Developmental Toxicity Effects Of Thalidomide Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

In the past 20 years more than 30 diverse platforms has been recommended for assessing developmental toxicity (DT) with the boundary of primary cell cultures, non mammalian embryos, and mammalian embryos or primordial. Invention of embryonic stem cells (ESC) explained the usage of blastocyst derived mouse ESC for in-vitro embryotoxicity testing Hofer, T. et al. (2004)(2). The rat primary embryonic cells based micro mass test was replaced with the differentiating ESC for the prediction of teratogenicity of test compounds Newall, D. R. et al. (1994)(3). In the development of mammalian in-vitro system for teratogenicity testing system, cytotoxicity end point was evaluated with dimethylthiazol- diphenyl tetrazolinm bromide (MTT) assay and it was concluded that ESC were more sensitive than fibroblast or differentiated cells in response to xenobiotics Laschinski, G. et al. (1991)(4). In terms of development of in vitro embryotoxicity system mouse ESC were differentiated in presence of toxicant such as retinoic acid which augmented skeletal muscle differentiation Heuer, J. et al. (1993)(10). However, when embryonic stem cell test (EST) method introduced a combination of cytotoxicity (IC50 for ESC and 3T3), inhibition concentration for differentiation (ID50) and inhibition of the differentiation of ES cells into contracting myocardial with linear biostatistical prediction model revolutionized the embryotoxicity screening Scholz, G. et al. (1999b), Scholz, G. et al. (1999a)(5, 8). In the further improvement of EST molecular end points such as FACS were introduced to reduce the time consumption and high-throughput screening. Seiler, A. et al. (2004)(6)

To improve the developmental toxicity testing, human embryonic stem cells were introduced for the consideration of safety of patients and consumers in terms of high efficiency in in-vitro toxicological tests and more importantly can avoid the inter species variability. As a first step early developmental markers and cardiac markers were used to determine the toxicity of know toxicant such as retinoic acid and 5 fluro uracil. Adler, S. et al. (2008)(13) But cardiac specific differentiation and analysis of few markers alone not sufficient to make the concrete decision for toxicity determination. Our recent publication regarding multilineage differentiation provided the evidence for the expression of endoderm, ectoderm and mesoderm lineage markers using high sensitive transcriptional profiling and interference of these germ layer markers. These findings will provide substantial evidence to classify the compounds for developmental toxicity Jagtap, S. et al. (2011).

However, in the context of developmental toxicity and species specificity the role of thalidomide is indiscriminate. Thalidomide, a hypnosedative drug causes stunted limb growth (dysmelia) during human embryogenesis Sauer, H. et al. (2000b and is now used to treat leprosy and multiple myeloma Knobloch, J. et al. (2007b. Although the embropathy of thalidomide has been in focus for over three decades,Stephens, T. D. (1988) its recent use in the inhibition of angiogenesis and immunomodulatory behavior Singhal, S. et al. (1999b has gained importance Sauer, H. et al. (2000a);D'Amato, R. J. et al. (1994b. It was clinically proven that thalidomide reduces the TNF-a and interleukin-12 in patients with chronic Crohn s disease Bauditz, J. et al. (2002). Anti angiogenic property of thalidomide is another possibility for the teratogenicity and was reported in bFGF induced-rabbit cornea micro pocket assay D'Amato, R. J. et al. (1994a). Clinical evidence also showed the role of thalidomide in advanced and refractory multiple myeloma due to its anti-angiogenic property. Recently it was deciphered that cereblon was identified as molecular target for thalidomide which forms an E3 ubiquitin ligase complex with DDB1 and Cul4A that is important for limb out growth Ito, T. et al. (2010)(20). Although many molecular mechanisms such as oxidative DNA damage, angiogenesis inhibitor, IGF-1, and FGF-2 antagonist proposed for the teratogenecity Stephens, T. D. et al. (2000a)(33), anti-angiogenic property of thalidomide regained the importance of thalidomide in the treatment of refractory multiple myeloma and significant molecule in pharmaceutical industry Zeng, X. et al. (2006), Singhal, S. et al. (1999a)(12, 21). The original application such as sedative effect of thalidomide was found that enantiomer specific and it was proved that enantiomer R or (+) has exhibited positive sedative effect compared to S or (-) (12). In vivo teratogenecity of thalidomide derivative EM12 was conducted with non human primate callithrix jacchus and found 30ug/kg was effective to induce severe skeletal abnormalities and 10ug/kg was considered no observed adverse effect level (NOAEL), this was the lowest dose reported ever for teratogenecity (7). The limb defects were observed in cynomolgus monkeys in response to 15-20 mg of thalidomide treatment during 26-28 days of gestation and transcriptional studies of 6hr treatment of thalidomide down regulated vasculature development gene ontology Ema, M. et al. (2010)(18). The congenital malformations of thalidomide were found prominently in rabbit but where as in rat and hamsters fetal changes were not observed significantly and in mice no fetal malformations were observed. Among the many species studied rabbit is more optimal to study thalidomide effect due to anatomical differences of placentas and its membrane permeability's Schumacher, H. et al. (1968), Teo, S. K. et al. (2004a), FRATTA, I. D. et al. (1965)(15, 23, and 24). Although many in vivo studies conducted with different concentrations of thalidomide, with best of our knowledge no dose response in vitro studies were performed with either primary cell line or ESCs Ema, M., Ise, R., Kato, H., Oneda, S., Hirose, A., Hirata-Koizumi, M., Singh, A. V., Knudsen, T. B., and Ihara, T. (2010), DELAHUNT, C. S. et al. (1964) (7, 18, and 22).

