YU Ziyue, LIU Shuxing, CHEN Xin, WARREN Alan, LIN Xiaofeng ,and LI Jiqiu ,

1) Guangzhou Key Laboratory of Subtropical Biodiversity and Biomonitoring, South China Normal University,Guangzhou 510631, China

2) Department of Life Sciences, Natural History Museum, Cromwell Road, London SW7 5BD, UK

3) The Fujian Provincial Key Laboratory for Coastal Ecology and Environmental Studies, College of the Environment and Ecology, Xiamen University, Xiamen 361005, China

4) Key Laboratory of the Ministry of Education for Coastal and Wetland Ecosystem, College of the Environment and Ecology, Xiamen University, Xiamen 361005, China

Abstract Ecotoxic effects of antibiotics or ammonium have been confirmed independently in aquatic animals, but few studies have investigated their combined effects. In aquaculture ecosystems, these pollutants frequently coexist, and often in high concentrations. In this study, the combined effects of antibiotic nitrofurazone and NH4Cl on the population dynamics and growth rates of two species of ciliated protists, Euplotes vannus and Pseudokeronopsis rubra, were investigated. Profiles of the dose-responses were visualized, and interactions between the two pollutants were quantified by the response surface method (RSM). Results showed that 1)the dynamics of the population growth differed significantly between the testd ciliates and varied with the concentrations of the pollutants; 2) the relative growth rate (RGR) of both ciliates decreased significantly with increased pollutant concentrations, while the difference in RGR between the two ciliates was not significant; 3) RSM analysis demonstrated an additive effect of nitrofurazone and NH4Cl on the RGR in both ciliates. In brief, ecotoxic effects can be caused by nitrofurazone and ammonium independently on the two test ciliates, and such effects can be strengthened when they present at the same time. These findings offer a valuable reference for evaluating combined ecotoxic effects caused by multiple pollutants in aquaculture ecosystems.

Key words ciliates; combined ecotoxicity; nitrofurazone; NH4Cl; population growth

1 Introduction

With the large-scale development of the aquaculture industry, increasing amounts of pollutants have been produced and discharged into the aquatic environment resulting in serious ecological problems (Van Ducet al.,2018). Two pollutants of particular concern are high nitrogen load and antibiotics. It has long been known that high nitrogen load can accelerate the deterioration of water quality by facilitating eutrophication, anoxia, and acidification (Groenveldet al., 2019), and is toxic to aquatic animals (Liet al., 2019; Kimet al., 2019). Among soluble nitrogen substances, ammonia is the most toxic one. Most of the ammonia produced by the aquaculture industry originates as a byproduct of protein metabolism after feeding (Ebelinget al., 2006). Notably, the deterioration of water quality can reduce the health status of farm animals and induce disease outbreak, which will lead to the use of a large amount of antibiotics to prevent and control the disease (Hanet al., 2020). Antibiotic abuse in the aquaculture industry and the ecological effects caused by antibiotics have attracted significant attention (Hanet al., 2020).The antibiotic resistance in bacteria has led to the emergence of ultra-virulent strains and a subsequent increased use of antibiotics in the aquaculture industry (Liuet al.,2017). This has resulted in high amounts of antibiotics being released into the aquatic environment. Furthermore,the toxic effects of antibiotics are not selective, and nontarget organisms can be negatively impacted (V?litaloet al.,2017). Consequently, both ammonia and antibiotics can induce the ecological toxicity at all scales, ranging from individual organisms to populations and communities, and thus affect the ecosystem significantly. It is noteworthy that the combined effects of pollutants pose a more realistic, but more complicated, scenario than the effects posed by the pollutants individually (Farzana and Tam, 2018).Although the ecological toxicity of ammonia or antibiotic have previously been evaluated by biological responses(Xianget al., 2010; Zhouet al., 2011; Honget al., 2015,2017; Yanget al., 2019), little is known about the combined effects of these two pollutants in the same ecosystem.

For sustainable development, it is imperative to carry out environmental quality monitoring and ecological risk assessment. Among the biological responses to environmental change, population growth rate is a reliable index since it integrates molecular, cellular, tissue and wholeorganism levels of responses (P?rtneret al., 2010). Population growth rate is therefore a commonly used biomarker for ecotoxicity assessment (Smolderset al., 2004).In addition, the derived parameters of population dynamics can reflect the effect of environmental stress on population growth and characterize the response profile. Such parameters include the upper asymptote value (K), the theoretical population size at time zero (N0), and the per capita growth rate (r) drawn from the Logistic model of population dynamics. Previous studies have shown that integrating the population growth rate and growth dynamics of certain model ciliates can evaluate the environmental stresses (Honget al., 2017; Ruanet al., 2018).

