Skip to main content

A new strategy for developmental toxicology in vitro using molecular and cellular markers in mammalian whole embryo culture to enhance sensitivity and selectivity


The goal is to devise an approach that will better predict the developmental toxic potential of a chemical, based upon an in vitro test. A reliable system would have wide application, reduce the use of animals, harmonise international testing, and bring economic savings to manufacturers. Our hypothesis is that using expression domains as end-points will enhance the sensitivity and predictivity of whole embryo culture. The rationale is: a) Domains can reveal regions of cellular commitment which have no morphological counterpart. b) Some toxicants may alter gene expression, directly or indirectly, to shift margins of expression to ectopic sites. c) Dysmorphogenesis must be preceded by cellular changes not observable at the gross level.
The plan is to treat embryos with prototype developmental toxicants and monitor the expression of selected genes, concentrating on the CNS as the target tissue. We are using rat and mouse for species comparison, and in vivo and cultured embryos for validation. We have selected six developmental control genes: Emx-2, Otx-2, Krox-20, Hoxb-4, Pax-3 & Pax-6.
We have now examined the actions of several prototype teratogens: valproic acid (VPA); all-trans retinoic acid (AT-RA); ethanol (EtOH); methyl mercury (MM); triazole; and hyperthermia. We have detected chemically-induced changes in expression domains by in situ hybridisation (ISH), whole-mount in situ hybridisation (WMISH) and whole-mount immunostaining, as summarised below. The studies show that the expression domains examined are undoubtedly very useful end-points in specific studies. However, whether they are practical end-points in routine screening is less clear, but will be resolved in the remainder of the project.
VPA on rat & mouse embryos in vivo (B, GB & I) A major part of the work has been devoted to the analysis of rat and mouse embryos exposed in utero to 0, 75 or 150 mg/kg VPA s.c. Some embryos were collected on day 9, examined for gross morphological abnormalities, fixed, and shipped to Louvain-la-Neuve and London, where they are being analysed for gene expression. Others were collected just prior to term and examined for structural defects, which showed greater sensitivity of the mouse and an apparent effect on axial skeleton specification.
AT-RA and 9-cis-RA effects on mouse embryos in vitro (B) Early somite mouse embryos have been exposed to AT-RA or 9-cis-RA for 8h and examined immediately or after further culture for 40h. Using ISH and 3D-reconstruction it has been shown that treatment induces ectopic expression of Hoxb-1 and Hoxb-2, as well as a specific down-regulation of Krox-20 in rhombomere 5.
VPA on mouse and rat embryos in culture (B, GB) To mimic effects in vivo, embryos have been exposed briefly to 1.5 mM VPA, then cultured for 24h untreated. Mouse embryos have then been analysed by ISH and rat embryos by WMISH. In mouse, analysis of Emx-2, Otx-2, Krox-20 and Pax-3 showed a reduction in Krox-20 expression in rhombomere 3, but no other major effects on spatial patterns. In rat, an analysis of Emx-2, Hoxb-4 and Pax-3 showed altered expression, but only commensurate with altered morphology. A refinement in culture technique was shown to increase sensitivity of mouse embryos to VPA by two-fold.
Whole-mount immunostaining to monitor ganglion development (B, GB) In a number of studies in both mouse and rat, it has been demonstrated that using antibodies to neurofilament proteins shows patterns of ganglion development, and reveals chemically-induced effects that are not visible from gross morphology. In culture, mouse embryos exposed to VPA, ethanol, AT-RA or MM, and rat embryos exposed to triazole or AT-RA, all develop abnormal cranial ganglia revealed by this approach. In some cases, this is preceded by alterations in the rhombomeres, which can be detected by in situ analysis of Krox-20, Hoxb-1 and Hoxb-2.
WHISH of rat embryos after hyperthermia (GB) At pre-somite stages, hyperthermia induces an apparent loss of cranial structures, much like AT-RA. To see if cell populations are lost, much-diminished or re-specified, treated embryos were examined for Emx-2, Otx-2, Krox-20, Hoxb-1 and Pax-6 expression. This showed that fore-brain populations are much-reduced, but not lost.

Funding Scheme

CSC - Cost-sharing contracts


MRC Experimental Embryology and Teratology Unit
St George's Hospital Medical School Cranmer Terrace
SW17 0RE London
United Kingdom

Participants (2)

5/3,Place Croix Du Sud 5/3
1348 Louvain-la-neuve
Università degli Studi di Milano
Via Celoria 26
20133 Milano