Our study was commenced to analyze the dose response assessment of thalidomide mediated DT on multilineage differentiated hESC based on the sensitive transcriptional profiling. Also the effect of thalidomide on multilineage differentiated mESCs and hESC and its cross species relationship of gene expression responses correspond to a powerful approach to study the molecular markers for thalidomide in mouse and human embryonic development. The dose response assessment of thalidomide will reveal the lowest and maximal toxicity responses during hESC embryonic development. A comparative analysis of mouse and human EB development in presence of thalidomide reveals the similarity and species specificity of thalidomide mediated teratogenecity in mouse and human. Finally the sensitive common markers for 14 day old EBs in H9 and CGR8 differentiation and markers for the respective treatment of thalidomide in embryonic development will be non-species specific in embryonic development. The residual species specific markers recapitulate the species specificity of thalidomide observed in in-vivo with human and mouse embryonic development.


hESC cell culture conditions and differentiation

H9 hESCs (WiCell, Madison, WI, USA) were cultured in knockout (KO)-DMEM-F12, 20 % KO serum replacement, 1 % non essential amino acids, penicillin (100 units/ml) / streptomycin (100 µg/ml), 0.1 mM β-mercaptoethanol supplemented with 4 ng/ml basic fibroblast growth factor (bFGF)and passaged with mechanical dissociation on irradiated mouse embryonic fibroblasts (MEF). Prior to random differentiation, cells were maintained for five days on hESC-qualified matrix (BD Biosciences, California, USA)-coated 60 mm tissue culture plates (Nunc, Langenselbold, Germany). Cultures were passaged and maintained in feeder-free conditions with conditioned medium Xu, C. et al. (2001 supplemented with 8 ng/ml bFGF. For multi lineage differentiation EBs were prepared as mentioned in our previous publication Jagtap, S., Meganathan, K., Gaspar, J., Wagh, V., Winkler, J., Hescheler, J., and Sachinidis, A. (2011) and EBs were maintained for 14 days on horizontal shaker in respective medium (Figure 1A).

For treatment, Thalidomide (0.01μM, 0.1μM, 1μM, 10μM, 70μM,) (Sigma, Steinheim, Germany) was added on day 0 in the medium, and vehicle DMSO (0.02%) was added to controls. Every alternate day, medium was replaced completely with fresh medium containing drug.

