IT’S not the most exciting experiment I’ve ever seen. An ordinary white mouse sits inside a sealed glass jar, breathing. But that’s enough for comparative physiologist Tony Hulbert. Each time the mouse exhales, it releases a trillionth of a gram of two gases, ethane and pentane. All animals, you and I included, puff this stuff out 24/7. It is nothing to us. But Hulbert believes that by measuring these two gases he can tell how quickly this candlewick we call life will burn down.
Across the spectrum of Noah’s ark, from the 2-gram shrew to the 200-tonne blue whale, much about an animal is determined by its size. In general, the larger the beast, the slower its metabolism and the longer its life, and vice versa. But the question of how nature imprints each creature with its assigned metabolic rate, and why some are destined to die sooner than others, is a long-standing mystery.
Now Hulbert and Paul Else, his collaborator at the University of Wollongong in New South Wales, Australia, think they have the answer. They say it is our membranes – the millionth-of-a-centimetre-thick envelopes that enclose our cells and the components within them – that determine all of these things. Gunky, watertight membranes saddle elephants and whales with slow metabolic rates, but also give them long lives. Runny, leaky membranes allow mice and hummingbirds to live fast – but also causes them to die young, their bodies ravaged by highly reactive free-radical oxygen molecules. Ethane and pentane are by-products of the radicals’ relentless gnawing. If Hulbert and Else are correct, their work could answer many questions raised by other studies into longevity, from why birds live longer than mammals of a similar size, to the puzzle of how restricting calorie intake can promote longevity. In short, it could be a grand unifying theory of life speed, life size and lifespan.
Advertisement
Although DNA and protoplasm have both been called the stuff of life, membranes also deserve that title. Many of the processes of life depend on the random diffusion and collisions of molecules within the water that makes up 90 per cent of us. It is membranes that impose some order on this confusion, and in so doing make life possible. Every cell is separated from the outside world by a membrane, and within a cell each individual structure is sealed in its own membrane like a sandwich in a zip-lock bag.
Membranes are made of molecules called fatty acids, which have greasy, water-repellent tails that pack lengthwise alongside each other, like pencils in a box, forming an oily seal that keeps water from getting across (see Graphic). All told, each of us has enough membranes in our body to cover about 75 soccer fields. And they are ideally situated for regulating metabolic rate, according to Else. “Because membranes are the barriers of life, everything has to move across them,” he says, “and the rates at which things move across them is fundamental to metabolism.”
Take, for example, the mitochondria. These cellular powerhouses contain a convoluted inner membrane, surrounded by a smooth outer membrane. Inside the inner membrane, the concentration of hydrogen ions – protons – is 10 times lower than outside. These fenced-out ions continuously stream in through tunnels in the inner membranes and the mitochondria harness the flow, like a waterwheel, to manufacture the ATP molecules that supply the rest of the cell with energy. The membranes that surround every cell in an animal’s body do a similar job, corralling the extra sodium ions concentrated inside the cell and harnessing their outflow to drive processes such as the importing of glucose and amino acids.
These cross-membrane ion differences have to be maintained constantly, and that job falls to protein machines that continuously chug ions back across the membrane – the so-called proton pump in the mitochondria and the sodium/potassium pump in the cell membrane. What is surprising – perhaps even beautiful – is that the amount of energy consumed in maintaining these cross-membrane ion differences corresponds perfectly to metabolic rate. Despite having very different metabolic rates, cells from a horse, a mouse or an iguana nonetheless spend 20 per cent of their energy running their proton pump and another 20 per cent running their sodium/potassium pump.
What is it about different animals’ cells that make their metabolic rate – and all of its individual components – scale perfectly up and down with the energy used by their trans-membrane pumps, like a set of Russian dolls? For a decade, Hulbert and Else pondered this question, becoming increasingly convinced that the answer must lie in the makeup of the membranes themselves. “From mice to horses,” says Hulbert, “the composition of the cells is relatively constant. The protein content doesn’t seem to vary and the lipid content doesn’t seem to vary. But the types of fatty acids in the membranes vary dramatically.”
