Way Forward Essay on Current and Future State of Developmental Toxicology Assays

Home / New Perspectives / Reproductive & Developmental Toxicity / Way Forward Essay on Current and Future State of Developmental Toxicology Assays

Reproductive & Developmental Toxicity

Way Forward Essay on Current and Future State of Developmental Toxicology Assays

Cindy X. Zhang, Julie Panzica-Kelly, and Karen Augustine-Rauch, Bristol-Myers Squibb

Published: July 29, 2009

About the Author(s)
Cindy Zhang received her B.S. in Wuhan, China and M.S. in Asian Institute of Technology in Bangkok, Thailand. Both degrees were in the area of Aquatic Biology with a focus on fish genetics, disease, and nutrition. In 1995, she obtained a M.S. in Immunology and Molecular Pathology from the University of Florida, where she evaluated the protective effects of interleukin-1 on hematopoietic cells treated with the anti-oncologic agent, L-phenylalanine mustard. From 1995 to 1998, she worked in the Center for Cancer Treatment and Research in Richland Memorial Hospital, which is affiliated with the University of South Carolina. Cindy engaged in various cancer research projects, which included working with modified stem cells for long-term hematopoietic reconstitution. Cindy has subsequently worked in industry at GlaxoSmithKline and Bristol Myers Squibb (BMS) in the departments of Drug Safety Assessment and Discovery Toxicology, where her research projects have been mainly focused on assay development in hematotoxicity testing and teratogen screening using in vivo animal models, primary cells, cell lines, as well as in vitro rat or zebra fish embryo cultures.

Currently Cindy is engaged in evaluation and selection of predictive statistical models for supporting refined in vitro teratogenicity screen assays and generating guideline documents describing the criteria associated with the morphological score systems used in the rat and zebrafish embryo culture screening assays developed by BMS. Cindy’s interests include advances in flow cytometry as well as studying mechanisms of teratogenesis and toxicity. Her memberships in Professional societies have included the International Society of Experimental Hematology, International Society for Hematotherapy & Graft Engineering, and International Society for Analytical Cytology.

Cindy X. Zhang, M.S.
Bristol-Myers Squibb
Discovery Toxicology
Pharmaceutical Candidate Optimization
311 Pennington Rocky Hill Rd
Pennington, NJ 08534-2130
Email: cindy.zhang@bms.com

Julieta Panzica-Kelly obtained a B.S. in biology from Pennsylvania State University. She subsequently joined BMS where she has worked in the investigative reproductive toxicology group for the last several years. During this time, she has been engaged in the development, optimization and validation of novel developmental toxicology assay and statistical model development including the zebrafish embryo culture assay and molecular embryonic stem cell assay (MESCA) described in this essay. She is also completing efforts toward obtaining a Masters’ degree in developmental and cell biology at Thomas Jefferson University in Philadelphia, Pennsylvania. Her thesis project is designed to pursue the molecular underpinning of adverse developmental outcome associated with methotrexate by applying data from experiments designed in MESCA. These data are likely to provide new, novel and potentially very important observations on adverse effects associated with a known human teratogen. Julie is currently engaged in completing the internal pre validation efforts supporting the MESCA assay and is completing her Masters thesis work. Julie’s research interests include molecular pathways of development and how they relate to mechanisms of teratogenicity. Julie is an associate member of the Teratology Society and a Member of the Mid Atlantic Teratology Society.

Julieta Panzica-Kelly, B.S.
Bristol-Myers Squibb
Discovery Toxicology
Pharmaceutical Candidate Optimization
311 Pennington Rocky Hill Rd
Pennington, NJ 08534-2130

Karen Augustine received her B.S. in biology from Stockton College in 1990 and Ph.D. in Cell Biology and Anatomy at the University of North Carolina at Chapel Hill in 1994. Her doctoral research established approaches to evaluate gene function by antisense oligonucleotide knockdown in cultured rodent embryos. With these methods, she evaluated functions and interactions of wnt signaling in axial development. She was a post doctoral fellow in Discovery at Amgen from 1994-1997 where she was engaged in gene discovery and generation of transgenic animal models and conducted expression and functional characterization of novel targets associated with cancer and diabetes. She joined SmithKline Beecham/GSK in 1997 where she headed the Molecular Teratology group. In 2002, she joined BMS as the Associate Director of the Reproductive Toxicology group and is currently a Research Fellow of the Investigative Reproductive Toxicology group in Discovery Toxicology. Her current research focuses are in the establishment of methods and strategies to proactively assess teratogenicity profiles of discovery compounds as well as oversee studies associated with evaluating mechanisms of teratogenicity. She has been engaged in a number of professional society stewardship activities including the editorial board for The New Anatomist and Publications Committee for the journals, Teratology and Birth Defects Research. She was recently a member of Council for the Teratology Society and a Member at Large for the Health and Environmental Sciences Institute (HESI). In association with HESI, her group, as well as a number of colleagues in industry, are currently engaged in a cross-Pharma pre validation effort to optimize and further evaluate the zebrafish teratogenicity assay for future use across sectors.