CGR8 culture conditions and differentiation

CGR8 mouse ESCs (ECACC 95011018) were cultured without feeders in GMEM supplemented with 10% FBS, 2mM L-Glutamine, 100units /ml leukemia inhibitory factor and 50μM beta mercaptoethanol in 0.1% gelatin coated flasks. The cells were passaged every alternative day with less than 70% confluence. The differentiation method such as hanging drop method and medium composition were mentioned in previous publication (ref) For treatment 30 μM thalidomide was added from day 0 to day 14. For control 0.001 % DMSO was added.

Cell viability assay

The cytotoxicity assay was performed as described in our previous paper ( Jagtap et al ) After 48 hours of cell seeding, medium was removed completely and fresh medium containing thalidomide (0.0001-100 µM) was added. In control wells, 0.1 % of dimethyl sulfoxide (DMSO) was added as a vehicle control. After 3 days, 20 µl of CellTiter 96® AQueous MTS and phenazine methosulfate (Sigma, Steinheim, Germany) solution was added to the medium as per the manufacturer's instructions. After 1-2 hours incubation at 37 °C the absorbance was measured with TECAN spectrophotometer at 490 nm. The study was performed in three independent experiments. The drug concentration at which 50 % and 10 % inhibition of cell growth was observed, were determined as IC50 and IC10 values respectively.

A colony formation assay was performed to determine cytotoxicity Kreja, L. et al. (2002. Approximately 150 mechanically disrupted Lerou, P. H. et al. (2008 clumps were seeded on matrix-coated (BD Biosciences, California, USA) tissue culture grade 6-well dishes as described before. After 48 hours, medium was removed and fresh medium containing test substances was added. Thalidomide was used in 8 different concentrations (0.0001 µM -100 µM). 0.1 % of DMSO was added as a vehicle control. Every alternate day medium was refreshed. On day 7 colonies were fixed with methanol for 15 minutes and 1 % aqueous crystal violet solution was added for 10 minutes at room temperature. Plates were washed 6 times with water and colonies were counted manually. Numbers of colonies were counted in triplicate and colony growth was calculated relatively to control wells. IC50 values derived from concentration at 50 % and IC10 values derived from 10 % inhibition of colony formation.

Microarray labeling and Hybridization:

To isolate total RNA, samples collected from undifferentiated hESCs, EBs treated with Thalidomide (0.01μM, 0.1μM, 1μM, 10μM, 70μM) and control treated with 0.02% DMSO till day 14 from three independent experiments. For CGR8 experiment samples collected from undifferentiated mouse ESCs, EBs treated with Thalidomide (30μM) and control treated with 0.001% DMSO till day 14 from three independent experiments. Samples were homogenized with Trizol (Invitrogen, Darmstadt, Germany) and RNA was extracted using Trizol and CHCl3 (Sigma, Steinheim, Germany). The total RNA was purified using RNeasy minelute cleanup kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Nanodrop (ND-1000, Thermo-Fisher, Langenselbold, Germany) was used for quantification and the quality of RNA was determined by denaturing agarose gel electrophoresis. For transcriptional profiling Affymetrix Human Genome U133 plus 2.0 arrays were used. All reagents and instrumentation regarding microarrays were acquired from Affymetrix (Affymetrix, Santa Clara, CA, USA, For RNA amplification 100ng total RNA was used with Genechip 3' IVT Express Kit as per the manufacturer's instructions (Affymetrix). For aRNA purification magnetic beads were used and 15 μg of aRNA was fragmented with fragmentation buffer as per the guidelines. For hybridization (Affymetrix HWS kit) 12.5 μg fragmented aRNA was hybridized with Affymetrix Human Genome U133 plus 2.0 arrays along with hybridization cocktail solution and then placed in Genechip Hybridization Oven-645 (Affymetrix) rotating at 60 rpm at 45 °C for 16 h. Affymetrix HWS kit was used for staining and washing according to the affymetrix washing and staining protocol using Genechip Fluidics Station-450 (Affymetrix). The stained arrays were scanned with Affymetrix Gene-Chip Scanner-3000-7G and the quality control analysis were performed with Affymetrix GCOS software.

