A twist on DNA

The rigors of physics meet the wild and woolly chromosome

He looks like a biologist.

Sitting at his lab bench, Paul Wiggins works with slightly impatient speed, preparing an experiment on yeast DNA by filling up a tray full of test tubes with genetic markers.

It’s true that the Whitehead Fellow got his doctorate in physics from the California Institute of Technology. The first paper on which he was lead author described interactions between stars and black holes.

Illustration: Stuart Bradford

But he decided to put the cosmos aside and instead seek ways to quantify biological forces and structures at the molecular level.

“In biology, structure is regulated very carefully; it’s not an accident,” Wiggins says. “Being physicists, we ask, What’s the structure, and how does it affect function?”

“We’re trying to do things very systematically,” he notes. “A lot of theories about this have been based on very circumstantial evidence. And biological models of physical interactions should be more than cartoons. We want to create mathematical models that distill which pieces are important and cut right through to the chase.”

“At the end of the day, numbers matter,” Wiggins emphasizes. “A factor of two in expression of a gene might make the difference between survival and death for an organism.”

All wound up, where does it go?

Most of us first learn to visualize DNA alone in its elegant double-helix form, picturing it floating idly in space.

Well, no. In the teeming chaos of our cells, two meters of DNA are crammed into a nucleus that’s one hundred-thousandth as wide. Like a string of beads, the DNA double helix is spooled around protein cores to create nucleosomes, and these are packed tightly together to make up a chromatin fiber. Then the chromatin fiber winds in wide loops around protein scaffolding to create a chromosome.

Enter proteins called transcription factors, which control gene expression by binding to DNA regulatory sequences. The binding sites may be next to the gene or far away on the genome. To affect gene expression, distant transcription factors must come into contact with the start of the gene, by bending the intervening DNA.

“Even in the simplest organisms, the genomic space (the length of the intervening DNA) can change the level of gene expression,” says Wiggins. “To understand the processes that the cell uses to convert genetic code to function, you must understand how DNA bends and twists.”

“The physics of DNA bending is very pretty,” he adds, using his belt to demonstrate how DNA twists.

“You basically take a look at every possible configuration. I find these problems very aesthetic because every configuration gets a vote—it’s a bit like a democracy, but some votes count more than others. If a configuration costs more energy it is less likely, and there is a famous formula that relates energy to probability. Essentially you can predict the outcome of a configuration by counting these weighted votes.”

Wiggins, Phil Nelson of the University of Pennsylvania, Cees Dekker of the Delft University of Technology and coworkers have just gathered results from the first experiments to directly and quantitatively measure the elastic properties of DNA on the five-nanometer scale. The researchers studied the bending of a long, random sequence of bacterial DNA with atomic force microscopy. They found that DNA can bend more easily than expected at high rates of curvature.

“We basically claim that the classic theory describing DNA bending fails at the 10-nanometer-length scale, just the scale where biological processes care about it,” Wiggins explains. “Our contribution was to make a direct measurement of the bendability of DNA. But due to the difficulty of pulling off this experiment, there is still skepticism about its results.”

Location, location, location

The level of gene expression is strongly affected by the gene’s precise location in the nucleus. “We want to systematically test this idea: How much does where you put something in the chromosome matter, quantitatively?” says Wiggins.

One key challenge is to create spatial measurements of genes in living cells. Joshua Martin, a graduate student in the Wiggins lab, is exploring the structure of chromatin (or rather, for the moment, the structure of DNA within yeast). To do this, he adds fluorescent tags to random locations on the genome. These locations are then sequenced in order to map out where the tags have landed, eventually helping to plumb the depths of gene structure. The work will test theoretical models that Martin has created with his advisor Jané Kondev of Brandeis University.

Another project in the Wiggins lab looks at the puzzle of heterochromatin regions. These are stretches of condensed chromatin that aren’t expressed most of the time, or are expressed at very low levels, while euchromatin regions are less condensed and more active. Graduate student Brian Ross is making a rigorous study of how the structure of both kinds of regions affects gene expression, warming up with studies in E. coli bacteria.

