How DNA Goes A'Courtin'
Simplified model catches essential details of how DNA complements find their matches
Until now, scientists have known little about how complementary single strands of DNA court one another before binding to form the classical double helix. But now, molecular dynamics simulations have identified that the binding—or hybridization—mechanism depends largely on the sequence of the DNA: Ordered sequences will meet and then slither lengthwise to find the correct match; but sequences that are random will connect at key sites then rapidly assemble along the molecule’s length.
“One would have thought that random sequences would have more difficultly hybridizing, and that is not necessarily the case,” says Juan J. de Pablo, PhD, professor of chemical and biological engineering at University of Wisconsin, Madison. The work was published in the October 5 issue of the Proceedings of the National Academy of Sciences.
Scientists have previously tried to simulate the pathways by which DNA strands combine, but the models they used included too much detail to enable sufficiently long computations, de Pablo says. So De Pablo’s group developed a highly simplified model, tested on experimental data, to capture essential details of the interactions between the base pairs of complementary strands of DNA. The researchers then simulated the process by which the single strands interact using molecular dynamics and Monte Carlo simulations, taking multiple “snapshots” of the double helix as it assembled. To the team’s surprise, the path to a successful union depended crucially on the sequences of the molecules.
When the sequences of both single strands are ordered or repetitive, any two sites of base pairs can come together and the two strands slowly “slither” lengthwise until complementary base pairs match along the entire chain, says de Pablo. When the sequences are random, however, single sites located toward the center of the strands unite early. “The moment they come together, then the molecule just assembles perfectly and it does so very quickly,” de Pablo says.
The results could influence the design of technologies that depend on the hybridization process, such as gene chips, de Pablo says. To engineer more efficient and reliable hybridization, researchers could use random sequences, which bind more efficiently and with fewer errors.
“This is an interesting step forward,” says Nadrian Seeman, PhD, professor of chemistry at New York University. “No one had taken the time to track the pathway previously.” Seeman has used the principle of random sequencing in his own hybridization studies, and he finds it reassuring to see it vindicated by the simulation data. “It does tell people who are designing sequences to avoid repetition in the sequences,” he says.