Data analysis and statistical procedures

The raw dataset was normalized with Quantile normalization method executable with R Bioconductor Bolstad, B. M. et al. (2003. To identify differentially expressed genes a linear model was implemented using R and the LIMMA package Smyth, G. K. (2004. F Statistic was applied to determine the differentially expressed transcripts (DET). The raw P-values were adjusted by Benjamini Hochberg procedure for controlling the false discovery rate (P ≤ 0.05) at probe level. Besides the above mentioned conditions, fold change value with the threshold value ≥ (+ / -) 1.0 was used to filter the significantly expressed transcripts which results 25433 DET. To further short list the DET and to obtain the sigmoid expression curve for the dose response data at 70μM concentration transcripts were filtered with the threshold value ≥ (+ / -) 2.0 resulting 1026 transcripts.

Gene expression and functional annotation analysis

Hierarchical Clustering analysis of the differential expressed transcripts has been performed using uncentered correlation and average linkage Eisen, M. B. et al. (1998 and was displayed using JAVA tree view Saldanha, A. J. (2004. To further investigate the differentially expressed genes, functional annotation and gene ontology (GO) clustering in Database for Annotation, Visualization and Integrated Discovery (DAVID) was used Dennis, G. et al. (2003. These analyses explore the biological process functional annotations for differentially expressed transcripts derived from control 7-days and 14-days untreated (control) EBs and Ara-C treatment. To derive the biological impact, dysregulation of significant genes were annotated based on EASE score enrichment P value (P ≤ 0.01) Hosack, D. A. et al. (2003 and clustering stringency at medium level as per DAVID.

BMD calculations and GO analysis for mean BMD:

For bench mark dose (BMD) identification for thalidomide, BMD express version 1.3 was used. The filtered 1026 transcripts from F statistics were used for BMD calculation, model selection and GO analysis in BMD software. The probe sets determined by the ANOVA (p value ≤ 0.05) were used with different statistical models such as Power, Linear (1°), Polynomial 2°, Polynomial 3° to fit with data. The best model was selected based on the model describing the data with the least complexity. The genes from the BMD calculations were categorized based on the GO classification and the average BMD and the Standard deviation (SD) were calculated for each GO (25)

Drugs and materials

Thalidomide was purchased from Sigma (Steinheim, Germany) and 300 mM stock solution was prepared in DMSO (Sigma, Steinheim, Germany). The final concentration of DMSO 0.02% in drug treatment will be added to untreated sample to normalize the effect of DMSO.


Morphology and Dose response assessment of thalidomide mediated gene expression profiling:

The effect of thalidomide on hESC derived EBs depend on the concentration of thalidomide used to treat the cells as shown in figure 1b. In the absence of thalidomide, called as untreated, hESC derived EBs exhibited a classic EB like appearance when cultured. In addition thalidomide at different concentrations i.e.: 0.01µM, 0.1µM, 1µM, 10µM and 70µM induced a considerable effect on the morphology of the EBs in dose dependent manner. There was no reduction in the size as well as no apoptosis was observed in the EBs till the IC10 concentration was reached. Although an IC50 concentration of 70µM had a profound effect on morphology of the EBs.

