Validating zebrafish for safety drug screening: A comparison of zebrafish and rodent toxicity data.

Rafael Miñana Prieto, Maria Rubio Brotons, Carles Cornet, Jone Ibarra, Sylvia Diballa, Tiziana Pederzani, Flavia de Santis, and Javier Terriente. ZeClinics SL, PRBB, Barcelona, SPAIN.


In recent years, the zebrafish has become a valuable model. It is gaining popularity in the field of safety pharmacology due to its unique biological advantages. There is a gap between high-throughput in-vitro studies and low-throughput in-vivo studies during the drug discovery pipeline. The use of zebrafish-based drug screening could fill that gap and become an important asset to assess toxicity and efficacy of novel drugs. Zebrafish has, from early developmental stages, fully functional organs from a physiological point of view and large transparent progeny, which develops externally and make them ideal for non-invasive approaches, such as in-vivo imaging. Despite these advantages, the use of zebrafish is still scarce in current drug discovery pipelines and requires further validation to achieve a full adoption by the regulatory agencies. Zebrafish assays will only be useful if they can show high predictive power. Thus, we used our toxicological data (LC50) obtained in zebrafish embryos and have compared with LD50 values obtained from rodents (mice and rats), to evaluate the predictivity of the model.

Besides, we performed drug bioavailability studies of the same compounds in zebrafish embryos to address the role of drug permeability in toxicity assessment. The aim of this study is to examine how the covariance of zebrafish and rodent toxicity is influenced by factors such as compound type, absorption, metabolism and mechanism of toxicity. For that, the three most used routes of administration in rodents (oral, intraperitoneal and intravenous) have been compared to our drug incubation. For the compounds examined here, we have seen that drug incubation better correlates with LD50 values obtained via intravenous route, than with intraperitoneal or oral respectively.

Furthermore, we have analysed if the log P (partition- coefficient) value could have any impact in the drug toxicity seen in the embryos. Our preliminary results suggest, that even though there is no strong correlation between LC50 and log P, those compounds which are more hydrophilic (less lipophilic) and therefore, more water-soluble; tend to be less toxic than the ones which possess high log P values (more lipophilic, less hydrophilic, less water-soluble).

Our results will allow placing the use of zebrafish-based drug screening in an even better position for biomedical drug discovery purposes. More work is required, but we hope that this and other translational validation studies could help to promote the zebrafish as a potential alternative model to toxicology testing.

Validating zebrafish for safety drug screening: A comparison of zebrafish and rodent toxicity data.