As Hulbert surveyed the membrane compositions of different animals, a pattern quickly emerged. The greasy tails of membrane fatty acids are made of chains of between 14 and 22 carbon atoms. In saturated fatty acids, all of the atoms in the chain are connected by single bonds. In unsaturated fatty acids, the chain includes between one and six double bonds, each causing a kink in the chain – and fatty acids with two or more double bonds are called polyunsaturated fatty acids, or PUFA. Hulbert found that the slower an animal’s metabolic rate, the more saturated its membrane fatty acids were. And the faster an animal’s metabolic rate, the more unsaturated its fatty acids were.
The correlation was especially strong for the most unsaturated PUFA of them all, known as 22:6 because it has six double bonds along its 22-carbon tail. The membranes of elephant muscle cells are only about 0.2 per cent 22:6, while those of mice, which have a metabolism that is 35 times faster, are about 20 per cent 22:6. Hulbert found the same trend in birds, with the percentage of 22:6 in membranes scaling in an almost linear way with metabolic rate. And reptiles and amphibians also fit the pattern: they had highly saturated membranes to match their sluggish, cold-blooded ways.
The connection is that the level of PUFA in a membrane affects the way it behaves. The highly saturated membranes of sluggardly animals allow only a few sodium and hydrogen ions to leak across, and so their sodium/ potassium pumps and proton pumps work slowly. But the highly unsaturated membranes in animals with zippy metabolisms are much more leaky, and even though they have the same number of sodium/potassium and proton pumps per gram of tissue as more sluggish animals, these pumps operate at a higher rate. Since both sodium/potassium and proton pumps sit within the membranes, Hulbert and Else couldn’t help but wonder: what if having more unsaturated fatty acids nestled against the pumps somehow caused them to work more frenetically? In 1999 they decided to test their idea.
Given that sodium/potassium pumps in each animal have their own characteristic pace, Hulbert and Else tried to alter that pace by taking pumps from rats, which have a fast metabolism, and swapping them into membranes from cane toads, whose metabolism is seven times slower. The rat pumps slowed down, running at half their original rate, once transplanted into more saturated toad membranes. Of course, slower pumping could also arise from the pumps being damaged during handling. The clincher, then, came when the researchers did the experiment in reverse, sticking slow-chugging toad pumps into rat membranes. The toad pumps sped up – a result that cannot be attributed to damage. Hulbert and Else have since obtained the same results by swapping pumps and membranes between warm-blooded cows and cold-blooded crocodiles.
“It’s not just the sodium pump,” says Else, “it’s the calcium pump, it’s the exchangers, it’s the ion channels, it’s the whole kit and caboodle. I think the properties of membranes can lift up or push down overall activities of membrane-bound proteins.” If he is right, then membranes could be setting the pace of metabolism in everything from walruses to whooping cranes.
The notion that membrane composition might affect the activity of proteins certainly seems plausible when you consider that cells whose job is to transmit signals rapidly contain especially high levels of 22:6 in their membranes. It plays a vital role in neurons and retinal cells, for example. Indeed, all mammals contain the same proportion of 22:6 in their brains – about 15 per cent – and animals deprived of 22:6 in experiments grow to be slow-witted and half-blind.
The key seems to be that 22:6 is far more flexible than other fatty acids. Nuclear magnetic resonance studies and computer simulations show that it can flip-flop between different shapes billions of times per second. The result is a membrane with “tumultuous disorder, and behaviour that is almost liquid-like”, says Klaus Gawrisch, who did some of the experiments at the National Institutes of Health (NIH) in Bethesda, Maryland.
To see what difference this makes, you need look no further than sight itself. When light strikes a retinal cell, the process of triggering a nerve impulse begins with activation of a protein called rhodopsin embedded in the cell membrane. Rhodopsin tilts in the membrane, pushing fatty acids out of the way, and then finds and activates a thousand individual G proteins, which are essentially messengers that wade around knee-deep in the membrane. Each G protein must in turn find and activate a thousand phosphodiesterase enzymes, which also wade in the membrane. For us to see things in motion, it all has to happen in two-thousandths of a second.
