Research Themes

Modeling the circadian clock : From molecular mechanism to physiological disorders

Rhythmic phenomena occur at all levels of biological organization, with periods ranging from less than a second to years. Among these, circadian rhythms occur with a period of about 24 h (from latin "circa" about and "dies" day) and play a key physiological role in the adaptation of living organisms to their periodically varying environment. Experimental advances during the last decade have permitted to largely unravel the molecular bases of circadian rhythms in a number of organisms, such as Drosophila, Neurospora, cyanobacteria, plants and mammals, including humans.

Oscillatory behavior often originates at the cellular level from regulatory feedback loops which involve many parameters and interacting variables. Relying only on sheer intuition to predict the dynamics of such complex regulatory systems rapidly meets with limitations. Analyzing the origin of oscillations has therefore much to gain from theoretical models closely related to experimental observations. Some of the roles and advantages of theoretical models in biology are listed here. These considerations on the use of theoretical models apply to the study of biological processes in general, but pertain with particular weight to biological rhythms which only occur in precise conditions. Determining these conditions is a primary goal of a modeling approach.

Computational models for circadian rhythms were at first borrowed from the physical literature, as exemplified by the use of the van der Pol oscillator for modelling properties of circadian oscillations. This line of research is still pursued to study, for example, the effect of light on the human circadian system. Besides these abstract models, a complementary approach rests on the study of molecular models - based initially on the Goodwin model - that are more directly related to the biochemical regulatory processes that underlie circadian rhythms.

In nearly all cases investigated so far, it appears that a central role in the mechanism of circadian rhythmicity is played by autoregulatory loops of negative feedback on gene expression. Based on the experimental observations available on the molecular mechanisms in fruit fly we initially examined computational models for the Drosophila circadian clock. Incorporating the effect of light in the circadian mechanism allows the comparison of theoretical predictions with experimental data in regard to several properties, including oscillations in continuous darkness or light, entrainment by light-dark cycles, and phase shifting by light pulses. We also use the model to address the transient or permanent suppression of circadian rhythmicity by critical pulses of light. Finally, the model show that sustained oscillations only occur in a restricted domain of parameter values. Thus the transcription of the various clock genes does not necessarily lead to circadian rhythmicity. The production of a robust circadian rhythm requires that these genes be expressed at the appropriate levels ensuring that the genetic control system operates within a domain of sustained oscillations.

Since the discovery of the first mammalian homolog genes in 1997, remarkable progress has been made in the understanding of the molecular mechanism of the mammalian circadian clock. Based on experimental data, we developed computational models for the mammalian circadian clock based on positive and negative regulatory feedback loops. These models account for a variety of experimental observations such as for the phase relationship between the mRNAs or proteins, the entrainment by light-dark cycles, the phase shifts induced by light pulses, or the multiple roles of phosphorylation in the clock. The analysis of our computational model also highlights the possible existence of multiple oscillators within the genetic regulatory network controlling circadian rhythms. Theoretical results show the possibility of sustained oscillations in the absence of the negative loop involving the PER proteins. Such prediction was confirmed experimentally several years later with the triple PER knockout mice showing some rhythmicity.

Based on genetic and biochemical advances on the molecular mechanism of circadian rhythms, we used our computational model for the mammalian circadian clock to examine the dynamical bases of circadian-clock-related physiological disorders in humans. We investigated "internal" perturbations of the clock leading to several sleep phase disorders such as the Familial Advanced Sleep Phase Syndrome (FASPS), the Delayed Sleep Phase Syndrome (DSPS), or the Non-24 h Sleep-Wake Syndrome. We also investigated "external" perturbations of the clock such as jet lag or chronic jet lag. It becomes important to clarify the conditions for entrainment and for its failure because perturbations or dysfunctions of the circadian clock may lead to numerous physiological disorders, which pertain not only to the sleep-wake cycle but also to mood and cancer.