PERSPECTIVES
of the day, wild-type
mice will quickly show
precisely timed bouts
Forebrain
rebrain
eb
of daytime activity that
anticipate the availNB
ability of food. The
per,
pe
per
er, ccrrryy
NPAS2-deficient mice
were slow to adapt
SCN
SC
C
their behavior to the
Activity
feeding change and as
CB
Sleep
per,
pe
per
er, ccry
rryy
a result lost weight and
Feeding
became sick.
Light
The molecular clock
in the forebrain responds rapidly to a reCooperation among circadian oscillators. Circadian oscillators in the
SCN and in the forebrain regulate rhythms of locomotor activity, sleep, stricted feeding signal
and feeding. In the SCN, CLOCK (C, orange) and BMAL1 (B, blue) form by altering the patterns
heterodimers that influence rhythmic transcription of target genes, such of gene expression in
as per and cry. In the forebrain, rhythmic transcription is regulated by the clock loop (while
NPAS2 (N, yellow), which forms heterodimers with BMAL1. Light acts on the clock loop of the
the SCN, whereas other rhythmic signals (such as food availability) in- SCN remains unfluence the forebrain clock. The two clocks are probably coupled, but the changed). These findings suggest that the
molecular details of this coupling remain unclear.
signals from restricted
periods of inactivity and only very short feeding are deciphered by the forebrain
bouts of sleep.
clock, whereas the daily light:dark cycle
The NPAS2-deficient mice also showed acts primarily on the SCN. It is probable
interesting changes in their ability to be- that the forebrain and the SCN, responding
come entrained to environmental signals. to different inputs, together regulate normal
Whereas wild-type animals took several entrainment of circadian behavior to diverse
days to adjust to an advancing shift of the temporal signals in the environment.
light cycle, the NPAS2-deficient mice
The circadian phenotype of NPAS2-defishowed rapid advances on the first day of cient mice is perhaps more subtle than those
the new regime. On the other hand, they of mice carrying other circadian mutations
showed delayed responses to restricted feed- or gene deletions that abolish rhythmicity
ing. Although mice normally feed during entirely or dramatically change its period.
their active period at night, if access to food The new phenotype is nonetheless instrucis restricted to a short period in the middle tive. Its very subtlety emphasizes the impor-
tance of painstaking analysis of whatever assay is being used to measure rhythmicity. In
previous studies, little attention has been
paid to the fine structure of locomotor behavior patterns, and important findings may
have been missed. More importantly, the
work of Dudley et al. reveals the first tissuespecific specialization within the circadian
molecular machinery in mammals. It seems
unlikely that this is the only such variation
on the circadian molecular theme. Oscillators in other areas of the central nervous
system and in the periphery may well vary in
subtle ways that affect their tissue-specific
functions or their coupling to the rest of the
circadian system. The findings of Dudley et
al. represent an important step in what will
be a difficult but rewarding analysis of the
system that has been called the “biological
clock” but might be better characterized as
the “circadian temporal program.”
References
1. S. Yamazaki et al., Science 288, 682 (2000).
2. M. R. Ralph, R. G. Foster, F. C. Davis, M. Menaker,
Science 247, 975 (1990).
3. S. M. Reppert, D. R. Weaver, Nature 418, 935 (2002).
4. D. M. Berson, Trends Neurosci. 26, 314 (2003).
5. M. Reick, J. A. Garcia, C. Dudley, S. L. McKnight, Science
293, 506 (2001).
6. C. A. Dudley et al., Science 301, 379 (2003); published
online 3 July 2003 (10.1126/science.1082795).
7. M. W. Young, S. A. Kay, Nature Rev. Genet. 2, 702
(2001).
8. N. Gekakis et al., Science 280, 1564 (1998).
9. M. H. Vitaterna et al., Science 264, 719 (1994).
10. M. K. Bunger et al., Cell 103, 1009 (2000).
11. J. A. Garcia et al., Science 288, 2226 (2000).
Published online 3 July 2003;
10.1126/science.1087824
Include this information when citing this paper.
