For years, evolution's critics picked on supposed gaps in the historical record--missing links between different forms or species in biologists' evolutionary lineages. Evolutionary leaps, say from dinosaurs to birds, are inconceivable without intermediates, so the reasoning went. Finding key fossils is no easy matter, but creationists interpreted the absence of evidence as evidence of absence--no links, no evolution, only supernatural design.
Paleontologists were patient, though. They predicted that the feathers so important in bird flight were probably co-opted from another function, most likely thermal insulation. If that's true, scientists should eventually find fossils of feathered flightless animals. Their patience paid off over the past few years as China's Liaoning province yielded spectacular specimens of feathered dinosaurs.1 And birds aren't alone. The same painstaking process of scientific discovery is illuminating the evolutionary history of flowering plants, whales, snakes, and--dare we say it--humans.
But never say die--if cats have nine lives, creationism has at least a dozen. Having lost the fossil wars, creationists turned to biochemical pathways and subcellular structures. How could a biochemical pathway, which may involve 20 or more separate steps catalyzed by a score of enzymes, evolve? They don't, according to a new breed of "neocreationists" rallying under the banner of Lehigh University's Michael Behe. Unfortunately, the argument is familiar. Ergo, surely a metabolic pathway with a specific function is "irreducibly complex,"2 making stepwise evolution unlikely. Remove the feathers--er, an enzyme--and it doesn't fly. Right?
Wrong, say biochemists and evolutionary biologists. Now philosopher of science Niall Shanks has added his two cents. With colleague Karl Joplin at the University of East Tennessee in Johnson City, Shanks argued that biological systems exhibit "redundant complexity," not irreducible complexity, which makes Behe's idea a big oversimplification.3 Enzymes come in multiple forms, or isozymes, that produce a variety of outcomes. Likewise, genes duplicate and specialize, thereby creating new pathways and functions.4 The evidence is obvious in a host of multigene families governing all sorts of processes. Redundancy is "like a scaffold" that supports pathways in the making, explains Shanks.
In essence, evolution co-opts parts from preexisting hardware--what Stephen J. Gould called exaptation.5 Behe proposed a "special kind of complexity that cannot be explained in naturalistic terms," says Shanks, but "it can be explained naturally without magic or hocus pocus."
"We were able to catch notothenioid AFGP evolution 'in action,' so to speak."
Eyes on Bacteria
The lens crystallins of eyes are some of the best examples of exaptations. Crystallins lend optical properties to lens cells important in light transmission. Various proteins have specialized as crystallins during eye evolution, some resulting from gene duplication followed by specialization, others retaining their original metabolic activities in a process Joram Piatigorsky of the National Eye Institute called recruitment by gene sharing.6 Lactate dehydrogenase, aldehyde dehydrogenase, and enolase have all been put to work as lens crystallins, while still acting as enzymes.
If the really important question in eye evolution isn't gross anatomy but molecular pathways, as Behe believes, the answer isn't in intelligent design or other supernatural handwaving, but more biochemistry and genetics. That also holds true for nitrogen-fixing rhizobial bacteria that inhabit the root nodules of legumes. According to J. Peter Young of York University, United Kingdom, rhizobia may have borrowed genes from each other, fungi, and even host plants to patch together new biosynthetic pathways for nod factors, signaling molecules that let roots know the bacteria are around.7
Biochemical Irreducibility--The Deep Freeze
A team led by Chi-Hing Cheng, senior research scientist in the Department of Ecology, Ethology and Evolution at the University of Illinois, Urbana, recently uncovered one of the nicest examples of biochemical exaptation, in fish that thrive in Antarctic waters where temperatures go below -2 degrees Celsius, lower than the freezing point of their blood. Plants and animals manufacture a variety of antifreeze proteins that block the growth of destructive ice crystals. Notothenioids like the Antarctic toothfish make antifreeze glycoproteins (AFGPs) that vary in molecular weight from 2,600 to 34,000 daltons. The fish maintain high levels of glycoprotein in the blood because they have multiple genes, each encoding a "polyprotein" that's chopped into numerous AFGPs.
AFGPs consist of a repeating three-amino acid sequence consisting of threonine alanine-alanine. With two sugars attached to each threonine, AFGPs are the fish's version of ethylene glycol. But how did such unusual proteins arise? How did notothenioids get their antifreeze genes as Antarctic waters froze 10-15 million years ago? Cheng's group discovered that AFGP genes evolved from an ancestral gene encoding trypsinogen, a pancreatic protein that cleaves to produce the digestive enzyme trypsin.8 The molecular footprints were obvious: AFGP and trypsinogen genes share significant sequence identity at several locations.
What clinched the story was Cheng's finding that trypsinogen contains a three-amino acid sequence with no known function in the enzyme. You guessed it: threonine alanine-alanine. In constructing AFGP, the tripeptide reiterated again and again, probably because the repetition had antifreeze properties strongly selected by ice cold water. Most of the rest of the trypsinogen gene was discarded. By deleting parts of the trypsinogen gene and recruiting and amplifying others, evolution did its borrowing act.
Now for the icing on the cake: The toothfish contains not only genes for AFGP and trypsinogen, but a hybrid gene--a missing link?--encoding both AFGP and trypsinogen.9 The AFGP part occurs exactly where expected, near the beginning of the trypsinogen portion of the gene that previously encoded the ancestral tripeptide. Says Cheng, "We were able to catch notothenioid AFGP evolution 'in action,' so to speak, because we believe the protease-AFGP split is a rather recent event in the evolutionary time scale." Cheng is now trying to find out if the hybrid protein has trypsin activity.
Barry A. Palevitz (email@example.com) is a contributing editor for The Scientist.
1. X. Xu et al., "A dromaeosaurid dinosaur with a filamentous integument from the Yixian formation of China," Nature, 401:262-6, Sept. 16, 1999.
2. M.J. Behe, Darwin's Black Box: The Biochemical Challenge to Evolution, New York, The Free Press, 1996.
3. N. Shanks, K.H. Joplin, "Redundant complexity: a critical analysis of intelligent design in biochemistry," Philosophy of Science, 66:268-82, June 1999.
4. S. Ohno, Evolution by Gene Duplication, Berlin, Springer-Verlag, 1970.
5. S.J. Gould, E.S. Vrba, "Exaptation--a missing term in the science of form," Paleobiology, 8:4-15, 1982.
6. J. Piatigorsky, G. Wistow, "The recruitment of crystallins: new functions precede gene duplications," Science, 252:1078-9, 1991.
7. J.P. Young, "The evolution of rhizobia and their nodulation genes," XVI International Botanical Congress, 1999.
8. L.B. Chen et al., "Evolution of antifreeze glycoprotein gene from a trypsinogen gene in Antarctic notothenioid fish," Proceedings of the National Academy of Sciences, 94:3811-6, April 15, 1997.
9. C.-H. Cheng, L. Chen, "Evolution of an antifreeze glycoprotein," Nature, 401:443-4, Sept. 30, 1999.