Genetics in Obesity
Article Written by Dr Alex Blakemore, Professor of Human Molecular GeneticsDisclaimer:
I am a geneticist, working for Imperial College London. The facts that I present will be as accurate and as up-to-date as humanly possible, but the commentary and opinions are my own. I'm writing this as plain old Alex (aifb) so if anything that I say upsets you (hope not!), you need to sue me personally, not the College!
Part One: Setting the Stage
Is fat bad?
Body fat is not a bad thing, it's necessary and normal. We need it for energy and water storage, insulation, support and protection of organs, proper reproductive and immune system function. It's also a strong sexual symbol. A healthy woman might have 25-35% of her body composed of fat - think about that - ONE QUARTER TO ONE THIRD! See here for desirable body fat charts….http://www.dietandfitnessresources.co.uk/info_charts/body_fat_chart.htm
It is very important that we have enough fat on our bodies. People without enough body fat are ILL - women are infertile, males and females are at higher risk of death (e.g. from infections) and of osteoporosis (weak and brittle bones). Because of this, our bodies have developed complex systems to make sure that we have just the right amount of body fat. Unfortunately, these can go wrong…..
Why are people fatter now than in the past? It must be environmental, rather than a genetic effect because genes can't change that fast….
We are the first generation to live with such an abundance of food (especially during the winter). Stone-age people had no access to supermarkets (M&S food, yum, yum!), high calorie prepared foods (Green & Blacks, yum, yum!) or fast food restaurants (variable yumminess). Also, we don't need to use as much energy to actually get food nowadays - not too much hunting and foraging in our modern lives.
Our bodies evolved in a very different environment: it used to be that we needed to be very motivated to find food so we have strong reward systems (pleasure) to encourage us to do this. In the distant past, those people who sought out the most food, ate the most calories and stored the most energy in their bodies (i.e. tended to get fat) were the ones who survived harsh times. In that environment, they (we) would be the healthiest people around and the most likely to reproduce. So, any small "normal" gene sequence differences that reinforce eating behaviour (by giving us extra pleasure from eating, for example) or increase fat storage would tend to be common in the population. These are not "bad" genetic variations - we have champion stone-age survival genes!
Our environment is now very different, but our bodies are still the same. Unfortunately, in the modern world those same gene variants are no longer advantageous - and people who carry them are much more vulnerable to our current unhealthy food and lifestyle: these people get fat. In this way, the changing environment exposes the effects of genes.
We obese people spend a lot of time fighting our bodies, hating our bodies, punishing ourselves and often thinking of our bodies as the enemy. But remember that these genes kept our ancestors alive in hard times. I find it helps to think of my body, not as the enemy, trying to make me fat to ruin my life, but like a daft old auntie always trying to get me to wear a woolly vest to keep me healthy through the winter. My body wants me to enjoy food, eat plenty and store fat so I will survive the winter (it doesn't know about M&S). It's trying to save me, not ruin my life!
What is the evidence that genes are involved at all? How big an effect do they have?
To see how important the effect of genes might be on any particular trait (in this case, fatness), geneticists look at twins. Identical twins are genetically identical (or as near as possible) whereas non-identical twins are no more similar than other brothers and sisters. This means that the differences between identical twins are environmental, whereas non-identical twin pairs have both genetic and environmental differences. To see how strong the genetic effects are, we measure how much more similar identical twins are than non-identical twins.A number of people have done this and the best evidence is that in adults, about 70% of the variation in fatness between people is due to genes. In children, the effects are even stronger: 77% for BMI and the same for waist circumference: here's an abstract and commentary for that work….
http://www.ajcn.org/cgi/content/full/87/2/275
http://www.ajcn.org/cgi/content/abstract/87/2/398
Another way to test this is to look at children who have been adopted: it turns out that they resemble their birth parents' body sizes and shapes much more than they resemble their adopted families. This confirms the importance of genetics in determination of fatness. In fact, obesity is highly genetic: for example much more so than high blood pressure (29%) or depression (50%) and comparable to height (81%).
Part 2: In which I introduce the main players - genes
If you sat in a crowded room and looked at the people around you, it's a fair bet that you would not look as much like the people around you as you look like your mum and dad, or brothers and sisters. That's because of the genes you share with your family members.So what exactly are genes?
