Introduction

Nowadays, water resources are becoming increasingly scarce and many of them are polluted by anthropogenic sources, such as household wastes and different water-dependent industrial activities. In this regard, it is known that wastewaters from domestic origin contain pathogens, suspended solids, nutrients (nitrogen and phosphorus), and other chemical pollutants, such as heavy metals, detergents, pharmaceuticals, and pesticides, whose toxicity has been widely discussed (Al-Jlil 2009; Rajasulochana and Preethy 2016). On the other hand, wastewaters produced by the textile industry contain large amount of dyes and other chemicals, some of which are carcinogenic non-biodegradable agents that may pose a major threat to health and to the environment (Ghaly et al. 2014 and references therein; Kharat 2015). Despite their composition and complexity, both domestic and textile wastewaters have been recognized as one of the major factors contributing for worldwide aquatic pollution, fundamentally because they contain chemicals with toxicological concern; in addition, these substances may block the sunlight, thereby impairing photosynthesis and increasing the biological oxygen demand in receiving waters, compromising the fundamental reoxygenation process (Kharat 2015; Seow et al. 2016). Therefore, it is of high importance to dispose of these liquid residues in a proper manner with the goal of providing adequate environmental and human health protection. In this sense, the treatment of wastewater before such effluents enter receiving natural water bodies is a critical point (Gogate and Pandit 2004; Seow et al. 2016).

In addition, one of the most relevant problems related to wastewater treatment is the incomplete reduction of the load of all the contaminants present in wastewaters, to attain specific limits that are established by governmental laws (Pawęska and Bawiec 2017). This drawback represents a potential risk to aquatic organisms exposed via the environment, living in receiving water bodies (Hernando et al. 2005). In this sense, phytoremediation has been proposed as an effective, low-cost, preferential cleanup option for moderately contaminated sites (Türker et al. 2014). Plants can directly participate in detoxification processes, through contaminant incorporation and subsequent metabolization, or immobilization within the plant, or indirectly, through the promotion or support of rhizospheric microorganisms that effectively carry out the detoxification process, and one of these processes is the phytofiltration (Ibañez et al. 2018). Particularly, phytofiltration involves the use of aquatic plants, either floating, submerged, or emergent, to remove pollutants from solution, mainly through their root system although in some cases, fronds are also involved directly in the removal process (Olguín and Sánchez-Galván 2012; Paisio et al. 2017). This strategy is of special relevance for the treatment of liquid industrial residues due to their action as “nutrient pumps” (Herath and Vithanage 2015). Taking into account this role of phytoremediation, it is possible to say that the use of aquatic macrophytes (e.g., Lemna spp.) for the complete treatment of wastewater derived from specific treatment plants can be important, particularly during the final tertiary treatment phase, allowing completely purified wastewaters to be released into the environment (Chaudhary and Sharma 2014; Saha et al. 2015). These plants that may be used on phytoremediation processes can be ultimately deposited in landfills, but may also be incinerated; incineration has been proposed as a way of reducing the plant volume and also to generate energy by means of heat (Souza and Silva 2019). In addition, these plants may also be used in the ceramic industry, incorporating the biomass used in the treatment of ceramic blocks, which corresponds to an efficient method of transformation of the obtained products (Lima et al. 2015).

Another topic to consider is that in order to assess the efficiency of an applied wastewater treatment strategy, adequate evaluation tools must be used. Regarding this aim, traditionally the quality of effluents is based on the control of chemical, biochemical, and physical parameters and even more in the detection of specific pollutants (Hernando et al. 2005). However, it is not sufficient to assess the environmental risk because they are not real measurements of the toxicity effects on the aquatic ecosystem (Chang et al. 2009; Rizzo 2011). Thus, with the aim of avoiding this drawback, toxicity tests have been used to evaluate whether effluent detoxification takes place (Klamerth et al. 2010; Lyu et al. 2018). However, it is necessary to take into account that toxicity and chemical measures are complementary analytical tools for monitoring wastewaters quality, which can contribute with reliable indices of the toxic impact of effluents in the aquatic environment. A very popular bioassay used internationally for toxicity screening of chemical compounds, and also for the monitoring of effluents, is undoubtedly the toxicity test with freshwater Daphnids, particularly with Daphnia magna (Persoone et al. 2009).

Therefore, this work evaluated the efficiency of phytoremediation of domestic and textile effluents, collected from wastewater treatment plants from north of Portugal using an aquatic macrophyte, Lemna minor, as an alternative treatment procedure. Then, the toxicity of the treated effluents was analyzed by applying D. magna tests. Additionally, key physiological and biochemical characteristic (catalase activity) of the plants were evaluated in order to evaluate sub-individual responses that may compromise additional biological functions of particular ecological importance.

Materials and methods

This work was divided in three steps: (1) phytoremediation process of effluents (textile and domestic) with L. minor; (2) quantification of toxicological endpoints in L. minor individuals used in the phytoremediation process; (3) toxicity characterization of effluents and phytoremediation process efficiency using D. magna ecotoxicological assays.

