Ned Seeman sees DNA as his own personal nanoscale erector set. Last summer, the chemistry professor’s group at New York University figured out how to make DNA assemble itself into a three-dimensional crystal. The results were published in Nature in September.
Seeman’s self-assembling 3-D DNA could change the way people think about making crystals, which chemists use to determine the structures of new molecules they’ve made. Growing high quality crystals has historically been more art than science. The key to a beautiful crystal is order. After all, a crystal is nothing more than an ordered arrangement of atoms or molecules. But getting atoms to behave themselves is no easy task. Conditions in the laboratory have to be perfect to grow a crystal with flawlessly ordered structure. Oftentimes, crystallographers, the scientists who grow crystals, spend painstaking hours trying to make the perfect crystal without any luck.
Seeman has changed all of that. Starting with tiny structural DNA building blocks called tensegrity triangles, he and his team got these structures to self-assemble into a rhombohedral crystal—a six-sided crystal a fraction of a millimeter in length. Unlike other crystallization techniques, their results were not based on luck.
“This is the first example of non-trial-and-error crystallization,” Seeman said during an interview in his office overlooking Washington Square Park. Sitting amidst a glut of books, papers and homemade DNA models, Seeman explained his lab’s accomplishment with a baseball analogy. He noted that if only 26 things go right on the ball field when you need 27, it makes the difference between a home run and a ground out. “The same is true of crystallization,” he said. “We designed a crystal from scratch and got it to self-assemble into a 3-D structure.” Essentially, he and his colleagues figured out how to control conditions such that they get a home run every time they come up to bat.
“This is a very visionary way of thinking,” said Helen Berman, director of the Protein Data Bank, an international catalog that documents new molecular structures. “[Seeman] had a vision, he persisted, and he was the first to engineer materials in this way, using DNA,” she said. Berman, who is a chemist at Rutgers and a long-time friend of Seeman, was surprised to learn of her friend’s invention. She encountered the new crystal when a researcher on her team was trying to enter the self-assembled 3-D structure into the Protein Data Bank. Berman and her team didn’t quite know how to categorize the crystal because it was unlike anything they’d seen before. “It’s the first of its type, so we had to figure out the right way to represent it so it’s meaningful biologically and chemically.”
Seeman built his first artificial DNA structure, a stick-cube, in the early 1990s. Since then, the relatively new field of structural DNA nanotechnology, which he helped invent, has grown rapidly. Seeman and other researchers in the field have used bits and pieces of genetic material to build a variety of flat structures, 3-D crystals, and even nano-sized robots.
“Until about 2000, no other labs were doing what we’re doing. Now there are over fifty,” Seeman said. The spread of the technology is what he’s most excited about. As he wrote in a feature article for Nature in 2003, “The exploitation of DNA for material purposes presents a new chapter in the history of the molecule.”