http://www.the-scientist.com/yr2001/jul/research_010709.html (you must registere to read this article)

The Scientist 15[14]:16, Jul. 9, 2001

RESEARCH

The Quest for Perfect Timing
Molecular intricacies of animal model clocks provide insight to humans

By Karen Young Kreeger

Researchers have pondered, and investigated, for decades why one person is
alert and productive at 6 a.m. while another can't even focus before noon.1
But now, their persistence is paying off: chronobiologists, those who
investigate circadian rhythms, or daily clocks, are finally making concrete
links between sleep patterns in humans and a menagerie of well-studied
animal models.

As with many behavioral studies, it's the extreme or unusual cases that
eventually inform scientists about normal processes. A few years ago, a
woman complained to investigators at the University of Utah in Salt Lake
City that she felt an overwhelming desire to fall asleep around 7:30 p.m.,
and wake up before dawn, around 4:30 a.m. And she was not the only
one--other members in her family had the same problem. This started Utah
investigators on a research project that culminated in the first study to
link a human genetic syndrome to what others had been discovering in animal
clock-gene investigations. These family members, eventually recruited into
the study, suffer from familial advanced sleep phase syndrome (FASPS).

Earlier this year, senior authors Louis Ptacek, a Howard Hughes Medical
Institute (HHMI) investigator and a professor in the departments of
neurology and human genetics at Utah, and university colleague Ying-Hui, a
research associate professor of neurobiology and anatomy, published their
study describing the exact mutation involved in FASPS.2 People affected by
FASPS have a mutation within a kinase-binding region of the hPer2 gene near
the telomere of chromosome 2. And, it turns out that hPer2 is a human
homologue of the period gene found in fruit flies, which encodes a key
protein in the Drosophila time-keeping mechanism. But understanding why
this connection is so significant requires a look at the genetics of the
fruit fly's internal clock.

Fruit Fly Chronobiology 101
 

What makes the Utah study especially interesting, says Michael Young, a
genetics professor at Rockefeller University in New York and director of
its National Science Foundation Center for Biological Timing, is what it
uncovers about the regulation of normal human wake-sleep cycles. "In the
beginning there was period," he quips. The fruit fly gene period was cloned
in his lab and that of Michael Rosbash and Jeffrey Hall at Brandeis
University in 1984. In the 1990s, a series of genes and proteins that make
up the fruit fly clock were also cloned. So far, seven of these have been
characterized in a number of labs3: period, timeless, clock, cycle (Bmal-1
in mammals), double-time (which encodes casein kinase 1Î, shaggy (GSK-3 in
mammals), and vrille.4­15

Central to the molecular mechanism of the fly's clock is the diurnal
production, movement into the nucleus, and breakdown of the two-protein
complex of TIMELESS and PERIOD. The transcription factors CLOCK, CYCLE, and
VRILLE control production; these factors flip the genetic switch on and off
for ultimately making TIMELESS and PERIOD proteins. Added to the mix are
the kinases DOUBLE-TIME and SHAGGY, which are enzymes that attach phosphate
groups to PERIOD and TIMELESS, respectively.

When the period and timeless genes are first activated, TIMELESS proteins
start to accumulate, but PERIOD proteins hold off a bit. Initially, PERIOD
proteins bind with the kinase DOUBLE-TIME because it's always abundant in
cells. This partnership leads to PERIOD'S phosphorylation and degradation,
which continues all day long. By early evening, TIMELESS proteins become so
plentiful that PERIOD has two potential partners--TIMELESS and DOUBLE-TIME.
When PERIOD binds with TIMELESS, it's protected from DOUBLE-TIME and
therefore from breakdown.

DOUBLE-TIME still binds to PERIOD, and eventually the now three-protein
complex makes it into the nucleus. Here the complex signals the cell to
start decreasing production of PERIOD and TIMELESS proteins. This process
finally ends at sunrise. Light-sensitive TIMELESS is eliminated from the
cell, leaving PERIOD and DOUBLE-TIME alone once again. This in turn leads
to the phosphorylation and breakdown of PERIOD, and the cycle starts anew.

