When repetita no-longer iuvant: somatic instability of the CAG triplet in Huntington's disease

Abstract

Trinucleotide repeats in DNA exhibit a dual nature due to their inherent instability. While their rapid expansion can diversify gene expression during evolution, exceeding a certain threshold can lead to diseases such as Huntington’s disease (HD), a neurodegenerative condition, triggered by >36 C–A–G repeats in exon 1 of the Huntingtin gene. Notably, the discovery of somatic instability (SI) of the tract allows these mutations, inherited from an affected parent, to further expand throughout the patient’s lifetime, resulting in a mosaic brain with specific neurons exhibiting variable and often extreme CAG lengths, ultimately leading to their death. Genome-wide association studies have identified genetic variants—both cis and trans, including mismatch repair modifiers—that modulate SI, as shown in blood cells, and influence HD’s age of onset. This review will explore the evidence for SI in HD and its role in disease pathogenesis, as well as the therapeutic implications of these findings. We conclude by emphasizing the urgent need for reliable methods to quantify SI for diagnostic and prognostic purposes.

Full text

Nucleic Acids Research , 2025, 53 , gkae1204
https://doi.org/10.1093/nar/gkae1204
Advance access publication date: 14 December 2024
Critical Reviews and Perspectives
When repetita no-longer iuvant: somatic instability of the
CAG triplet in Huntington’s disease
Elena Cattaneo
1 , 2 ,
* , Da vide Scalz o
1 , 2
, Martina Zobel
1 , 2
, Raffaele Iennaco
1 , 2
,
Camilla Maffezzini
1 , 2
, Dario Besusso
1 , 2 and Simone Maestri
1 , 2
1
Department of Biosciences, University of Milan, street Giovanni Celoria, 26, 20133, Milan, Italy
2
INGM, Fondazione Istituto Nazionale Genetica Molecolare ‘Romeo ed Enrica Invernizzi’, street Francesco Sforza, 35, 20122, Milan, Italy
*
To whom correspondence should be addressed. Te l: +39 02 50 32 58 42; Email: elena.cattaneo@unimi.it
Abstract
Trinucleotide repeats in DNA exhibit a dual nature due to their inherent inst abilit y. While their rapid expansion can diversify gene expression during
e v olution, e x ceeding a certain threshold can lead to diseases such as Huntington’s disease (HD), a neurodegenerative condition, triggered by > 36
C–A–G repeats in e x on 1 of the Huntingtin gene. Notably, the disco v ery of somatic inst abilit y (SI) of the tract allows these mutations, inherited
from an affected parent, to further expand throughout the patient’s lifetime, resulting in a mosaic brain with specic neurons exhibiting variable
and often extreme CAG lengths, ultimately leading to their death. Genome-wide association studies have identied genetic variants—both cis
and trans , including mismatch repair modiers—that modulate SI, as shown in blood cells, and inuence HD’s age of onset. This review will
e xplore the e vidence f or SI in HD and its role in disease pathogenesis, as well as the therapeutic implications of these ndings. We conclude
by emphasizing the urgent need for reliable methods to quantify SI for diagnostic and prognostic purposes.
Gr aphical abstr act
42CAG
42CAG
42CAG
95CAG
42CAG
46CAG
800CAG
42CAG
50CAG
150CAG
42CAG
48CAG
Introduction
The paradox of C–A–Gs: from advantage to
adversity
Tandem repeats are a signicant component of DNA, partic-
ularly in primates, comprising up to seven percent of the total
genome. Among these, triplet repeats are blocks of three base
pairs (bp) repeated many times one after the other ( 1 ), with the
cytosine–adenine–guanine (C–A–G) triplet being among the
most abundant in exons of the human genome ( 2 ). Through-
out evolution, the size of these repeats can gradually increase
within existing genes, contributing to diversication of gene
expression, regulation and function, without the need for new
genes. For instance, in humans, > 1500 tandem repeats have
been found to be specically expanded compared to non-
human primates, and this expansion has been associated with
differential isoform usage in genes containing the repeats ( 3 ).
Received: October 2, 2024. Revised: November 8, 2024. Editorial Decision: November 14, 2024. Accepted: December 2, 2024
©The Author(s) 2024. Published by Oxford University Press on behalf of Nucleic Acids Research.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License
(https: // creativecommons.org / licenses / by-nc / 4.0 / ), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the
original work is properly cited. For commercial re-use, please contact reprints@oup.com for reprints and translation rights for reprints. All other
permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact
journals.permissions@oup.com.

2 Nucleic Acids Research , 2025, Vol. 53, No. 1
This increase in repeat size is thought to facilitate faster evo-
lutionary changes by generating a broader range of pheno-
types than other types of genetic variations—such as single
nucleotide polymorphisms, which produce only two variants,
i.e. mutated or not mutated. This process may lead to new
shades of traits or behaviors that are subject to natural selec-
tion ( 4–8 ). However, in some cases, repeats in genes encounter
a paradox, shifting from being evolutionarily advantageous to
becoming pathogenic in human adults.
A notable example of the paradoxical role of tandem re-
peats is that of the CAGs in the Huntingtin ( HTT ) gene, where
expansion beyond a certain threshold leads to Huntington’s
disease (HD). In early phylogeny, the CAG sequence rst ap-
peared in the Echinodermata HTT gene with a small number
of repeats (2 CAGs). Inserted at a specic point in the gene (at
the 18th triplet, as in the human gene), the CAG sequence be-
gan to lengthen throughout evolution, increasing to 4 repeats
in amphibians, shes, reptiles and birds, 7 in mice, 8 in rats
and reaching 11–18 repeats in pigs and non-human primates
( 9 ). Only humans show a considerably higher number of CAG
repeats, with normal range variability between 9 and 35 ( 10 ).
This evidence suggests that evolution has been permissive re-
garding this tract and its elongation. Initially considered func-
tionally neutral, the CAG tract in HTT was recently shown to
be under purifying selection ( 9 ). A stepwise increase in CAG
size—from 0 to 2, to 4, to 7—was found to inuence posi-
tively neuronal parameters in cells in vitro , indicating that the
tract is functionally relevant ( 9 ). The CAG repeats in the HTT
gene of healthy individuals also shape brain structure, with
more CAGs in the normal range correlating with an increase
in grey matter ( 11 ). Furthermore, longer CAG repeats in HTT
have been associated with better cognitive outcome ( 12 ). In-
terestingly, individuals in their teens with CAG repeats in the
pathological range have been shown to exhibit improved cog-
nitive parameters during the disease-free phase ( 13 ).
These data indicate that the CAG tract in HTT is evolu-
tionarily advantageous, a nding that seems counterintuitive
for a gene associated with disease. In fact, when > 36 CAG
repeats in the rst of the 67 exons of the HTT gene are trans-
mitted through the germline, HD, a genetically dominant neu-
rodegenerative disease, will manifest ( 14 ). Each CAG triplet
in the gene is translated into glutamine (Q) in the resulting
HTT protein, with the mutant version therefore incorporating
> 36Qs ( 14 ). Symptoms typically appear in adulthood, four to
ve decades after birth, despite the gene’s presence since con-
ception, marking the characteristic age of onset (AOO). The
clinical presentation includes uncontrolled movements, cog-
nitive decline and psychiatric disturbances. Within the brain,
the medium spiny neurons (MSNs) and cortical neurons de-
generate ( 14 ). Longer inherited CAG tracts are linked to an
earlier AOO, though other genetic and environmental fac-
tors inuence this timing and the progression of the disease
( 15 ). The exact relationship between pathological CAG ex-
pansion in the HTT gene, the late onset of symptoms, and
the tissue- and cell-type-specic vulnerability in HD, remains
unclear. This question similarly applies to other adult-onset
neurodegenerative conditions, such as spinal bulbar muscu-
lar atrophy and spinocerebellar ataxias, which also stem from
CAG expansions in different genes and affect various neuron
types ( 16 ).
