by Nick Lane
1
The Deepest Evolutionary Chasm
The void between bacterial and eukaryotic cells is greater than any other in biology. Even if we begrudgingly accept bacterial colonies as true multicellular organisms, they never got beyond a very basic level of organization. This is hardly for lack of time or opportunity—bacteria dominated the world for two billion years, have colonized all thinkable environments and more than a few unthinkable ones, and in terms of biomass still outweigh all multicellular life put together. Yet for some reason, bacteria never evolved into the kind of multicellular organism that a man on the street might recognize. In contrast, the eukaryotic cell appeared much later (according to the mainstream view) and in the space of just a few hundred million years—a fraction of the time available to bacteria—gave rise to the great fountain of life we see all around us.
The Nobel laureate Christian de Duve has long been interested in the origin and history of life. He suggests in a wise final testament, Life Evolving, that the origin of the eukaryotes may have been a bottleneck rather than an improbable event—in other words, their evolution was an almost inevitable consequence of a relatively sudden change in the environmental conditions, such as a rise in the amount of oxygen in the atmosphere and oceans. Of all the populations of proto-eukaryotes living at the time, one form simply happened to be better adapted and expanded rapidly through the bottleneck to take advantage of the changing circumstances: it prospered, while less well-adapted competing forms died out, giving a misleading impression of chance. This possibility depends on the actual sequence of events and selection pressures involved, and can’t be ruled out until these are known with certainty. And of course, when we are talking of selection pressures exerted two billion years ago, it is unlikely that we can ever be certain; nonetheless, as I mentioned in the Introduction, it is possible to exclude some of the possibilities by considering modern molecular biology, and to narrow down a list of the most likely possibilities.
Despite my enormous respect for de Duve, I don’t find his bottleneck thesis very convincing. It is too monolithic, and the sheer variety of life weighs against it—there seems to be a place for almost everything. The whole world did not change at once, and many varied niche environments persisted. Perhaps most importantly, environments lacking in oxygen (anoxic or hypoxic environments) persisted on a large scale, and do so to this day. To survive in such environments calls for a very different set of biochemical skills from those needed to survive in the new oxygenated surroundings. The fact that some eukaryotes already existed should not have precluded the evolution of a variety of different ‘eukaryotes’ in different environments, such as the stagnant sludge at the bottom of the oceans. Yet this is not what happened. The astonishing fact is that all the single-celled eukaryotes that live there are related to the oxygen-breathing organisms living in fresh air. I find it highly improbable that the first eukaryotes were so competitive that they annihilated all competition from every environment, even those most unsuited to their own character. Certainly the eukaryotes are not so competitive that they annihilated the competition of bacteria: they took their place alongside them, and opened up new niches for themselves. Nor can I think of any parallels elsewhere in life, on any scale. The fact that the eukaryotes became the masters of oxygen respiration did not lead to its disappearance among the bacteria. And more generally, many types of bacteria persisted for billions of years despite unceasing and unforgiving competition for the same resources.
