University of LeedsTel +44 113 343 3112 Leeds LS2 9JT Fax +44 113 343 3167
Obesity and the metabolic syndrome are increasingly serious medical problems throughout the developed world. This reflects in part previous selection pressure for "thrifty" genes that could protect individuals against frequent famines. There is, however, a markedly reduced life expectancy from all causes in overweight individuals, which affects their life assurance premiums. About 20% of North American adults are now clinically obese, and obesity is growing in children. Europe, North America and Japan are following similar paths.
know the definition and incidence of obesity
describe the main diseases associated with obesity
understand the importance of pro-inflammatory cytokines and low grade inflammation
appreciate the constancy of adult weights
understand the local controls within the GI tract
enteric nervous system
local hormones: CCK, GLP1, GIP, PPY, PYY
local gut systems are supervised by the hindbrain
food intake and energy output are modulated by the hypothalamus
know the principal input signals to the hypothalamus
know the main output signals from the hypothalamus
to the hindbrain
to the anterior pituitary
to the autonomic nervous system
know the main output signals from the pituitary
understand the control of fat mobilisation
understand adaptive thermogenesis
know about diet, drug and surgical therapy
There is an excellent open access review by Cummings & Overduin (2007) Gastrointestinal regulation of food intake. J Clin Invest 117(1), 13-23 which is well worth reading. Click the link to download the paper.
Back to the top
Obesity is often defined in terms of the body mass index (BMI):
BMI = weight (kg) / height (m)2
A BMI of 25-30 is overweight, above 30 is obese. This is useful as a guide, but BMI does not adequately distinguish fat from lean muscle mass. Separate norms should be used for men, women, children and for different races, but this is rarely done. Skinfold measurements with callipers are better, but require skilled observers and repeated tests. The definitive method is based on dual wavelength X-ray absorption, but this is too complex for routine use. Weighing people under water and measurments of electrical impedance have also been employed. Ordinary waist measurements may be the best guide to abdominal fat depostion, which seems to carry a greater risk of disease than peripheral, subcutaneous fat.
Back to the top
The most serious links are those involved in the metabolic syndrome between obesity, type 2 diabetes, dislipidaemia, hypertension and cardiovascular diseases (principally heart attacks and stroke) since these conditions are major causes of death in the general population. The pathogenic mechanisms in diabetes seem to involve the non-enzymic glycation of connective tissue proteins, leading to microangiopathy followed by kidney, retinal and neurological problems. Diabetics also have an adverse blood lipid profile that is associated with atherosclerosis and large vessel disease. The key to successful diabetic therapy is good glycaemic control.
Type 2 diabetes has recently been reviewed in Science 307, 366-387. [Issue of 21 January 2005] This collection of articles also discusses many of the other topics (e.g. mitochondria, cytokines, AMPK, UCP1) that we are covering in these lectures. It includes the first report of visfatin, a new insulin-like hormone secreted specifically by visceral fat depots. Leeds students can click the link in this box to read these articles on screen, or save and print out a PDF version.
Pathogenic mechanisms in hypertension are disputed, but may involve (1) reactive oxygen species (ROS) interfering with nitric oxide signaling, (2) damage to the vascular endothelium, increasing the risk of clotting and (3) increased myocardial oxygen demand because the heart has more work to do.
Atherosclerosis and cardiovascular disease are inflammatory processes, and many of the effective drugs (aspirin, statins) have anti-inflammatory effects.
Obesity, hypertension and diabetes are not the only risk factors for the development of cardiovascular disease. A complete list of cardiovascular risk factors would also include:
Age [over 45 years for men, 55 years for women].
Blood clotting disorders [fibrinogen / plasminogen activator].
Family history [affected male relatives under 55 years, females under 65 years].
Homocysteinaemia [new research has uncovered a surprisingly common condition].
Hypercholesterolaemia [high risk] and other dyslipidaemias.
Lack of physical exercise.
Male gender, post menopausal women; [oral contraceptives: low risk].
Poverty and low socio-economic status.
Tobacco smoking [very high risk].