Collection and characterization of effluents

Samples of effluents of both domestic and textile origins were collected at effluent treatment plants in the north of Portugal (Guimarães and Vieira do Minho). Both effluents were subjected to a complete (preliminary and secondary) treatment process, after which treated effluents, complying will all regulatory requirements, are released into freshwater streams. Samples were collected directly from the pumping stations of both plants, properly accommodated in 30-L plastic jars, and transferred to the laboratory where they were evaluated in terms of some characteristics, such as chemical oxygen demand (COD) and pH. After the determination of the following measurements, samples were accommodated at 4 °C until further processing.

Chemical oxygen demand (COD) determination

The organic matter content was determined by measuring COD. For this purpose, a HACH commercial kit (2125915 COD HR) was used, following the method 410.4 of US EPA (EPA 1993). This method is based on the ability of the oxidizable organic compounds to reduce the dichromate ion to chromium ion, which was quantified according to its absorbance at 610 nm. An HANNA multiparametric spectrophotometer (model HI 83214) was used to determine absorbance values, which correspond directly to COD concentrations (mg/L).

Phytoremediation process

The effluents were subjected to phytoremediation by L. minor in order to provide additional treatment (tertiary), to optimize the decontamination initially generated in the corresponding treatment plants. L. minor was cultivated according to the procedures described in Alkimin et al. (2019) and the phytoremediation process was conducted in glass flasks (n = 8), with 1 L of effluent, in which L. minor were inoculated so as to cover 70% of the surface. The process was conducted during 5 days in controlled conditions (23 ± 1 °C; 16hL:8D light intensity 5500 lx) and the same conditions were adopted for both effluents. For comparison of results, a similar exposure was conducted with L. minor cultured in modified Steinberg medium (phytoremediation control—PC) (OECD 2006). At the end of the exposure period, L. minor individuals were removed, and specific endpoints were analyzed to assess the effects of both effluents on L. minor individuals, namely those that were used during the phytoremediation process. For the determination of pigment levels and catalase activity (as described below), the plants were stored in Eppendorf® microtubes at − 80 °C until the analysis. The resulting effluents were used in ecotoxicity tests with D. magna to assess their ecotoxicological profile, after being subjected to the phytoremediation process.

Lemna minor analyses

Morphological parameter

Root length was used as morphological parameter. This parameter was measured in at least 3 plants per replicate (n = 8), and for all conditions (PC, and RTE and RDE after phytoremediation process). In addition, this same parameter was also analyzed in non-exposed plants from the stock culture (pre-treatment—PE).

Pigment content determination

Chlorophyll a (Chl a) and b (Chl b), carotenoids (Car), and anthocyanins were extracted using approximately 25 mg of L. minor tissue. Biomass was macerated in 1.5 mL of acetone:water (9:1 v/v) for Chlor a and b and Car extraction, and in 1.5 mL of methanol 1%:HCl:water (90:1:1) for anthocyanin (AA) extraction (Sims and Gamon 2002). After this procedure, samples were centrifuged (Eppendorf 5810 Refrigerated Centrifuge) during 5 min at 4 °C and 15,000×g, and absorbance readings (470, 529, 537, 647, 650, and 663 nm) were performed in 96-well plates, in microplate reader Thermo Scientific, model Multiskan GO, version 1.00.40, with SkanIt Software 3.2. The calculation was performed according the following Eqs. 1, 2, 3 (Sims and Gamon 2002), and 4 (Wellburn and Lichtenthaler 1984). The results were expressed in milligrams per gram of fresh weight (mg/g FW).

$$ \mathrm{Chlor}\ \mathrm{a}=0.01373\times {\mathrm{A}}_{663}-0.000897\times {\mathrm{A}}_{537}-0.003046\times {\mathrm{A}}_{647} $$
(1)
$$ \mathrm{Chlor}\ \mathrm{b}=0.02405\times {\mathrm{A}}_{647}-0.004305\times {\mathrm{A}}_{537}-0.005507\times {\mathrm{A}}_{663} $$
(2)
$$ \mathrm{AA}={\mathrm{A}}_{529}-\left(0.28{8\mathrm{A}}_{650}\right) $$
(3)
$$ \mathrm{Car}=1000\times {\mathrm{A}}_{470}-2.27\times \mathrm{Chl}\ \mathrm{a}-81.4\times \mathrm{Chl}\ \mathrm{b}/227 $$
(4)

Catalase determination

Catalase (CAT) was chosen as biochemical marker since this enzyme is an important option to prevent the occurrence of oxidative damage (through reactive oxygen species accumulation) that can compromise the plant physiology, by causing severe oxidative damage, thus inhibiting growth and grain yield (Caverzan et al. 2016). CAT activity was determined in a 96-well microplate. Spectrophotometric readings were performed in the previously described microplate reader. CAT activity was assayed by the procedure described by Aebi (1984). This activity was quantified based on the degradation rate of the substrate H2O2, monitored at 240 nm for 5 min. The results were expressed by considering that 1 U of CAT activity equals the number of moles of H2O2 degraded per minute per milligram of protein. Protein quantification was performed at 595 nm using the Bradford method (Bradford, 1976), adapted to microplate with bovine γ-globulin as standard, in order to express enzymatic activities per milligram of protein on the analyzed samples.

