I. Identification and characterization of novel factors involved in the antiviral RIG-I / IRF-3 pathway

 Figure 2– Workflow of the screening approach used to identify novel factors of the RIG-I / IRF3 pathway. GFP-tagged IRF3 was employed as readout for the activation of the innate antiviral pathway: once the RIG-I system is triggered, this transcription factor translocates from the cytoplasm into the nucleus. By automated microscopy and image processing, more than 32.000 frames were acquired and analyzed. Advanced statistical evaluation was required to identify genes that interfere with (or enhance) IRF3 activation upon RIG-I stimulation (Knapp 2011). 22 high-confidence hits could be validated (Willemsen 2017).

Figure 2– Workflow of the screening approach used to identify novel factors of the RIG-I / IRF3 pathway. GFP-tagged IRF3 was employed as readout for the activation of the innate antiviral pathway: once the RIG-I system is triggered, this transcription factor translocates from the cytoplasm into the nucleus. By automated microscopy and image processing, more than 32.000 frames were acquired and analyzed. Advanced statistical evaluation was required to identify genes that interfere with (or enhance) IRF3 activation upon RIG-I stimulation (Knapp 2011). 22 high-confidence hits could be validated (Willemsen 2017).
© dkfz.de

In order to sense a viral infection and trigger an appropriate defensive response, the cell has developed a set of sensor molecules. One of the more prominent sensors is retinoic acid inducible gene I (RIG-I, aka DDX58). Upon recognition of viral RNA (Binder 2011), RIG-I triggers a signaling cascade through the crucial innate adapter molecule MAVS (aka CARDIF, IPS1 or VISA), eventually leading to activation of the latent transcription factor IRF-3. This transcription factor in turn drives the production and secretion of type I (and III) interferons, which are highly potent mediators of the antiviral response, rendering the infected and the surrounding cells into an alert antiviral state, which, in most cases, is sufficient in restricting the replication and spread of the virus. In this project, we performed a high-throughput siRNA-based screen for cellular kinase genes that regulate this antiviral signaling in the RIG-I / IRF-3 axis (Willemsen 2017, Knapp 2011). We could validate 22 genes regulating antiviral IRF-3 signaling but also the inflammatory response via the transcription factor NFkB. One of the most prominent hits was DAPK1, a kinase that was previously known for its role in apoptosis, autophagy and other cellular processes. We could demonstrate, that DAPK1 also constitutes a negative feedback regulator of RIG-I signaling: upon virus infection, RIG-I not only activates its canonical downstream factors (incl. IRF3 and NFkB), but- with a certain time delay- also its own inhibitor DAPK1. We could further show that activated DAPK1 in turn directly interacts with and phosphorylates RIG-I, which as a consequence loses its ability to sense viral RNA. Like this, the antiviral pathway is slowly shut down again (Willemsen 2017). We are currently following up on the biological significance of virus-induced activation of DAPK1 and its family members DAPK2 and DAPK3. Moreover, we are investigating the roles of some other kinases out of the 22, which might link so far unrelated cellular pathways and systems to the antiviral response.

  • Willemsen, J., Wicht, O., Wolanski, J.C., Baur, N., Bastian, S., Haas, D.A., Matula, P., Knapp, B., Meyniel-Schicklin, L., Wang, C., Bartenschlager, R., Lohmann, V., Rohr, K., Erfle, H., Kaderali, L., Marcotrigiano, J., Pichlmair, A. and Binder, M. (2017) Phosphorylation-Dependent Feedback Inhibition of RIG-I by DAPK1 Identified by Kinome-wide siRNA Screening. Molecular Cell 65 (3):403–415

  • Binder, M., Eberle, F., Seitz, S., Mücke, N., Hüber, C. M., Kiani, N., Kaderali, L., Lohmann, V., Dalpke, A. & Bartenschlager, R. (2011). Molecular mechanism of signal perception and integration by the innate immune sensor retinoic acid-inducible gene-I (RIG-I). J Biol Chem 286, 27278–27287

  • Knapp, B., Rebhan, I., Kumar, A., Matula, P., Kiani, N. A., Binder, M., Erfle, H., Rohr, K., Eils, R. & other authors. (2011). Normalizing for individual cell population context in the analysis of high-content cellular screens. BMC Bioinformatics 12: 485

II. Understanding the dynamics of HCV RNA replication and its interplay with the interferon system

 Figure 1– Schematic representation of the mathematical model of intracellular HCV RNA replication (Binder 2013). A system of 11 differential equations is used to in silico simulate HCV replication under conditions that would be hard or impossible to achieve experimentally.

Figure 2– Schematic representation of the mathematical model of intracellular HCV RNA replication (Binder 2013). A system of 11 differential equations is used to in silico simulate HCV replication under conditions that would be hard or impossible to achieve experimentally.
© dkfz.de

Hepatitis C virus (HCV) is a positive strand RNA virus, but unlike almost all other representatives of this class of viruses, it establishes a prolonged (measured in decades) persistent infection. This is characterized by a steadily ongoing low-profile replication within liver cells, which requires a tightly controlled mode of intracellular replication. While many aspects of its life-cycle have been studied in detail, we still lack a comprehensive understanding of the RNA replication cycle inside a cell. In this project, we quantitatively assess the highly dynamic processes occuring directly after a viral genome entered its host cell. Based on previous knowledge and on new quantitative and dynamic data, we developed a mathematical description of these intracellular events (Binder 2013). This mathematical model allows for an in silico simulation of viral replication, even under conditions that would not be possible to examine in vitro. Currently, we are extending this model to comprise the full life-cycle of the virus, including particle production and secretion, as well as receptor binding and cell entry. Furthermore, we are very much interested in how viral replications interacts with the host cellular antiviral defense system (RIG-I/MDA5 -> IRF3 -> interferon). We have previously characterized several interactions of HCV with the innate antiviral system already, both, how HCV impacts the induction of the interferon system (Hiet 2015, Bender 2015), as well as how interferon can impact on HCV replication (Metz 2011, Grünvogel 2015, Grünvogel 2016). Currently, we want to get a better understanding of the molecular mechanisms how interferon really interferes with HCV replication. Therefore, we are treating HCV replicating cells with defined doses of interferons and measure the inhibitory effect in a time-resolved and quantitative manner. The mathematical model will then be used to interpret these data and predict the most likely mode of action of itnerferon.

