Burning to keep warm: the biology of brown fat

Part 1: The fat that makes heat

1Mammals are endotherms. They keep their core body temperature within a narrow range, largely by using heat released from their own metabolism. When the surroundings are cold, extra heat can be produced in two ways. Shivering uses rapid, involuntary contractions of skeletal muscle. Non-shivering thermogenesis takes place mainly in a specialised tissue called brown adipose tissue, or brown fat. Newborn babies, small rodents and hibernating mammals rely heavily on brown fat, and for most of the twentieth century it was assumed that adult humans had almost none. Brown fat in a newborn baby is found mainly between the shoulder blades, around the kidneys and along the spine, close to major blood vessels.

2Under a microscope, brown adipocytes look very different from the white adipocytes that make up most body fat. A white adipocyte holds a single large droplet of triglyceride that squeezes the nucleus and cytoplasm into a thin rim. A brown adipocyte contains many small lipid droplets and an unusually large number of mitochondria. The iron-containing cytochromes in these mitochondria give the tissue its brown colour. Brown fat also has a dense network of capillaries and is supplied by many sympathetic neurones.

3Brown and white adipocytes do not share the same origin. Studies tracing cell lineages in mice show that classic brown adipocytes develop from the same group of stem cells in the embryo that gives rise to skeletal muscle. As these cells differentiate, a small number of master regulatory proteins switch on the genes needed for a heat-producing cell and switch off genes needed for muscle. White adipocytes come from a different group of precursor cells.

A tissue rediscovered

4Interest in adult brown fat revived by accident. Doctors examining whole-body scans of patients being checked for cancer noticed symmetrical patches of high activity in the neck and above the collarbones. These scans combined positron emission tomography (PET) with computed tomography (CT). Before the scan, each patient was injected with a glucose analogue labelled with fluorine-18, a radioactive isotope that emits positrons. The patches were far more common in scans carried out in winter and in thinner patients.

5The glucose analogue is taken into cells by the same carrier proteins that transport glucose. Once inside, it is phosphorylated but cannot be broken down any further, so it becomes trapped. Cells that take up glucose fastest therefore build up the most radioactivity, and the scanner detects where this is happening. Tumours show up because they take up glucose rapidly. So did the patches in the neck, which biopsies later confirmed were brown fat.

Switching on the heater

6When skin temperature falls, thermoreceptors in the skin send impulses to the hypothalamus. The hypothalamus responds by increasing impulses in sympathetic neurones that end in brown fat. These neurones release the neurotransmitter noradrenaline, which binds to β3-adrenergic receptors in the cell surface membrane of brown adipocytes. Binding activates an enzyme that produces cyclic AMP inside the cell. Cyclic AMP acts as a second messenger and activates lipase enzymes that hydrolyse the stored triglycerides.

7The fatty acids released are broken down inside the mitochondria to acetyl coenzyme A, which enters the Krebs cycle. Reduced NAD and reduced FAD then deliver electrons to the electron transport chain in the inner mitochondrial membrane. As electrons pass along the chain, energy is used to pump protons from the matrix into the intermembrane space. In most cells, the protons flow back through ATP synthase and the energy is used to make ATP.

8Brown adipocytes are different because their inner mitochondrial membrane contains large amounts of uncoupling protein 1, known as UCP1 or thermogenin. UCP1 lets protons leak back into the matrix without passing through ATP synthase. Respiration is said to be uncoupled from ATP production. Oxygen is still used and substrates are still oxidised, often very quickly, but much of the energy is released as heat rather than being transferred to ATP. The blood flowing through the capillaries carries this heat around the body.

9In an illustrative study, twelve healthy adult volunteers each spent 24 hours in a sealed room on two separate occasions, once at 27 °C and once at 19 °C. Their energy expenditure was calculated from their oxygen consumption and carbon dioxide production. The mean energy expenditure was 7 410 kJ per 24 hours at 27 °C and 7 690 kJ per 24 hours at 19 °C. A PET-CT scan at the end of each session detected active brown fat in 3 of the 12 volunteers after the warm session and in 10 of the 12 after the cool session.

