How the Camel Got Its Hump, One Protein at a Time

Scientists study enzymes for clues to the way organisms adapt

By LILA GUTERMAN

Life sometimes flaunts its ability to evolve. Camels carry around conspicuous humps that store fat for times of need. Giraffes sport impossibly long necks that provide access to the leaves atop trees. Bats employ a sonar system that allows them to catch prey under the cover of darkness.

But the end point of evolution often doesn't reveal how such adaptations came about. Paleontologists have traditionally tried to reconstruct the story by looking for fossils of organisms that came before today's creatures. Now, some other scientists are taking a different approach, by thinking small. Since protein molecules called enzymes carry out the basic functions of life, those molecules themselves must change for an animal, plant, or microbe to develop new adaptations.

"It's how we got where we are," says Gregory A. Petsko, a professor of biochemistry and chemistry at Brandeis University.

How did the enzymes of life change so that bats developed sonar and camels grew humps? Several biologists think that they have deciphered the general answer and that they can use the new knowledge for a variety of purposes: to make predictions about the functions of newly discovered proteins, to understand why microorganisms can degrade some pollutants but not others, and to watch evolution happening in the laboratory.

Enzymes display exquisite complexity as the chains of amino acids fold into spirals, barrels, flat sheets, and other shapes. They speed the chemical reactions necessary for life by grabbing onto another molecule, called a substrate, and guiding it through a chemical transformation such as removing chlorine or adding water. Most enzymes are acutely specific: They will catalyze only a certain reaction and perform it only on certain substrates.

"Enzymes are capable of accelerating almost any reaction you can possibly imagine," Mr. Petsko says. And yet, nature has evolved a limited number of enzyme shapes -- perhaps a thousand, estimates Patricia C. Babbitt, an associate professor of pharmaceutical chemistry at the University of California at San Francisco. "That means that new enzyme functions must have evolved from old enzymes, from the same structural scaffolds," Mr. Petsko says. Most likely, organisms duplicate the gene that produces an enzyme, thereby keeping one working copy, while the other copy undergoes some mutation.

To understand how that second copy changes, think of a growing business. A company can expand in one of two ways: It can offer its regular service to new customers, or it can provide its established customers with some new service. In much the same way, an altered enzyme can perform its regular reaction on a new substrate or it can keep binding to the same substrate but perform a new reaction. "For the last 50 years, people thought the way that worked is ... [that enzymes] kept the ability to bind a substrate," says Ms. Babbitt. But she, Mr. Petsko, John A. Gerlt, an enzymologist at the University of Illinois at Urbana-Champaign, and others have shown that far more often, evolving enzymes catalyze the same reactions as their forerunners but work on different substrates.

To tell the difference between the two mechanisms of enzyme evolution, the scientists looked at the functions of distantly related enzymes. The researchers can tell enzymes are related because they have similar structures.

If enzymes evolve by subjecting the same substrate to new reactions, then distantly related enzymes with similar shapes should bind to the same substrates. "There are some examples of that," says Mr. Petsko. But in "the overwhelming major ity of cases," enzymes evolve by doing the same sort of chemistry on a very different substrate. The researchers found that in most groups of structurally similar, related enzymes, every member catalyzes the same or similar chemical reactions.

That knowledge could help scientists trying to predict the functions of enzymes newly discovered from the sequencing of many organisms' genomes. Unless the enzyme is obviously very similar to one of known function, Ms. Babbitt says, researchers have had difficulty guessing its use. But now, she says, "if that protein looks very distantly like a group of proteins that do a range of different functions, we can look at that range of functions and try to fit that new guy in."

Although the new findings contradict longstanding assumptions about how enzymes evolved, they have not sparked much controversy, says Ms. Babbitt. The reason is that researchers had previously lacked sufficient information on protein sequences and structures to study the question systematically.

Besides, says Mr. Petsko, the chemistry-first, substrate-second route makes more sense from an evolving organism's point of view. "Suppose I'm a bacterium living in the soil, and all of a sudden people dump a certain toxic substance into the soil," he says. "Let's say I evolve to eat it. How would I do that? There are two models. One is if it turns out this toxic substance looks like some substrate I already use ... I could tinker around with the chemistry to do something useful." But during the time it takes for those adaptations to develop, the bacterium would get nothing out of the process and would have no competitive advantage over other bacteria. So natural selection wouldn't favor such a change.

"Now try the second strategy," Mr. Petsko says. "I find an enzyme that I already have that by accident takes this toxic substance and at a very low level chews it up. That immediately gives me a selective advantage." Even if the enzyme doesn't bind to the toxic substrate well, it degrades at least some of the material and the bacteria can survive while the enzyme evolves to deal with the substrate more efficiently.

That's exactly what Shelley D. Copley, an associate professor of molecular, cellular, and developmental biology at the University of Colorado at Boulder, found when she looked at bacteria that have evolved to degrade pentachlorophenol, a highly toxic wood preservative first introduced to the environment in 1936. Before then, bacteria had no reason to have enzymes to degrade it. She has found that two enzymes that help break down the foreign substance both show signs that they have recently evolved and may still be undergoing change.

The first one "is an incredibly poor enzyme," she says. The reaction it catalyzes is 100 times slower than reactions sped by other enzymes in its family. All of the enzymes in this family work to add an oxygen and a hydrogen to their substrates, and sure enough, so does this enzyme. "This is an enzyme that's really struggling," she says.

"We're trying to figure out why the bacterium is having a problem making this a better enzyme."

The second enzyme speeds a reaction that removes chlorine atoms from pentachlorophenol. It also seems to be at an early stage of evolving. The enzyme actually slows down if more substrate is present -- the reverse of the typical situation.

Mr. Petsko has captured some still-evolving enzymes in another organism. He reported at an American Chemical Society meeting last month that he found a yeast enzyme that has been duplicated three times, and it's clear that the three new enzymes have recently diverged since they share 99.9 percent of their amino acids. He thinks the new enzymes don't serve any function yet because when he engineered yeast not to make the enzymes, the altered and normal yeast appeared the same.

Mr. Petsko hopes to watch evolution happen in his laboratory with these three new enzymes. He plans to "mutate the living daylights" out of them and then put the yeast under stresses that will select for different characteristics such as heat tolerance, ability to survive damaging radiation, or resistance to drugs.

If it works, the effects won't be as striking as a camel's hump or a giraffe's neck, but they will provide a clear example of evolution in action.

Easy-to-print version

E-mail this article