Ciliated protists (ciliates) are important components of aquatic and terrestrial ecosystems and play significant roles in material circulation and energy flow (Payne, 2013). Furthermore, ciliates are widely used as model organisms in many fields related to biological research because of their biological properties, such as high diversity, short life cycle, cosmopolitan distribution, ease of isolation and cultivation, and rapid responses to environmental disturbance(Montagneset al., 2012; Liet al., 2019). Different ciliate species may not, however, respond in the same way to different stresses. In order to address this problem, two or more species are commonly used as test organisms in the same study (Honget al., 2015; Ruanet al., 2018). In the present study,Euplotes vannusandPseudokeronopsis rubrawere chosen as the model organisms. Both species are commonly found in marine aquatic environments but have different morphological features, phylogenetic positions,behaviors, and ecological niches (Hu and Song, 2001; Esteban and Finlay, 2007; Chenet al., 2011). Under normal culture conditions,Euplotes vannusgrows faster thanPseudokeronopsis rubra.Generation time ofEuplotes vannusis approximately 10 hours (Xuet al., 2011) and shorter than that ofPseudokeronopsis rubra(approximately 40 h, our unpublished data).

Nitrofurazone is a broad-spectrum antibiotic. Despite the introduction of policies to prohibit the use of nitrofurazone because of its toxicity to humans, other mammals, shellfish, and microeukaryotes (Martinezet al., 1995;Vlastoset al., 2010; Zhouet al., 2011), it is widely employed in the aquaculture industry because of its low cost and high efficiency for controlling bacterial infections in farm animals (Duet al., 2014). It is difficult to determine the actual intake of nitrofurazone by animals through the determination of parent compounds or metabolic residues because it is metabolized rapidly and its half-life timein vivois only a few hours (McCrackenet al., 1995). Additionally, there is no effective method for the detection of nitrofuran residues in animals (Bendallet al., 2019), which might be another important reason that such drugs are used illegally and difficult to be regulated. Most of the incidences of nitrofuran metabolites in food are connected with illegal usage of these drugs (McCrackenet al., 1995).For example, it was reported that nitrofurazone was recently detected in fluid milk and dairy powders (Bendallet al., 2019). In brief, both the research of nitrofuran antibiotic detection method and its ecological toxicity have important scientific value and practical significance.

The ecotoxicity of nitrofurazone on ciliates has been previously evaluated (Zhouet al., 2011; Honget al., 2015;Honget al., 2017). It has been shown that nitrofurazone and ammonium each has its own unique toxic mechanism(Zenget al., 2017). However, since few studies have investigated the toxicity of nitrofurazone and ammonium in combination to aquatic organisms, it is not clear whether the interactive effects between these pollutants depend on their toxic mechanisms. In the present study, we hypothesize that the combined ecotoxic effects of nitrofurazone and NH4Cl, whether they are antagonistic, synergistic or additive, depend on an inherent intersection of the toxic mechanisms between them. In order to test this hypothesis,we measured the population dynamics and growth rates ofEuplotes vannusandPseudokeronopsis rubraexposed to different concentrations of nitrofurazone and NH4Cl independently or in combination.

2 Materials and Methods

2.1 Chemical Reagents and Solutions

Nitrofurazone (5-nitro-2-furfural semicarbazone) and ammonia chloride (NH4Cl) were obtained from Sigma-Aldrich Shanghai Trading Co. Ltd., Shanghai, China. Stock solutions of nitrofurazone (100 mg L?1) and NH4Cl (1000 mg L?1) were prepared by dissolving the reagents in sterilized artificial marine water (AMW, 28 g of NaCl, 0.8 g of KCl, 5 g of MgCl2·6H2O, and 1.2 g of CaCl2in 1000 mL of distilled water, pH 8.2, salinity 30). Test solutions with different concentrations were prepared by diluting the stock solution with appropriate volume of AMW.

