DOMAIN SHUFFLING IN THE REAL WORLD
On which side of the edge of evolution would domain shuffling be expected to fall in nature, rather than in the lab? Is it biologically reasonable to think that random mutation and natural selection could build new, coherent genetic circuits from old protein-binding sites? One big difficulty in coming to a firm conclusion on that question is that, unlike the situation with respect to shape space and protein-protein binding sites, there have been no good experiments that show what fraction of mutations would work—nothing like the experiments of Greg Winter’s lab and others that showed that between ten and a hundred million binding sites have to be searched in a shape space library to find one that will bind with a modest affinity to a second protein. So any conclusion we reach will be less quantitative and more tentative than for the development of brand new protein-protein binding sites.
Nonetheless, there is information available that can help us make an informed judgment. First, in all of the experiments, Lim’s lab didn’t just splice two genes together in a single step; they took several additional steps as well. For example, in the case of the hybrid mating factor/concentrated salt scaffold protein they added further mutations to knock out the original pathway, to ensure there was no cross-reaction where, say, one signal would activate both the mating response and the high salt response. Remember, the more steps that have to occur between beneficial states, the much less plausible are Darwinian explanations.
Second, in joining together various protein-binding domains to control actin assembly, Lim’s group found quite complex results:
Switches could be divided into diverse behavioral classes. At the extremes, five switches showed little or no basal repression, and nine were extremely well-repressed, but could not be activated under any of the tested conditions. Most constructs, however, showed some type of gating behavior…. Heterologous switch behavior was also dependent on affinity of the autoinhibitory interactions…. Linker length also affected switch behavior…. [I]ncreasing interdomain linker length did not uniformly reduce coupling, which suggests that these effects are context-dependent…. The combinatorial switch library also yielded switches with the unexpected behavior of antagonistic or negative input control…. This unanticipated class of switches highlights a striking feature of the library: Subtle changes in switch parameters can lead to dramatic changes in gating behavior.9
In other words, the system behavior is chaotic and incoherent, depending on many conflicting factors. Which of the various possibilities would be harmful to an organism? Which of the very few that might be helpful for the moment would be evolutionary dead ends, single steps to local peaks in a rugged evolutionary landscape? In the mating/salt tolerance experiment, the poor mutated yeast was sterile, unable to mate, and could only resist high salt concentrations if supplied with mating factor. To say the least, such a response would be unlikely to help in nature.
The third and most important factor in judging how helpful domain shuffling is likely to be is that P. falciparum seems to have made no use of it. In a hundred billion billion chances, when the malarial parasite was in a life-or-death struggle with chloroquine, domain shuffling was nowhere to be seen. Writes Lim: “By allowing the establishment of novel regulatory connections between molecules with no previous physiological relation, such recombination events would be a powerful force driving evolution of novel cellular circuitry.” Yet the fancied “powerful force” wasn’t as helpful as a few, simple, run-of-the-Darwinian-mill point mutations in PfCRT.
Domain shuffling would be an instance of the “natural genetic engineering” championed by James Shapiro, where evolution by big random changes is hoped to do what evolution by small random mutations can’t. But random is random. No matter if a monkey is rearranging single letters or whole chapters, incoherence plagues every step. Although we have a less secure quantitative base for deciding, and new data might bear on the question one way or another, it’s likely that domain rearrangement is similar to everything else that random mutation does. One step might luckily be helpful on occasion, maybe rarely a second step might build on it. But Darwinian processes in particular and unintelligent ones in general don’t build coherent systems. So it is biologically most reasonable to conclude that, like multiple brand-new protein-protein binding sites, the arrangement of multiple genetic elements into sophisticated logic circuits similar to those of computers is also well beyond the edge of Darwinian evolution.
COMPUTER ASSUMPTIONS
What about computers themselves, though? If some aspect of biology can be mimicked accurately on a computer, wouldn’t that allow us to probe the edge of evolution in greater detail? In principle, it would. The problem is that living things are so complex that all descriptions of them, whether in computers or books, require the kind of drastic simplification that can lead to serious error if we’re not careful. A prominent example is Avida, an “artificial life” computer program that, according to its inventors, explains “how complex functions can originate by random mutation and natural selection.”10
In Avida, an “organism” is a sequence of computer instructions coupled with a processor that executes these instructions in sequence. Just as we burn calories with every activity we engage in, these artificial organisms burn computational “energy” with each instruction executed. They, like us, have to feed themselves if they want to survive. In Avida, these artificial organisms are awarded extra computer “food” if they manage to acquire a set of instructions that performs a simple computational task. (Let’s not worry about the computer details of how instructions are acquired or lost.) And, as you may have guessed, random mutation and natural selection seem to be perfectly capable of delivering the needed instruction sets.