In this review, we will examine recent evidence showing
that, in addition to germline transmission, the pathological
CAG repeat expands further in somatic tissues, particularly
in the neurons of patients’ brains. HD and its CAG tract in
the HTT gene will serve as a case study. We will summarize
the discovery of somatic instability (SI) of the CAG, outline
the mechanisms that contribute to this phenomenon, and dis-
cuss its signicance in pathogenesis. In these studies, the pre-
cise measurement of CAG size and composition in individual
brain cells, along with the corresponding transcriptional pro-
les, has become increasingly important. These aspects are dis-
cussed in detail in the accompanying article by some of the au-
thors ( 17 ). Finally, we will review strategies aimed at reducing
SI with the goal of ghting the disease.
Neurons in the HD brain: where CAG instability
spirals out of control
Like many trinucleotide repeat disorders, the CAG repeat in
HTT is prone to expansion ( 18 , 19 ). CA G repeat expansion
is a phenomenon that can occur in human germ cells and be
passed on to the next generation. Although rare, this typically
happens during paternal transmission within the germline. It is
associated with anticipation, a phenomenon where the disease
phenotype manifests earlier in offspring than in the affected
parent. This earlier onset is due to a signicant increase in
CAG repeat length, for example, from 42 repeats in the parent
to 60 or more in the child ( 16 ). Germline instability is also
responsible for new mutations in families with no history of
HD, where an allele in the intermediate CAG repeat range (e.g.
between 27 and 35) expands into the pathological range in the
offspring ( 20 ).
Notably, CAG instability also manifests within the tissues
most affected by the disease. It has been discovered that the
inherited pathological CAG repeat in the HTT gene is unsta-
ble in various tissues and can expand throughout an individ-
ual’s life, particularly within the affected brain ( 21 ). This phe-
nomenon, known as SI, is pronounced in specic cell types
and brain regions, particularly in MSNs of the striatum, the
most affected brain area ( 22 ,23 ). The foundations for this piv-
otal discovery were laid by Kennedy and colleagues in 2003,
who identied extreme somatic expansion in the striatum of
HD patients using early CAG sizing technologies, after simi-
lar observations in HD mouse models ( 24 ,25 ). Concurrently,
as instability is linked to altered DNA repair, Wheeler and
colleagues made signicant contributions, expanding upon
work by the Messer laboratory ( 26 ), by demonstrating that
knocking out the mismatch repair (MMR) gene Msh2 modi-
es CAG repeat instability in an HD mouse model, delaying
brain pathology ( 27 ). At the time, these ndings were not fully
appreciated, perhaps due to technological limitations and the
prevailing notion that such expansions merely increased the
toxicity of inherited pathological alleles without directly caus-
ing cell death ( 28 ). In the following years, more studies con-
rmed the primary role of SI in A OO ( 29 , 30 ). Seminal work
from the McMurray laboratory, for example, showed a delay
in pathogenesis upon suppressing SI across multiple mouse
models ( 31 ,32 ).
Genome-wide association studies (GWAS) provided critical
insights into the relevance of SI by exploring the HD genome
at the population level. These studies identied both cis -acting
polymorphisms in HTT exon 1 and trans -acting polymor-
phisms in other loci as modiers of AOO, while at the same
time affecting SI in peripheral cells ( 33 ,34 ). They conrmed
that CAG repeat length inherited through the germline ac-
counts for only ∼60% of individual AOO variability ( 33 ),

Nucleic Acids Research , 2025, Vol. 53, No. 1 3
Microglia
Astrocytes
Medium Spiny
Neurons
Cortical
Neurons
Time
B
A
no. CAG
in HTT
(CAG)42CAGCAGCAGCAGCAGCAGCAGCAGCAACAG
G
)
4
2
CA
G
CA
G
)
4
C
(C
)
CA
G
C
(CAG)42CAGCAGCAGCAGCAACAG
(CAG)42CAACAG
(CAG)42CAGCAGCAGCAGCAGCAACAG
Somatic instability
with further expansion
Cortical Neurons
Medium Spiny Neurons
150 CAG
150 CAG
Figure 1. SI causes cell-type specic vulnerability. ( A ) Recent studies have shown that somatic expansions are not only tissue-specic, but also cell-t ype
specic; ( B ) Vulnerable cell types preferentially undergo somatic expansion over the course of the patient’s lifetime, ultimately leading to transcriptional
dysregulation and cell death.
with the remaining variance explained by an increasing num-
ber of cis - and trans -modiers. For instance, variations in the
nucleotide composition of the HTT CAG repeat region and
polymorphisms in MMR genes were found to correlate with
differential SI in the blood of HD patients exhibiting accel-
erated clinical manifestations ( 34–36 ), supporting a link be-
tween modiers, SI and HD onset.
A new theory, the ‘two-sequential components’ hypothesis,
further links SI to AOO ( 37 ). According to this hypothesis,
the inherited germline-transmitted HTT CAG repeat, once ex-
ceeding a certain instability threshold, undergoes somatic ex-
pansion at a rate modulated by trans- and cis -acting modiers
of AOO. Upon exceeding a second toxicity threshold, CAG
triplets in vulnerable cell types become harmful ( 37 ). This the-
ory challenges the longstanding hypothesis that mutant HTT
(mHTT) exerts cumulative damage throughout the patient’s
life, with SI merely exacerbating that damage ( 32 ,37 ). More
recently, two pioneering studies analyzing postmortem human
HD brains demonstrated that SI rates are not only organ and
tissue-specic [SI is marked in the brain striatum, but only
moderate in the cerebellum ( 21 )], but also cell-type specic,
with MSNs—the most vulnerable cells in HD—showing the
highest levels of instability ( 22 ,28 ) (Figure 1 ).
These advancements have been made possible by continu-
ous improvements in genomic and CAG sequencing technolo-
gies, as detailed in the accompanying paper by some of the
authors ( 17 ). For example, Mätlik and colleagues developed
a uorescence-activated nuclear sorting (FANS) approach to
isolate different cell populations from ve post-mortem HD
brains based on marker gene expression. The isolated cells
were then subjected to bulk RNA-seq and HTT CAG sizing,
providing matched transcriptional and instability proles at
the subpopulation level ( 22 ). In a different study, Handsaker
and colleagues pushed the technological frontier further by de-
veloping a sophisticated single-cell RNA-sequencing method
based on long-reads, allowing simultaneous acquisition of
transcriptional prole and HTT CAG size from the same cells.
