S 8 | N A T U R E | V O L 4 9 7 | 2 3 M A Y 2 0 1 3

B
urning the midnight oil can leave you
tired and grumpy the next day, dull-
ing your mind and slowing your reac-
tion times. But lack of sleep has consequences
beyond the brain as well, with long-term sleep
disturbances leading to metabolic problems.
Matthew Brady, a biologist at the University
of Chicago in Illinois who studies the links
between sleep and metabolism, puts it simply:
Fat cells need their sleep as well.
Many large population studies over the
past decade have found that people who sleep
poorly are more likely to be obese and to have
metabolic disorders such as type 2 diabetes.
One study of more than 1,000 people found that
sleeping for only 5 hours rather than 8 hours
results in a 3.6% increase in body mass index.
There is evidence that its this lack of sleep
that causes metabolic disorders and weight
gain. In 1999, Eve Van Cauter and her col-
leagues at the University of Chicago showed
that sleep restriction in healthy young men led
to signs of insulin resistance, which can lead
to type 2 diabetes. This finding changed the
entire world of sleep research, says Fred Turek,
who studies the biology of circadian rhythms at
Northwestern University in Evanston, Illinois.
Before Van Cauters research, the only known
effect of a lack of sleep was that you were tired,
Turek says, but the study revealed that people
getting too little sleep are at risk of obesity, dia-
betes and other metabolic disorders.
The underlying biology is slowly becoming
clear. Van Cauters work helped to show that
two hormones leptin and ghrelin are
likely to be involved in the link between sleep
and weight gain. Leptin is produced by fat cells
and is a signal of satiety; ghrelin is produced
by the stomach and signals hunger. Together,
these hormones regulate hunger and appetite.
Van Cauters team later showed that restrict-
ing the sleep time of healthy young men in the
lab caused their leptin levels to fall and their
ghrelin levels to rise, increasing their appetite,
especially for fatty and sugary foods
1
.
The biochemical pathway that leads from
lack of sleep to changes in leptin and ghrelin
levels is still being investigated, says Esra Tasali
of the University of Chicago, who worked with
Van Cauter on the leptinghrelin study. Tanta-
lizing clues emerged from a study published in
2012 by Tasali, Brady and Van Cauter showing
that reduced sleep leads to increased insulin
resistance in fat cells. The effect was huge:
after four nights of just 4.5 hours of sleep, the
fat cells of young, healthy volunteers showed
a 30% reduction in insulin sensitivity
2
. Thats
the equivalent of metabolically ageing them
10 to 20 years, says Brady.
So reducing sleep can have a profound effect
on individual cells. The fat cells getting some
signal saying Im not sleeping much, and thats
going to alter its biology, says Turek.
Metabolically, sleep quality is just as impor-
tant as sleep duration. The onset of slow-wave
sleep coincides with hormonal changes that
OBESI TY
Heavy sleepers
A growing body of evidence shows that getting a good
nights sleep plays an important role in regulating the
bodys metabolism.
BY BRI AN OWE NS
E
M
I
L
I
A
N
O


P
O
N
Z
I
SLEEP OUTLOOK
2013 Macmillan Publishers Limited. All rights reserved
2 3 M A Y 2 0 1 3 | V O L 4 9 7 | N A T U R E | S 9
affect glucose regulation, such as the release of
growth hormone. When people are allowed to
sleep for 7 or 8 hours, but are prevented for sev-
eral consecutive nights from going into deep,
slow-wave sleep, they show the same insulin
signalling response as if they had been sleeping
for only 4 hours.
The link between sleep and metabolism
might be controlled by a part of the autonomic
nervous system called the sympathetic nerv-
ous system, says Tasali. Activity of the sympa-
thetic nervous system inhibits digestion, and
so suppresses the secretion of insulin from the
pancreas. Brady adds that because leptin levels
are proportional to insulin-stimulated glucose
uptake in fat cells, a reduction in insulin sen-
sitivity could lead to less leptin production,
which would in turn stimulate appetite and
potentially lead to weight gain.
WEVE GOT RHYTHM
The interaction between sleep and metabolism
is complicated by the poorly understood
relationship between sleep and the bodys
natural circadian rhythms. If you disrupt
circadian rhythmicity, youre going to have
effects on the sleepwake cycle, says Turek.
And if you disrupt the sleepwake cycle,
youre probably having effects on various
circadian rhythms as well.
The circadian rhythm is determined by a
collection of interrelated biochemical clocks
that influence when we sleep, when we eat, and
many other biological activities (see Stepping
out of time, page S10). Mice that lack a work-
ing copy of a protein called Clock develop high
blood sugar and cholesterol levels and low
insulin, eventually becoming obese
3
.
The Clock protein is active in a brain region
known as the suprachiasmatic nucleus (SCN),
which is linked to the lightdark cycle of day
and night. The SCN synchronizes all the bodys
24-hour clocks, either directly, for example by
stimulating the pineal gland to produce the
hormone melatonin, or indirectly, by influenc-
ing the time of feeding, which activates pancre-
atic function. But, says Turek, humans are the
only species that doesnt pay attention to their
biological clock we eat whenever we want.
