The applicability of the "read-across hypothesis" for assessing the effects of human pharmaceuticals on fish

Abstract

The presence of human pharmaceuticals in the environment has raised concerns regarding their potential adverse effects on non-target aquatic organisms. Pharmaceuticals are designed to target specific molecular pathways in humans in order to produce known pharmacological and physiological responses, before toxicological effects are seen. The “Read-Across Hypothesis” stipulates that pharmaceuticals can produce similar biological effects in fish, as in humans, if the molecular target is conserved, and the internal (blood plasma) concentrations are similar. The read-across hypothesis was tested using ibuprofen, a non-steroidal anti-inflammatory drug, and the model fish test species, the fathead minnow (Pimephales promelas), to determine if ibuprofen can cause similar target-mediated effects in teleost fish and humans, at comparable blood plasma concentrations. Fathead minnows were exposed, using continuous flow-through systems, for ≤96 hours to a range of ibuprofen water concentrations (100, 270, 370 and 500 µg/L) to determine if plasma concentrations similar to human therapeutic plasma concentrations (HTPCs, or Cmax) could be established in fish blood plasma. The mode of action of ibuprofen was used to identify relevant endpoints (i.e. cyclooxygenase (COX) enzyme) in order to examine target-mediated effects following drug exposure. The water and plasma ibuprofen concentrations were determined using LC-MS/MS. The measured ibuprofen plasma concentrations in individual fish were linked to target-mediated effects on COX gene expression, COX enzyme activity and prostaglandin E2 (PGE2) synthesis (products of COX activity), which were quantified using molecular (QPCR) and biochemical (colourimetric and enzyme immunoassay) assays, and linked with the Cmax of ibuprofen. It was demonstrated that in fish with a mean ibuprofen plasma concentration 1.8-fold below the Cmax, PGE2 concentrations (the most robust endpoint) was significantly inhibited following ibuprofen exposure. However, in fish exposed to an ibuprofen concentration closer to (2 to 3-fold above) environmentally relevant water concentrations (i.e. 9 µg/L), when the mean plasma concentration was 224-fold below the Cmax, fish did not respond to ibuprofen exposure. This study provides qualitative and quantitative evidence for the applicability of the “read-across hypothesis”, and highlights its potential utility for prioritising pharmaceuticals for environmental risk assessment.