Science, E. coli, and the Edge of Evolution

Dear Readers,

As I wrote in The Edge of Evolution, Darwinism is a multifaceted theory, and to properly evaluate the theory one has to be very careful not to confuse its different aspects. Unfortunately, stories in the news and on the internet regularly confuse the facets of Darwinism, ignore distinctions made in The Edge of Evolution, or misstate the arguments of intelligent design. The disregard for critical distinctions blurs the issues badly. Over the next few days I will briefly respond to four separate stories

1) A few months ago an interesting paper in Science, “Adaptive mutations in bacteria: high rate and small effects”, by the group of Isabel Gordo demonstrated that beneficial mutations in E. coli were more frequent than had been thought. In fact, the authors remark that “We found a rate on the order of 10(-5) per genome per generation, which is 1000 times as high as previous estimates, and a mean selective advantage of 1%.” They show that the previous underestimates of the beneficial mutation rates were likely due to clonal interference — accumulation of beneficial mutations in large bacterial populations which then interfere with each other to dominate the population, making beneficial mutations seem less frequent. Does this new result mean that Darwinian evolution can construct molecular machinery much easier than thought?

No. While the result is interesting, readers of The Edge of Evolution will not be very surprised by it. As I showed for mutations that help in the human fight against malaria, many beneficial mutations actually are the result of breaking or degrading a gene. Since there are so many ways to break or degrade a gene, those sorts of beneficial mutations can happen relatively quickly. For example, there are hundreds of different mutations that degrade an enzyme abbreviated G6PD, which actually confers some resistance to malaria. Those certainly are beneficial in the circumstances. The big problem for evolution, however, is not to degrade genes (Darwinian random mutations can do that very well!) but to make the coherent, constructive changes needed to build new systems. The bottom line is that the beneficial mutations reported in the new Sciencepaper most likely are degradatory mutations, and so don’t address the challenges outlined in The Edge of Evolution.

This is the second in a series of responses I’m posting this week, this one regarding the Darwinian website The Panda’s Thumb, where a woman named Abbie Smith questioned whether results from HIV research actually square with the claims I made that little fundamental change has occurred in the virus, even though it attains enormous populations sizes and has a much increased mutation rate. 

Although she calls herself a “pre-grad student,” the tone of the post is decidedly junior high, the tone of someone who is trying hard to compete with all the other Mean Girls on that unpleasant website. I’ll pass over all that and try to stick to the substance.

Her post mainly concerns a small protein coded for by HIV-1 called Vpu. She first points out that the amino acid identity between the homologous chimp SIV protein with HIV Vpu is 39%, much less than that of other homologous viral proteins, and she seems to regard that fact by itself as remarkable. Yet the alpha and beta chains of human hemoglobin are only about 44% identical, and have virtually superimposable structures and very similar functions. The fact that the chimp and human versions of VPU have 39% identity indicates they are structurally virtually identical. That doesn’t seem like a fundamental change to me.

She goes on to write that Vpu acts to degrade CD4 molecules by binding to them and recruiting the pathway that degrades CD4. Unfortunately, she seems not to have read the beginning of chapter 8 of The Edge of Evolution (“Objections to the Edge”), where I make some careful distinctions:

This chapter makes some important distinctions and addresses potential objections. It considers counter-arguments to my attempt to define the edge of evolution — not philosophical ones, about the “other side” of that boundary, but technical and logical ones about the line itself….

Another, more important point to note is that I’m considering just cellular proteins binding to other cellular proteins, not to foreign proteins. Foreign proteins injected into a cell by an invading virus or bacterium make up a different category. [emphasis added here] The foreign proteins of pathogens almost always are intended to cripple a cell in any way possible. Since there are so many more ways to break a machine than to improve it, this is the kind of task at which Darwinism excels. Like throwing a wad of chewing gum into a finely tuned machine, it’s relatively easy to clog a system — much easier than making the system in the first place. Destructive protein-protein binding is much easier to achieve by chance.

So the example she chose is from exactly the category that I excluded in the above paragraph. My exclusion isn’t arbitrary. As I wrote, there are many more ways to cripple a machine than to build one, so destructive Darwinian processes can appear to accomplish more. Yet The Edge of Evolution is concerned with how molecular machinery is constructed, not destroyed. One can’t ignore such critical distinctions and make progress. But, in my experience, many Darwinists overlook important differences.

She goes on to list several other properties of Vpu, but, while interesting, none at all are what one should call “fundamental” changes. For example, she notes, the HIV Vpu has several sites that are negatively charged by virtue of being phosphorylated. She continues, “Yet some SIVcpz Vpus have only one [phosphorylation] site, and instead utilize a simple string of negatively charged amino acids in place of the second site. Different ways of performing similar tricks with totally different amino acids. I think that’s biochemically significant as well.”

