The brain and obesity | 91TV
Transcript
- Thank you, Anne, for that kind and terrific introduction. It's a great honour and privilege
- for us both to be here. It's unusual for the Croonian to be delivered as a duet. Looking at
- Sadaf and I as both endocrinologists and studiers of nutrition, we were intrigued and indeed daunted
- to see of the three previous duets, one was given by Gowland Hopkins, the father of nutrition,
- and another by Bayliss and Starling, who essentially both named and founded the
- science of endocrinology. So it's a hard act to follow, but thank you all for coming this evening.
- As a physician scientist, we've viewed this problem through a biomedical perspective,
- and the principal disorder which is impacted by the processes we're going to talk about today,
- is obesity, which has become a health challenge, as you can see on a global scale.
- It carries with it a huge burden of adverse metabolic, cardiovascular and oncological
- consequence. It's become increasingly common in our country, and indeed throughout the
- developed and developing world. People with severe obesity are most severely afflicted,
- and not only suffer from medical disbenefits, but all of the social, educational, economic,
- and indeed stigma, that is frequently associated with carrying large excesses of body fat.
- Obesity, as I say, has become more common, and I can show you one of
- a thousand slides to demonstrate that, but I particularly like this one. It's a study where
- the authors have found a dataset from Northeastern America of US war veterans
- in 1890, and found in an identical demographic 100 years later, and compared the prevalence of
- BMI as indicated by a body mass index of greater than 30. You can see that over 100 years that the
- prevalence of obesity went from 3 per cent to 3 per cent over a 100-year period. A dramatic
- increase, and a particular increase in the severe obesity end of the spectrum.
- So it's clear that our genetics hasn't changed over that period, and that this must be driven
- by the environmental changes, both in terms of food intake and the
- lack of a need for physical activity in our daily lives that have occurred over the past 100 years,
- and indeed has accelerated over the past 20 or 25 years. However, that has led some people to
- rather simplistic conclusions, and I think it is important to point out that obesity is not a new
- disease. It's inconceivable that the Palaeolithic sculpture of this Venus of Willendorf could
- possibly have created this image without having had a visual input of somebody who was obese,
- and that long ago. Hippocrates was much exercised by how he could help his obese
- patients. He told them they should eat only once a day, take no baths and sleep on a hard
- bed and walk naked as long as possible. Some cynics will say we haven't advanced very far
- in our treatment of obesity since those days. Unless we think we're the first age of doctors to
- be worried about obesity, Thomas Short wrote his discourse on corpulency in 1728, where he opined
- no age had seen more instances of corpulency than our own. So obesity has always been with us.
- I mentioned the obesogenic environment, the food environment, the activity environment,
- but if we are living in an obesogenic environment, why aren't we all obese? Why are some lean,
- why are some thin, and some obese? Is it entirely an issue of voluntary choice,
- or are lean people naturally morally superior? Of course, I laugh, and I wouldn't think that.
- Of course, what we're supported in our thoughts, are that the evidence for heritability of obesity
- is extraordinarily high. Not quite as high as that, for example, for height, but far
- higher than for traits such as IQ and others. It's a really high heritability, and I won't bore
- you with all the details of the heritable estimates, but really the most compelling
- work comes from identical twins, separated at birth and brought up in different environments,
- where the correlation between the identical twin that they've never seen is far higher than any
- family member with whom they've been brought up. That heritability has remained very high,
- even in modern studies. Jane Wardle, a beautiful study in the early 2000s - Jane, who sadly is no
- longer with us - did a fantastic twin study in the UK, and showed that that degree of heritability
- was still the same, pretty much, in contemporary Britain. So it's still a very heritable trait.
- So if it's heritable, there must be genetic factors influencing it,
- and of course, genetic factors don't influence metaphysical properties,
- but they influence physical issues. What do they what do they influence?
- Well, they must do a pretty good job, because we have an extraordinary amount of caloric intake
- and expenditure going through our little 70kg, on average, frame throughout our lives, the amount
- of energy intake and food output, what we do, and yet we do not oscillate between a supermodel and a
- sumo wrestler throughout our lives. Most people follow trajectories, reasonable trajectories,
- similar to what they've been in their early life. There is, of course, some drift.
- So this leads us to consider that there must be a physiological process
- which regulates body weight, and actually, the first experiments that really proved that,
- were very simple ones. Simple ones done by actually Gordon Kennedy, who was at Mill Hill
- at one point, and others around that time, who did very simple experiments on rats.
- They took healthy rats, and the rats continued to grow throughout their lives, and were just healthy
- rats, and then gavage-fed them for a couple of weeks, and when you start the gavage feeding,
- they put on weight. But then if you put them back in their home cage with their voluntary food,
- and the same amount of food, they stop eating and they return to exactly the trajectory they were on
- before. Similarly, when you underfeed them, you get a lowering of body weight, and then you put
- them back in their home cage with the voluntary food, and they start to eat and return to exactly
- the same trajectory. So that was the first body of experimental evidence that says, actually,
- body weight is a regulated phenomenon. There must be a physiological regulator of our energy stores.
- So if there's a physiological regulator, where is it sited? What we know is that the key
- site for the prime regulator is in the brain, and in particular it's in the hypothalamus.
- We know that, first, from the clinical observations of Frohlich and others, which
- observed that people who developed hypothalamic tumours early in life,
- often rapidly develop severe obesity and when they died were found to have
- destructive lesions within the hypothalamus. Then, similarly, just using rodent models, scientists
- in the in the US in the 40s and 50s showed that lesioning the hypothalamus could either result
- in a very severely obese animal, or indeed, if you lesioned the right part of the hypothalamus,
- you could lead to a very thin animal, who didn't eat and lost a tremendous amount of weight.
- I think the key experiment kicking off the modern era in obesity was done only a couple of miles
- from where I currently work, in the old Addenbrooke's Hospital, in a small MRC unit,
- by a chap called Harvey, who was actually the PhD student of David
- [?McCants 0:07:38.2]. McCants wasn't on the paper - Harvey just wrote a single-author
- paper - and to me this is the seminal experiment of the modern era.
