Areas of Interest


Cancer is a disease of altered communication of cells on multiple levels. Tumor formation is initiated by genomic changes in individual cells capable to disrupt their physiological growth control. The common property of cancer cells is uncontrolled proliferation in their organismal environment. Tumor progression can be fostered by adaptations to the tumor micro-environment culminating in metastatic spread of the tumor through the organ and finally the entire body. These processes are driven by a complex array of extracellular cues such as growth factors, cytokines and extracellular matrix components that bind to cell surface receptors and activate intracellular signaling networks, thereby controlling cell proliferation, differentiation and survival. Due to the non-linearity of the underlying reactions and their distinct dynamics, it is highly challenging to dissect key mechanisms contributing to cancer formation. We employ a Systems Biology approach that combines mathematical modeling with quantitative data generation to gain insights into mechanisms that determine cellular decisions and to identify cancer-relevant perturbations.

From the regulation of erythropoiesis to erythroleukemia

In our body, the production of red blood cells from erythroid progenitor cells, known as erythropoiesis, is a very robust process. Every second 2.5 million red blood cells are generated in a healthy human individual. Despite the high proliferative capacity of erythroid progenitor cells, malignancies in this lineage, erythroleukemia, are very rare.

The key regulator of the formation of red blood cells is the hormone erythropoietin (Epo), which is secreted by the kidney. Erythroid progenitor cells at the colony-forming unit-erythroid (CFU-E) stage entirely depend on Epo, which regulates cell survival, cell proliferation and cell differentiation. Since erythroid progenitor cells are readily accessible and can be cultivated in vitro, the system enables us to establish mathematical modeling approaches to answer biological questions and later on apply these approaches to other more complex systems. By developing a data-based mathematical model of the interaction of Epo with the Epo receptor (EpoR), we uncovered that very rapid ligand turnover enables the system to respond to a wide range of Epo concentrations (Becker et al., Science, 2010). By data-based mathematical modeling we showed that the JAK2-STAT5 signal transduction pathway is regulated by two negative feedback regulators that share the labor and control the pathway at low or high Epo concentrations, respectively (Bachmann et al., Mol Syst Biol, 2011). Recently, we established by an integrative mathematical model of Epo-induced MAP kinase und PI3 kinase signaling pathway activation that the abundance of the signaling components determines cell context-specific information processing through intracellular signaling networks. We could further show that snapshot information on protein abundance obtained from patient samples is sufficient to adapt the mathematical model to the human context and to predict the response of patient cells to pathway inhibitors (Adlung et al., Mol Syst Biol, 2017). These advances open new avenues to derive quantitative information from limited, snapshot patient material to pinpoint processes deregulated in erythroleukemia and to guide personalized treatment.

By combinging experimental data and mathematical modeling, cell type-specific impacts of the AKT and MEK inhibitors were predicted and validated. Adapted from Adlung et al., Mol Syst Biol, 2017.

From perturbation of liver regeneration to liver cancer

The liver has a remarkable capacity to regenerate, and the regeneration process is tightly regulated to avoid uncontrolled growth. Upon damage, non-parenchymal cells such as Kupffer cells and hepatic sinusoidal endothelial cells are activated to secrete interleukin (IL)-6 and tumor necrosis factor alpha (TNFα). These factors prime the quiescent hepatocytes in the liver for the proliferative response. Subsequently, hepatic stellate cells and hepatic sinusoidal endothelial cells secrete hepatocyte growth factor (HGF), the major mitogen for hepatocytes. A key factor involved in the termination of liver regeneration is transforming growth factor beta (TGFβ), which is secreted by the stellate cells and suppresses cell cycle progression. Liver damage occurs for instance in response to drugs, alcohol, an unhealthy lifestyle or viral infections. Based on experimental data generated in primary mouse hepatocytes, we established dynamic pathway models for IL-6, HGF, TGFβ and TNFα signaling pathways. With our Systems Biology approach we identified hepatocyte-specific cross-talk mechanisms of HGF-induced signal transduction (D'Alessandro et al., PLoS Comput Biol, 2015) and a gatekeeper that ensures proliferation only in response to a true HGF signal (Mueller et al., Mol Syst Biol, 2015). Additionally, the mathematical model of TGFβ-induced signal transduction enabled the identification of relevant Smad complexes. Further, a data-based mathematical model of IL-6 signal transduction enabled us to predict optimized inhibitor treatment regarding how medical compounds such as Paracetamol (Acetaminophen, APAP) or a high-fat and high-caloric diet influence the regenerative potential of liver cells and thereby contribute to liver failure. Further, we are developing a mathematical model of hepatitis B virus (HBV) infection and host interaction to prevent persistent infection of the liver and thereby reduce the risk of patients to develop liver cancer.

Selection of mathematical models based on quantitative experimental data generated in primary mouse hepatocytes was performed to identify cross-talk mechanisms of HGF-induced signal transduction. Adapted from D'Alessandro et al., PLoS Comput Biol, 2015.

From mechanisms contributing to lung cancer progression towards personalized treatment optimization

Due to early metastatic spread, high mutation load and development of therapy resistance, lung cancer remains the leading cause of cancer related deaths worldwide. A major mediator inducing epidermal to mesenchymal transition (EMT) and thereby fostering early metastasis is the transforming growth factor beta (TGFβ). We identified that TGFβ-induced signal transduction is frequently deregulated in lung cancer. One of the underlying mechanisms is down-regulation of the negative regulator of TGFβ signal transduction BAMBI, and we showed that reconstitution of BAMBI in lung cancer cells reduces the tumor burden in mice (Marwitz et al., Cancer Res, 2016). Currently, we are developing a data-based mathematical model to unravel regulatory mechanisms altered in the context of lung cancer. We exemplarily compared the dynamics of Epo-induced signal transduction in erythroid progenitor cells and lung cancer cells and successfully established a mathematical modeling approach to predict promising targets that could be utilized to specifically target cancer cells (Merkle et al., PLoS Comput Biol, 2016). Therapeutic inhibitors targeting the EGFR are successfully applied in a subgroup of lung cancer patients harboring an activating mutation in the EGFR. However, almost all patients develop resistance against the targeted therapy. We utilized mathematical modeling and unraveled mechanisms how MET contributes to therapy resistance against EGFR inhibitors. A widely applied therapeutic option is chemotherapy. However, systemic chemotherapy induces anemia that strongly reduces the quality of life of lung cancer patients. Two therapeutic options, blood transfusions and erythropoiesis stimulating agents (ESAs), are available to manage chemotherapy-associated anemia. We developed a mechanistic mathematical model to guide clinical decisions based on the prediction of the response and the associated risk to the available therapeutic options.

Reconstitution of the pseudoreceptor BAMBI, a negative regulator of the TGFβ pathway, abrogates TGFβ-induced cancer cell migration. Adapted from Marwitz et al., Cancer Res, 2016.

to top