Novel biomarker platform for a 3Rs approach to assessment of developmental and reproductive toxicity

Any chemical that is released into the environment has the potential to be harmful, including those used in the agro-and petrochemical industries which form a significant percentage of the UK GDP. Current regulations stipulate that such compounds must be tested for their potential to cause harm to the developing foetus and to reproduction and so chemicals are routinely tested on rats and rabbits over one or two generations to determine any effects. This uses a large number of animals (at least 2500 per compound) and is not entirely predictive of effects in man. We aim to develop an alternative testing approach to identify compounds with a high risk of developmental and reproductive toxicity (DART) using non-verterbrate model organisms, the nematode worm (C. elegans) and social amoebae (Dictyostelium). This approach will provide a sensitive and robust assay system and the use of two organisms in combination increases the predictive power massively. We will establish a collection of reporters for different developmental events in both organisms, clustering into useful groups informative of particular developmental and reproductive processes. The system will be educated by testing on compounds known to cause, or not to cause, DART culminating in the development of a predictive medium throughput pipeline. Ultimately this will enable the elimination of potentially harmful chemicals early in product development, without testing on rats and rabbits.

Developmental and reproductive toxicity (DART) assessment presents significant challenges during the compound discovery pipeline, involving both one and two-generational testing of rats and rabbits, thus utilizing at least 2500 animals per compound. Our objective is to establish a novel non-vertebrate biomarker-based alternative testing platform to replace animals for compounds with medium to high risk of DART by identifying and eliminating them prior to animal testing. A combinatorial approach using Caenorhabditis elegans and Dictyostelium discoideum is envisaged in order to synergistically enhance predictive power. These simple eukaryotic organisms are genetically tractable and molecular developmental processes are well defined and conserved with mammals. We will establish a library of fluorescent reporter strains in both organisms that will be used as biomarkers of DART. These will represent a broad range of developmental and reproductive processes and pathways to define a molecular signature of development. These reporters will be arrayed in clusters defining molecular and developmental pathways and changes in fluorescence (levels and/or expression patterns) will act as read outs for potential DART. Phenotypic anchoring will inform cluster applicability to fine tune the assay where appropriate. Ultimately we will establish a novel biomarker based pipeline that will serve to replace vertebrate animals in the early stages of testing for compounds with high risk of DART as they can be eliminated without recourse to experiments on rats and rabbits. In the long term use of non-vertebrate models has high potential to replace vertebrates within the regulatory framework in DART identification.

The testing platform generated in this proposal is designed to replace the use of rats and rabbits in identifying compounds with high risk of DART. Currently the regulatory framework demands one (OECD 414) and two (OECD 416) generational studies in rats and rabbits to assess developmental and reproductive toxicity respectively. Each study uses at least 2500 animals. Currently the introduction of new compounds in Europe requires 90,000 animals and costs 60 million Euros each year. In 2009 DART toxicity tests accounted for up to 90% of the animals used for testing and 70% of the costs. It has been suggested that compliance with European REACH (Registration, Evaluation, Authorisation and Restriction of Chemicals) legislation on chemical testing may use 20 times more animals and cost 6 times as much as previously estimated (Hartung and Rovida, 2009). A high percentage of these compounds do show some aspect of DART and therefore do not make it to market. If our system can identify half of these, then we will replace 45,000 animals per year in Europe. If only a single compound is removed this will save 2500 animals.

Academic scientists studying drug action in relation to developmental processes currently use mice as the organism of choice, in particular because of the availability of transgenic animals to facilitate identification of targets. As both the organisms we propose to use are genetically tractable they will provide a non-vertebrate alternative. Breeding of transgenic mice to study a single compound would use around 100 mice per year and such studies are undertaken in around 10 labs in the UK, therefore we estimate this could replace 1000 mice per year in the UK.

Reference: Hartung and Rovida (2009) Nature 460, 1080