Numerous other conditions are associated with obesity, including: gallstones; gastro-oesophageal reflux disease; degenerative joint diseases (knees, hips or spine); psychological problems; reproductive problems; snoring and sleep apnoea (not breathing); varicose veins. Associations do not necessarily imply causality, and some links are disputed, especially the asthma and cancer risks.
In February 2009 the UK Food Standards Agency embarked on a national advertising campaign to persuade people to reduce their dietary intake of saturated fat. This is backed up by press releases and in depth links although some of the detail seems a bit sparse.
In contrast to this, recent medical research does not support changing the composition of the diet among those trying to lose weight. Calorie reduction, by any route, seems to be the only dietary intervention that counts:
Katan (2009) Weight loss diets for the prevention and treatment of obesity. New England Journal of Medicine 360, 923-925.
Sachs et al (2009) Comparison of weight loss diets with different compositions of Fat, Protein and Carbohydrates. New England Journal of Medicine 360, 859-873.
Shai et al (2008) Weight loss with a Low-Carbohydrate, Mediterranean or Low-Fat Diet. New England Journal of Medicine 359, 229-241.
Malik & Hu (2007) Popular weight-loss diets: from evidence to practice. Nature Clinical Practice: Cardiovascular Medicine 4(1), 34-41.
Elwood et al (2007) Milk and dairy consumption, diabetes and the metabolic syndrome: the Caerphilly prospective study. Journal of Epidemiology and Community Health 61, 695-698.
Chan et al (2008) Effect of weight loss on markers of triglyceride-rich lipoprotein metabolism in the metabolic syndrome. European Journal of Clinical Investigation 38(10), 743-751.
Pro-inflammatory cytokines: (TNF-α IL-1 & IL-6) act on the hypothalamus to reduce appetite and raise body temperature in response to infection, and many other serious illnesses. They have anti-insulin effects. They were first identified as products of the immune system, especially macrophages, but it is now recognised that "obese" adipocytes also secrete these proteins, and cause low-grade inflammation. A major negative feedback loop stabilises the immune system, involving IL-1, hypothalamus, corticotropin releasing hormone, corticotropin (=ACTH) from the pituitary, and corticosteroids from the adrenal cortex.
Interleukin-6 [IL-6] is now seen to have a complex role. It elicits the acute phase response from liver following injury or infection. The liver secretes various acute phase proteins including C-reactive protein [CRP] and caeruloplasmin that put the body on a "war footing" before the threat has been properly identified. Exercising muscle produces IL-6 which helps to mobilise fat stores and has a generally beneficial effect.
Leptin, a cytokine released by fat cells acts on the arcuate nucleus within the hypothalamus to suppress eating behaviour and also increases energy expenditure. Leptin production rises with increasing fat cell mass. Mice that are homozygous for the ob mutation fail to make leptin and are grotesquely obese, as are db/db mice that lack the leptin receptor. Leptin has at least two signalling pathways: JAK/STAT via nuclear gene expression, and an effect on ATP-sensitive potassium channels in glucose-responsive neurons. This affects the neuronal firing rate. [The same K channel is involved in islet cells and cardiac muscle.] Leptin also has major effects on reproductive behaviour. Sexual maturation is delayed by lack of food. Starving women, female athletes and anorexics with low fat stores experience secondary amenorrhea. Leptin signalling defects lead to gross obesity, but these are very rare in humans. Other factors must be responsible for obesity in the general population.
Adiponectin is a mixture of anti-inflammatory peptide hormones also secreted by adipocytes that regulate energy homeostasis and the metabolism of glucose and lipids. Adiponectin stimulates the phosphorylation and activation of 5'-AMP-activated protein kinase (AMPK) in skeletal muscle and liver. This leads to phosphorylation of acetyl coenzyme A carboxylase (ACC), fatty-acid oxidation, glucose uptake and lactate production in myocytes, and phosphorylation of ACC and reduced gluconeogenesis in liver. Adiponectin increases insulin sensitivity in target tissues, but also stimulates fatty acid oxidation and blocks the differentiation of new adipocytes in bone marrow. Obese patients often have lower adiponectin levels than thin people.