- What Harvey did, was he did a parabiosis experiment. He basically sewed the neonatal skins
- of two rats together, so that small molecules could transfer from one circulation to the other,
- but there wasn't enough circulation transfer to do bulk transfer of nutrients.
- He then performed the lesion experiment in the brain, making one animal hyperphagic and eating
- a large amount of food and becoming obese, and then he watched what happened to the other animal.
- Essentially, the other animal lost interest in eating, stopped eating, and became extremely thin.
- They concluded that there must be a circulating factor produced in the obese animal, which
- transferred across to the other animal, and the parabiotic partner was responsive to it.
- But that science kind of stalled, because either the biochemical or genetics
- ability to find such a molecule, the time was not ripe for that. It took about a decade or
- more later at Bar Harbour, Maine, with the Jackson Labs, for Doug Coleman to study
- genetic forms of obesity in mice, and apply the same parabiotic paradigms, and show essentially,
- that the ob/ob mouse lacked a circulating factor, to which the db/db mouse was highly resistant.
- It then took a further period of time with the power of molecular genetics,
- at a time when it was far from trivial, and the murine genome was only partly mapped,
- for Jeff Friedman at the Rockefeller Institute to positionally clone the causative gene in the
- ob/ob mouse. What Jeff discovered was that it was a genetic defect in a circulating protein,
- made in adipose tissue and made in adipose tissue only, and going to the brain acting on
- brain receptors, and it was the mutation in those receptors that was responsible for the obesity of
- the db/db animal. So we had at last had, if you like, a real paradigm shift, understanding that
- the brain has a way of sensing nutritional state of the body, and altering appetite
- and energy expenditure in response to that state. The subsequent 20-or-so-years has really showed a
- fantastic refinement of that concept. Now, as well as leptin coming from coming from adipose tissue,
- we have a range of hormones that we know come from other tissues, including the
- pancreatic islets, the large and small intestine, and even the stomach, that provide a range of
- signals to the brain of both long- and short-term nutritional status, and not just caloric content,
- but actually the macronutrient distribution within the food that's just been eaten.
- So there's a symphony orchestra of signals coming to the brain of both long-term and
- short-term signals about what we're eating and what our status of our nutrient stores are.
- So we entered this field through a Homo sapiens as our experimental organism, if you like, of
- principal interest. We were physician scientists, and I was a physician scientist returning
- from the United States to Cambridge in 1991, inspired by my mentor Jeff Flier, and also by
- people like Brown and Goldstein, who had discovered phenomenally powerful information
- about the regulation of cholesterol through studying rare humans with extreme disorders.
- So the focus of my small lab when I came back ,was to study rare extreme disorders of metabolism,
- and having done so, first of all, in relation to how insulin worked, and then moving into obesity,
- in a kind of annus mirabilis of 1997, we found in quick succession the first two genes, the
- disruption of which caused severe human obesity. Rudy LaBelle, a leading scientist in the field,
- described that as a kind of grail of obesity research. So how did we get there? Well,
- our first gene we found was actually a genetic defect in a patient that I'd been seeing in my own
- regular NHS clinic for several years. We'd figured out that she was likely to have a prohormone
- processing disorder, and a wonderful NHS clinical biochemistry registrar, Rob Jackson, spent years
- cloning and characterising the PCSK1 gene before we finally found the causative defect in that
- gene. Shortly after Sadaf joined the lab, Sheila Muhammad, a geneticist in Oxford, had sent us
- a patient with severe obesity, and we found that that patient and their cousin had
- a defect in the leptin gene itself, resulting in unsecreted leptin and severe hypoleptinaemia.
- So at that time, I think it's important to understand that there was a lot of,
- not cynicism I'd say, but certainly a lot of conservatism thinking that biology was really very
- unimportant for adiposity, and most people's obesity was rather
- the consequence of poor personal choices, and I think it was an important step
- to be actually able to say that severe obesity can result from mutations in single genes.
- So we then had a very exciting couple of years, and Sadaf and I
- had great excitement trying to find out the
- physiological consequences of human leptin deficiency, and reverse it. What's fascinating
- about the leptin system, is that really most of its major effects, certainly on energy balance,
- is very well conserved between mice and humans. So the effects on food intake, energy expenditure on
- obesity are really very similar between mice and humans. Another very important feature,
- which we'll come back to in a moment, is the fact that, in the absence of leptin, neither mice
- nor humans have normal reproductive development. They have a severe central defect in reproductive
- onset. They can occasionally go into puberty, but very late, 20, 30 years later than normal.
- They also have other abnormalities, such as impaired cellular immunity.
- One of the questions that always comes up when we show how dramatic leptin is as benefiting human
- congenital leptin deficiency is, if leptin is so fantastic, why isn't it a blockbuster
- obesity drug? The reason it isn't a blockbuster obesity drug is that the dose-response
- relationships between leptin and its output, i.e. body-fat stores, is very, very non-linear.
- You can see here normal BMI ranges, where you have plasma leptin, you're already reaching
- the levels at which changing leptin levels are really having very little effect on body weight.
- This actually makes sense, because a beautiful paper, soon after the discovery of leptin, by my
- old boss Jeff Flier, with Rex Ahima, really showed that leptin, in evolutionary terms, is a signal of
- fasting and starvation, not excess. So the principal purpose of leptin is in its rapid
- decrement when body-fat stores fall, to signal a whole range of both behaviours i.e. food-seeking,
- switching off reproductive axis, switching off energy-burning processes, so the principal purpose
- of leptin is to signal the transition from the adequately-nourished to the inadequately-nourished
- state by dropping. So it's a bit like Sherlock Holmes' dog that didn't bark in
- the night. It's doing most of its signalling in its absence rather than its presence.
- How does leptin work? Well, leptin primarily works in the central nervous system. That's
- predominantly where its long form of its receptor is located, within the hypothalamus and a few
- other nuclei within the brain. There, it works, again, not exclusively, but very dominantly,
- in a small region of the arcuate nucleus of the hypothalamus, through a few thousand neurones
- only in the mice that produce, on the one hand pro-opiomelanocortin system agonists, on the
- other, an antagonist called AGRP. Roger Cohen, I think is definitely worth celebrating for his
- contribution to both identifying the family of melanocortin receptors in the first place,
- and being the first person to show that the melanocortin 4 receptor was likely to be the
- key receptor downstream of leptin, regulating energy, food intake, and energy balance.
