The effect of silica-magnetite nanoparticles on the ecotoxicity of the antibiotic ciprofloxacin

The increase in the production and application of engineered nanomaterials, including nanoparticles (NPs), leads to their discharge into the environment, where they can interact with coexisting antibiotics from wastewater, causing a complicated joint effect on organisms that need to be studied. Herein, a typical engineered nanomaterial, silica-magnetite NPs modified with tetraethoxysilane and 3-aminopropyltriethoxysilane (MTA-NPs, 1–2 g/L), and common antibiotic ciprofloxacin (CIP, 0–5 mg/L) were selected as the analytes. Their joint toxicity to a model of ciliates infusoria, Paramecium caudatum was specifically investigated. The impact of CIP, MTA-NPs, and humic acids (HA) was tracked for 24 h, individually and collectively, on the mortality of infusoria. The addition of MTA-NPs and HA at the studied concentrations leads to 40% mortality of organisms. The combined presence of the MTA-NPs at a concentration of 1.5–2 mg/L and HA at a concentration of 20–45 mg/L has a multiplier effect and allows to reduce the mortality rate of ciliates > 30% due to the enhanced removal of CIP. That finding demonstrated a clearly detoxifying role of dissolved organic matter (here, humic substances) in case of complex water pollution where pharmaceuticals and nanomaterials are presented.


Introduction
Currently, the world has a problem of pollution of natural waters and soils with drugs, for example, antibiotics. (Gothwal and Shashidhar 2014). One of the most widely distributed antibiotics is ciprofloxacin (CIP), which belongs to the group of second-generation fluoroquinolones. Fluoroquinolones (FQs) exhibit a broad antibacterial spectrum by inhibiting bacterial DNA gyrase activity. They are frequently employed to treat many bacterial infections in human and veterinary medicine. (Davis et al. 1996;Acaröz and Sözbilir 2020;Acaröz et al. 2021).
The CIP content in streams and wastewater has been found to be approximately < 1 mg/L. However, the concentration of the antibiotic in hospital wastewater was significantly higher from -3 to 87 mg/L, and from enterprises that produce drugs, -31 mg/L (Larsson and Pedro 2007) .
Getting into aquatic natural objects, CIP affects the viability of living organisms to varying degrees. Thus, the concentrations of CIP at which various algal species lost 50% of their original population in the above toxicity studies ranged from 7.9 to 23,000 µg/L (Fu et al. 2017;Martins et al. 2012;Robinson et al. 2005). Eukaryotic microalgae are generally less sensitive than prokaryotes (Martins et al. 2012;Robinson et al. 2005). For Daphnia magna, 36.49 mg/L of the antibiotic leads to death of 50% of animals (Dionísio et al. 2019). Data on the threshold concentrations of CIP for protozoa such as ciliates are not available. Therefore, one of the first tasks of this study was to examine the ecotoxicity of CIP on ciliates.

