On Science: DNA supports Darwin
This article first appeared in the St. Louis Beacon, Feb. 11, 2009 - On Feb. 12, we celebrate Charles Darwin's 200th birthday. Perhaps no individual has had such a sweeping influence on our views of who we are and how we relate to the world around us.
Darwin saw humankind as but one strand of a much larger web of life, a view very important today as we face the impact of human activities on our globe's climate and health.
Darwin's idea was that evolution explains life's diversity, the challenges of living favoring those members of a species better able to survive and thrive. Over time the population changes as the more "fit" individuals become more common. This is what Darwin called "natural selection."
Eventually the population becomes so different we call it a new species. In Darwin's view, all living things are related to one another by ancestors we have in common, we humans are part of a much larger family of life sharing the resources of planet earth.
A simple idea, Darwin's proposition of evolution by natural selection now forms the intellectual backbone of the science of biology, a subject I taught at Washington University for 30 years.
I began my doctoral work - 40 years ago! I had red hair then and was a bit thinner. For all those years, I have been a Darwinist - just so you know my bias.
So, why make such a fuss about Darwin's anniversary?
This theory by Darwin has led directly to major advances in medical research (an understanding of why bacteria become resistant to antibiotics, for example) and crop improvement, and it has a great deal to say about humanity's distant past and the world's ecological future.
In Missouri, you know of the negative view of Darwin's idea held by a broad slice of Americans. Darwin's theory of evolution, while almost universally accepted by the scientific community -- as surely as gravity and electricity -- is not accepted by many non-scientists. Darwin is as controversial now in Missouri as during the Scopes trial in 1925.
It is not my goal here to delve into this controversy, because among biologists there is no controversy. Darwin's case is firmly established by an overwhelming body of evidence. Instead, what I want to do is focus on one narrow aspect of that evidence: what DNA tells us about evolution.
In recent years, biology has been awash in DNA data. Completion of the Human Genome Project opened the door, allowing researchers to read the genome -- all the DNA -- of almost any species. Now, instead of trying to infer what has happened from the few fossil bones that happen to survive, we can directly compare the DNA of different species to search for traces of a species' evolutionary past. As you will see, these new data fully support Darwin's theory. But what is really interesting about them is the light they shed on the currently fashionable version of "Darwin-is-wrong-ism," Intelligent Design.
First, let's look at the DNA evidence. Then we will come back to Intelligent Design.
As an old man, Darwin said, "The fact of evolution is the backbone of biology" - the same statement I just made. So, let's start with that. Is Darwin right? Is evolution a fact?
To address this, it's important to start where Darwin did, with a clearly defined hypothesis to test. Darwin defined evolution as "descent with modification." By this he meant that one species, over time, may include among its descendants a new, different species. The fossil record is full of examples of this. Many of you may have read of the discovery of new fossil whales - species that evolved from four-footed hooved mammals in a series of steps preserved for us to see as fossils in rock.
Darwin was very clear about this "descent with modification." A sketch he made in his notebook in 1837 shows diversification of a species from a single stock. He did not view "descent with modification" as a linear process, (monkey - ape - Australopithicene - Neanderthal - human) but rather as a branching process, with many of the branches now extinct. The fossil evidence supports this view very well. Look how well the 2007 "tree" of human evolution matches Darwin's 1837 sketch!
So what does DNA tell us about the tree of life? If you look at genomes, at all the DNA organisms possess, do you find evidence of evolution? Let's see.
If you think about it, if Darwin's assertion is correct - that organisms like a bird evolved from ancestral species like a dinosaur - we should be able to track evolutionary changes in our DNA.
The genes of dinosaurs reflect adaptations to the environmental challenges of the world they lived in. Over millions of years, the world changed, and new solutions were required to meet new challenges. At every stage, changes were made to the DNA, and over time they add up!
This hypothesis, that evolutionary changes reflect accumulated changes in DNA, leads to the following prediction: Two species that are more distantly related (for example, humans and birds) should have accumulated a greater number of evolutionary differences than two species that are more closely related (say, humans and chimpanzees).
So have they? Let's compare vertebrate species to see.
