Although much of this book will be concerned with retelling the minutiae of biological detail in support of the general thesis of the ubiquity of evolutionary convergence and, what is more important, its implications, here is a brief overview of what strike me as the basic tensions in evolution. The first is what, for want of a better name, I might term 'inherency'. A hard-boiled reductionist will dismiss this as a non-problem, but I am not so sure. Perhaps the first obvious clue was the result, surprising at the time, of the minimal genetic difference between ourselves and the chimps. In terms of structural genes the much-quoted difference amounts, it is said, to about 0.4%. If there were any residual doubt of the closeness between Homo and Pan, then other indicators of similarity, such as the fact that the string of amino acids that make up the protein haemoglobin is identical in number and sequence, are surely a sufficient indicator of our evolutionary proximity. This, of course, confirms the obvious: we and the chimps share an ancestor, probably between about 6 and 12 million years ago, and indeed there is much we have in common. But in other respects we are poles apart. I'm told that chimps driving cars (or at least go-karts) have the time of their lives, but we are neither likely to see a chimp designing a car, nor for that matter mixing the driest of Martinis, let alone being haunted by existentialist doubts.
This problem of inherency, however, is far more prevalent and pervasive than the local quirk that chimps and humans are genom-ically almost identical, but otherwise separated by an immense gulf of differences. Let us look, for example, at a much deeper stage in our evolution, effectively at the time of the ancestors of the fish. Enter the moderately undistinguished animal known as the lancelet worm or amphioxus (Branchiostoma and its relatives, Fig. 1.2). By general agreement this beast is the nearest living approximation to the stage in evolution that preceded the fish, which in turn clambered on to land, moved to using the egg, grew fur, and in one lineage developed into socially alert arborealists. All these changes and shifts must have been accompanied by genetic changes, but if we look back to am-phioxus we see a genetic architecture in place that seemingly has no obvious counterpart in its anatomy. To give just one example: the central nervous system of amphioxus is really rather simple. It consists of an elongate nerve cord stretching back along the body, above the precursor of the vertebral column (our backbone, consisting of a row of vertebrae) and a so-called brain. The brain can only be described as a disappointment. It is little more than an anterior swelling (it is called the cerebral vesicle) and has no obvious sign in terms of its morphology of even the beginnings of the characteristic threefold division seen in the vertebrate brain of hind-, mid- and fore-sections. Yet the molecular evidence,6 which is also backed up by some exquisitely fine studies of microanatomy,7 suggests that, cryptically, the brain of amphioxus has regions equivalent to the tripartite division seen in the vertebrates.
The clear implication of this is that folded within the seemingly simple brain of amphioxus is what can almost be described as a template for the equivalent organ of the vertebrates: in some sense amphioxus carries the inherent potential for intelligence. Quite how the more complex brain emerges is yet to be established. The evidence that a key development in the molecular architecture of the vertebrates was episodes of gene duplication,8 that is, doubling up of a gene, could well give one clue. This is because the 'surplus' gene is then potentially available for some new function. It could alternatively be claimed that amphioxus is secondarily simplified (the condition sometimes referred to as regressive), but it retained genes for
figure 1.2 The amphioxus animal. Upper, entire animal. The anterior end is to the right, with the 'brain' located towards upper side. Prominent white units are gonads. Lower, detail of anterior with prominent feeding (buccal) tentacles and more posteriorly gill bars. The notochord is the longitudinal structure slightly above the mid-line, with closely spaced vertical lines. The nerve cord lies above the notochord, with minimal enlargement at the anterior. (Courtesy of Dale Stokes, Scripps Institution of Oceanography (upper) and Thurston Lacalli, University of Victoria, British Columbia (lower).)
vital functions, although ones no longer specifically connected to the coding for a complex brain. Unfortunately the rather limited information on the earliest amphioxus-like animals, from the Cambrian period (c. 545-500 Ma (million years) ago)9 does not extend to seeing their brains. In general, however, the genomic evidence suggests that the living amphioxus is not in some sense degenerate but is genuinely primitive.
