In a paper that has been accepted for publication in the Journal of the American Chemical Society, and posted on its Web site, Silverman and graduate student Chandrasekhar Miduturu begin with a piece of unfolded RNA. Through specific chemical reactions, they attach two strands of DNA, each resembling one side of a ladder. The two DNA strands spontaneously bind together, then the researchers add magnesium ions to initiate folding of the RNA.
"Folding of the RNA structure competes with formation of the DNA constraint until a chemical balance is reached," Silverman said. "In some cases, the DNA is like a barnacle, just stuck onto the RNA without perturbing its structure. In other cases, the DNA changes the RNA structure. We can predict which situation will occur based on the shape of the RNA and on the attachment points of the DNA constraint."
In cases where the normal RNA shape and the DNA constraint cannot co-exist simultaneously, the balance between competing RNA and DNA structures is controlled by the concentration of magnesium ions, Silverman said.
In work not yet published, the researchers have also shown that the effects of the DNA constraint on the RNA structure can be modulated by external stimuli such as DNA oligonucleotide strands, protein enzymes and chemical reagents.
While Silverman and Miduturu are currently using RNA as a proof of principle for their DNA constraint studies, they also plan to use the new technique to more effectively study the folding process of RNA. Because they can control RNA structure precisely, they could generate and examine biologically relevant folded and misfolded RNAs. They could also hook the DNA constraints to other molecules, including non-biological macromolecules, to control their folding.
Importantly, the process of manipulating macromolecules with DNA constraints can be either reversible or irreversible, depending on which chemical trigger is used. Like a switch, a particular molecular shape could be turned on and off.
"Another key aspect of DNA constraints is their programmability," Silverman said. "By placing two or more constraints on one molecule, we could generate multiple molecular states that would be programmable by DNA sequence. In other efforts, we would like to control macroscopic assembly processes by influencing the shapes of self-assembling molecular components."
The David and Lucile Packard Foundation and the University of Illinois funded the work.
Journal of the American Chemical Society