Well, I disagree. I don’t think that’s biochemically fundamental at all. In each case one has a blob of negative charge. Since the mutation rate of HIV is so extremely high, kinase sites are likely replaced every day in some virus with multiple glutamates or aspartates and vice versa. She also points out that some virus Vpu subtypes have altered the location of modification sites or acquired signals to localize protein in particular subcellular compartments. But again, because of the virus’s extremely high mutation rate, such sites would be expected to come and go frequently. As I emphasized in The Edge of Evolution, HIV’s enormous numbers and very high mutation rate cause immense variation. The question, however, is to what extent the immense variation has produced novel virus systems or machinery? And, as I indicated, the answer is very little. Butler at al (HIV Genetic Diversity: Biological and Public Health Consequences, Current HIV Research, 2007, 23-45) remark under the subheading “Biological Consquences of HIV Diversity”:

With such breadth of genetic diversity among HIVs, one might expect significant biological differences between the clades. Although interesting variations can be seen, much of the data concerning biological implications of HIV diversity is contradictory.

Plenty of differences do exist, and some are “interesting”, but not all that great.

Darwinists overlook the considerable power of the example of the relatively minor changes in HIV: there have been a truly astronomical number of copies produced in just the past fifty years or so. And because of its much increased mutation rate, it has undergone in the past half century as many of some kinds of mutations as all the cells have undergone in the history of the world. If Darwinism had the power that its boosters claim, we should expect to see truly fundamental changes. Yet despite the enormous number of opportunities, only minor changes have appeared. That is very strong evidence of the strict limits on what Darwinian processes can accomplish.

This is the third in a series of responses I’m posting this week.

In “Evolution of the Bacterial Flagellum” (Microbe Magazine, July 2007), Wong et al seek to counter arguments of intelligent design proponents such as myself that the flagellum did not evolve by random mutation and natural selection. Unfortunately, their otherwise-fine review misunderstands design reasoning and so fails to engage that issue. The critical passage from Wong et al is the first paragraph:

Proponents of the intelligent design (ID) explanation for how organisms developed claim that the bacterial flagellum (BF) is irreducibly complex. They argue that this structure is so complicated that it could not have emerged through random selection but had to be designed by an intelligent entity. One part of this claim is that each flagellar component is used solely for the purpose of making a flagellum that, in turn, is used only for motility. Further, each flagellar protein is assumed to have appeared independently of the other component proteins.

Although the first two sentences are correct, the last two sentences are quite wrong. (The authors cite no references for these latter claims.) It is no part of the design argument that each component of an irreducibly complex structure must be used solely for that purpose, nor that each part must arise independently. In my 1996 book Darwin’s Black Box, which brought the concept of irreducible complexity to wide public attention, I pointed out the fact that, for example, proteins of the blood clotting cascade share sequence homology with each other and with other serine proteases, and the fact that ciliary proteins such as tubulin are involved in other tasks in the cell. Yet I explained that neither sequence homology nor multiple functions showed how integrated systems containing many parts could be put together by small random steps. Unfortunately, Wong et al spend their efforts addressing their own erroneous assertions. They fail to address the only pertinent question, the question of whether random, unintelligent processes — even when filtered by natural selection — could plausibly build a structure such as the flagellum.

To address the adequacy of random processes plus selection would require rigorous experiments or calculations showing that the intricate, functional structures are not too improbable given the evolutionary resources available. Recent work bears negatively on this difficult question. In long term laboratory evolution experiments over tens of thousands of generations (Lenski, R.E. 2004. Phenotypic and genomic evolution during a 20,000-generation experiment with the bacterium Escherichia coli. Plant Breeding Reviews 24:225-265), cultures of E. coli were repeatedly seen to lose the ability to make ribose and maltose, and to repair their DNA. Some mutations shut down expression of their flagellar genes, apparently to conserve energy. No selected mutations were observed which could plausibly be argued to be the incipient stages of some new, complex functional system. Similar kinds of results are seen in other well-studied evolutionary systems. For example, in response to strong pressure from the malarial parasite, the human genome has suffered a handful of positively-selected-yet-degradative mutations (Carter, R. and Mendis, K.N. 2002. Evolutionary and historical aspects of the burden of malaria. Clin. Microbiol. Rev. 15:564-594), including ones that render nonfunctional the genes for glucose-6-phosphate dehydrogenase, the alpha and beta chains of hemoglobin, band 3 protein, and others. Again, no selected mutations were observed which could plausibly be argued to be the incipient stages of some new, complex functional system.

To a skeptic such as myself, this does not look like the sort of process which could build complex molecular machinery.