What are we to make of this apparent contradiction? If, as we have seen, random mutation is incoherent and severely constrained in our best evolutionary studies of real biological organisms, how can a process that is supposedly analogous to Darwinism work for a computer program? The simple answer is that the conclusions drawn from an analogy are only as good as the analogy. Although Avida is lifelike in a few respects, it only takes one critical departure for the overall analogy to fail.
Let’s look at just one example to illustrate the point. In Avida, acquiring new abilities is only one way for an organism to get computer food. Another way is by simply acquiring surplus instructions, whether or not they do anything. In fact, instructions that aren’t ever executed—making them utterly useless for performing tasks—are beneficial in Avida because they provide additional food without requiring any additional consumption. It’s survival of the fattest!
It’s also very unrealistic. Biological organisms show the opposite behavior—genes that are useless in the real world are not rewarded; the genes are rapidly lost or degraded by mutation. Why, then, was Avida programmed to do the opposite—to reward organisms for carrying useless instructions? As explained on the Avida website, the counterbiological feature was needed, “Otherwise there is a strong selective pressure for shorter genomes.”11 In other words, otherwise the program wouldn’t give the desired results. The computer programmers remark, “This isn’t the most elegant fix, but it works.”
Computers can be useful tools in science when the assumptions built into programs are realistic. But if assumptions are wrong, computer simulations can be misleading. That’s why the most informative evolutionary studies by far are ones of real organisms such as malaria. The million-murdering death makes no assumptions.
Notes
2 Arms Race or Trench Warfare?
1.The phrase is from a poem found in a letter written by the British scientist Ronald Ross to his wife on August 20, 1897, which he called “Mosquito Day.” Ross (1857–1932), who discovered that malaria is transmitted by mosquitoes, won the 1907 Nobel Prize in Physiology or Medicine. The poem is reproduced in Sherman, I. W. 1998. Malaria: parasite biology, pathogenesis, and protection. Washington, D. C., ASM Press, 6.
2.Born into an affluent Grenad
ian family, the well-educated Noel came to the United States in 1904 to study dentistry at the Chicago College of Dental Surgery. Despite occasional hospitalizations for the effects of his unrecognized sickle cell disease, Noel graduated in 1907 and returned to his native Grenada, where he set up a successful dental practice. He died at the age of thirty-two from “asthenia from pneumonia,” probably as a secondary result of sickle cell disease. Savitt, T. L., and Goldberg, M. F. 1989. Herrick’s 1910 case report of sickle cell anemia. The rest of the story. JAMA 261:266–71.
3.Pauling, L., Itano, H. A., Singer, S. J., and Wells, I. C. 1949. Sickle cell anemia, a molecular disease. Science 110:543–48.
4.Ingram, V. M. 1958. Abnormal human haemoglobins. I. The comparison of normal human and sickle-cell haemoglobins by fingerprinting. Biochim. Biophys. Acta 28:539–45; Hunt, J. A., and Ingram, V. M. 1958. Abnormal human haemoglobins. II. The chymotryptic digestion of the trypsin-resistant core of haemoglobins A and S. Biochim. Biophys. Acta 28:546–49.
5.Although some authors think the sickle gene arose independently more than once, Cavalli-Sforza argues for a single origin (Cavalli-Sforza, L. L., Menozzi, P., and Piazza, A. 1994. The history and geography of human genes. Princeton, N.J.: Princeton University Press.)
6.Forget, B. G. 1998. Molecular basis of hereditary persistence of fetal hemoglobin. Ann. N.Y. Acad. Sci. 850:38–44.
7.Bookchin, R. M., Nagel, R. L., and Ranney, H. M. 1967. Structure and properties of hemoglobin C-Harlem, a human hemoglobin variant with amino acid substitutions in 2 residues of the beta-polypeptide chain. J. Biol. Chem. 242:248–55.
8.Except under extreme conditions not found in a normal
9.Let’s compare one population in which half of the genes are normal and half sickle to another population in which half of the genes are normal and half C-Harlem. Assuming random inheritance, one-quarter of both populations would have two normal genes and be vulnerable to malaria, and one-half of both populations would have one normal and one mutant gene, which in both cases would confer resistance to malaria. Two copies of sickle are for all intents and purposes lethal. Two copies of C-Harlem would not be lethal in themselves, but might not provide much protection against malaria. So the remaining quarter of the first population would have two copies of the sickle gene and die, presumably before reproducing. The remaining quarter of the second population would have two copies of the C-Harlem gene and live for a while, but with presumably little or no resistance to malaria. Assuming their vulnerability to malaria leads to a 50 percent mortality rate before reproducing, then the advantage of the second population would be in the one-half of the quarter of the population with two copies of the mutant gene that managed to survive malaria. This amounts to a one-eighth selective advantage of C-Harlem over sickle—a very large edge in evolutionary terms.
