From Femtosecond Physics to Yeast Genetics

By Jason Socrates Bardi

Rene Descartes once remarked that as a young man, he read every book he could get his hands on—literally. The absurdity of making such a statement today is a measure of the immense explosion of information in the last few centuries.

This explosion of information has had a paradoxical effect on modern science. The more we know, the less aware we become of what we know. Disciplines have become so specialized that it is difficult to keep abreast of the discoveries in one sub-specialty, let alone several fields.

This is problematic for scientists because sometimes the best answers for the most important questions are already known—but not by the people who ask the question. And the people who know the answers are simply unaware that others are asking.

"Interdisciplinary science encourages you to look at problems in a unique way," says Floyd Romesberg, assistant professor in the Department of Chemistry at The Scripps Research Institute (TSRI), shortly before offering a tour of his lab. "That's what attracts me to science and to Scripps."

TSRI fosters interdisciplinary approaches through formal ties and infrastructure that bring scientists from multiple backgrounds together.

"The same people [in my laboratory] synthesize molecules and biophysically analyze them," says Romesberg.

Antibodies, Shaken Not Stirred

One area of research in Romesberg's laboratory involves studying the flexibility and dynamics of proteins using spectroscopy—a new, specialized application of an old tool.

"This is absolutely standard spectroscopy that people have done on small molecules for years," says Romesberg. For years laboratories have routinely used light spectrometers, say, to measure protein concentration or to follow enzymatic reactions.

However, Romesberg's spectrometer is not the kind you might find in any catalog of equipment lying around the lab. It is a custom-built femtosecond (10-15 second) laser spectrometer that takes up nearly an entire room. Nor is the application he is using it for a routine measurement—he is directly probing the flexibility of proteins in solution.

Protein flexibility is an important area in biology because of the role of flexibility in protein–protein recognition. Flexibility may be an important quality that characterizes the recognition of antigens by antibodies or helps an enzyme catalyze a reaction, for instance. Antibody recognition, says Romesberg, may not be the simple, fixed lock-and-key mechanism introductory texts elude to, but one in which the keys and the locks are vibrating and changing their shape as they come together in solution.

However, this sort of flexibility is difficult to characterize experimentally. In proteins, it involves bond vibrations that ever-so-slightly displace atoms a million times every millionth of a second.

These tiny vibrations are important for understanding how a protein recognizes its target with high affinity. And they are what gave Romesberg the idea to try spectroscopy, even though his own background is largely in bioorganic chemistry.

"If the research leads us to [something like] spectroscopy, we will follow the research," he says.

The Reluctant Spectroscopist

Romesberg has built a femtosecond laser to measure protein flexibility. This laser emits a burst of photons in a roughly 17 femtosecond pulse—which is billions of times faster than the fastest shutter speed on a good camera.

This incredible speed is necessary, though, because just as a fast shutter speed captures a fast movement on film, a fast laser captures a fast movement within a protein.

"[The laser] allows us to take 'photographs' of a protein vibrating," says Romesberg.

The femtosecond pulses excite the molecules in the sample, depositing energy, which is absorbed by vibrating bonds within the protein. The electron distribution in these bonds may then change, depending on how much they vibrate. By comparing an excited, "spectra" readout to a normal spectrum, Romesberg and his colleagues can assess how flexible particular parts of a protein are.

This is not always simple, however, since proteins are large molecules with lots of vibrating bonds—so many bonds that a spectrum may have overlapping vibrations that are impossible to differentiate.

So Romesberg selectively incorporates deuterium—so-called "heavy hydrogen" because it has an extra neutron—in the place of normal hydrogen atoms. This extra neutron changes the physics of the vibration and shifts it to a region where it can be observed distinct from other vibrations. This allows him to discriminate between vibrations without affecting the overall shape of the protein.

Expanding the Genetic Code

Another area of research in the group involves expanding the genetic code by adding additional letters (bases) to the genetic alphabet.

This concept is simple, but putting it into practice is not. To add bases, a scientist must have polymerase enzymes that can recognize the unusual bases and replicate them with the four standard bases—something that polymerases were not designed by nature to do.

Romesberg and his group have been tackling this challenge by using a combination of mutagenesis and phage display to invent polymerases that can recognize additional bases.

The lab has designed DNA polymerases using a innovative phage system that involves placing variant polymerase enzymes on a phage particle next to a piece of DNA with unnatural bases attached. The polymerases that are able to replicate the DNA will do so, incorporating biotin labels at the same time. These labels can then be used to distinguish the functional mutants from the rest of the polymerases, which cannot replicate the DNA with the unnatural bases.

And it works.

As a proof of principle, Romesberg's group, in collaboration with the group of Professor Peter Schultz, evolved a DNA polymerase into an RNA polymerase, demonstrating that directed evolution could be used to make an enzyme with a new function.