The "family tree" of vertebrates shows how biologists believe 18 different vertebrate species are related, based on comparing fossils, embryos and many other similarities and differences. If you look at the tree for a moment, you can see building a tree like this is not rocket science. Apes and monkeys, because they are in the same order (primates), are considered by biologists to be more closely related to each other than either is to members of another order, such as mice and rats (rodents), and so are placed closer together on the tree.
The wealth of genomes (again, a genome is all the DNA that an organism possesses) that have been sequenced since completion of the human genome project allows us to directly compare the DNA of these 18 vertebrates, and in this way to see just how different they are.
To reduce the size of the task, investigators at the National Human Genome Research Institute working at the University of California Santa Cruz, have focused on 44 so-called ENCODE regions. These regions are scattered around the vertebrate genomes. Corresponding to roughly 1 percent of the total human genome, they were selected to be representative of the genome as a whole, containing protein-encoding genes as well as non-coding "junk" DNA.
I'm not very bashful when it comes to this sort of inquiry, so I contacted the investigators directly and enlisted their help. For each vertebrate species, I asked the investigators to determine the similarity of its DNA to that of humans -- that is, the percent of the nucleotide letters in the DNA of that organism's 44 ENCODE regions that match those of the human genome.
As Darwin's theory predicts, the closer the relatives, the less the genomic difference we see. The chimpanzee genome is more like the human genome (91 percent similarity for these ENCODE regions) than the monkey genomes are (72 - 79 percent). Furthermore, these five genomes, all order Primates, are more like each other than any are to those of another order, such as rodents (mouse and rat).
In general, as you proceed through the taxonomic categories of the vertebrate family tree from very distant relatives (some in the same class as humans) to very close ones (in the same family), you can see clearly that genomic similarity increases as taxonomic distance decreases - just as Darwin's theory predicts. The prediction of evolutionary theory is solidly confirmed.
The analysis does not have to stop here. The evolutionary history of the vertebrates is quite well known from fossils. And because many of these fossils have been independently dated using tools such as radioisotope dating, it is possible to recast the analysis in terms of concrete intervals of time, and to assess directly whether or not vertebrate genomes accumulate more differences over longer periods of time as Darwin's theory predicts.
For each of the 18 vertebrates being analyzed, we can plot genomic similarity -- how alike the DNA sequence of the vertebrate's ENCODE regions are to those of the human genome--against divergence time (that is, how many millions of years have elapsed since that vertebrate and humans shared a common ancestor in the fossil record).
Thus, the last common ancestor shared by chickens and humans was an early reptile called a dicynodont that lived about 250 million years ago; since then the genomes of the two species have changed so much that only 7 percent of their ENCODE sequences are still the same. So that's one point on the graph, where 7 percent intersects with 250 million years.
For each vertebrate, I carried out a similar analysis, seeking out in the literature the age of the common ancestor of that vertebrate to humans. In each instance this involves a different paleontologist establishing the age of the common ancestor, often working decades and continents away from the other researchers. So there is no possibility of collusion among the researchers to get a desired result.
If there is a God overseeing the grand design of the universe, as I believe there is, then he is surely yelling in my ear here! Over their more than 300 million year history, vertebrates have steadily accumulated more and more genetic change in their DNA. "Descent with modification" was Darwin's definition of evolution, and that is exactly what the graph is showing us. The evolution of the vertebrate genome over time is not a theory, but an observation - without any wiggle room, a fact. Just as Darwin said.
So that's the first line of evidence that the new DNA data provides us about evolution.
DNA also provides us with another line of evidence, even more direct. We have seen that DNA has accumulated many changes over the course of vertebrate evolution. If Darwin is right, and these changes represent descent with modification, rather than total replacements of one species with another by an intelligent designer, then some of the DNA changes of ancestor forms should still be present in the DNA of today's vertebrates - Darwin's footprints in the DNA.
If Darwin is right, we should be able to see traces of our evolutionary past preserved in the DNA of present-day species. Can we?
To see, we will examine a key evolutionary advance among the animals, the invention of a way to carry out embryonic development not as one process, but rather as a series of sub-routines. Why is this important? Being able to break the process up into components allowed animals to run some of the subroutines in one part of the body, and other subroutines in other parts. For the first time, different segments of the animal body could develop different structures!
