Power, Sex, Suicide

by Nick Lane

  Bennett and Ruben argued that the resting metabolic rate was eventually elevated to the point that internal heat production could raise body temperature permanently. At this point, the advantages of endothermy—niche expansion, and so on—were selected for their own benefit. Selection was now directed at maintaining internally generated heat, favouring the evolution of insulatory layers such as subcutaneous fat, fur, down, and feathers.

  Sizing up to complexity

  For the aerobic capacity hypothesis to work, both the maximal and the resting metabolic rate of mammals and birds need to be substantially higher than those of lizards. This is well known to be the case.2 Lizards become exhausted quickly and have a low capacity for aerobic exercise. While they can move very fleetly (when warmed up) their muscles are mostly powered by anaerobic respiration to produce lactate (see Part 2). They can sustain a burst of speed for little more than 30 seconds, enabling them to dart for the nearest hole and hide, whereupon they often need several hours to recover. In contrast, the aerobic performance of similarly sized mammals and birds is at least six to tenfold greater. While not quicker off the mark or fleeter of foot, they can sustain the pace for far longer. As Bennett and Ruben put it in their original Science paper: ‘The selective advantages of increased activity are not subtle but rather are central to survival and reproduction. An animal with greater stamina has an advantage that is readily comprehensible in selective terms. It can sustain greater levels of pursuit or flight in gathering food or avoiding becoming food. It will be superior in territorial defense or invasion. It will be more successful in courtship or mating.’

  What must an animal do to improve its stamina and speed? Above all else, it has to augment the aerobic power of its skeletal muscles. To do so requires more mitochondria, more capillaries and more muscle fibres. We immediately run into a difficulty with space allocation. If the entire tissue is taken up with muscle fibres, there is no room left over for mitochondria to power muscle contraction, or for capillaries to deliver oxygen. There must be an optimal tissue distribution. To a point, aerobic power can be improved by a tighter packing of these components, but beyond that improvements can only be made by greater efficiency. This is indeed what happens. According to Australian researchers Tony Hulbert and Paul Else, at the University of Wollongong, New South Wales, mammalian skeletal muscles have twice as many mitochondria as the equivalent lizard muscles, and these are in turn more densely packed with membranes and respiratory complexes. The activity of respiratory enzymes in rat skeletal muscle is also about twice that of the lizard. In total, the aerobic performance of rat muscle is nearly eight times that of the lizard—a difference that wholly accounts for its greater maximum metabolic rate and aerobic capacity.

  This deals with the first part of the aerobic capacity hypothesis: selection for endurance raises the mitochondrial power of muscle cells, leading to a faster maximum metabolic rate; but what about the second part? Why is there a link between maximal and resting metabolic rate? The reason is not clear, insofar as none of the possible explanations has been proved. Even so, there is a good intuitive reason to expect a connection. I mentioned that lizards may often take several hours to recover from exhaustion, even after a few minutes of vigorous exertion. Such a slow recovery is less dependent on muscles than on organs, such as the liver and kidneys, which process the metabolic waste and other breakdown products of vigorous exercise. The rate at which these organs operate depends on their own metabolic power, which in turn depends on their mitochondrial power—the more mitochondria, the faster the recovery. Presumably the advantages of endurance also apply to recovery time: given the eightfold rise in aerobic power of mammalian muscles, if there were no compensating changes in organ function it would take a whole day, rather than a few hours to recover from exercise.

  Unlike muscles, organs are not faced with a dilemma of space allocation—while the density of mitochondria doesn’t change with size in muscles, it does in the organs. As animals get bigger, the power laws that we discussed in the last chapter mean that their organs become more sparsely populated with mitochondria. This is an opportunity in waiting. For the organs of a large animal to gain power, the tissue architecture doesn’t need to be restructured as it does in muscle: it can simply be repopulated with mitochondria. This opportunity seems to have given rise to endothermy. In their classic comparative studies, Hulbert and Else showed that the organs of mammals contain five times as many mitochondria as an equivalent lizard, but in all other respects the mitochondria are no different. For example, the efficiency of their respiratory enzymes is exactly the same. In other words, for every hard-won increment in muscle power, it’s relatively simple to counterbalance the new power by filling up the half-empty organs with a few more mitochondria, so as to ensure speedy recovery from aerobic exertion. The important point is that the function of organs like the liver is coupled to muscle demand, and not with the need to keep warm.

  Proton leak

  But there is a diabolical catch. We have seen that the muscles contribute little to the resting metabolic rate: the danger of oxygen toxicity means that blood is diverted away from the muscles and into the organs, where there are relatively few mitochondria to cause damage. So what might have happened in the first mammals? They had extra mitochondria in their organs to compensate for their higher aerobic capacity, but nowhere to divert the blood, which had to pass through either the organs or the muscles.