This allowed the grouping of cells not only by transcriptional

4 Nucleic Acids Research , 2025, Vol. 53, No. 1
(CAG)42
(CAG)42
Glutamine Proline
(CAG)42 CAACAG CCG CCA (CCG)n
Reference
Loss of
interruption
(LOI)
CAACAG
Duplication
(DUP)
QQQQ
QQP
QPPP
QQQQ
QQP
QPPP
QQQQ
QQP
QPPP
QQ
(Q)42+2 (P)n
(P)n
(P)n
CCG CCG(CCG)n
CAGCAG
(CAACAG)2CCG CCA (CCG)n
(Q)42+2
(Q)42+4
Figure 2. HTT allele str uct ures inuence HD A O O. T he upper part of the diagram represents a reference HD allele with 42 CAG repeats f ollo w ed b y the
typically human ‘CAA–CAG’ tract, leading to a protein with 42Q + 2Q (both CAA and CAG translate to glutamine, Q). The CCG–CCA pair [representing
the initial tract of the proline-rich domain (PRD)] f ollo wing the CAGs is also shown. Middle and bottom sections: GWAS-identied variants in the HT T
allele nucleotide sequence that alter A O O; specically, (middle) the LOI disease haplotype, an A-to-G synonymous mutation in the polyQ tract, leads to
the same protein as the reference HD allele (42Q + 2Q), but accelerates disease onset. Con v ersely (bottom), in the DUP disease haplotype, the
inclusion of an additional ‘CAA–CAG’ tract dela y s disease onset despite adding two extra Qs to the protein (42Q + 4Q).
prole (and hence cell type) but also by acquired CAG length
( 28 ). They found that (i) SI is especially pronounced in MSNs,
the most vulnerable neurons in HD; (ii) within-patient, MSNs
must exceed 150 CAGs to undergo overt and cell-autonomous
transcriptional dysregulation; (iii) a portion of the MSNs ex-
hibited extreme expansions in the HTT gene, with > 800 CAG
repeats; and (iv) as a consequence of such extreme expansions,
many cell types in the brain become transcriptionally dysreg-
ulated through non-cell autonomous mechanisms. Their data
support a multi-phase model termed ‘ELongATE’, where so-
matic expansion rates in neurons accelerate signicantly once
they exceed 80 CAGs, with a threshold of about 150 CAG re-
peats required to trigger cell-autonomous transcriptional dys-
regulation in MSNs, and the consequent cell death ( 28 ). Ac-
cording to this model, atrophy and de-vascularization of the
caudate, both typical of HD, occur only after MSN loss and
despite the other cell types show modest SI ( 28 ). Future studies
will determine whether the ELongATE model, which provides
a plausible explanation for the central role of SI in disease
pathogenesis, will stand the test of time.
DNA modiers shaping the fate of CAG repeat
expansion
Although SI at the HTT DNA locus has been known for some
time, GWAS data have revitalized its study by identifying ge-
netic variants involved in DNA repair and replication, two
processes that signicantly inuence AOO. These ndings,
along with increasing evidence of somatic variability in CAG
repeat expansion within the brain, suggest that DNA itself is
a crucial biotype linked to disease pathogenesis. This idea is
further supported by recent evidence associating variations at
the HTT locus with both AOO and SI ( 33 ).
Cis -modiers, i.e. genetic variants located near the CAG
repeat, have been identied that inuence HD progression
( 33 , 34 , 36 , 38 ). One such variant affects the penultimate triplet
of the CAG repeat. In humans, the repetition of pure CAGs
typically ends with a C AA–C AG segment, where the C AA,
along with CAG, encodes glutamine (Q) (Figure 2 , reference).
This penultimate CAA is considered an interruption in the
pure CAGs stretch. However, the resulting protein contains
a continuous polyglutamine (polyQ) stretch encoded by the
pure CAGs sequence, the penultimate CAA and the nal CAG.
A typical patient with 42 CAG repeats followed by the CAA–
CAG produces a protein with 42 + 2Qs (Figure 2 , reference).
One group of patients deviated from the reference HD cases
by having the penultimate CAA codon replaced by CAG (loss
of interruption, LOI), creating a longer uninterrupted CAG
stretch, while producing the same protein. Specically, such
patients with, supposedly, 42 CAG repeats would acquire an
additional C AG–C AG segment in place of C AA–C AG, thereby
creating a longer stretch of pure CAGs. Despite this, the num-
ber of Qs in the resulting protein remains at 42 + 2Qs (Figure
2 , LOI). Remarkably, patients with the LOI haplotype exhib-
ited a hastening of AOO of 25 years on average. This LOI
variant was also associated with a variant in the CCA codon
of the PRD, also inuencing AOO ( 34 ,36 ,39 ).
Another group of patients carried a duplication of the
penultimate C AA-C AG tract (DUP, Duplication). In a hypo-
thetical patient with 42 uninterrupted CAG repeats, this du-
plication will result in a protein with 42 + 4Qs, which could
theoretically increase protein toxicity (Figure 2 , DUP). How-
ever, this variation is now recognized as a benecial cis -acting
modier, leading to an average delay of 4.2 years in AOO
( 34 ).
Importantly, both the LOI and DUP cis- variants have been
associated with SI, at least in blood cells. Increased and de-
creased CAG instability , respectively , have been observed in
these genetic conditions, pointing to a direct link between
HTT locus cis -variants, CAG instability and disease pheno-
types ( 33 , 34 , 36 , 38 , 39 ). Still, the full impact of this relation-
ship within a cohesive framework of SI-driven HD pathol-
ogy by cis -acting modiers remains to be explored, and data
conrming that these modiers inuence instability in human
brain neurons are still needed. In addition, data on Africa-
ancestry individuals contradicts the increased SI associated
with the LOI variants, suggesting that other mechanisms may
contribute in driving the pathogenesis ( 36 ,39 ).
A second set of data from GWAS points to trans -modiers
involved in DNA repair. Polymorphisms in genes such as
PMS1, MLH1, MSH3, PMS2, FAN1 and LIG1, which are in-
volved in the MMR pathway, have been implicated in DNA
repair and replication events leading to instability ( 33 ). These
polymorphisms, which typically induce a non-synonymous
amino acid change, alter protein levels and either accelerate or
delay disease onset, depending on the gene involved. Genetic
perturbations in MMR genes have been shown to affect SI in

Nucleic Acids Research , 2025, Vol. 53, No. 1 5
mouse models of HD and other trinucleotide expansion disor-
ders ( 19 , 30 , 40 ). The GeM-HD consortium, leveraging GWAS
data, conrmed the association between a polymorphism in
MSH3 and SI in blood ( 33 ,36 ). Newer GWAS with larger co-
horts not only conrmed previous modiers, but also identi-
ed new ones, such as POLD1 and MED15 ( 36 ). Notably,
among its various activities, MED15 has been found to bind
SREBP and regulate cholesterol biosynthesis ( 41 ), a process
whose normalization in the HD rodent brain proved bene-
cial ( 42 ,43 ).
These ndings highlight the critical role of both cis and
trans variants to instability (in blood cells) and AOO, although
the contribution of RNA-related mechanisms to instability
cannot yet be ruled out.
From guardians to contributors: MMR in somatic
instability mechanisms
During replication or transcription in post-mitotic neurons,
single-stranded DNA can form secondary structures, espe-
cially when expanded repeats are present ( 19 ,35 ). Long CAG
repeats, in particular, can give rise to complex, higher-order
non-canonical structures, such as DNA triplexes, hairpins, G-
quadruplexes and R-loops ( 18 ,44 ,45 ). The formation of these
structures depends on various factors, including sequence mo-
tifs, the presence of interruptions (i.e. purity), length and possi-
bly methylation status ( 18 , 44 , 45 ). Once these structures form,
they often create large loops (slip-outs) that are recognized
by MMR proteins, which attempt to repair the DNA dam-
age. However, rather than repairing the damage, these pro-
teins can contribute to the somatic expansion of pathogenic
repeats ( 35 ) (Figure 3 ).