- Around that time, in the late 90s, early 2000, Sadaf and I were working very closely together
- in this field. We were joined by two people who have stayed in Cambridge long term, and with whom
- we've interacted and collaborated continuously ever since, and that's Tony Coll and Giles Yeo,
- and we'd like to just use this opportunity to thank them for being fantastic colleagues
- over the years. Giles came from Sydney Brenner's lab, having been a PhD student
- with Sydney Brenner, and brought real molecular biological expertise to our group, and Tony,
- as a clinician scientist, very bravely took the leap into our animal phenotyping, and creating
- our ability to study mouse models and model human disease in animals. Together, their contributions
- to this body of work have been phenomenal. So I'd like to use the opportunity to thank them both.
- After the discovery of the first two human genes, the third one came along, and simultaneously with
- the lab of Philippe Froguel, we reported that humans who lacked melanocortin 4 receptors also
- became obese in a very clear manner. I think probably the next most
- interesting thing that Sadaf and I discussed - we won't talk about the O'Rahilly food factor,
- though. We have ways of investigating and thinking about how we might express this data.
- What we were very interested in is why obesity occurred in these patients with these mutations,
- and what was remarkable, really, is that bringing people into a clinical research facility and
- measuring their food intake, one could relate - particularly in the middle here you'll see that
- that we have melanocortin 4 receptor complete and partial loss of function - the molecular
- properties of the mutation dictated how much food was eaten when people sat at a test meal.
- A kind of remarkable correlation between a molecular property and a complex human behaviour.
- So we established that this was probably the commonest
- single-gene mutations in which could lead to penetrant obesity,
- and more recently, working with the University of Bristol and the children of the 90s cohort,
- on a birth cohort, we could see that about 200,000 people in Britain are likely to carry heterozygous
- mutations in MC4R, and when they do, they're on average - with the beautiful data that we have
- from Bristol - on average, they carry about 15 kilos of excess fat, so about a 200,000
- people going into their adult lives just due to the mutations in this single gene alone.
- To delve a little deeper into the melanocortin pathway briefly,
- the melanocortin peptides come from a complex pro-peptide precursor proopiomelanocortin,
- or POMC. POMC is processed in a tissue-specific manner to generate its hormonal outputs,
- and briefly, their ACTH from the anterior pituitary, which acts on the adrenal gland
- to control cortisol production, and in the skin POMC is produced and controls both pigmentation
- and sebum gland activity through the MC1 and 5 receptor. PCSK1, the gene I mentioned at the very
- beginning, is actually a critical convertase that does this chopping and this proteolytic processing
- in these tissues to produce the bioactive peptides that control melanocortin production.
- To control energy balance, we're clearly doing it through the central nervous system. So
- where is that happening? Well, there are only two of the receptors expressed in the CNS. There's the
- melanocortin 4 receptor I told you about, and the melanocortin 3 receptor, the kind of quiet
- cousin of the melanocortin receptors that no one's been quite clear what it what it does.
- Two mouse models were generated by two different groups over the last period of time,
- and when they were knocked out, the mice lacking MC3 are not obese, but they do have a
- high fat-to-lean mass ratio. They're also a bit shorter. They have impaired linear growth and low
- IGF-1 concentrations, but, really, there's been no clarity about what melanocortin 3 receptor does in
- human biology. Recently, we had an opportunity - and I think this exemplifies where we are now in
- contemporary science, with the power of largescale sequencing in the UK Biobank, with the power of
- consanguineous cohorts - how we can really now start to ask fundamental physiological questions
- using human genetics. Almost for the first time, we can start getting there. I'll just briefly,
- in one slide, tell you the MC3R story. So what we did first, was look in UK Biobank and
- find enough people with enough mutations, and then do the functional studies in vitro. That's Alice
- Williamson in particular, a PhD student in the lab ,did beautiful work that's assisted by Brian Lamb.
- Then, working with our colleagues in the MRC Epidemiology Unit,
- we were able to show that those individuals who carried loss of function mutations in UK Biobank,
- actually had a very significant delay at menarche. They had a lower adult height.
- Their measure of muscle mass was reduced compared to normal muscle, and they had low circulating
- IGF-1, and this largely correlated with the degree of mutational dysfunction. We really wanted to
- find somebody who didn't have any MC3R, and that's where having access to consanguineous cohorts
- comes in. Here, Genes & Health, led by David van Heel - but I see Richard Trembath here in
- the audience, who was another begetter of this study - and working with Sarah Finer, we were
- able to look into the Genes & Health study, find an individual who had a homozygous MC3 mutation,
- make it in the test tube, show that it was dead, and then go back to the patient themselves. Who,
- indeed, went into puberty at the age of 23, was extremely short, and had an exceptionally low
- lean mass for their degree of body weight. I should point out that there's no evidence
- that MC2r links to body mass index or adiposity. There's not a single milligram
- of excess fat in the heterozygous carrier. So it's not like MC4, which does the body weight.
- We turn to colleagues in the US, Roger Cohen and Richard Simile, who had the MC3, or knockout,
- mice and were able to show the following. When you take a wild-type mouse and fast it,
- much like a young woman who might go on a diet has a delayed next period, if you fast a mouse,
- you increase the cycle. We know that nutrition influences cycle length, and it does so exactly in
- the wild-type mice, but that's totally dependent on having a functional MC3R. In the absence of
- MC3R, nutritional signals are not transmitted to the reproductive axis in those mice.
