Studies in Europe, Japan, and North America have reported that wild fish exposed to treated sewage effluents can exhibit significant physiological and reproductive abnormalities consistent with exposure to hormonally active chemicals. The main objective of this research project was to examine the estrogenic and androgenic activity in treated sewage to determine the risk associated with treated sewage discharges in Australia and New Zealand. Several bioassays, including a sheep estrogen receptor and a rainbow trout androgen receptor binding assay, were set up and validated with model compounds. The assays were then used to measure the estrogenic and androgenic activity in sewage samples from 15 municipal sewage treatment plants (STP) utilizing a variety of treatment technologies. Raw sewage samples contained high levels of both estrogenic and androgenic activity, up to 185 ng/L estradiol equivalents (EEq) and up to 9330 ng/L testosterone equivalents (TEq), respectively. Secondary treatment processes such as activated sludge had the greatest impact on removal of biological activity from the wastewater. The estrogenic and androgenic activity in final treated effluents were <1 to 4.2 ng/L EEq and <6.5 to 736 ng/L TEq, respectively. Based on lowest observable effective concentrations reported in the literature, these levels are unlikely to induce biological effects in exposed fish in the short term. To examine potential long-term effects, resident mosquitofish chronically exposed to undiluted treated sewage were sampled. Several morphological biomarkers indicative of endocrine disruption were measured and compared with mosquitofish captured at a reference site. Mosquitofish captured in a constructed wetland for tertiary treatment of secondary treated sewage exhibited morphological differences such as elongated anal fins consistent with exposure to androgenic chemicals, although this effect was not measurable in fish collected at sites further downstream or at any of the other sites. Based on these results, it is unlikely that mosquitofish populations would be significantly affected by exposure to final treated sewage. A reverse transcription real-time polymerase chain reaction (RT-PCR) method to measure the production of a female-specific protein (vitellogenin) mRNA in adult male mosquitofish was developed, and this could be used as a rapid test to detect early changes in individuals exposed to estrogenic activity.

Endocrine disrupting chemicals (EDCs) are ubiquitous in the environment and their presence in water bodies is documented. They discharge into surface water (SW) unmonitored, posing a threat to both aquatic and terrestrial lives. This is a challenge as not all populations have access to treated drinking water (TDW). The EDC contaminated serves as a route of exposure, together with ineffective treatment plants. Given the complexity of the endocrine system, EDCs may mimic or antagonise natural hormones or disrupt their synthesis, metabolism and excretion. The associated health effects include testicular dysgenesis syndrome, metabolic disorders and cancers. Policy and internationally standardised test methods are however sti ll limited. This study therefore aimed to assess the suitability of two assays used for screening estrogenic activity and one for androgenic activity in different water sources. The study consisted of two phases. In phase 1, water sample (tap, surface and treated wastewater) were collected from a catchment area in Pretoria. The samples and a spiked MilliQ laboratory water sample were extracted with solid phase extraction (SPE) and sent to Germany for distribution to participating laboratories. Samples (n=24) from six different countries were received to test for androgenic activity in the MDA-kb2 reporter gene assay. In phase 2, SW and TDW samples were collected from April 2015 until March 2016. The samples were filtered, extracted using SPE and assayed with the YES assay, T47D-KBluc reporter gene assay for estrogenic activity and MDA-kb2 reporter gene assay for androgenic activity. In phase 1, androgenic activity was detected in 4 out of 24 (21%) samples and ranged from 0.23 ± 0.040 ng/L to 0.008 ± 0.001 ng/L DHTEqs. In phase 2, estrogenic activity was detected in 16 out of 24 (67%) SW samples in the T47DKBluc reporter gene assay and ranged from 0.31 ± 0.05 pg/L to 10.51 ± 5.74 pg/L EEqs. It was below the detection limit (dl) in the YES assay. Androgenic activity was detected in 4 out of 24 (17%) SW samples, ranging from 0.0033 ± 0.0050 ng/L to 0.090 ± 0.040 ng/L DHTEqs. Androgenic and estrogenic activity was higher i n pretreatment samples compared to post-treatment in both treatment plants. In phase 1, the MDA-kb2 reporter gene assay was successfully applied to water samples from different sources. Androgenic activity was highest in treated wastewater. In phase 2, treatment plants proved to be effective in removing estrogens detected in the SW samples, as the TDW samples were below the dl. Estrogenic activity is within the ranges reported in other studies. Positive samples were below the 0.7 ng/L proposed trigger value for health risk assessments. Detected androgenic activity was lower in TDW samples compared to the SW samples supplying the two treatment plants indicating that they were both effective in removing the androgenic activity detected. Few studies have reported androgenic activity in tap water. This study strengthens the argument for using a battery of assays when monitoring endocrine activity as EDCs occur at low concentrations in mixtures. / Dissertation (MSc)--University of Pretoria, 2017. / School of Health Systems and Public Health (SHSPH) / MSc / Unrestricted