Karen Augustine-Rauch, Ph.D.
Bristol-Myers Squibb
Discovery Toxicology
Pharmaceutical Candidate Optimization
311 Pennington Rocky Hill Rd
Pennington, NJ 08534-2130
Email: karen.augustine@bms.com


Formalized in vivo study designs were incorporated into reproductive safety assessment of pharmaceutical and environmental agents in the mid 1960s soon after the discovery that the sedative, thalidomide, had caused severe limb and craniofacial malformations in about 10,000 European babies (Franklin, 1962; Gillis, 1962; Lenz & Knapp, 1962). Due to the complex nature of the embryo-fetal development as well as maternal-fetal interactions during gestation, reproductive safety assessment has historically involved in vivo testing in at least two species. Under the International Conference on Harmonisation (ICH) guidelines, most medicinal products for teratogenicity testing require three studies: fertility and early embryonic development; pre-and postnatal development, including maternal function; and embryo-fetal development (ICH-S5, 2005). These studies are very expensive and time-consuming, and require extensive use of animals.

In the past two decades, there has been increasing focus on seeking in vitro alternatives for assays that can provide a less expensive and higher throughput means to assess compounds and reduce animal use. Various in vitro developmental toxicity assays have been explored and developed over this period (Bournias-Vardiabasis & Teplitz, 1982; Fort, et al., 1998; Raabe, et al., 1998; Fort, et al., 2004).

Since the early 1990s, the European Centre for the Validation of Alternative Methods (ECVAM) has had a leading role in identifying promising developmental toxicity assays. With collaboration between academic expert laboratories and industry, ECVAM has led an effort in the selection of test compounds and validation activities. Three rodent-based developmental toxicity assays including the mouse embryonic stem cell test (EST), the rodent micromass assay (MM), and rodent whole embryo culture (rWEC) were considered as promising alternative tests for developmental toxicity assays and were validated throughout 5 years of intensive studies among several laboratories in Europe (Spielmann, et al., 2001; Genschow, 2002). Such accomplishments have been successful as a means to address increasing pressures from animal rights and legislative changes, such as the European Union’s Registration, Evaluation, Authorisation and Restriction of Chemical substances legislation (REACH), which will encourage/require replacement of in vivo testing with in vitro approaches for safety assessment of cosmetics and some consumer products (Lilienblum, et al., 2008).

Efforts continue to improve these assays and explore additional alternatives such as the zebrafish embryo-larva model. In this review, the current state of three in vivo developmental toxicity assays that are commonly in use will be described: the rat whole embryo culture (rWEC) assay, the zebrafish embryo-larva developmental toxicity assay and the embryonic stem cell test (EST). The advantages and limitations of each assay, current efforts for improvements, and future direction of the assays will be discussed in this essay.

A Review of Developmental Toxicology Assays of Current Interest to the Field
Rodent Whole Embryo Culture

Over the past 30 years, the rodent whole embryo culture system has been developed, optimized, and validated as an in vitro teratogenicity assay. The method was first developed to culture rat embryos by New’s group at Cambridge University in England in the early 1970s (New, 1978). A method to support culture of mouse embryos soon followed in the mid-1970s (Sadler, et al. 1982). The original application was to develop a rodent embryo culture method to function as an in vitro approach to examine chemicals that have teratogenic potential, in lieu of testing in vivo (New, 1978). The basic concept of this system is to culture early organogenesis-stage rodent embryos (late headfold- early somite stage) in rotated culture bottles, typically over a 48 hour period, although the cultures can be extended for up to 72 hours. The culture medium is supplemented with serum and the culture bottles are constantly supplied with specific percentages of oxygen over the culture period. Test compounds are added into the culture medium at the beginning of the culture. At the end of the culture, embryos are examined and evaluated for abnormalities of developmental morphology and overall embryonic growth after treatment with test compounds.

By the 1980s, the WEC technique had been widely used in many groups for different applications including as teratogenicity screens (Schmid, et al., 1981; Sadler, et al., 1982; Schmid, et al., 1983; Cicurel & Schmid, 1988), mechanistic studies in teratogenesis as well as studies of embryo physiology, development, and organogenesis (Ellington, 1983; New, 1983; Steele, et al., 1983; Ellington, 1987; Hunter, et al., 1987; Cockroft, 1988; Fujinaga, et al., 1988). During this time, various methods were used in different groups to evaluate ontogeny and mechanism of abnormal embryonic development. In order to quantitate embryonic growth and development, Brown and Fabro (1981) developed a score system which provided a more accurate measure of morphological development. In this system, six developmental stages of each feature are defined and scored from 0 to 5. The total score for an individual embryo is taken as a total morphological score (TMS).

