P-glycoprotein (P-gp; abcb1) is one of the major ABC transport proteins that mediates multixenobiotic resistance (MXR) defense in fish. In order to offer a sound evaluation of its ecotoxicological relevance it is critical to characterize substrate specificity of fish P-gp. Measurement of the ATPase activity is a reliable approach often used to discern type of interaction of various drugs with mammalian P-gp. A similar assay has never been used for characterization of P-gp in aquatic organisms and the main goal of this study was to develop a specific ATPase assay for characterization of fish P-gp. For this purpose we have used P-gp enriched membrane vesicles isolated from fish hepatoma PLHC-1/dox cells characterized by high overexpression of P-gp. As additional demonstration of a P-gp specific phenotype, we have quantified transcript expression of a series of eight ABC efflux transporter genes constitutively expressed in PLHC-1 wild type and PLHC-1/dox cells. Transcript expression analysis confirmed high and specific P-gp transcript overexpression in PLHC-1/dox cells. Provided that the transcript abundance is translated to protein, the development of ATPase assay is enabled. Using this model we determined Km ATP of 0.4 mM, baseline ATPase activity from 35–50 nmol/mg PROT/min, and maximal activation of ATPase activity obtained for fish P-gp in our system was 1.8–2.5-fold over baseline. All these values were in good agreement with data previously reported for mammalian P-gp. In order to perform a more detailed characterization of fish P-gp substrate specificity, in the next step of our study we used the developed ATPase assay to test 50 different compounds for their interaction with fish P-gp. The same set of compounds was also tested with calcein-AM (Ca-AM) transport activity assay both using PLHC-1/dox cells and NIH 3T3/MDR1 fibroblast cells overexpressing human P-gp. Our results showed that there is a clear difference for some substances—five compounds specifically interacted only with fish P-gp, while seven compounds exhibited interaction with human P-gp only. Most of the compounds tested in this study showed similar behavior in respect to fish or human P-gp and relatively high correlation in the interaction potency was found between fish and human P-gp. In summary, the described results represent the first in depth insight into substrate specificity of an important xenobiotic efflux transporter in fish. In addition, our study showed that combination of Ca-AM assay and the developed ATPase assay using inside/out vesicles isolated from PLHC-1/dox cells, offers a high-throughput and reliable approach for identification of environmentally relevant pollutants that interact with fish P-gp.