Responsible Editor: Mohamed M. Abdel-Daim
The fate of CIP within aquatic systems is likely dependent on two most important mechanisms for its elimination from water: photodegradation (Cardoza et al. 2005) and sorption (Golet et al. 2002;Belden et al. 2007). When photodegraded, the half-life of CIP is approximately 2 h in natural water, which probably leads to loss of antibiotic activity (Cardoza et al. 2005). The presence of organic matter in the aquatic environment prolongs half-life of CIP (Lin et al. 2017) , which may result in antimicrobial activity on the solid matrix or on organisms that consume the solid matrix (Belden et al. 2007).
Dissolved organic matter (DOM) has a significant effect on the fate of antibiotics in the environment, being present in almost all aquatic ecosystems with a content of 0.1 to 10 mg/L. Humic matter (HM) is the most important part of the organic matter in surface waters (Nebbioso and Piccolo 2013) . Essentially, humic substances (HS) can be viewed as supramolecular structures composed of several thousand different molecules (Piccolo 2001) . To understand how antibiotics migrate in the environment, it is important to understand the mechanism of interaction between HS and antibiotics. This process is not well understood. Thus, the second task of this study was to evaluate CIP binding to humic acids (HA), which represents naturally occurring HS in aquatic systems, prior to the use of nanoparticles of magnetite NPs. Another question to be examined was whether binding would affect the bioactivity of test organisms like ciliates. This was accomplished by measuring the sorption rate of CIP in a laboratory study with and without the presence of the HS.
Like antibiotics, NPs are also a class of new contaminants that can accumulate and enter the environment. NPs are currently used in consumer products such as cosmetics, therapeutics, drug delivery systems, food packaging, diagnostics, and biosensors (Ray et al. 2009;Pereira and Oliveira 2012;Pastrana-Martínez et al. 2015;Laurent et al. 2008). Due to increased use, NPs are being released into the environment more and more and are already a new class of pollutants of concern (Ray et al. 2009). Magnetite NPs undergo many chemical and physical changes when released into the environment, changing their surface. (Philippe and Schaumann 2014;Aiken et al. 2011). It has been found that nearby chemicals from the environment readily coat the surfaces of the nanoparticles via surface ligand exchange. It is assumed that, in an aqueous medium, Fe and O atoms on the surface of magnetite NPs will adsorb OH − and H + ions. Additionally, it is possible that due to the hydroxyl-rich surface, magnetite NPs will bind CIP at the carboxylic acid moiety (Ma et al. 2003;Rehana et al. 2015).
The use of NP magnetite and antibiotics continues to grow. This is likely to escalate their environmental impact. Based on this, it is important to study not only the effect of individual pollutants but also the collective ones. The engineered silica-magnetite NPs are a prototype (model) of natural inorganic colloids mainly composed of iron oxides (Wigginton et al. 2007), they are also presented in almost all surface waters that are a part of aquatic ecosystems as well (Philippe and Schaumann 2014). The high mobility and surface area of humic substances are known to play a key role in pollutant transportation (Wigginton et al. 2007) as well as interaction with microbial communities (Bonneville et al. 2006;Neal et al. 2005) and ciliates (Li et al. 2012). The behavior of colloidal nanoparticles depends on the humic substances, as they are able to change their surface properties and hence their stability (Aiken et al. 2011), along with soil transportation ). In addition, HS strongly influences the adsorption of various ecotoxicants in colloids (Philippe and Schaumann 2014).
Therefore, this study aimed to estimate the joint effect of CIP and magnetite NPs in the presence of HS toward ciliates Paramecium caudatum. P. caudatum is a model organism that has made important contributions to molecular and cellular biology. P. caudatum are visible to the naked eye due to their rather large size (50-300 µm in length) (Houten 2019) . R. Mayne et al. (2018) showed that P. caudatum cells consumed starch that was coated with magnetite NPs in amounts exceeding 5-12% of their body volume. This proves that P. caudatum is a candidate organism for nanomaterial manipulation and delivery. Magnetic restraints of Paramecium were also shown by S. Furukawa and T. Kawano through the internalized magnetite (particles about 3 µm in diameter) (Furukawa and Kawano 2012) . The surface modification of magnetite NPs with silica (tetraethoxysilane and 3-aminopropyltriethoxysilane, MTA-NPs) presents a model of inorganic colloids.

Chemicals"
CIP (P98%) was obtained from the pharmaceutical factory "Kelun-Kazpharm" (Kazakhstan). Leonardite-standard humic acids (HA, Powhumus, Humintech, Germany) was used as acquired. The sample of silica-magnetite NPs modified with tetraethoxysilane and 3-aminopropyltriethoxysilane (MTA-NPs) has the following characteristics: average particle size of ~ 12 nm and adsorption surface area of − ~ 120 m 2 /g. All of the other chemicals and reagents employed in this work were of analytical and molecular grade and were acquired from commercial sources.

Ciliates acute toxicity test
Ciliates acute toxicity test based on ciliates mortality was conducted by using Paramecium caudatum Ehrenberg as described previously (Bondarenko et al. 2020). Stock cultures of P. caudatum were maintained at room temperature in Petri dishes containing sterile mineral medium Lozin-Lozinskiy (with the following composition, mg/L: NaCl = 100.0, KCl = 10.0, CaCl 2 ·2H 2 O = 10.0, MgCl 2 ·6H 2 O = 10.0, and NaHCO 3 = 20.0) and a boiled rice grain or a yeast suspension as feed.
For biotest analysis, culture is used at the beginning of the stationary growth phase, which occurs 2-3 days after reseeding into a new portion of the nutrient medium. In this case, the culture density will be (4000 ± 1000) cells/mL. The replicates represented by the 3-5 wells of the immunological plate included from 8 to 10 units of unicellular ciliates. The total sample size of ciliate populations in the analysis of the toxicity of each sample was at least 30 unicellular individuals.

