Pharmaceuticals in drinking water sources have raised significant concerns due to their persistent input and potential human health risks. The seasonal occurrence of 25 pharmaceuticals including 23 antibiotics, paracetamol (PAR), and carbamazepine (CMZ) in Taihu Lake was investigated; meanwhile, the distribution and removal of these pharmaceuticals in two drinking water treatment plants (DWTPs) and a constructed wetland were evaluated. A high detection frequency (>70%) in the Taihu Lake was observed for nearly all the 25 pharmaceutics. Chlortetracycline (234.7 ng L), chloramphenicol (27.1 ng L), erythromycin (72.6 ng L), PAR (71.7 ng L), and CMZP (23.6 ng L) are compounds with both a high detection frequency (100%) and the highest concentrations, suggesting their wide use in the Taihu Basin. Higher concentrations of chloramphenicols, macrolides, PAR, and CMZP were observed in dry season than in wet season, probably due to the low flow conditions of the lake in winter and the properties of pharmaceuticals. The overall contamination levels of antibiotic pharmaceutics (0.2-74.9 ng L) in the Taihu Lake were lower than or comparable to those reported worldwide. However, for nonantibiotic pharmaceutics, PAR (45.0 ng L) and CMZP (14.5 ng L), significantly higher concentrations were observed in the Taihu Lake than at a global scale. High detection frequencies of 25 pharmaceuticals were observed in both the two DWTPs (100%) and the wetland (>60%) except for florfenicol and sulfapyridine. The removal efficacies of the studied pharmaceuticals in DWTP B with advanced treatment processes including ozonation and granular activated carbon filtration (16.7-100%) were superior to DWTP A with conventional treatment processes (2.9-100%), except for sulfonamides. Wetland C with the constructed root channel technology was efficient (24.2-100%) for removing most pharmaceuticals. This work suggests that the application of cost-effective technologies such as constructed wetlands should be considered as an efficient alternative for removing pharmaceuticals from water supply sources.

Background: The successful application of Fish plasma model (FPM) will greatly simplify the risk assessment of drugs. At present, the FPM has been applied to the risk assessment of several human drugs with high hydrophobicity. However, its applicability to a wide variety of compounds needs to be tested. Filed work about distribution characteristics of endocrine disrupting compounds (EDCs) in water and in fish plasma plays a key role in promoting the successful application of FPM. However, there are few reports on the distribution of EDCs in fish, especially in wild fish plasma. Results: The distribution of 31 EDCs, including seven estrogens, eight androgens, six progesterones, five glucocorticoids and five industrial compounds, in water and in plasma of five categories of wild fishes in Taihu Lake was studied and the typical FPM was tested by the field data. The detected concentration of most compounds is relatively low (< 10 ng/L), and the industrial compounds are the predominant pollutants with the highest concentration up to 291.7 ng/L (Bisphenol A) in water. In general, glucocorticoids were detected at highest concentrations in plasma of all kinds of fishes, followed by industrial compounds and estrogens. Except for glucocorticoids (up to 43.61 for Cortisone), the average concentrations of other four categories of EDCs in fish plasma were generally low (< 10 μ g / L). The available measured bioaccumulation factors (BAFs) of 20 kinds of EDCs were in the range of 0 to 5626. The bioaccumulation of EDCs in fish plasma is not only hydrophobicity dependent, and it is both fish species-specific and compound-dependent. The classical FPM was tested and the results indicated that the good coincide of measured logBAFs and the theoretical logBCFs was only observed for limited EDCs (7 out of 20). The present available FPM generally underestimated the LogBAFs of most hydrophilic EDCs (logKow <3.87, 11 out of 20) whereas overestimated the LogBAFs of several high hydrophobic EDCs (LogKow >3.87) in fish plasma. Conclusion：Although the present FPM is barely satisfactory, it is still promising for predicating the accumulation of EDCs in fish plasma and for further environmental risk assessment.

