The mean EC50 (48h) value for Ceriodaphnia silvestrii exposed to caffeine was 45.94 mg L-1. Studies by Di Lorenzo et al. (2019) and Bang et al. (2015) recorded EC50 values (48h) of 395 mg L-1 and 445.3 mg L-1 for D. magna, respectively, while Li (2013) recorded a LC50 (96h) of 377.6 mg L-1 for Dugesia japonica. Comparing these data with the EC50 (48h) obtained in the present study, C. silvestrii is among these species the most sensitive to caffeine, with regard to acute toxicity.
The results of the chronic toxicity tests showed no impairment on the survival and reproduction of C. silvestrii under tested conditions. However, previous studies have shown adverse effects to other aquatic biota organisms, such as, a recent study demonstrated negative effects on zebra fish (Danio rerio) after gestational exposure to caffeine, such as significant damage to ciliated cells (97.0 mg L-1) and toxicity in the early stages of development (48.5 mg L-1 and 97.0 mg L-1) (Rah et al. 2017). Rodriguez et al. (2014) also reported chronic toxicity of caffeine in D. rerio embryos. Embryos exposed to 970.95 mg L-1 developed tremor, uncoordinated movement and severe body curvature within 6 hours of exposure to caffeine, while longer exposure caused irreversible body curvature and lethality. Similarly, Lee and Wang (2015) showed that caffeine affected the development of fish embryos of the species Oryzias latipes at concentrations of 9.7 mg L-1 and 97.0 mg L-1. Furthermore, as demonstrated by Di Lorenzo et al. (2019), caffeine represented an environmental risk in all water bodies investigated in Spain. These data suggest the potential environmental risks associated with the presence of caffeine in the aquatic environment.
In our study, caffeine was the only drug that reached a moderate level of environmental risk, due to the greater MEC (357.0 µg L-1) found in the literature, in the studies of Ferreira (2005). Although for most available data the risk was classified as low and insignificant.
Ceriodaphnia silvestrii was more sensitive to diclofenac sodium salt in the acute test (EC50, 48h = 14.59 mg L-1) in comparison to the other drugs tested. Cleuvers (2003) and Cleuvers (2004) also showed that the cladoceran Daphnia magna, the algae Desmodesmus subspicatus and the macrophyte Lemna minor were more sensitive to diclofenac than to other anti-inflammatories, such as acetylsalicylic acid, ibuprofen and naproxene, in the evaluation of acute toxicity. Therefore, ours results support findings by others studies, showing that diclofenac is probably the most acutely toxic compound within the NSAID class (Fent et al. 2006).
Chronic toxicity results showed no implications on the survival and reproduction of C. silvestrii exposed to sublethal diclofenac sodium salt concentrations considering an exposure period of eight days. This was also observed by Oliveira et al. (2018) for the same species at concentrations of 0.0625 to 2 mg L-1 and same exposure time. The same response was also observed for D. magna at 21 days of exposure at concentrations of 29.5 to 72 mg L-1 (Oliveira et al. 2015a). However, the exposure of D. magna to lower concentrations of diclofenac (0.0005 to 7.2 mg L-1) over a period of 48 hours caused adverse effects on neuron regulation biomarkers such as total cholinesterases (ChEs) by a significant decrease of ChE activity in concentrations of 0.0005 mg L-1 and 7.2 mg L-1, and on the defense of the enzymatic oxidative stress, e.g., by a decrease in the activity of glutathione peroxidase selenium-dependent (Se-GPx) in concentrations of 0.0005 mg L-1 and 0.5 mg L-1 (Oliveira et al. 2015b). Adverse effects have also been observed on the fish species Oncorhynchus mykiss when exposed to concentrations of 1 to 500 µg L-1 of diclofenac for 28 days. Cytological changes in the liver, kidneys and gills were observed even at the lowest concentration tested (Triebskorn et al. 2004), as well as kidney lesions and gill changes in concentrations above 5 µg L-1 (Schwaiger et al. 2004).
Although the environmental risk was classified as insignificant, these literature data suggest the potential sublethal effects of diclofenac sodium salt on non-target aquatic organisms, as the concentration levels that were related with toxic effects have been reported in the aquatic environment (see Table 2). This may also implicate long-term risks to the aquatic biota.
The mean EC50 (48h) value for ketoprofen was 24.84 mg L-1. As this concentration is over thousand fold bigger than the levels of ketoprofen registered, for instance, in effluent from sewage treatment plants in Sweden (0.33 µg L-1) (Bendz et al. 2005) and Canada (0.09 µg L-1) (Lee et al. 2005), and in rivers in Europe (239 ng L-1) (Loos et al. 2009) and Brazil (0.62 µg L-1) (Ide et al. 2017), it is very unlikely to observe acute effects of this drug on the aquatic biota.