- Then we started to look in the brains of both mice and humans by single-cell RNA seq,
- and by single-molecule fish. Brian Lamb, in the first case,
- did beautiful work showing that that MSH is highly and selectively co-expressed in central neurones
- that regulate both growth in the one hand, growth hormone releasing hormone, and in the peptides
- that release the gonadotrophin releasing hormone, the classic trigger of puberty, and they're highly
- coexpressed and selectively coexpressed in those neurones. Of course, the next body of work will be
- to take those out of those neurones and finally, absolutely prove the functional link between those
- findings. So I think this can allow us to suggest a bifurcating model of nutrient sensing by the
- central melanocortin pathways, such that the MC4R is concerned with appetite regulation and energy
- storage. These neurones all take in signalling from leptin, from insulin, from amino acids,
- and then send a bifurcating signal regulating what we do with excess nutrient. We grow,
- we accrue lean mass and we time sexual maturation. It's kind of the signal of the good times.
- This is when nutrition is abundant, these are the things that we want to do.
- So I think what we can say is, that this bifurcating signal regulates through the MC4 the
- acquisition and retention of calories, but through MC3R regulates the disposition of caloric energy.
- This, I think does have some explanatory power. For example, we know that both obese
- leptin-deficient mice and humans have hyperphagia, obesity and reduced energy
- expenditure, all of which are replicated by people and mice with MC4R deficiency.
- Ob/ob mice, and particularly mice, have reduced linear growth. Less clear in humans,
- but both have extreme pubertal delay, but these are not found in MC4R deficiency, and we believe
- that this is the case because MC3R is what is mediating the link to both growth through GHRH
- and the central control of reproduction. In my final slide, I'd like to say that this
- pathway could provide some of the neuroanatomical substrate for phenomena that have really been
- globally seen over the last century. In other words, over the last century, we have all become
- taller as a species, but as in most countries, we've continued to grow, - it's plateauing out
- now - but always been attributed to nutrition, but not very clear how that happens.
- Similarly, our age at puberty has massively decreased in the vast majority of countries,
- and so we think that the pathway, the neuroanatomical pathway we've just defined,
- is likely to be the pathway that underpins that. So to summarise, hypothalamic melanocortin system
- senses nutrient status through these various signals. It acts through MC3R to regulate the
- hormones controlling growth, lean mass, and onset of puberty, and provides a substrate for those
- profound cellular changes in height and pubertal timing. I'm going to hand over to Sadaf.
- Thank you, Steve. So these genetic discoveries together have provided a
- molecular framework for trying to understand how human energy homeostasis is regulated.
- These hypothalamic melanocortin neurones project widely throughout the brain,
- and it's through those neural connections that you regulate energy intake, energy expenditure
- and substrate utilisation, how much carbohydrate and fat do we oxidise. Now, what we learned from
- our studies in patients with congenital leptin deficiency, is that the major way in which
- the leptin melanocortin pathway regulates energy homeostasis is through the regulation of appetite.
- So you can see the children on the top-left there with severe obesity.
- When we first observed these children, we found that the major problem was an incredible drive to
- eat. They're incredibly hungry, constantly asking for food. They behave as if they're starving and
- they're happy with any kind of food, even hospital food, which is obviously a disorder of appetite.
- Now, what we realised is, when we treated these children with recombinant leptin injections,
- there was a dramatic change in their eating behaviour. Their hunger went down,
- they became much more choosy, and they actually experienced satiety for the first time.
- Now we were really intrigued to try to see what would happen in the brain in response to images
- of food, and to that extent, we did some imaging studies using functional MRI to see the brain's
- response to pictures of food, versus not food as a control. A fairly straightforward study design.
- What we found was that leptin-deficient children, when they see any picture
- of food, it doesn't matter what it is, they get huge activation in the striatum
- and the accumbens, areas of the brain that respond to reward or pleasure,
- and when we treat them with recombinant leptin, even after just seven days, that neural activation
- goes down. Now, it's very interesting, that reward pattern in the brain is very specific to food,
- because whilst they can't tell between broccoli and a burger, they can discriminate between
- a Ferrari and a Fiesta. So it's a really specific response in the brain to food.
- Now, their brain activation changes, but also so does their behaviour. We recorded that very
- simply using liking ratings, and you can see here in the untreated state, before treatment,
- they really rate their cake very highly, but they also rate cauliflower very highly, which
- is obviously abnormal. Then, after seven days of leptin administration, there's a kind of, quite
- like the cake, but now cauliflower is zero out of ten, which is more of a normalised response.
- So a really fundamental observation here was that a hormone such as leptin can regulate a really
- complex behaviour, such as how much you like food. Now this would make a great deal of sense,
- because if you're starving or you're lacking in nutrients, which is what leptin is there to
- signal to the brain, then you would want to be looking for food and craving food. Of course,
- this is also what happens when people go on a diet and they're in the weight-reduced state.
- So leptin, clearly, regulates food reward, but it also seems to have a role in macronutrient
- preference, so that's the choice of food that you make whether you choose fat, carbohydrate
- or protein. Now, there's been evidence for quite some time that macronutrient preference is highly
- variable in animals - there's a lot of variation in inbred strains - but it's also highly variable
- and inherited in humans. Studies of identical twins have shown considerable heritability for
- macronutrient preference, and some really lovely studies done in ob/ob mice, actually
- before leptin was discovered, showed that these mice, that we subsequently found out lack leptin,
- definitely prefer fat at the expense of carbohydrate. Then, more recent studies,
- focusing on the melanocortin 4 receptor. showed very nicely that if you take any mice and you
- switch them to a high-fat diet, they will eat more, but the melanocortin 4 receptor-lacking mice
- will definitely eat more, and they will continue to eat more. When you challenge these mice with
- a forced choice, either you can have this or this, they always prefer high fat and not really sugar.
- A series of studies in animals using either genetic manipulations or pharmacological
- manipulations, have shown very clearly that, if you disrupt or you tone down the melanocortin
- pathway, you increase the preference for fat and you decrease the preference for sucrose,
- and the opposite happens if you manipulate the pathway in the other direction.
- Again, that makes evolutionary sense. This pathway is there to defend us against starvation.