2.2 Organisms and Culture

The marinehypotrich (s. l.) ciliatesEuplotes vannusandPseudokeronopsis rubra,which are common in coastal waters, were used as the test organisms. Cultures of both species were obtained from the Laboratory of Protozoology, Ocean University of China, Qingdao, China, and had previously been identified based on their morphology and small subunit rDNA sequences (Hu and Song, 2001; Chenet al., 2011). For the present study, clonal cultures were maintained in AMW at 17℃. Rice grains were added as a food source for the ciliates to facilitate the growth of bacteria.

2.3 Determination of Population Growth Dynamics

The toxicity tests were conducted in flat-bottomed culture plates, each with 24 wells (height: 10 mm; diameter:16 mm) at 25℃. After an acclimation period of 24 h, ciliates were inoculated into a series of culture systems with initial population of 10 cells in 1 mL medium per well. To evaluate the ecotoxic effects of nitrofurazone and NH4Cl independently or in combination, a two-factor random experimental design was used as follows: three concentration gradients of nitrofurazone (0, 6, and 18 mg L?1) and four concentration gradients of NH4Cl (0, 150, 300, and 600 mg L?1) were combined freely between the two factors to form 12 culture systems (treatments). Hereinto, the maximum concentration of nitrofurazone used here is the usual dosage used to treat bacterial diseases in fish (Colorni and Paperna, 1983), while the maximum concentration of NH4Cl was approximately twice the 24 h LD50in both tested ciliates (our unpublished date). Each treatment was randomly assigned to triplicate wells. The duration of each exposure was 48 h.

During the 48-h exposure period, half of the culture media was renewed after 24 h, while all the ciliates were retained in the original wells by screen filtration (Yanget al.,2019). The bacteriumEscherichia coliwas used as food for the ciliates, the initial inoculation density of which was regulated to approximately 107cells mL?1by using a hemocytometer under a light microscope. The population sizes of ciliates were determined by enumerating cells at 6-h intervals under a dissecting microscope (20× – 30×).Those ciliate cells unable to swim or creep on the bottom of the well, together with disappeared cells, were regarded as dead (Madoni and Romeo, 2006).

2.4 Calculation of Population Growth Parameters

The population growth dynamics were characterized by parameters derived from the Logistic equation (1). The Logistic growth curves were iteratively fitted to numbers of ciliate cells (population size) using the Marquardt-Levenberg least squares algorithm. Parameters were then determined from the following equations using SigmaPlot 11.1 (Systat Software Inc., Germany):

In Eq. (1),Ntis the population size (number of ciliate cells) at a given time pointt;ris per capita growth rate(h?1);Kis the upper asymptote of the population growth curve; whenr≥ 0, or bothr＜ 0 andK ≤ N0,Kis the carrying capacity; whenr＜ 0 andK ＞ N0,Klose its biological meaning and is just a parameter of the equation;N0is the theoretical population size of ciliates at the beginning of experiment derived from Eq. (1):

In Eq. (2),RGRis the relative growth rate of tested ciliates;Twas defined in this study as the time period from inoculum to extinction, and if the population was not extinct during the 48-h exposure period,T= 48 h;Niis the number of ciliate cells at the beginning of the experiment(i.e., the inoculum);Ntis the number of ciliate cells (population size) at a given time point (t).

2.5 Evaluation on Interactive Effects by Using Response Surface Methodology (RSM)

In order to visualize theRGRprofile and evaluate the interactive effects between the tested variables, the response surface methodology (RSM) was performed and polynomial regression equations were developed to describe the effects of nitrofurazone (X1) and NH4Cl (X2).Independent variables of nitrofurazone concentration (X1;0, 6, and 18 mg L?1) and NH4Cl concentration (X2; 0, 0.15,0.30, and 0.60 g L?1) were designed to investigate their effects on the dependent variables ofRGR. The general model of the quadratic polynomial regression equation was employed as shown in Eq. (3). By using this equation, the linear (X1,X2), quadratic (X21,X22) and interactive (X1×X2) effects of the independent nitrofurazone (X1) and NH4Cl (X2) on the dependent variable (RGR) were determined by

In Eq. (3),β0is a constant (intercept);βiis the linear coefficient;βiiis the quadratic coefficient;βijis the coefficient of interactive effects, and;XiandXjare independent variables. The RSM was used to analyze the experimental data using Design Expert Version 8.0.6 software(Stat-Ease Inc., Minneapolis, MN). Response surface and contour plots in two dimensions were developed in the second order polynomial models. The analysis of variance(ANOVA) test was applied to identify interactions between the independent variables using Design-Expert program.