Snapshots of Water at Work
William H. Robertson, Eric G. Diken, Mark A. Johnson
he molecular structure of water is
simple. But compared to most molecular liquids, the degree of deformation of the H2O molecule changes dramatically as it fluctuates in and out of the cooperative hydrogen bonds afforded by extended networks. Add an ion, which can further
distort the sheath of water surrounding it,
and you have a seriously complex situation.
Two reports in this issue address the dynamical aspects of how water responds to
the presence of a charged solute and how the
mechanism of a chemical reaction between
two negatively charged reactants depends on
their initial separation in solution. On page
T
The authors are at the Sterling Chemistry Laboratory,
Yale University, New Haven, CT 06520, USA. E-mail:
[email protected]
320
347, Omta et al. (1) use ultrafast lasers to establish the unique character of water molecules in the hydration shell around an anion.
And on page 349, Rini et al. (2) show how
the hydration shell participates in a photoinitiated acid-base reaction by shuttling
the proton between reaction partners. Both
studies demonstrate the power of time-resolved infrared spectroscopy to isolate the
contribution of different regions of the solution to bulk behavior (in this case viscosity
and reaction rate, respectively).
Omta et al. (1) call into question the
meaning of the traditional notion (3) that
some aqueous ions are “structure makers”
while others are “structure breakers.” To
quantitatively address the issue of soluteimposed structure on water, they argue that
a measurable quantity related to structure is
18 JULY 2003
VOL 301
SCIENCE
the viscosity. The viscosity usually controls
diffusive motions, including orientational
diffusion. Hence, interrogation of the orientational relaxation of water molecules as
a function of distance from an ion—in this
case, ClO4−—should reveal variations in
the local structure.
By carrying out this measurement in the
infrared region of the OH stretching vibrations, Omta et al. selectively probe water
molecules that reside either in the primary
hydration shell or in the more distant bulk
water. For water molecules near the ClO4−
ion, the relaxation times are long, as one
might expect for molecules that are strongly anchored to the anion. Surprisingly,
however, for all other water molecules detected in the experiment (away from the
ion), the rate is found to be identical to that
of the pure liquid and independent of the
concentration of the solute.
The dominant effect of the ion on water
structure thus appears to be restricted to the
immediate vicinity of the ion, rather than
propagated into the fabric of the surround-
www.sciencemag.org
CREDIT: KATHARINE SUTLIFF/SCIENCE
C H E M I S T RY
PERSPECTIVES
−
ing liquid. But the ClO4 ion is known to the acetic acid product. They thus provide
increase the macroscopic viscosity of an unusually complete picture of how all
aqueous solutions. This paradox is resolved species in the reaction evolve.
by noting that a larger effective radius of
The resulting time profiles clearly difthe ions would account for the increased ferentiate between reactions that occur
viscosity (4). The required radius corre- through initially hydrogen-bonded reactant
sponds nicely to the size of the ion and its pairs and those that occur by diffusion of
associated hydration shell. Thus, the hydra- the reactants to form a collision complex.
tion shell seems to act like a chemically At low concentration, Rini et al. observe a
distinct species whose overall size controls third pathway, in which the photo-acid dothe macroscopic viscosity.
nates a proton to the water molecules in the
The localized character of the hydration hydration shell to form the hydronium ion.
shell around a solute anion naturally raises This species then migrates to find and neuthe important question of its molecular tralize the acetate ion long after the appearstructure. Recent studies (5) of ions sur- ance of the photo-acid’s conjugate base.
rounded by a controlled number of water Water thus plays an integral role in the remolecules have shed light on this issue. For action by shuttling the proton between the
example, Robertson et al. (6) have identi- reaction partners.
fied the intrinsic structures of the
hydration shells around the hydroxide and fluoride ions, and
have started to map how water interacts with complex anions such
as acetate (see the figure) (7).
These organic species are more
challenging because they possess
both hydrophilic and hydrophobic
domains (3).
Water structure also plays a key
role in the kinetics of acid-base reactions. Rini et al. (2) present an
elegant ultrafast study of a photoinitiated proton transfer between
an anionic proton donor and the Water in action. Schematic illustration of how the hydraacetate-ion proton acceptor. With a tion shell around the anionic part of the acetate ion paruniversal infrared probe, the au- ticipates in the reaction with H3O+. Rini et al. (2) provide
thors can independently monitor evidence for this proton shuttle mechanism. Omta et al.
the decay of the photo-acid and the (1) show that the effect of an anion on water structure is
rise of both its conjugate base and highly localized.