Well, actually it's not as easy as you might think to define this. Perhaps the simplest thing is to say that "a gene is a piece of genetic material that codes for a useful product". In most cases that useful product is a protein (eg a hormone like insulin, or a structural protein like the collagen found in skin and bones).You can think of the genes as a set of instructions, in code, which tell the cell all the "recipes" for the different proteins it needs to make. These "recipes" are cleverly encoded in the chemical structure of the genes themselves.
For most genes, you get one copy from your mum and another from your dad, so that you have some "insurance" if one copy is not working properly.
In humans (and most things, but not all viruses) the genetic material is an extremely long thin molecule called deoxyribose nucleic acid (DNA to its friends). You all know what it looks like because the pictures are on telly all the time: a spiral staircase shape. The "steps" or "rungs" of the staircase are composed of different chemicals, called bases.
The four main bases are usually represented as the letters A,C,T and G. Two bases fit together to make one complete "rung", and if you know what base is on one side, you can tell what its "partner" will be as the A always pairs with T, and the C always pairs with G (this is important as it is how genes are copied when cells divide - the whole DNA molecule gets "unzipped" and each half is filled in with new, matching bases, so that you end up with two double-stranded DNA molecules instead of one).
For more details of the structure and function of DNA, look here:
http://learn.genetics.utah.edu/
It's a great site for learning the basics of genetics (and also has some interesting stuff on the genetics of addiction).
Another interesting site for learning genetics is here:
http://www.dnaftb.org/dnaftb/15/concept/index.html
So what are chromosomes then?
When cells want to divide (either for growth or wound repair, or to make eggs and sperm, or whatever), they need to share out the DNA between the two baby cells so that each gets the proper complement of genes.You can imagine how tricky that is with long, long strands of DNA all tangled up together. Just like sorting out a messy basket of knitting wool, the only sensible thing to do is to wind it up into compact "balls of wool". That's exactly what a chromosome is - DNA all packaged up neatly (with some special proteins to keep it in position).
"Normal" humans have 23 pairs of chromosomes, including the sex chromosomes, X and Y (females have two X chromosomes and males have one X (from their mother) and a Y chromosome (from their father).
The biggest pair of chromosomes is called chromosome 1 and the next biggest is chromosome 2, etc. all the way down to teeny-weeny chromosome 22.
There's nice stuff on chromosomes in the two sites mentioned above, but if you want to play at actually analysing problems with human chromosomes, you can do it here:
http://www.biology.arizona.edu/human...yotyping2.html
What about genetic differences between people?
A mutation is, strictly speaking, any heritable change in DNA sequence, although it is usually used where such a change results in disease.The loss or gain of a whole chromosome (like in the exercise in the link above) has very major effects (often it's fatal before birth), and having a big chunk of chromosome missing or duplicated is also very serious. Some of these chromosomal problems include obesity among their range of effects - study of this can help us find important genes influencing fatness.
There are a range of much smaller changes in DNA sequence, though, that are less catastrophic. Depending on exactly what the change is, it can sometimes stop a gene working completely causing a genetic disease (like cystic fibrosis or haemophilia) or impaired function (like colour-blindness). These things are usually individually quite rare. Don't feel too unusual though, if you find out you have such a mutation - each of us carries a number of these (I have heard the estimate that everyone carries about 300 major disease-causing mutations). Luckily, in most cases we don't get the disease because our "insurance copy" can compensate. There are a few known examples where obesity (in humans or in animal models) can be caused by a major mutation in a single gene. I'll tell you about these later on.
There are also thousands and thousands of "normal" variations in the DNA sequence too (for example causing differences in eye colour, height, blood group and all the other small differences between us all). These small differences (called polymorphisms - meaning "multiple forms") affect exactly how our bodies work. They do not directly cause disease on their own, but each variation has a subtle effect and they may add up to contribute to obesity, diabetes, risk of cancer, etc. (often in an interaction with lifestyle and environment). We'll come to discussion of this later.
How can we find the key genes for obesity?
There is a range of ways, including:- study of obese animals, including particular strains of mice
- study of extreme childhood cases of obesity to find what genes might be affected by major mutations
- study of more common types of adult obesity to see which genes might contribute in more subtle ways.
