond water and a known number of M. aeruginosa and the chlorophyte A. falcatus. Three bottles were dosed with 100 Cyclops, while the remaining bottles were not dosed with Cyclops and used as a control. The estimated average initial concentrations were approximately 12 000 cells ml-1 for M. aeruginosa, and 10 000 cells ml-1 for A. falcatus. Both treated and control bottles were incubated for 10 days under the same conditions for growing M. aeruginosa as described above. The number of M. aeruginosa cells in treated and control bottles was monitored daily by withdrawing an aliquot of culture under septic conditions. The cells were counted using a Sedgewick Rafter counter under a light microscope according to the method used by Hötzel and Croome (1999). The grazing rate of Cyclops on M. aeruginosa was estimated using the following equation according to Devetter and Seda (2006). The grazing rate (Gr) is given as the number of prey eaten by one predator per day:
where Nc is the number of prey in the control, Ne is the number of prey in experimental enclosures and Ce is the number of C. vicinus carnivorous stages in the enclosures.
In order to determine the effect of grazing by Cyclops on MC production in M. aeruginosa, 5 ml of algal suspension was removed daily from both control and treated cultures. The culture samples were filtered on Whatman GF/C filter paper, and filters with attached cells were extracted overnight in 80% aqueous methanol while grinding with 0.5 mm silica beads. The extracts were then centrifuged twice at 10 000 g for 10 min at 4°C, and the supernatants of each extract were pooled and subjected to a sterilized air stream to evaporate the organic solvent. The concentration of microcystins in the remaining aqueous fraction was determined by ELISA according to the method of Carmichael & An (1999) using a commercial Abraxis kit (Abraxis, USA). To detect microcystins potentially accumulated in the body of Cyclops during grazing, 10 individuals of Cyclops were sampled from each bottle daily. The animals were ground with a known volume of 80% methanol in a household mortar and left overnight for complete extraction of MCs. This step was repeated twice. The methanolic extracts were centrifuged at 10 000 rpm, and the relevant supernatants were combined together. The organic solvent was then evaporated through a sterilized air stream, and MC concentration in the remaining aqueous fraction was detected by ELISA as described above. The toxicity of MCs accumulated in the body of Cyclops was confirmed by a protein phosphatase inhibition assay (PPIA) according to Carmichael & An (1999).
Statistical analysis
Data analysis was performed using SPSS for Windows (SPSS Inc., V17). The one-way ANOVA test was used to compare significant differences (p < 0.05) in M. aeruginosa cell density, the ingestion rate of Cyclops, and microcystin concentrations between the treatment and control groups.
Results
The results of the present study have revealed that Cyclops caused a significant reduction in the growth of M. aeruginosa as estimated by the number of cells (p < 0.05), compared to control cultures (Fig. 1).
The cell density of M. aeruginosa declined steeply in cultures containing cyclops and was reduced to zero after seven days of incubation, while the cell density in control cultures (i.e. not-containing cyclops) increased during the incubation period and reached its exponential phase at 4-day incubation (Fig. 1).
Figure 1
Changes in cell density of M. aeruginosa (cells ml-1) over the course of the grazing experiment with Cyclops vicinus
The grazing rate of Cyclops on Microcystis significantly differed during the experiment and ranged from 68.9 to 606.3 M. aeruginosa cells (Cyclops)-1 d-1 (Fig. 2). This ingestion rate seems to be correlated with the cell density of M. aeruginosa in the cultures (r = 75). The highest grazing rate was obtained on the fourth day of incubation, when the number of Microcystis cells reached its maximum (67 186 cells ml-1) in the control cultures. Whereas, the lowest rate was determined after a 1-day incubation with cell density of 15 562 cells ml-1 (Figs 1, 2). Meanwhile, Cyclops also fed on edible food (A. falcatus) in addition to M. aeruginosa, but with lower grazing rates (71.83–385.43 Anksitrodesmus cells (Cyclops)-1 d-1) (Fig. 2). In addition to grazing on M. aeruginosa, the copepod Cyclops induced microcystin production within Microcystis cells during the feeding period, compared to control cultures (i.e. without Cyclops). MC concentrations in grazer-treated cultures were 1.67–12.5 times higher than those in control cultures (Fig. 3). These concentrations increased with the duration of cell exposure to a grazer, with the highest level obtained on the ninth day of incubation (189.5 pg MC cell-1). Additionally, our results showed that Cyclops accumulated MCs at levels ranging from 0.05 to 3.21 μg Cyclops individual-1 in their body during the feeding on M. aeruginosa.
