August 18, 2007
Locusts, Cannibalism & Human Obesity
By Simpson, Steve
Steve Simpson explains how his basic research on the behaviour of locusts has surprising ramifications for understanding a worrying human epidemic. Locusts have the extraordinary capacity to change from harmless grasshoppers into mass swarming pests. The trigger for this remarkable and devastating transition is being touched by other locusts on the hind legs - something that can be simulated by tickling them with a paintbrush.
Underlying cannibalism in crickets and locusts is a specific and powerful appetite for protein. These are not the only animals that regulate their protein intake - so too do flsh, rodents, birds and humans. In fact, our protein appetite may well lie at the heart of the modern obesity epidemic: protein has the power both to drive obesity and to cure it.
I will take you on a strange journey. We begin in the midst of a desert locust swarm in northern Africa (Fig. 1). This animal is one of a dozen or so species of grasshopper that are called locusts, some of which live in Australia. They are defined by having one peculiar feature that distinguishes them from all other grasshoppers: they are essentially two animals packed inside the same genotype.
To illustrate this let us take the two juvenile desert locusts shown in Figure 2. The one on the left has been reared on her own. She is a beautiful green colour, which allows her to blend in with the vegetation where she lives and avoid the attention of predators. She would live a solitary existence, being shy in her behaviour and repelled by other locusts.
In contrast, her sister on the right was reared in a crowd. She is brightly coloured and would be a very active animal, being attracted by other locusts and tending to aggregate. These aggregations form marching bands and, when they become adults, flying swarms containing billions of locusts that migrate hundreds of kilometres and cause devastation to agriculture in vast areas of Africa and Asia.
The process of changing from the solitary to the gregarious phase is initiated by crowding, and begins to occur really quickly. Behaviour is the first feature to change. If you took the animal on the left and put her in a crowd, within an hour she would be attracted rather than repelled by other locusts, and within a few hours she could be part of a marching band. This transition is known as "density-dependent behavioural phase change", and my research group has spent the past 15 years trying to understand it.
What is it about being in a crowd that causes the change? What stimuli are provided by other locusts in a crowd that might trigger the transition? On reflection, it could be the sight of other locusts, their smell, being contacted by them, or perhaps their sound.
As it turns out, touch is critical. What happens is that the environment brings together solitary locusts, against their predisposition to avoid each other, at limited patches of food plants. The congregated locusts jostle with each other, and it is this jostling that causes the change from repulsion to attraction. But simply being touched just anywhere won't work - which is where the paintbrushes come in.
We spent many happy hours in Oxford in 2001 tickling locusts on various body parts with paintbrushes and demonstrated that, unless you stimulate hairs on the hind legs of the animal, you don't cause the change from solitary to gregarious behaviour (Fig. 3). This was an important discovery because it allowed us to delve into the nervous system and to focus our search for the controlling neural pathways and the molecular changes that accompany the process of phase transition.
Let us return to a recently aggregated group of locusts, which at this point are simply milling around. They are moving around and staying together but not going anywhere in particular until suddenly, as if of one mind, the entire group becomes highly aligned and starts to march. As it marches, the mass of locusts looks like a river flowing through the environment. Of course, up close it is not a fluid but is made up of particles - and those particles are locusts - which gave us a clue about how we might study the collective behaviour of locusts.
In 2005 we turned to our statistical physics colleagues, who have developed a set of models called "self-propelled particles models", in which they take individual particles, program them to behave in simple ways with respect to their neighbouring particles, and then simulate what happens in clouds or swarms of such particles. Using these models we found that the collective decision to start marching emerges within a crowd from simple local interactions between locusts.
There is no leader of locusts. There is no hierarchical control. Marching emerges because the locusts are following a very simple rule: "align with your moving neighbours". Once a critical density of locusts is reached, just adding one or two more to a local area will suddenly cause the transition to collective aligned movement. The march has begun.
We were able to explore the phenomenon in the laboratory in Oxford and Sydney by causing locusts to march endlessly round and round an arena shaped like a Mexican hat. We could measure the interactions between individual locusts in the crowd and, by adding more and more locusts into the arena, demonstrate very good accordance between what the animals did and what the self-propelled particles model was predicting.
A Different Mormon
This raises the next question: why do locusts align with their moving neighbours and what causes such alignment? The answer to this question came from a related animal called the Mormon cricket, and it turned out that the answer was really rather sinister.
The Mormon cricket is a large flightless cricket that lives in America's northwest, and forms vast marching swarms extending kilometres (Fig. 3). They are called Mormon crickets because they began to devastate the first crops planted by the Mormon pioneers after arriving at Salt Lake in 1848. The community was powerless to stop the destruction and were facing starvation until a flock of seagulls came to the rescue and ate the crickets.
During fleldwork in Utah in 2005 Greg Sword, Pat Lorch, lain Couzin and I discovered that the Mormon crickets are on a forced march to find protein. If they don't keep moving they don't find more protein, but worse than that is that if they stop moving they become somebody else's protein meal (Fig. 4). Were you or I to be incapacitated in the face of these animals I think that we would become a protein meal as well (Fig. 5).