Ecotoxicological tests—D. magna maintenance and exposure

To analyze the efficiency of phytoremediation of L. minor, ecotoxicological tests were performed using the macroinvertebrate D. magna (clone K6) that is routinely kept at the Center of Environmental and Marine Studies at University of Aveiro, Aveiro, Portugal and cultivated according Daniel et al. (2018) and ASTM (2014). Tests were carried out using the raw effluents and effluents resulting from phytoremediation treatment.

Daphnia magna acute test

Independent experiments were used to assess the acute toxicity of the effluents to D. magna. The tests were performed in agreement with a standard protocol (OECD 2004), under the same laboratory conditions described above. The acute assays were conducted in glass flasks containing 50 mL of test solution and 50 mL of clean ASTM medium in the control. A total of 25 animals (< 24 h old) were divided into five groups of five animals per treatment concentration. For acute tests, a range of dilutions between 1% and 100% (being 100% pure effluent and 1% the highest dilution) of both effluents were used to calculate EC50; these tests were performed in triplicate. After EC50 calculation, the obtained values also were transformed in acute toxic units (TUa = 100/EC50) following the specifications: < 1, low toxic; > 1 to < 4, toxic; > 4, very toxic (Verma 2008).

Subchronic reproduction test

Subchronic reproduction test was conducted with D. magna, based on the OECD guideline 211 (OECD 2012); the exposure period was modified according to the studies of Ribeiro et al. (2011) with D. magna, and of Lameira (2008) and Vacchi et al. (2016) with Daphnia similis. The test was ended when the exposed daphnids were in their fourth brood, approximately 17 days of exposure. This method measured the subchronic toxicity in terms of reproduction, using ten neonates (<24 h old) for each effluent concentration. Exposure media were totally renewed every other day. This test used sublethal dilutions of each effluent, calculated from data obtained in acute tests: 25%, 50%, and 100% of raw (RDE) and treated (TDE) domestic effluent; 5%, 10%, and 15% of raw textile effluent (RTE); and 25%, 50%, and 100% of treated textile effluent (TTE).

Swimming behavior test

The behavior assay was performed with 16 animals, exposed to sublethal concentrations of both effluents, with and without phytoremediation treatment (same as described above) plus control (clean ASTM), distributed in six-well plates (one organism per well) with 10 mL of solution per well. To develop this test, animals < 24 h old were used and exposure was of 96 h duration, with behavioral analysis being made at 48 and 96 h after the onset of exposure in order to evaluate the effluents’ effects through the period of exposure. Medium was renewed every other day. Immediately before each behavioral analysis, the animals were transferred to 24-well plates with 2 mL of the respective contaminated solution. Each plate was composed by 16 treated (exposed) animals, plus 4 control animals and 4 animals subjected to the phytoremediation control (PC). The behavior analysis started after 10 min of adaptation to new conditions. Movement of the animals was tracked using the Zebrabox (Viewpoint, Lyon, France) tracking system using a 25-frame-per-second infrared camera over a period of 20 min. Movement was stimulated by alternating light and dark periods during two cycles with 5L:5D min (300L:300D s) each cycle (Dionísio et al. 2020). Typically, D. magna individuals are less active during dark periods and move more during light phases. Data outputs were obtained at each 5 min and the following parameters were calculated: total distance moved (TD) and total swimming time (TT).

Statistical analysis

To calculate the EC50 value, we used the software Probit in IBM SPSS Software version 25, and a one-way analysis of variance (ANOVA, with appropriate post hoc test) was conducted for the other analysis. The type of ANOVA (parametric or non-parametric) and post hoc test (Dunnett’s, Dunn’s, or Tukey test) was chosen depending on whether normality and homogeneity of data were demonstrated by analysis of the residuals with the Shapiro–Wilks test. Test statistics (ANOVA) and analysis of normality were conducted using the software SigmaPlot version 12.5 and a significance level of 0.05 was adopted.

Results

Physical–chemical parameters and L. minor toxicity

Chemical oxygen demand (COD—Fig. 1) and pH were the physical–chemical parameters analyzed; COD values in the controls were close to 20 mg/L, while raw effluents of both types (domestic and textile) had higher levels of this parameter. On the other hand, COD in effluents after the phytoremediation process returned to basal values, close to control. The other parameter was pH, for control treatment this value was 5.53 (± 0.1); for RDE, TDE, RTE, and TTE, the values were 7.79 (± 0.1), 7.55 (± 0.15), 7.96 (± 0.12), and 7.77 (± 0.17), respectively; and in general, control had lower pH values than treated and non-treated effluents. Despite not having being quantified, our observations lead us to conclude that the here-proposed treatment was responsible for a significant reduction of the effluent’s color intensity, that was not quantified.