Let’s consider a single example, the methanogens. These bacteria (more technically, Archaea) scratch a living by generating methane gas from hydrogen and carbon dioxide. We’ll consider this briefly as the methanogens are important to our story later on. The problem for methanogens is that, though carbon dioxide is plentiful, hydrogen is not: it quickly reacts with oxygen to form water, and so is not found in oxygenated environments for any long period. The methanogens can therefore only survive in environments where they have access to hydrogen gas—usually environments totally lacking in oxygen, or with constant volcanic activity, replenishing the source of hydrogen faster than it is used up. But the methanogens are not the only type of bacteria that use hydrogen—and they are not particularly efficient at extracting hydrogen from the environment. Another type of bacteria, so-called sulphate-reducing bacteria, makes a living by converting (or reducing) sulphate into hydrogen sulphide—the gas that stinks of rotten eggs (in fact rotten eggs reek of hydrogen sulphide). To do so, they too can use hydrogen gas, and they usually out-compete the methanogens for this scarce resource. Even so, the methanogens have survived for three billion years in niche environments, where the sulphate-reducing bacteria are penalized in some other respect—usually for the lack of sulphate. For example, because freshwater lakes are impoverished in sulphate, the sulphate-reducing bacteria can’t establish themselves; and in the sludge at the bottom of such lakes, or in stagnant marshes, the methanogens live on. The methane gas they emit is known as swamp gas, and at times it sets alight with a mysterious blue flame that plays over the marshes, a phenomenon known as ‘will-o-the-wisp’, which explains many ‘sightings’ of ghosts and UFOs. But the productions of the methanogens are far from insubstantial. Anyone who advocates switching from exploiting oil reserves to natural gas can thank the methanogens—they are responsible for essentially our entire supply. Methanogens are also found in the guts of cattle and even people, as the hindgut is exceedingly low in oxygen. The methanogens thrive in vegetarians because grass, and vegetation in general, is low in sulphur compounds. Meat is much richer in sulphur; so sulphate-reducing bacteria usually displace methanogens in carnivores. Change your diet, and you will notice the difference in polite company.
The point I want to make about methanogens is that they were the losers in the race through a bottleneck, yet nonetheless survived in niche environments. Similarly, on a larger scale, it is rare for the loser to disappear completely, or for the latecomers never to gain at least a precarious foothold. The fact that flight had already evolved among birds did not preclude its later evolution in bats, which became the most numerous mammalian species. The evolution of plants did not lead to the disappearance of algae, or indeed the evolution of vascular plants to the disappearance of mosses. Even mass extinctions rarely lead to the disappearance of whole classes. If the dinosaurs disappeared, the reptiles are nevertheless still among us, despite stiff competition from birds and mammals. It seems to me that the only bottleneck in evolution comparable to that which de Duve postulates for the eukaryotic cell is the origin of life itself, which may have happened once, or perhaps numerous times with only one form ultimately surviving—in which case this, too, was a bottleneck. Perhaps, but this is not a good example, for we simply don’t know. All we can say for sure is that all life living today ultimately shares the same ancestor, and so sprang from the same progenitor. Incidentally, this rules out the view, expressed by some, that our planet has been populated by successive waves of invasions from outer space—such a view is not compatible with the deep biochemical relatedness of all known life on earth.
If the origin of the eukaryotic cell was not a bottleneck, then it was probably a genuinely unlikely sequence of events, for it happened only once. Speaking as a multicellular eukaryote, I might be biased, but I do not believe that bacteria will ever ascend the smooth ramp to sentience, or anywhere much beyond slime, here or anywhere else in the universe. No, the secret of complex life lies in the chimeric nature of the eukaryotic cell—a hopeful monster, born in an improbable merger 2000 million years ago, an event still frozen in our innermost constitution and dominating our lives today.
Richard Goldschmidt first advanced the concept of a hopeful monster in 1940—the year that Oswald Avery showed that the genes are composed of DNA. Goldschmidt’s name has since been derided by some writers, and held up as an anti-Darwinian hero by others. He deserves neither epitaph as his theory is neither impossible nor anti-Darwinian. Goldschmidt argued that the gradual accumulation of small
genetic changes, mutations, was important, but could only account for the variation within a species: it wasn’t a sufficiently powerful source of evolutionary novelty to explain the origin of a new species. Goldschmidt believed that the big genetic differences between species could not be mounted by a succession of tiny mutations, but required more profound ‘macro-mutations’—monster leaps across ‘genetic space’, which is to say the gulf between two different genetic sequences (the number of changes required to get from one to another). He appreciated that random macro-mutations, sudden large changes in gene sequence, were far more likely to produce an unworkable mutant, and so he christened his one-in-a-million success a ‘hopeful monster’. For Goldschmidt, a hopeful monster was the lucky outcome of a large and sudden genetic change, rather than a succession of tiny mutations—the kind of thing an archetypal mad scientist might produce in the laboratory after a lifetime of dedicated and deranging failure. With our modern understanding of genetics, we now know that macro-mutations don’t account for speciation, at least not in multicellular creatures (though they may in bacteria, as argued by Lynn Margulis). However, it seems to me that the fusion of two whole genomes to create the first eukaryotic cell is better viewed as a macro-mutation to create a ‘hopeful monster’ than purely as a succession of small genetic changes.