MMR proteins form dimer complexes to address DNA
damage. MutS α(MSH2–MSH6) primarily recognizes small
base mismatches, while MutS β(MSH2–MSH3) detects larger
loops (2–10 base pairs), which form in pathogenic con-
texts. Upon recognizing a loop, MutS βrecruits either
MutL α(MLH1–PMS2) or MutL γ(MLH1–MLH3), forming
a ternary complex. Instead of breaking the damaged strand,
MutL is believed to create a break in the opposite strand ( 35 ).
An endonuclease then cuts the opposite strand, leading to the
longer repeat on the damaged strand being used as template
for DNA repair. This may result in the erroneous incorpora-
tion of new repeats into the gene ( 35 ).
The mechanisms by which these trans -modiers affect AOO
may not be the same across all genes, and this is still an active
research area. Moreover, multiple distinguishable AOO mod-
iers have been associated with polymorphisms in genes from
MMR pathway, such as PMS1 , PMS2 , MSH3 and LIG1 . Also
polymorphisms in FAN1 can inuence AOO in both direc-
tions ( 33 ,36 ). Although FAN1 is not a canonical MMR fac-
tor, it binds MLH1 and competes with MSH3 for ternary com-
plex formation, mitigating the effects of SI associated with ex-
panded CAG repeats. Its overexpression reduces SI, contrast-
ing the effects of other MMR proteins ( 46–48 ).
Given this mechanism of action, down-regulating MMR
proteins is expected to reduce SI. This effect has been ob-
served in multiple HD mouse models decient in MMR genes
( 26 , 27 , 30 , 40 , 49 ). However, caution is necessary when manip-
ulating these proteins, as deciencies in MSH2, MSH6, MLH1
and PMS1 have been associated with cancer ( 35 , 50 , 51 ).
Targeting somatic instability to treat HD
In recent years, therapeutic efforts have largely focused on re-
ducing mHTT levels ( 52 ), while also raising awareness and ex-
pertise in the most promising treatment approaches ( 53 ). The
proposed mechanism of MMR proteins suggests that manip-
ulating the expression of MMR genes—mimicking the effects
of naturally occurring trans -modiers—could offer promising
targets to slow repeat expansion rates ( 19 ). In support of this
therapeutic avenue, polymorphisms identied by GWAS are
well tolerated, as they have not been negatively selected by
evolutionary pressure.
The eld is now poised to expand gene silencing approaches
( 52 ,53 ) to target molecules affecting CAG instability. Animal
studies have demonstrated the therapeutic potential of com-
pounds that reduce the expression of proteins in the MMR
pathway. These studies differ in the targeted gene, the method
used to modulate translation efciency, the HD model and the
phenotypic readouts, however, they all agree on the value of
this approach ( 50 , 51 , 54 ).
O’Reilly and colleagues used RNA interference to reduce
MSH3 protein levels, demonstrating the efcacy of their ap-
proach in both in vitro and in vivo models ( 54 ). They iden-
tied two compounds that achieved a 55%–60% reduction
in MSH3 protein levels in the striatum two months post-
injection. Remarkably, this level of silencing was sufcient to
prevent somatic CAG expansion for up to 4 months in the
striatum of HD mouse models, as conrmed by capillary elec-
trophoresis fragment analysis ( 54 ). Notably, when they tested
a siRNA targeting the HTT gene itself, despite achieving high
silencing efciency, they found no measurable impact on so-
matic repeat expansion ( 54 ).
Ferguson and colleagues explored the effect of targeting
MMR genes identied as HD modiers of AOO using one
human HD induced pluripotent stem cell line ( 50 ). They
used CRISPR interference to reduce the expression of MSH2,
MSH3, MSH6, MLH1, PMS1, PMS2, MLH3 and LIG1 tran-
script levels by 60%–80%. The study assessed the impact of
silencing each target on SI, in both proliferating cells and
post-mitotic neurons. Using capillary electrophoresis frag-
ment analysis—therefore a method with limited sensitivity for
rare alleles ( 55 )—they observed a signicant reduction in SI
over 2 months when targeting MSH2, MSH3 and MLH1. A
moderate reduction was also observed with PMS1, PMS2 and
MLH3 , with PMS1 being proposed as a novel target for slow-
ing CAG repeat expansion ( 50 ).
Recently, Wa ng and colleagues studied the impact of knock-
ing out different MMR genes in mice ( 51 ). They generated ho-
mozygous and heterozygous KO alleles for 9 MMR genes in
a HD knock-in mouse model with extremely long CAG tract,
including Msh3, Mlh1, Pms1, Pms2, Ccdc82, Tcerg1, Msh2,
Msh6 and Polq . After 6 months, capillary electrophoresis frag-
ment analysis revealed in knock-in mice a signicant linear in-
crease in SI in the brain regions most affected by HD, along
with transcriptional alterations and mHTT aggregation, a typ-
ical HD hallmark ( 56 ,57 ). Notably, KO for Msh3 , Msh2 and
Pms1 showed a gene-dosage-dependent rescue of HD pheno-
types ( 51 ). Specically, Msh3 KO not only reduced SI com-
pared to unperturbed knock-in mice, but also led to a tran-
scriptional rescue in MSNs, with the effect being more pro-
nounced when both gene copies were knocked out. To further
assess cell-type-specic mosaicism of the CAG repeats, Wa ng
and colleagues performed fragment analysis on MSNs puri-

6 Nucleic Acids Research , 2025, Vol. 53, No. 1
Polymerase
Ligase 1
Normal Mismatch Repair
Mismatch binding by MutSβ
M
S
H
3
M
S
H
2
MutL recruitment
and nick formation
Gap formation
Repair synthesis and ligation
Mismatch correction
Exonuclease
MutSβ MutL CAG
BA
MLH1
PMS2
MLH3
M
S
H
3
M
S
H
2
Hypothetical expansion mechanism
Hairpin binding by MutSβ
MutL recruitment
and nick formation
on opposite strand
Gap formation
Repair synthesis and ligation
Aberrant correction leads to expansion
M
S
H
2
M
S
H
3
Ligase 1
Polymerase
PMS2
MLH3
MLH1
M
S
H
2
M
S
H
3
Exonuclease
M
S
H
3
M
S
H
2
PMS2
MLH3
MLH1
Figure 3. Proposed model for expansion and comparison to the canonical MMR pathway. ( A ) In canonical MMR, MutS βrecognizes small breaks and
recruits MutL to perform an excision in the strand carrying the mismatch; an exonuclease then forms a gap, and the DNA polymerase repairs the
damaged strand using the intact strand as template. ( B ) In the proposed model for somatic expansion, MutS βrecognizes bigger loops and recruits
either MutL αor MutL γ, forming a ternar y complex; however, excision occurs in the strand opposite to the one carrying the mismatch and the
e x onuclease creates a gap. As a result, the strand carrying the damage is used as a template for re-synthesis, leading to repeat expansion.

Nucleic Acids Research , 2025, Vol. 53, No. 1 7
ed using FANS. Compared to the bulk striatal tissue, they
observed a much narrower distribution of CAG sizes, with un-
perturbed knock-in mice almost doubling the initial germline
CAG number over 16 months ( 51 ).
Recent studies emphasize MSH3 and PMS1, key compo-
nents of MutS βand MutL β, as safe and effective targets for
reducing SI and ameliorating HD phenotypes. Ta rge ti ng both
genes simultaneously may also yield synergistic effects ( 50 ).