- If you're starving, you would want to have more fat. You get twice as many calories per gram
- of fat than you do of carbohydrate, and you can store fat more easily. So it makes a great deal
- of sense that the pathway should also govern food preference. Now, we set about trying to see if we
- could test whether the same actually happens in humans. This is quite challenging, and a really
- seminal study is led by Agatha van der Klaauw with colleagues, Julia Keogh and Elana Henning,
- who really took on the challenge of trying to find an experiment that we could do in people
- that would test these parameters. Now, it's quite challenging in people, because we
- eat food that's a combination of fat, carbohydrate and protein, but after a lot of experimentation,
- I was personally involved in some of the eating. We managed to find that a British classic,
- which is chicken korma and rice, was the perfect experimental tool for this. What we could do was
- we could manipulate the fat content, giving 20 per cent, 40 per cent, or 60 per cent fat,
- without altering the appearance, the taste, the texture, or the sensory properties of the
- meal. Then, when we give people, lean people, obese people, and people with MC4R mutations ad
- libitum amounts - they were quite large trays - of the chicken korma, and we see how much they eat,
- what we find is that people tend to eat similar amounts, but actually the people
- with the MC4R mutation ate significantly more from the high-fat chicken korma without realising it.
- Now, when we did the similar experiment, but this time for sucrose preference,
- using a true British classic, Eton mess, what we found is that most people like the sweeter version
- of the Eton mess, but actually the people with the MC4R mutations really did not like
- the sweeter versions. In fact, they hardly liked the taste at all. So really, as in mice,
- it does look like this pathway is also important in food preference in humans. Really, what this
- whole body of work shows, is that nutritional signals and hormonal signals, such as leptin,
- are really playing a critical role in telling the brain about nutrient availability
- and in governing our eating behaviour, and several aspects of that behaviour.
- Now, of course, our behaviour is also modified by external cues in the environment,
- the sight of food, the taste of food, the smell of food, and how much food is available, but I think
- it shouldn't be underestimated what a powerful role these biological factors have to play, and
- in certain people, they can have a dominant role in how much they eat and what they eat. Really,
- this leads to the fact that in humans we can see that, as in animals, eating behaviour has both an
- innate or hardwired component, but also a learned component that you pick up from the environment.
- Now, what about energy expenditure? Most of the work we've done has been on energy intake.
- We didn't really find any major effects in basal metabolic rate, but we did see
- some evidence of impaired sympathetic tone which would affect substrate utilisation,
- and really, the clinical clue came from the finding of impaired sympathetic tone, coupled
- with low blood pressure and low heart rate, even when people are given a physiological stress,
- where here they're being infused with intravenous insulin as part of a clamp. So people who are
- lacking the melanocortin 4 receptor have low blood pressure and low heart rate,
- despite severe obesity. Now, interestingly, at the same time, colleagues at Eli Lilly had developed a
- first-generation melanocortin 4 receptor agonist, so stimulate the pathway. They found that if you
- stimulate the pathway, people lose weight, but actually their blood pressure goes up. So, sadly,
- we were responsible for the closure of this drug development program, because of this work,
- but it did highlight the critical link between the melanocortin pathway, weight and blood pressure.
- Now, what we showed is actually it's broader than that. It's the leptin melanocortin pathway as a
- whole that couples weight and blood pressure. In some very elegant studies led by our colleague
- Michael Cowley, we showed in humans and he showed in mice, that lacking leptin or the leptin
- receptor, these mice, despite the fact that they are severely obese, have a low blood pressure,
- even when you challenge them with a high-fat diet. When you give ob/ob mice that are lacking leptin
- back leptin, they lose weight, but their blood pressure goes up. So together, this really fits
- with the fact that the leptin melanocortin pathway is the primary pathway coupling weight change with
- blood pressure. We know when we treat patients that if people gain weight, their blood pressure
- goes up. If they lose weight, their blood pressure goes down. This is the explanation for that.
- So as people gain weight, they have more fat mass. Leptin is released, goes to the brain
- through the melanocortin circuits, drives up sympathetic tone, and drives up blood pressure,
- and in our patients who have genetic disruption of this pathway, you don't get that rise in blood
- pressure, and so they have relatively normal blood pressures, despite severe obesity.
- Now, moving forward, what we've found is that this core melanocortin pathway
- really seems to be central, and is also important in other genetic disorders that we've identified.
- Now, we've been very fortunate that over the years in work led by Julia Keogh, Elana Henning and
- Rebecca Bounds, we've really recruited a large cohort of patients, with the help of clinical
- collaborators across the world, into the Genetics of Obesity Study, or GOOS, and genetic studies in
- that cohort have allowed us to find new genes that seem to regulate the function of this pathway in
- different ways. So this is just to focus in on the POMC neurones that respond to leptin,
- and we've more recently identified genes in which there are rare variants in our patients
- that directly affect the transcription of POMC, or indirectly affect its transcription by interacting
- with other transcription factors. The example here is SRC1. On the bottom of this slide here,
- you can see that when we model human mutations in mice, we find that the SRC1 variants not
- only suppress the transcription of POMC, but also the depolarisation of POMC neurones.
- Now, these neurones are so critical that you can affect not only their function,
- but also their development, and in work led by Agatha van der Klaauw,
- using exome sequencing in the GOOS cohort, we identified rare variants in a cluster of 14 genes
- that together play a critical role in the development of these neurones.
- So these are the semaphorin 3s, their receptors and their co-receptors,
- and they're critical for axon guidance. They tell the axons where to go during development. What we
- found is that variance in these family of genes will prevent the axons going to the right place,
- and when we model that in mice, we can see that they lack the right neural circuits,
- because they're basically being misdirected, and the mice gain weight as a result of that.
- Now, what we've learned is that actually quantitative variation
- in this melanocortin pathway can contribute to the regulation of weight in a number of ways.
- So in collaboration with Nick Wareham and Claudia Langenberg, we looked at the UK Biobank,
- half a million people who volunteered for research in the UK, and we focussed in on the melanocortin
- 4 receptor, which we've already studied. We found 61 different variants, and here we
- characterise them functionally in great detail, studying the canonical pathway,
- which involves G protein signalling, where you measure how much cyclic AMP is produced,
- but also an alternative pathway mediated by beta arrestin recruitment. What we found was that there
- were some variants that caused a loss of function, so MC4R is not working, and those people tended
- to be obese, and that's similar to our earlier finding in the children with severe obesity. We
- also found there were other variants that caused a gain of function, the receptor was too powerful,
- and those gain of function variants were associated with protection from obesity,
- and indeed a 50 per cent reduced risk of type two diabetes.
