In recent years, a growing number of human pharmaceuticals have been detected in the aquatic environment, generally at low concentrations (sub-ng/L to low μg/L). These compounds are characterised by highly specific mechanisms of action, high potency and prolonged activity in order to minimise dosing requirements and potential toxicity in patients. Among the various classes of pharmaceuticals, steroids and anti-steroids are widely used, as shown by the analysis of their clinical use carried out at the beginning of this Ph.D. project. Although the amounts used are much lower than the amounts of some other pharmaceuticals (e.g. analgesics), their ability to affect important physiological processes in fish (e.g. reproduction) at very low concentrations (ng/L) suggest that this class of compounds should represent a high priority for ecotoxicological research. In particular, this Ph.D. project addressed the question of whether or not dutasteride, a human pharmaceutical mainly used to treat benign prostatic hyperplasia, may cause adverse effects in the teleost fathead minnow (Pimephales promelas) by inhibiting the activity of both isoforms of 5α-reductase (5αR), the enzyme which convert testosterone into dihydrotestosterone (DHT). The theoretical framework used to guide the design of the experimental studies was based on the combination of several conceptual approaches, including the study of the evolutionary degree of conservation and functionality of the drug target in non-target species, and the cross-species extrapolation of pharmacological and toxicological information generated during pre-clinical and clinical studies in mammals during drug development. The results obtained during the first phase of this Ph.D. project strongly suggested that DHT has a physiological role in the fathead minnow. In fact, 5αRs are evolutionary conserved in this species, 5αRs genes are expressed in tissues such as the testis, and DHT circulates in fathead minnow plasma at concentrations similar to those detected in humans. These findings represented the rationale for testing the effects of dutasteride in the fathead minnow. Dutasteride caused significant adverse effects in all the in vivo studies performed in order to evaluate its potential toxicity on fish, including early life stage and short term reproduction studies, and all the tested life stages were sensitive to the inhibition of 5αRs activity; however, none of the observed adverse effects occurred at concentrations of exposure lower than 32 μg/L (measured concentration). The results also showed that female fish are highly sensitive to disruption of the androgenic pathways, highlighting their utility for the evaluation of potential adverse effects caused by anti-androgens on fish. In conclusion, the results presented in this Thesis suggest that, at present, the potential presence of dutasteride in the environment does not represent a risk to wild fish populations, due to the high concentrations required to elict significant adverse effect (LOEC = 32 μg/L) and the low volume of drug prescribed every year (5.07 kg in UK in 2006). However, the high bioaccumulation factor of dutasteride suggest that further studies should be conducted to elucidate the role played by the bioaccumulation process in the toxicity responses observed in fish.

The toxicological screening of the numerous chemicals that we are exposed to requires significant cost and the use of animals. Accordingly, more efficient methods for the evaluation of toxicity are required to reduce cost and the number of animals used. Computational strategies have the potential to reduce both the cost and the use of animal testing in toxicity screening. The ultimate goal of this thesis is to develop computational models for the prediction of toxicological endpoints that can serve as an alternative to animal testing. In Paper I, an attempt was made to construct a global quantitative structure-activity relationship (QSAR)model for the acute toxicity endpoint (LD50 values) using the Munro database that represents a broad chemical landscape. Such a model could be used for acute toxicity screening of chemicals of diverse structures. Paper II focuses on the use of acute toxicity data to support the prediction of chronic toxicity. The results of this study suggest that for related chemicals having acute toxicities within a similar range, their lowest observed effect levels (LOELs) can be used in read-across strategies to fill gaps in chronic toxicity data. In Paper III a k-nearest neighbor (k-NN) classification model was developed to predict human ether-a-go-go related gene (hERG)-derived toxicity. The results suggest that the model has potential for use in identifying compounds with hERG-liabilities, e.g. in drug development.