Back to the top
Ignoring short term fluctuations (which are mostly gain and loss of water) our body weight stays remarkably constant. A person who gains 20kg over 10 years has ingested about 590MJ above their energy requirements, but this represents only about 1.3% excess in their total energy consumption of about 44GJ over the same period. Weight regulation is achieved through precise and effective negative feedback loops, making it very difficult to lose weight by dieting.
There are two components to this regulation:
Hunger varies inversely with body weight
Metabolism varies directly with body weight
In other words, if you lose weight the desire to eat more food becomes very powerful, while at the same time your basal metabolic rate declines to conserve the limited fat stores that remain. [Several of the following figures have been adapted from Schwartz et al (2000) Central nervous system control of food intake Nature 404(6778), 661 - 671.]
The gut contains approximately as many neurons as the spinal cord, and this complex nervous system can manage the entire digestive process, controlling enzyme secretion, absorption and peristalsis without assistance from the brain. Entero-endocrine cells (similar to taste buds) within the gut wall also secrete peptide and other hormones, which regulate the production of digestive juices, feed forward to stimulate insulin secretion (the incretin effect), and also indicate satiation to the central nervous system.
|hormone||produced by||acts on||controls|
growth hormone, signals hunger
duodenum, in response to fatty or high protein meals
gall bladder, liver, pancreas, CNS
gall bladder contraction, hepatic bile secretion, pancreatic enzyme secretion, minor effect on insulin release, satiation (terminates meal)
insulin release – incretin effect
hedonic aspects of feeding?
satiety (delays next meal)
The brain uses the autonomic nervous system to supervise the gut. There is a two-way flow of information and the brain modulates the volume and composition of the digestive juices, the rate of passage through the pylorus and the total residence time within the gut.
In all vertebrates information about smell, taste and gastric fullness is conveyed by the cranial nerves. Smell receptors transmit information via the olfactory nerve (I) which is really a specialised brain tract. The taste buds are mainly innervated by the facial nerve (VII) and the glossopharyngeal (IX) while the liver, stomach and duodenum are served by the vagus nerve (X). This system does not merely detect that the gut is full, but can also monitor what kinds of food have been eaten, their digestibility and their likely calorific yield.
Fundamental feeding assessments ranging from "unbearable hunger" to "stuffed to capacity" are made by the hindbrain, in the nucleus of the solitary tract or possibly in the parabrachial nucleus nearby. The corresponding emotional and physical sensations are communicated to the higher centres and influence our conscious behaviour.
A second tier of control systems in the hypothalamus modulate the basic meal feeding reflexes according to the body's fat stores and blood sugar levels. The hypothalamus is a small volume of nervous tissue surrounding the third ventricle, with important neural connections to the hindbrain, and also to the pituitary gland. Some of this tissue lies outside the blood - brain barrier, and is able to respond directly to circulating hormones, and sense the glucose concentration in the blood. It is uniquely placed to control both energy inputs and outputs.
To locate your own hypothalamus, put your fingers in your ears, then move upwards about 20mm and forwards about 20mm. The line joining your fingertips now passes through your hypothalamus, in the middle of your skull. The total volume of human hypothalamic tissue is only about 4ml but it contains at least a dozen distinguishable nuclei which collectively regulate blood pressure, osmolarity and thirst, temperature, diurnal rhythms, sexual activity and emotional responses in addition to food intake and metabolic activity. The "wiring diagram" for this tiny area is extremely complex, and in addition it contains major nerve tracts connecting the midbrain and hindbrain to the higher centres. The areas concerned with feeding are coloured green on the diagrams below. Point at the figure with your mouse for additional information.
The hypothalamic controls preserve the natural feeding cycle, but if food is available meals become progressively larger and more frequent when signals indicate that fuel reserves are running low. At first hungry animals search assiduously for food, but there is a second stage during longer term starvation when activity declines and metabolism closes down to preserve the remaining fuel stocks, in the hope that things might improve. Conversely, a generous food supply stimulates metabolism and provides an ideal opportunity to breed. There are important links between the fuel gauges and the reproductive system.
