Directly to content
  1. Publishing |
  2. Search |
  3. Browse |
  4. Recent items rss |
  5. Open Access |
  6. Jur. Issues |
  7. DeutschClear Cookie - decide language by browser settings

Data-driven modeling of the dynamic competition between virus infection and the antiviral interferon response

Rinas, Melanie

[thumbnail of Thesis_Melanie_Rinas.pdf]
Preview
PDF, English
Download (17MB) | Lizenz: Creative Commons LizenzvertragData-driven modeling of the dynamic competition between virus infection and the antiviral interferon response by Rinas, Melanie underlies the terms of Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 Germany

Citation of documents: Please do not cite the URL that is displayed in your browser location input, instead use the DOI, URN or the persistent URL below, as we can guarantee their long-time accessibility.

Abstract

The interferon system functions as a first line of defense against viral infections. The cellular recognition of viruses leads to the production of interferons (IFN) by the infected cells. Secreted IFN stimulates an antiviral response in an autocrine and paracrine manner: While autocrine IFN action inhibits virus production in infected cells, paracrine IFN signaling induces an antiviral protective state in naïve cells. Although central molecular components of the IFN system have been characterized, our quantitative understanding of its dynamics remains limited. In particular, it is not precisely known which molecular processes are decisive for the outcome of virus-host interactions.

Together with our experimental cooperation partners, we have studied virus-induced IFN signaling at single-cell resolution in communicating cell populations after infection with a non-spreading and a spreading virus. On this basis, we established two complementary mathematical models. First, we developed a stochastic multi-scale model accounting for the intracellular dynamics in individual cells and the cell-to-cell communication via secreted IFN. Second, we constructed a delay differential equation model to analyze the competition between viral spread and IFN-induced antiviral defense in a cell population. Both models were parameterized on the basis of original experimental data and the numerical analyses of the models aimed at deriving testable predictions for new experiments.

By live-cell imaging, we showed that key steps of the IFN pathway including virus-induced signaling, IFN expression, and induction of IFN-stimulated genes are stochastic events in individual cells. To relate the single-cell data after primary infection to antiviral protection at the cell-population level, we established a stochastic model which combines the heterogeneous IFN signaling in single cells with the intercellular communication through released IFN. The parameters describing the virus-induced activation of the transcription factors of the IFN gene were estimated from a distribution of observed single-cell activation times using as objective function Neyman’s chi-squares statistic. The minimization of this objective function by simulated annealing revealed that virus-induced signaling is cooperative. Moreover, fitting of the measured time delays between transcription factor activation and IFN gene induction in individual cells with a gamma distribution by applying the maximum-likelihood method implies that IFN gene induction downstream of transcription factor activation is a slow multi-step process. Notably, mathematical modeling and experimental validation indicate that reliable antiviral protection in the face of multi-layered cellular stochasticity can be achieved by paracrine propagation of the IFN signal. Therefore, a few IFN-producing cells are able to protect a large number of naïve cells (Rand, Rinas et al. 2012).

To investigate the competition between viral spread and IFN-induced antiviral defense, we examined virus-host interactions after infection with spreading Dengue virus. For this purpose, our experimental partners generated data showing the antiviral response dynamics of fluorescent reporter cells after infection with a fluorescently labeled Dengue virus. Based on these kinetic data, we established a delay differential equation model with time delays for virus replication, virus production and IFN secretion. Using data-driven least-squares fitting and profile likelihood analysis, we identified the model parameters within narrow confidence bounds and found that the timing of virus production and IFN secretion after infection are almost identical. This direct competition together with the highly heterogeneous IFN response in single cells fosters the coexistence of IFN-induced protection of naïve cells and viral spread in non-protected cells. To analyze which components of the antiviral IFN system have the greatest influence on viral spread, we compared the infection dynamics of the wild-type Dengue virus with the attenuated spread of the vaccine candidate Dengue virus E217A mutant. We quantified the differences between wild-type and mutant Dengue virus infections using data-driven parameter optimization constrained by the determined wild-type virus related parameter values. In this way, we identified two mutant virus specific parameters, which explain the attenuation of the mutant through a reduced virus production rate and an accelerated IFN secretion taking place much earlier than virus production. By mathematical modeling and validation experiments, we predict that rapid IFN action curbing virus production in infected cells is critical for the attenuation of the Dengue virus E217A mutant. Thus, a fast acting autocrine IFN signal could limit viral spread in such a way that accelerated paracrine IFN response has only a minor impact on the spread of Dengue virus (Schmid, Rinas et al. submitted).

In conclusion, our work demonstrates that mathematical modeling is an essential tool to integrate data and mechanisms from the molecular to the cell-population level. The research on understanding which molecular mechanisms shape virus-host interactions might inform the development of new antiviral therapies and vaccines.

Document type: Dissertation
Supervisor: Marciniak-Czochra, Prof. Dr. Anna
Date of thesis defense: 1 July 2015
Date Deposited: 11 Sep 2015 09:27
Date: 2016
Faculties / Institutes: The Faculty of Mathematics and Computer Science > Dean's Office of The Faculty of Mathematics and Computer Science
DDC-classification: 500 Natural sciences and mathematics
Controlled Keywords: mathematical modeling, parameter estimation, multi-scale model, stochastic modeling, delay differential equations, virus-host interaction, interferon system, antiviral response
About | FAQ | Contact | Imprint |
OA-LogoDINI certificate 2013Logo der Open-Archives-Initiative