Bacterial acute toxicity test
A bacterial acute toxicity test was conducted as described previously (Yakimenko et al. 2022). Bacterial toxicity was determined by a bioluminescence inhibition test as described by Zarubina et al. (2015). We used the lyophilized bacteria with a genetically modified strain of luminous Escherichia coli K12 TG1 carrying the lux operon of luminous soil bacteria, Photorhabdus luminescens ZM1. The strain was produced and stored in the collection of the Microbiology Department of the Faculty of Biology, Moscow State University (Danilov et al. 2002).
Bacterial suspension (0.1 mL) and 0.9 mL of test solutions (or distilled water as control) were placed into vials, and the luminescence intensity (I) was measured after 30 min of exposure using a Biotoks-6MS luminometer (Russia), recording it is counted per second.
The sample toxicity was assessed using the toxicity indicator (T): T = (I 0 -I/I 0 ) × 100, where I 0 and I are luminescence intensities of the control and experimental samples in counts per second, respectively, at the fixed exposure time (30 min) at room temperature (22 °C). T was determined automatically using the in-built software of the luminometer.
The toxicity was classified by using a generally recognized approach: T < 20 permissible toxicity; 20 ≤ T < 50 toxic; T ≥ 50 highly toxic. If T < 0 (a sample stimulates bacterial luminescence), the sample is assumed to be nontoxic (Danilov et al. 2002;Zarubina et al. 2015).

Equilibrium sorption studies
A sorption technique was used to quantify CIP and MTA-NPs sorption in the presence of HA. 1 and 2 g/L of MTA-NPs and/or HA of 0.01 and 0.05 g/L was weighed in and placed into a 200-mL conical flask and filled with 100 mL of distilled water with CIP concentration of 5 mg/L to conduct the adsorption experiments. All flasks were rotated at 30 rpm in the dark on a rotary shaker for 24 h at 298 K. After the reaction time of 24 h, the MTA-NPs and the supernatant were separated as described above. At the end of the equilibration period, the suspensions were centrifuged at 1000 rpm for 10 min, and the supernatants were extracted by syringe and filtered through a hydrophilic membrane (0.45 lm) for CIP quantification by the UV-Vis spectrophotometry. After filtration, the pH of the supernatant was measured. Its values are given in Table 1. Equilibrium CIP concentrations were determined by transferring 2 mL aliquots to a quartz cuvette. Samples were analyzed on a Solar PB2201 UV-Vis spectrophotometer in cuvettes.

Experiment design
Response surface methodology (RSM) is a combination of statistical and mathematical approaches to identify the best conditions for conducting an experiment. Design-Expert 13 software was used to generate statistical models. The surface response methodology was used to evaluate the relationship between a number of independent factors and responses in order to optimize the latter.
The mortality of infusoria, Paramecium caudatum, (%) is the response of the system (Y), and the two parameters including MTA-NPs and HA concentrations, with three levels [low (− 1), high (+ 1), and median (0)], were independent variables. The typical concentration of HA ranges from 1 to 50 mg L −1 (carbon) in soils and groundwater . The ranges of factors and levels of independent variables are shown in Table 2.

Statistical analyses
Comparisons among EC50 values for different concentrations of humic acids and different pH were analyzed using a two-factor analysis of variance (ANOVA; C(HS) as factor one and pH as factor two). Comparison among treatments was conducted by using a Fisher's probable least-squares difference test (PLSD; p < 0.05). Individual comparisons between pH levels for each C(HS) were also performed (t-test) in order to enhance the description of a positive interaction noted for the two-factor ANOVA. All computations were made by using StatView (SAS Institute, Version 5.01, Cary, NC, USA).