Figure 2
Grazing rate of Cyclops vicinus feeding on M. aeruginosa and A. falcatus
Figure 3
Changes in microcystin content in M. aeruginosa cells (pg cell-1) in the presence of the grazer Cyclops vicinus and control cultures
Discussion
Our results confirm the ability of the copepod C. vicinus to coexist with and feed on the toxic cyanobacterium M. aeruginosa. We fed Cyclops mixtures of M. aeruginosa with good food algae (A. falcatus) rather than pure diet (M. aeruginosa), as pure diet of Microcystis reduces fitness and survival of zooplankton, and it is unlikely to occur in nature (Wilson et al. 2006; Ger et al. 2010). Despite the presence of edible food in the cultures, Cyclops were able to graze down toxic M. aeruginosa with grazing rates higher than those for the edible alga A. falcatus. These results are thus not consistent with previous studies showing that copepods can graze more on cyanobacteria when alternative food is scarce (Gorokhova, Engstrom-Ost 2009; Engström-Öst et al. 2011). Our results also contradict the hypothesis that cyanobacteria are generally not considered as preferred food for copepods (DeMott et al. 1991). However, these results are consistent with those obtained by Kozlowsky-Suzuki et al. (2003) reporting high rates of toxic cyanobacteria ingestion by copepods. Furthermore, Urrutia-Cordero et al. (2015) demonstrated the suppression of phytoplankton assemblages dominated by Microcystis, Anabaena, and Planktothrix by grazing of cyclopoid copepods. In response to grazing in our experiments, M. aeruginosa exhibited a several-fold increase in MC content (189.5 pg MC cell-1) compared to that detected in control cultures (15.2 pg MC cell-1). This indicates that M. aeruginosa induces toxin production as a way to combat or prevent grazing by copepods. These results are in line with those of previous studies reporting that toxic cyanobacteria increase toxin production as a defense strategy when they face the grazing pressure (Jang et al. 2003; Lurling 2003; Mohamed, Al-Shehri 2013; Selander et al. 2015). Interestingly, no mortality of Cyclops was observed during the feeding on M. aeruginosa. Also, we did not notice any reduction in the growth of the Cyclops body in this study, as the body size of Cyclops fed with M. aeruginosa did not differ significantly from that of Cyclops fed with the edible alga A. falcatus in initial Cyclops cultures (data not presented in this paper). The results of our study are therefore not in line with those of previous studies that demonstrated a reduction in the growth of different zooplankton species during feeding on toxic cyanobacteria (Smutná et al. 2014; Herrera et al. 2015; Kaczkowski et al. 2017; Kosiba et al. 2018). However, our results are in agreement with those of Paes et al. (2016), who found that the biomass of calanoid copepods positively correlates with the microcystin concentrations in the environment. The absence of a negative effect of toxic Microcystis on Cyclops strongly supports the hypothesis that some herbivores are able to feed on toxic cyanobacteria and develop resistance to their cyanotoxins (Panosso et al. 2003). However, this resistance seems to be determined by the history of exposure to these toxins, as copepods that experience interaction with toxic cyanobacterial blooms could develop resistance to cyanotoxins and may be able to graze better on toxic cyanobacteria than unexposed copepods (Paes et al. 2016). Our data best fit this explanation, where Cyclops individuals used in the study were collected from eutrophic fishponds containing heavy blooms of toxic cyanobacteria. Otherwise, the discrepancy in the resistance of copepods to cyanotoxins could be attributed to species-specific differences through the capacity of some copepod species for detoxification of cyanotoxins (Ger et al. 2010).