It happens that Mormon crickets have a specific appetite for protein, and if that appetite is satiated the animals stop cannibalising and don't march. A specific protein appetite is not restricted to Mormon crickets, but is found in most animals - including locusts.
When a locust is offered the opportunity to mix its own diet it will ingest a very precise "intake target" of protein and carbohydrate. But if you don't allow it to get to that point by feeding it on nutritionally unbalanced diets, then the locust prioritises protein intake over carbohydrate intake. As a result, on high carbohydrate diets they over-consume carbohydrate in an effort to extract their target level of protein, while on high-protein diets they under-eat carbohydrate rather than eat much more than the target protein intake.
There are costs for not reaching the intake target: if a locust eats too much carbohydrate it becomes obese and dies early. Granted, it is pretty hard to tell that a locust is obese because it has an exoskeleton, but it is chubby on the inside: rather like an overweight knight being wedged into his suit of armour. In contrast, locusts end up very lean on higher than optimal protein diets; they lose fat mass and become susceptible to starvation, which is ecologically very relevant to a locust.
When David Raubenheimer and I looked at other animals we found a very similar situation to what we saw in locusts: protein intake is tightly regulated and takes priority over the intake of other nutrients. We have found the same pattern in rodents, fish, birds and humans. In an experiment in 2003 in which we incarcerated a group of people in a chalet in the Swiss Alps and provided them with a buffet of foods we found that, as in locusts, protein intake was prioritised when the diet forced a trade-off between protein and non- protein energy (carbohydrate and fat).
What are the implications of having a dominant protein appetite? If there were to be a shift towards including more high-fat and high- carbohydrate items in the diet, our powerful protein appetite would cause us to eat too much fat and carbohydrate to gain limiting protein. Unless you get rid of those excess calories, your body composition will change and you may become tipped into a vicious cycle towards morbid obesity.
Fat deposits, especially those packed around our internal organs, release compounds called free fatty acids into our bloodstream. These compounds interfere with insulin's role of inhibiting protein breakdown in the liver and muscles. As a result, our bodies start to metabolise protein for energy - something which normally only happens during starvation. Protein metabolism leads to an increase in the need for protein in the diet - which, on a low protein diet, will drive even further over-consumption of fat and/or carbohydrate and result in even greater gain in body fat. It becomes a vicious cycle. To give an idea of the power of this "protein leverage" effect, if your diet changed from 14% protein to 12.5% (as a percentage of total energy), which sounds a trivial reduction, you would have to consume 14% more non-protein energy to get the same absolute quantity of protein. This is exactly what has happened over the past 40 years in the United States. The effect is made worse when the requirement for non-protein energy diminishes as a result of reduced levels of exercise.
In contrast, if your diet shifts towards a higher percentage of protein then your body will not let you over-consume protein to a great extent, the result being that there is an under-consumption of non-protein energy, a negative energy balance, and the potential to lose weight. This is why high-protein diets help people to lose weight. (The Atkins and CSIRO diets are examples that have attracted both popularity and scientific controversy.)
The protein leverage effect might explain why people typically regain weight rapidly after coming off a low-calorie diet (the yo- yo diet effect). During dieting they have lost muscle mass, which increases the need for protein in the diet and drives overeating of fat and carbohydrate. In this regard it is telling that studies have shown that people are less prone to regaining weight after a period of calorie restriction if their diet is higher in protein.
The protein leverage effect might even explain why some populations are more at risk of obesity than others. Hunter gatherers and those living on oceanic islands are especially susceptible to obesity and related metaboli c disorders when shifting to a westernised diet The ancestors of these people did not go through the agricultural revolution 10,000 years ago, which saw a massive incorporation of starch into the diet, but rather continued to live on a diet relatively high in protein. These people would have a higher need for protein in their diet than populations who had evolved for 10,000 years on a higher-carbohydrate diet - which would make protein leverage effects especially strong when faced with a modem western diet that is high in fat and carbohydrate.
A recent experiment of ours on caterpillars supports the idea that animals - and perhaps humans too - evolve metabolic responses that match their natural diet. After only eight generations of rearing them on a high-carbohydrate, low-protein diet, caterpillars had evolved so that they no longer stored excess ingested carbohydrate as body fat. In other words, they had evolved resistance to obesity. In contrast, caterpillars reared in a low- carbohydrate, highprotein world evolved to store any excess carbohydrate that came their way as body fat.
It appears that protein has the power to drive obesity and also to ameliorate it. We are currently testing this hypothesis in tightly controlled human trials, as well as in animal studies. Meanwhile, our locust research continues apace in an effort to improve the management and control of this terrible pest.
Steve Simpson is Professor and ARC Federation Fellow in the School of Biological Sciences at the University of Sydney. This article was developed from his presentation on being admitted as a Fellow of the Australian Academy of Science.
Copyright Control Publications Pty Ltd Aug 2007