  Once our prototype mammal has digested his food, caught so easily with his newfound aerobic prowess, he goes to sleep. Beyond replenishing his reserves of glycogen and fat, there is little call to expend energy. His mitochondria fill up with electrons extracted from food. This is a dangerous situation. The respiratory chains in the mitochondria become packed with electrons, because there is only a sluggish electron flow. At the same time, there is plenty of oxygen around, as the blood flow can’t be diverted. In these conditions, electrons easily escape from the respiratory chains to form reactive free radicals, which can damage the cell. What might be done?

  According to Martin Brand in Cambridge, one answer might be to waste energy by keeping the whole system ticking over. The danger from free radicals is at its greatest when there is no electron flow down the chain. Electrons pass most readily on to the next complex in the chain, and so tend to react with oxygen only when that complex is choked up with electrons, blocking normal flow. Restarting electron flow usually requires the consumption of ATP.3 If there is no demand for ATP, the whole system clogs up and becomes reactive. This is the situation when resting after a large meal. One possible escape is to uncouple the proton gradient, so electron flow is not tied to ATP production. In Part 2, we compared this to a hydroelectric dam, in which an overflow channel prevents flooding in times of low demand. In the case of the respiratory chains, instead of passing through the ATPase to generate ATP (the main dam gates), some protons pass back through other pores in the membrane (the overflow channels), so that part of the energy stored in the gradient is dissipated as heat. By uncoupling the proton gradient in this way, slow electron flow is maintained, and this restricts free-radical damage (just as the overflow channel prevents flooding). The fact that such a mechanism does protect against free-radical damage was verified in a fascinating study of mice by John Speakman and his colleagues in Aberdeen, working with Martin Brand. Their title said it all: ‘Uncoupled and surviving: individual mice with high metabolism have greater mitochondrial uncoupling and live longer.’ We’ll look into this further in Part 7, but in short they live longer because they accrue less free-radical damage.

  In resting mammals, perhaps a quarter of the proton gradient is dissipated as heat. The same is true of reptiles, but they have barely a fifth the mitochondria in each cell and so generate five times less heat per gram. Their organs are also relatively small, so reptiles have fewer mitochondria altogether, adding up to a tenfold difference in heat production. In the first large mammals, proton leak may well have generated enough heat to raise body tempe
rature by a significant degree, simply as a side-effect of aerobic health. Once heat is generated in this way, selection can take place for endothermy for its own sake—for keeping warm. In contrast, small animals can only generate enough heat to maintain their body temperature if they insulate themselves better, and even step up the rate of heat production. These properties probably evolved in the descendents of animals that were already endothermic—otherwise we’re back to the problems of raising body temperature for its own sake. In other words, it’s likely that endothermy evolved in animals that were large enough to balance heat production with heat loss, while the smaller descendents of the first endothermic mammals had to make further adjustments to offset their heat-retention problem. Small mammals like rats need to supplement their normal heat production with brown fat, rich in mitochondria, which are dedicated to heat production—here, all the protons leak back across the mitochondrial membranes to give off heat. This in turn means that the resting metabolic activity of small mammals doesn’t correlate with muscular capacity for hard work, but rather with the rate of heat loss.

  These ideas explain several long-standing puzzles, and scotch any fond lingering notions of a universal constant (metabolic rate scales with mass3/4 across the entire living world) once and for all. It is plain why small mammals and birds (none of which approach large mammals in size) scale with an exponent of about 2/3: the greater part of their metabolic rate is linked not to muscle function, but to maintaining body heat instead. In contrast, for larger mammals and reptiles, heat generation is not a top priority—quite the contrary, overheating is much more of a problem—so the metabolic power of their organs only needs to balance muscular demand, and not heat production. Because the maximal metabolic rate scales with an exponent of 0.88, so too does the resting metabolic rate.

  How closely animals meet these expectations depends on other factors, such as diet, environment, and species. So, for example, marsupials have a lower resting metabolic rate than most other mammals, as do desert-dwellers and all ant-eaters: we can predict that they should take longer to recover from vigorous physical exertion, or will be less likely to partake in it at all; and generally this is the case.4 It seems that the scaling of energetic efficiency presents an opportunity that can be met in different ways, from new pinnacles of aerobic power in energetic birds and mammals, to varying degrees of sloth in well-protected but less energetic animals, whether armadillos or tortoises.

  First steps up the ramp

  Generating energy with mitochondria enables eukaryotic cells to be much larger than bacteria—on ‘average’, perhaps 10 000 to 100 000 times the size. Large size brings with it the gift of energetic efficiency. Within limits that are probably defined by the efficiency of the supply network, the bigger the better. This is an immediate payback for an immediate advantage, and is likely to counterbalance the immediate disadvantages of larger size—the requirement for more genes, more energy, and better organization. The immediate reward of energetic efficiency may have helped push eukaryotes up the ramp of ascending complexity.

