By Sherri Chasin Calvo
Drawing on electron-beam lithography techniques from the semiconductor industry, a multinational team led by Princeton University physicist Robert Austin is developing an instrument to make high-resolution DNA maps in a tenth of a second.
As algorithms become increasingly robust, the limiting factor for DNA mappers has become spatial resolution. Ordinary microscopes can get down to 0.3 micrometers. And powerful near-field optical and atomic force microscopes can see down to the molecular level. But as anybody who has used a camera knows, more zoom means a smaller field of view. In other words, to get more detail you have to look at a smaller piece of the picture, which in turn means longer scanning times to complete the map.
But using microfluidic channels, the researchers have built a device that detects spatial resolutions as fine as 200 nanometers without sacrificing time.
Instead of running the DNA through gels, Austin etched micron-sized channels on a silicon chip, forming an obstacle course that traps the molecules and keeps them in the field of view.
A near-field scanner then maps the molecules as they pass over a narrow slit, using laser light to activate fluorescent tags and generate a signal. With ordinary microscopes the resolution is limited to the wavelength of light. With the near-field scanner, however, the resolution is only limited to the width of the slit. The researchers use three slits in each experiment to avoid errors from background noise, discarding signals that do not appear at all three.
The team’s ultimate goal, says Austin, is to develop an instrument that can look at all the DNA from a single cell at a particular stage in its life cycle.
In addition to the mapping instrument, the team has also adapted their microfluidic technology to speed up the separation of DNA fragments by size. For example, says collaborator Edward Cox, a Princeton molecular biologist, they have shortened the process of separating BAC and PAC inserts from 48 hours to less than 10 seconds.
Now Austin and his cohorts are trying to figure out how to keep the DNA molecules stretched out long enough to get a useful scan. They tried placing tiny posts along the path to help extend the molecules, but the molecules still bunched up as they meandered through the channels. To keep them in line, the researchers are now working to shrink the microfluidic etchings a hundred-fold. At 50 nm the channels would be too narrow for the taut DNA to relax before it passes over the slits. The technology to create such refined channels exists, says Austin.
Another factor is the speed of the molecules as they move through the channels, measured by the time they arrive at each of the three slits. The faster they move along, the less time they have to clump up.
The group recently got a $1 million annual DARPA grant, renewable for three to five years, to further develop the DNA mapping instrument. Austin plans to use some of that money to recruit electrical engineers and computer scientists who can connect the microfluidic device to the macroscopic world. “We understand the basic physics,” he says. “Now we’re trying to get it to the point where we can get it into the lab.”