Back to the top
Many local peptide hormones conveying information in the gut are also detected in the brain, where they act as neurotransmitters. Although there is some similarity between the signals conveyed at these two locations, both systems have evolved independently for about 500 million years. There is now great diversity of function and we should not expect an exact match. Neural pathways have been traced by dissection, and more recently by following axonal transport of peroxidase and virus particles. Activation of nerve cells leads to transiently increased expression of the fos gene, which can be detected immunologically. This has greatly assisted the identification of the neural pathways involved in appetite regulation. Gene knockout on the other hand has been less useful. These systems have considerable redundancy, and knocking out individual genes often has only minor effects on the animal's behaviour. Some of these genes are imprinted, and disruption of the maternal and paternal copies produces opposite effects.
The gut lacks the kidney's ability to generate hyper- and hypo-osmolar solutions. Digestion and absorption are generally isotonic processes. Isotonic in this context means having almost the same osmotic pressure as blood. In addition to the fluids that we drink, considerable volumes of liquid are secreted by the upper gastrointestinal tract. Both the solutes and the water are resorbed in the lower gut. The osmotic pressure of nutrient solutions is a practical issue where patients must be fed intravenously.
Calculate the quantities of water and alkali that are required to dissolve the hydrolysis products of a typical meal as a neutral isotonic solution (300 milliosmoles/litre). Assume that the subject consumes 12MJ/day in three main meals, and that 10% of the energy is derived from protein, 60% from starch and 30% from fat. The energy yields are 17kJ/g (dry) from both proteins and polysaccharides and 37kJ/g from fat. The average molecular weight for an amino acid residue in a protein is 120, glucose is 180, a C16 fatty acid is 256 and glycerol is 92.
It should work out to rather a large volume. You can click here to check your answers.
You could not drink so much water with each meal, nor could you eat sufficient sodium salts to drive glucose and amino acid uptake, if each ion were used only once. The process works because the water and the ions are recycled many times. It is essential to conserve water by re-assembling the digested foodstuffs back into fat globules or glycogen as rapidly as possible. The insolubility or high molecular weight of these storage materials ensures a very low osmotic pressure. It is also desirable to keep the solute concentrations within the intestinal lumen as low as possible to discourage bacterial growth. Local hormones from the mid-gut stimulate the production of digestive juices by the liver and pancreas and restrain the stomach from releasing food too quickly. Sensory cells in the gut walls also warn the downstream processing systems in the pancreas, liver and adipocytes of the imminent arrival of osmotically active sugars and calorigenic fats. The insulin response to oral glucose with all these sensory cues intact is more accurate and better timed than the response to the same amount of intravenous glucose. The local gut hormones that signal this information are known as incretins.
Blood glucose monitors: There are at least five independent sensors that monitor blood glucose concentration and contribute to the overall regulation of food intake.
Pancreatic b cells
Portal vein glucose sensor
Nucleus of the solitary tract
These sensors differ in their cross-sensitivities to other metabolites, but they appear to have some features in common, including low-affinity GLUT2 transporters mediating glucose entry to the cells and the low-affinity glucokinase enzyme. This makes their intracellular ATP supply exquisitely sensistive to glucose availability, and forms the basis for their signalling systems. Low blood glucose (hypoglycaemia) triggers a massive response from the sympathetic nervous system, and is associated with a pounding heart, trembling, anxiety, sweating and hunger. The effects of hyperglycaemia on feeding and satiation are much less obvious, and it may even lead to a paradoxical increase in food intake.
There is an excellent brief review by Korner & Leibel (2003) To Eat or Not to Eat — How the Gut Talks to the Brain NEJM 349, 926-928 which includes a very useful diagram. You are strongly recommended to read this article, and the associated report on Peptide YY in the same issue of this journal. They are both available electronically, using the normal Leeds library password.