"Our evolved mutants synthesize RNA as efficiently as the DNA polymerase synthesize DNA," says Romesberg.

The group is currently working toward evolving a DNA polymerase that can highly efficiently replicate DNA with unnatural nucleosides. The team has worked with a number of novel bases, including many that form complementary pairs because of hydrophobic forces (as opposed to the hydrogen bonds that keep normal bases together). Lately, they have also been designing novel perfluorinated hydrocarbon bases, which have the strange property of preferring to form their own liquid phase, unlike most other substances, which are either soluble in water or oil.

The goal of this research is to develop an in vivo system—a strain of bacteria with the mutant polymerase that is able to use these unnatural bases and replicate, for instance.

So far, this can only be done in vitro, and Romesberg and his colleagues are working out the conditions that will allow it to work in vivo as well. This involves determining whether the unnatural nucleosides can be taken up into cells and be transported to the right place, as well as being processed into nucleotides along the way by metabolic enzymes which add phosphates.

The Mutation is the Message

The last area of research in Romesberg's laboratory is, as he puts it, "an effort to understand the genes that drive evolution."

This is an odd way of stating the problem, perhaps, because people have always understood evolution to be the force that drives genes. Mutations introduced into the DNA of an individual would propagate through the generations and eventually become part of the germline if the mutation was of benefit to the organism.

But most mutations are not beneficial to an organism, and life has generally evolved to make as few as possible. One of the main controls is the ubiquitous polymerase molecule, which has the task of replicating the genome.

Evolution has provided life with the ability to replicate genomes extremely well, with multiple, redundant repair and proofreading mechanisms that would make even NASA jealous. Nevertheless all organisms are subject to a certain level of spontaneous mutations—mistakes that escape repair and become part of the DNA of the cell in which they occur. Slowly over time, these mutations accumulate and species diverge.

However, mutations can be fast-tracked as well. When cells are subjected to ultraviolet radiation, for instance, the rate of mutations increases because the energy of the light is enough to cause bases of DNA to cross-link and introduce errors. When the DNA is later replicated, these errors are propagated into mutations.

However, this may not be the only way that UV radiation causes mutations. In an experimental result that Romesberg calls "amazing" mutations may actually increase because of the cells' response to the radiation and not solely from the radiation itself.

The experiment is a simple one: researchers take healthy phage particles—a virus that infects bacteria—and infect bacterial cells that have been previously subjected to UV radiation. As you would expect, the DNA of the bacteria is mutated because of its exposure to the radiation. However, the results of the experiment show that the DNA of the virus is also mutated. Since the viral DNA was never subjected to UV radiation, it had to have been mutated by the cell.

What the experiment may be pointing to is that cells may have a way to mutate themselves when they need to—such as when they are threatened with extinction.

Take the bacterium Escherichia coli for instance. When E. coli cells are subjected to damage, they upregulate repair enzymes, which then go to work trying to fix the problem. If the damage persists, the cell upregulates recombination enzymes, which are tasked with recombining the DNA—another way to repair it.

And, says Romesberg, if the damage still persists, the cells upregulate enzymes whose sole task is to make mutations. Presumably this is an effective evolutionary strategy for dealing with environmental changes that might otherwise wipe them out. In order to evolve, organisms have to mutate, so they turn on the mutation process when they are threatened with extinction.

"There is a growing belief now that all organisms have evolved these mechanisms to facilitate evolution when they have to," says Romesberg. "We are now working to understand these processes at the biochemical level."

Romesberg's group, in collaboration with Associate Professor Elizabeth Winzeler, is using a complicated competitive growth assay involving the yeast Saccharomyces cerevisiae, to do this.

They use a yeast library where every gene was deleted one by one and replaced with a tag. The assay screens the yeast for genes that are not necessary for growth but whose deletions makes the yeast resistant to mutations when irradiated. Then they can search for human homologues of these genes. Using a similar screen, they have recently identified six genes involved in the DNA damage response.

"That's what I love about science," says Romesberg as he describes the discovery. "Being on the steep edge of the learning curve."






Go back to News & Views Index

Investigator Floyd Romesberg directs a diverse program of research in his laboratory in the Department of Chemistry. Photo by Jason S. Bardi.




















This laser emits a burst of photons in a roughly 17 femtosecond pulse—a speed billions of times faster than the fastest shutter speed on a good camera.

















In a phage display system to evolve unnatural polymerase activity, polymerases are displayed on phage with covalently attached DNA oligonucleotide substrates. Desired mutants that synthesize DNA with the unnatural bases biotinylate the attached oligonucleotides and may be selectively isolated. Also shown is the triphosphate of a previously developed unnatural base. Image courtesy of Jodie K. Chin.