  Still a couple of conundrums tease me, but I think they can be explained. First, energetic efficiency is often dismissed as a target for natural selection, on the grounds that large animals still have to eat more food than small animals: the energetic savings are only apparent on a cell-by-cell, or gram-weight basis. The critic is quick to point out that natural selection works at the level of individuals, usually, and certainly not on a gram-weight basis. This is obviously true, but the environment and the needs of an organism still relate to its size. We saw that a rat is seven times as hungry as a human being: in relation to its body size it must find and eat seven times as much food as we do. But a rat is no stronger or faster in relation to its surroundings than we are. The term relative here is real. Clearly, a rat can’t hunt buffalo, but we can, or anything else down to the size of a rat and beyond. The world that animals inhabit is shaped by their size, and in our own world we need to eat seven times less food than does a rat, day in day out. On the same basis, we can survive seven times as long without food or water. The scale of the advantage can be seen even more clearly if we think in terms of how much we need to eat relative to our body weight. A mouse, for example, must eat half its body weight every day to avoid starvation, whereas we need only consume about 2 per cent of our own body weight. Surely this is a genuine advantage. That is not to say size is invariably a dominant advantage—in many circumstances, small size may offer strong advantages, leading to different evolutionary trends; but the energetic efficiency of greater size does seem to have had a deep influence on the direction of eukaryotic evolution.

  The second conundrum that teases me relates to the very pervasiveness of energetic advantage. In Part 4, we’ve considered mostly the mammals and reptiles. We have broken down the energy savings into their components, to conclude that they offer genuine opportunities, rather than merely the constraints of a fractal network. On the other hand, I have also pointed out that bacteria are limited by their surface area-to-volume ratio, and that this is a constraint, not an opportunity. Do single eukaryotic cells, such as amoeba, really have an advantage with larger size? Do trees, or shrimps? Have we, in rejecting a universal constant, also relinquished any right to generalize beyond the example of mammals?

  I don’t think so. I have left other examples aside until now because the answers are less certain—they have received far less attention than mammals and reptiles. Nonetheless, I suspect that most organisms, including single cells, gain the same benefits. In larger organisms, these benefits are the familiar economies of scale: it’s cheaper by the dozen. As in society, such benefits depend on the set-up costs, operational costs, and distribution costs, and these impose outer limits on the economies of scale. But within these limits, the benefits ought to apply widely. This is because living organisms are highly conservative in their operational principles. In particular, their organization is invariably modular. Both single cells and multicellular organisms are made up from a mosaic of functional parts. In multicellular organisms, the organs perform particular functions, such as breathing or detoxification; within cells, discrete functions are carried out by organelles like mitochondria. Modular functions within single cells include genetic transcription, protein synthesis, packaging, membrane synthesis, pumping salts, digesting food, detecting and responding to signals, generating energy, moving around, trafficking of molecules, and so on. I imagine that the economies of scale apply as much to these modular aspects of single cells as they do to multicellular organisms.

  This idea brings us back to the question of gene number, which I touched on at the start of the chapter. We noted that complex organisms need more genes, and thought about Mark Ridley’s argument that the invention of sex enabled the accumulation of genes, opening the gates to complexity. But sex, as we saw, may not have been the gatekeeper, and certainly didn’t limit gene number in bacteria or single eukaryotic cells. I wonder whether the accumulation of genes in eukaryotes is better explained in terms of the energetic efficiency of larger cells. Larger cells usually have a larger nucleus. It seems that balanced growth during the cell cycle requires the ratio of nuclear volume to cell volume to be basically constant—another power law! This means that, over evolution, the nuclear size, and with it the DNA content, adjusts to changes in cell volume for optimal function. So as cells grow larger they adjust by developing a larger nucleus with more DNA, even if this extra DNA does not necessarily code for more genes. This explains the C-value paradox discussed in Chapter 1, and is why cells like Amoeba dubia have 200 times more DNA than a human being, albeit coding for fewer genes.

  The extra DNA is often dismissed as junk, and may be purely structural, but it can also be called upon to serve useful purposes, from forming the structural scaffolding of chromosomes, to providing binding sites that regulate the activity of many genes. This extra DNA also forms raw material for new genes, building the foundations of complexity. The sequences of many genes betray their ancest
ry as junk DNA. Might it be that the origin of complexity was as simple as a scale? As soon as eukaryotic cells became powered with mitochondria, there was a selective advantage to them being bigger. Bigger cells need more DNA, and with this they had the raw material needed for more genes and greater complexity. Notice that this is the reverse of bacteria: whereas a heavy selection pressure to lose genes oppresses bacteria, eukaryotes are under pressure to gain them. If Ridley is correct that sex postpones mutational meltdown, it might be that the requirement for more DNA with larger size was an underlying selection pressure that gave rise to sex itself.

  For eukaryotic cells, the possession of mitochondria raised the ceiling on the possibilities of life. Mitochondria made larger size probable, rather than staggeringly unlikely, inverting the constrained world of bacteria. With larger size came greater complexity. But there were disadvantages too, arising from a conflict between the mitochondria and the host cells. The consequences of this long battle were equally pervasive, marking life forever with deep scars; yet even these scars had the power to create and destroy. Without mitochondria we would have had no cell suicide, but no multicellular ‘individuals’; no ageing, but no sexes. The dark side of mitochondria had even more power to rewrite the script of life.