2.6 Statistical Analyses

All the statistical analyses were carried out using SPSS software (version 17.0 for Windows). Results are presented as means ± SD (standard deviation of means). The difference is statistically significant whenP≤ 0.05. Before statistical analyses, the raw data were transformed by positive number and logarithmic transformations. Subsequently, these data were tested for distribution of normality (Kolmogorov-Smirnov test) and homogeneity of variance (Levene test). In cases of normal distribution and homogeneity of variance, an analysis of variance (ANOVA) ort-test was performed on the variables to analyze the significance of any differences. When significant differences were detected, differences between means were compared by the least significant difference (LSD) multiple range test. If the conditions are not met for ANOVA ort-test, the nonparametric test was used to determine the significance of differences between samples.

3 Results

3.1 Population Growth Dynamics and Derived Parameters

The dynamics of population growth differed betweenEuplotes vannus(Figs.1A, C, E, and G) andPseudokeronopsis rubra(Figs.1B, D, F, and H) and varied with exposure concentrations of nitrofurazone and NH4Cl both independently and in combination. Results intuitively showed that the population growth dynamics of the two ciliates differed in all tests.

Characteristic parameters were derived from population growth dynamics forEuplotes vannus(Figs.2A, C, and F)andPseudokeronopsis rubra(Figs.2B, D, and G). The upper asymptote value (K) differed significantly among nitrofurazone-treatedE. vannuscells when the NH4Cl concentrations were 150 mg L?1and 300 mg L?1(Fig.2A;P ＜0.05). InP. rubra, significant differences caused by nitrofurazone were detected at all the tested concentrations of NH4Cl (Fig.2B;P≤ 0.05).

Fig.1 Growth dynamics of Euplotes vannus (A, C, E, and G) and Pseudokeronopsis rubra (B, D, F and H) exposed into different concentrations of nitrofurazone in combination with serial concentrations of NH4Cl. The four concentration gradients of NH4Cl were 0 mg L?1 (A and B), 150 mg L?1 (C and D), 300 mg L?1 (E and F), and 600 mg L?1 (G and H).

With respect to the per capita growth rate (r), exposure to 600 mg L?1NH4Cl did not result in significant differences in nitrofurazone-treatedE. vannuspopulations. However, significant differences inrwere detected among other populations exposed to the pollutants in combination (Fig.2C;P ＜0.05). Similarly, no significant differences inrwere detected in nitrofurazone-treatedP. rubrapopulation apart from those cells treated with nitrofurazone in combination with 150 mg L?1NH4Cl (Fig.2D;P≤ 0.05).

In regard to the theoretical population size of ciliates at the beginning of experiment (N0), generally there were significant differences in the tested ciliates exposed to different combinations of nitrofurazone and NH4Cl (Figs.2F and G;P ＜0.05). The exceptions were nitrofurazone-treatedE. vannusin combination with 300 mg L?1NH4Cl, and nitrofurazone-treatedP. rubrain combination with 300 or 600 mg L?1NH4Cl (Figs.2H and G;P ＞0.05).

Results of M-W U test demonstrated significant differences inrandN0between the two ciliate species (Figs.2E and H;P ＜0.05).

3.2 Relative Growth Rate (RGR)

Comparative analyses were performed on relative growth rates (RGRs) between the two species exposed to nitrofurazone and NH4Cl, both independently and in combination (Fig.3). Firstly, results of Welch’st-test showed no difference inRGRbetweenE. vannusandP. rubrain the absence of nitrofurazone and NH4Cl (Fig.3A;P ＞0.05).Significant differences were detected inE. vannusexposed to different concentrations of nitrofurazone (Fig.3B;P ＜0.05),whereas no significant differences were detected inP. rubra(Fig.3C;P ＞0.05). For both species, RGRs of NH4Cltreated populations differed significantly at each concentration of nitrofurazone (Figs.3D and E;P ＜0.05).

Significant differences inRGRwere detected between the two species when exposed into different concentrations of NH4Cl (Figs.3F and H;P ＜0.05). Furthermore, at each concentration of NH4Cl, RGRs of nitrofurazone-treated ciliates differed significantly.

3.3 Effects of Interaction Between Nitrofurazone and NH4Cl on RGRs

The response and contour plot of RGRs changed with the concentrations of either nitrofurazone or NH4Cl (Fig.4).Overall, RGRs decreased with the increasing concentrations of nitrofurazone and NH4Cl (Figs.4A, B, C, and D).Furthermore, at higher concentrations of NH4Cl, increasing the concentration of nitrofurazone caused a dramatic decrease inRGR, especially at the lower concentration range of nitrofurazone (Figs.4A and C). This phenomenon was more pronounced inP. rubrathan inE. vannus(Figs.4B and D).

