Funded Grants

Nonlinear systems dynamics in differentiation circuits

Complex interactions between genes and proteins form genetic circuits that underlie biological processes in all organisms from simple bacteria to humans. These genetic circuits can control nonlinear dynamic behaviors such as cellular decision-making and differentiation, by integrating extracellular signals and computing appropriate responses. Comprehensive understanding of these biological processes cannot be achieved by investigating single proteins or genes one at a time, but rather require a systems-level analysis of the underlying genetic circuit dynamics. It is therefore essential to uncover the design principles of these gene regulatory circuits and develop a theory that explains the interactions within them and how they function at the systems-level. This information will help us better comprehend and predict circumstances in which gene circuit operation may fail and lead to complex diseases. Such knowledge will be particularly valuable in developing systems-level based treatment of diseases by identifying optimal targets for regulating gene circuit operations. This type of directed therapy could have the added benefit of reducing side effects commonly associated with treatment.

In cells, genetic circuits commonly regulate each other, constituting an additional level of complexity. For simplicity, systems-level studies to date have investigated genetic circuits that control distinct biological processes in conceptual isolation. This approach has been fruitful, and in recent years several groups including us, have been able to develop mathematical theories on how small genetic circuits generate specific cellular behaviors. However, genetic circuits we consider in conceptual isolation interact with each other in the cell, establishing uncharted higher levels of complexity and behavior. Here we aim to develop a mathematical theory describing how two complex gene regulatory circuits that each control mutually exclusive cellular processes, interact with each other. Specifically, this work will establish a mathematical understanding of how cross-regulation between genetic circuits ultimately determines cellular behavior.

Multipotent cellular differentiation is a particularly striking example where interactions between genetic circuits determine cell fate. During this developmental process, individual cells can differentiate into one of several possible states. For example, a hematopoietic stem cell can differentiate among others, into a B-cell, lymphocyte or red blood cell. The choice and progression towards a specific cell fate is typically regulated by complex interactions between distinct differentiation circuits. An ideal simple model system to study how complex cross-regulation between gene regulatory circuits controls multipotent differentiation, is the stress response of the soil bacteria Bacillus subtilis. Under nutrient limiting conditions, individual B. subtilis cells can differentiate into a spore or become competent and take up extracellular DNA. The distinct genetic circuits that control sporulation and competence have been well characterized. However, we currently lack critical understanding as to how these circuits interact with each other and determine the outcome of differentiation.

Previously we utilized a combination of mathematical modeling and single cell quantitative fluorescence time-lapse microscopy, to develop a systems-level theory describing how the competence circuit controls differentiation into competence (Süel et al, Nature, 2006). Through this approach, we demonstrated that the competence circuit constitutes, what is known in the field of nonlinear dynamics as an excitable system. We have shown that, consistent with excitable dynamics, competence is a transiently differentiated state that is initiated in a probabilistic manner. Interestingly, action potentials in neurons are also triggered by an excitable system, emphasizing the generality of nonlinear dynamics concepts in elucidating the behavior of complex biological processes.

Most recently, we performed a comprehensive multi-dimensional analysis of competence circuit function across parameter values, noise levels, and circuit architectures (Süel et al, Science, 2007). The results of this study provided an understanding of the dynamic competence system well beyond its normal operating regime. We showed that the competence circuit possesses remarkable, and inter-related, properties of tunability, robustness, and noise-dependence. This work also provided direct experimental evidence that a differentiation circuit is triggered by noise. Such a probabilistic view of biological processes is a new concept in biology, and bares resemblance to the quantum mechanical revolution in physics that occurred in the beginning of the 20th century.

The systems-level understanding of the competence circuit we have obtained provides a unique starting point to investigate how interactions between the competence and sporulation circuits regulate multipotent cellular differentiation in B. subtilis. We will use continuous and discrete simulations to develop a theoretical systems-level framework describing the interactions between the competence and sporulation circuits. The mathematical modeling will be utilized to generate specific hypothesis regarding the cross-regulation between the two circuits that can be tested experimentally.

On the experimental side, we will measure the dynamics of multiple competence and sporulation circuit components simultaneously in single B. subtilis cells. This will be accomplished by generating B. subtilis strains that contain multiple fluorescent reporter proteins with distinguishable colors to follow the activities of various circuit components. Using our expertise in quantitative multi-color fluorescence time-lapse microscopy we will then record movies of these strains as cells choose and execute distinct differentiation programs. Through image analysis of these movies, we will measure the interaction dynamics of the underlying genetic circuits during differentiation. Additionally, we will perform specific genetic perturbations to “re-wire” the competence and sporulation circuits to experimentally test high-level properties such as circuit memory and cell fate outcome. We are particularly interested in identifying whether genetic circuits that interact with each other, retain memory of their circuit state prior to their interaction. Perhaps, circuits can also retain a memory of the cross interaction itself, that could affect future interactions between circuits. Answers to such high-level questions regarding circuit cross-regulation and dynamics will be extremely valuable in developing a theoretical foundation that describes genetic circuit interactions at the systems-level.

Developing a “theory of genetic circuits” will have the additional benefit of bringing together researchers from many diverse fields. The broad appeal of such a theory is partially based on the fact that this proposed project combines elements from many diverse scientific fields. Aside from biologists, a genetic circuit theory would also be of interest to scientists from the fields of Mathematics, Physics, Computer Sciences and Electrical Engineering.

Finally, one of the most important long term benefits to come from this work would be in the development of novel strategies for the treatment of a broad range of diseases. Among the current problems of drug therapy are the severe side effects associated with treatment. Side effects commonly arise when we target a process in the cell such as cell death, by designing a drug to a key protein that often has numerous interactions with many different proteins and genes in the cell. Due to the many interactions the drug target has with different proteins, however, other unintended critical processes in cells can also be affected. A systems-level understating of genetic circuits could not only improve our understanding and prediction of side effects, but more importantly could also identify new targets with fewer side effects. For example, instead of trying to alter a cellular process by designing a drug to a protein with several critical interactions, we could perturb the same cellular process by identifying and targeting another component of the circuit with far fewer vital interactions. Medical treatment of complex diseases such as cancer would especially benefit from such a novel approach to identifying drug targets for therapy. This work is designed to make complex biological processes such as development, conceptually accessible by building a theoretical systems-level framework to describe the underlying genetic circuit interactions. Ultimately, such high-level insight into fundamental biological processes may lead to new and improved methods of treatment for diseases.