Fig. 1
figure 1

Chemical oxygen demand (COD) of domestic and textile effluents before and after the phytoremediation process. C: control RDE: raw domestic effluent, TDE: treated domestic effluent, RTE: raw textile effluent; TTE: treated textile effluent. Different letters represent significant differences between treatments (p < 0.05). For each parameter, mean and standard error are shown

Full size image

Root length (Fig. 2) was the vegetal morphological parameter analyzed; exposure to both effluents caused a significant increase (p < 0.05) of this parameter in exposed plants, when compared with pre-treatment (PE) plants and subjected to control.

Fig. 2
figure 2

Root length of L. minor plants exposed to domestic and textile effluents and to uncontaminated control (Steinberg’s solution), during 5 days. PE: pre-treatment, C: control, RDE: raw domestic effluent, RTE: raw textile effluent. Equal symbols represent significant differences between treatments (p < 0.05). For each parameter, mean and standard error are shown

Full size image

Pigments contents (depicted in Fig. 3), such as Chlor a and Chlor b levels, were not affected in L. minor exposed to RDE. On the other hand, when plants were exposed to RTE, the contents of these pigments significantly decreased (p < 0.05). Similar results were obtained for the pigment carotenoid. However, AA levels showed different patterns: in RDE-exposed plants, this pigment was significantly increased (p < 0.05), while in RTE-treated plants, exposure led to a decrease in their content (p < 0.05).

Fig. 3
figure 3

Pigments of L. minor plants exposed to domestic and textile effluents and to uncontaminated control (Steinberg’s solution), during 5 days. C: control, RDE: raw domestic effluent, TE: raw textile effluent. “*” represents significant differences between treatments (p < 0.05). For each parameter, mean and standard error are shown

Full size image

The oxidative stress responses on L. minor plants used in the phytoremediation process were measured by quantification of CAT activity (Fig. 4). In plants exposed to the TE, the activity of this enzyme was significantly increased (p < 0.05).

Fig. 4
figure 4

Catalase activity of L. minor plants exposed to domestic and textile effluents and to uncontaminated control (Steinberg’s solution), during 5 days. C: control, RDE: raw domestic effluent, RTE: raw textile effluent. *Significant differences between control group (p < 0.05). For each parameter, mean and standard error are shown

Full size image

Ecotoxicological tests—D. magna

The tests with D. magna started with the determination of the effective concentration value. This determination was impossible for RDE since the effluent was not capable of causing effect (immobilization) even before treatment. Consequently, and even when submitted to the treatment by L. minor, this effluent was not capable of causing immobilization to juveniles of D. magna. However, it was possible to calculate an EC50 value for the RTE, which was found to be between 53.82% and 66.89% of effluent, and with a TUa calculated between 1.85 and 1.49, allowing to classify the effluent as “toxic.” This effluent, after being treated by L. minor, was not causative of any mortality. Respect to control, it did not exert acute toxicity on D. magna.

The number of neonates from the first and the fourth broods (Table 1) were not affected by effluent exposure (p > 0.05); however, the total neonate number suffered an increase when exposed to RTE 10% and 15%, and to TTE 50% and 100% (p < 0.05), showing a lower toxicity of TTE respect to RTE. The age of release of the first brood was only significantly changed after the exposure to a TDE 25% (p < 0.05).

Table 1 Results obtained in the D. magna subchronic reproduction assay exposed to raw and treated domestic and textile effluents
Full size table

After 48 h (Fig. 5) of exposure to RDE, D. magna individuals showed some differences in their swimming behavior: TT was the most affected parameter, with significant differences (p < 0.05) being reported in two light and dark cycles. On the other hand, TD was only significantly different (p < 0.05) in comparison with the control treatment, in organisms exposed to the concentration of 25%, in the first light/dark cycle, with a significant decrease (p < 0.05) in the swimming distance. After 96 h of exposure (Fig. 6), the TT followed the same pattern, with significant differences (p < 0.05) being registered for all cycles; TD was affected in the first light cycle, while in the second dark cycle, the swimming distance was significantly impaired.