ANOVA was performed on the response surface quadratic model, equation, and the correlation coefficients for both nitrofurazone and NH4Cl (Table 1). Correlation coefficients (R2) in theRGRmodel ofE. vannuswas calculated to be 0.9522, which implied 95.22% of variation could be explained by this model. Similarly,R2in theRGRmodel ofP. rubrawas calculated to be 0.8503, which implied that 85.03% of variation could be explained by this model. Overall, these two quadratic models fitted well with the experimental data. Inferred from theF-value andP-value, the variablesX1(nitrofurazone),X2(NH4Cl), and their squares () were all significant in theRGRof bothE. vannusandP. rubraand presented negative effects; In contrast, the negative effect ofX1×X2onRGRwas not statistically significant, suggesting there was no significant interaction between nitrofurazone and NH4Cl.

Fig.2 Parameters of the upper asymptote value (K), the per capita growth rate (r) and the population at time zero (N0) derived from logistic growth dynamics for Euplotes vannus (A, C and F) and Pseudokeronopsis rubra (B, D and G). E, H:Comparison between the two ciliate species. Columns bearing the same superscript letter are not significantly different.Outlier: the outlier without statistical significance.

Fig.3 Relative growth rate of Euplotes vannus (B, D, F, and H) and Pseudokeronopsis rubra (C, E, G and I) exposed to different concentrations of nitrofurazone in combination with serial concentrations of NH4Cl. A, Relative growth rates were compared between the two ciliate species. Columns and boxes bearing the same superscript letter are not significantly different.

Fig.4 Response surface (A and C) and contour plots (B and D) for interactive effects between nitrofurazone and NH4Cl on relative growth rates in the two ciliate species. A and B from Euplotes vannus; C and D from Pseudokeronopsis rubra.Con., concentration.

Table 1 Analysis of variance (ANOVA) for response surface quadratic model

4 Discussion

The toxic effects induced by nitrofurazone and NH4Cl have been confirmed in many animals, for example, the ecotoxicity of nitrofurazone on mice (Kariet al., 1998),rats (Itoet al., 2003), the insectMuscadomestica(Macri and Sbardella, 1984), and ciliates (Honget al., 2015, 2017);and the ecotoxicity of NH4Cl on fish (Hegaziet al., 2010),crabs (Romano and Zeng, 2009),Daphnia(Xianget al.,2010), and microalgae (Liet al., 2019). Few studies, however, have focused on the toxic effect of NH4Cl on ciliates or on the effects of nitrofurazone and NH4Cl in combination. The present study demonstrated that both nitrofurazone and NH4Cl exposure independently or in combination had a negative effect on the population dynamics and growth rates ofEuplotes vannusandPseudokeronopsis rubra.

In terms of ciliate growth dynamics, it can be concluded intuitively that the profiles of the population sizes differ among the treatments. However, the differences between profiles can only be quantified by the function of the corresponding parameters. The growth curves of each species were modeled by the Logistic model of population dynamics. The effects of different concentrations of nitrofurazone and NH4Cl were demonstrated by comparing the upper asymptote value (K), the per capita growth rate(r), and the derived population size at time zero (N0). These derived parameters have specific biological meanings in theoretical ecology. For example,randKsignify the quality and quantity of the offspring respectively;Ksignifies the carrying capacity; and tradeoffs betweenrandKindicate the adaptation strategy of organisms under environmental pressure (Pianka, 1970). It is important to note that any model or theory has specific application conditions. Without the conditions, it will lose its biological significance. For this study, the population growth curve of the ciliates showed a downward trend. Consequently, in some casesKloses its biological significance because its corresponding culture time is negative. For example, ifr＜0 andK ＞ N0, the value of the upper asymptote (K) has no biological meaning. Therefore, when discussing the adaptability of the ciliates to their environment using these parameters, they have to fall within a biologically significant range. In theory, ciliates should be typicalr-strategists because of their biological traits, such as high fecundity, small body size, short generation time, and high capacity for dispersal (Weinbauer and H?fle, 1998). Thus, in the present study,Kcan, to a certain extent, play an auxiliary role in exploring the adaptability of ciliates to different concentrations of pollutants. The derived population size at time zero (N0) has been shown to be an important factor impacting the outcomes of interspecific competition for limiting nutrients among microorganisms such as freshwater algae (Hu and Zhang, 1993) and herbivorous rotifers (Sarmaet al., 1996). In theory, the value ofN0can indirectly reflect the survival ability of species under environmental stress (Tilman, 1982; Holtet al., 1994). This implies that species with lowN0values will have higher survival ability when the same growth performance is achieved under the same levels of stress. In the present study, significant differences were detected between these two ciliate species in terms ofrandN0in populations exposed to different concentrations of nitrofurazone and NH4Cl. This is consistent with the findings of Honget al.(2015) who reported differences betweenEuplotes vannusandPseudokeronopsis rubrain terms of theirbiological response (e.g., levels of gene damage) when exposed to nitrofurazone.

