Second, coherence. A telltale signature of planning is the coherent ordering of steps toward a goal. Random mutation, on the other hand, is incoherent; that is, any given evolutionary step taken by a population of organisms is unlikely to be connected to its predecessor.
I discuss evolutionary steps over the next three sections, and coherence in the subsequent three. In this chapter, we’ll continue to examine the molecular level of life. Later, we’ll extend the analysis to higher levels.
MONKEYS, TYPEWRITERS
A few years ago a curious fellow decided to test the old saw that, given typewriters and enough time, an army of monkeys would eventually produce the works of Shakespeare. A computer with keyboard was placed in a cage containing six macaques in a British zoo and left for four weeks. The result? “The macaques—Elmo, Gum, Heather, Holly, Mistletoe and Rowan—produced just five pages of text between them, primarily filled with the letter S. There were greater signs of creativity towards the end, with the letters A, J, L and M making fleeting appearances, but they wrote nothing even close to a word of human language.”3 The five pages have been published under the ironic title “Notes towards the Complete Works of Shakespeare.”
Because the names of amino acids that constitute the building blocks of proteins are abbreviated as single letters (L for leucine, S for serine, and so forth), as are the nucleotides that are the building blocks of DNA (A, C, G, and T) proteins can be likened to words and paragraphs, and the information contained in human DNA can be likened to an encyclopedia. Extending the analogy, evolution can be pictured as monkeys at a typewriter—not actually writing the words from scratch, but occasionally introducing a spelling change at random into a pre-existing text. If the change is a misspelling or ungrammatical, it is tossed out. However, if the spelling change leads to some new meaning, then (the analogy suggests) natural selection might preserve it. This analogy is crude and strained, but it has been popular among Darwinian popularizers. For example, in Darwin’s Dangerous Idea, the philosopher Daniel Dennett considers the first line of the classic novel Moby Dick: “Call me Ishmael.” A change or insertion of just one character can change the sense of the sentence, notes Dennett. For example, inserting a comma gives “Call me, Ishmael,” making it seem as if another person is addressing Ishmael, rather than he himself speaking. Switching another letter, writes Dennett, yields “Ball me Ishmael,” changing the meaning drastically.
For now let’s overlook the fact that neither of these changes, which redefine Ishmael as someone other than the narrator, fits easily with the point of view of the next sentence (“Some years ago—never mind how long precisely—having little or no money in my purse, and nothing particular to interest me on shore, I thought I would sail about a little and see the watery part of the world”) or the rest of the book. Let’s just concentrate on that first sentence. The critical point for our present purpose is that a change of one character—the addition of a comma, the switch of a C for a B—can alter the meaning of the sentence. As with written sentences, so too with biology. The change of a single amino acid or nucleotide “character” in protein or DNA can alter its “meaning”—its biological activity—as it does for sickle hemoglobin.
The eminent evolutionary biologist John Maynard Smith, who died in 2004, addressed this point over thirty years ago and reached an important conclusion.
The model of protein evolution I want to discuss is best understood by analogy with a popular word game. The object of the game is to pass from one word to another of the same length by changing one letter at a time, with the requirement that all the intermediate words are meaningful in the same language. Thus WORD can be converted into GENE in the minimum number of steps as follows:
WORD WORE GORE GONE GENE
Because mutations are relatively rare, the monkey’s typing is almost always judged after a single keystroke. Bad changes are quickly eliminated. So, reasoned Smith, evolution has to slog along one tiny, beneficial step at a time. If it needs two changes to help, it gets stuck. University of Rochester evolutionary biologist H. Allen Orr recently seconded John Maynard Smith’s reasoning:
Given realistically low mutation rates, double mutants will be so rare that adaptation is essentially constrained to surveying—and substituting—one-mutational step neighbors. Thus if a double-mutant sequence is favorable but all single amino acid mutants are deleterious, adaptation will generally not proceed.5
If two mutations have to occur before there is a net beneficial effect—if an intermediate state is harmful, or less fit than the starting state—then there is already a big evolutionary problem. For example, changing “Call me Ishmael” to “Call me Israel” might be beneficial in some context, but it would require several changes, and thus would appear to be beyond what Smith and Orr allow. Yet Smith and Orr actually overstate the case. Although multiple evolutionary changes are unlikely, as we saw in Chapter 5 they can occur, at least for the prolific malarial parasite. Several changes apparently did occur together very infrequently in the protein that conferred resistance to chloroquine. So the Smith-Orr criterion of two changes, while reasonable as a rule of thumb, is not a hard and fast law. But what allows exceptions to the rule of thumb? And what if instead of two mutations, three mutations were needed at once? Or more?
