Chapter 5: The Engine of Evolution
Chapter 5 is the only part of Coyne’s book that attempts to demonstrate the causal power of the proposed mechanisms driving evolutionary change, the key point of contention in the origins debate. Coyne concedes that “everywhere we look in nature, we see animals that seem beautifully designed to fit their environment…It is no surprise that early naturalists believed that animals were the product of celestial design, created by God to do their jobs.”
According to Coyne,
Darwin dispelled this notion in The Origin. In a single chapter, he completely replaced centuries of certainty about divine design with the notion of a mindless, mechanistic process — natural selection — that could accomplish the same result. It is hard to overestimate the effect that this insight had not only on biology, but on people’s worldview. Many have not yet recovered from the shock, and the idea of natural selection still arouses fierce and irrational opposition. (p. 115)
Natural selection’s key selling point is that it can act as a designer substitute. But what evidence does Coyne offer us? He spends much of the chapter explaining the concepts of random mutation, natural selection and genetic drift, and demonstrating that each of these processes really operates in nature (which no credible biologist denies). With regard to random genetic drift, Coyne explains that “genetic drift is not only powerless to create adaptations, but can actually overpower natural selection.” This point was brought out recently in a paper by Michael Lynch in PNAS (Lynch, 2012). Lynch argues that “random genetic drift can impose a strong barrier to the advancement of molecular refinements by adaptive processes.”
Evolution in the Test Tube
After introducing readers to the evolutionary processes in operation and giving examples of natural selection in the wild and artificial selection, Coyne turns his attention to instances of evolution occurring in the laboratory. His first exhibit is Richard Lenski’s long-term experiments with E. coli at the University of Michigan. Coyne doesn’t cover Lenski’s work in great depth. Biochemist Michael Behe has offered detailed reviews of Lenski’s work in his book The Edge of Evolution, his Uncommon Descent blog, here at ENV, and in a 2010 peer-reviewed paper published in the Quarterly Review of Biology (Behe, 2010).
Coyne also mentions the work of Barry Hall and his colleagues at the University of Rochester (Hall, 1998; Hall, 1997; Hall, 1982; Hall, 1978). For a detailed discussion of Barry Hall’s work and its significance, I refer readers to Michael Behe’s article here.
Resistance to Drugs and Poisons
Coyne mentions the resistance of Staphylococcus aureus to penicillin and methicillin. The mechanisms of such resistance, however, are less than impressive. Staphylococcus aureus acquires methicillin resistance by means of the mecA gene (Niemeyer et al., 1996), which sits on the mec operon. The mecA gene codes for a modified penicillin-binding protein (PBP2′), of high molecular weight, that possesses a decreased binding-affinity for ?-lactams such as penicillins. In the presence of methicillin, PBP2′ takes over the functions of methicillin-sensitive PBPs and the bacterial peptidoglycan cell wall is structurally altered such that it has fewer oligomeric peptides. Since the mecA gene is located at the same chromosomal locus in all isolates of MRSA, it is probable that this adaptation occurred only a single time.
Finally, Coyne cites DDT resistance in insects as evidence of what the evolutionary mechanisms can accomplish. But the evolution of resistance to DDT is actually fairly trivial. Indeed, the most common way this has happened is by the simple modification of the relevant sodium channel by alteration of its amino acid sequence such that it interferes with the action of the insecticide. Actually, so far as I am aware, in every single known case this is accomplished by replacing the amino acid leucine (position 1014) with phenylalanine (Williamson et al., 1996). Granted, there are some insects that occasionally enhance this resistance by means of an additional mutation in the same protein. But it does not appear to occur independently of the first. It thus seems that the very same alterations have occurred independently in many different cases. This is to be expected, however. If we assume that the probability of mutating a particular nucleotide is something like 10^-9, there are easily adequate probabilistic resources to account for this — there is a very high probability that most of the possible single-nucleotide substitutions will occur in a large enough insect population. The more coordinated/non-adaptive mutations you require to facilitate a novel innovation in function, however, the less likely it is that mutation will stumble upon such an adaptation.
A second point worthy of note is that the resistant mutants are rendered less fit overall than the “wild type.” That is to say, the mutation incurs a fitness cost. The mutant insects, though resistant to DDT, are also more prone to paralysis at mildly raised temperatures.
Coyne fails to address any of the challenges to his view that evolutionary mechanisms are able to accomplish what he requires of them, such as the rarity and isolation of stable protein folds in sequence space (Axe, 2010a; Axe, 2004; Gauger and Axe, 2011), or the mathematical limits of complex adaptation (Axe, 2010b), or the challenge of irreducibly complex systems like the bacterial flagellum.
I will continue my review of Coyne’s book in a post tomorrow.