Circadian clocks are seeing the systems biology light
© BioMed Central Ltd 2005
Published: 28 April 2005
Skip to main content
© BioMed Central Ltd 2005
Published: 28 April 2005
Circadian rhythms are those biological rhythms that have a periodicity of around 24 hours. Recently, the generation of a circadian transcriptional network - compiled from RNA-expression and promoter-element analysis and phase information - has led to a better understanding of the gene-expression patterns that regulate the precise 24-hour clock.
More than 30 years ago, Seymour Benzer started on the quest for the holy grail of behavioral neuroscience, the elucidation of behavior at a molecular and systems level . In part because of his efforts, this quest is perhaps furthest along in the study of biological rhythms, which in mammals are most clearly manifested in the regulation of a very primitive behavior, the sleep-wake cycle. Elegant genetic and biochemical studies in several species have revealed that the circadian clock that controls such daily rhythms is a cell-autonomous transcriptional/translational feedback loop (reviewed in ). In mammals, the master circadian clock resident in the suprachiasmatic nucleus (SCN) of the hypothalamus functions to synchronize other oscillators that drive physiological outputs to a 24-hour rhythm. Despite increasingly refined models of how individual clock components function together as a self-sustaining oscillator, the link between their action, transcriptional oscillations (or clock output), and dependent processes such as physiology and behavior has remained elusive.
Enter systems biology, which can be broadly defined as the integration and synthesis of information from various subfields to inform a biological question . In this field, changes to biological systems are observed at multiple levels under a set of experimental condition(s). Integration of complex data, such as RNA and protein levels together with phenotypes, facilitates the construction of prospective models, which can inform and be informed by experimental data. The methodologies used may include, but are not limited to, transcriptional profiling, differential proteomics, cell-based screening, and whole-organism phenotypic screening . These studies often produce information-rich datasets that necessitate the use of bioinformatics tools to organize and manage the information and to synthesize testable hypotheses.
As a nascent field, many of the initial studies could be viewed as hypothesis-generating experiments. Transcriptional profiling, proteomic screens, and in silico studies in themselves merely capture a snapshot of data coincident with a biological process. As the field has progressed, however, studies have become more refined, and involved the interplay between hypothesis generation and testing.
After initial studies in model systems such as Arabidopsis and Drosophila, several authors have applied a popular systems-biology tool - transcriptional profiling - to the study of mammalian circadian transcriptional output [4–8]. Transcriptional profiling, which is usually accomplished using DNA arrays, was employed to identify batteries of genes and biological systems that are controlled by the master clock in the SCN [4, 9], as well as rhythms regulated by peripheral clocks in liver, kidney, and heart [4–8]. Although the core circadian activators, Bmal1 and Clock, and the core repressors, the cryptochromes, function analogously in these tissues, their downstream targets vary between different tissues. For example, analysis of rhythmic genes in liver revealed their principal roles to be regulation of metabolism, whereas genes cycling in the SCN were primarily involved in signaling and neurosecretion. These studies also uncovered, to a varying degree, a central battery of genes that are clock-regulated in every tissue and can be thought of as first-order clock-controlled genes (CCGs); some of these, such as Bmal1, the cryptochrome Cry1, the period homolog Per2, and the nuclear receptor Nr1d1 are components of the clock itself [4, 6–8]. More recent profiling studies have used genetic models to refine both the roles of these core components and their outputs. Using animals with a functionally deficient locus for several of the known circadian regulators, such as Clock mutant and cryptochrome-deficient Cry1 -/- Cry2 -/- mice, studies have shown both the specificity and redundancy of various core components and the effect of their deletion on rhythmic behavior and transcription .
An important aspect of circadian biology is the requirement for rhythmic output throughout the entire 24-hour day. How does the clock regulate gene expression throughout the entire day given a limited number of regulatory elements? This is known as the phase-control problem. Ueda et al.  elegantly demonstrate how complex phase regulation can be accomplished using combinations of three clock-regulated elements: E/E' boxes, D boxes, and RREs. These experiments show how constructive and destructive interference can be used to generate new phases and amplitudes of circadian transcription. Ueda et al.  used in silico methods to construct models that accurately reflect the observations that cycling genes can have low-amplitude or high-amplitude oscillations. The phase of these oscillations can shift forwards or backwards depending on which cohort of genes is regulating their expression. Using this foundation of knowledge, it was possible to construct a model of the circuit of the circadian feedback system. This model was then probed to find the Achilles' heel of the transcriptional circuit underlying the mammalian circadian clock. These experiments supported the proposed model, and concluded that the E/E' box plays the critical role in the regulation of circadian transcription.
The advent of systems biology has allowed the elucidation of biological features such as behavior. Complicated feedback loops can be decoded, allowing the identification of central regulators of the system and those controlling more subtle processes. With respect to circadian behavior, these studies have reinforced the importance of the E/E' box regulators Clock, Bmal1, Cry1 and Cry2. The true importance of these studies, however, lies in the construction of sophisticated models of regulatory systems that can be experimentally tested. The continued application of these approaches, coupled with rigorous experimental testing to confirm prospective modeling, provides a remarkable opportunity to explain how behavior can result from relatively simple transcriptional networks. Benzer's quest continues, and indeed, the magnitude of the task is only now becoming apparent. Despite this, Ueda et al.  are leading the way in helping us see how systems biology can shed light on the circadian clock.