*2.3. Biotransformation*

To evaluate biotransformation, the protozoan *S. ambiguum* was incubated with the antidepressant solution (100 μg/L) in darkness for two days. The Orbitrap ™ high-resolution UPLC-MS/MS was used to determine the potential metabolites of the antidepressants in both medium and the protozoan cell. The tentative metabolites of the antidepressant were detected by Compound Discoverer Software (Thermo Fisher Scientific).

The tests were performed twice, and the relative area of the chromatogram peaks are presented in Table 2. The chromatograms of the tested samples were compared to that of the control samples. The peaks observed in two replicates of the samples and not visible in two controls were shown. The predicted transformation products and the di fference between the measured and theoretical mass are given. As controls, the antidepressant solutions without the protozoa were incubated under the same conditions. No transformation products were observed in the control samples (data not presented), which confirms the previous findings that these compounds are stable in the aquatic environment [42,43]. Derivatives of only two drugs (fluoxetine and paroxetine) were detected in the protozoa homogenates, whereas four to six transformation products were observed in aquatic media for each antidepressant. The very low levels inside the protozoan cells may be caused by the method of sample preparation. Because of their very low volume, the cell homogenates were analyzed without any enrichment techniques, while the medium was concentrated 100-fold by passing it through Hydrophilic-Lipophilic Balance (HLB) cartridges. The lack of metabolites inside the cells could also be caused by their better solubility in water, high elimination rate from the cells, and lower bioconcentration in the cells than those of the parent compounds.


**Table 2.** Biotransformation of the tested antidepressants by *S. ambiguum*. Relative abundance of the compounds in the protozoan *S. ambiguum* and in the medium after two days incubation of the protozoans with the parent compounds. The test was performed in duplicate.

Five mianserin derivatives were observed in the tested samples, and these were N-demethylation and oxidation products (Table 2). The major mianserin metabolites that are formed in the liver in humans are N-desmethylmianserin, 8-hydroxymianserin and mianserin N-oxide (www.drugbank.ca). Similar products, formed probably by oxidation and oxidative desmethylation, were observed for sertraline, but not fluoxetine (Table 2). Because of the low abundance of these compounds, it was not possible to confirm their structure by fragmentation. Three main sertraline metabolites have been reported in humans: desmethylsertraline, sertraline ketone and sertraline N-carbamoyl glucuronide [44]. In humans, fluoxetine and sertraline are mainly metabolized to N-desmethyl products, which retain their pharmacological activity [18]. N-desmethyl metabolites were also found in aquatic organisms. Silva et al. [18] presented several findings on the occurrence of norfluoxetine and norsertraline in many freshwater fish. These metabolites are more stable than their parent compounds and less polar; thus, their levels in many cases were higher than those of their parent compounds, especially in the liver and brain. However, the authors did not provide the source of these metabolites in aquatic organisms. In organisms collected from the environment, the most probable source of these compounds was the accumulation of metabolites of human origin. Only laboratory tests can prove the occurrence of biotransformation processes in aquatic organisms. Rodriguez et al. [45] detected residual norsertraline in crab cultures incubated with sertraline for two days. Chu et al. [46] found increased concentrations of norfluoxetine in fish incubated with fluoxetine. The mussel *M. galloprovincialis* was exposed to a nominal concentration of fluoxetine (75 ng/L) for 15 days [47]. The authors observed that the concentration of fluoxetine and norfluoxetine increased from 2.53 and 3.06 ng/g dry weight after 3 days up to 9.31 and 11.65 ng/g after 15 days, respectively. These results sugges<sup>t</sup> that fluoxetine accumulated in mussel tissues is likely to be metabolized into norfluoxetine with the increase in the time of exposure.

In humans, paroxetine is metabolized to paroxetine catechol, which is methylated and conjugated into second phase metabolites [42,48]. Cleavage of the paroxetine is also possible, which leads to the formation of the metabolite with a molecular mass of 209 Da [48]. The latter compound was also observed in our studies (Table 2).

Two identical derivatives of SSRIs were observed, which resulted from the addition of CO and C15H22O (Table 2). To the best of our knowledge, such transformation products have not been described either for humans or for aquatic organisms. Their structures were not proposed in the current study because of their very low abundance to perform fragmentation studies. However, this will be the subject of future studies.
