The cells of the hematopoietic system are continuously renewed in a tightly controlled growth and differentiation process. Dysregulation of hematopoiesis results in leukemias or anemia. Key regulator of the erythroid lineage is the hormone erythropoietin (Epo) that binds to the erythropoietin receptor (EpoR), a hematopoietic cytokine receptor specifically present on erythroid progenitor cells. Tight coordination of signal strength and duration in response to receptor activation controls the regeneration of erythrocytes from erythroid progenitor cells. To identify regulatory mechanisms controlling cellular decisions in response to EpoR signaling, we are employing a systems biology approach based on simultaneous quantitative monitoring of multiple signaling components in combination with dynamic pathway modeling.
Since signaling networks are extremely complex, description by mathematical models can provide a useful tool to gain insights into their key regulatory mechanisms. However, to establish meaningful mathematical models it is essential that these models are based on high-quality quantitative data acquired under standardized conditions. Therefore, we established strategies for error reduction and automated data processing to reliably apply quantitative immunoblotting for acquiring time-course data sets and determining the stoichiometry of signaling pathway components (Schilling et al., FEBS Journal 2005; Schilling et al., IEE Proc Systems Biology). To extend the possibilities to simultaneously detect alterations in signaling components in small sample volumes we developed quantitative protein arrays based on detection in the near infrared range (NGFN).
Furthermore, to examine gene regulatory networks, we are currently developing strategies to use microarray analysis and quantitative real time PCR for quantitative data generation.
To elucidate principal mechanisms determining cellular decisions, we are addressing how signal transduction through the EpoR coordinates proliferation and differentiation of erythroid progenitor cells by a systems biology approach. Based on available biological knowledge we are establishing mathematical models for signaling pathways activated by the EpoR estimating the parameters of the model using time-resolved quantitative data. A major signaling cascade triggered upon activation of signal transduction through the EpoR is the JAK2-STAT5 signaling pathway that involves the activation of a cytoplasmic tyrosine kinase (JAK2) and a latent transcription factor (STAT5). In collaboration with our modeling partners from the group of Jens Timmer (University of Freiburg), we established a mathematical model capturing the dynamic behavior of the core module of the JAK2-STAT5 signaling pathway (Swameye et al., PNAS 2003). Applying this model enabled us to identify rapid nucleocytoplasmic cycling as an important systems property and to successfully predict steps in the cascade most suitable for intervention. Since the dynamic behavior of a signaling pathway is critically determined by feedback loops, we are currently extending the model to address the role of the negative regulatory adapter protein CIS and the tyrosine phosphatase SHP-1 for controlling extent and duration of STAT5 activation in collaboration with our modeling partners from the groups of Jens Timmer (University of Freiburg) and Thomas Höfer (DKFZ, Heidelberg) (COSBICS). In addition, we are establishing data-based mathematical models for the Epo-mediated activation of MAP kinase and PI3 kinase signaling cascades and aiming at connecting the activation of the Epo-induced signaling network with activation of gene regulatory networks (SBCancer).
Besides the temporal regulation, signal processing by signaling networks is critically determined by the subcellular organization. Furthermore, characteristic dynamic behavior such as oscillation or adaptation may only be detectable at the single cell level. Therefore, we employ fluorescence imaging methods ranging from qualitative immunofluorescence studies in fixed cells to complex quantitative time-lapse microscopy of living cells to gain a better understanding of how signals are propagated from the cell surface to the nucleus. As an example we are examining in the case of the JAK2-STAT5 pathway the dynamics of the import and export processes. We monitor the dynamics of fluorescently labeled STAT5 by time lapse imaging and photobleaching techniques such as Fluorescence Recovery After Photobleachig (FRAP) and Fluorescence Loss In Photobleaching (FLIP). The parameters determined by quantitative live cell imaging are then combined with the quantitative immunoblotting data and used to improve the mathematical models describing the pathways under investigation (COSBICS). To study local interactions and concentrations of signaling molecules we are planning to use Fluorescence Energy Resonance Transfer (FRET) and Fluorescence Correlation Spectroscopy (FCS) in collaboration with groups at the DKFZ and in the BIOQUANT building.