Figure 3 shows eisen plot for the dose response for the three germ layer markers including markers for stress response. We observed a dose dependent down regulation of these markers with IC50 concentration repressed the markers up to 2 fold change. The common signatures between the thalidomide doses were analyzed with 5 set venn diagram. The comparative analysis (figure 2a) for the common genes between different concentrations of thalidomide revealed no unique transcripts that were universally dysregulated. COL1 A1 was found to be uniquely dysregulated between 0.01µM -1µM and HES5 between 0.1µM -70µM concentrations. Mesoderm markers COL6A3 and SCN2A were present in all concentrations except 0.1µM. MAP1B, the microtubule associated protein 1B was down regulated in 0.1µM, 10µM and 70µM (t statistics), however F statistics results showed MAP1B consistently down regulated in all the concentrations. A maximum of 92 transcripts were commonly dysregulated between 10µM and 70µM groups. A complete list of transcripts is provided in supplementary table XXX. To find the global expression pattern of the 2D principal component analysis (PCA) (Figure 2b) was performed, it was found out that pluripotent hESC was distinctly away from treated and untreated samples. The lowest concentrations such as 0.01µM, 0.1µM, 1µM closely placed with control which showed low number of DET in figure 2c. The long distance of 10µM and 70µM corresponding to the high number of DET. Further insight into the transcripts for the dose response, sensitive empirical t statistics was performed and we observed an increase in the number of dysregulated transcripts as the concentration of thalidomide increases. Figure 2c shows the number of dysregulated transcripts at fold change 2 and FDR p value ≤ 0.05. The maximum dysregulated transcripts observed at IC50 concentration were 803 up regulated genes and 223 downregulated genes. The two lowest concentration, 0.01µM and 0.1µM, showed a plateau, where the number of transcripts between them did not differ significantly. At 0.01µM concentration of thalidomide dysregulated sensitive markers such as COL1A1, SRGN, COL6A3, SCN2A, and SPOCK3 and among these 5, mesoderm markers such as COL1A1 and COL6A3 were down regulated.

Gene ontology and toxicogenomics signatures for thalidomide at various concentrations.

To determine the effect of thalidomide across the doses F statistics analysis was performed. At fold change 1 and FDR p value ≤ 0.05, 25433 transcripts (supplementary table XXX) were filtered. To find out the global expression pattern of these transcripts hierarchical clustering was performed. The 25433 transcripts were filtered at SD gene vector ≥1 and the filtered 14613 transcripts were clustered with uncentered correlation and average linkage clustering (Supplementary figure 1). To obtain the sigmoid dose response curve and to further explore the dose response genomic data the transcripts were further filtered at fold change 2 in the highest concentration 70µM which results 1026 transcripts (supplementary table XXX). To analyze the expression pattern of the markers across the concentration average linkage clustering was performed with an uncentered correlation without filtering the transcripts (Figure 4). The cluster analysis reveals the dose dependent repression of the transcripts for the down regulated transcripts. To find the functional knowledge of these transcripts DAVID functional annotation was performed for up and down regulated transcripts.

Figure 3 summarizes the up regulated GO observed were for cell communication, signal transduction, cell differentiation and cell development and some significant down regulated GO such as developmental process, anatomical structure development present in parent BP, and response to stress, limb morphogenesis and skeletal development were present in child BP.

Bench mark dose Assessment and GO analysis for the determination of mean BMD:

The genes from the BMD analysis categorized based on the respective GO classifications. Less than 10 genes present in the respective GOs were filtered. The GO classification (Table 1a) found that mammary gland development is most sensitive BP at the concentration of 0.88±1.19 µM (Mean BMD±SD). Subsequently skeletal system morphogenesis, limb development, embryonic skeletal system development was affected at the concentration of 2.8±4.99 µM, 3.04±4 µM, 3.1±4.95 µM respectively. The BP heart development, urogenital system development, striated muscle development was affected at the concentration of 6.13±12.62 µM, 6.19±10.92 µM, 6.64±11.44 µM respectively. The parent BP such as anatomical structure morphogenesis and organ development consists of 120 and 150 genes affected at the concentration of 8.18±13.11, 8.75±13.37 µM respectively. At 10.74±7.98 µM concentration DNA repair BP has been affected in response to thalidomide. The cytotoxicity mediated BP such as cellular response to DNA damage stimulus provides the mean BMD 10.40±8.9 µM. The genes mediated for the bone development consists of the BMD mean 13.95±25.83 µM. The ectoderm related cellular component (CC) (Table 1b) GO such as dentrite, axon, microtubule cytoskeleton mediated genes provide the mean BMD 9.67±9.84 µM, 18.39±28.47 µM, 9.65±11.13 µM respectively. The mesoderm related CC such as actin cytoskeleton, extracellular matrix, collagen mediated genes generate the mean BMD 13.15±14.93, 9.99±13.65, 12.76±16.91 respectively.