- So how do those variants work? Well, they work by keeping the receptor on the cell surface
- so that it continues to send a signal to suppress appetite, and that would explain why people remain
- protected from obesity. Now, the crystal structure of MC4R has recently been described in both the
- inactive and active state, and we hope, therefore, that it might be possible to mimic the effects of
- those natural mutations and design drugs that do the same thing, resulting in safe and effective
- weight loss. Now, our target has been to try to see if we can find new ways
- to harness the power of genetics for the discovery of weight loss therapy. So we've already focussed
- a lot of attention on the rare variants that cause severe obesity. We know that from the
- work of many colleagues, that common and rare variants in the population can either increase
- your risk of obesity or confer protection against obesity. But what about these people at the other
- end of the spectrum, the people who remain thin in an otherwise obesogenic environment?
- Now, we've been interested in these individuals for some time. I must admit to not being very
- sympathetic about their plight, but what we decided to do was this would be a really
- interesting thing to study. So what we did was approach GPs across the country,
- from over 1000 GP practices, a real testament to the power of the NHS to do this kind of research,
- and we recruited over 3000 people who are healthy but thin. That's with a BMI of 18,
- and we really enriched for people with a family history of thinness, and we took great care to
- exclude people who had eating disorders, exercise too much, or any medical condition that could
- explain their thinness. In collaboration with Inȇs Barroso, we did a genome-wide association study in
- the STILTS cohort, comparing them to severely obese people and 10,000 normal weight people.
- We found that the thin people have less of the obesity susceptibility alleles. So they're thin
- because they're biologically lucky, not, Steve, because they're morally superior.
- Now, we're doing some exome sequencing in the thin people, because we're very keen to find some rare
- variants that might be associated with their thinness, and if we can find those variants,
- we want to see how those genes work. What do they actually do? Now, it's quite likely that
- some people are thin because they have smaller appetites, but we're really fascinated by a group
- of people who tell us that they can eat what they like, and they still don't gain weight.
- Really keen to find out how that works, and to do that, we're going to bring people in for some
- physiological studies. We have state-of-the-art facilities for these studies in Cambridge,
- and we can bring people in and challenge them by overfeeding them, and see how they respond and how
- come they don't gain weight. So watch this space. Now, this is really the sort of mechanistic work,
- and it provides, really, the framework for how we think about the problem of obesity, but our
- interest has always been to try to translate that knowledge to improve the lives of patients.
- Already now, we have many genes that cause severe obesity, and genetic testing is now
- recommended in clinical guidelines worldwide. This is important for many conditions,
- but it's particularly important in severe obesity, where people are hugely stigmatised because of a
- condition that everybody thinks is their fault. This is particularly the case for children with
- severe obesity, whose parents are often blamed for causing this, with no consideration that there may
- be biological factors to play. I just wanted to highlight the headlines of this newspaper here,
- about one of the children that I looked after, who turned out to have a leptin receptor mutation,
- and the family were, effectively, hounded from their home, because the media, and actually a
- government white paper, blamed them for causing the death of their child with severe obesity.
- I've been involved in at least 30 of such cases, where people have tried to take children away from
- their families and into social care, by blaming the families for causing obesity. I think one,
- we hope, impact of our work, has been to try to stop that kind of behaviour and encourage a more
- passionate and considered view towards patients. We're now entering an era where we have treatments
- for these patients, because we understand the mechanism that's driving their obesity.
- Very recently, recombinant leptin, which we've been using for over 20 years, has been licensed
- for the treatment of congenital leptin deficiency. It's also being trialled to license for another
- disorder where leptin is very low due to a lack of fat cells. That's lipodystrophy.
- There's a new drug that we've, again, recently had approval for in the US, and I hope very soon
- by NICE, which is setmelanotide. Now, this is a second-generation melanocortin 4 receptor
- agonist, that causes weight loss without changing blood pressure, and we're very
- excited by the results of clinical trials that we've been involved in. This is data from the
- very first patient with POMC deficiency who was treated in Berlin, who lost 25 kilos in 12 months,
- because you're targeting the mechanism that's driving her obesity. We have data now from
- other patients with POMC deficiency and leptin receptor deficiency, and those patients are
- doing very well. As we found several other genes that are converging on that same pathway, we're
- really optimistic that a number of those patients will benefit from this drug in clinical trials.
- So really, just to conclude, together and with many colleagues, we've shown that the central
- melanocortin pathway integrates hormonal signals to modulate food intake and body weight in humans.
- This pathway also is fundamental to coupling nutritional status to growth, body composition,
- and pubertal timing. There's a lot of different types of genetic variation, but what's clear,
- whether it's rare variants, common variants, obesity or thinness, is that the genetic variation
- affects the central pathways that are involved in the regulation of weight, and plays a considerable
- role in susceptibility to human obesity. We're entering an era now with this molecular basis.
- There's an increased chance of success for precision-medicine approaches to treat patients
- with specific subtypes of obesity, and in doing so, we provide proof of principle that the central
- mechanisms of appetite regulation can be targeted safely and effectively, and that's potentially
- generalisable to more common forms of obesity. So really, it's my pleasure now, just to conclude
- by thanking the many people who've helped us with this work over 25 years. So firstly, I wanted to
- start by thanking our colleagues who have referred patients to the GOOS cohort from around the world,
- and of course, our many academic collaborators, a number of whom are here, and we're really
- delighted to share this occasion with them. I'd particularly like to thank our funding bodies,
- who've really given us longstanding support over the years, which has allowed us to take on some
- of these risky and somewhat challenging questions. I'd particularly like to thank our many colleagues
- that we've collaborated with in the Institute of Metabolic Science. I'm delighted to see
- Inȇs Barroso here. We've collaborated for over 20 years on many different types of genetic studies,
- but we're also closely co-located with the MRC Epidemiology Unit. I think I saw Nick
- Wareham here also as well, and we similarly have collaborated and continue to benefit
- from this expertise in genetic epidemiology across the range of metabolic disease.