CLIMBING THE TOWER
As I mentioned earlier, I work in Iacocca Hall at Lehigh University’s branch location known as the “Mountaintop Campus.” One prominent feature of Iacocca Hall is an observation tower that stands about six stories high. At the top of the tower is a room with large windows all around. On a clear day you can see all the way east to New Jersey, west to Allentown, and north to the Poconos. It’s a great room for university dinners, and it’s rented out occasionally for wedding receptions. Elevators lead from the third floor to the tower room; there are no intermediate floors. If the elevators are out of service, the only access is through a narrow staircase.
Suppose you were standing outside the tower with a friend, and he asked how long it would take to reach the top. Jokingly, you might say that, from the outside, it’d take forever, because no one could walk up the sheer outside walls. Even with a running jump, the best athlete on earth couldn’t get up more than a small portion of the height. Ha, ha, the friend would respond, what a droll fellow you are. But he knows there are stairs inside, and wanted to know the time needed to climb the stairs.
For me, walking up the stairs to the tower might take ten minutes. For a younger person who’s in somewhat better shape, maybe one minute. In either case, it’s just a moment of time compared to the struggle faced by a jumper on the outside of the building. Without stairs (or a ladder or mountain-climbing equipment and so on) the tower room is effectively beyond reach. Only breaking the climb into many small steps with a staircase makes the task possible.
Suppose, though, that a person walking up the inside stairs encountered a missing step, so to get to the next step he had to climb twice the normal step-to-step distance. How long would it then take to climb the stairs? It would depend. If the climber were a frail old man, the missing step might be equivalent to a brick wall—virtually impassable. If the climber were just an out-of shape, middle-aged couch potato (like me), one missing step wouldn’t be too much of a problem. Swinging a leg up to the next step might induce some puffing and wheezing, but could be done, albeit slowly. If the climber were an athletic twenty-something, she would bound over the break. But suppose two steps were missing, or three. With three missing steps, the couch potatoes would likely be left behind with the frail old men, but the jocks could still go on. With more missing steps, even the athletes would have trouble. Some might need multiple tries before successfully making the jump, so their progress would be slower. If a whole flight of steps were missing between floors, then even the athletic twenty-somethings would be stymied. At some point, with enough steps missing, even the most athletic person on the planet couldn’t pass.
Of course the stairway to the tower room in Iacocca Hall is an
analogy for Darwinian evolution, one that presents the problem for evolution in a different way than monkeys and typewriters do. As with scaling physical heights, so too with ascending biological heights. If there are many closely spaced steps leading from one level of biology to another, then moving between them is trivial. If there are no such steps, the task is effectively impossible. Charles Darwin realized the distinction from the beginning. In The Origin of Species he emphasized that his new theory of evolution by natural selection had to explain changes in biological systems by “numerous, successive, slight modifications” of old ones. He insisted that, for his theory to be correct, evolution had to be a gradual, step-by-tiny-step process leading from one working arrangement to a new working arrangement through a series of intermediate states, all of which also worked, and all of which were just a small biological distance from their preceding and succeeding steps. Since Darwin lived before the discovery of the molecular basis of life, he didn’t realize, as John Maynard Smith and Allen Orr did, that those steps were actually tiny changes in molecules. Darwin knew that if there were steps between biological levels, his idea would work. He also knew that if the stairs were missing, “my theory would absolutely break down.”