Gene expression profiling: comparison between the toxicogenomics of human and murine groups.

Cross-species gene expression assessment of H9 (human) and CGR8 (murine):

To determine the thalidomide mediated gene expression profiling, human and mouse orthologous genes were identified between Affymetrix Human U133 plus 2.0 array and Mouse 430 array (Figure 5). Total 20481 unique orthologous genes were commonly present between the two Affymetrix whole genome arrays. To identify molecular signatures that may characterize the diversity or similarity, gene expression profiles in the human and murine ESC groups were compared. The human system identified 3294 unique genes for 14 day old EBs and 232 unique genes for thalidomide treatment (fold change 2; FDR p value ≤ 0.05).We were able to identify 2148 unique genes (fold change 2; FDR p value ≤ 0.05) for 14 day old EBs and 1463 unique genes (fold change 2; FDR p value ≤ 0.05) for thalidomide treated for the murine system. Further the unique DET for murine and human systems were compared to the common orthologs genes obtained from two Affymetrix arrays and common orthologs were identified between the 14 day old EBs and thalidomide treated groups. Common species non specific orthologs for 14 day old EBs between murine and human was 559 and between thalidomide treatment was 59. There are 1253 and 1460 unique orthologous species specific genes were present for 14 old EBs signatures in murine and human respectively. There are 1208 and 115 unique orthologous species specific genes were present for thalidomide signatures in murine and human respectively (Figure 5). A hierarchical cluster analysis of the 559 genes for 14 day old EBs and 59 genes for thalidomide treatment is presented in figure 6a and 6c respectively. For the 14 day old EBs 559 genes, lung development, respiratory tube development, embryonic organ development, sensory organ development, cartilage development, angiogenesis, kidney development, neuron development, cardiac muscle tissue development, chordate embryonic development, vasculature development was the significant categories identified (Supplementary table XXX) and for thalidomide mediated 59 genes chordate embryonic development, cardiac muscle tissue development, cardiac cell differentiation, heart morphogenesis (Supplementary table XXX) was identified. 41 unique orthologs were identified that were common between the species non specific 14 day old EBs and species non specific thalidomide treated markers in human and murine systems. Among 41, 32 transcripts expressing more than 2 fold change were down regulated in thalidomide treatment (Supplementary table XXX). The 41 genes including AFP, FGA, GATA6, HOXA9 and genes related to cell proliferation and/or signal transduction such as FABP1, IGFBP7, RBP4 and IGFBP3 (Figure 6b). The results of an eisen plot and the biological process associated with these genes for human and murine systems is presented in figure 6. There were 1253 (Figure 5A) and 1460 (Figure 5 B) unique orthologous species specific genes present on 14 days old differentiated EB signatures in murine and human Affymetrix arrays respectively. There were 1208 (Figure 5 C) and 115 (Figure 5 D) unique orthologous species specific genes present on thalidomide signatures in murine and human respectively (Figure 2). To analyse the biological significance of these genes (Figure 5: A, B, C, D) DAVID gene ontology was performed. CGR8 specific 14 day differentiated EBs shows developments such as vascular, skeletal system, bone, kidney (Table 2a) and H9 specific 14 differentiated old EBs exhibits neuron development, embryonic skeletal system and forebrain (Table 2b). CGR8 specific thalidomide mediated genes shows developments such as- spleen, muscle cell, skeletal muscle fibre, inner ear and salivary gland (Table 2c). H9 specific thalidomide mediated genes exhibited heart, limb and chordate embryonic development (Table 2d). (Supplementary table XXX). Pathway analysis between species; non specific markers of 559, ECM receptor interaction was observed which consists of 16 transcripts. In response to thalidomide treatment between species; non specific markers of 59 revealed ECM receptor interaction was affected with 4 markers.