So what kind of a monster was the first eukaryote, and why was its origin so improbable? To understand the answers, we need to think first about the nature of eukaryotic cells, and the many striking ways in which they differ from bacteria. We have already touched on this in the Introduction; here, we need to focus on the magnitude of the differences, the wide yawn of the chasm.
Differences between bacteria and eukaryotes
Compared with bacteria, most eukaryotic cells are enormous. Bacteria are rarely larger than a few thousandths of a millimetre (a few microns) or so in length. In contrast, although some eukaryotes, known as the pico-eukaryotes, are of bacterial size, the majority are ten to a hundred times those dimensions, giving them a cell volume about 10 000 to 100 000 times that of bacteria.
Size is not the only thing that matters. The cardinal feature of eukaryotes, from which their Greek name derives, is the possession of a ‘true’ nucleus. This nucleus is typically a spherical, dense mass of DNA (the genetic matter) wrapped up in proteins and enveloped in a double membrane. Here, already, are three big differences with bacteria. First, the bacteria lack a nucleus at all, or else have a primitive version that is not enclosed by a membrane. For this reason bacteria are also termed ‘prokaryotes’, from the Greek ‘before the nucleus’. While this is potentially a prejudgement—some researchers argue that cells with a nucleus are just as ancient as those without—most specialists agree that prokaryotes are well named: they really did evolve before cells with a nucleus (the eukaryotes).
The second big difference between bacteria and eukaryotes is the size of their genomes as a whole—the total number of genes. Bacteria generally have far less DNA than even simple single-celled eukaryotes such as yeast. This difference can be measured either in terms of the total number of genes—usually adding up to hundreds or thousands—or the total DNA content. This latter value is known as the C-value, and is measured in ‘letters’ of DNA. It includes not only the genes, but also the stretches of so-called non-coding DNA—DNA that does not code for proteins, and so can’t really be called ‘genes’. The differences in both the number of genes and the C-value are revealing. Single-celled eukaryotes like yeasts have several times as many genes as most bacteria, whereas humans have perhaps twenty times as many. The difference in the C-value, or total DNA content, is even more striking, as eukaryotes contain far more non-coding DNA than bacteria. The total DNA content of eukaryotes spans an extraordinary five orders of magnitude. The genome of a large amoeba, Amoeba dubia, is more than 200 000 times larger than that of the tiny eukaryotic cell, Encephalitozoon cuniculi. This enormous range is unrelated to complexity, or the total number of genes. Amoebae dubia actually has 200 times more DNA than do humans, even though it has far fewer genes and is obviously less complex. This odd discrepancy is known as the C-value paradox. Whether all this non-coding DNA has any evolutionary purpose is debated. Some of it certainly does, but a large part remains puzzling, and it is hard to see why an amoeba should need so much (we will return to this in Part 4). Nonetheless, it is a fact, requiring an explanation, that eukaryotes generally have orders of magnitude more DNA than prokaryotes. This is not without a cost. The energy required to copy all this extra DNA, and to ensure it is copied faithfully, affects the rate and circumstances of cell division, with implications that we will explore later.