Rhodopsin works efficiently because 22:6 doesn’t pack in membranes snugly, says NIH biophysicist Drake Mitchell. “You get voids, and rhodopsin, when it changes conformation, seems to exploit those voids to push against the membrane.” And for those membrane-wading G proteins and phosphodiesterases, having lots of 22:6 around probably makes the difference between wading in water and wading in molasses. This becomes apparent when you replace 22:6 with a similar PUFA that has just one fewer double bond. Then you get a 50 per cent decrease in signalling intensity.
The take-home message is that speed requires runny membranes. And growing evidence suggests that a whole multitude of cross-membrane signalling systems require membranes of just the right consistency to function normally. For example, Leonard Storlien from AstraZeneca in Gothenburg, Sweden, has found that, in humans, abnormally low PUFA content in membranes correlates with insulin resistance, suggesting a link between modern diets and the growing incidence of adult-onset diabetes, high blood pressure and heart disease. Also, low dietary intake and low blood levels of 22:6 correlate with depression, bipolar disorder and violent behaviour, while several clinical trials indicate that dietary 22:6 supplements can treat depression and bipolar disorder (żěè¶ĚĘÓƵ, 24 August 2002, p 34).
So membranes crammed full of PUFA set us free to hot-rod down the metabolic highway of life, but those PUFA also exact a horrible price: the more of them our membranes contain, the shorter our life. In all mammals and birds that have been looked at, membrane PUFA content correlates inversely with lifespan. This is because those carbon-carbon double bonds that make our membranes well greased and souped-up also happen to be vulnerable to attack by the free-radical oxygen molecules that leak out of every mitochondrion as it produces energy. As a result, highly unsaturated membranes will deteriorate and age faster than more saturated ones. Worse still, the oxidised PUFA created when free radicals attack release two nasty chemicals – HHE and HNE – that drift to other parts of the cell, where they damage DNA and proteins.
Evidence for this came when Reinald Pamplona Gras from Lleida University in Spain, used dietary experiments to increase unsaturation by 30 per cent in the membranes of rats and found a 50 per cent increase in membrane and protein damage. And the membrane oxidation theory could also explain the observation that restricting calorie intake lengthens lifespan. It turns out that calorie-restricted animals remove PUFA from their membranes – protecting the membranes from oxidation. The theory could also explain why birds live so long – birds with the zippiest metabolisms still die soonest, but for any given metabolic rate, birds live twice as long as mammals. Bird membranes contain lower levels of omega-3 PUFA, the kind that includes 22:6 and is most vulnerable to oxidation.
The ultimate test
Given its far-reaching implications, it is not surprising that Hulbert and Else’s theory is provoking plenty of interest. “If their idea flies,” says George Somero, a comparative physiologist at Stanford University, California, “it will have some very important effects on people’s thinking about rates of living and evolution of warm-blooded animals.” But other researchers won’t be convinced overnight. For example, it is not clear that the increased protein and membrane damage Pamplona Gras found actually compromises cell function. And while the ultimate proof would be to lengthen an animal’s lifespan by changing membrane composition, in practice doing this makes animals sick. “It’s an interesting idea because so much depends on membranes,” says Brian Merry, a zoologist studying lifespan at the University of Liverpool, UK. “But it’s a hypothesis at the moment, which is waiting to be either disproved or verified.”
Hulbert is gearing up for experiments to address these questions. Working with blowflies, which can easily be manipulated to change their PUFA content or alter longevity-associated genes, he will assess how these changes alter the production of ethane and pentane — the telltale gases that reveal the amount of membrane damage caused by oxygen free radicals – and whether changing the fatty-acid composition of membranes can increase lifespan.
But for now, the task falls to a tiny mouse, sealed inside a glass jar, to reveal the depths of our own mortality, and to help explore the rules of our allotted days here on Earth.
- “Life, death and membrane bilayers” by Tony Hulbert, Journal of Experimental Biology, vol 206, p 2303