To challenge and validate model predictions we are developing tools for targeted over-expression and down-modulation as well as using primary cells from conditional knockout mice. We are employing the Cre-loxP system and established a mouse line that enables us to specifically eliminate genes in the erythroid lineage (Heinrich et al., Blood 2004).
A unique feature of hepatocytes is the nearly unlimited potential for regeneration. In a tightly regulated growth process differentiated hepatocytes are primed to reenter proliferation. During the "priming" phase of the tightly regulated growth process cytokines such as interleukin(IL)-6 rapidly trigger signaling pathways that enable the previously quiescent hepatocytes to enter cell cycle progression. Proliferation of hepatocytes is primarily stimulated by the hepatocyte growth factor (HGF). Major factors involved in termination of hepatocyte regeneration and thereby ensuring liver mass conservation are TGF-beta and activin which suppress cell growth. Critical for hepatocyte regeneration is the coordinated activation of multiple signaling pathways and the timing, extent and duration of activation. To identify key regulatory mechanisms we are applying a systems biology approach that combines quantitative data generation with dynamic pathway modeling and facilitates prediction regarding systems behavior (HepatoSys).
To study the dynamic behavior of the IL-6 induced JAK1-STAT3 cascade and determine its general design principles, we stimulate primary murine hepatocytes based on experimental design predictions. Different stimuli as well as inhibitors are applied. Dynamical properties like signal amplitude, duration and their impact on biological read-outs are studied applying techniques like quantitative immunoblotting, quantitative Real Time (RT)-PCR, chromatin immunoprecipitation (ChIP)-on-chip and in combination with mass spectrometric (MS) analysis. Negative feedback-loops as well as cross-talk with other pathways are monitored to understand how different signals are integrated within cells and how they determine specific biological responses.
The Hepatocyte Growth Factor (HGF) is a pleiotropic cytokine that has been discovered as inducer of dissociation and motility in epithelial cells and as a mitogen for hepatocytes. In the liver, HGF is produced by the stellate cells in response to hepatic injury. HGF secretion by stellate cells activates the Met receptor present on the hepatocytes membrane in a paracrine fashion. Met-receptor activated signalling pathways enhance the entry of the hepatocytes into the cell cycle, and into a proliferative state. The regulation of cell proliferation is mainly driven by PI3K and PKB/AKT signalling pathway. We are generating time-resolved data for receptor, Akt and GSK3 phosphorylation and are establishing a data-based mathematical model to dissect mechanisms regulating the proliferation of hepatocytes.
TGFbeta induced activation of the SMAD signaling cascade plays a crucial role in the termination phase of liver regeneration. Applying newly developed methods for data acquisition and processing numerous key players of the TGFbeta-SMAD signaling system including extracellular ligand, phosphorylated SMAD2 and activated SMAD2/4 complexes are analyzed in parallel. Additionally, the induction of target genes is monitored both on the mRNA and protein level.
Based on these data of the pathway components and their functional states ordinary differential equation (ODE) models including two different layers of negative regulation are being built. Our models will help to elucidate how the regulatory mechanisms are integrated to modulate amplitude and duration of TGFbeta-SMAD signaling and thereby control termination of hepatocyte regeneration.
The Hepatitis C Virus is the major cause of chronic liver disease and hepatocellular carcinoma worldwide. By modulation of cellular signaling pathways the virus evades the cellular antiviral response, leading to a persistent infection in more than 80% of cases. We want to use a combination of biochemical and fluorescent imaging methods to study how the virus interferes with the Jak1-STAT1/3 signalling pathway and thus disrupts the cell’s response to the crucial antiviral cytokine Interferon alpha (Viroquant).