Cholecystokinin (CCK) is the best known member of a group of hormones secreted by the duodenum in response to the partially digested output from the stomach, which is called chyme. Fatty meals are particularly effective. CCK delays gastric emptying, stimulates pancreatic enzyme production, causes the gall bladder to contract, promotes insulin release by the pancreas and produces a sensation of fullness or satiation. Other incretins in the same group are gastric inhibitory peptide (GIP), glucagon-like peptide 1 and glucagon-like peptide 2. All have been identified as drug development targets.
CCK signals to the brain via the vagus nerve. In addition to the chemical information about the meal contents, the body also measures gastric distension, and the metabolic situation in the liver. Although the details remain to be elucidated, it is clear that this system can judge what you have eaten, and arrive at a fairly accurate estimate of its likely energy yield. It matches the output of the foregut to the digestive capacity of the midgut, controls the duodenal pH and osmolarity, and warns the downstream processors of exactly what to expect. It is not so easily fooled, and it can tell a low-calorie slimming product from a normal meal.
Pro-inflammatory cytokines: (TNFa IL-6 & IL-1) act on the hypothalamus to reduce appetite and raise body temperature in response to infection, and many other serious illnesses. They were first identified as products of the immune system, especially macrophages, but it is now realised that many other tissues (including adipocytes) can secrete these compounds. There is a major negative feedback loop involving the hypothalamus, corticotropin releasing hormone, corticotropin (=ACTH) secreted by the pituitary, and corticosteroids from the adrenal cortex which damp down pro-inflammatory cytokine production. This loop normally acts to stabilise immune system activity, but it also has spill-over effects on appetite and weight regulation.
Ghrelin is a 28-residue peptide secreted by endocrine cells within the gastric sub-mucosa. It acts on the hypothalamus to stimulate growth hormone release by the pituitary. Ghrelin is also produced locally by neurons within the hypothalamus, and in other parts of the intestine. It antagonises leptin, increases metabolic efficiency and stimulates eating behaviour, resulting in weight gain. "When your stomach’s growling its making ghrelin." The control of gastric ghrelin production is complex. The hormone normally indicates hunger and is released by the empty stomach, but paradoxically it may also be released in response to high-protein meals. It is thought that the effective weight loss achieved by gastric surgery may reflect a fall in ghrelin output.
Insulin: In addition to its role in blood glucose homeostasis, insulin reduces food intake and plays a major part in appetite regulation. Gene knockout experiments have confirmed that lack of either brain insulin receptors or insulin receptor substrate 2 (IRS2) results in hyperphagia, obesity and female infertility. It is believed that insulin promotes phosphorylation of leptin receptors and JAK2, which enhances the phosphorylation of STAT3 in the presence of leptin. Insulin levels are often raised in type 2 diabetes which is associated with insulin resistance and obesity.
Amylin: Pancreatic b cells co-release a second polypeptide hormone called amylin at the same time as they release insulin. Amylin produces a feeling of satiation, and may assist in the regulation of food intake.
Apolipoprotein A-IV is a glycoprotein synthesised by enterocytes in response to long chain dietary fat. Short chain fats are ineffective. It is a component of chylomicrons, which are microscopic globules of fat and proteins secreted by the cells lining the small intestine into the lymph draining from the gut. Apolipoprotein A-IV may be regulated by PYY (see below). It is thought to regulate food intake, possibly by stimulating CCK production. It may be effective in its own right because it is also present in the brain.
Peptide YY: (PYY) is a 36-residue peptide secreted by endocrine cells within the lower small intestine, pancreas and colon. It slows down digestion and reduces the consumption of food. PYY inhibits gastric acid secretion, gastric emptying, pancreatic enzyme secretion and gut motility. It acts on the arcuate nucleus in the hypothalamus to suppress appetite and reduce food intake. There has recently been great interest in PYY after it was found to reduce food intake by 33% in obese subjects, who are normally leptin resistant. PYY shows sequence homology to orexigenic NPY described below and to pancreatic polypeptide (PPY) from F cells in the pancreatic islets. PPY inhibits the secretion of pancreatic enzymes and bicarbonate.