A seemingly even better solution to the problem hasn’t turned up yet in nature. If a normal beta gene and a sickle gene could be brought together on the same chromosome (similar to something like Hb anti-Lepore: Efremov, G.D. 1978. Hemoglobins Lepore and anti-Lepore. Hemoglobin 2:197–233) and be equally expressed, then a population in which that arrangement was universal would be superior to either of the two mixed populations discussed above. If everyone in the population expressed both normal and sickle hemoglobin, then one would effectively have a population where everyone had sickle trait, with no cases of either sickle disease (lethal in itself) or vulnerability to malaria.
10.Tishkoff, S. A., Varkonyi, R., Cahinhinan, N., Abbes, S., Argyropoulos, G., Destro-Bisol, G., Drousiotou, A., Dangerfield, B., Lefranc, G., Loiselet, J., Piro, A., Stoneking, M., Tagarelli, A., Tagarelli, G., Touma, E. H., Williams, S.M., and Clark, A. G. 2001. Haplotype diversity and linkage disequilibrium at human G6PD: recent origin of alleles that confer malarial resistance. Science 293:455–62.
11.Another interesting hemoglobin mutation that I won’t discuss is called hemoglobin E. Like HbC, it has a glutamic acid to lysine change in the beta chain. However, the position in the beta chain that’s altered is different. In HbE, amino acid number 26 is changed. HbE appears mostly in Asian populations and is thought to have antimalarial properties because it is found frequently in malarious regions.
12.Modiano, D., Luoni, G., Sirima, B. S., Simpore, J., Verra, F., Konate, A., Rastrelli, E., Olivieri, A., Calissano, C., Paganotti, G. M., D’Urbano, L., Sanou, I., Sawadogo, A., Modiano, G., and Coluzzi, M. 2001. Haemoglobin C protects against clinical Plasmodium falciparum malaria. Nature 414:305–8.
13.The threat of death from malaria diminishes for those who are constantly exposed to it—not for adults who have only infrequently or never been exposed.
14.Darwin married his cousin Emma, with whom he had ten children.
15.Children with one C gene and one sickle gene have symptoms similar to, but milder than, sickle cell disease.
16.Some things that could greatly complicate the outcome include: the development of other hemoglobin mutations (like C-Harlem); occurrence of other, nonhemoglobin, mutations in humans; the rate of flow of normal hemoglobin genes (HbA) into the populations; the effects of inbreeding; changes in the malarial parasite P. falciparum; and changes in the mosquitoes that carry the parasite.
17.Fortin, A., Stevenson, M. M., and Gros, P. 2002. Susceptibility to malaria as a complex trait: big pressure from a tiny creature. Hum. Mol. Genet. 11:2469–78; Kwiatkowski, D. 2000. Genetic susceptibility to malaria getting complex. Curr. Opin. Genet. Dev. 10:320–24; Mazier, D., Nitcheu, J., and Idrissa-Boubou, M. 2000. Cerebral malaria and immunogenetics. Parasite Immunol. 22:613–23; Kwiatkowski, D. P. 2005. How malaria has affected the human genome and what human genetics can teach us about malaria. Am. J. Hum. Genet. 77:171–92.
18.Strachan, T., and Read, A. P. 1999. Human Molecular Genetics, 2nd ed. Wiley: New York, Table 17.2.
19.The alpha+ thalassemias. (Carter, R., and Mendis, K. N. 2002. Evolutionary and historical aspects of the burden of malaria. Clin. Microbiol. Rev. 15:564–94; Allen, S. J., O’Donnell, A., Alexander, N. D., Alpers, M. P., Peto, T.E., Clegg, J. B., and Weatherall, D. J. 1997. Alpha+-thalassemia protects children against disease caused by other infections as well as malaria. Proc. Natl. Acad. Sci. USA 94:14736–41; Flint, J., Hill, A. V., Bowden, D. K., Oppenheimer, S. J., Sill, P. R., Serjeantson, S. W., Bana-Koiri, J., Bhatia, K., Alpers, M.P., and Boyce, A. J. 1986. High frequencies of alpha-thalassaemia are the result of natural selection by malaria. Nature 321:744–50.
20.Greene, L. S., and Danubio, M. E. 1997. Adaptation to malaria: the interaction of biology and culture. Gordon and Breach Publishers: Amsterdam.