10"People are surprised that mild cold, not freezing cold, is enough," says Dr Mira Lindqvist, a physiologist who led the illustrative study. "Most of our volunteers did not shiver at all at 19 °C. The extra energy they used came from tissues that were quietly producing heat. When we repeated the scans after ten days of daily cool exposure, the active areas were larger and brighter." "Brown fat seems to respond to regular cold in much the same way that muscle responds to training."

11The amount of UCP1 in brown adipocytes is not fixed. Over several days of cold, transcription of the UCP1 gene increases. Part of this response depends on transcription factors activated by the noradrenaline signal. Thyroid hormone is also needed. It is a lipid-soluble hormone that crosses the cell surface membrane and binds to a receptor inside the nucleus, and the hormone and receptor together act as a transcription factor that binds to DNA near the UCP1 gene.

12Long periods of cold can also cause beige adipocytes to appear within deposits of white fat. Beige cells contain UCP1 and many mitochondria, but they develop from precursor cells within the white fat rather than from the muscle-like lineage of classic brown fat. This process, called browning, can partly reverse when the cold exposure stops, showing that the cells can change their pattern of gene expression in both directions.

13Gene technology has been central to this research. Mice in which the UCP1 gene has been knocked out cannot keep their body temperature steady if they are suddenly moved into the cold, although they survive if they are cooled gradually. Researchers have used DNA microarrays and, more recently, sequencing of all the mRNA in a tissue to compare brown, beige and white adipocytes. Thousands of genes differ in expression, even though the three cell types contain the same genome.

Part 2: Can brown fat help with obesity?

14Because active brown fat removes large amounts of fatty acids and glucose from the blood, there is interest in whether it could help treat obesity and type 2 diabetes. In type 2 diabetes, body cells respond poorly to insulin and blood glucose concentration stays high after meals. A tissue that takes up glucose without needing much insulin could, in principle, help to bring blood glucose concentration down. Brown fat may also release signalling molecules that affect the liver and muscles, although this idea is still being tested in human studies.

15One approach uses drugs. A drug that activates β3-adrenergic receptors, already approved to treat an overactive bladder, has been tested in small trials. In an illustrative trial of 14 volunteers, a single dose increased resting energy expenditure by a mean of 6 per cent and doubled the uptake of the glucose analogue by brown fat. However, some volunteers also had a faster heart rate and higher blood pressure, because related adrenergic receptors are found in the heart and blood vessels.

16A second approach uses regular mild cold. In another illustrative study, volunteers who spent two hours a day at 17 °C for six weeks increased their brown fat activity and lost a small amount of body fat. "It is a very gentle intervention," says Dr Tomás Reyes, who ran that study, "but it only works if people keep doing it. When the cold sessions stopped, the effect faded within a few weeks."

17Many scientists remain cautious. An adult has at most a few hundred grams of brown fat, so its contribution to total daily energy expenditure is modest. People who are obese tend to have less active brown fat than lean people. Much of the research has been done in mice, which have a much larger surface area to volume ratio than humans and lose heat much faster, so mice depend on brown fat far more than people do.

18Some populations living in very cold climates have variants of genes involved in heat production that are rare elsewhere. One explanation is that, over many generations, individuals whose brown fat produced more heat were more likely to survive cold winters. Over time, the frequency of these alleles in the population would have increased. Whether such variants could be harmful in a warm climate with plentiful food is still being investigated, and studies comparing populations must also allow for differences in diet and lifestyle.

19Hibernating mammals show how powerful brown fat can be. During hibernation, the body temperature of a ground squirrel may fall to just a few degrees above freezing and its heart rate falls to a few beats per minute. Every few days the animal rewarms itself to about 37 °C in under three hours. Most of this heat comes from brown fat between the shoulder blades, which warms the blood flowing to the heart and brain first, before shivering begins in the skeletal muscles.

20Even if brown fat never becomes a cure for obesity, it has changed how biologists think about energy balance. "We used to treat fat as a passive store," says Dr Lindqvist. "Now we know that some fat cells are among the most metabolically active cells in the body, and that the brain can turn them up or down." Researchers are now asking why some people have far more active brown fat than others, and whether age, body mass, diet or the temperature of the homes they live in explains the difference.