Neuropeptide Y (NPY) is a 36-residue, highly conserved peptide that is widely distributed throughout the vertebrate nervous system. NPY has multiple neurosecretory and cardiovascular functions. There are at least five classes of NPY receptor distributed over a wide range of tissues. NPY delivers a powerful orexigenic (appetite-promoting) signal within the hypothalamus, which activates a neural pathway leading to the nucleus of the solitary tract. NPY is over-produced in leptin-deficient mice, but double knockout mice show less severe symptoms, suggesting that NPY mediates the overfeeding observed in leptin-deficient animals.
Pro-opiomelanocortin (POMC) is a protein that is cleaved to yield a variety of important signals. It is the precursor of the pitutitary hormones MSH and ACTH, and in addition transmits a variety of messages within the brain.
At least four distinct 7-transmembrane G-protein linked receptors recognise the core heptapeptide sequence of the melanocortins. MC1R controls skin pigmentation, MC2R is the ACTH receptor, while MC3R and MC4R control appetite and energy expenditure. Gene knockout experiments show that both MC3R and MC4R transduce anorexic (appetite reducing) signals, but double knockouts are significantly heavier than single knockouts, showing that MC3R and MC4R have distinct roles. MC3R appears to control the conversion of dietary fuel into fat whereas MC4R regulates food intake and possibly energy expenditure.
Agouti-related protein (AGRP) is named after a related gene controlling hair colour in rodents. It is a powerful antagonist of the MC3R and MC4R melanocortin receptors in the hypothalamus. Obese patients have elevated plasma levels of AGRP, and over-expression of AGRP in animal models leads to obesity.
Cocaine and amphetamine regulated transcript (CART) is a neuropeptide precursor protein that is abundant in the hypothalamus. It is upregulated after cocaine or amphetamine administration. CART-derived peptides reduce food intake when injected into the third cerebral ventricle, and probably have several other functions within the central nervous system.
Pituitary outputs: In addition to the neural outputs modulating feeding reflexes in the hindbrain, the hypothalamus controls energy expenditure via the anterior pituitary and the sympathetic nervous system. The principal output signal to the pituitary is thyrotropin releasing hormone (TRH) a small peptide secreted into the hypophyseal portal system which causes the anterior pituitary to release thyroid stimulating hormone (TSH). Thyroid hormone acts on peripheral tissues, increasing alertness, heat production and basal metabolic rate. Conversely, the temperature regulating system interacts with energy homeostasis, and low ambient temperatures induce a powerful urge to eat. In addition, there is interaction between energy homeostasis and the reproductive systems. Sexual maturation is delayed, and sexual activity generally diminishes when food supplies are inadequate.
Thyroid hormone production and a "normal" level of alertness and physical activity require adequate food intake. Thyroxin output falls during prolonged starvation when metabolism and physical activity are severely restricted to conserve fuel stocks. The strategy is a sensible one: if you don't find food fairly quickly, the best option is to slow down and hope for the best, instead of squandering precious resources on a fruitless search. Two other pituitary hormones are associated with fasting and gluconeogenesis: corticosteroids (which participate in a feedback loop involving pro-inflammatory cytokines) and growth hormone, which defends the body's protein and glycogen reserves, and promotes the breakdown of fat.
Autonomic outputs: Hypoglycaemia and hypothermia both lead to sustained sympathetic responses. Subjects feel hungry and eat if possible, but they refine their other actions to suit the circumstances. Hypoglycaemia requires hepatic glycogenolysis and gluconeogenesis, while hypothermia requires increased heat production and a redistribution of blood flow. Sympathetic activity is controlled by the hypothalamus, which instructs the adrenal medulla to secrete adrenalin. This is a rather blunt instrument, and localised sympathetic responses (such as blood flow regulation) are mediated by individual nerves. Parasympathetic activity can also respond to the hypothalamus, which controls the nucleus of the solitary tract.