The design of the animal body -- decisions about what segment becomes what part -- is made by a small set of genes called (for reasons that totally escape me) HOX genes. The HOX genes act as a series of switches that activate alternate programs of development in different parts of the body.
You can look at flies to see how this works. There is one HOX gene guiding each fly body segment, one specifying a pair of wings, another the eyes and so forth. All of these genes are located on one chromosome, in a cluster in the same order as the segments of the body.
Now get this: These same HOX genes are found in virtually all animals, including us. Like the mouse, we humans have four clusters of the HOX genes, each on a different chromosome, each gene corresponding to a particular segment of the human body.
When you look at the HOX gene DNA sequences, each human HOX gene is a close match to a HOX gene in the fly. This evolutionary advance only happened once. We humans have inherited this ability to develop body segments differently from ancient ancestors - we carry their "HOX gene" in our DNA - Darwin's footprints.
For a similar and even more powerful example of Darwin's footprints in our DNA, we have only to pull back and compare human embryos to those of other vertebrates.
More than a hundred years ago, German embryologist Ernst Haechel pointed out similarities in early vertebrate embryos. Because he over-emphasized the similarities in the drawings he published, his observation was criticized, but his basic point is quite sound: reptiles, birds and humans all have gill slits and tails as embryos.
But could this just be subjective interpretation? Not at all. Look at the DNA.
Biologists have sequenced the HOX and other genes that govern embryo development. And in fish, bird, amphibian and mammal, all the same genes are used, in the same order, and they all have nearly identical DNA sequences! The same developmental tool kit has been preserved throughout the evolution of the vertebrates, strong evidence of their common ancestry. Again, Darwin's footprints.
But maybe such similarities are somehow a coincidence? A butterfly and an eagle both have wings, but this doesn't mean they are close relatives. Ken Miller, a biology professor at Brown University, answers this objection very clearly. He first describes an instance of cheating on a final exam, two students copying each other's work. Challenged, they denied cheating, pointing out to Miller that they were roommates and studied together from the same notes - any similarities in their answers must just reflect this and doesn't mean they copied.
Then Professor Miller placed the exam papers in front of the students, and called attention to a series of words, each circled by Miller in red ink. The two students had misspelled the same six words, in exactly the same way! They had to have copied from each other - while there is just one way to spell a word correctly, there are an infinite number of ways to get it wrong. When the errors match perfectly, there can be no doubt that the errors must have a common source.
What does this have to do with Darwin and DNA?
Consider a cluster of genes in human chromosome 16: the beta-globin gene, expressed at different times in embryonic, fetal and adult development. All have nearly identical DNA sequences. In the center of the cluster is one non-working gene, called a "pseudo-gene." It has the same DNA sequence as other members of the cluster, except for a series of eight errors that keep it from working. Think of them as "misspellings."
We humans are not the only species with a beta-globin gene cluster. Gorillas and chimpanzees have them, too - five working copies, surrounding a single pseudogene, just as in humans - and the gorilla and chimpanzee pseudogenes have exactly the same set of DNA errors! Just like Professor Miller's cheating students, they have matching mistakes. Eight identical mistakes could not have turned up independently in three unrelated species. The only sensible way to look at this is to conclude that the errors arose in a single common ancestor of the three species. Our genome is a copy of an earlier genome that preserves even the mistakes. Darwin's footprints.
All in all, the new DNA data build a very convincing case for Darwin and evolution.
But hold on. Perhaps an intelligent designer put everything together this way, footprints and all, just to test our faith, or for some other reason we can't appreciate! The key thing to appreciate, today's critics of Darwin cry, is that Darwinian evolution is impossible, (leaving an intelligent designer the only other possibility).
Impossible? How could that be? Intelligent design proponents argue that the intricate molecular machinery of our cells is so elaborate, our body processes so interconnected, that they cannot be explained by evolution from simpler stages. The molecular machinery of the cell is "irreducibly complex."
An irreducibly complex system is defined by intelligent design proponent Michael Behe as "a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning." Like a mechanical mouse trap, each part plays a vital role. Remove just one and cell molecular machinery cannot function.