The two studies (1, 2) illustrate how detailed mechanistic insights into aqueous
processes can be obtained with ultrashort
time resolution, which allows researchers
to look beneath the averaging inherent in
bulk measurements. Together with cluster
studies, these ultrafast techniques provide
rigorous tests of widely accepted molecular-level pictures (8, 9).
One of the longstanding challenges in
aqueous chemistry, raised by the work of
Rini et al., is the molecular-level mechanism of proton transport (8, 10, 11). An
important step toward this goal is to extend the ultrafast infrared methods demonstrated in the bulk studies reported here (1,
2) to the cluster regime. The advantages of
the small systems are that the reactions
can be initiated with precisely controlled
energy and starting geometry, and the dynamics can be treated with the powerful
theoretical tools refined in the study of
molecular photochemistry.
References
1. A. W. Omta, M. F. Kropman, S. Woutersen, H. J. Bakker,
Science 301, 347 (2003).
2. M. Rini, B.-Z. Magnes, E. Pines, E. T. J. Nibbering,
Science 301, 349 (2003).
3. Y. Marcus, Ion Solvation (Wiley, Chichester, UK, 1985).
4. J. K. G. Dhont, An Introduction to Dynamics of
Colloids (Elsevier, Amsterdam, 1996).
5. P. Ayotte, G. H. Weddle, J. Kim, M. A. Johnson, Chem.
Phys. 239, 485 (1998).
6. W. H. Robertson, E. G. Diken, E. A. Price, J.-W. Shin, M.
A. Johnson, Science 299, 1367 (2003).
7. W. H. Robertson et al., J. Phys. Chem. A, in press.
8. L. Onsager, Science 166, 1359 (1969).
9. W. L. Jorgensen, J. Chandrasckhar, J. D. Madura, R. W.
Impey, M. L. Klein, J. Chem. Phys. 79, 926 (1983).
10. U. W. Schmitt, G. A. Voth, J. Chem. Phys. 111, 9361
(1999).
11. J. P. Cowin et al., Nature 398, 405 (1999).
PA L E O N T O L O G Y
Making the Best
of a Patchy Fossil Record
Andrew B. Smith
aleontologists have been striving to
document the history of species diversity through geological time for nearly
150 years. Until recently, the favored method
involved nothing more complex than counting the fossils recorded from each time interval, and a great effort has gone into compiling taxonomic and stratigraphic data to
quantify diversity trends through time.
However, this simple approach comes
with a price: It is only valid if each time in-
P
The author is in the Department of Palaeontology,
The Natural History Museum, London SW7 5BD, UK.
E-mail: [email protected]
terval has been sampled to the same extent.
On page 358 of this issue, Crampton et al.
(1) show that this may not be the case.
Working from a massive database of fossil
marine molluscs and fossil localities in
New Zealand covering the past 60 million
years, they point out that sampling effort is
far from uniform. Furthermore, sampling
effort correlates with the amount of rock
that crops out at the surface.
Paleontologists have long been aware
that the more thoroughly a rock unit is
sampled, the more specimens are found.
This, in turn, leads to higher recorded diversity: All other things being equal, if on-
www.sciencemag.org
SCIENCE
VOL 301
ly a small amount of rock is present at outcrop, then the total diversity recorded is
likely to be much lower than if a large area
of outcrop is available for sampling.
Rarefaction, a statistical technique that
uses subsampling to reduce populations to
the same size, can be used to remove such
bias. Initial large-scale studies with this approach (2) suggest that global biodiversity
patterns may indeed be substantially biased
by unequal sampling. However, rarefaction
requires more data than are generally available in global taxonomic compilations.
Hence, more indirect ways of estimating
sampling levels are required. Two have recently been used.
First, estimates of how extensively a time
interval has been sampled have been based
on the number of named geological formations (that is, mappable units of rock laid
down in a distinct past environmental setting). The more geological formations that
are recorded in a time interval, the greater
the range of environments that exist to be
18 JULY 2003
321