From 1996 to 2000, ECVAM and the European Teratology Society (ETS) organized an extensive intra-laboratory validation of three in vitro embryotoxicity tests to assess general concordance and provide guidance on which assay could be considered for inclusion in regulatory testing in the European Union (Spielmann & Liebsch, 2001; Genschow, et al., 2002). The post-implantation rat WEC assay was included in the validation study along with two other assays: the rat micromass test using primary cultures of dissociated limb bud cells, and the embryonic stem cell test using the mouse D3 embryonic cell line (Genschow, et al., 2004). In the rat WEC, rat embryos were prepared with the visceral yolk sac and ectoplacental cone intact, and cultured for 2 days with various concentrations of compounds. The embryos were then morphologically assessed using the TMS system based on developmental staging parameters (Brown & Fabro, 1981). In addition, the rate of malformation and dead embryos of each test group were recorded and analyzed to determine the concentration of no effect (ICnoec), concentration of 50% effect (IC50), and concentration of maximum effect (ICmax). The IC50 cytotoxicity information was also obtained from each test compound by using mouse fibroblast cell line, NIH3T3 cells. In comparisons among the three assays, the rat WEC assay presented the highest cumulative concordance for correctly classifying 80% of the 20 test compounds, which included non, weak, and strong in vivo teratogens (Genschow, et al., 2004).

The rWEC assay is considered a well established in vitro model for identifying and characterizing teratogenic properties of test compounds. A primary advantage of the WEC assay is that it represents a whole organismal model system of the test species that is most commonly used in vivo for developmental toxicity evaluation. In this regard, it substantially reduces the number of animals required for toxicological testing. For instance, a standard rodent embryo-fetal development study to support regulatory filing of pharmaceutical agents, requires approximately 22 dams per dose group; typically at least 88 animals are utilized for this evaluation supporting 3 dose groups and a vehicle control group (ICH guidelines, S5(R2), Nov. 2005 ). In contrast, only 6-8 dams are required to support evaluation of a compound in WEC, encompassing 3 test concentrations and a vehicle group with 10 embryos per treatment group.

Because of its in vitro nature, WEC also provides well controlled experimental conditions and the embryos can be readily accessed and manipulated during the time course of the culture period. Furthermore, specific properties of the compounds can be evaluated. For instance, if the compound produces metabolites in vivo, developmental toxicity can be determined empirically by evaluation of synthesized metabolites in the assay. Rodent or human-specific metabolites matched to in vivo plasma concentrations can be incorporated into WEC experimental designs as a means to compare the drug effects directly with human parameters to better assess and characterize relative teratogenic liabilities against pharmacokinetic differences between humans and rodent test species.

The limitations of the WEC assay include the relatively short timespan during which whole rodent embryos can be cultured (2-3 days during the organogenesis period). This developmental window represents the most sensitive period for teratogenic insult associated with a majority of known teratogens. However, although the critical window of sensitivity of some teratogens may cover this period, the insult may not manifest morphologically until a later time in embryonic development. For instance, compounds that affect neural crest populations active during the organogenesis period may produce malformations that are not visually apparent during the early organogenesis period. Examples include compounds that may cause cleft lip/palate, such as various anti-seizure medications, or compounds that produce crest-related conotruncal septation defects, such as endothelin receptor antagonists (Asada, et al., 1997; Spence, et al., 1999; Treinen, et al., 1999). Lack of maternal-fetal interactions or maternal influences during drug testing is another limitation that this in vitro assay can not overcome in comparison to in vivo studies. Finally, because WEC requires the use of whole embryos, this assay entails euthanasia of the dams. In context of supporting the Three Rs (reduction, refinement and replacement) in animal welfare, that attribute should be considered against the goal of the experiment to determine, in an ethical context, whether WEC is essential or whether an alternative in vitro model that does not require animal sacrifice would be sufficient.

Zebrafish Embryo Culture

Fish, in both adult and embryo-larva form, have had a long history of being useful in ecological toxicology testing due to their viability status and morphological integrity serving as sensitive indicators of water supplies tainted with pollutants/toxicants. The fish model was subsequently discovered as a valuable model for developmental genetics research. In the 1980s, Joseph Streisinger, a bacteriophage geneticist who studied transcription and protein production in Escherichia coli, identified the zebrafish as a promising vertebrate model for developmental genetics due to the simplicity associated in maintaining and breeding the adults and the rapid development of the embryos. Streisinger and colleagues conducted seminal studies that pioneered the transgenic animal field, which ultimately enabled manipulation of the genome and subsequent evaluation of phenotype in various species.