Future strategies will likely focus on enhancing the potency,
stability and duration of gene knockdowns in in vivo mod-
els, alongside developing high-throughput, long-term screen-
ing systems ( 58 ). Neuronal organoid cultures could serve as a
valuable tool, enabling co-culturing and perturbation screen-
ing of various cis and trans -modiers ( 59 ). Since HD pheno-
types, such as SI, transcriptional dysregulation and mHTT
aggregation formation ( 28 ,51 ) are often co-modulated, SI
in HD-vulnerable cell types may serve as an effective early
marker for potential therapies. However, to ensure consistency
across studies, rigorous standards for target enrichment, se-
quencing methods and bioinformatics protocols will be essen-
tial ( 17 ), potentially guiding the development of clinical guide-
lines for SI monitoring and HD prognosis assessment.
The two sides of the coin: heads - HD is driven by SI
Transcriptional proling and matched CAG sizing in post-
mortem HD brains has highlighted a primary role of SI in
HD pathogenesis ( 28 ). The proposed ‘ELongATE’ model—
enabled by advancements in cutting-edge technology ( 17 )—
provides a plausible explanation to some unresolved ques-
tions. Firstly, the progressive nature of MSN loss during HD
neurodegeneration may be attributed to the time required for
the accumulation of extreme expansion events in these neu-
rons over a lifetime, with up to 840 CAGs detected in post-
mortem HD brains ( 28 ). Relevant work from the Heintz lab
reviewed SI in cell types within the striatum and found SI in
MSNs but also in cell types that do not degenerate ( 22 ,60 ).
They conclude that although SI may be important, it alone
is not sufcient for cell death. However, these studies used
short-read sequencing methods (technically limited to approx-
imately 110 CAG) and may have overlooked extreme expan-
sion events above 150 CAGs, which seem to be the key driver
of pathogenesis ( 28 ). Secondly, the prolonged period of de-
generation observed in HD patients—typically 10–20 years
from diagnosis to death ( 53 ,61 )—seems compatible with the
fast but asynchronous neuronal degeneration proposed in the
‘ELongATE’ model. According to this model, despite all cells
in the patient’s brain express mHTT, only mHTT with > 150
CAG is indeed toxic. As such, at any given time, only a few
MSNs may produce this extremely expanded and toxic ver-
sion of the mHTT protein, thereby contributing to the slow,
progressive degeneration. Thirdly, the limited efcacy of the
initial ASO strategies targeting HTT—despite persistent, dose-
dependent decreases in mHTT levels in cerebrospinal uid
( 53 )—may be due to the small number of MSNs produc-
ing the extremely expanded and toxic mHTT protein at any
given time, which are not preferentially targeted by the ASO
( 28 ). Most of the mHTT present may actually be harmless,
thus providing minimal clinical benet from its depletion ( 28 ).
Conversely, the small fraction of toxic mHTT should probably
be depleted at higher efciency, to see a consistent therapeutic
benet.
The two sides of the coin: tails - HD is not driven by
SI
The ‘ELongATE’ model must also address potentially conict-
ing evidence gathered over the past 15 years. Firstly, recent
GWA studies have identied certain disease haplotypes that
accelerate AOO without increasing SI in blood and, poten-
tially, in the HD brain ( 36 ,39 ). If SI in MSNs is not involved
in these cases, alternative SI-independent pathogenic mecha-
nisms may be at play, such as those associated with the pro-
duction of truncated HTT exon1 transcript or protein ( 53 ).
Secondly, extreme SI in HD patients has been described in pe-
ripheral tissues, such as the liver, which do not exhibit HD
pathology ( 21 , 55 , 62 ). Although this may be associated with
the liver’s regenerative potential or its clearing capacity which
may prevent mHTT accumulation, this aspect requires fur-
ther investigation. Thirdly, some evidence suggests that CAG
length in HTT inuences brain development ( 13 ,63 ), with
mHTT potentially providing an early advantage that is fol-
lowed by an accelerated aging process ( 64 ). Since these phe-
notypes are likely associated with transcriptional changes oc-
curring very early during neurodevelopment, they are unlikely
to be driven by SI. Accordingly, it can be concluded that the
CAG tract in HTT appears to inuence phenotype even before
reaching the proposed toxicity threshold of 150 CAGs. Lastly,
numerous studies have shown transcriptional changes in dif-
ferentiated neurons derived from human pluripotent stem cell
lines with < 150 CAGs ( 59 ,65–69 ). According to the ‘ELon-
gATE’ model, transcriptional changes at lower CAGs—at least
in the HD striatum—are considered part of a neurodegenera-
tion process resulting from the death of the co-existing MSNs
carrying extreme expansions ( 28 ). If this holds true, no tran-
scriptional changes are expected in HD neurons, if none of
those neurons had enough time to reach the proposed toxic-
ity threshold. This hypothesis may be tested in telencephalic
organoids, that recapitulate the microarchitecture of the brain.
Data from our lab showed transcriptional differences in 56Q
versus 20Q cell lines after only 45 days of in vitro differenti-
ation ( 59 ), presumably a period insufcient to reach the pro-
posed CAG toxicity threshold. Further studies are needed to
verify the impact of extreme SI in HD and to explore the mech-
anisms by which neurons below the toxicity threshold degen-
erate.
Conclusions
In conclusion, while many aspects remain to be fully eluci-
dated, SI is emerging as a critical pathogenic mechanism in
HD, with implications for other neurodegenerative disorders
involving triplet repeats ( 33 ). Harnessing advanced technolo-
gies to monitor SI is becoming essential for both diagnos-
tic and therapeutic approaches in HD. As we further inves-
tigate the mechanisms driving CAG repeat expansion in spe-
cic brain regions, particularly in neurons, it is evident that SI
not only inuences AOO but also shapes disease progression.
By identifying and targeting key cis- and trans - modiers, new
therapeutic strategies can be developed to slow or even halt
the expansion process. This innovative direction underscores
the need for precise diagnostic tools that go beyond simple
CAG counts, capturing the complexities of genetic variability
to offer more tailored and effective treatments. With advance-
ments in gene silencing and genome editing technologies, the
potential to directly modify SI could pave the way for transfor-

8 Nucleic Acids Research , 2025, Vol. 53, No. 1
mative interventions, offering hope to patients, scientists and
physicians confronting this devastating condition.
Data availability
No new data were generated or analysed in support of this
research.
Funding
European Research Council, Advanced Grant [742436];
NSC-Reconstruct Consortium, European Union’s Hori-
zon 2020 Research and Innovation Program [874758];
C.H.D.I. Foundation, New York, U.S.A. [JSC A11103];
Leslie Gehry Prize for Innovation in Science from the
Hereditary Disease Foundation (New York, U.S.A.); Fon-
dazione Telethon [GMR23T1059 and GMR23T1216];
Ministero dell’Istruzione, dell’Università e della Ricerca
[2022LBENTH]. Funding for open access charge: H2020
European Research Council Grant [742436].
Conict of interest statement
None declared.
This paper is linked to: doi:10.1093/ nar/ gkae1155 .
References
1. Gymrek, M. , Willems, T. , Guilmatre, A. , Zeng, H. , Markus, B. ,
Georgiev, S. , Daly, M.J. , Price, A.L. , Pritchard, J.K. , Sharp, A.J. , et al.
(2016) Abundant contribution of short tandem repeats to gene
expression variation in humans. Nat. Genet., 48 , 22–29.
2. Kozlowski, P. , de Mezer, M. and Krzyzosiak, W. J . (2010)
Trinucleotide repeats in human genome and exome. Nucleic Acids
Res., 38 , 4027–4039.
