Realistic assessment of the effects of bisphenol A on a fish population in ecosystemic conditions
Résumé
There is a clear lack of information on the impact of bisphenol A (BPA) on ecologically relevant endpoints, at both population and community scale. The pattern of effects is complex and ranges from behavioral disorder to reproductive impairments, therefore the resulting effects on a population can only properly be assessed through long term exposure in an ecosystemic context. Mesocosms studies provide such information for small fish like Gasterosteus aculeatus, a known suitable fish species to study endocrine disruption. INERIS mesocosms are designed for long term study of communities, in an ecosystemic context characterized by spatial heterogeneity, species heterogeneity and density-dependent effects. The mesocosms were set up in November with the selected nonfish species, introduced in quantities and at locations defined in a uniform protocol for each treatment. In the beginning of March, initial tagged populations were introduced in each mesocosm. 15 females and 10 males per mesocosm were selected. Based on literature data, three nominal concentrations of BPA were selected: 1, 10 and 100 µg/L. Treatment began in mid-April and ended by the end of September. Five groups of fish were identified: juvenile (< 25 mm), non-founder fish (? 25 mm without tag) male or female, founder fish male or female (? 25 mm with tag). The populations were compared based on the abundance of these different groups of fish, on the sex-ratio, on the condition index, and on the length frequency distribution of the fish population. The value of pH was significantly lower in 100 µg/L of BPA than in the other treatments. This observation could be explained by the more important number of invertebrates such as gastropods and/or the lower biomass of macrophyte. Concentrations of BPA measured were close to the nominal concentrations at 0 m from the inlet of the water in mesocosm then, concentrations decreased and were stable 5m after the inlet. Populations exposed to 100 µg/L of BPA showed a number of non-founder females and males and a juvenile length significantly higher than in the other populations. In our study, BPA perturbed the structure of the population at the maximum exposure concentration. Other trophic levels observed during the same experiments are being analysed to complete these results. We believe that the outcomes of the whole study would be a right basis for ecologically relevant risk assessment of bisphenol A.