Such accomplishments included development of in vitro fertilization techniques in zebrafish that enabled haploid embryo development and the generation of homozygous mutant lines of fish (Streisinger, et al., 1981). This work lead to additional ground breaking efforts in the 1990s where researchers adapted mutagenesis strategies to the zebrafish and established the N-ethyl-N-nitrosourea (ENU)-based mutagenesis approach to generate zebrafish mutants that enabled phenotypic changes to be readily mapped to genotype, which has led to thousands of characterized mutants (Driever, et al., 1996; Haffter, et al., 1996). With these advances, the zebrafish has become increasingly popular as a model system for developmental biologists and geneticists and the entire zebrafish genome has recently been sequenced (Sanger Institute, 2007).

The earliest application of using the fish embryo for toxicology-based testing has remained relatively unchanged until recently, when a number of laboratories began studying the zebrafish embryo-larva as a potential model for developmental toxicology assessment (Chapin, et al., 2008). Dovetailing with the developmental genetics field, the zebrafish has gained attention as a developmental toxicology model due to the relative ease in husbandry and breeding of the adults and in culturing the externally fertilized eggs, as well as the rapid development of the embryo-larva—the entire organogenesis period is complete by 72 hours post fertilization and can be readily observed and/or manipulated in culture.

The zebrafish embryo-larva is generally transparent for several days, which facilitates morphological assessment of various structures and organ systems. In addition, much of the embryological processes and molecular pathways associated with zebrafish organogenesis are conserved between the teleost and mammalian species (Knaut, et al., 2002; Nusslein-Volhard, et al., 2002). Together with the comprehensive sequence characterization of the genome as well as a growing database of mutations on both the genotype and phenotypic level, the zebrafish has much promise as a valuable model for studying the mechanisms of teratogenicity of selective developmental toxicants. In this regard, a malformation “phenotype” produced by a test compound could be potentially linked to altered genes or pathways by evaluating existing databases describing zebrafish mutant phenotypes and their associated genetic alteration.

In 2007, the first Developmental Toxicology Assay Workshop supported by the Developmental and Reproductive Toxicology (DART) subsection of the Health and Environmental Science Institute (HESI) was held, focusing on the current and future state of the EST, WEC and Zebrafish as developmental toxicology assays. The use of the zebrafish model was discussed extensively and some of the highlights of these discussions are summarized in this essay (as well as in Chapin, et al., 2008). Currently only a limited number of validated zebrafish developmental toxicology assays have been published and a small number of biotech companies commercially offer zebrafish-based teratogenicity assays (Brannen, et al., 2008). For the most part, these developmental toxicology assays are reported to have very good to excellent concordance, typically reporting between 80-90% total concordances. Most have used chorinated embryos and have reported low false negative rates (in vitroassay (Chapin, et al., 2008). At Bristol-Myers Squibb, we have developed a zebrafish assay that uses dechorinated embryos and we have had similar results (Brannen, et al., submitted). Each respective laboratory has unique protocols, but in general these laboratories typically calculate a ratio that is defined as the concentration causing general toxicity (typically the concentration causing 50% lethality (LC50) against a concentration that produces either the lowest or no adverse effect (LOAEC or NOAEC, respectively); then a threshold value is established that optimally classifies teratogens from non teratogens. In our hands, we have found the LC25 ratio against the embryo-larva NOAEL as a robust prediction model, where a ratio ≥10 classifies compounds as positive for teratogenicity (Chapin, et al., 2008; Brannen, et al., submitted).

One area that would benefit the field is reaching consensus related to the criteria for defining death. A number of groups defined lethality as lack of evidence of a heartbeat. However, the zebrafish embryo-larva can potentially remain viable for up to 24 hours post cessation of heart beat. Such effects can complicate the use of LD50 or LD25 values since viability measurements are essential. In our own experience, zebrafish embryo-larva that die during the culture period rapidly undergo necrosis and frequently, there is little evidence of tissue found in the culture well. We defined death as absence of a heart beat, and/or evidence of necrosis and severe dysmorphology throughout the embryo.

There are differences in opinions related to whether it is necessary to remove the chorion, which is a shell-like coating that protects the embryo. We have removed the chorion to facilitate compound delivery; however this is a time-consuming process that would not be desirable for a high throughput assay design. Some laboratories include drug content analysis in their assays using chorinated embryos and have reported a substantial difference in respective compound uptake by the embryos, where such evaluation is recommended to confirm results, especially in cases where a negative (non teratogen) classification is calculated. Compound uptake analysis has the additional benefit of confirming efficient delivery of the compound and aiding in interpreting relative teratogenic potency.