We'll talk about what we have learned from obese mice later.
Part 3
Many people are sceptical about the role of genes in obesity and over the last few months I've given a lot of thought to why this is. I think it might be that because obesity is caused by overeating, many people think that since it's due to behavioural and psychological factors, it cannot be genetic.I think that sometimes people think that:
1) if you say something is "genetic" that means it's a fixed thing and there is no point in even trying to change it ("its an excuse")
or
2) Since behaviour changes all the time in response to the environment, and since we can (for a while at least) consciously control our behaviour, it can't be genetic ("I can affect it, so it can't be predetermined")
I hope to convince you all that neither is true.
What can obese mice tell us about human obesity?
You know, it's really rare to see a mouse that is fat. Usually lab mice (like wild mice) manage to balance their food intake and activity levels so that they stay nice and sleek. In fact, if wild mice got really fat, they wouldn't last long because they'd be too slow to run away from predators.
Every so often though, a genetically different (mutant) mouse arises. About 50 years ago, a fat mouse spontaneously appeared within a colony of laboratory mice. This mouse was fat because it was apparently hungrier than its littermates. It ate much more and got fatter and fatter until it was twice the weight of its brothers and sisters: see this picture:
http://images.google.co.uk/imgres?im...%3Den%26sa%3DN
This was the first evidence that severe obesity could have a completely genetic basis. The mouse did not have a distressing childhood, alcoholic parents, a history of emotional or sexual abuse, a depressing job or an unsatisfactory marriage. It didn't go to McDonalds or spend too much time playing computer games. It had the same food as the other mice, but just ate more of it. Something had gone wrong with whatever mechanism usually helped mice balance their food intake with their energy expenditure.
The scientists all those years ago tried to find out what was going on. Breeding experiments produced more obese mice and showed that the effect was caused by a mutation in a single gene (you can tell this by the patterns of inheritance). It was a recessive effect, which meant that both copies of the gene (one from each parent, remember) were mutated. The scientists speculated that there must be some kind of signaling molecule that was not working properly. There were two main possibilities:
1) that the signal up-regulated eating behaviour and it was overactive in the obese mouse OR
2) the signal down-regulated eating behaviour and was broken in the obese mouse. This was the most likely scenario since most mutations screw things up rather than making them work harder.
To test whether (1) or (2) was the explanation, they did a rather disgusting experiment (don't shout - it wasn't me - and they were allowed to do these things all that time ago). They joined the circulatory systems of an obese (ob/ob) mouse with a normal mouse. As a result of this, the obese mouse reduced its eating and became leaner.
This experiment showed that there was some kind of signaling molecule carried in the bloodstream, which controlled appetite. After many more years of work, the molecule was isolated - it was a small protein hormone which they called leptin (after leptos, which is Greek for thin apparently).
Leptin is produced by fat cells. The more stored fat we have, the more leptin is produced. It is released into the bloodstream, and travels up to the hypothalamus (part of the brain that controls feeding behaviour). There, the signal molecule is detected by receptors on the surface of specialised brain cells. This causes a chain reaction of different further signals, with the end result of decreased appetite.
It's a neat system for allowing the brain to monitor the amount of body fat present and alter food intake accordingly. When there is a lot of leptin around, appetite is decreased.
To cut a long story short the ob/ob mouse doesn't make any functional leptin. This means that the hypothalamus "thinks" that there is no fat at all on the body. This would be an emergency survival situation, so the brain sends out emergency "EAT" signals. No matter how much fat the mouse has, it can't send the leptin signal. Consequently, the emergency signal to eat is constantly on - the poor mouse literally feels starving.
You know that to survive is our most basic instinct. Those emergency survival signals to eat override all other things.....the poor mouse is always desperate for food.
As well as controlling satiety (feeling satisfied when you have eaten), leptin has a range of other roles in the body, including effects on bone growth, immune function and, very importantly, reproduction. Obese mice are infertile and, in humans too, leptin has important effects on fertility. Without a certain level of body fat stores, a woman could not safely carry a pregnancy and breastfeed a child, so leptin levels also regulate the onset of puberty and, even after puberty, regulate ovulation.
How does this relate to humans?