RGRis a commonly used index for measuring the growth rates of populations and individuals, and is expressed as the increase in size per time unit during a specified period of time (Hoffmann and Poorter, 2002). In the present study,RGRgenerally showed downward trends with increasing concentrations of either nitrofurazone or NH4Cl. However,no significant difference inRGRwas detected between the two ciliate species. The effects of multiple pollutants in combination can be divided into three types,i.e., antagonism, synergy and addition (Murphyet al., 2018). In the present study, the antagonistic and synergistic effects of nitrofurazone and NH4Cl on theRGRof either species was not significant although there was a slight negative impact due to synergy. In previous studies, it was suggested that the combined effect may depend on the mechanism of the pollutants to a large extent (Taoet al., 2018).For ciliates, the toxic mechanisms of ammonium nitrogen can be attributed to effects such as inducing the depolarization of ATP depletion leading to cell death (Rodrigueset al., 2014) or ammonia excretion impairment (Kimet al.,2019). In the case of nitrofurazone, oxidative superoxide anion radicals and hydrogen peroxide are produced when it is metabolized (Ghersi-Egeaet al., 1998). These can react with Cu(I) to produce the primary reactive species capable of causing DNA damage (Hirakuet al., 2004).Thus, there seems to be no common toxic mechanisms between nitrofurazone and NH4Cl. Their effects on the two ciliate species when applied in combination is additive rather than synergistic or antagonistic. This finding is supported by the fact that there were no significant effects on theRGRof ciliate due to antagonism or synergy (see Table 1).

RSM is an effective mathematical and statistical tool for measuring not only the effect of independent variables but also their interactive effects (Draper, 1997). RSM has been used in a variety of fields to design experiments for optimizing processes and production including food, medicine, and chemicals, and for successfully predicting the effects of technological parameters (Belwalet al., 2016).In the present study, profiles of the dose-responses (RGRs)induced by nitrofurazone and NH4Cl were visualized by utilizing surface and curve options in RSM. The experimental data were fitted to a second-order polynomial model and the regression coefficients were obtained by multiple linear regression whereby the various variables and their interactions were quantified using statistical analyses. Results of ANOVA for the response surface quadratic model showed that coefficients of interaction between nitrofurazone and NH4Cl were negative in both ciliates,though the effects were not statistically significant (Table 1). This indicates that there is a significant additive effect on the toxic mechanism between the two pollutants, but there is no significant synergistic effect. To date, no evidence suggests there is an interaction of toxic mechanisms between ammonium nitrogen and nitrofurazone in aquatic animal (Martinezet al., 1995; Vlastoset al., 2010;Zhouet al., 2011; Groenveldet al., 2019; Kimet al., 2019;Liet al., 2019). The findings of the present study provide further evidence for the potential of RSM as a powerful tool for evaluating the combined ecotoxic effects induced by multiple pollutants.

In summary, nitrofurazone or NH4Cl can independently negatively impact the population growth rates and dynamics ofEuplotes vannusandPseudokeronopsis ruba,althoughtheir responses differ in terms of their derived growth dynamic parametersK,randN0. When applied in combination, nitrofurazone and NH4Cl produced additive effects and negatively impacted the relative growth rates of both ciliates. However, the difference between two species of ciliates was not significant. RSM can be used as a powerful tool for evaluating the combined ecotoxic effects induced by multiple pollutants.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos. 31971519, 31430077, 41476128)and in part by a grant from the Dedicated Fund for Promoting High-Quality Economic Development in Guangdong Province (Marine Economic Development Project:GDOE(2019)A23).