21.The gene for G6PD occurs on the X chromosome. Women have two X chromosomes but men have only one (in addition, they have a Y chromosome). So a woman can have a broken G6PD gene on one of her X chromosomes but a working copy on the other. If a man inherits a broken G6PD gene, he has no second copy to back it up.
22.Ruwende, C., Khoo, S. C., Snow, R. W., Yates, S. N., Kwiatkowski, D., Gupta, S., Warn, P., Allsopp, C. E., Gilbert, S. C., and Peschu, N. 1995. Natural selection of hemi- and heterozygotes for G6PD deficiency in Africa by resistance to severe malaria. Nature 376:246–49.
23.Alberts, B. 2002. Molecular biology of the cell, 4th ed. New York: Garland Science, pp. 604–5.
24.Carter and Mendis. 2002.
25.Kennedy, J. R. 2002. Modulation of sickle cell crisis by naturally occurring band 3 specific antibodies—a malaria link. Med. Sci. Monit. 8:HY10-HY13.
26.Pogo, A. O., and Chaudhuri, A. 2000. The Duffy protein: a malarial and chemokine receptor. Semin. Hematol. 37:122–29. The nucleotide at position–33 in the gene is changed from a T to a C. The mutation abolishes the binding site for h-GATA-1 erythroid transcription factor.
27.Dawkins, R. 1986. The blind watchmaker. New York: Norton, p. 178.
28.Ibid., p. 181.
3 The Mathematical Limits of Darwinism
1.Hastings, I. M., Bray, P. G., and Ward, S. A. 2002. Parasitology. A requiem for chloroquine. Science 298:74–75.<
br />
2.Plowe, C. V. 2005. Antimalarial drug resistance in Africa: strategies for monitoring and deterrence. Curr. Top. Microbiol. Immunol. 295:55–79.
3.Wellems, T. E., Walker-Jonah, A., and Panton, L. J. 1991. Genetic mapping of the chloroquine-resistance locus on Plasmodium falciparum chromosome 7. Proc. Natl. Acad. Sci. USA 88:3382–86.
4.Su, X., Kirkman, L. A., Fujioka, H., and Wellems, T. E. 1997. Complex polymorphisms in an approximately 330 kDa protein are linked to chloroquine-resistant P. falciparum in Southeast Asia and Africa. Cell 91:593–603.
5.Gardner, M. J., et al. 2002. Genome sequence of the human malaria parasite Plasmodium falciparum. Nature 419:498–511.
6.Bray, P. G., Martin, R. E., Tilley, L., Ward, S. A., Kirk, K., and Fidock, D. A.2005. Defining the role of PfCRT in Plasmodium falciparum chloroquine resistance. Mol. Microbiol. 56:323–33.
7.Fidock, D. A., Nomura, T., Talley, A. K., Cooper, R. A., Dzekunov, S. M., Ferdig, M. T., Ursos, L. M., Sidhu, A. B., Naude, B., Deitsch, K. W., Su, X. Z., Wootton, J. C., Roepe, P. D., and Wellems, T. E. 2000. Mutations in the P. falciparum digestive vacuole transmembrane protein PfCRT and evidence for their role in chloroquine resistance. Mol. Cell 6:861–71.
8.Two different origins in South America, one in Papua–New Guinea, and one in Asia that spread to Africa. (Wootton, J. C., Feng, X., Ferdig, M. T., Cooper, R. A., Mu, J., Baruch, D. I., Magill, A. J., and Su, X. Z. 2002. Genetic diversity and chloroquine selective sweeps in Plasmodium falciparum. Nature 418:320–23.)
9.Kublin, J. G., Cortese, J. F., Njunju, E. M., Mukadam, R. A., Wirima, J. J., Kazembe, P. N., Djimde, A. A., Kouriba, B., Taylor, T. E., and Plowe, C. V.2003. Reemergence of chloroquine-sensitive Plasmodium falciparum malaria after cessation of chloroquine use in Malawi. J. Infect. Dis. 187:1870–75; Cooper, R. A., Hartwig, C. L., Ferdig, M. T. 2005. Pfcrt is more than the Plasmodium falciparum chloroquine resistance gene: a functional and evolutionary perspective. Acta. Trop. 94:170–80. Drug resistance mutation in pfmdr, the other protein involved in chloroquine resistance, also incurs a fitness cost (Hayward, R., Saliba, K. J., Kirk, K. 2005. pfmdr1 mutations associated with chloroquine resistance incur a fitness cost in Plasmodium falciparum. Mol. Microbiol. 55:1285–95).
The Edge of Evolution Page 30