3. Sulovari, A. , Li, R. , Audano, P. A . , Porubsky, D. , Vollger, M.R. ,
Logsdon,G.A. and Human Genome Structural Varia ti on
ConsortiumHuman Genome Structural Var ia ti on Consortium,
Warren, W. C . , Pollen, A.A. , Chaisson, M.J.P. , et al. (2019)
Human-specic tandem repeat expansion and differential gene
expression during primate evolution. Proc. Natl Acad. Sci., 116 ,
23243–23253.
4. Kashi, Y. , King, D. and Soller, M. (1997) Simple sequence repeats as
a source of quantitative genetic variation. Trends Genet., 13 ,
74–78.
5. Kashi, Y. and King, D.G. (2006) Simple sequence repeats as
advantageous mutators in evolution. Trends Genet. , 22 , 253–259.
6. Fondon, J.W. , Hammock, E.A.D. , Hannan, A.J. and King, D.G .
(2008) Simple sequence repeats: genetic modulators of brain
function and behavior. Trends Neurosci. , 31 , 328–334.
7. Zuccato, C. and Cattaneo, E. (2016) The Huntington’s paradox.
Sci. Am., 315 , 56–61.
8. Wright, S.E. and Todd, P. K . (2023) Native functions of short
tandem repeats. Elife , 12 , e84043.
9. Iennaco, R. , Formenti, G. , Tro vesi, C. , Rossi, R.L. , Zuccato, C. ,
Lischetti, T. , Bocchi, V. D . , Scolz, A. , Martínez-Labarga, C. ,
Rickards, O. , et al. (2022) The evolutionary history of the polyQ
tract in huntingtin sheds light on its functional pro-neural
activities. Cell Death Differ. , 29 , 293–305.
10. MacDonald, M.E. , Ambrose, C.M. , Duyao, M.P. , Myers, R.H. ,
Lin, C. , Srinidhi, L. , Barnes, G. , Taylor, S.A. , James, M. , Groot, N. ,
et al. (1993) A novel gene containing a trinucleotide repeat that is
expanded and unstable on Huntington’s disease chromosomes.
Cell , 72 , 971–983.
11. Mühlau, M. , Winkelmann, J. , Rujescu, D. , Giegling, I. ,
Koutsouleris, N. , Gaser, C. , Arsic, M. , Weindl, A. , Reiser, M. and
Meisenzahl,E.M. (2012) Vari at io n within the Huntington’s disease
gene inuences normal brain structure. PLoS One , 7 , e29809.
12. Lee, J.K. , Conrad, A. , Epping, E. , Mathews, K. , Magnotta, V. ,
Dawson, J.D. and Nopoulos, P. (2018) Effect of trinucleotide repeats
in the Huntington’s gene on intelligence. EBioMedicine , 31 , 47–53.
13. Schultz, J.L. , Saft, C. and Nopoulos, P. C . (2021) Association of CAG
repeat length in the Huntington gene with cognitive performance
in young adults. Neurology , 96 , e2407–e2413.
14. Caron, N.S. , Wright, G.E. and Hayden, M.R. (1993) Huntington
Disease. In: Adam, M.P. , Mirzaa, G.M. , Pagon, R.A. , Wallace, S.E. ,
Bean, L.J. , Gripp, K.W. and Amemiya, A. (eds.) GeneReviews(®) .
University of Washington, Seattle, Seattle (WA).
15. Genetic Modiers of Huntington’s Disease (GeM-HD)
Consortium (2015) Identication of genetic factors that modify
clinical onset of Huntington’s disease. Cell , 162 , 516–526.
16. Budworth, H. and McMurray, C.T. (2013) A brief history of triplet
repeat diseases. Methods Mol. Biol. , 1010 , 3–17.
17. Maestri, S. , Scalzo, D. , Damaggio, G. , Zobel, M. , Besusso, D. and
Cattaneo,E. (2024) Navigating triplet repeats sequencing:
concepts, methodological challenges and perspective for
Huntington’s disease. Nucleic Acids Res.,
https:// doi.org/ 10.1093/ nar/ gkae1155 .
18. Rajan-Babu, I.-S. , Dolzhenko, E. , Eberle, M.A. and Friedman, J.M.
(2024) Sequence composition changes in short tandem repeats:
heterogeneity, detection, mechanisms and clinical implications.
Nat. Rev. Genet., 25 , 476–499.
19. Schmidt, M.H.M. and Pearson, C.E. (2016) Disease-associated
repeat instability and mismatch repair. DNA Repair (Amst) , 38 ,
117–126.
20. Ranen, N.G. , Stine, O.C . , Abbott, M.H. , Sherr, M. , Codori, A.M. ,
Franz, M.L. , Chao, N.I. , Chung, A.S. , Pleasant, N. and Callahan, C.
(1995) Anticipation and instability of IT-15 (CAG)n repeats in
parent-offspring pairs with Huntington disease. Am. J. Hum.
Genet., 57 , 593–602.
21. Mouro Pinto, R. , Arning, L. , Giordano, J.V. , Razghandi, P. ,
Andrew, M.A. , Gillis, T. , Correia, K. , Mysore, J.S. , Grote
Urtubey, D.-M. , Parwez, C.R. , et al. (2020) Patterns of CAG repeat
instability in the central nervous system and periphery in
Huntington’s disease and in spinocerebellar ataxia type 1. Hum.
Mol. Genet., 29 , 2551–2567.
22. Mätlik, K. , Baffuto, M. , Kus, L. , Deshmukh, A.L. , Davis, D.A. ,
Paul, M.R. , Carroll, T. S. , Caron, M.-C. , Masson, J.-Y. , Pearson, C.E. ,
et al. (2024) Cell-type-specic CAG repeat expansions and
toxicity of mutant Huntingtin in human striatum and cerebellum.
Nat. Genet., 56 , 383–394.
23. Ciosi, M. , Cumming, S.A. , Chatzi, A. , Larson, E. , Tottey, W. ,
Lomeikaite, V. , Hamilton, G. , Wheeler, V. C . , Pinto, R.M. , Kwak, S. ,
et al. (2021) Approaches to sequence the HTT CAG repeat
expansion and quantify repeat length variation. J. Huntingtons
Dis., 10 , 53–74.
24. Kennedy, L. , Evans, E. , Chen, C.-M. , Craven, L. , Detloff, P. J . , Ennis, M.
and Shelbourne,P .F . (2003) Dramatic tissue-specic mutation
length increases are an early molecular event in Huntington
disease pathogenesis. Hum. Mol. Genet., 12 , 3359–3367.
25. Kennedy, L. and Shelbourne, P .F . (2000) Dramatic mutation
instability in HD mouse striatum: does polyglutamine load
contribute to cell-specic vulnerability in Huntington’s disease?
Hum. Mol. Genet., 9 , 2539–2544.
26. Manley, K. , Shirley, T.L. , Flaherty, L. and Messer, A. (1999) Msh2
deciency prevents in vivo somatic instability of the CAG repeat in
Huntington disease transgenic mice. Nat. Genet., 23 , 471–473.
27. Wheeler, V. C . , Lebel, L.-A. , Vrbanac, V. , Teed, A. , te Riele, H. and
MacDonald,M.E. (2003) Mismatch repair gene Msh2 modies the
timing of early disease in hdh(Q111) striatum. Hum. Mol. Genet.,
12 , 273–281.
28. Handsaker, R.E. , Kashin, S. , Reed, N.M. , Ta n, S. , Lee, W.-S. ,
McDonald, T. M . , Morris, K. , Kamitaki, N. , Mullally, C.D. ,

Nucleic Acids Research , 2025, Vol. 53, No. 1 9
Morakabati, N. , et al. (2024) Long somatic DNA-repeat
expansion drives neurodegeneration in Huntington disease.
bioRxiv doi: https:// doi.org/ 10.1101/ 2024.05.17.592722 , 20 May
2024, pre-print: not peer-reviewed.