Essentially, says Young, "it's a very delicate balance between the two
kinases to set the right rate of the clock." The phosphorylation of
TIMELESS by SHAGGY has the opposite effect of what DOUBLE-TIME does to
PERIOD protein. PERIOD's phosphorylation makes the clock go slower early in
the cycle by removing PERIOD protein and SHAGGY phosphorylation of
TIMELESS, then makes the clock go faster by pushing the protein complex
into the nucleus more quickly.

A mutation in the casein kinase 1Î binding site of the hPER2 protein
somehow advanced the clocks of the FASPS family. Utah's Fu theorized how
this might work, considering the scenario worked out in flies. "Our working
hypothesis is that usually hPER2 is highly regulated by CK1Î through
phosphorylation," he says. "In the normal case, CK1Î will just
phosphorylate PERIOD and make it degrade really fast." With the mutation,
the hPER2 protein becomes a poor substrate for CK1Î, causing fewer
phosphates to be added. This makes the hPER2 protein more stable than in
the normal cell. The hPER2 protein accumulates faster, enters the nucleus
prematurely, thereby advancing the internal clock.

Although the clock genes present in flies have human homologues, salient
differences exist. In mammals, the clock is more complex, having more of
each type of gene­three periods, for example. Then, rather than TIMELESS
being PERIOD's partner in the protein complex, it seems most likely that a
mammalian protein called cryptochrome is PERIOD's partner. In the fly,
cryptochrome is a photoreceptor for the clock, connecting the clock to
light from the surrounding environment, but in a different way. In fact,
Amita Sehgal, associate professor of neuroscience and an HHMI assistant
investigator at the University of Pennsylvania School of Medicine (and a
former postdoc with Young) is currently looking into cryptochrome's role in
the breakdown of the TIMELESS protein.

Jump to Mammals

It wasn't until the late 1990s that researchers cloned mammalian clock
genes. "There are now at least eight that we know are intimately involved
in the internal clockwork," says Steve Reppert, chairman of neurobiology at
the University of Massachusetts in Worcester. Reppert's lab at the
Massachusetts General Hospital, where he is also a professor at Harvard
Medical School, was involved in figuring out that mammalian timeless
doesn't perform the same role as timeless in fruit flies and that the
cryptochromes have taken over timeless functions in mammals.16,17

The master clock in mammals is located in the suprachiasmatic nucleus
(SCN), a group of about 10,000 cells in the hypothalamus at the brain's
base. About five years ago, Reppert's lab began to discover how the SCN
clock was put together at the molecular level. They found that the several
feedback loops in mammals are very similar as to what is proposed in
flies.18 The SCN receives direct enervation from the retina, and the
light-dark cycle is the primary way that the clock is coordinated to a
24-hour day. In May, Reppert's group described how the three period genes
in mammals function in the SCN clock.19 "The bottom line is that the Per1
and Per2 genes have nonoverlapping, essential functions for the clockwork,
while Per3 is not necessary," he says.

These incremental discoveries about how the SCN works at a molecular level
might one day help in understanding the sleep problems of Alzheimer's
disease patients. In another recent study, David Harper, assistant director
of research at the geriatric psychiatry program at McLean Hospital in
Belmont, Mass., which is part of the Harvard Medical School, tracked core
body temperature and activity levels of 38 elderly people with probable
Alzheimer's disease (AD). After they died, he examined their brain tissue
and found that 11 of them had varying types of non-AD dementia diseases of
the fronto-temporal region of the brain.20

"What was surprising was when we saw the data for the non-AD patients, they
had a normal pattern of core body temperature, but the rest activity was
entirely disorganized," he says. "It almost looked like they had SCN
lesions. The activity for these patients looked like their circadian system
wasn't working, but their core body temperature regulation was. But in AD
patients, it looked like the circadian system was working but was unable to
keep the correct time." Core body temperature is a direct measurement of
activity in SCN cells; whereas, rest activity measures movement of patients
every five minutes and is dependent on many factors.

