Screening for potential effects of endocrine-disrupting chemicals in peri-urban creeks and rivers in Melbourne, Australia using mosquitofish and recombinant receptor-reporter gene assays.

Sexually mature male mosquitofish (Gambusia holbrooki) were collected from various sites around Melbourne in 2009 to evaluate the performance of gonopodial indices as a biomarker for endocrine disruption in Melbourne's waterways. The mosquitofish indices assessed were body length (BL), gonopodial length (GL)/BL ratio, ray 4:6 ratio and the absence or presence of hooks and serrae, and these varied between sites. The study was complemented by measurements of estrogenic, retinoid, thyroid and aryl hydrocarbon (AhR) receptor activities of the water. Male mosquitofish were 16.3-21.5 mm in length, and although there was a statistically significant positive relationship showing that bigger fish had longer gonopodia than small fish (r2 = 0.52, p < 0.001), there were few significant differences in GL/BL ratio of fish between sites. Measured estrogenic activity was mostly in the range 0.1-1.7 ng/L EEQ, with one site having much higher levels (~12 ng/L EEQ). Aryl hydrocarbon (AhR) receptor activity was observed in all water samples (7-180 ng/L βNF EQ), although there was no consistent pattern in the level of AhR activity observed, i.e., 'clean' sites were as likely to return a high AhR activity response as urban or wastewater treatment plant (WWTP)-impacted sites. There was no correlation between measurements of receptor actvity and gonopodial length (GL):BL ratio and BL. We conclude that the mosquitofish gonopodia only fulfills part of the criteria for biomarker selection for screening. The mosquitofish indices assessed were cheap and easy-to-perform procedures; however, there is no baseline data from the selected sites to evaluate whether differences in the morpholical indices observed at a site were a result of natural selection in the population or due to estrogenic exposure.