The susceptibility of developing embryo and fetus to chemical exposures during prenatal and early postnatal life may result in important effects on gene expression, thus ensuing functional defects and increased risks of disease later in life Grandjean, P. et al. (2008. These chemicals are often being chemotherapeutic drugs Sioka, C. et al. (2009. Since the damage these chemicals may cause is irreversible it is important to understand the toxicity of these chemotherapeutic drugs. Embryonic stem cells offer an effective tool to assess the toxicological profile of these drugs and to understand and predict the damage for these therapeutic agents Peters, A. K. et al. (2008.(27) ESCs have a unique ability to differentiate into all somatic cell types. Toxicogenomics approaches with respect to ESCs are proposed as an alternative to the traditional approaches for drug safety testing (Winkler et al., 2009). Ara C mediated dysregulation of hESC multi-lineage differentiation has employed to test the ability of hESC in toxicity testing. Recapitulation of embryonic development can be at least partly done in hESC, multilineage differentiation of hESC and toxicogenomics can be used as a model for the developmental toxicity assessment (Jagtap et al).

The developmental toxicity of thalidomide is illustrated in different studies in vivo and in vitro Stephens, T. D. et al. (2000b);Knobloch, J. et al. (2007a);Knobloch, J. et al. (2008. (8, 13, 28)We demonstrate the toxicity of thalidomide with respect to interspecies differences; dose dependant response and gene ontology using hESC and mESC mediated multilineage differentiation. With this approach we made an effort to addresses the gap between the in-vivo embryonic development and in-vitro embryonic development (includes mouse and human origin) in toxicological application.

For dose response transcriptomic analysis 0.01µM, 0.1µM, 1µM, 10µM and 70µM concentrations were selected which includes IC10 (10µM) and IC50 (70µM) concentrations (S.F 1). At 70µM concentration morphological evaluation results cytotoxicity and these results correlates with the gene expression analysis where the stress response markers were highly expressed (figure 1a and 1b). One of the studies was to find out the lowest observed adverse effect levels concentration (LOAEL) and the most sensitive markers for thalidomide. The sensitive t statistic approach found that 5 markers were dysregulated at the lowest concentration 0.01 µM (S.T 1) among, mesoderm marker COL6A3 was repressed, which is classified in organ development in GO. To find out the common markers in different concentrations, 5 set venn diagram was used. No common signature for all the concentration was observed but the common markers among the concentrations are discussed in the results section.

As a first step to find the teratogenecity, selected germ layer markers expression was represented in Eisen plot, Figure 2a explains the teratogenecity of thalidomide in germ layer markers, and the dose response model reveals endoderm; ectoderm and mesoderm markers were under represented from 1 µM, where as the stress response markers were repressed significantly on 70 µM. The number of dysregulated transcripts shows the pattern that at 0.01µM and 0.1µM had an impact on only few transcripts, where as at 1µM concentration showed steep, increase of markers but from 10 µM to 70 µM there is marginal number of transcript changes due to the apoptosis markers expression (Fig 2a). The toxicogenomics approach has encompasses the altered gene expression pattern in response to toxicant insult because these altered gene expression level can be measured prior to manifestation at the organismal level in traditional approach (27). In this present study thalidomide treatment for 14 days during the multilineage differentiation altered 25433 transcripts and further stringent filtration of markers results 1026 transcripts. Figure 3a explains the pattern for the serial down regulation of the developmental markers with concentration vice a versa for up regulated markers. The biological interpretation for these up and down regulated markers in DAVID analysis reveals thalidomide specific BP such as limb development and skeletal development in child BP, multicellular organismal development and organ development in parent BP. In response to toxic insult BP such as response to DNA damage stimulus, response to stress, DNA repair were present in GO analysis (Figure 3a).