Back to the top
Lipoprotein lipase (= clearing factor lipase) is the enzyme responsible for the breakdown of circulating chylomicrons and VLDL in the fed state, permitting lipid uptake into cells and triglyceride deposition within the tissues. In healthy subjects, lipoprotein lipase activity is increased by insulin, but this regulation is lost in type 2 diabetes.
Mobilising lipase (= hormone-sensitive lipase) is the enzyme responsible for the breakdown of stored triglyceride during fasting in response to circulating adrenaline. It is controlled by cAMP / PKA. It is active mainly within white adipose tissue (where it releases free fatty acids into the bloodstream) but is also found in other tissues (e.g. type 1 muscle fibres, brown adipose tissue) with significant lipid stores. There is evidence that lipid mobilisation in response to adrenaline is decreased in some obese subjects.
Three enzymes are generating huge research interest at present, because they might allow people to lose weight while eating a normal diet. They are AMPK, which we mentioned in a previous lecture, acetyl CoA carboxylase and fatty acid synthase.
Malonyl CoA is an intermediate in lipid biosynthesis, but it also blocks the carnitine-dependent entry of acyl CoA into mitochondria, and thereby inhibits lipid oxidation, as we discussed in lecture 3. There are two isoforms of acetyl CoA carboxylase, ACC1 in adipocytes and ACC2 in muscle. ACC2 is inactivated by AMPK (as is cholesterol biosynthesis via HMG-CoA reductase) and gene knockout mice lacking ACC2 lose weight while eating and breeding normally. Malonyl CoA is also the substrate for fatty acid synthase. Shimokawa et al (2002) PNAS 99(1), 66-71 have reported that inhibition of fatty acid synthase acts via the hypothalamus to reduce food intake. It provokes rapid weight loss in normal and obese mice, through a mechanism upstream of NPY and POMC. All these lipogenic enzymes are, to say the least, potential targets for some extremely profitable drugs...
Back to the top
Animals have several strategies for keeping warm. They can seek out warm places, fluff up their fur or feathers, reduce skin blood flow, fidget, shiver or indulge in non-shivering thermogenesis by increasing their resting metabolic rate. The first three methods are counterproductive if the main objective is to burn off surplus fuel without raising body temperature. Such behaviour might confer a selective advantage if animals were obliged to eat an unbalanced diet, where essential nutrients are diluted in a sea of calories.
In many species, and in new-born babies, brown adipose tissue is an important site of heat production. This tissue is coloured brown because it is rich in mitochondria, which can be temporarily uncoupled under the control of the sympathetic nervous system. This superficially futile dissipation of mitochondrial chemiosmotic gradients produces heat rather than ATP. Heat production has an absolute requirement for thyroid hormone. There is not much brown adipose tissue in adult humans, and attention has therefore focused on muscle and white adipose tissue, which seem to be more closely involved in thermogenesis in adults.
Five uncoupling proteins (UCP1, UCP2, UCP3, BMCP1 and UCP4) have been characterised in mammals. The first three have apparently evolved from the adenine nucleotide carrier. UCP1 or thermogenin is found in brown adipose tissue and requires oxidised CoQ as a cofactor. Expression of the UCP1 gene is stimulated by catecholamines. Catecholamines also activate the UCP1 protein, through a mechanism involving cyclic AMP, hormone sensitive lipase, and free fatty acids.
UCP2 is widely expressed in many tissues but mRNA levels are unexpectedly increased by fasting. Gene expression seems to have little connection with thermogenesis, but it is apparently associated with a shift away from carbohydrates towards fat oxidation. The supply of reducing agents the respiratory chain seems to be subtly different when fats are the major substrate, and this seems to increase the production of toxic compounds such as superoxide and hydroperoxide from the incomplete reduction of oxygen to water. UCP2 production may form part of a mechanism that minimises the unwanted generation of these reactive oxygen species (ROS). UCP2 reduces ATP levels in pancreatic islet cells, and this may mediate some anti-insulin actions of free fatty acids.