Bactericidal action of ciprofloxacin
CIP is a bactericidal drug affecting gram-negative and grampositive organisms (Yakovlev 1997). Therefore, first of all, it is important to trace how much this antibiotic affects the persistence and activity of environmental bacteria. Environmental bacteria are incredibly ubiquitous and diverse and play a crucial role in the cycling of elements within our environment (Knoll et al. 2012). No effect of CIP on luminous bacteria was observed at CIP concentrations of 10 mg/L. The toxicity index was 0 (Fig. 1). The luminescence of the biosensor increased after the addition of predetermined concentrations of the antibiotic. When the concentration of CIP was increased to 100 mg/L, the toxicity index reached a value of 29.11, which indicates a toxic effect.
The calculation of effective concentrations by using a probit regression model showed that the EC50 for the bacterial test culture is 135.73 mg/L of CIP and the EC10 is 66.7 mg/L.

Toxicity of CIP, MTA-NPs, and HA on infusoria
The CIP that has been evaluated for the current study's toxic effects was toxic to ciliates infusoria P.caudatum in the mortality test; the EC50 value was reached at 1.1 mg/L. The mortality of ciliates proportionally increased by increasing the antibiotic concentration, reaching a peak at 5 mg/L CIP (mortality ~ 80%). A dose of 0.05 mg/L CIP resulted in mortality in less than 20% of ciliates and can be called not harmful (Fig. 2).
Comparison of the toxicity of CIP in relation to ciliates and bacteria showed that ciliates were much more sensitive to CIP than bacteria. This justifies the choice of ciliates as test objects in detoxification experiments. Probably, the artificial modification of the genome of the E. coli bacterial strain by introducing the lux operon of the luminous soil bacteria Photorhabdus luminescens ZM1 leads to its resistance to the antibiotic CIP and possibly to other drugs. Obviously, natural bacterial strains are more suitable for such experimental control of the safety of pollution of natural environments with pharmaceuticals.
The impact of CIP, MTA-NPs, and HA was tracked for 24 h, individually and collectively, on the mortality. The addition of MTA-NPs and HA at the studied concentrations  did not lead to the death of more than 40% of living organisms ( Fig. 3) (p-value for different concentrations less than 0.05). According to numerous published data, iron oxide NPs functionalized with amine-containing silanes exhibit different responses with respect to various test organisms. The mechanism of toxicity of magnetite NPs with APTES can be described with the function of the positive charge of the NH 3 + ions, which presumably interacts with negative charges on the surface of microorganism cells (phospholipids). That leads to a change in membrane permeability and leakage of intracellular components (Bieser and Tiller 2011;Fernandes et al. 2013). Similar effects were also observed by the authors (Hoskins et al. 2012) for iron oxide NPs coated with polyethyleneimine (PEI) in comparison with the same NPs additionally coated with polyethylene glycol (PEG), which affect the negatively charged cell membrane and enhance endocytosis. Primary amines on the surface of NPs create a large positive surface charge (+ 55.6 mV) and, as previously reported, cause a cytotoxic effect (Fernandes et al. 2014).
The combined presence of MTA-NPs and dissolved organic matter in a solution led to the mortality of ciliates by less than 20%, which is lower than for a single MTA-NP and HA (Fig. 3).
This made it possible to choose the following concentrations for a multifactorial experiment to assess the detoxifying ability of silica-coated, magnetically controlled MTA-NPs in the presence of HA. Further work was devoted to the evaluation of the effect of different concentrations of MTA-NPs in combination with different concentrations of HA on the detoxification of CIP using the design of experiments (the response surface methodology, RSM).

Regression models and statistical testing
The results of mathematical and statistical processing by using the surface response methodology made it possible to obtain a second-order equation showing the dependence of the mortality of ciliates on the concentration of nanoparticles and dissolved organic matter: where A and B refer to the concentrations of MTA-NP and HA, respectively. Equation (1) shows how the mortality of ciliates changes when one of the particular factors or their combination changes. The sign in front of the coefficient of the regression equation indicates the nature of the change in the response. A positive value means that the responsethe mortality of ciliates-decreases with an increase in this factor, and a negative value-vice versa, respectively. In this case, both factors are negative. Thus, a decrease in the concentration of nanoparticles and dissolved organic matter leads to an increase in the mortality of ciliates. The value of the coefficient of the regression equation makes it possible to estimate which change in which of the factors leads to the most significant change in the response. So, according to the obtained data, it is the change in the concentration of humic substances that most affects the mortality of ciliates.