“There have been a lot of theories about which structural mechanisms drive gene expression, but we’re still really missing the big picture,” Wiggins stresses. “Many things could affect this, such as whether the gene is close to the nuclear envelope or to the center of the nucleus.”

“Genes that are silenced are very frequently found right next to the nuclear envelope—the membrane around the nucleus,” comments Andreas Hochwagen, another Whitehead Fellow, who is collaborating with Wiggins. “It could be that the envelope itself is setting up some kind of territory where genes are kept quiet. You can turn on a gene and watch it being moved into the center of the nucleus, which is pretty cool.”

“It’s not just a ball of wool in there,” Hochwagen says. “But we don’t know much about it. It’s a very basic cell biology problem.”

“No one has mapped out the complete physical locations of an entire chromosome,” says Wiggins. The two Fellows are planning an ambitious effort to take steps toward making that map, and perhaps to eventually extend it to the entire genome.

“In experiments in this field, the results seem to depend on how the material is isolated,” Wiggins says. “For instance, some researchers have found that chromatin fibers are 30 nanometers in diameter, while others say that it’s 10 nanometers. We’d like to look at things in live cells if possible, and if not, to perturb them as little as possible.”

Physical power

His background as a physicist gives Wiggins a different perspective in thinking about the business of the cell. While biologists often think of it as a chemical reactor, he likes to imagine it as a factory, with proteins as machines on interlocking assembly lines.

As experimental tools and mathematical models evolve, we’re getting much better ways to understand how these incredibly tiny machines crank along. “We can now do experiments where we watch a single molecule do things,” Wiggins points out. “Many technologies such as x-ray crystallography start out as physics,” he adds cheerfully. “Then they become useful, and they turn into biology.”

*************

Whole lotta shakin’ going on

“In everyday life, you and I have a very good intuition for what happens in the physical world,” says Paul Wiggins. “This intuition turns out to be completely wrong for molecular biology.”

Our intuition begins to fall apart at a certain scale, perhaps around one millimeter, the size of a gnat, he suggests. A cell is about 10 microns across, and an enzyme only 10 nanometers (billionths of a meter) wide.

Living in a fluctuating world [view movie] 220 kbps Taken under a microscope, this movie shows a filament attached to a rotating molecular motor (F1-ATPase) that exploits thermal fluctuations. The fluorescent filament serves as a reporter that tells us about the position of the molecular crank in the motor. Although on long time scales the motion progresses counter clockwise, on short time scales the motion is random as the filament is bombarded by molecules. When a fluctuation drives the filament through a complete step, this fluctuation is captured by ATP hydrolysis and the motor is reset, fluctuating around a new equilibrium position. Although the motor borrows energy from thermal fluctuations, it must repay the debt by hydrolyzing ATP (the fuel). Without ATP, the molecular motor cannot "lock-in" favorable fluctuations.  Video: Lodish et al. Molecular Cell Biology. & Noji et al. Nature 386:299.

“If what you’re interested in shrinks down enough, suddenly funny things start to happen,” Wiggins says. For one, the pressure surrounding a truly tiny object isn’t standard. That means that “no matter what you look at, if it’s small enough, it moves,” he says. “Ever since we invented microscopes, people have been doing this experiment by accident.”

This is the world of thermal fluctuations in which DNA lives, constantly bouncing against water molecules and everything else around it. And these aren’t soft bounces. At the size of proteins, “atoms are veritable bullets,” Wiggins says. “Enzymes are being hit by things going hundreds of meters per second.”

This is hugely important in biology. “Proteins are continually buffeted by thermal fluctuations, and favorable fluctuations are captured for standard biological functions. But how can you make predictions about the behavior of something that’s being pelted all the time by bullet-speed molecules?”

Fortunately, all these objects are at the same temperature, and “everything is so well-sampled that you can use some powerful mathematical tools that work with random events,” he says. Physicists can examine a molecule, analyze the ways in which it is free to be flexible, and create useful predictions of how these thermal fluctuations will push it around.