Researchers Unravel DNA Tangles And Enzyme Seamstresses

Untangling the heaps of DNA strings during cell division is the job of special enzymes called topoisomerases. How they achieve this feat may be simpler than previously thought, says U of T research. In a study published in the current online issue of Biophysical Journal, researchers used computer simulations to mimic the DNA mess inside the nucleus with a series of billions of linked and unlinked loops. Their calculations indicated that whether DNA molecules are interlinked is shown by the way they touch each other. Interlinked DNA loops tend to touch in an easily recognizable hook-like way, fitting together perfectly; whereas strands of unlinked DNA molecules tend to curve away when they touch each other. The findings could have implications in designing new drugs to treat cancer and infectious diseases as uncontrolled DNA linking and tangling often result in cell death.

"The exciting part is that these seemingly abstract physical principles we work with can be useful some day to tie up DNA in cancer cells and kill them off," says U of T biochemist Hue Sun Chan, the study's co-author and a Canada Research Chair in proteomics, bioinformatics and functional genomics.The study is co-authored by Professor Lynn Zechiedrich of Baylor College of Medicine in Houston, Texas, who proposed the notion in an earlier conceptual report, and the lead author, Zhirong Liu, is a U of T postdoctoral fellow in Chan's research group. "These are the same general principles that can be applied to other areas of science and engineering to address various entanglement problems," Chan says.

The curved distinctions between DNA strands may allow the seamstresses of the process -- the topoisomerases -- to identify linked DNA loops, cut a strand apart, let another strand pass through, and then reconnect the cut strand so that DNA can separate into untangled lengths that are the chromosomes: the topoisomerases only have to cut and reconnect at hooked-like but not other touching points. If this process were disrupted, however, the cell would be in serious trouble. "One link could keep the cell from dividing. Two links are even more lethal," says Zechiedrich, who notes that "the results of these computer simulations are very striking."

While Chan and his collaborators stress their results represent only a quantitative proof of concept, they do see the finding as particularly relevant for understanding diseases such as cancer, where cell multiplication goes haywire. "Besides further elucidating the principles we found, what needs to be done now is to test these findings experimentally and ultimately apply them to real-life cell division and target the developmental processes that lead to disease."

The study was funded by the Canadian Institutes of Health Research, the Canada Research Chairs Program, the National Science Foundation, the National Institutes of Health and the Burroughs Wellcome Fund.