- We have strong links with our colleagues in the hospital, and I also wanted to really flag,
- most importantly, all the people who did the work, so our current and former members of both labs and
- teams, for their hard work, for their creativity, whether it's experiments or chicken korma,
- and for their dedication to really improving the lives of patients through our science. Most
- importantly of all, I wanted to finish by thanking the patients and their families. Thank you.
- Thank you both very much indeed for absolutely outstanding lectures. Absolutely wonderful. We
- have about ten minutes or so for questions, both live in the audience - there are microphones,
- please wait until a microphone arrives at you before you start speaking - and I would
- like to invite anyone who's online. I am holding a tablet in front of me, and I'm hoping it will work
- for anyone online. If you'd like to put your questions on Slido, I will view them
- and choose ones. I thought if we could start with the live audience. So at the front here.
- Well, that was terrific. Thank you both very much. I'd like you to
- say a bit more about the separation of the blood pressure regulation
- and the obesity, because you did spell out rather nicely that these were both downstream of leptin,
- but that didn't quite seem to fit with Steve's model, that leptin, once you were
- within a normal weight range, wasn't doing much. Also, I'm intrigued by the newer
- melanocortin 4 receptor agonist, which seems to have separated the two effects.
- Okay, so if I could start? So, clearly, we don't know all the answers to those questions. What's
- clear is that leptin clearly does have a role in the regulation of blood pressure and in coupling
- weight to blood pressure. Some of those effects are mediated through the melanocortin pathway.
- Some are probably mediated through other pathways. One pathway we know about involves
- something called GPR10, which is expressed in the dorsomedial hypothalamus and seems to
- be another pathway. So I agree, it's not all of the story. I think it's part of the story. Then,
- when it comes to why this particular drug doesn't increase blood pressure, that's very intriguing.
- It's pretty clear it does not, and we and many other colleagues have been following
- those poor patients obsessively to see if it does. So really unknown,
- because we've tried various things. It's not a question of biased signalling, for example,
- so does this particular drug have different signalling properties
- and send the signal through one particular pathway versus the others? Whether the drug is
- agonising specific receptors in specific brain regions, or subpopulations, we don't yet know.
- So I think the simple answer is we don't yet know, and we'd have to actually do the
- experiments to figure that out.
- what's very clear is that the absence of leptin and melanocortin signalling alters the normal
- relationship between BMI and blood pressure. What's not yet entirely clear is how much
- that contributes to the obesity-associated blood pressure seen in the general population. That's
- a trickier question.
- Thank you Steve. Thank you Sadaf. I just wanted to ask about obesity and malignancy,
- and I don't know whether anybody, or you, I don't know whether you've done any work on this,
- but there's a paper in this week's Nature, actually, which shows different cytokine pathways
- in obese and non-obese mice, which might underlie the increased development of
- malignancies in people who are obese. I don't know whether you've got any comments on that.
- I'll briefly comment on that, because there are numerous potential mechanisms.
- There are some that are relatively trivial. The biggest increase in
- obesity-associated carcinomas is the oesophageal adenocarcinoma, which is almost certainly
- increased intraabdominal pressure and reflux oesophagitis being the dominant
- causal factor in increasing your risk of having a lower oesophageal cancer as visceral adiposity
- increases. Then there are non-trivial mechanisms, and I think for some of the cancers, like
- endometrial cancer, it's pretty clear that it's hyperinsulinaemia is the major... So as you become
- obese, you definitely become hyperinsulinaemic. There's a very strong - and insulin is a very
- potent anti-apoptotic factor, and particularly for the endometrium. So there are some of the cancers,
- and then there's a whole range of possible steroid metabolism issues, and then we can get into many,
- many, many putative mechanisms. I think it's an important question, one that we'd like to, in the
- Institute, do a bit more to start addressing.
- from Emmanuelle. In an obesogenic environment where nutrition transfer and sedentary lifestyle
- are the major modifiable drivers of obesity, what effort is being made to discourage
- the consumption of overprocessed foods and encourage the intake of wholemeal foods?
- Neither of us wants to take that. No.
- So I think, of course, as you all know, there are many public health efforts being undertaken
- to try to tackle our obesogenic environment, because we know that across the population,
- the amount of food we eat and the type of activity that we do is a major driver of that. So,
- of course, there are many population initiatives. I think one of the challenges is that some of
- those initiatives assume, that if everybody just follows the rules, it'll all be okay,
- and I think what we've clearly shown is that it's not as simple as that, and there are a lot
- of reasons why some people will have difficulties in, for example, eating certain foods, undertaking
- certain levels of activity. So I think, broadly speaking, and I think sometimes people think that
- we're arguing against those kinds of public health policies, and clearly the answer is we're not,
- but I think what we are united in advocating for, is that there needs to be a more nuanced
- understanding of the complexity, and that you need both public health policy changes and
- treatments to care for people with severe obesity.
- Right at the back. Yes, on the end. Apologies for the microphone.
- Thank you for a very splendid lecture. I was wondering, obviously, you're looking
- at the people who have extreme obesity that's, obviously, very genetic, but the common garden
- variety of obesity that most people have, is probably not so clear-cut how this is
- caused, I guess. I just wonder, do you have any thoughts about how your research relates to
- the ordinary person who has obesity and how we'll better treat that?
- I'm sure both of us may want to say something about that. What I would say is that he
- common forms of obesity still have a very strong heritability. So the
- heritability doesn't disappear when you go away from the extremes. It remains the case across the
- entire spectrum. Now that we have genome-wide studies in over a million people worldwide,
- we've got a pretty good idea of what that architecture of the genetic susceptibility is,
- and rather remarkably, it's very focussed in the similar areas of the brain that we see in our rare
- individuals. You might ask, what have we learned from the rarities? Well, we've learned that it's
- largely a disorder of appetite, and if we fix the appetite, we fix the obesity. What what we're
- now seeing for the very first time, with the development of the GLP-1 receptor agonists and
- other related molecules, that common or garden obesity, even when pretty intractable and severe,
- is responding very well to a similar strategy. So if you can find a safe and effective molecule that
- helps people control their appetite, you get really effective weight loss in a way that's
- never been possible through simply providing people with dietary and exercise advice.