A confirmatory method for the simultaneous detection of 29 pharmaceuticals in fish muscle and plasma was developed by using solid-phase extraction combined with ultra-high performance liquid chromatography and tandem mass spectrometry. Fish samples were extracted with methanol and enriched using Oasis HLB solid-phase extraction columns in one step. Twenty-nine target pharmaceuticals were quantified by the internal standard method and the calibration curves showed good linearity in a wide range with determination coefficients of greater than 0.913. The detection limits of the pharmaceuticals ranged from 0.01 to 2.00 μg/kg (μg/L). The applicability of the method was checked by precision and recovery experiments. The average recoveries of the 29 pharmaceuticals were between 61 and 111%, and all the relative standard deviations were below 25%. Our reported method has been demonstrated to be sensitive, convenient, rapid, and reliable for the simultaneous determination of 29 pharmaceuticals in fish muscle and plasma. Real sample determination showed that 25 and 9 of the 29 compounds were detected in fish muscle and plasma, respectively.

Background: The successful application of Fish plasma model (FPM) will greatly simplify the risk assessment of drugs. At present, the FPM has been applied to the risk assessment of several human drugs with high hydrophobicity. However, its applicability to a wide variety of compounds needs to be tested. Field work about distribution characteristics of endocrine disrupting compounds (EDCs) in water and in fish plasma plays a key role in promoting the successful application of FPM. However, there are few reports on the distribution of EDCs in fish, especially in wild fish plasma. Results: The distribution of 31 EDCs, including seven estrogens, eight androgens, six progesterones, five glucocorticoids, and five industrial compounds, in water and plasma of five categories of wild fishes in Taihu Lake was studied and the typical FPM was tested by the field data. The detected concentration of most compounds is relatively low (< 10 ng/L), and the industrial compounds are the predominant pollutants with the highest concentration up to 291.7 ng/L (Bisphenol A) in water. In general, glucocorticoids were detected at the highest concentrations in plasma of all kinds of fishes, followed by industrial compounds and estrogens. Except for glucocorticoids (up to 43.61 for Cortisone), the average concentrations of the other four categories of EDCs in fish plasma were generally low (< 10 μg/L). The available measured bioaccumulation factors (BAFs) of 20 kinds of EDCs were in the range of 0 to 5626. The bioaccumulation of EDCs in fish plasma is not only hydrophobicity-dependent, but is also both fish species-specific and compound-dependent. The classical FPM was tested and the results indicated that the good coincides of measured log BAFs and the theoretical log BCFs were only observed for limited EDCs (7 out of 20). The present available FPM generally underestimated the Log BAFs of most hydrophilic EDCs (log Kow < 3.87, 11 out of 20) whereas overestimated the Log BAFs of several high hydrophobic EDCs (Log Kow > 3.87) in fish plasma. Conclusion: Although the present FPM is barely satisfactory, it is still promising for predicting the accumulation of EDCs in fish plasma and for further environmental risk assessment.

Solid phase extraction combined with ultra‐performance liquid chromatography‐tandem mass spectrometry was developed for the simultaneous determination of 31 endocrine‐disrupting chemicals in fish plasma. The strong anion exchange/primary‐secondary amine cartridge and the mixed cation exchange cartridge were used in tandem instead of using a single mixed cation exchange cartridge for sample purification. Suitable eluents were selected for each of the two cartridges: 4.5% ammonia/acetonitrile solution for cartridges in tandem and acetone:n‐hexane (V:V = 3:7) for the strong anion exchange/primary‐secondary amine cartridge alone. With this optimized Solid phase extraction method, the recoveries of 31 endocrine disrupting chemicals were between 43.0% and 131.3%, the method detection limits were 0.45 to 1.35 ng/ml, and the limits of quantitation were 1.50–4.50 ng/ml. The innovative pretreatment method that connects two cartridges in tandem is well positioned to mitigate the matrix effects of fish plasma, thereby improving the accuracy of multiclass endocrine‐disrupting chemicals determination. The significance of this method is to facilitate the application of the fish plasma model for the environmental risk assessment of endocrine‐disrupting chemicals.