29. Swami, M. , Hendricks, A.E. , Gillis, T. , Massood, T. , Mysore, J. ,
Myers, R.H. and Wheeler, V. C . (2009) Somatic expansion of the
Huntington’s disease CAG repeat in the brain is associated with an
earlier age of disease onset. Hum. Mol. Genet., 18 , 3039–3047.
30. Tomé , S. , Manley, K. , Simard, J.P. , Clark, G.W. , Slean, M.M. ,
Swami, M. , Shelbourne, P .F . , T illier, E.R.M. , Monckton, D.G . ,
Messer, A. , et al. (2013) MSH3 polymorphisms and protein levels
affect CAG repeat instability in Huntington’s disease mice. PLoS
Genet., 9 , e1003280.
31. Kovtun, I.V. , Liu, Y. , Bjoras, M. , Klungland, A. , Wilson, S.H. and
McMurray,C.T. (2007) OGG1 initiates age-dependent CAG
trinucleotide expansion in somatic cells. Nature , 447 , 447–452.
32. Budworth, H. , Harris, F. R . , Williams, P. , Lee, D.Y. , Holt, A. , Pahnke, J. ,
Szczesny, B. , Acevedo-Torres, K. , Ayala-Peña, S. and McMurray, C.T.
(2015) Suppression of somatic expansion delays the onset of
pathophysiology in a mouse model of Huntington’s disease. PLoS
Genet., 11 , e1005267.
33. Genetic Modiers of Huntington’s Disease (GeM-HD)
Consortium (2019) CAG repeat not polyglutamine length
determines timing of Huntington’s disease onset. Cell , 178 ,
887–900.
34. Wright, G.E.B. , Collins, J.A. , Kay, C. , McDonald, C. , Dolzhenko, E. ,
Xia, Q. , Be
ˇ
canovi
´
c, K. , Drögemöller, B.I. , Semaka, A. , Nguyen, C.M. ,
et al. (2019) Length of uninterrupted CAG, independent of
polyglutamine size, results in increased somatic instability,
hastening onset of Huntington disease. Am. Hum. Genet., 104 ,
1116–1126.
35. Rajagopal, S. , Donaldson, J. , Flower, M. , Hensman Moss, D.J. and
Tabrizi,S.J. (2023) Genetic modiers of repeat expansion
disorders. Emerg. Top. Life Sci., 7 , 325–337.
36. Genetic Modiers of Huntington’s Disease (GeM-HD)
Consortium, Lee, J.-M. , McLean, Z.L. , Correia, K. , Shin, J.W. , Lee, S. ,
Jang, J.-H. , Lee, Y. , Kim, K.-H. , Choi, D.E. , et al. (2024) Genetic
modiers of somatic expansion and clinical phenotypes in
Huntington’s disease reveal shared and tissue-specic effects.
bioRxiv doi: https:// doi.org/ 10.1101/ 2024.06.10.597797 , 18 June
2024, pre-print: not peer-reviewed.
37. Hong, E.P. , MacDonald, M.E. , Wheeler, V. C . , Jones, L. , Holmans, P. ,
Orth, M. , Monckton, D.G . , Long, J.D. , Kwak, S. , Gusella, J.F. , et al.
(2021) Huntington’s disease pathogenesis: two sequential
components. J. Huntingtons Dis., 10 , 35–51.
38. Ciosi, M. , Maxwell, A. , Cumming, S.A. , Hensman Moss, D.J. ,
Alshammari, A.M. , Flower, M.D. , Durr, A. , Leavitt, B.R. ,
Roos, R.A.C. , Holmans, P. , et al. (2019) A genetic association study
of glutamine-encoding DNA sequence structures, somatic CAG
expansion, and DNA repair gene variants, with Huntington
disease clinical outcomes. EBioMedicine , 48 , 568–580.
39. Dawson, J. , Baine-Savanhu, F. K . , Ciosi, M. , Maxwell, A. ,
Monckton, D. G. and Krause, A. (2022) A probable cis -acting
genetic modier of Huntington disease frequent in individuals
with African ancestry. HGG Adv. , 3 , 100130.
40. Dragileva, E. , Hendricks, A. , Teed, A. , Gillis, T. , Lopez, E.T. ,
Friedberg, E.C. , Kucherlapati, R. , Edelmann, W. , Lunetta, K.L. ,
MacDonald, M.E. , et al. (2009) Intergenerational and striatal CAG
repeat instability in Huntington’s disease knock-in mice involve
different DNA repair genes. Neurobiol. Dis., 33 , 37–47.
41. Nakatsubo, T. , Nishitani, S. , Kikuchi, Y. , Iida, S. , Ya mad a, K. ,
Tanaka, A. and Ohkuma, Y. (2014) Human mediator subunit
MED15 promotes transcriptional activation. Drug Discov. Ther.,
8 , 212–217.
42. Val en za, M. , Birolini, G. and Cattaneo, E. (2023) The translational
potential of cholesterol-based therapies for neurological disease.
Nat. Rev. Neurol., 19 , 583–598.
43. Cattaneo, E. and Barker, R.A. (2024) Brain cholesterol therapy for
Huntington’s disease—Does it make sense? Clin. Tran sl. Med., 14 ,
e1746.
44. Barbé, L. and Finkbeiner, S. (2022) Genetic and epigenetic interplay
dene disease onset and severity in repeat diseases. Front. Aging
Neurosci. , 14 .
45. Khristich, A.N. and Mirkin, S.M. (2020) On the wrong DNA track:
molecular mechanisms of repeat-mediated genome instability. J.
Biol. Chem., 295 , 4134–4170.
46. Goold, R. , Hamilton, J. , Menneteau, T. , Flower, M. , Bunting, E.L. ,
Aldous, S.G. , Porro, A. , V icente, J.R. , Allen, N.D. , Wilkinson, H. , et al.
(2021) FAN1 controls mismatch repair complex assembly via
MLH1 retention to stabilize CAG repeat expansion in
Huntington’s disease. Cell Rep. , 36 , 109649.
47. Goold, R. , Flower, M. , Moss, D. H. , Medway, C. , Wood-Kaczmar, A. ,
Andre, R. , Farshim, P. , Bates, G.P. , Holmans, P. , Jones, L. , et al. (2019)
FAN1 modies Huntington’s disease progression by stabilizing the
expanded HTT CAG repeat. Hum. Mol. Genet., 28 , 650–661.
48. McAllister, B. , Donaldson, J. , Binda, C.S. , Powell, S. , Chughtai, U. ,
Edwards, G. , Stone, J. , Lobanov, S. , Elliston, L. , Schuhmacher, L.-N. ,
et al. (2022) Exome sequencing of individuals with Huntington’s
disease implicates FAN1 nuclease activity in slowing CAG
expansion and disease onset. Nat. Neurosci., 25 , 446–457.
49. Kovalenko, M. , Dragileva, E. , Claire, J.S. , Gillis, T. , Guide, J.R. ,
New, J. , Dong, H. , Kucherlapati, R. , Kucherlapati, M.H, Ehrlich, M.E,
et al., (2012) Msh2 acts in medium-spiny striatal neurons as an
enhancer of CAG instability and mutant huntingtin phenotypes in
Huntington’s disease knock-in mice. PLoS One , 7 , e44273.