Another substantial difference between zebrafish protocols is the stage to initiate compound treatment as well as the stage to morphologically assess the embryo-larva. Some groups, such as Phylonix initiate compound treatment at 24 hours post fertilization (hpf), the equivalent of early organogenesis. Other groups, such as those from Danio Labs (now Summit/Evotec) initiate compound treatment at the 2-cell stage (early cleavage stage). Concern has been raised regarding this latter timing for compound administration as maternal gene transcription dominates the early embryo (up to 4-6 hpf), which may influence teratogenic response associated with early treatment.

We initiate compound treatment at ~5-6 hpf, which is the equivalent of early gastrulation and reflects the equivalent stage when mammalian animal models are treated with compound in reproductive safety assessment evaluations. There was general agreement regarding the potential importance of covering the entire window of organogenesis in a teratogenicity assay, which would suggest that treatment in the zebrafish embryos should commence at the gastrulation stage (approximately 5-10 hpf) and the embryo remain exposed to compound throughout the organogenesis phase. As such, most laboratories morphologically assess the embryos when organogenesis is completed—sometime at or beyond 72 hpf when they are at the larva stage.

We have conducted morphological evaluations at approximately 30 hpf and 5 dpf (days post fertilization) and have found the dysmorphology NOAELs identified at 5 dpf to produce superior concordance (Brannen, et al., submitted). Particular morphological structures as well as how malformations are identified vary between laboratories as well. Indeed the field would benefit by establishment of a consortium to harmonize an assay protocol and evaluate this protocol with a large set of compounds including intra laboratory assessment to determine repeatability.

Some additional areas of assay refinement include evaluation of zebrafish strain differences and selection of a preferred strain to support a harmonized assay. Establishing means to monitor and control pH in the test media is relevant since the zebrafish embryo-larva are very sensitive to pH changes and free acid or base forms of compounds at sufficient concentrations could alter the pH of the test media.

Consensus surrounding the rationale for dose selection would also be beneficial. In the DART symposium round table discussions, limited compound solubility was the typical basis for top dose concentration. It was generally agreed that compound solubility limitation was one of the most significant draw backs of the zebrafish assay. However, an alternative delivery approach could be the use of microinjecting the compounds directly into the yolk ball. If such approaches were utilized, consensus surrounding test concentration ranges would be valuable to incorporate in a harmonized assay. Other areas that need to be better characterized include historical tracking and recording of background anomalies associated with respective wild type strains of fish utilized in these assays as well as DMSO vehicle historical control data.

The Embryonic Stem Cell Assay

The embryonic stem cell test was developed with a primary focus to reduce the use of animals used in characterizing compounds for teratogenic potential (Spielmann, et al., 1997; Spielmann, 2005). In the EST, mouse embryonic stem cells are allowed to aggregate and differentiate into embryo-like structures called embryoid bodies (EBs); in the presence of compound, and 10 days later, the cells are visually scored for beating cardiomyocytes. These EBs undergo gastrulation and eventually differentiate into the different types of cells that make up an organism. The test uses discriminant analysis as a prediction model to distinguish between strong, weak and non-embryotoxic chemicals. After a well designed inter-laboratory validation lead by ECVAM, the EST was determined to be 73%, 69%, and 100% accurate in delineating none, weak and strong teratogens, respectively. The test’s limitation is differentiating between weak and non-teratogens (Genschow, 2004).

For the past few years, there has been a general interest in the field to shorten the duration of the EST, as well as to enhance the throughput capabilities of the assay by incorporating molecular endpoints (zur Nieden, et al., 2004; Festag, et al., 2007; van Dartel, et al., 2008). For example, one laboratory has been utilizing flow cytometry techniques in order to detect changes in myosin heavy chain and α-actinin genes after compound treatment on day 7 of the EST. The group found that these endpoints were just as sensitive as the conventional microscopic scoring of cardiomyocyte differentiation (Seiler, et al., 2004).

In this same laboratory, additional quantitative RT-PCR endpoints have been developed to detect changes in gene expression at different time points, and in different cell types (osteoblasts, chondrocytes and neuronal cells). These experiments have been successful in detecting the types of defects produced by specific teratogens by analyzing gene expression changes in the types of cells that are affected by these specific compounds (zur Nieden, et al., 2004). Another laboratory has investigated changes in gene expression by using microarray analysis after compound treatment. This group focused on detecting changes between days 3 and 4 of the assay, and found 43 genes that were deregulated after monobutyl phthalate treatment that may be indicative of embryo toxicity (van Dartel, et al., 2008). Although this is promising data, there was only one compound evaluated in this analysis and teratogens act via many different mechanisms that may not be covered by the specific gene list identified in this study.

