What forces DNA molecules through tight spaces? Not elasticity but disorder, Cornell researchers discover

The fluidic device consists of a quasi-two-dimensional gap between a floor and ceiling approximately 60 nm in height. Some regions of the device are populated with nanopillars.

A new understanding of how large biological molecules behave in tiny spaces could lead to a method for separating DNA strands by length. It also could throw light on the way molecules move in living cells.

Using a forest of nanofabricated pillars so small that DNA molecules can only slip through lengthwise, Cornell University researchers have demonstrated the existence of an entropic recoil force that causes the molecules to move from a tight space into a more open one.

The findings, published in Physical Review Letters (March 25, 2002), are by Stephen Turner, a postdoctoral research assistant at Cornell; graduate student Mario Cabodi; and Harold Craighead, the Charles W. Lake Jr. Professor of Engineering, professor of applied and engineering physics and interim dean of the Cornell College of Engineering.

This work follows previous advances by Turner, Craighead and others in the same field that shed new light on how DNA molecules are inserted into confined spaces. Now they are the first to demonstrate how DNA strands are ejected from confined spaces.

In water, strands of DNA or other long-chain molecules tend to coil into a roughly spherical shape. Previously the researchers found that when a DNA molecule in a spherical configuration comes up against an opening too small for the sphere to pass through, some small part of the chain is first pulled into the opening, causing the rest to uncoil and follow.

In these experiments the DNA molecules are pulled into the dense array of pillars by an electric field. If the field is removed before a molecule is all the way in, it will recoil back into the open space and resume its spherical shape. What is the force that causes this behavior? Physicists have theorized that it is an entropic force related to the confinement of the molecule in a narrow tube. Entropy is a measure of the amount of disorder in a system, and an entropic force would tend to move things toward the most disorderly arrangement. In this case, Turner explains, that would be the one in which the molecule can assume many different configurations -- that is, free in water -- rather than the one in which it is confined in a narrow tube.

(a) DNA molecules are driven into the dense-pillar region, which occupies the bottom 80% of each frame. Six or seven molecules are shown, one (on the left) having been completely lodged in the unfavorable region. Occasionally molecules enter the dense-pillar region by herniation. However, these hernias quickly unfold in the electric field once an end of the molecule has been incorporated. The scale bar represents 5 mm. (b) Entropic recoil at various stages ending when all the molecules have recoiled except the leftmost. Without the interaction between the two regions, this molecule experiences no center-of-mass motion and only slight contraction. This effect shows the contribution from entropic elasticity is negligible and immediately suggests a method for separation of long-strand polymers in accordance to length.

In the new experiment, the researchers used electron beam lithography equipment at the Cornell Nanofabrication Facility to build a device consisting of a flat open space next to a forest of tiny pillars. Each pillar is about 35 nanometers (nm) in diameter, with the pillars spaced 125nm apart. (A nanometer is one-billionth of a meter, or three times the diameter of a silicon atom.) The device is made of silicon nitride, which is transparent to visible light. The DNA molecules themselves are too small to be seen by visible light, but they are stained in a way to make them fluorescent so that the light they give off can be observed.

An electric field was applied to the ends of the experimental device, pulling the DNA molecules toward the pillared region. The field was applied in short pulses so that the molecules were first driven into the pillared region and than allowed to recoil. Molecules that had moved entirely into the pillared region did not recoil.

Analysis of videomicrographs showed that the recoil was not elastic, the researchers say. "Elastic recoil is initially rapid followed by a gradual slowing," they write in their paper. "Here, the recoil is initially slow and gradually increases in speed."

The recoil happens, Turner explains, because atoms within the chain molecules are always in motion, always colliding with water molecules and the surrounding pillars. Inside the pillared space, he says, these collisions happen in all directions, tending to cancel each other out. But at the interface between the pillars and the open space, the links in the chain just outside the pillared space can only collide with the pillars in one direction, and the reaction to these collisions exerts a force that tends to pull the chain back out.

"What we've seen here is a new way in which disorder can force something to move," Turner says.

From the geometry of the system and estimates of the drag exerted on the molecules by water, the researchers estimate the minimum entropic force at 5.7 femtoNewtons. (A femtoNewton is one-quadrillionth of the force it takes to support the weight of a medium-sized apple.) These conclusions should apply to all long-chain molecules, or polymers, the researchers say. And they suggest that entropic forces might play a role in the movement of such molecules in living cells. Since molecules that have moved all the way into confinement do not recoil, they say, this method could be used to separate molecules by length.

The paper in Physical Review Letters is titled "Confinement-Induced Entropic Recoil of Single DNA Molecules in a Nanofluidic Structure."

Media Contact

Media Relations Office