Fig. 5
figure 5

Effects of raw domestic effluent (RDE—48 h of exposure) on D. magna locomotion in dark and light period. Values are mean values ± standard error. 300 and 900 s corresponding to light periods, 600 and 1200 s corresponding to dark periods. C: control, PC: phytoremediation control, TT: total swimming time, TD: total distance. “*” significant differences between treatment and control (p < 0.05). For each parameter, mean and standard error are shown

Full size image
Fig. 6
figure 6

Effects of raw domestic effluent (RDE—96 h of exposure) on D. magna locomotion in dark and light periods. Values are mean values ± standard error. 300 and 900 s corresponding to light periods, 600 and 1200 s corresponding to dark periods. C: control, PC: phytoremediation control, TT: total swimming time, TD: total distance. “*” significant differences between treatment and control (p < 0.05). For each parameter, mean and standard error are shown

Full size image

Exposure for 48 h to TDE (Fig. 7) was not generally causative of significant changes (p > 0.05) in terms of light/dark cycles; however, animals exposed to PC during the first light cycle had their TT decreased, and their TD increased after the second light cycle (p < 0.05). However, 96 h (Fig. 8) of exposure to this effluent caused a decrease (p < 0.05) in TT on both light cycles and also following the second dark cycle. TD was significantly impaired (p < 0.05) after light cycles in animals exposed to PC; TD was also impacted (p < 0.05) after the dark cycle in daphnids exposed to PC and TDE.

Fig. 7
figure 7

Effects of treated domestic effluent (TDE—48 h of exposure) on D. magna locomotion in dark and light periods. Values are mean values ± standard error. 300 and 900 s corresponding to light periods, 600 and 1200 s corresponding to dark periods. C: control, PC: phytoremediation control, TT: total swimming time, TD: total distance. *Significant differences between treatment and control (p < 0.05). For each parameter, mean and standard error are shown

Full size image
Fig. 8
figure 8

Effects of treated domestic effluent (TDE—96 h of exposure) on D. magna locomotion in dark and light periods. Values are mean values ± standard error. 300 and 900 s corresponding to light periods, 600 and 1200 s corresponding to dark periods. C: control, PC: phytoremediation control, TT: total swimming time, TD: total distance. *Significant differences between treatment and control (p < 0.05). For each parameter, mean and standard error are shown

Full size image

In general, 48 h of exposure to RTE (Fig. 9) caused a decrease (p < 0.05) in TD in all cycles, being significant in dark cycles. TT was not so affected, showing significant reductions in PC-exposed animals in first light cycle and in organisms exposed to 15% (p < 0.05). After 96 h (Fig. 10), TT was only significantly inhibited in PC-exposed organisms (p < 0.05) in the first light cycle; TD was not compromised (p > 0.05).

Fig. 9
figure 9

Effects of raw textile effluent (RTE—48 h of exposure) on D. magna locomotion in dark and light periods. Values are mean values ± standard error. 300 and 900 s corresponding to light periods, 600 and 1200 s corresponding to dark periods. C: control, PC: phytoremediation control, TT: total swimming time, TD: total distance. *Significant differences between treatment and control (p < 0.05). For each parameter, mean and standard error are shown

Full size image
Fig. 10
figure 10

Effects of raw textile effluent (RTE—96 h of exposure) on D. magna locomotion in dark and light periods. Values are mean values ± standard error. 300 and 900 s corresponding to light periods, 600 and 1200 s corresponding to dark periods. C: control, PC: phytoremediation control, TT: total swimming time, TD: total distance. *Significant differences between treatment and control (p < 0.05). For each parameter, mean and standard error are shown

Full size image

In general, after 48 h of exposure to TTE (Fig. 11), daphnids did not suffer behavior alterations in terms of TT (p > 0.05), with the exception of PC-exposed organisms in light cycles, a parameter that was significantly decreased (p < 0.05); light cycles too provoked a significant decrease (p < 0.05) in TD-exposed daphnids. On the other hand, after 96 h of exposure (Fig. 12), TT was significantly affected (p < 0.05), with a decrease in this parameter in the first light cycle and in both dark cycles when animals were exposed to PC. In addition, TTE exposure increased swimming distance (p < 0.05) in animals exposed to 25% in the first light cycle. The same parameter was affected by PC and 100% exposition (p < 0.05) decreasing TD in both dark cycles and just in the second dark cycle, respectively.

Fig. 11
figure 11

Effects of treated textile effluent (TTE—48 h of exposure) on D. magna locomotion in dark and light periods. Values are mean values ± standard error. 300 and 900 s corresponding to light periods, 600 and 1200 s corresponding to dark periods. C: control, PC: phytoremediation control, TT: total swimming time, TD: total distance

Full size image
Fig. 12
figure 12

Effects of treated textile effluent (TTE—96 h of exposure) on D. magna locomotion in dark and light periods. Values are mean values ± standard error. 300 and 900 s corresponding to light periods, 600 and 1200 s corresponding to dark periods. C: control, PC: phytoremediation control, TT: total swimming time, TD: total distance. *Significant differences between treatment and control (p < 0.05). For each parameter, mean and standard error are shown