As an example of such an irreducibly complex system, they often point to the series of more than a dozen blood-clotting proteins that act in our body to cause blood to clot around a wound. Take out any step in the complex cascade of reactions that leads to coagulation of blood, and your body's blood would leak out from a cut like water from a ruptured pipe. Remove a single enzyme from the complementary system that confines the clotting process to the immediate vicinity of the wound, and all your lifeblood would harden. Either condition would be fatal.
The need for all the parts of such complex systems to work leads directly to their criticism of Darwin's claim that natural selection has led to the evolution whose traces we see about us today. If dozens of different proteins all must work correctly to clot blood, how could natural selection act to fashion any one of the individual proteins? Like the mousetrap, our blood clotting system must have been designed all at once, as a single functioning machine.
What's wrong with Behe's argument, as evolutionary scientists have been quick to point out, is that a complex molecular machine does not evolve as that machine, but rather by putting together bits and pieces that have evolved for other reasons, pieces that were doing other things in the cell.
This is a surprisingly simple idea to demonstrate, as many biology professors now do in their classes.
How about a mousetrap? I have removed two parts, leaving just the platform, the spring and the hammer. With the spring, I can shoot a spitball - a very effective weapon. Or how about using it as a tie clasp? And the hold-down bar makes a nice toothpick!
You see the point. Each part of the mouse trap can have a different job. So the mouse trap is not irreducibly complex after all.
That's the fundamental fallacy in the irreducible complexity argument. Natural selection can assemble a complex system from parts with other functions
The mammalian blood clotting system, for example, has evolved in stages from much simpler systems. The core of the vertebrate clotting system, called the "common pathway", evolved at the dawn of the vertebrates approximately 600 million years ago, and is found today in lampreys, the most primitive fish.
As vertebrates evolved, proteins were added to the clotting system, improving its efficiency. The so-called "extrinsic pathway", triggered by substances released from damaged tissues, was added 500 million years go. Each step in the pathway amplifies what goes before, so adding the extrinsic pathways greatly increases the amplification and thus the sensitivity of the system.
Fifty million years later, a third component was added, the so-called "intrinsic pathway." It is triggered by contact with the jagged surfaces produced by injury. Again, amplification and sensitivity were increased to ultimately end up with blood clots.
Another "molecular machine" often cited as being irreducibly complex is the rapidly-rotating flagellum that bacteria use to propel themselves through the water. The flagellum has a very complex structure - a super-complicated mousetrap, if you will. But it turns out that the DNA sequences of genes encoding the proteins that make up the base of the flagellum are nearly identical to those of the genes making the proteins of a non-rotating membrane transport system in bacteria called the type III secretory system. The fact that a subset of the proteins from the flagellum is fully functional doing a different job in the type III secretory system shows clearly that the flagellum is NOT irreducibly complex.
So, there you have it.
Looking at DNA places our view of evolution in increasingly clear focus. Darwin's view that the diversity of life on earth reflects a long history of descent is completely verified, the footprints of our ancestors clear to see in our genes. The suggestion made by Darwin's critics that molecular machines within our cells are too complex to have evolved is shown by today's molecular data to be false -- a result that I am sure would not have surprised Darwin. Ken Miller's new book "Only a Theory," which has guided my critique of intelligent design, is a wonderful resource for any wishing to look into the matter further.
Next November will mark the 150th anniversary of the publication of "On the Origin of Species." It is a wonderful book to read, just as convincing now as it was a 150 years ago - the new DNA data just makes it shine all the more brightly.
George B. Johnson's "On Science" column looks at scientific issues and explains them in an accessible manner.
Johnson, Ph.D., professor emeritus of Biology at Washington University, has taught biology and genetics to undergraduates for more than 30 years. Also professor of genetics at Washington University’s School of Medicine, Johnson is a student of population genetics and evolution, renowned for his pioneering studies of genetic variability. He has authored more than 50 scientific publications and seven texts.
As the founding director of The Living World, the education center at the St Louis Zoo, from 1987 to 1990, he was responsible for developing innovative high-tech exhibits and new educational programs.
Copyright George Johnson