Vid riskbedömning av enskilda ämnen utvärderas riskkvoten, kvoten mellan exponeringsnivå och ett hälsobaserat riktvärde. Flertalet toxikologiska studier styrker antagandet att en blandning av kemikalier, vars koncentrationer var och en för sig inte leder till någon skadlig effekt, kan samverka och leda till skadliga effekter. I detta examensarbete har jag som ett led i den hälsobaserade miljögiftsövervakningen vid Miljöförvaltningen, Stockholms stad, utvecklat ett beräkningsverktyg för att uppskatta kombinationseffekter vid blandningar av kemikalier som förekommer i inomhusmiljö. Beräkningsverktyget bygger på modellen för koncentrationsaddition under antagandet att inga avvikelser förekommer, där riskkvoter för enskilda ämnen summeras till ett översiktligt riskindex. Exponeringsvägarna inandning och oralt intag beaktas så att respektive riskindex adderas till ett multipelt riskindex. Vidare bygger verktyget på antagandet att koncentrationsaddition, som gäller för lika toxikologisk verkan på molekylär nivå även gäller på systemnivå, så att ämnen kan delas in i grupper efter de hälsoeffekter som de orsakar och därmed möjliggöra en utvärdering av multipelt riskindex för respektive hälsoeffekt. Bristen på hälsobaserade riktvärden för ämnen i luft har hanterats genom beräkning av uppskattat dagligt intag via inandning, samt genom att använda öppna resurser för identifiering av toxikologiskt lika ämnen. För PFAS-ämnen och PAH:er har relativa potensfaktorer använts vid beräkning av riktvärden. Utvärdering har även gjorts med toxikologiska referenspunkter justerade med osäkerhetsfaktorer. Beräkningsverktyget har utvecklats i Excel med ingående ämnen i en tabell med indelning efter hälsoeffekt. Test av kriterier för gruppering inom det aktuella beräkningsverktyget har gjorts genom jämförande analyser med två andra verktyg för riskbedömning av blandningar. Beräkningsverktyget användes sedan för att utvärdera 225 unika ämnen med kvantitativa analysdata från mätningar av inomhusluft respektive damm i Stockholms stads förskolor. Utvärderingen visar att kombinationseffekter har störst inverkan för hälsoeffekterna irritation och andningsvägar respektive utveckling. / <p>Inom examensarbetet har ett excelverktyg utvecklats, vilket beskrivs i publikationen. </p>

Brominated flame retardants (BFRs) are chemicals used in e.g. electronic equipment, textiles and plastics for the prevention of fire. Over recent decades, concern has been raised regarding some heavily used BFRs, since the levels in the environment have been increasing. In the present thesis, persistent, bioaccumulative, and toxic (PBT) properties were studied for a structurally varied set of BFRs selected to represent more than 60 organic BFRs. The studied BFRs include: 2,4,4'-tribromodiphenyl ether (BDE 28), 2,2',3,4,4',5',6-heptabromodiphenyl ether (BDE  183), decabromodiphenyl ether (BDE 209), hexabromocyclododecane (HBCD), tetrabromobisphenol A (TBBPA), tetrabromobisphenol A 2-hydroxyethyl ether (TBBPA OHEE), 2,4,6-tribromophenol (246BrPh), 1,2-dibromo-4-(1,2-dibromoethyl)cyclohexane (TBECH), and hexabromobenzene (HxBrBz). It is likely that soil will act as a sink for BFRs. Therefore, studies of BFRs were performed on persistence in soil, and on bioaccumulation from soil in the earthworm Eisenia fetida. Large variation in the biodegradability in soil among the tested BFRs was observed. The studied brominated diphenyl ethers (BDE 28 and BDE 209) were very persistent under both aerobic and anaerobic conditions, while 246BrPh and TBECH degraded quickly in both aerobic and anaerobic soil. The bioaccumulation in earthworm from soil was high for HxBrBz, TBECH and for tested brominated diphenyl ethers with 4-6 bromine atoms (BDE 47, BDE 99, and BDE 153). Bioaccumulation was also studied in zebrafish after dietary exposure to a mixture of BFRs. It was shown that several metabolites were formed and retained in zebrafish, which highlights the importance of also searching for and identifying persistent degradation products. Maternal transfer was shown for all BFRs present in the female zebrafish. This shows that zebrafish young (fry) are exposed to these BFRs at approximately the same concentrations as female zebrafish during the early-life stages, when fish are usually most sensitive to organic contaminants. Toxicity of individual BFRs and a BFR mixture was studied in Nitocra spinipes using a silica gel-based system. Highest toxicities were observed for BDE 28, TBBPA, and TBBPA OHEE. In the mixture toxicity study, simultaneous exposure to low concentrations (individually causing no significant effect) of six BFRs significantly affected the survival of Nitocra spinipes. The results from the PBT studies presented in the thesis and literature data were compared with the criteria for PBT classification, as set in the European REACH legislation. Further, some BFRs with physico-chemical properties similar to those of identified PBTs were suggested to be prioritized for future PBT testing.