One closely related project, funded within an EraNET, extends our HCV model to the life-cycles of other plus-strand RNA viruses. Together with high-throughut screening data, these models will form the basis for the in silico identification of novel antiviral compounds active against different viruses at the same time.

Mathematical modeling is performed in a very close collaboration with the biomathematicians of the group of Lars Kaderali (University Medicin Greifswald).

  • Grünvogel, O., Esser-Nobis, K., Windisch, M.P., Frese, M., Trippler, M., Bartenschlager, R., Lohmann, V. and Binder, M. (2016) Type I and type II interferon responses in two human liver cell lines (Huh-7 and HuH6). Genomics Data 7:166–170

  • Grünvogel, O., Esser-Nobis, K., Reustle, A., Schult, P., Müller, B., Metz, P., Trippler, M., Windisch, M.P., Frese, M., Binder, M., Fackler, O., Bartenschlager, R., Ruggieri, A. and Lohmann, V. (2015) DDX60L Is an Interferon-Stimulated Gene Product Restricting Hepatitis C Virus Replication in Cell Culture. J Virol, 89 (20):10548–10568

  • Bender, S., Reuter, A., Eberle, F., Einhorn, E., Binder, M. and Bartenschlager, R. (2015) Activation of Type I and III Interferon Response by Mitochondrial and Peroxisomal MAVS and Inhibition by Hepatitis C Virus. PLoS pathogens 11 (11):e1005264

  • Hiet, M.-S., Bauhofer, O., Zayas, M., Roth, H., Tanaka, Y., Schirmacher, P., Willemsen, J., Grünvogel, O., Bender, S., Binder, M., Lohmann, V., Lotteau, V., Ruggieri, A. and Bartenschlager, R. (2015) Control of temporal activation of hepatitis C virus-induced interferon response by domain 2 of nonstructural protein 5A. Journal of Hepatology 63 (4):829–837

  • Binder, M., Sulaimanov, N., Clausznitzer, D., Schulze, M., Hüber, C. M., Lenz, S. M., Schlöder, J. P., Trippler, M., Bartenschlager, R., Lohman, V. and Kaderali L. (2013). Replication vesicles are load- and choke-points in the hepatitis C virus lifecycle. PLoS Pathog 9, e1003561

  • Metz, P., Dazert, E., Ruggieri, A., Mazur, J., Kaderali, L., Kaul, A., Zeuge, U., Windisch, M.P., Trippler, M., Lohmann, V., Binder, M., Frese, M. and Bartenschlager, R. (2012) Identification of type I and type II interferon-induced effectors controlling hepatitis C virus replication. Hepatology 56 (6):2082–2093

III. System biological description of host cell antiviral responses

Figure 3– Understanding RIG-I:dsRNA interaction and activation of the signaling pathway. We have previously shown that RIG-I does not only recognize and bind to the 5’-terminus of double-stranded RNA (dsRNA), but in fact along the entire length of the dsRNA molecule (A-C) (Binder 2011). In collaboration with the group of Prof. Kaderali in Greifswald, we are currently developing a mathematical model of this cooperative binding event (D), which will be extended to cover the full RIG-I / IRF3 pathway.

Figure 3– Understanding RIG-I:dsRNA interaction and activation of the signaling pathway. We have previously shown that RIG-I does not only recognize and bind to the 5’-terminus of double-stranded RNA (dsRNA), but in fact along the entire length of the dsRNA molecule (A-C) (Binder 2011). In collaboration with the group of Prof. Kaderali in Greifswald, we are currently developing a mathematical model of this cooperative binding event (D), which will be extended to cover the full RIG-I / IRF3 pathway.
© dkfz.de

Where the above described project identifies and characterizes the influence of individual factors on the quantitative outcome of RIG-I / IRF-3 signaling, this project aims at a comprehensive systems biological description of this central antiviral pathway. For that purpose, we assess the kinetic behavior of the signaling cascade as a whole, in addition to its individual sub steps. In a bottom-up-approach, we start with a rudimentary, descriptive mathematical model, simply recapitulating the activation behavior of the pathway upon stimulation with viral RNA. Step by step, we then refine the model by implementing individual signal transduction and regulatory events on a molecular level. Our long term goal will be a comprehensive mathematical model of innate antiviral signaling, which will allow for an in silico simulation of the complete signaling cascade that can include situations which are difficult and/or impossible to study in vitro or in vivo. This model can then be combined with the model of viral replication (e.g. HCV, see above) in order to study the complex interplay of virus infection, triggering of host defenses and the action of viral antagonists of immune signaling (e.g. HCV NS3/4A protease or influenza NS1) (Pichlmair 2012).

  • Pichlmair, A., Kandasamy, K., Alvisi, G., Mulhern, O., Sacco, R., Habjan, M., Binder, M., Stefanovic, A., Eberle, C.-A., other authors & Giulio Superti-Furga. (2012). Viral immune modulators perturb the human molecular network by common and unique strategies. Nature 487, 486–490.

to top