There is a complex interaction bewteen insulin and fats. In the short term, free fatty acid metabolism increases the intracellular ATP concentration in islet tissue and raises insulin output. This may help us to cope with the immediate effects of fatty meals. However, in the longer term insulin secretion is depressed by fatty acids, possibly because of the UCP2 induction decribed above. This may be important during starvation. Fat oxidation competes with glucose oxidation in most peripheral tissues, and fats therefore have an anti-insulin effect on the rest of the body. Whatever the mechanism, obese people tend to have high circulating levels of glucose, insulin and fats, and this may progress to overt type 2 diabetes. The average circulating insulin concentration is a good proxy for the total body fat.
UCP3 is found mainly in skeletal muscle and mRNA levels are increased by thyroid hormone and fasting. Transgenic mice that over-express human UCP3 are thin and hyperphagic, have lower fasting plasma glucose and insulin levels and an increased glucose clearance rate. However, gene knockout mice are apparently normal, and the function of this protein remains a mystery.
Back to the top
Randomly choosing a few low-fat or supposedly "healthy" foods does not work. Feedback systems easily compensate for this behavior and most people who cut back at one meal simply eat more at subsequent meals. There are two alternatives which do work:
A) Conventional wisdom: calorie-controlled low-fat, high-starch, physically bulky diets, rich in fresh fruit, green vegetables and unrefined carbohydrates. Total energy intake should be slightly reduced. It is essential to count the calories and restrict intake, because many subjects feel hungry all the time. It helps to join a support group.
B) "Atkins" diets: very low carbohydrate, unrestricted protein and fats. There is no need to count calories because most subjects do not feel hungry on these diets and there is a natural limit to how much one can eat.
Most clinical trials show Atkins diets work better than the conventional wisdom. Fears about heart disease have not been substantiated, but there may be a high drop-out rate.
Weight loss diets should include an exercise program to increase energy demand, leading to a gradual weight loss over several months. Crash diets should be avoided because they chiefly affect body water content, rather than fat. They may also promote muscle catabolism for gluconeogenesis. The altered lifestyle must be maintained indefinitely: if the person ceases to control their food intake then most of the fat will be rapidly replaced.
There is no entirely satisfactory drug treatment for obesity, and it is normally recommended that moderately overweight patients (BMI < 30) are treated with diet and exercise alone. Several compounds proved to have serious side-effects, and manufacturers are competing to bring an effective product to this lucrative market place. Development is proving difficult because multiple signalling pathways are involved in the regulation of body weight. Some existing drugs are tabulated below:
blocks pancreatic lipase
blocks 5HT uptake / sympathomimetic
dry mouth, headache
glucagon-like peptide 1
delays gastric emptying, more insulin produced
natural product, so probably fairly safe
use in type 2 diabetes
cannabinoid receptor antagonist
effective drug but causes depression
withdrawn: suicide risk
multiple CNS effects, anti-epileptic drug
psychomotor problems unpleasant for patient
poor patient compliance
high abuse potential
limited duration of action
blocks 5HT uptake
heart valve defects
blocks cholecystokinin breakdown
not properly tested, not yet in clinical use
None of these drugs is particularly effective, and a 10% weight loss is considered very good. Patients tend to re-accumulate the lost weight when the therapy is stopped. Nevertheless, even a modest weight loss can lead to a marked improvement in cardiovascular risk. There are two recent excellent reviews of anti-obesity drugs and therapeutic targets, by Crowley et al (2002) Nature Review Drug Discovery 1(4), 276-286 and Yanovksy & Yanovsky (2002) NEJM 346(8), 591-602.
Liposuction to remove excess fat is not currently recommended, but gastric surgery is proving surprisingly effective. "Stomach stapling" operations include Roux-en-Y gastric bypass, and a variety of banding procedures. They probably reduce ghrelin output. It is also possible to insert a water-filled intragastric balloon, and there are proprietary dietary equivalents using natural hydrophilic gels. Surgery normally produces much larger weight losses than drug therapy, but carries a significant risk. Excellent results have recently been reported using implanted gastric pacemakers. This minor operation has a negligible risk.
HTML by John Illingworth