Random mutation is the perfect tool for the evolutionary job when steps are continuous and close together. When there are some broken stairs, with small gaps between steps, it’s a potential tool. The seriousness of the breach in the steps depends on the health of the climber. In evolutionary terms, roughly, the larger the population of a species, the “healthier” it is. Species with tiny populations are the frail old men of biology, foiled by a missing step. Ones with abundant population are the athletes, able to leap multiple missing steps. Yet, as with human athletes and missing stairs, there comes a point where even the most abundant population on earth cannot jump an evolutionary barrier. Random mutation is almost certainly useless, even for the largest populations, when a flight of stairs is missing between biological floors.
GOING STRAIGHT
The number of missing steps between one biological structure and another is the first major criterion I’ll use for drawing a line marking the edge of Darwinian evolution. Recall the example of sickle cell disease. The sickle cell mutation is both a life saver and a life destroyer. It fends off malaria, but can lead to sickle cell disease. However, as we saw in Chapter 2, hemoglobin C-Harlem has all the benefits of sickle, but none of its fatal drawbacks. So in western and central Africa, a population of humans that had normal hemoglobin would be worst off, a population that had half normal and half sickle would be better off, and a population that had half normal and half C-Harlem would be best of all. But if that’s the case, why bother with sickle hemoglobin? Why shouldn’t evolution just go from the worst to the best case directly? Why not just produce the C-Harlem mutation straightaway and avoid all the misery of sickle?
The problem with going straight from normal hemoglobin to hemoglobin C-Harlem is that, rather than walking smoothly up the stairs, evolution would have to jump a step. C-Harlem differs from normal hemoglobin by two amino acids. In order to go straight from regular hemoglobin to C-Harlem, the right mutations would have to show up simultaneously in positions 6 and 73 of the beta chain of hemoglobin. Why is that so hard? Switching those two amino acids at the same time would be very difficult for the same reason that developing resistance to a cocktail of drugs is difficult for malaria—the odds against getting two needed steps at once are the multiple of the odds for each step happening on its own.
What are those odds? Very low. The human genome is composed of over three billion nucleotides. Yet only a hundred million nucleotides seem to be critical, coding for proteins or necessary control features. The mutation rate in humans (and many other species) is around this same number; that is, approximately one in a hundred million nucleotides is changed in a baby compared to its parents (in other words, a total of about thirty changes per generation in the baby’s three-billion-nucleotide genome, one of which might be in coding or control regions).6 In order to get the sickle mutation, we can’t change just any nucleotide in human DNA; the change has to occur at exactly the right spot. So the probability that one of those mutations will be in the right place is one out of a hundred million. Put another way, only one out of every hundred million babies is born with a new mutation that gives it sickle hemoglobin. Over a hundred generations in a population of a million people, we would expect the mutation to occur once by chance. That’s within the range of what can be done by mutation/selection.
To get hemoglobin C-Harlem, in addition to the sickle mutation we have to get the other mutation in the beta chain, the one at position 73. The odds of getting the second mutation in exactly the right spot are again about one in a hundred million. So the odds of getting both mutations right, to give hemoglobin C-Harlem in one generation in an individual whose parents have normal hemoglobin, are about a hundred million times a hundred million (1016). On average, then, nature needs about that many babies in order to find just one that has the right double mutation. With a generation time of ten years and an average population size of a million people, on average it should take about a hundred billion years for that particular mutation to arise—more than the age of the universe.