To study the metabolic effect of eating at
different times, Turek altered the natural feed-
ing time of mice. He fed one group of mice a
high-fat diet only during the day, when these
nocturnal animals would not normally be eat-
ing, and fed another group the same diet only
at their biologically normal time, at night. The
results sound a cautionary note to those of
us who consistently fall out of sync with our
bodys natural eating schedule. Although both
groups of mice consumed the same amount
of calories and had
similar levels of activity,
after six weeks the ani-
mals fed during the day
had gained significantly
more weight and had
more body fat than those fed at night. Eating
at the wrong time of day leads to alterations
in something maybe basal metabolic rate,
maybe body temperature that changes how
the body processes the energy that it takes in,
he says. If youre eating at the wrong time of
day, youre more likely to gain weight.
So when humans override the SCNs signal
by eating late at night, they throw their feed-
ing cycle and various downstream rhythms
out of alignment with the central nervous sys-
tem, with possible knock-on consequences for
their metabolism. Its not only what you eat,
its when you eat, says Turek.
SLEEP IT OFF
Its clear then that bad or shortened sleep
causes metabolic problems. But would getting
more and better sleep reverse these effects
for example, by helping obese people to lose
weight? Research by Tasali and others suggests
that such therapeutic effects are possible.
Tasali has focused on improving sleep quality
in pre-diabetic patients who suffer from sleep
apnoea, a breathing
disorder that leads to
disturbed sleep. We
can give them back
their sleep quality by
treating their sleep
apnoea, and see what
happens to their glu-
cose metabolism,
she says. In a study presented at the American
Thoracic Society meeting in Philadelphia, she
found that a two-week regimen of the standard
treatment for sleep apnoea, called continuous
positive airway pressure, led to greatly reduced
insulin resistance
4
.
Similarly encouraging results have come
from weight-loss research. Obesity specialist
Giovanni Cizza at the National Institute of
Diabetes and Digestive and Kidney Diseases in
Bethesda, Maryland, has been testing whether
overweight people who are short sleepers can
lose weight if they are coached to sleep longer
5
.
The study ended in 2012 and Cizza is now ana-
lysing the data. He says he is pleased with the
results, despite a major hiccup in the trial.
The subjects were split into an intervention
group, in which subjects were coached about
good sleep hygiene and how to improve their
sleep habits, and a comparison group, given no
coaching. But simply making the comparison
group more aware of their sleep habits, by
asking them to fill out sleep diaries, helped
them extend their sleep duration by almost
as much as the intervention group. Basically,
Cizza says, my placebo group took the drug
too. The results have not yet been published,
so he would not say how much weight the
subjects lost only that because both groups
extended their sleep, there was no significant
difference between them in sleep duration or
weight loss.
Clearly, the relationship between sleep and
metabolism is complicated. A host of interact-
ing biological, behavioural and psychological
factors influence both sleep and metabolic
function, says Jim Horne, a sleep researcher
at Loughborough University, UK. People who
sleep less have more time to spend eating, for
example. The stress and tiredness caused by a
bad nights sleep can lead to comfort eating and
less activity during the day. In addition, obesity
is known to cause breathing problems, such
as sleep apnoea, that can interrupt sleep. And
both poor sleep and obesity are often associ-
ated with other disorders, such as depression
(see The dark night, page S14).
As a result, it is unclear whether short sleep
duration causes obesity, or vice versa, says
Horne. I challenge this idea that short sleep
makes you fat. I would say being fat causes
short sleep and bad sleep.
But it may not be necessary to untangle
cause and effect fully before testing the simple
proposition of whether better sleep can help
people lose weight. Because sleep is not a drug,
testing its effects should be a less daunting
process than that required for pharmaceutical
development, says Cizza. After all, he points
out, if youre not getting enough sleep, getting
more will have no adverse side effects, and will
have many added benefits in terms of mood
and neuropsychological function, for example.
Where there is reasonable evidence for effi-
cacy and there is no harm, I think we should
bring this to the public sooner than a drug.
Brady points out that people often find it
difficult to change their diet and take more
exercise. Telling them to sleep more, we think
that might be a little more palatable.
Horne, however, urges caution. In most
cases, he says, the effect of extra sleep on metab-
olism is vanishingly small. Fifteen minutes of
brisk walking every day is far better at regu-
lating your body weight than an extra hour of
sleep, so we have to put things in perspective.
He worries that people will be tempted to avoid
healthy exercise in favour of the less strenuous
option of popping sleeping pills, which do have
side effects. That could be dangerous, and no
one working on the complex relationship
between sleep and metabolism would recom-
mend trying such a shortcut.
But there is mounting evidence that getting
the right amount of healthy sleep can be just as
important as diet and exercise in controlling
your metabolism, and can do a world of
good beyond making you wake up happy in
the morning.
Brian Owens is a freelance writer based in
St Stephen, New Brunswick, Canada.
1. Siegel, K. et al. Ann. Intern. Med. 141, 846850
(2004).
2. Broussard, J. et al. Ann. Intern. Med. 157, 549557
(2012).
3. Turek, F. et al. Science 308, 10431045 (2005).
4. Pamidi, S. et al. American Thoracic Society Meeting,
20 May 2013.
5. Cizza, G. et al. Clin. Trials. 7, 274285 (2010).
NATURE.COM
For more on the link
between sleep and
metabolism:
go.nature.com/wrp4ku
If youre eating
at the wrong
time of day,
youre more
likely to gain
weight.
OUTLOOK SLEEP
2013 Macmillan Publishers Limited. All rights reserved