EDITOR'S CHOICE IN STRUCTURAL BIOLOGY
In 1962, researchers at the National Institutes of Health identified peculiar twists of DNA shaped into four-stranded structures, rather than the double helix that had come to define DNA. For much of the 50 years since the discovery of these structures, now known as G-quadruplexes, “it was felt that those findings were a laboratory curiosity, an artifact if you will,” says Stephen Neidle of University College London. Still, researchers were intrigued by these test-tube structures because they were made exclusively from guanines and were stable at physiological conditions. Yet evidence for their existence in human cells remained elusive. “It’s almost become more religion than science,” says Steve Jackson of the University of Cambridge. “Some believed in them, some didn’t.”
To end the debate, Jackson’s lab teamed up with the lab of Shankar Balasubramanian, also at Cambridge. They used a small molecule called pyridostatin, which binds to G-quadruplexes in vitro, to try to ferret out these structures in human cells, and found that, like other small molecules that bind quadruplexes in vitro, pyridostatin induces a DNA damage response. The team took advantage of this response by exposing cells to pyridostatin and cross-linking the DNA to a damage-response protein, a histone called ?H2AX. After zeroing in on the genomic foci of this damage response, the group used high-throughput sequencing to determine which genes pyridostatin had targeted and determined that they were indeed regions with a high tendency toward G-quadruplex formation. “It shows that G-quadruplexes really [do exist] in human cells in culture,” Jackson says.
The findings are a triumph for those who had believed that G-quadruplexes exist in vivo. Pyridostatin doesn’t induce G-quadruplexes to form, Jackson points out, but binds to those that already exist.
What G-quadruplexes are doing in the genome still remains unanswered. “I think probably in some cases G-quadruplexes are problems that need to be resolved by the cell,” says Jackson. For instance, others have reported that in yeast it appears that the helicase Pif1 unwinds G-quadruplex structures to maintain genomic stability. Jackson’s group also found overlap between pyridostatin damage and Pif1 targets.
“I think in other cases, the idea that they can have positive functions is very appealing,” says Jackson. Given that telomeres can form G-quadruplexes, it’s possible that the structures are involved in facilitating telomeres’ unique structure or preventing them from being recognized as broken bits of DNA, Jackson speculates. Or perhaps G-quadruplexes are involved in regulating transcription, since they also form in promoter regions, making them possible targets for small-molecule therapies to arrest cancer’s cell cycle. Although G-quadruplex research has been conducted for half a century, Jackson says, “it’s still early days.”
R. Rodriguez et al., “Small-molecule–induced DNA damage identifies alternative DNA structures in human genes,” Nat Chem Biol, 8:301-10, 2012.