Knowing exactly how studies like Utah's genetic investigation of FASPS and
basic research in model animals will affect sleep disorder treatments is
still far off, says Ptacek. "The simple answer is that we can't really
manipulate the system. We can't adjust the clock unless we understand it
better. But our hope is that in understanding the ticking of the clock, the
working of those gears, then we'll be able to test compounds in
combinatorial libraries, and maybe design rational strategies from this
knowledge." Karen Young Kreeger (kykreeger@aol.com) is a contributing
editor for The Scientist. References 1. R. Lewis, "Chronobiology
researchers say their field's time has come," The Scientist, 9[24]:14, Dec.
11, 1995.

2 K.L. Toh et al., "An hPer2 phosphorylation site mutation in familial
advance sleep phase syndrome," Science, 291:1040­3, 2001.

3. Clock and cycle were isolated independently by Rosbash and Hall; Steven
Kay from the Scripps Research Institute; and Isaac Edery from Rutgers
University. The remaining genes were described by Young's lab; and in the
case of timeless in collaboration with Amita Sehgal from the University of
Pennsylvania and Charles Weitz from Harvard Medical School.

4. J.E. Rutila et al., "CYCLE is a second bHLH-PAS clock protein essential
for circadian rhythmicity and transcription of Drosophila period and
timeless," Cell, 93:805­14, 1998.

5. R. Allada et al., "A mutant Drosophila homolog of mammalian Clock
disrupts circadian rhythms and transcription of period and timeless," Cell,
93:791­804, 1998.

6. T. K. Darlington et al., "Closing the circadian loop: CLOCK-induced
transcription of its own inhibitors per and tim," Science, 280:1599­603, 1998.

7. K. Bae et al., "Circadian regulation of a Drosophila homolog of the
mammalian Clock gene: PER and TIM function as positive regulators,"
Molecular Cell Biology, 18:6142­51, 1998.

8.M. Myers et al., "Positional cloning and sequence analysis of the
Drosophila clock gene timeless," Science 270, 805­8, 1995.

9. A. Sehgal et al. "Rhythmic expression of timeless: A basis for promoting
circadian cycles in period gene autoregulation," Science, 207:808­11, 1995.

10. N. Gekakis et al., "Isolation of timeless by per protein interaction:
Defective interaction between timeless protein and long-period mutant
perl," Science, 207:811­15, 1995.

11. J.L. Price et al., "double-time is a novel Drosophila clock gene that
regulates period protein accumulation," Cell, 94:83­95, 1998.

12. B. Kloss et al., "The Drosophila clock gene double-time encodes a
protein closely related to human casein kinase 1Î, Cell, 94:97­107, 1998.

13. N. Nadoo et al., "A role for the proteosome in the light response of
the timeless clock protein, Science, 285:1737­41, 1999.

14. S. Martinek et al., "A role for the segment polarity gene shaggy/GSK-3
in the Drosophila circadian clock," Cell, 105:769­79, 2001.

15. J. Blau and M.W. Young, "Cycling vrille expression is required for a
functional Drosophila clock," Cell, 99:661­71, 1999.

16. A.L. Gotter et al., "A time-less function for mouse Timeless," Nature
Neuroscience, 3: 755­6, 2000.

17. K. Kume et al., "mCRY1 and mCRY2 are essential components of the
negative limb of the circadian clock feedback loop," Cell, 98: 193­205, 1999.

18. L.P. Shearman et al., "Interacting molecular loops in the mammalian
circadian clock," Science, 288:1013­9, 2000.

19. K. Bae et al., "Differential functions of mPer1, mPer2, and mPer3 in
the SCN circadian clock," Neuron, 30, 525­36, 2001.

20. D.G. Harper et al., "Differential circadian rhythm disturbances in men
with Alzheimer disease and frontotemporal degeneration," Archives of
General Psychiatry, 58(4):353­60, 2001.
 

The Scientist 15[14]:16, Jul. 9, 2001