Probabilistic Modeling and Analysis of Embryonic Stem Cell Regulatory Network Dynamics and Ethanol Effects On Differentiation towards a Neural Cell State

Thursday, November 11, 2010: 1:30 PM
255 D Room (Salt Palace Convention Center)
Rajanikanth Vadigepalli, Joshua Ogony and Helen Anni, Department of Pathology, Anatomy and Cell Biology, Thomas Jefferson University, Philadelphia, PA

Molecular understanding of alcohol toxicity during embryo development is critical to tackle fetal alcohol spectrum disorders, which are mostly associated with prenatal alcohol exposure in utero. The present study focuses on characterizing the alcohol effects on the regulatory network interactions mediating embryonic stem (ES) cell differentiation towards a neuroectodermal cell state. A considerable body of work addressed the alcohol toxicity using in vivo animal models as well as in vitro differentiating neural precursor cells. However, there is limited understanding of how alcohol affects the onset of embryogenesis at the molecular level. To address this issue, we take an integrated approach combining high-throughput experimental techniques and probabilistic network modeling to analyze alcohol effects on ES cell differentiation networks.

Our experimental system consists of mouse ES cells and retinoic acid-directed differentiation to neuroectodermal fate in adherent monolayer cultures in the presence of ethanol. A key aspect of our experimental approach is the high-throughput measurement of multi-parametric single-cell data by high-end flow cytometry. We analyze these multiplex data sets using a probabilistic network modeling approach by adapting the Algorithm for the Reconstruction of Accurate Cellular Networks (ARACNE: Margolin et al., BMC Bioinformatics 2006) that was previously developed for gene expression data analysis to instead utilize the single-cell multiplex data on protein abundances. Through this approach, we identify key network interactions and their dynamics as altered by ethanol exposure during differentiation. Our use of ARACNE is facilitated by the large data sets obtained from flow cytometry on single cells, where each cell is treated as an independent observation with correlated multi-protein measures.

We have obtained single-cell multiplex data on the expression of the ES core transcription factors (TFs) Sox2, Oct4 and Nanog in RA-directed differentiating ES cells treated 06 days with ethanol. In addition, we obtained single-cell multiplex data sets for comparison from matched controls. We constrained our analysis to the data from live cells gated by a fixable Live/Dead near infra-red dye to distinguish subpopulations of cells. The flow cytometry simultaneous measurement of the expression of the core TFs is based on the intracellular staining with a cocktail of differentially-conjugated fluorescent antibodies for Nanog, Oct4 and Sox2 vs. that of the non-specific isotype control.

Our analysis of the multiplex data on core TFs indicates that: (1) the core TFs were down regulated during differentiation, and ethanol amplified the down regulation of Nanog and Sox2, but does not further alter Oct4 levels; (2) the strength of the correlative interactions between the core TFs decreases during RA-mediated differentiation compared to that in the undifferentiated ES cells; (3) ethanol modulates the RA-mediated differentiation network by further decreasing these interactions over time. Adverse effects of ethanol on ES cell differentiation may be mediated by a synergy between the changes in the TF levels as well as the network interactions. Guided by these results, we propose the hypothesis that ethanol adversely affects ES cell differentiation through a synergistic decrease in the levels of the core TFs as well as the network interactions over time. These perturbations may lead to altered lineage commitment and the developmental defects of fetal alcohol spectrum disorders observed in mice and humans.

Research Support: NIAAA T32 AA007563 grant, the Graham fund, and the Department of Pathology, Anatomy and Cell Biology, Thomas Jefferson University.


Extended Abstract: File Not Uploaded
See more of this Session: Mathematical Approaches in Systems Biology I
See more of this Group/Topical: Food, Pharmaceutical & Bioengineering Division