The current study has also provided evidence for MC accumulation by Cyclops – the total MC concentrations detected in Cyclops tissues increased during the period of feeding on M. aeruginosa (0.05 to 3.21 μg Cyclops individual-1). The exponential accumulation of MCs during feeding on toxic M. aeruginosa can be interpreted as an indirect indication of viability of Cyclops in this experiment as suggested by Shams et al. (2014). Despite several studies on MC accumulation in a variety of aquatic organisms, including bivalves, crustaceans, zooplankton and fish (Mohamed 2001; Mohamed et al. 2003; Zhang et al. 2009; Wojtal-Frankiewicz et al. 2013), to our knowledge this is the first study to address the accumulation of MCs by the copepod Cyclops. In this regard, Johnson et al. (1999) stated that cladocerans and copepods contain large lipid reserves, which could allow lipophilic MCs to accumulate in their bodies. Since Cyclops is omnivores with a high protein level, they occupy an important position in aquatic food webs and serve as food for certain fish (Piasecki et al. 2004). Therefore, the detection of MCs in the Cyclops body in our study suggests the transfer of such potent toxins to higher trophic levels. Many studies have provided evidence for the transfer of MCs accumulated in zooplankton grazers to higher trophic levels of the food web (Ibelings et al. 2005; Oberhaus et al. 2007; Lehman et al. 2010; Sotton et al. 2014). The presence of these toxins in the aquatic food web not only poses a risk to fish and other aquatic organisms, but also represents a threat to human health through the consumption of MC-contaminated fish and other aquatic organisms. However, MC concentrations in zooplankton, particularly Daphnia, may decrease despite a constant influx of toxins from the environment (Smith, Haney 2006; Shams et al. 2014). This decrease was attributed to the detoxification and elimination of MCs via excretion systems in these animals (Shams et al. 2014). Accordingly, our study indicates that the MC detoxification process may not be efficient in Cyclops, probably due to high MC concentrations in the ambient environment. This is consistent with the results of Chen et al. (2005) reporting that high MC concentrations cause a reduction in antioxidant enzyme activities, possibly resulting from overburdening of the detoxification system by MC metabolism in Daphnia. Similarly, Shams et al. (2014) demonstrated that higher proportions of toxic Planktothix rubescens could shift the balance between accumulation and excretion/detoxification toward greater accumulation of MC in grazers.
In conclusion, this study has clearly demonstrated the effective and significant grazing of the copepod C. vicinus on the toxic cyanobacterium M. aruginosa in the presence of the edible food alga, A. falcatus. The study also revealed that M. aeruginosa exhibited induction in MC production in response to grazing by Cyclops with 1.67–12.5 times higher toxin concentrations compared to control cultures. However, Cyclops survived such high toxin concentrations without any noticeable lethal effects. Our results not only have important implications for the understanding of grazer-prey interactions, but also provide important information on the future use of biomanipulation with Cyclops to control cyanobacterial blooms in eutrophic water sources. Nevertheless, the accumulation of MCs in the Cyclops body upon feeding on toxic Microcystis cells may hinder the use of Cyclops in the biomanipulation of Microcystis blooms, because it represents a major source of food for fish and consequently can act as a transfer vector of cyanotoxins to higher trophic levels. However, the balance between MC accumulation and its detoxification in copepods seems to be dependent on toxin concentrations in the diet (Shams et al. 2014), and MC detoxification pathways in grazers can be more efficient at low MC concentrations compared to high MC concentrations, which could cause partial or full inhibition of detoxification pathways (Chen et al. 2005; Shams et al. 2014). Therefore, further in situ studies are needed to investigate the ability of Cyclops to assimilate and detoxify MCs at environmentally relevant concentrations. This will be the key determinant for the use of Cyclops as a potential biological control agent against Microcystis blooms.