Full size image

Discussion

L. minor phytoremediation-based process

Results from the phytoremediation process using L. minor showed to be efficient, by decreasing the COD in both effluents. COD is one of the most widely used parameters indicating organic pollution applied to both wastewater and surface water, and it is defined as the amount of oxygen required to oxidize the organic matter present in wastewater (Arrojo 2006). The higher the COD, the higher the amount of organic pollution in the water sample. Thus, COD is considered one of the most important quality control parameters of an effluent in wastewater treatment facilities (Wu et al. 2011). In other words, L. minor was capable of significantly decreasing the amount of organic compounds present in both effluents. This decrease was extremely important since the post-treatment effluents presented COD values of the same order of magnitude of those presented by the control treatment (Steinberg’s media). Moreover, both effluents prior to the phytoremediation process also exhibited low COD values and therefore they would not require additional treatment. From this point of view, it is important to highlight that the treatment by L. minor reduced these values even further, which is a positive aspect of phytoremediation since these treated effluents would produce a lesser impact on the environment after being discharged, in relation to their untreated form. Organic wastes mineralize after being discharged into the receiving water bodies and the resulting nutritive elements stimulate plant production, leading to eutrophication (Kanu and Achi 2011). In this situation, the biomass increases considerably and surpasses the assimilation limit by herbivores. The excessive production of organic matter leads to the buildup of “sludge” and the mineralization process consumes all dissolved oxygen from the water column, which is responsible for fish mortality (Kanu and Achi 2011). Thus, L. minor could be considered an efficient phytoremediator plant for DE and TE treatment, being potentially used to perform the tertiary treatment of these effluents in order to ensure their final purification, as it was demonstrated by the COD reduction achieved in this work.

On the other hand, it is important to note that the mentioned reduction in COD values by the treatment by L. minor was higher for RTE, reaching removal efficiencies around 73%, while for RDE this reduction was of 48%. These variations can be due to the nature of the organic compounds contained in each effluent, which is a decisive factor for their degradability. Thus, more easily degradable organic compounds could be present in higher amounts in textile compared to domestic effluents. Moreover, it is important to mention that in textile samples, a coloration was detected, which was reduced during the here-tested phytoremediation process. In this regard, it is not negligible the data showing that L. minor was able to discolor them since it is known that dyes are complex and environmentally troublesome contaminants. The toxic effects of dyes towards the aquatic environment result from their long persistence in the environment, accumulation in sediments and in biota, decomposition of pollutants into carcinogenic or mutagenic compounds, and also low aerobic biodegradability (Samchetshabam et al. 2017). The here-obtained results are in line with previously published data. Chaudhary and Sharma (2014) emphasized that duckweeds are functionally simple, yet easy to maintain, and they can provide a highly effective tertiary treatment, with a performance that is equal or even superior to conventional wastewater treatment systems now recommended for large-scale operations. This efficiency was again demonstrated by Sivakumar (2014) who found a decrease in COD and in color of industrial effluents suggesting the possibility of using L. minor for the phytoremediation of various parameters in selected textile industry effluents and in any type of textile industry effluents. In addition, Patel and Kanungo (2010) found a high nutrient removal rate by L. minor–based phytoremediation of DEs. These works confirm the possibility of a wide use and improved efficiency of L. minor in effluent phytoremediation processes, which are in agreement with our study. Finally, the here measured pH values were decreased after the phytoremediation process, and both effluents became more neutral than alkaline. This pH reduction might be due to microbial action under anaerobic conditions. During microbial respiration, organic matter decomposes and releases CO2 which may be responsible for decreasing the pH (Mahmood et al. 2005). These results are in line with other phytoremediation studies that used macrophyte species (such as Eichhornia crassipes and Chara vulgaris) to this purpose (Saha et al. 2017; Mahajan et al. 2019).

Toxic effects on L. minor

Chemical environmental stress may occur due to the presence and exposure of biota to chemicals whose metabolism leads to the formation of highly unstable oxygen derivatives, known as reactive oxygen species (ROS). Their accumulation in the cells of plants may cause severe oxidative damage, thus inhibiting growth and grain yield (Caverzan et al. 2016), and to protect themselves from the deleterious effects of ROS, plants express antioxidant mechanisms (Racchi 2013). One option to prevent the occurrence of oxidative damage is through enhancement of CAT activity (Racchi 2013), an enzyme which very efficiently promotes the conversion of hydrogen peroxide (resulting from the dismutation of the superoxide anion by superoxide dismutase) to water and molecular oxygen (Valko et al. 2006). In this study, exposure to RDE was not able to cause alterations in CAT activity; on the other hand, RTE exposure increased the activity of this enzyme, suggesting the establishment of a pro-oxidative scenario. These patterns were also obtained in previous works. Singh et al. (2008) reported a CAT activity increase in L. minor after being exposed for 7 days to different concentrations of industrial effluents with high amount of metals. Basiglini et al. (2018) did not find CAT activity alterations after effluent exposition, but reported significant changes in another antioxidant enzyme responsible for hydrogen peroxide degradation in plants, namely, ascorbate peroxidase (APX). Radić et al. (2010), after exposing L. minor to an industrial effluent for 7 days, observed a significant increase in peroxidase activity (POD), an enzymatic form also involved in the antioxidant defense mechanism of plants. In general, it is possible to suggest that the activation of antioxidant defense system linked to oxidative stress in plants occurs after exposure to industrial effluents; however, the pattern of response of the species depends on its level of tolerance, plant growth stage, and concentration of contaminants (Gill and Tuteja 2010).

