Skip to main content

"""Role of Nonlinear Dynamics of NF-kB in Inflammation"""

Final Report Summary - NONLIN-KB ("Role of Nonlinear Dynamics of NF-kB in Inflammation")

Inflammation is a self-limiting response to infection and injury, and it is essential both to limit damage and to promote repair. Importantly, excessive or sustained inflammation leads to extensive tissue damage and to the development and progression of several diseases with a huge social impact, including septic shock autoimmune diseases and cancer. Multiple steps of inflammation are regulated by transcription factors (TFs) that are members of the Nuclear Factor kappa B (NF-kB) family. It was recently shown that upon stimulation with different biochemical substances, the transcription factor NF-kB undergoes sustained nucleus-to-cytoplasm oscillations. By analyzing the mathematical models of the NF-kB system using my background on nonlinear dynamics and chaos we hypothesized that such oscillations could be more complex in certain situations, in particular if the cells were stimulated following more temporally complex patterns (as for example with periodic stimuli). Furthermore we hypothesized that the cells periodically stimulated could develop regimes of coordinated oscillations, which could provide important clues on how the cells organize the inflammatory response at a tissue level. To observe such dynamics, time-lapse imaging of the cells for different stimuli was the main experimental tool to use. Hence, the main two objectives of this project was to provide a faithful description of the dynamics of NF-kB in different situations, both at a single cell level and at a population level, and to try to unveil the complexity that the nonlinear dynamics of the NF-kB signalling could display. We believed that our interdisciplinary approach could provide new insights on how to interfere with NF-kB and by extension with the inflammatory process.

Being a physicist, with a background in nonlinear dynamics, since the beginning of the project I have received an extensive training in microscopy and in molecular biology. Once the experimental setup for long imaging of cells with fluorescently tagged NF-kB upon stimulation was ready, in order to have a complete database on the dynamics of NF-kB, we developed a software able to quantify the dynamics of NF-kB for hundreds of cells overcoming the distortions that might be present in this type of experiments. Using this software, we have generated a database in which the behaviour of hundreds of cells upon different stimulations (TNF-alpha and LPS) is analyzed. Furthermore, we have elaborated a framework in order to perform a rigorous time series analysis of the data obtained, that allow us to characterize properly the dynamics of NF-kB in different situations.

In parallel, provided that most mathematical models of NF-kB dynamics were high dimensional systems we decided to propose a simpler low dimensional NF-kB model based in simply three layers of regulation, to have an insight on the origin of the different dynamics observed and to make predictions on NF-kB-driven transcription.

In order to perform experiments with cells periodically stimulated, we have set up a experimental system in which cells can be stimulated through a microfluidic perfusion platform while being imaged. With this system, we have performed experiments on different conditions and used our software in order to quantify NF-kB dynamics. We have used quantifiers from nonlinear dynamics, such as measures of phase synchronization, to describe the dynamics observed. Our mathematical models have been adapted and have been shown to reproduce faithfully the phenomena observed.

As a result of this interdisciplinary work, that included techniques from microscopy, molecular biology, image processing, time series analysis and mathematical models, we have obtained a number of results that soon will be followed by others:

- We have found that the dynamics of NF-kB, even upon simple stimuli, is more complicated than what was previously reported: not only oscillating behaviours are possible, but also non-oscillating. Furthermore, we have found spontaneous activations of the oscillations in absence of stimulation.

- We have shown that a low dimensional NF-kB model based in simply three layers of regulation is able to reproduce the heterogeneity of the dynamic observed. We have also found that the role of stochasticity in gene expression is fundamental, making also theoretical contributions to understand the properties of these systems.

- We have found that for many different conditions it is possible to find coordinated dynamics, i.e. that NF-kB oscillates synchronously. The picture that is emerging from our experiments is more complex than what we expected: in presence of periodic stimuli we find coordinated behaviour that at the same time can be quite heterogeneous among cells. We expect to be able to provide a complete description of these phenomena, also making use of our mathematical models.  

- Finally, we have provided a characterization of a drug, dexamethasone,  in the dynamics upon simple stimuli. We have also found that these drugs not only modify the dynamics, but also the transcription profile. Our final aim is to show how these inflammatory drugs can distort the coordinated dynamics, which is probably the key to understand how to correct the inflammatory response at a tissue level.

Increasing the competitiveness of European research requires multidisciplinary approaches. I obviously share this view and, for this reason, I believe that the present project, which integrates novel concepts of Biology with ideas from Nonlinear Dynamics, has contribute to the increase European reasearch competitiveness. By bringing my ideas to a prestigious institution devoted to Life Science but with less tradition in quantitative biology approaches, like San Raffaele Scientific Institute, I hope to have contributed to foster interdisciplinary research. In this sense, I have made efforts to facilitate the transfer of knowledge by participating in the seminars organized in the institution (also as a speaker), also organizing a seminar on mathematical modeling in biology and even teaching some lessons in the university to Medical and Biotechnology students. On the other hand, I have talked about my research in conferences of Nonlinear Dynamics so I hope to have been able to attract the attention of researchers from those fields to this kind of problems. Concerning wider social implications of this project, chronic inflammatory pathologies have an increasing incidence in Europe, whose population is ageing quickly. I believe that our insights on NF-kB will eventually provide clues for the design of smart therapies against related illnesses, that are expected to have a huge social impact in the next future.


Samuel Zambrano, PhD
San Raffaele University and Scientific Institute
via Olgettina 58
I-20132 Milano, Italy
phone +39 02 2643 3660
fax +39 02 2643 5544