In our hands, genes were identified via microarray analysis as well as by literature searches focused on potential functional relevance in early embryonic development. The dynamics of transcriptional expression was assessed during the course of the 10 day assay. From this data, a panel of genes and one time point was selected to validate the assay with known in vivo teratogens and non-teratogens using quantitative RT-PCR. A statistically significant prediction model was developed in order to test pharmaceuticals in a high throughput manner.

Some pros of working with the EST include: the reduction of animals and trained personnel required for hazard identification as well as the costs associated with in vivo animal testing, and the ease of possible automation of the assay as opposed to whole animal systems. The cons of working with the EST include: the labor intensive hanging drop culture, the time-consuming scoring of the beating EBs, and the duration of the assay. The EST is constantly being improved providing resolutions to these negative attributes, thus making the EST a promising approach to test embryo toxicity of chemicals in a high throughput method.

Future Directions

Efforts are being conducted by various laboratories including our group to continue to develop and improve the three assays described in this essay. The current state of improvements applied to these assays and potential future direction of this work are described as follows:

Rodent Whole Embryo Culture Assay

We have engaged in efforts to further optimize the WEC assay in context of improving concordance of the assay and reduce animal requirements. A novel morphological score system called the Dysmorphology Based Score System (DBSS) has been developed to identify dysmorphology and define the respective degree of severity, so that more subtle or selective structure-specific abnormalities can be delineated to identify compounds that induce structure or organ-selective effects (score system and examples of application described in Augustine-Rauch, et al., 2004a and 2004b).

We are in the process of developing a simplified experimental design of the WEC assay to reduce time, labor and animal requirements. This method eliminates the need to conduct the IC50 cytotoxicity assay in NIH3T3 cells and reduces the number of required test concentration groups to one, due to the use of novel statistical model algorithms (unpublished data). To this end, using the DBSS in assessment of approximately 40 compounds of known in vivo outcome (teratogens and non-teratogens), statistical prediction models have been developed to support an abbreviated teratogenicity screen assay which maintain a similar level of concordance as the original ECVAM assay.

It is our intention to publish a detailed description of the DBSS and the results of the statistical model study and make the methods publically available in the near future. It will be beneficial to continue to explore improvements in the WEC assay from the perspective of refining the assay design and/or incorporating statistical models as well as conducting broader validation efforts by utilizing larger test sets of compounds. Such applications may lead to an in vitro assay of sufficient robustness that could be considered by regulatory agencies as an acceptable routine in vitro testing alternative in the future.

Zebrafish Assays

Researchers who have worked with zebrafish embryo-larva as a developmental toxicology model generally are in agreement that it presents considerable promise as a system to test teratogenic liability of compounds. At this point, a number of researchers have completed assessments with various zebrafish assay study designs and the field would benefit by having a working group evaluate available protocols and datasets and work together to generate a protocol that is viewed to be practical for execution and have robust concordance metrics using a similar approach as was conducted by ECVAM (Spielmann, et al., 2001). Assessment across a number of laboratories would also be beneficial to determine inter laboratory/inter-experimental variability. Such a consortium effort is currently underway, with results intended to be shared within a forum such as the HESI DART committee to facilitate harmonization and broader application amongst industry.

Embryonic Stem Cell Assay

We have been successful in generating a molecular-based embryonic stem cell test that is shorter in duration (4 days) and measures transcriptional response of a small number of gene targets to support a statistical prediction model that has comparable total concordance (78%) to the EST. Thus, the assay has potential to be less labor intensive and time consuming. Approximately 40 compounds were enrolled in our efforts to validate this assay internally. Therefore, an inter-laboratory validation of the molecular EST with a larger test set should be completed in order to determine the assay inter-experimental and inter-laboratory variability. The results of this large validation could then be compared to the results of the other tests described in this essay in order to determine how these assays can be used together to test new compounds for their teratogenic potential.


Much progress has been made in generating in vitro alternative approaches for supporting reproductive safety assessment of test compounds. Devising a recommended testing strategy awaits a comparative assessment of the concordance and precision metrics of the respective assays as well as a consideration of the bioethics of animal use associated with these assays. It is not anticipated that these in vitro approaches would completely replace the need to assess reproductive safety profiles of test compounds in vivo. However, such approaches may be useful as a means to proactively optimize potential drug candidates prior to moving forward into development, which may streamline the number of compounds requiring in vivo safety assessment. Also, a robust in vitro developmental toxicology assay may be considered as a replacement of specific in vivo studies (such as in accord with REACH legislation) or as a trigger for comprehensive in vivo reproductive toxicology assessment to better safeguard women of child-bearing potential populations from exposure to potential teratogens during exploratory clinical studies.