The antioxidant activity of carotenoids arises primarily as a consequence of the ability of the conjugated double-bonded structure to delocalize unpaired electrons (Mortensen et al. 2001). This is primarily responsible for the excellent ability of β-carotene to physically quench singlet oxygen without degradation and for the chemical reactivity of β-carotene with free radicals (Valko et al. 2006). The here-obtained results showed that DE was not capable to cause alterations in Car content, but RTE exposure significantly decreased this pigment content. This result is in agreement with the previously discussed data concerning CAT activity, for which RDE was not capable to cause alterations in this enzyme’s activity, while RTE exposure was responsible for its decrease, which may suggest a mechanism defense against ROS increase. The excess of released ROS that occurs in response to the metabolism of chemicals found in the analyzed effluents is capable of causing the oxidation of Car, thereby decreasing their overall content in plants exposed to this effluent. Decrease in Car content was also reported by Brkanac et al. (2010) after analyzing a long-term exposure of L. minor to different surface water samples contaminated with effluent discharge (namely municipal and industrial effluent from Croatia). Considering the here-obtained data, and from the literature, it is possible to conclude that this pigment may be considered a sensible parameter to assess abiotic environmental stress.

Anthocyanin is a secondary metabolite that can be produced in response to oxidative stress, performing important protection roles (Juszczuk et al. 2004), such as a scavenger of a wide array of reactive oxygen and nitrogen species (Dauphinee et al. 2017) and other mechanisms of defense (sunscreen and metal chelating) (Landi et al. 2015). An increase in its content might indicate the occurrence of oxidative stress; on the other hand, a decreasing trend of its levels means a failure of the (antioxidant) defense system (Miguel 2011). In this study, L. minor exposed to RDE had an increase in AA content indicating the occurrence of pro-oxidative alterations that were, however, effectively counteracted by this defensive mechanism. On the contrary, plants exposed to RTE had decreased levels of these pigments suggesting that the oxidant process was capable of significantly oxidizing these compounds. This set of results regarding AA levels suggests a potential failure on the antioxidant defense system. Since the Car content in these RTE-exposed plants was not affected, alternative mechanisms of defense (e.g., CAT) must have been activated—indeed, this parameter was significantly increased following exposure to this effluent. Normally, AA is not a pigment commonly analyzed to determine effluent toxicity, but this study demonstrated its usefulness and sensitivity to address the toxicity of both domestic and/or industrial effluents.

However, under oxidative stress, if it is mild or intermediate, organisms usually block general programs of their life cycle (such as reproduction or extensive biosynthesis), to develop responses to prevent or neutralize negative ROS effects (Lushchak 2014). Photosynthetic pigments (namely chlorophylls) are included in the group of physiologically important substances whose biosynthesis may be compromised by chemical insult, as demonstrated by some studies addressing the effects of different classes of chemical contaminants. Roy et al. (2015), after exposing L. minor for 96 h to 10–25% of tannery industrial effluent, found a significant decrease in Chl content. Tkalec et al. (2008) obtained similar results with the same species after 6 and 12 days of cadmium exposure, and a decrease in L. minor Chls was reported by Fekete-Kertész et al. (2015) after exposure to chemicals such as benzophenone and bisphenol A during 7 days. This inhibitory effect in terms of Chlor levels was reported after exposing L. minor to RTE, but not to RDE; a failure of the antioxidant defense mechanisms was closely followed by significant decreases in the contents of both Chlor a and b. On the other hand, and despite oxidative damages, the exposure to effluents caused an increase in root size, which may be associated with a large amount of nutrients, normally present in effluents, that favored their own uptake and enhanced the plant’s metabolism and growth, characteristics described in the literature for some plant genera/species, including Lemna spp. (Cedergreen and Madsen 2002). This response has already been reported after exposure to industrial effluents (Patel and Kanungo 2010). The results about the increase of the root size in plants exposed to the effluents may be in conflict with the decrease of pigments levels, an effect that was observed in these plants as well, as shown above. However, stressed plants (as those analyzed in this study, which were subjected to stressful biotic and/or abiotic factors) are capable to allocate energy to different physiological systems according to their necessities (Nguyen et al. 2016; Mundim and Pringle 2018). This can be an explanation for our findings since plants spent more energy in favoring their root growth to enhance nutrient absorption, in relation to the energy invested to produce pigments. This assumption is explained by Bazzaz et al. (1987) since, at an ecological level, energy budget allocation includes the relationship between investment in one function and investment in others, such as the relationship between defense and growth; at a physiological level, energy allocation entails the partitioning of resources within the plant and the consequences of this partitioning to favor gain or loss of critical resources.