Validity of the model
The adequacy of the model describing the effect of nanoparticles and HA as a main part of dissolved organic matter (DOM) concentrations on ciliate mortality was assessed by using the analysis of variance. Multiple regression analysis made it possible to obtain coefficients adequately described by a quadratic model. The results of the analysis of variance for quadratic response surface models are presented in Table 3.
The value of the coefficient of determination (R 2 ) equal to 0.86 indicates that more than 86% of the total data can be explained by a quadratic model and less than 14% cannot be described by the resulting regression analysis model. The adjusted regression values for the obtained data were greater than 0.83, indicating a sufficient correlation between the response and the influencing factor. The predicted R 2 of 0.8046 is in reasonable agreement with the adjusted R 2 of 0.8376, i.e., the difference is less than 0.2. Adequate precision compares the range of the predicted values at the design points to the average prediction error. Ratios that are greater than 4 indicate adequate model discrimination. In this particular case, the value is well above 4. The p-value for the model is less than 0.05, which indicates that the model is significant and therefore has a significant effect on the response under study (ciliates mortality). In the same manner, for the mortality, all factors, their squares, and linear combination (A, B, B 2 , and AB) except for the value of the squared factor A 2 (MTA-NPs concentration) are significant model terms. These insignificant model terms (not counting those that are required to support the hierarchy) can be removed and may result in an improved model. According to the F-criterion of the model equal to 36.06, the model can be called significant.
With a probability of 99.99%, such a large value of the F-criterion cannot arise due to the noise.
The constructed graphs of the predicted, actual, and normal probabilities of student residuals make it possible to assess the eligibility of the developed statistical model. The relationship between the experimental and predicted values of mortality in the presence of MTA-NP and HA is shown in Fig. 4.
According to Fig. 4a, there is a good correlation between the data obtained experimentally and the values predicted by the statistical model for the ciliate mortality values. The  4 Actual and predicted values of response for evaluation of infusoria's (a) and normal probability plot (b) of the studentized residuals normal probability plot can be used as a method for graphical evaluation of the adequacy of residuals, that is, the difference between experimental and model data (Fig. 4b). Figure 4b shows that the residual behavior follows a normal distribution and is linear, which indicates the adequacy of the model; therefore, this model can be used to assess the detoxification of MTA-NPs in the presence of HA.

Effect of variables on responses
Effect of the concentrations of MTA-NPs on the detoxication process Figure 5 represents the dependence of the value of inhibition of ciliates on various concentrations of the MTA-NPs as 0-2 g/L in the absence of HA (HA concentration equals 0 g/L). The absence of the MTA-NPs leads to death of 79.1% on average due to the toxic effect of 5 g/L CIP, according to experimental data, which is in good agreement with the model values of 76.3% and literature data (Dionísio et al. 2019). The addition of the MTA-NPs leads to an almost linear decrease in the inhibition of infusoria from 79.1 to 44.6% with an increase in the MTA-NP concentration from 0 to 2 g/L. These results indicate the sorption of CIP on the MTA-NPs surface and a decrease in the concentration of CIP in the solution after removal of the precipitate. Sorption of CIP is also confirmed by UV-Vis spectrophotometry data, demonstrating the removal of about 50% of CIP with the addition of 1 g/L MTA-NPs.
The mechanism of sorption can be explained by the forces of electrostatic attraction between MTA-NPs and CIP. For the pKa of CIP, the carbonyl group gets deprotonated at 5.9 (pKa1) while the amine group gets protonated at 8.9 (pKa2) (Balarak et al. 2016;Bizi and Bachra 2020) . CIP can thus exist as a cation, anion, and zwitterion (Carabineiro et al. 2011). The position of a functional group of MTA-NPs and CIP can be observed in Fig. 6 in green and red. At the pH close to 7, the MTA-NPs with negative surface charge due to the presence of NH 2 -groups on the surface (zeta potential of about − 25 mV at pH about 7) have a complex interaction by the deprotonated carboxylic groups of CIP in the pKa interval 5.9-8.8 (pKa1 = 5.90 ± 0.15; pKa2 = 8.89 ± 0.11 (Carabineiro et al. 2012) (Fig. 6)).
Previously, Rakshit and colleagues showed an electrostatic pH-dependent sorption mechanism for ciprofloxacin (Rakshit et al. 2013). CIP (pKa1 = 6.1; pKa2 = 8.7) and magnetite (approx. isoelectric point of NPs is 6.5) are positively charged at acidic pH, which limits their interaction. The increase in pH from 4.0 to 6.0 leads to an increase in the proportion of neutral CIP and to the possibility of interaction between CIP and MTA-NPs. Adsorption decreased at pH above 6.5 because CIP and MTA-NPs became increasingly negatively charged with increasing pH.
The EC50 value during detoxification of CIP in the presence of MTA-NPs is 1.7 g/L. The resulting regression (Eq. 1) indicates that an increase in the concentration of MTA-NPs will lead to a decrease in the mortality of protozoa: the addition of MTA-NPs at a concentration of 3.2 g/L will lead to the mortality of ciliates. Figure 7 demonstrates the dependence of the mortality of ciliates on the HA concentration in the absence of MTA-NPs (0 g/L). The graph shows that the dependence is nonlinear: the mortality decreases with an increase in the concentration of HA from 0 (mortality of 63% of ciliates) to 0.029 g/L (mortality of 31% of ciliates). An increase in the concentration of HA leads to an increase in the mortality of ciliates up to 46% at the maximum concentration of HA used at 0.05 g/L. The predicted data shows that a further increase in the concentration of HA to 0.07 g/L will lead to the death of 100% of ciliates.