The third big difference lies in the packing and organization of DNA. As we noted in the Introduction, most bacteria possess a single circular chromosome. This is anchored to the cell wall, but otherwise floats freely around the cell, ready for quick replication. Bacteria also carry genetic ‘loose change’ in the form of tiny rings of DNA called plasmids, which replicate independently and can be passed from one bacterium to another. The daily exchange of loose plasmids in this way is equivalent to shopping with loose change, and explains how the genes for drug resistance spread so quickly in a population of bacteria—just as a coin may find itself in twenty different pockets in a day. Returning to their main gene bank, few bacteria wrap their main chromosome in proteins—rather, their genes are ‘naked’, making them easily accessible—a current account rather than a savings account. Bacterial genes tend to be ordered in groups that serve a similar purpose, and act as a functional unit, which are known as operons. In contrast, eukaryotic genes give no semblance of order. Eukaryotic cells possess quite a number of disparate, straight chromosomes, which are usually doubled up to give pairs of equivalent chromosomes, such as the 23 pairs of chromosomes found in humans. In eukaryotes, the genes are strung along these chromosomes in virtually a random order, and to make matters worse they are often fragmented into short sections with long stretches of non-coding DNA breaking up the flow. To build a protein, a great tract of DNA often needs to be read off, before it is spliced up and melded together to form a coherent transcript that codes for the protein.
Eukaryotic genes are not just randomized and fragmented, they are also tricky to get at. The chromosomes are tightly wrapped in proteins called his-tones, which block access to the genes. When the genes are being replicated during cell division, or copied to make transcripts for building proteins, the configuration of the histones must be altered to allow access to the DNA itself. This in turn has to be controlled by proteins called transcription factors.
Altogether, the organization of the eukaryotic genome is a complicated business that fills library after library with footnotes. We’ll come to another aspect of this complicated set-up in Part 5 (sex, which is not found in bacteria). For now, though, the most important take-home point is that there is an energetic cost to all of this complexity. Where bacteria are almost always ruthlessly streamlined and efficient, most eukaryotes are lumbering and labyrinthine.
A skeleton and many closets
Outside the nucleus, eukaryotic cells are also very different to bacteria. Eukaryotic cells have been described as cells with ‘things inside’ (Figure 2). Most of the things inside are membrane structures, made of a vanishingly thin sandwich of fatty molecules called lipids. The membranes form into vesicles, tubes, cisterns, and stacks, enclosing spaces—closets—that are physically separated, by the lipid barrier, from the watery cytosol. Different membrane systems are specialized for various tasks, such as building the components of the cell, or breaking down food to generate energy, or for transport, storage, and degradation. Interestingly, for all their variety of size and shape, most of the eukaryotic closets are variations on the simple vesicle: some are elongated and flattened, others are tubular, and some are simply bubbles. The most unexpected is the nuclear membrane, which looks like a continuous double membrane enclosing the nucleus, but is in fact a series of large flattened vesicles welded together, and rath
er surprisingly, continuous with other membrane compartments in the rest of the cell. The nuclear membrane is therefore distinct in structure from the external membranes of any cell, which are always a continuous single (or double) layer.
2 Schematic illustrations of a bacterial cell (a) compared with a eukaryotic cell (b). The illustrations are not drawn to scale; bacteria are about the same size as the mitochondria in (b). The membrane structures are in fact sparsely depicted in the eukaryotic cell, for clarity, and in reality the differences in internal structure are even more marked. Bacteria are remarkable for their inscrutability, even by electron microscopy.
Then there are the tiny organs within cells, the so-called organelles, such as the mitochondria and the chloroplasts in plants and algae. The chloroplasts are worth a special mention. They are responsible for photosynthesis, the process by which solar energy is converted into the currency of biological molecules, which possess their own chemical energy. Like the mitochondria, the chloroplasts derive from bacteria, in this case the cyanobacteria, the only group of bacteria capable of true photosynthesis (to generate oxygen). It is notable that both mitochondria and chloroplasts were once free-living bacteria, and still retain a number of partially independent traits, including a contingent of their own genes. Both are involved in energy generation for their host cells. Both these organelles are tangibly different from the other membrane systems of eukaryotic cells, and these differences set them apart. Like the nucleus, the mitochondria and chloroplasts are enclosed in a double membrane, but unlike the nucleus their membranes form a true continuous barrier. Along with their own DNA, their own ribosomes and protein assembly, and their semi-autonomous manner of division, the double membranes of the mitochondria and chloroplasts are among the features that point an incriminating finger at their bacterial ancestry.