To determine the teratogenic potential of thalidomide and its derivative various in-vivo assessments were made in rat, mice and rabbit and dog and subsequently reported the NOAEL based on traditional clinical and histopathological measurements. In Beagle dogs the NOAEL was 200mg/kg/day, the embryonic development NOAEL for New Zealand White Rabbits treated females mated to untreated males was < 10 mg/kg, the NOAEL for mice and female rats were 3000 mg/kg and 30 mg/kg for male rats. Teo, S. K., Denny, K. H., Stirling, D. I., Thomas, S. D., Morseth, S., and Hoberman, A. M. (2004a), Teo, S. K. et al. (2001), FRATTA, I. D., SIGG, E. B., and MAIORANA, K. (1965), Teo, S. K. et al. (2004b), (23, 26, 24, 28, 29). Although for thalidomide derivative EM12 the NOAEL was observed at 10μg/kg in the non-human primate Callithrix jacchus. Similarly the integration of Bench mark dose (BMD) estimation with dose response genomic data can determine the detrimental effects of the toxicant and in the present study to make high stringent GO contain <10 transcripts were filtered out Thomas, R. S. et al. (2007)(25). This dose response thalidomide data reveals the most sensitive GO and the respective concentration such as mammary gland development (0.88±1.19 µM). The sensitive gene expression data shows 1.5μg/ml (6.1μM), 3.5 μg/ml (14 μM) 1.6 μg/ml (6.2 μM) and 1.9 μg/ml (7.7 μM) is sufficient to alter the heart development, bone development, urogenital system development and embryonic development BP respectively (Table XXX). This information provides the prior knowledge about the toxicological observation with the traditional morphological observation. The cellular component GO in BMD shows the ectoderm mediated CC such as neuron projection, dentrite, axon, microtubule cytoskeleton and mesoderm mediated CC such as collagen, actin cytoskeleton, extracellular matrix were altered <20 μM thalidomide (Table XXX). The spectrum of malformations after consumption of thalidomide in human produces the detrimental effect in dysmelia and systemic anomalies such as kidney malformations, cardiac anomalies and central nervous system complications Miller, M. T. (1991) (reviewed 30). These anomalies are corresponding to the GO processes in thalidomide treatment during the hESC embryonic development.

To further analyze the cross species assessment of thalidomide effects in hESC and mESC embryonic development, we used comparative toxicogenomics approach for 14 day old EBs and thalidomide mediated transcripts. Although the stages of embryonic development for human and mouse differs, from the unique orthologous signatures 27% common transcripts for hESC and mESC 14 day old EBs were observed (Figure XX). The homeodomains are highly conserved among the species; it is rapidly evolving, and HOXA10 is a transcription factor expressed prominently during embryonic uterine development along with the cofactors such as PBX and MEIS Sarno, J. L. et al. (2005), Ting, C. T. et al. (1998)(31, 32). In the present study 14 day old mouse and human EBs expressed HOXA10, PBX1, PBX3, MEIS1 and MEIS2, explains that the development of human and mouse embryonic stem cells were conserved to the optimum extend with the 27% common signatures. And further this study explores the unrevealed common signatures for thalidomide for mouse and human species. And the hierarchical analysis represents the similar expression pattern of the common DET for 14 day old EBs, thalidomide treatment and common non species specific genes for treated and untreated. (Figure XXX). Species specific genes for 14 day old EBs, thalidomide treatment for human and mouse EBs were investigated for the GO analysis, the results reproduced the in-vivo results such as the limb deformities were not observed in mice (24) for 200mg/kg and the limb defects were observed in cynomolgus monkeys in response to 15-20 and transcriptional studies showed vasculature development gene ontology was down regulated (18). The limb development and heart development GO was found only for human and not in mice in response to thalidomide treatment. The selected GOs observed for CGR8 cells can not be observed in in vivo studies reported. (Table XXX).

In sum combination of functional genomics, pluripotent stem cells and in vitro toxicity revealed novel signatures for thalidomide mediated toxicity in human and mouse embryonic development and the lowest concentration of thalidomide for mesoderm markers inhibition is reported. The dose response model further discovered the specific concentrations of thalidomide needed to alter the developmental biological process. Further the cross species comparison showed the unrevealed common thalidomide signatures for mouse and human as well as the species specific signatures reproduce the in vivo assessments. Hence the sensitive genomic approach in combination with stems cells as an alternative to in vivo assessment for developmental toxicity.