- So I think it is generalisable, to some extent, and as I said, that does not change the fact that
- we need to change the obesogenic environment. There are all sorts of health reasons for why
- we should do that. It's very difficult, and it's politically very difficult, because
- obesity isn't a knowledge-deficiency disorder. We won't cure it by dropping leaflets around
- telling people not to be fat. We need to do much more than that. So I don't know whether
- Sadaf wants to add any more to that.
- but I think so. I think, other than to say that I think it really is just about
- a variation. We've been operating at the extremes, but the same factors contribute at the
- middle of the distribution, and clearly both environment and genetics and their interaction,
- so how much weight you gain in a given environment with a certain degree of overeating or loss of
- activity. So it's clearly more nuanced than I think our public health policies suggest.
- I'm just going to ask one more question online, and then there will be two final ones from the
- audience, because we're close to the end of the time. The online one is linked in to what you
- were just saying, I think, I hope, and this is from Wojciech. I've chosen it because there's a
- whole heap of likes online associated with it. For people with weight, appetite, metabolic issues,
- like myself, what practical solutions can we count on? The two or three medications you listed,
- is that all? I know you were just saying something about that. How does one determine
- the underlying basis of the issues and subsequent [?therapies 0:54:46.7]? I think, fundamentally,
- how can all of these people be helped right now? What can you tell them to do?
- Obviously if I had a simple answer to that, I would probably be in the Bahamas
- with my simple answer. I think it's difficult to know, and I think it's hard to comment on
- personal circumstances, because I don't know that individual's personal circumstances.
- I think, broadly speaking, the things that we have mentioned
- are relevant. So if people have severe obesity and really struggle with their weight,
- then there needs to be a medical assessment of those individuals to consider genetic and other
- causes of their severe obesity, and plan the treatments for those individuals. For people
- in the population who struggle with their weight, and that's a lot of people, as we mentioned at the
- very beginning, it's clearly a combination of the diet we eat, the amount of activity that we do,
- but also increasingly, there is a recognition that telling people and guiding people and leaving them
- to it, is not really very effective. So it's about people being able to access the support,
- which I think in this country, traditionally, we haven't been so good at that, and there is
- a greater drive now to make wider access to everything from behavioural interventions to
- medical and surgical interventions. So I'd say probably go and see your doctor.
- Thank you. So two more. The gentlemen in the blue shirt towards the back. Thank you.
- Thank you very much. I was just wondering from a gene therapies perspective, would that also be an
- interesting avenue to explore as a therapeutic, as well as the pharmacologic interventions?
- Yes, I think with the sort of almost renaissance, or re-emergence of gene therapy,
- I think there certainly is interest. We're talking to a number of companies, actually,
- who are interested in this space. Obviously, leptin would be a very good candidate for gene
- therapy because it's localised to the adipose tissue, but we don't really need it because
- recombinant leptin works very well, but for people with very severe intractable hyperphagia
- due to dysfunction of these pathways, then yes, I do think it's potentially something
- that could be explored. There are now ways of targeting drugs to the right place,
- by virtue of MRI imaging, and also using particular types of tools for delivery. So
- I think that is an exciting new frontier.
- Thank you. Is there a mechanistic understanding of middle-age spread and the sort of the
- age-related increase in BMI?
- Broadly, the answer is everything gets worse. The control of body mass index by the hypothalamus is
- an active process. So if you have a car accident and you damage your hypothalamus, by and large,
- it won't lead you not to eat, it will lead you to eat more. So the control of energy
- homeostasis is a highly active process, and in the nihilistic view that everything gets worse,
- that's probably - but what you actually said is not entirely right throughout life,
- because as you get very much older, then body mass index tends to drop. Maybe we think we have
- some insights, molecular insights, into why that might be, because there are certain TGF-beta-like
- peptides, cytokines, GDF15, etc., which do go up dramatically with age, and do have a suppressive
- effect on food intake. So there may be changes later in life, which actually explain why the
- people over 80, for example, find it hard to keep on weight rather than keep off weight.
- I guess as endocrinologist, we should add, there are obviously hormonal changes
- around that time, that particularly influence both men and women, and differently.
- Okay, so thank you very much indeed for everyone who contributed questions and helped a really nice
- discussion. Don't go away yet, because, before I say a final thank you and ask everybody to do the
- usual applause - I'll try not to drop that on the floor - I have the great pleasure of presenting
- you with your medal and scroll. If it had been the olden days, I'd be giving you a cheque check
- as well, but cheques no longer exist, or sort of thing. So in the order, Sir Stephen.
- Thank you. Thank you very much. Thank you.
- Oh, you haven't sawn it in two.
- Thank you again for an absolutely wonderful lecture.
- Those who know they are invited to the reception, can you please move
- to the City of London Rooms? Thank you.
The Croonian Medal and Lecture 2022 given by Sir Stephen O'Rahilly FMedSci FRS and Professor Sadaf Farooqi FMedSci FRS.
Globally, over 1 billion people are overweight or obese. Obesity and its complications are associated with significant morbidity and mortality due to type 2 diabetes, cardiovascular disease and some forms of cancer. While the widespread availability of high calorie, palatable food and physical inactivity are major environmental drivers of the rise in prevalence of obesity, heritable factors play a substantial role in influencing a person’s propensity to gain weight (or not), within an obesogenic environment.
In this lecture, Professors O’Rahilly and Farooqi will discuss how the identification of genes and the pathways they regulate has highlighted the fundamental role of the brain in modulating eating behaviour and body weight. They will discuss how studies of the leptin-melanocortin pathway have provided a mechanistic framework for understanding how body weight is regulated in humans, how energy status is coupled to reproduction and growth and how disruption of these neural mechanisms causes obesity. They will also discuss the impact of this work on societal perceptions of obesity and how targeting these mechanisms has provided new treatments for people with severe obesity.
About the Royal Society
91TV is a Fellowship of many of the world's most eminent scientists and is the oldest scientific academy in continuous existence.
/
Subscribe to our YouTube channel for exciting science videos and live events.
Find us on:
Bluesky:
Facebook:
Instagram:
LinkedIn:
TikTok: