A tug-of-war at the DNA: how Huntington’s repeats grow and shrink

Ever since large genetic studies in Huntington’s disease (HD) revealed that the longer the CAG expansion, the earlier symptoms appear, we’ve known that repeat length matters. Recent work has highlighted just how that repeat length increases within vulnerable brain cells — from about 50 CAGs to over a thousand. 

Understanding how these expansions happen, and how they influence the disease, is crucial for developing the right therapeutic strategies. Can we correct the expanded DNA in affected cells? Well… maybe the cells can do it themselves!

The players in a molecular tug-of-war

DNA-repair genes can strongly influence when HD symptoms begin. For years, researchers have been asking: what do these genes actually do to the faulty stretch of DNA that causes HD? And can we harness this knowledge to delay symptom onset — perhaps long enough that the disease never develops?

A new study in Nature Communications from Petr Cejka’s group reconstructs, molecule by molecule, how two opposing DNA-repair teams compete inside our cells. One team mistakenly lengthens the CAG repeat when trying to fix it, while another trims it back. This elegant biochemical dissection finally shows why players such as MSH3, MLH3, and FAN1 have such a strong impact on disease onset — and opens new routes to slow or even prevent HD.

Why DNA repair matters in HD

DNA is stored in the nucleus forming a double helix, the letters on each strand pair precisely, like the matching teeth of a zipper.

But when the CAG sequence in the huntingtin (HTT) gene becomes too long, the strands no longer line up perfectly. One side can end up with extra “teeth,” creating a mismatch that bulges out from the helix — what scientists call an extrusion loop (imagine a zipper with a kink on one side!).

Everyone inherits some CAG repeats in their HTT gene, but in general, people with 40 or more eventually develop the disease. When these repeats get longer, the DNA can’t zip up neatly anymore, and the cell’s repair machinery rushes in to fix it. And here is where the tug-of-war game starts. Repair can go two ways: some machinery complexes smooth things out and stabilize the DNA, while others accidentally make the repeat longer and longer.

The “expansion crew”: MutSβ and MutLγ

DNA repair usually acts like a spell-checker, scanning for errors, mismatches, or small loops that appear when our DNA is copied. In HD, however, part of the repair team creates the problem.

It’s a literal tug-of-war between two DNA-repair pathways acting on the same repeat. Which side wins likely determines whether CAGs grow or shrink in a given cell.

Two complexes — MutSβ (made of MSH2 + MSH3) and MutLγ (MLH1 + MLH3) — recognize the extrusion loop that forms when there are lots of CAG repeats. Instead of removing the loop, the expansion crew uses the loop as a template and fills in extra CAGs.The result? The repeat grows even longer. MutSβ and MutLγ turn a normal repair job into a “copy-and-paste” mistake that expands the CAG number.

The “contraction crew”: FAN1 to the rescue

Enter FAN1, a nuclease — essentially a pair of molecular scissors — that can do the opposite. FAN1 recognizes these DNA loops and cuts them directly at the site of the problem. Working with helper proteins, the FAN1 crew removes extra repeats instead of adding new ones.

FAN1 also has a clever second trick: it physically blocks MutLγ from partnering with MutSβ, stopping the expansion machinery before it even starts.

A molecular tug-of-war

By setting up both reactions side by side in a test tube, the team revealed a literal tug-of-war between two DNA-repair pathways acting on the same HTT repeat. Which side wins likely determines whether CAGs grow or shrink in a given cell.

Connecting biochemistry to human genetics

The discovery that DNA repair genes affect when symptoms appear didn’t come out of the blue — it started with genome-wide association studies (GWAS) enabled by donated DNA samples from thousands of people with HD. These large-scale studies searched the entire genome for genetic variations that modify the age of onset. The clear message was that genes involved in DNA repair — like MSH3, MLH3, and FAN1 — are major players.

This new biochemical model beautifully explains why those GWAS signals point to repair genes. Variants that boost MutSβ or MutLγ activity (in MSH3 or MLH3) speed up CAG expansion and lead to earlier symptoms, while variants that enhance FAN1 activity can slow expansion and delay onset.

Scientists had long seen these correlations — now, thanks to the Cejka team’s molecular reconstruction, we can finally connect the dots between human genetics and the actual DNA chemistry that could be driving Huntington’s disease.

Why this matters

Understanding these precise mechanisms isn’t just fascinating biology — it’s a roadmap for how we could develop therapies. If we can tilt the balance toward contraction or stabilization, we might slow or even halt the disease process itself.

Some companies are already pursuing this idea:

• ASOs targeting MSH3 or inhibitors of MutSβ aim to reduce expansion activity are being developed by Ionis Pharmaceutical, LoQus23 Therapeutics and Pfizer
• Harness Therapeutics is trying to boost FAN1 function, or mimicking its blocking effect on MutLγ, could offer another route to protect HTT from runaway expansion

What’s next?

Although strong evidence suggests that somatic repeat expansion drives when symptoms begin, this remains a working model. Researchers are now trying to map how these repair processes differ across brain cell types and how they interact within living tissue.

Learning how cells naturally correct their own DNA errors could inspire treatments that let them fix Huntington’s disease from within.

The key challenge is balance: the same DNA-repair systems that sometimes lengthen the HTT repeat also protect the rest of our genome. The ultimate goal will be to fine-tune these pathways to suppress CAG expansion without compromising DNA integrity elsewhere.

If this model holds true, it could open an entirely new therapeutic avenue — targeting DNA repair itself to delay or even prevent Huntington’s disease.

Summary

• HD onset is strongly influenced by genes involved in DNA repair.
• MutSβ (MSH2–MSH3) and MutLγ (MLH1–MLH3) cooperate to nick CAG DNA, adding extra repeats.
• FAN1 and its crew cut the CAG loop instead, removing excess repeats. FAN1 also blocks the MutSβ–MutLγ partnership, preventing expansions.
• These opposing reactions explain why enhancing FAN1 or reducing MLH3/MSH3 activity could delay HD onset.

Learn more

Mechanism of trinucleotide repeat expansion by MutSβ–MutLγ and contraction by FAN1.