DNA can be more than just a double helix. When four strands are folded into a compact bundle, they form a G-quadruplex (G4). These bundles form in guanine-rich regions across the genome, including telomeres at the ends of chromosomes and the control regions of genes, where the proteins that dock onto them help decide which genes are switched on. G4 does not act the same way everywhere. What it does depends on where it forms and which proteins bind to it.

This is why understanding the interactions surrounding G4 is so important.

Biochemical assays give us more information about the intricacies of these interactions. These assays assess the way DNA and proteins interact, which occurs in processes such as DNA replication, transcription, and repair. However, previous assays haven't been able to capture the full picture.

A team led by Kazumitsu Onizuka and Shinichi Sato of Tohoku University, with Takanori Oyoshi of Shizuoka University, has developed a photocatalytic proximity labeling method that overcomes previous pitfalls and captures partners that earlier methods missed. Furthermore, the technique revealed hexokinase-1 to be an unexpected player in these interactions.

"The problem is that identifying the proteins that bind G4 is not at all straightforward," explains Sato. "Conventional bait-and-capture methods miss weak or transient binders. Newer chemical probes also have downsides, as they bind into the quadruplex themselves and block some of the proteins they are meant to catch."

The researchers tested their new method, attaching a photocatalyst (which reacts to light) directly to a human telomere G4 and adding a small tagging reagent. Thirty seconds of blue light generated singlet oxygen, a reactive form of oxygen that survives only a few nanometers, so only proteins sitting immediately beside the structure received a chemical tag that could later be identified by mass spectrometry.

The new method flagged more than a thousand candidate binders, but the one that ranked first was unexpected. It was hexokinase-1 (HK1) - an enzyme that carries out the first step of glucose metabolism and normally works at the outer membrane of mitochondria.

"This finding was a genuine surprise," remarks Onizuka. "HK1 is a textbook metabolic enzyme usually associated with glycolysis, and as far as we know, no direct interaction with a folded nucleic acid had been reported for it before."

The binding was confirmed with purified protein by two independent assays. Electrophoretic mobility shift assays showed that HK1 forms a complex with G4 DNA, but not with double-stranded DNA, and microscale thermophoresis confirmed that the interaction is a tight one. It is an exciting finding that suggests that metabolism and gene regulation, which are usually studied separately, may be connected more directly than we assumed.

While the biological relationship between HK1 and G4 is still unclear, the team is excited to tackle this question next, as it may help us better understand cancer and metabolic disease.

"The method we developed is not just a one-off," says Ahmed Mostafa Abdelhady, who worked extensively on this project when he was a doctoral student at Tohoku University. "It's more like a general platform that could work for other nucleic acid structures. We are excited to test other candidates as well."

The findings were published in Communications Chemistry on August 27, 2026.