A genetic mutation could help unlock the secrets of the molecular clock
By Jeffrey M. Perkel | September 20, 2006
A human sleep disorder than disrupts circadian rhythms has provided new insights into the post-translational controls governing the molecular clock, according to a new paper in Genes and Development.
Achim Kramer, professor of chronobiology at Charité Universitätsmedizin Berlin, and colleagues mapped 21 in vivo phosphorylation sites in the protein PERIOD (Per) 2, one of which corresponds to the site of a known mutation in human familial advanced sleep phase syndrome (FASPS), a disorder that advances the internal clock by four or five hours, causing people to go to sleep very early in the evening and rise very early in the morning.
The FASPS mutation changes a conserved serine residue (Ser659 in the mouse) to glycine, and the authors demonstrated in a cell culture system that lack of phosphorylation at this site accelerates nuclear export and degradation of the Per2 protein. The team also demonstrated that alternate phosphorylation events can destabilize the protein and developed a mathematical model to explain how perturbations to each of these events can either shorten or lengthen an organism's circadian rhythm.
"This is interesting work," Louis John Ptacek, a Howard Hughes Medical Institute investigator at the University of California San Francisco who was not involved in the study, told The Scientist. "We know that period proteins are phosphorylated, and in multiple spots. What these guys have done is take a systematic approach to figuring out where these sites are."
"How it is that these clock proteins are regulated and how they produce the periodicity that they do is a big question in circadian biology," according to Amita Sehgal, HHMI Investigator at the University of Pennsylvania Medical School, who was not involved in the study. "While we have a basic understanding of the molecular clock, we still don't know how circadian periodicity is generated, and this paper is an important step in that direction. Similar studies with other clock proteins may eventually allow us to build a complete picture of how the circadian period is generated."
PER2, one of a handful of proteins implicated in the molecular clock, forms a complex with cryptochrome proteins that translocates to the nucleus to shut down transcription of a host of clock proteins, including both period and cryptochrome. The delay between when these proteins are expressed and when they are turned off governs the periodicity of the molecular clock. Phosphorylation of period proteins has long been recognized as a key regulator of these events, says Ptacek. In 2001 Ptacek, with UCSF colleague Ying-hui Fu and David Virshup of the University of Utah, mapped the genetic lesion that leads to FASPS, suggesting that the mutation caused under-phosphorylation of Per2 by casein kinase I (CKI)-epsilon, thereby stabilizing Per2 and fast-forwarding the molecular clock. "We had the molecular mechanism backward," Virshup told The Scientist.
"We found that, yes, there are phosphorylations that trigger degradation, but this [Ser659 phosphorylation event] does not," Kramer said. "We propose discernable roles for different phosphorylations." The Ser659 phosphorylation event actually stabilizes the protein, whereas another, currently unidentified phosphorylation event acts to promote cytosolic degradation via the proteasome, he explained.
Kramer and his team postulate that, in the absence of cryptochrome, PER2 is rapidly phosphorylated by CKI-epsilon at this second site and degraded. When complexed to cryptochrome, however, the complex is phosphorylated at the FASPS and downstream sites, thereby holding it in the nucleus and stabilizing one or both proteins. Kramer's lab is now working to pin down which of the 20 other phosphorylation sites on Per2 could serve this second, degradation-promoting function.
Kramer used a mathematical model to understand how the interplay between two phenotypically distinct phosphorylation events could affect circadian rhythm, and applied that model to some confusing circadian mutations. It was long unclear, for instance, how two mutations in a single gene, the Drosophila CKI-epsilon homolog, DOUBLETIME, could lead to completely different phenotypes: one mutation lengthens the circadian rhythm while the other shortens it. Kramer's data suggests that these mutations differentially affect the kinase's ability to phosphorylate at one of the two sites, thereby leading to distinct circadian phenotypes. The model likewise explains the hamster tau mutant as a defective CKI-epsilon kinase that can phosphorylate the degradation-promoting modification site, but not the stabilizing (FASPS) site - a prediction that differs somewhat from a recent report from Virshup's group, which shows that tau is actually a gain-of-function mutant.
"Our hypothesis was that the tau mutation increased the degradation rate of the period protein by increasing phosphorylation at a different site, and Kramer's conclusion seems to be that tau is a loss of function, but that it's differentially affecting nuclear export versus degradation," Virshup said. He added, "Since we showed experimentally that there's an increase in activity, I think their data can be understood by an increase on the first site, rather than a decrease on the second site."
Noting that caution must be taken when applying cell culture-derived data to living organisms, Ptacek said his lab has now generated a mouse model of the FASPS mutation. "We're finding some things that agree with Kramer, and some things that are a little different," he said. "This is an example where a mutation in a cell culture system agrees with some of the things we learn from a living mouse, but there are differences as well."
Jeffrey M. Perkel
Links within this article:
K. Vanselow et al., "Differential effects of PER2 phosphorylation: molecular basis for the human familial advanced sleep phase syndrome (FASPS)," Genes Dev, published online Sept. 18, 2006. DOI: 10.1101/gad.397006
Louis John Ptacek
M. Gallego et al., "An opposite role for tau in circadian rhythms revealed by mathematical modeling," Proc Natl Acad Sci, 103:10618-23, 2006.