After the discovery of leptin deficiency in mice, people wondered whether a similar thing could happen in people. Professor Steve O'Rahilly and his co-researcher Sadaf Farooqi, who work in Cambridge, were looking after a family where affected children had very severe, early-onset obesity, with hyperphagia (overeating). They had immune problems and did not undergo puberty.....sound familiar?When the children's blood was tested, no leptin was detected. When their leptin genes were sequenced, it was found that, like the obese mice, they could not make functional leptin (the recipe, encoded in the sequence of bases in that part of the DNA molecule, was changed - and not in a good way!).
The good part of this story is that when Steve and Sadaf found out what was wrong with these children, they immediately knew that there was a chance to treat them. These pictures show pictures of leptin-deficient children before and after treatment with leptin hormone - it was pretty miraculous - they not only got thinner, but the immune function improved and they had normal pubertal development!
http://images.google.co.uk/imgres?im...%3Den%26sa%3DN
And here are pictures of Steve and Sadaf - heroes of this story. They are really nice people.
http://images.google.co.uk/imgres?im...%3D1%26hl%3Den
http://www.neuroscience.cam.ac.uk/di...file.php?isf20
Once we found leptin, we were able to find other genes in the same pathway - mutations in these genes also cause single-gene forms of obesity.......more about this lter!
OK, so I bet you're all out there saying "where can I get me some leptin?" aren't you?
Well the company that owned the rights to the leptin gene (Amgen paid $20 million for them), thought the same thing. They hoped that leptin could be used to treat obesity in humans, just as in the obese mouse. Unfortunately, though, the first clinical trials were unsuccessful and when we look at the levels of leptin in the people's blood, it's not hard to see why. There is a positive correlation between leptin levels and BMI - most fat people already had loads of leptin. I'll try to upload a graph of some data produced by my research group in about 1995, that shows this very clearly(in the gallery). My blood was tested too - I'm circled - and you can see that I'm awash with leptin - and yet I had no desire to stop eating at all!
The problem lies with the leptin receptor on those specialised brain cells in the hypothalamus. For some reason, they are not detecting the signal, so my body thought I looked like this:
http://images.google.co.uk/imgres?im...%3Den%26sa%3DN
and was sending out emergency "EAT, EAT, EAT!" instructions. It's a bit like in type 2 diabetes where people can actually be making loads of insulin, but the insulin receptors seem to have "gone to sleep" so they don't respond normally to it.
Just as rare families with catastrophic mutations in the leptin gene had been found, pretty soon people were found with a deficiency of the leptin receptor. Their symptoms were very similar, with early onset obesity, raging appetite and failure to go through puberty. Here's the report of the first such family:
http://www.nature.com/nature/journal.../392398a0.html
One of the key researchers was Philippe Froguel, whom I didn't know at the time, but is now my (rather brilliant) boss. Sadly, the children with leptin receptor deficiency don't respond to treatment with leptin (they were already making it themselves and had extremely high leptin levels, but with no receptor, it wasn't detected by the hypothalamus).
Here's a rather amusing picture of Philippe - it makes us all laugh because he's wearing a lab coat and holding a pippette for publicity purposes only - we would never allow him to play around in the lab - he's useless at that stuff!
http://images.google.co.uk/imgres?im...%3Den%26sa%3DN
Well, I guess now you're probably thinking "But those people with major leptin and leptin receptor mutations are really rare - what's that got to do with most peoples' obesity?" The answer is this: when researchers find important genes in obesity, we then look to see whether more subtle genetic variations in the sequence of the same genes have an effect on obesity too - increasing appetite in "normal people" so that they gain weight.
The way that we do this is to take a set of obese people and a set of similar (age/sex/ethnicity, etc.) non-obese people and look to see whether some "normal" genetic variations are more common in one group than the other. When we first started doing this sort of study, we did not use enough people (the work was expensive and difficult), so results were unreliable and contradictory. Nowadays, we use much bigger groups (techniques to detect the variations are now easier and about 1/100th of the cost) and the upshot is that subtle variations in the leptin and leptin receptor genes ARE associated with higher BMI. The effect is rather small though - we guess that for most fat people (at least those with adult-onset obesity), several genes are acting together - each increasing appetite a little bit.
So what are the other genes? And are there any with major effects that are NOT rare?