50. Ferguson, R. , Goold, R. , Coupland, L. , Flower, M. and Tabrizi, S.J.
(2024) Therapeutic validation of MMR-associated genetic
modiers in a human ex vivo model of Huntington disease. Am. J.
Hum. Genet., 111 , 1165–1183.
51. Wan g, N. , Zhang, S. , Langfelder, P. , Ramanathan, L. , Plascencia, M. ,
Gao, F. , Va ca , R. , Gu, X. , Deng, L. , Dionisio, L.E. , et al. (2024) Msh3
and Pms1 set neuronal CAG-repeat migration rate to drive
selective striatal and cortical pathogenesis in HD mice. bioRxiv
doi: https:// doi.org/ 10.1101/ 2024.07.09.602815 , 15 July 2024,
pre-print: not peer-reviewed.
52. Tabrizi, S.J. , Ghosh, R. and Leavitt, B.R. (2019) Huntingtin lowering
strategies for disease modication in Huntington’s disease.
Neuron , 101 , 801–819.
53. Tabrizi, S.J. , Estevez-Fraga, C. , van Roon-Mom, W. M . C . ,
Flower, M.D. , Scahill, R.I. , Wild, E.J. , Muñoz-Sanjuan, I. ,
Sampaio, C. , Rosser, A.E. and Leavitt, B.R. (2022) Potential
disease-modifying therapies for Huntington’s disease: lessons
learned and future opportunities. Lancet Neurol. , 21 , 645–658.
54. O’Reilly, D. , Belgrad, J. , Ferguson, C. , Summers, A. , Sapp, E. ,
McHugh, C. , Mathews, E. , Boudi, A. , Buchwald, J. , Ly, S. , et al.
(2023) Di-valent siRNA-mediated silencing of MSH3 blocks
somatic repeat expansion in mouse models of Huntington’s
disease. Mol. Ther., 31 , 1661–1674.
55. Lee, J.-M. , Zhang, J. , Su, A.I. , Wa lk er , J.R. , Wiltshire, T. , Kang, K. ,
Dragileva, E. , Gillis, T. , Lopez, E.T. , Boily, M.-J. , et al. (2010) A
novel approach to investigate tissue-specic trinucleotide repeat
instability. BMC Syst. Biol., 4 , 29.
56. DiFiglia, M. , Sapp, E. , Chase, K.O. , Davies, S.W. , Bates, G.P. ,
Vonsattel, J.P. and Aronin, N. (1997) Aggregation of huntingtin in
neuronal intranuclear inclusions and dystrophic neurites in brain.
Science , 277 , 1990–1993.
57. Scherzinger, E. , Lurz, R. , Tur maine, M. , Mangiarini, L. ,
Hollenbach, B. , Hasenbank, R. , Bates, G.P. , Davies, S.W. , Lehrach, H.
and Wanker,E.E. (1997) Huntingtin-encoded polyglutamine
expansions form amyloid-like protein aggregates in vitro and in
vivo . Cell , 90 , 549–558.
58. Pinto, R.M. , Murtha, R. , Azevedo, A. , Douglas, C. , Kovalenko, M. ,
Ulloa, J. , Crescenti, S. , Burch, Z. , Oliver, E. , V italo, A. , et al. (2024)
Identication of genetic modiers of Huntington’s disease somatic
CAG repeat instability by in vivo CRISPR-Cas9 genome editing.

10 Nucleic Acids Research , 2025, Vol. 53, No. 1
bioRxiv: https:// doi.org/ 10.1101/ 2024.06.08.597823 , 9 June
2024, pre-print: not peer-reviewed.
59. Galimberti, M. , Nucera, M.R. , Bocchi, V. D . , Conforti, P. , Vezzoli, E. ,
Cereda, M. , Maffezzini, C. , Iennaco, R. , Scolz, A. , Falqui, A. , et al.
(2024) Huntington’s disease cellular phenotypes are rescued
non-cell autonomously by healthy cells in mosaic telencephalic
organoids. Nat. Commun., 15 , 6534.
60. Pressl, C. , Mätlik, K. , Kus, L. , Darnell, P. , Luo, J.-D. , Paul, M.R. ,
Weiss, A.R. , Liguore, W. , Carroll, T. S. , Davis, D.A. , et al. (2024)
Selective vulnerability of layer 5a corticostriatal neurons in
Huntington’s disease. Neuron , 112 , 924–941.
61. Nopoulos,P.C. (2016) Huntington disease: a single-gene
degenerative disorder of the striatum. Dialogues Clin. Neurosci.,
18 , 91.
62. Duncan, A.W. , Dorrell, C. and Grompe, M. (2009) Stem cells and
liver regeneration. Gastroenterology , 137 , 466.
63. Barnat, M. , Capizzi, M. , Aparicio, E. , Boluda, S. , Wennagel, D. ,
Kacher, R. , Kassem, R. , Lenoir, S. , Agasse, F. , Braz, B.Y. , et al. (2020)
Huntington’s disease alters human neurodevelopment. Science ,
369 , 787–793.
64. Neema, M. , Schultz, J.L. , Langbehn, D.R. , Conrad, A.L. , Epping, E.A. ,
Magnotta, V. A . and Nopoulos, P. C . (2024) Mutant huntingtin
drives development of an advantageous brain early in life:
evidence in support of antagonistic pleiotropy. Ann. Neurol., 96 ,
1006–1019.
65. Ring, K.L. , An, M.C. , Zhang, N. , O’Brien, R.N. , Ramos, E.M. , Gao, F. ,
Atwood, R. , Bailus, B.J. , Melov, S. , Mooney, S.D. , et al. (2015)
Genomic analysis reveals disruption of striatal neuronal
development and therapeutic targets in Human Huntington’s
disease neural stem cells. Stem Cell Rep. , 5 , 1023–1038.
66. HD iPSC Consortium (2017) Developmental alterations in
Huntington’s disease neural cells and pharmacological rescue in
cells and mice. Nat. Neurosci., 20 , 648–660.
67. V ictor, M.B. , Richner, M. , Olsen, H.E. , Lee, S.W. , Monteys, A.M. ,
Ma, C. , Huh, C.J. , Zhang, B. , Davidson, B.L. , Ya ng , X.W. , et al.
(2018) Striatal neurons directly converted from Huntington’s
disease patient broblasts recapitulate age-associated disease
phenotypes. Nat. Neurosci., 21 , 341–352.
68. Mehta, S.R. , To m, C.M. , Wa ng, Y. , Bresee, C. , Rushton, D. ,
Mathkar,P .P ., Tan g, J. and Mattis,V.B. (2018) Human Huntington’s
disease iPSC-derived cortical neurons display altered
transcriptomics, morphology, and maturation. Cell Rep., 25 ,
1081–1096.
69. Ooi, J. , Langley, S.R. , Xu, X. , Utami, K.H. , Sim, B. , Huang, Y. ,
Harmston, N.P. , Tay, Y.L. , Ziaei, A. , Zeng, R. , et al. (2019) Unbiased
proling of isogenic Huntington disease hPSC-derived CNS and
peripheral cells reveals strong cell-type specicity of CAG length
effects. Cell Rep., 26 , 2494–2508.
Received: October 2, 2024. Revised: November 8, 2024. Editorial Decision: November 14, 2024. Accepted: December 2, 2024
©The Author(s) 2024. Published by Oxford University Press on behalf of Nucleic Acids Research.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https: // creativecommons.org / licenses / by-nc / 4.0 / ), which permits
non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact reprints@oup.com for reprints and
translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact
journals.permissions@oup.com.