Effect of HA concentration on the detoxication
Our earlier assessments of the toxicological ability of HA (Bondarenko et al. 2020) showed no toxicity toward ciliates in the concentration range from 0.1 to 100 mg/L. This allows to conclude that an increasing inhibition of test organisms with an increase in HA concentration is not associated with the intrinsic toxicity of the HA but is related to the sorption process. UV-Vis spectrophotometry data show that an increase in the concentration of HA from 0.01 to 0.05 g/L leads to sorption from 90 to 60% of CIP.
The nonlinear dependence of the concentration of adsorbed CIP and the sequence of mortality of ciliates on the concentration of HA may be associated with steric hindrance during sorption as well as with more difficult processes.
The ability to dissolve in alkaline and neutral solutions depends on the chemical composition of HA (Stevenson 1994). Carboxyl and phenolic groups of HA in an alkaline medium are deprotonated, and at the same time negatively charged groups are repelled; due to this fact, HA molecules acquire an elongated shape. With a decrease in pH, phenolic and carboxyl groups are protonated, and the repulsive effect decreases as a result of the molecules acquiring a helical conformation. In this case, the hydrophobic areas are located in the inner part of the structure, and the hydrophilic parts are in contact with the aqueous medium. At the same time, an increase in the concentration of HA can lead to steric hindrances in the sorption of CIP due to the inaccessibility of reactive groups. Fang et al. (Fang et al. 2013) suggested that the HA could act as an electron shuttle during the reaction, promoting CIP degradation due to accelerated electron transfer. For example, the decomposition of CIP occurred predominantly at a lower concentration of dissolved HA. This increased the removal of CIP from the solution. At high HA concentrations, HA reactive centers were depleted and CIP removal dropped to 0.