Revealing the foundations of the molecular architecture that underpins our brains and sentience gives us not only a feeling of emergence, but underlines how little we really know about why and how organic complexity arises. Nor is this example of the amphioxus brain and its molecular inherency in any way unusual. Equally instructive examples can be culled from the most primitive animals, such as the sponges and Hydra (the latter is a relative of the sea-anemones and corals), in which genes (or proteins) that are essential for complex activities in more advanced animals are already present. Doubtless they have their functions, but what these are and how they have been redeployed, co-opted, or realigned in more advanced animals is for the most part still unknown. The unravelling of these evolutionary stories is going to be one of the most fascinating episodes in recent biological history, but what will almost certainly be more extraordinary is how much of organic complexity will be seen to be latent in more primitive organisms. Or perhaps not that extraordinary: it is sometimes forgotten that the main principle of evolution, beyond selection and adaptation, is not the drawing of new plans but relying on the tried and trusted building blocks of organic architecture.
the navigation of protein hyperspace Life, then, is full of inherencies. We might legitimately enquire whether there is anything in the human condition that could prefigure some future level of complexity that with the virtue of hindsight will, no doubt, seem to have been inevitable, but to us remains unimaginable. Yet whatever privileges exist for untrammelled speculation, there is a story to be told which will occupy the rest of this book. My critics will, I imagine, complain at its eclectic, if not unorthodox, nature; and given that the topics covered will include such matters as extraterrestrial amino acids and ants pursuing warfare it is advisable to try to explain the underlying thread of the argument. Here we can do no better than to look at a stimulating and thoughtful essay written by Temple Smith and Harold Morowitz,10 which is an exploration of the tension between the predictabilities of physical systems and the seemingly contingent muddle that we call history. In brief, and their paper contains many other insights, they remind us of the simplicity of the basic building blocks of life, by which is meant such molecules as amino acids (which go to build the proteins, perhaps familiar as collagen or haemoglobin), or sugars (which when joined together can form carbohydrates). In the case of amino acids, however, even with the rather modest total of the 20 available variants and taking a relatively simple protein - consisting, say, of 100 such building blocks - it is immediately apparent that the potential number of combinations in which this protein could be assembled is absolutely gigantic. Specifically it is 20100, which is equivalent to 10130.
This is an uncomfortably large number,11 and, as Smith and Morowitz see it, this immensity of possibilities confers an inherent unpredictability on the process of evolution. Taking the figures given above, that is the 20 available amino acids and their random inclusion in a protein composed of a total of 100 amino acids, Smith and Morowitz then apply some apparently stringent criteria to the selection of those proteins that will actually work. The specific function they have in mind for proteins is as natural catalysts, that is, to function as the organic molecules known as enzymes,12 which serve to accelerate metabolic processes. The alternative, of course, is that a hypothetical protein will be non-functional, failing in one way or another. We know that in principle this is perfectly feasible, because there are many examples known where only a handful of changes, and sometimes even the substitution of a single amino acid for another one, will render the entire protein inoperative and thereby biologically useless. Let us then suppose that only one in a million proteins will be soluble, a necessary prerequisite for the watery milieu of a cell. Let us further suppose, and again the figure seems reasonable, that of these again only one in a million has a configuration suitable for it to be chemically active. How many potentially enzymatically active soluble proteins with an amino acid length totalling 100 could we expect to be available to life? A few thousand, perhaps even a few million? In fact, the total far exceeds the number of stars in the universe.
As Smith and Morowitz dryly note, 'It is quite clear from such numerology that the domain of possible organisms is enormously large if not infinite',13 especially when we recall that many proteins are substantially longer than 100 amino acids. The only way we can begin to envisage such a protein domain is in the abstract terms of a kind of hyperspace. Mathematically this will encompass all the measurements that together serve to define the totality of this 'protein space'. As Smith and Morowitz point out, with such an immense number of potential possibilities the number of proteins known to exist on Earth can only be an infinitesmally small fraction of this vast total. As they say, notwithstanding 'the immensity of the dimensionality of the descriptive hyperspace', the world we know and the evolutionary processes that define it have 'produced a very sparsely sampled hyperspace in the actual living world'.14
One inference that might be immediately drawn from this is that in principle the likelihood of any other world employing an area of 'protein space' that is even remotely close to that found on Earth should be vanishingly small. At this early stage of the argument we can leave aside, for the moment, the distinct likelihood that Earth-like planets are going to be in exceedingly short supply (Chapter 5), and simply remind ourselves that even as our net of exploration spreads first across the Milky Way and then from galaxy to galaxy, so each time a protein chemist steps on to the surface of a new planet only another tiny fraction of this immense 'hyperspace' will be documented. The combinatorial possibilities are so much more immense than all the planets with all their biospheres that most proteins will for ever be only hypothetical constructs. That, at least, is the expectation and it would seem difficult to refute. All other worlds might be expected to be truly alien, at least in so far as the occupation of protein 'space' is concerned. That is, at least, the assumption.
Was this article helpful?