海角大神

Nadrian Seeman sees a future filled with extremely small factory workers.

By small, the New York University chemistry professor means a billionth of a meter.

That鈥檚 the scale that he and others who work in the field of nanotechnology deal with on an everyday basis. By manipulating molecules, they attempt to build new materials and microscopic robots, possibly small and smart enough to move through human bodies.

But nanotechnology鈥檚 key obstacle has always been how to mass-produce these exotic molecules used as building blocks.

At this early stage, nanoscale manufacturing mistakes are pretty common. It would be as though a factory churned out cars where the rearview mirror was attached to the hood 鈥� and did so a third of the time.

But Professor Seeman has found a way around that. He and a Chinese team at Nanjing University have built a nanoscale factory worker. The tiny machine is made of DNA, the molecule that governs the way cells make proteins. But this DNA isn鈥檛 like that found in human cells 鈥� it鈥檚 synthetic and can鈥檛 reproduce by itself.

But like all DNA, it holds information in the form of genetic code. Seeman 鈥減rograms鈥� his tiny machine by stringing the right combinations of DNA 鈥� much in the way computer engineers use binary code.

鈥淲e鈥檙e prototyping the notion of programmable patterns,鈥� Seeman says.

The machine has two 鈥渁rms鈥� made of strings of DNA that are attached by another chain of DNA.

Each arm has a molecule on the end that attaches to other molecules and aligns them in a set order. These sticky ends only connect with a particular building block, and Seeman can program them to specify which molecule he wants.

This allows him to arrange pieces and form specific molecules with some precision 鈥� similar to the way a robotic automobile factory can be told what kind of car to make.

After the arms create the desired molecules one at a time, the whole mixture is heated and cooled, which causes the correct molecules to displace the incorrect ones.

This fixes any errors and is what makes the 鈥渇actory鈥� reliable enough to mass-produce.

Thus far, Seeman鈥檚 team has made molecules with various shapes 鈥� squares and triangles 鈥� that don鈥檛 have a specific function. The next step is constructing functional molecules 鈥� but he is mum on the details.

He imagines building several tiny DNA machines and programming them to work in harmony, creating more complex substances such as a fiber or even an electronic device.

Seeman has been working on nanorobotics for several years. He first perfected a one-armed version in 2006. It was the first time anyone had put together such a device in a DNA array, he says.

Now, that he鈥檚 pulled off a two-arm design, Seeman says that his team can finally build things.

The big leap in Seeman鈥檚 work is the ability to 鈥渞emote control鈥� the DNA arms, says Milan Stojanovic, a professor of medicine at Columbia University and director of the National Science Foundation鈥檚 Center for Molecular Cybernetics. Finally, his team can set up a protocol to fix errors along the way.

鈥淭hey start with the same basic structure, but then can 鈥榖uild鈥� on this basic structure depending [on] what they have in solution,鈥� Dr. Stojanovic says in an e-mail. Being able to get high yields is also important for making future progress more rapidly.

Seeman鈥檚 work is rather unique, Stojanovic says. 鈥淣ed is unpredictably creative in his approach to science,鈥� he says.

Another key contribution to future nanotechnology work is how Seeman manipulates the strand of DNA that connects the arms, says Paul Rothemund, a research associate at the California Institute of Technology in Pasadena.

Seeman borrowed a technique called 鈥淒NA origami鈥� to act like a pegboard for the arms, which can be reconfigured. DNA is expensive, Mr. Rothemund explains. So, Seeman鈥檚 adaptive, elastic system uses fewer pieces of DNA to make a given molecular configuration. Rothemund likens it to having a set of clamps or vice grips that could be rotated to hold differently shaped objects, rather than using a whole new set of tools for every project.

Chengde Mao, an associate professor at Purdue University, is particular hopeful that Seeman鈥檚 work 鈥� and that of other nanotech scientists 鈥� could allow for breakthroughs in DNA, building three dimensional structures.

Seeman credits an art print by mathematical illusionist M.C. Escher for starting him on the road to using DNA as a way into nanotechnology in the early 1980s.

After trying unsuccessfully to grow crystals for experiments, he spoke with a colleague who was doing work in recombinant DNA, a brand-new field at the time. Seeman was 鈥渢hinking about the Escher print 鈥楧epth,鈥� 鈥� where six-finned fish float through space, he says. 鈥淚 started thinking about a six-armed, three-dimensional junction.鈥�

He began picturing how to shape DNA that way. At the time, nanotechnology was in its infancy, and most chemists were working with inorganic molecules. Seeman, however, decided that DNA was a better way to start because it has a built-in structure.