Toxic effects on D. magna

The effect of dilutions of the effluent (raw and post-treatment) on daphnids was examined with acute (mortality and behavior) and sub-chronic (reproduction) toxicity tests. Acute test data showed that RDE and TDE had no significant acute toxicity. Considering these data, it is possible to suggest two possibilities: (1) the treatment applied to wastewater by plants (namely L. minor) is effective enough to remove most toxic compounds, or (2) this specific effluent that was here-tested did not receive a considerable load of toxic compounds. This is not in agreement with the literature since in most cases, RDE are toxic to aquatic organisms, such as Selenastrum capricornutum (Ra et al. 2007), D. magna (Ra et al. 2007; Gholami-Borujenia et al. 2018), and Danio rerio (Zhang et al. 2012).

The toxicity of RTE towards D. magna was possible to be converted into an EC50 value and into TUa, showing the evident toxicity of this effluent. This was a somehow expected result, considering its great complexity, as mentioned previously. In addition, these results are in agreement with those from the literature (Karthikeyan and Meyer 2006; Verma 2008; Gebrati et al. 2011), which show the correlation between EC50 and TUa with industrial effluent toxicity. On the other hand, after the phytoremediation process, these values were no longer possible to be calculated since TTE did not show any measurable acute toxicity, thus indicating the efficiency of the phytoremediation process with L. minor to treat this effluent. The increased efficacy of the treatment process, yielded a significant improvement in the quality of the effluent, which certainly can be attributed to the capacity of L. minor to absorb substances from the medium (Ugya 2015) and to its capacity to decrease the levels of those chemicals that were initially responsible for the higher COD values that were measured in this effluent.

These results are in agreement with the results of the reproduction test. D. magna is well known for its modulation of reproductive features, showing different reproduction strategies according to the environmental conditions in which they are kept, food amount and sources, presence of toxic substances, and other factors, such as genetic variability (Ebert 1993; Enserink et al. 1993; Viganò 1993). In this study, only the RTE effluent (raw and treated) caused alterations in the number of neonates. Part of this result was already expected since it was not possible to calculate values of EC50 and TUa for RDE. These results may, however, be indicative, but not a statement, of the absence of toxicity by this effluent. On the contrary, such parameters were indeed calculated for TE, indicating its likely toxicity. Furthermore, the composition of the here-tested effluents may be a factor influencing these results since as described in other section of this discussion, in general, RDE had a potentially higher concentration of organic compounds (such as nutrients), while RTE is most likely a complex mixture with different organic and inorganic compounds, including metals, dyes, and detergents that are toxic to the here-used model organism, D. magna (Westlake et al. 1983; Rodriguez et al. 2006; Gholami-Borujenia et al. 2018).

The here-tested phytoremediation process significantly decreased COD in RTE, but as a complex mixture, it may have a high variety and concentration of compounds, such as dyestuffs, salts, acids, bases, surfactants, dispersants, humectants, oxidants, and detergents, which render these waters esthetically unacceptable and unusable (Khandare and Govindwar 2015). However, these substances cannot be completely removed by phytoremediation processes (Bokhari et al. 2016; Khataee et al. 2012; Mkandawire and Dudel 2007) and may subsist in the treated effluent even after the treatment; in this case, it is plausible to find a mixture of chemicals that includes organics, metals, and azo dyes. Thus, reproduction alterations provoked by TTE on D. magna can be associated to these compounds. Similar results have been already demonstrated in previous studies from the literature. Flohr et al. (2012) obtained similar results (reproduction increase) after exposing D. magna for 21 days to the soluble fraction of different industrial wastes. This result is not entirely surprising since an increase in reproduction after exposure to an environmental contaminant may suggest a forced attempt of assuring the conservation of the species when challenged (Terra et al. 2008).

In relation to D. magna behavior, in general, effluent exposures (RDE, TDE, RTE, and TTE) reduced swim distance of the test organisms. Behavioral modifications are normally associated with neural disorders (Tierney 2011). A number of substances potentially present in DE and in TE are capable of causing changes in behavior of aquatic organisms, such as metals (Semsari and Megateli 2010), pharmaceuticals (Rivetti et al. 2016), and azo dyes (Barot and Bahadur 2015). Thus, behavioral traits may be considered highly integrative parameters that react to the presence of a large array of environmental chemicals, especially those that are part of a complex matrix such as treated effluents.

Conclusion

Both effluents were capable of altering several key parameters of L. minor, namely morphological (root length) and physiological/biochemical (pigment content and catalase activity) endpoints. Despite these findings, the here used plant species showed to be highly promising in terms of its effectiveness to be used in tertiary treatment (phytoremediation) of contaminated effluents. This efficacy was reflected by a significant decrease of COD (in both effluents) and, for RTE, a noteworthy decrease of the acute toxicity parameters measured in D. magna. In addition, RTE (raw and treated) was shown to have higher toxicity since it caused reproduction alterations in D. magna, while RDE did not. Lastly, behavior alterations in D. magna were elicited by both effluents pre- and post-treatment, suggesting the existence of neurotoxic chemicals in the effluents, even after treatment.