The results of chronic toxicity tests showed a high sensitivity of C. silvestrii to ketoprofen. The reproductive EC50 (8 days) of 1.94 mg L-1 for ketoprofen was the lowest among the drugs evaluated. A significant reduction in fecundity by 2.5 mg L-1 (LOEC) and total inhibition in neonate production by 5 mg L-1 was also observed. EC50 values (8 days) for ketoprofen were on average 12 times lower than EC50 values (48h) for the acute toxicity tests. Thus, as stated by Du et al. (2016), it is likely that the degree of toxic effects of pharmaceuticals to this organism may depend not only on the drug concentrations in the aquatic environment, but also on other factors such as the time of exposure.
Although ketoprofen has been detected in the aquatic environment, there is little information in the literature about its ecotoxicological effects on non-target organisms, especially invertebrates. The risk classification determined from environmental levels of ketoprofen has resulted in negligible risk. However, we demonstrated the potential of ketoprofen to cause adverse effects on the biota and due to the lack of information about potential toxic effects of ketoprofen, concerns about its environmental risks should continue to grow.
The mean EC50 (48h) for C. silvestrii exposed to paracetamol in the present study was 34.49 mg L-1 and the sensitivity to the drug was very close to that observed in the studies by Oliveira et al. (2018) for the same species (EC50, 48h = 40.3 mg L-1). For D. magna, EC50 (48h) values of 30.1 mg L-1 (Kim et al. 2007) and 50 mg L-1 (Henschel et al. 1997) have been described. Acute effects have also been reported on the fish species Oryzias latipes (LC50, 96h > 160 mg L-1) (Kim et al. 2007) and the freshwater Dugesia japonica (LC50, 96h = 150.8 mg L-1) (Li 2013).
The results of chronic toxicity tests with C. silvestrii exposed to paracetamol showed toxicity in the reproduction of the organisms with the LOEC value of 10 mg L-1 and reproductive EC50 (8 days) of 8.19 mg L-1. The 8-day EC50 values were 4 times lower than the 48-hour EC50 values found in acute toxicity tests. This confirms observations showing that zooplankton toxicity is closely related to the exposure time (Du et al. 2016).
Studies in the literature reported the occurrence of toxic effects on fecundity and rate of population growth in C. silvestrii exposed to 2 mg L-1 of paracetamol for 8 days (Oliveira et al. 2018), and in D. magna exposed to concentrations of 0.01 mg L-1 and 1 mg L-1 (48h) significant decrease in ChE and Se-GPx activities, respectively (Oliveira et al. 2015b). Comparing these results with the levels at which paracetamol has been detected in natural waters in the USA (10 µg L-1) (Kolpin et al. 2002), Brazil (above 13 µg L-1) (Sodré et al. 2007; Campanha et al. 2015), Taiwan (above 100 µg L-1) (Lin et al. 2008) and UK (above 69 µg L-1) (Roberts and Thomas 2006), we suggest that the there is a risk for aquatic organisms. Kim et al. (2007) reported possible environmental effects of paracetamol on the aquatic environment and advised further investigations. Authors encourage a more consistent approach to access the environmental risk of drugs using more sublethal parameters in toxicity tests, especially those based on biomarkers with enfaze to the observations in vitro bioassays.
Thus, although the environmental risk for paracetamol in this study was classified from low to insignificant, we agree that investigations to evaluate the long-term adverse effects of paracetamol on the aquatic ecosystems must be conducted.
Salicylic acid was the least acutely toxic drug for C. silvestrii, with an EC50 (48h) of 69.15 mg L-1, about 2 to 5 times less than the observed toxicity for the drugs diclofenac sodium salt, ketoprofen and paracetamol. Henschel et al. (1997) reported for salicylic acid EC50 (72h) > 100 mg L-1 for Scenedesmus subspicatus algae, EC50 (24h) of 118 mg L-1 for D. magna and LC50 (48h) of 37 mg L-1 for zebra fish embryos (Brachydanio rerio). Considering the concentrations necessary to cause acute effects, we do not expect that the registered environmental levels of salicylic acid have a direct impact on the survival of aquatic organisms.
Although reproductive toxicity was observed in the chronic exposures, C. silvestrii had a lower sensitivity to salicylic acid. Chronic toxicity tests performed by Marques et al. (2004) with D. magna showed no reproductive impairment after 21 days of exposure to concentrations of 1 to 10 mg L-1 of salicylic acid. It is important to note that our results and those described in the literature are limited to the target species and the endpoints investigated. The effects of salicylic acid and the other drugs were evaluated on isolated exposures and not consider the influences of other factors such as the co-occurrence of substances and fluctuations in environmental variables, which can maximize toxic effects.
In conclusion, although the drug concentrations that caused immobility and adverse effects to C. silvestrii in the present study were higher than their environmental levels, the assays consisted of exposures to a single drug with specific exposure times and endpoints. Therefore, the potential risk to tropical aquatic biota in the long-term and in exposure scenarios using multiple stressors remains to be investigated. In addition, SSD curves allowed the recognition of the absence of data in the literature for non-target invertebrate organisms in aquatic biota when exposed to drugs, especially, ketoprofen and salicylic acid.