(Some readers might be wondering if this is a fair analysis of double mutations in general, since I’m focusing just on the set that gives C-Harlem, yet other sets of mutations might arise in nature that might be as helpful as C-Harlem. It turns out that consideration won’t affect matters much. First, as we saw with the response of malaria to chloroquine, there may be very few useful evolutionary responses possible, even taking two steps at a time. Out of a hundred billion billion parasites, only one effective response was produced, a change in PfCRT. Such limited ability to respond is also seen in resistance to warfarin by rats and the similar responses of flies and mosquitoes to insecticides, and appears to be typical. Second, and more important, the odds against obtaining a cluster of mutations increases exponentially the more sites that have to be matched, but decreases only linearly with the number of combinations that are helpful. Even if there were a hundred possible double mutations that would help, that would decrease the average waiting time in the example above only linearly, just by a factor of a hundred, from a hundred billion years to a billion years. The general point would remain, that the need to mutate two or more sites together to get an effective evolutionary response immediately makes the problem much more difficult than having to match just one.)
Of course, C-Harlem did arise, relatively recently, in New York City. It happened exactly the way Darwin envisioned, by “numerous [well, two, anyway], successive, slight modifications,” each of which in turn was beneficial. After the sickle mutation first appeared, it began to increase in the population because of its beneficial effects, until millions of Africans had a copy of the gene. Now, instead of two mutations having to appear simultaneously in the DNA of one unbelievably lucky child, just one more mutation would have to happen in the offspring of any one of the millions of people who were already one step toward the goal. Because there were many more people playing the sickle Powerball lottery, all of whom had to match only one number instead of two, the jackpot went off relatively quickly.
Hemoglobin C-Harlem would be advantageous if it were widespread in Africa, but it isn’t. It was discovered in a single family in the United States, where it doesn’t offer any protection against malaria for the simple reason that malaria has been eradicated in North America. Natural selection, therefore, may not select the mutation, and it may easily disappear by happenstance if the members of the family don’t have children, or if the family’s children don’t inherit a copy of the C-Harlem gene. It’s well known to evolutionary biologists that the majority even of helpful mutations are lost by chance before they get an opportunity to spread in the population.7 If that happens with C-Harlem, we may have to wait for another hundred million carriers of the sickle gene to be born before another new C-Harlem mutation arises.
>
Suppose, however, that the first mutation wasn’t a net plus; it was harmful. Only when both mutations occurred together was it beneficial. Then on average a person born with the mutation would leave fewer offspring than otherwise. The mutation would not increase in the population, and evolution would have to skip a step for it to take hold, because nature would need both necessary mutations at once. For frail old men, the missing step would be a prohibitive barrier. The Darwinian magic works well only when intermediate steps are each better (“more fit”) than preceding steps, so that the mutant gene increases in number in the population as natural selection favors the offspring of people who have it. Yet its usefulness quickly declines when intermediate steps are worse than earlier steps, and it is pretty much worthless if several required intervening steps aren’t improvements.
Smith and Orr’s prohibition of double mutations may be wrong for malaria, but it is right for species like humans and other large animals. Our measly population size of millions to billions puts us in the “frail old man” class. Plasmodium falciparum, however, with a yearly population size on the order of a hundred billion billion (1020) or so is in the “couch potato” class, and can jump a missing stair or two.
COHERENCE
The second basic criterion for distinguishing between random and nonrandom mutation is coherence. Darwinian evolution cannot pursue a future goal. So envisioning Darwinian evolution as akin to climbing a solitary staircase—even one with missing steps—risks a subtle, yet fatal misconception. It is all too easy to think of the top of the stairs as the target, and to focus exclusively on the path leading to it, ignoring all other possibilities. If a Darwinist visualizes steps leading to some biological feature, the temptation is to conclude the route would be easily traveled by unaided nature. However, as Coyne and Orr emphasized, we need to ask whether a process is not just theoretically possible, but also biologically reasonable. Because random mutation and natural selection have no goal, Darwinian evolution faces the huge problem of incoherence: Like a drunkard’s walk, the next evolutionary step a population of organisms takes is very likely to be unconnected to the last step. The upshot is that even if a gradual route toward a complex structure exists—even one with no missing steps—if the route is lengthy enough, the likelihood of reaching it by random mutation is terrible.
The Edge of Evolution Page 12