Effect of joint presence of MTA-NPs and DOM on the detoxication process
The presence of HA as a dissolved organic matter during the sorption of CIP by MTA-NPs has a complex effect.  regardless of the concentration of the MTA-NPs, the survival of ciliates is suppressed by 40% or more; in the range from 20 to 45 mg/L of HA, the inhibition is also controlled by the concentration of the MTA-NPs and reaches 30% or less. As it appears, the highest survival rate of ciliates (more than 70%) is observed at a concentration of HA from 10 to 45 mg/L and at a concentration of MTA-NPs from 1.25 to 2 mg/L. The presence of HA at a concentration of less than 20 mg/L, regardless of the concentration of the MTA-NPs, leads to the mortality of ciliates up to 40% or more, which correlates with data on detoxification in the absence of HA (Fig. 5) or in the absence of MTA-NPs (Fig. 7). This indicates the absence of multiplicative sorption of CIP with the simultaneous presence of HA and MTA-NPs; it can be assumed that CIP competes with HA adsorbed on the MTA-NPs surface by occupying their reactive centers.
On the other hand, the presence of HA at a concentration above 45 mg/L also leads to the death of more than 40% of ciliates, regardless of the concentration of the MTA-NPs, which confirms that the key role of HA is in the detoxification of CIP (which is also indicated by the values of the regression coefficients in Eq. 1). The absence of any effect of NPs on the mortality of ciliates at HA concentrations above 45 mg/L also indicates the likelihood of adsorption of HA on the surface of MTA-NPs with overlapping of its reactive centers. Singh et al. (2009) also studied the sorption of radionuclides on the surface of magnetite in the presence of HA but did not see a significant change in the sorption capacity at a concentration of HA from 2 to 20 mg/L. The only explanation can be the fact that even at the highest HA concentration (100 µ mol/L), HA may be completely sorbed by magnetite at all pH values, as was also observed by Illés and Tombácz 2004. Despite that HA have a negative charge in the entire range of pH (Tombacz et al. 2015) and NPs are negative in the study area (zeta potential of about − 25 mV at pH about 7), it is possible to assume that the MTA-NPs surface is completely covered with HA at their concentration above 45 mg/L. As a result, the magnetically active MTA-NPs in the presence of HA adsorbed on its surface behave in accordance with the nature of the surface they are consistent with. Saei et al. (2017) stated that it is surface characteristics that stand out as one of the most significant determinants of biological performance, as the NP surface is the most prominent and earliest point of exposure (Saei et al. 2017).
However, the combined presence of the MTA-NPs at a concentration of 1.5-2 mg/L and HA at concentrations of 20-45 mg/L has a multiplier effect and allows to lower the mortality rate of ciliates to 30% or more due to the removal of a higher concentration of CIP. Luo et al. (2019) also demonstrated a decrease in the sorption capacity of biochar in relation to CIP in the presence of high concentrations of HA. However, according to the presented data, the sorption of CIP on biochar in the presence of HA decreases from 66.7% at 5 mg/L of HA to 0% at 20 mg/L and remains the same with an increase in the concentration of HA to 50 mg/L. Luo and colleagues explained this is due to the fact that a large amount of HA in solution is adsorbed on biochar, which can lead to blockage of pores or competition with CIP for sorption sites. On the other hand, as previously described, HA can act as an electron shuttle (Fang et al. 2013) and can lead to the decay of the CIP molecule during the electron transfer reaction.

Conclusions
One of the important ecological results of the work was the fact that the combination of MTA-NPs and HA in solution prevented the mortality of ciliates from the presence of CIP, but separately these substances acted differently on ciliates. MTA-NPs are often considered harmless, but the results demonstrate the potential activity of these particles in an aqueous solution. The addition of MTA-NPs to a solution with CIP (5 mg/L; the mortality rate of the ciliates is more than 80%) in a concentration of 3.2 g/L leads to a complete survival of the ciliates. Adding HA to a solution with CIP (5 mg/L, mortality of ciliates of more than 80%) first leads to a decrease in the mortality of ciliates at the concentration of HA of 29 mg/L, and then to an increase in the mortality of the ciliates, reaching a maximum (100%) at 70 mg/L HA. The lowest mortality of infusoria (more than 30%) is observed with the concentration of HA from 10 to 45 mg/L and the concentration of MTA-NPs from 1.25 to 2 mg/L. The infusoria P. caudatum were sensitive enough to CIP. EC50 for infusoria amounted to 1.1 mg/L. This is much less than for the bacteria Photorhabdus luminescens ZM1 EC50 (EC50 = 135 mg/L).
Thus, the performed experiments demonstrated clearly that humic substances that were applied as detoxifying agents toward complex contamination were able to detoxify water polluted with pharmaceuticals and nanomaterials. On the other hand, the detoxifying ability of humic substances was shown to be determined by both their binding ability toward ecotoxicants and their own mitigating activity. Understanding the interaction of humic substances, nanoparticles, and antibiotics may be crucial to fully elucidate the mechanism of transportation of both colloidal nanoparticles and pollutants in the environment.