Evaluating developmental osteotoxicity in the embryonic stem cell test

The teratogenic potential of a compound can devastate its chances as a drug candidate. Currently, embryotoxicity tests are performed in living animals. The drug industry has been in need for a practical in vitro embryotoxicity model for the last two to three decades. Over ten years ago, the currently most promising in vitro embryotoxicity test, the embryonic stem cell Test (EST) has been called to life. In contrast to other in vitro embryotoxicity tests, the EST uses two permanent cell lines, murine embryonic stem cells (ESCs) and differentiated 3T3 fibroblasts and thus does not require the use of pregnant animals. The EST may be useful to identify developmental inhibition particularly in cardiac tissue. However, the analysis of cardiomyocyte differentiation alone as an endpoint does not suffice for the determination of embryotoxic effects. Recently, we improved the EST to include other tissue endpoints, such as bone, cartilage and neurons. ESC-derived osteoblasts reach a state of full maturity after 30 days of differentiation in vitro as assessed by the expression of osteoblast-specific genes (i.e. osteocalcin). However, such a long test duration is not desirable for an in vitro assay. In the present study, we have thus begun to explore new means of evaluating the degree of osteogenesis. First, we have commenced to culture the differentiating ESCs in a monolayer instead of plating intact embryoid bodies. Recently, we further began to utilize IMAGE analysis to quantify osteogenesis, a process in which cultures are photographed and the percentage of black pixels is calculated. This is only possible because osteocalcin-expressing mineralized osteoblasts appear black in phase contrast microscopy. Furthermore, we analyzed osteoblast yield by measuring the amount of mineralized calcium incorporated into the matrix and ALP enzyme activity and compared this with quantitative real-time PCR for osteocalcin. A test set of chemicals, including Penicillin G and 5-fluorouracil, was then assayed using both mouse and primate ESCs. By using a monolayer type culture approach, mineralized cells were more evenly distributed fascilitating analysis already by differentiation day 14. Both the calcium assay and the morphometric read-out proofed to be valuable tools in detecting embryotoxicity. IC50 values taken from dose-response curves were comparable between these new endpoint assays and osteocalcin marker gene expression. Using primate ESCs for the differentiation, we also detected species-specific responses to bone teratogens. Ultimately, we were able to substitute expensive endpoint assays for less costly, but equally robust evaluation methods, and to reduce the test duration. This accomplishment should make the test more attractive for routine usage in industrial screening.