©2009 Cindy X. Zhang, Julie Panzica-Kelly, and Karen Augustine-Rauch

Asada, H., Kawamura, Y., et al. (1997). Cleft palate and decreased brain gamma-aminobutyric acid in mice lacking the 67-kDa isoform of glutamic acid decarboxylase. Proc. Natl. Acad. Sci. U.S.A. 94(12), 6496-6499.

Augustine-Rauch, K.A., Zhang, Q., Kleinman, M., Lawton, R. & Welsh, M.J. (2004a). A study of vehicles for dosing cultured rodent embryos with non-aqueous soluble compounds. Reprod. Toxicol. 18, 391-398.

Augustine-Rauch, K.A., Zhang, Q.J., Leonard, J.L., Chadderton, A., Welsh, M.J., Rami, HK, et al. (2004b). Evidence for a molecular mechanism of teratogenicity of SB-236057, a 5-HT1B receptor inverse agonist that alters axial formation. Birth Defects Res. Part A Clin. Mol. Teratol. 70, 89-807.

Bournias-Vardiabasis, N. & Teplitz, R.L. (1982). Use of Drosophila embryo cell cultures as an in vitro teratogen assay. Teratog. Carcinog. Mutagen. 2(3-4), 333-341.

Brannen, K.C., Panzica-Kelly, J., Charlap, J.H. & Augustine-Rauch, K.A. (2008). Zebrafish: A non-mammalian model of developmental toxicology. Chapter 8. Developmental Toxicity in the Target Organ Series. Raven Press; B. Abbott and D. Hansen, eds.

Brannen, K.C., Panzica-Kelly, J., Danberry, T., & Augustine-Rauch, K. (Submitted). Validation of a Zebrafish embryo teratogenicity assay utilizing a quantitative prediction model.

Brown, N.A. & Fabro, S. (1981). Quantitation of rat embryonic development in vitro: A morphological scoring system. Teratology. 24(1), 65-78.

Chapin, R., Augustine-Rauch, K., et al. (2008). State of the art in developmental toxicity screening methods and a way forward: A meeting report addressing embryonic stem cells, whole embryo culture, and zebrafish. Birth Defects Res. B Dev. Reprod. Toxicol. 83(4), 446-456.

Cicurel, L. & Schmid, B.P. (1988). Postimplantation embryo culture for the assessment of the teratogenic potential and potency of compounds. Experientia. 44(10), 833-840.

Cockroft, D.L. (1988). Changes with gestational age in the nutritional requirements of postimplantation rat embryos in culture. Teratology. 38(3), 281-290.

Driever, W., Solnica-Krezel, L., et al. (1996). A genetic screen for mutations affecting embryogenesis in zebrafish. Development. 123, 37-46.

Ellington, S.K. (1983). Effects of calcium on the growth of the rat conceptus during organogenesis in vitro. J. Reprod. Fertil. 67(2), 327-334.

Ellington, S.K. (1987). In vitro analysis of glucose metabolism and embryonic growth in postimplantation rat embryos. Development. 100(3), 431-439.

Festag, M., Viertel, B., et al. (2007). An in vitro embryotoxicity assay based on the disturbance of the differentiation of murine embryonic stem cells into endothelial cells. II. Testing of compounds. Toxicol. In Vitro. 21(8), 1631-1640.

Fort, D.J., Stover, E.L., et al. (1998). Phase III interlaboratory study of FETAX, Part 2: Interlaboratory validation of an exogenous metabolic activation system for frog embryo teratogenesis assay—Xenopus (FETAX). Drug Chem. Toxicol. 21(1), 1-14.

Fort, D.J., Rogers, R.L., et al. (2004). Comparative sensitivity of Xenopus tropicalis and Xenopus laevis as test species for the FETAX model. J. Appl. Toxicol. 24(6), 443-457.

Franklin, A.W. (1962). Thalidomide babies. Memorandum from the British Paediatric Association. Br. Med. J. 2(5303), 522-524.

Fujinaga, M., Mazze, R.I., et al. (1988). Rat whole embryo culture: An in vitro model for testing nitrous oxide teratogenicity. Anesthesiology. 69(3), 401-404.

Genschow, E., Spielmann, H., et al. (2002). The ECVAM international validation study on in vitro embryotoxicity tests: Results of the definitive phase and evaluation of prediction models. European Centre for the Validation of Alternative Methods. Altern. Lab. Anim. 30(2), 151-176.

Genschow, E., Spielmann, H., Scholz, G., Pohl, I., Seiler, A., Clemann, N., et al. (2004). Validation of the embryonic stem cell test in the international ECVAM validation study on three in vitro embryotoxicity tests. Altern. Lab. Anim. 32(3), 209-244.

