Researchers map 50,000 of DNA's mysterious 'knots' in the human genome

Innovative study of DNA's hidden structures may open up new approaches for treatment and diagnosis of diseases, including cancer.

DNA is well-known for its double helix shape. But the human genome also contains more than 50,000 unusual 'knot'-like DNA structures called i-motifs, researchers at the Garvan Institute of Medical Research have discovered.

Published today in The EMBO Journalis the first comprehensive map of these unique DNA structures, shedding light on their potential roles in gene regulation involved in disease.

In a landmark 2018 study, Garvan scientists were the first to directly visualise i-motifs inside living human cells using a new antibody tool they developed to recognise and attach to i-motifs. The current research builds on those findings by deploying this antibody to identify i-motif locations across the entire genome.

"In this study, we mapped more than 50,000 i-motif sites in the human genome that occur in all three of the cell types we examined," says senior author Professor Daniel Christ, Head of the Antibody Therapeutics Lab and Director of the Centre for Targeted Therapy at Garvan. "That's a remarkably high number for a DNA structure whose existence in cells was once considered controversial. Our findings confirm that i-motifs are not just laboratory curiosities but widespread -- and likely to play key roles in genomic function."

Curious DNA i-motifs could play a dynamic role in gene activity

I-motifs are DNA structures that differ from the iconic double helix shape. They form when stretches of cytosine letters on the same DNA strand pair with each other, creating a four-stranded, twisted structure protruding from the double helix.

The researchers found that i-motifs are not randomly scattered but concentrated in key functional areas of the genome, including regions that control gene activity.

"We discovered that i-motifs are associated with genes that are highly active during specific times in the cell cycle. This suggests they play a dynamic role in regulating gene activity," says Cristian David Peña Martinez, a research officer in the Antibody Therapeutics Lab and first author of the study.

"We also found that i-motifs form in the promoter region of oncogenes, for instance the MYC oncogene, which encodes one of cancer's most notorious 'undruggable' targets. This presents an exciting opportunity to target disease-linked genes through the i-motif structure," he says.

I-motifs hold promise for new type of therapies and diagnostics

"The widespread presence of i-motifs near these 'holy grail' sequences involved in hard-to-treat cancers opens up new possibilities for new diagnostic and therapeutic approaches. It might be possible to design drugs that target i-motifs to influence gene expression, which could expand current treatment options," says Associate Professor Sarah Kummerfeld, Chief Scientific Officer at Garvan and co-author of the study.

Professor Christ adds that mapping i-motifs was only possible thanks to Garvan's world-leading expertise in antibody development and genomics. "This study is an example of how fundamental research and technological innovation can come together to make paradigm-shifting discoveries," he says.

Read more at Science Daily

High-throughput experiments might ensure a better diagnosis of hereditary diseases

Researchers at the Department of Biology, University of Copenhagen, have now contributed to solving this problem for a specific gene called GCK. The study has just been published in Genome Biology.

Figure: GCK gene

Rasmus Hartmann-Petersen, Professor at the Department of Biology, explains:
- “The GCK gene, which codes for the enzyme glucokinase, regulates the secretion of insulin in the pancreas. GCK gene variants can therefore cause a form of hereditary diabetes. Although the connection between GCKand diabetes has been known for several years, we have, until now, only known the effect of a few percent of the possible variants of this gene”.

Together with colleagues at the PRISM centre, UCPH, who are currently studying the effects of genetic variations, the researchers measured the effect of all of the possible variants of GCK.

PhD student Sarah Gersing, who is the first author of the article, explains:
- “We used yeast cells to measure the activity of over 9000 different GCK variants. In this way, we were able to generate a list of the effects — both of already known variants, but also of variants that patients might carry, but that have not yet been discovered. This provides us with a reference for future GCK diagnostics”.

Prof. Kresten Lindorff-Larsen, who heads the PRISM centre, continues:
- “Our results are quite unique; not only have we measured the effect of several thousand variants, but for many of the variants, we can now explain what they do to the glucokinase protein. In our centre, we have gathered researchers working across a range of research fields, bridging from data analysis and biophysics to cell biology and medicine, and it is now clear how this broad approach pays off in explaining how diseases arise”.

Gene variants of GCK can, among other things, cause a form of hereditary diabetes called "GCK maturity onset diabetes of the young" (GCK-MODY).

Professor of genetics, dr. med. Torben Hansen, who is also a member of the PRISM centre, says: - "Although GCK-MODY patients exhibit elevated blood glucose levels, this is often not associated with complications. Hence, unlike other forms of diabetes, most GCK-MODY patients might therefore not need to be treated with medication. However, due to missing or inaccurate genetic data, more than half of the GCK-MODY patients are classified with having either type 1 or type 2 diabetes – and are therefore unnecessarily medicated. We estimate that approx. 1% of those who have recently been diagnosed with type 2 diabetes in Denmark have a variant in the GCK gene, meaning that they don’t need treatment, or need to be treated differently. Our new map of GCK variants can hopefully help give these patients a more correct diagnosis.”

The next step for PRISM is to transfer these methods to other genes and diseases.
- "We are already well underway with genes involved in e.g., neurodegenerative diseases, and we are trying to develop precise methods that can provide us with insights on disease mechanisms", says Rasmus Hartmann-Petersen.

 Kresten Lindorff-Larsen continues:
- "Our data gives us the opportunity to test and develop computational models for variant effects, which will then be transferable to other genes and diseases."

Read more at Science Daily