Gillis, L. (1962). Thalidomide babies: Management of limb defects. Br. Med. J. 2(5305), 647-651.

Haffter, P., Granato, M., et al. (1996). The identification of genes with unique and essential functions in the development of the zebrafish, Danio rerio. Development. 123, 1-36.

Hunter, III, E.S., Sadler, T.W. & Wynn, R.E. (1987). A potential mechanism of DL-beta-hydroxybutyrate-induced malformations in mouse embryos. Am. J. Physiol. 253(1 Pt 1), E72-80.

Knaut, H., Steinbeisser, H., et al. (2002). An evolutionary conserved region in the vasa 3’UTR targets RNA translation to the germ cells in the zebrafish. Curr. Biol. 12(6), 454-466.

Lenz, W. & Knapp, K. (1962). Thalidomide embryopathy. Arch. Environ. Health. 5, 100-105.

Lilienblum, W., Dekant, W., et al. (2008). Alternative methods to safety studies in experimental animals: Role in the risk assessment of chemicals under the new European Chemicals Legislation (REACH). Arch. Toxicol. 82(4), 211-236.

Nagel, R. (2002). DarT: The embryo test with the Zebrafish Danio rerio—a general model in ecotoxicology and toxicology. ALTEX. 19, Suppl 1, 38-48.

New, D.A. (1978). Whole-embryo culture and the study of mammalian embryos during organogenesis. Biol. Rev. Camb. Philos. Soc. 53(1), 81-122.

New, D.A. (1983). In vitro culture of embryo and fetus. Proc. Annu. Symp. Eugen. Soc. 19, 163-176.

Raabe, M., Flynn, L.M., et al. (1998). Knockout of the abetalipoproteinemia gene in mice: Reduced lipoprotein secretion in heterozygotes and embryonic lethality in homozygotes. Proc. Natl. Acad. Sci. U.S.A. 95(15), 8686-8691.

Sadler, T.W., Horton, W.E., et al. (1982). Whole embryo culture: A screening technique for teratogens? Teratog. Carcinog. Mutagen. 2(3-4), 243-253.

Schmid, B.P., Goulding, E., et al. (1981). Assessment of the teratogenic potential of acrolein and cyclophosphamide in a rat embryo culture system. Toxicology. 22(3), 235-243.

Schmid, B.P., Trippmacher, A., et al. (1983). Validation of the whole-embryo culture method for in vitro teratogenicity testing. Dev. Toxicol. Environ. Sci. 11, 563-566.

Seiler, A., Visan, A., et al. (2004). Improvement of an in vitro stem cell assay for developmental toxicity: The use of molecular endpoints in the embryonic stem cell test. Reprod. Toxicol. 18(2), 231-240.

Spence, S., Anderson, C., et al. (1999). Teratogenic effects of the endothelin receptor antagonist L-753,037 in the rat. Reprod. Toxicol. 13(1), 15-29.

Spielmann, H., Pohl, I., Doering, B., Liebsch, M. & Moldenhauer, F. (1997). The embryonic stem cell test, an in vitro embryotoxicity test using two permanent mouse cell lines: 3T3 fibroblasts and embryonic stem cells. In Vitro Toxicology. 10, 119-127.

Spielmann, H., Genschow, E., et al. (2001). Preliminary results of the ECVAM validation study on three in vitro embryotoxicity tests. Altern. Lab. Anim. 29(3), 301-303.

Spielmann, H. & Liebsch, M. (2001). Lessons learned from validation of in vitro toxicity test: From failure to acceptance into regulatory practice. Toxicol. In Vitro. 15(4-5), 585-590.

Spielmann, H. (2005). Predicting the risk of developmental toxicity from in vitro assays. Toxicol. Appl. Pharmacol. 207, S375-S380.

Steele, C.E., New, D.A., et al. (1983). Teratogenic action of hypolipidemic agents: An in vitro study with postimplantation rat embryos. Teratology. 28(2), 229-236.

Streisinger, G., Walker, C., et al. (1981). Production of clones of homozygous diploid zebra fish (Brachydanio rerio). Nature. 291(5813), 293-296.

Treinen, K.A., Louden, C., et al. (1999). Developmental toxicity and toxicokinetics of two endothelin receptor antagonists in rats and rabbits. Teratology. 59(1), 51-59.

van Dartel, D.A., Pennings, J.L., et al. (2009). Early gene expression changes during embryonic stem cell differentiation into cardiomyocytes and their modulation by monobutyl phthalate. Reprod. Toxicol. 27(2), 93-102.

zur Nieden, N.I., Kempka, G., et al. (2004). Molecular multiple endpoint embryonic stem cell test—a possible approach to test for the teratogenic potential of compounds. Toxicol. Appl. Pharmacol. 194(3), 257-269.