GB2605845A

GB2605845A - Somatic expansion inhibitors

Info

Publication number: GB2605845A
Application number: GB2105484.6A
Authority: GB
Inventors: Joanna Tabrizi Sarah; Balmus Gabriel; Goold Robert
Original assignee: Cambridge Enterprise Ltd; UCL Business Ltd
Current assignee: Cambridge Enterprise Ltd; UCL Business Ltd
Priority date: 2021-04-16
Filing date: 2021-04-16
Publication date: 2022-10-19
Also published as: GB202105484D0

Abstract

A composition for use in treating a repeat expansion disease comprising an MLH1 inhibitor and/or a FAN1 derived nuclease. The repeat expansion disease may include Huntington’s disease, Fragile X Syndrome, myotonic dystrophy (DM1 and DM2), amyotrophic lateral sclerosis and/or frontotemporal dementia caused by somatic expansion in the C9ORF72 gene, spinocerebellar ataxias (SCAs 1, 2, 3, 6, 7 and 17), Friedreich’s ataxia (FRDA), Fragile X Tremor Ataxia Syndrome (FXS/FXTAS), dentatorubral-pallidoluysian atrophy (DRPLA), spinal and bulbar muscular atrophy (SBMA), and Unverricht-Lundborg myoclonic epilepsy (EPM1). The MLH1 inhibitor may be a cyclic peptide, a peptide comprising the amino acid sequence SPYF, or a peptide having at least 70% identity to residues 120-140, 73-165, 73-190, 73-349, 1-165, 1-190 or 1-349 of SEQ ID NO.2. Also claimed is a method of identifying MLH1 inhibitors comprising culturing cells expressing MLH1 and MSH3 in the presence of a test agent, purifying MLH1 and proteins bound thereto, determining the level of MSH3 bound to MLH1 and comparing to a control level in the absence of the test agent.

Description

SOMATIC EXPANSION INHIBITORS

FIELD OF THE INVENTION

This invention relates to somatic expansion inhibitors, to methods of producing the same and to therapeutic applications thereof. More specifically, the invention relates to MLH1 inhibitors and FAN1 derived nucleases for treating, preventing or delaying the onset of repeat expansion diseases.

BACKGROUND OF THE INVENTION

Repeat expansion diseases are caused by the somatic expansion of sequence repeats, e.g. trinucleotide repeats. Faster somatic expansion rates correlate with earlier age at onset and faster and more severe disease progression. There are currently more than 50 known repeat expansion diseases, including Huntington's disease (HD), Fragile X Syndrome (FXS), myotonic dystrophy (DM1 and DM2), and Friedreich's ataxia (FRDA).

Recent genetic studies have shown that somatic expansion of the CAG repeat is the key pathogenic process driving HD onset and progression. HD is a monogenic neurodegenerative condition arising due to inheritance of 36 CAG trinucleotide repeats in exon 1 of the huntingtin (HTT) gene. Expansion of CAG repeats occurs in selected somatic and meiotic tissues, but the neurodegeneration is primarily due to loss of neurons in the striatum and cortex. The expanded CAG repeat may be pathogenic through several mechanisms, including at the protein level through translation into a longer, more toxic polyglutamine tract; at the RNA level through RAN translation or RNA secondary structure, and at the DNA level through an effect on transcription and DNA repair activity. Targeting downstream pathogenic processes, including protein homeostasis and metabolic dysfunction, has proved ineffective in therapy of HD. Somatic expansion of the CAG repeat is the key pathogenic process driving HD onset and progression, and so therapies acting on somatic expansion are crucial for effective therapy of HD.

Several genome-wide association studies (GWAS) have identified DNA repair genes as main modifiers of repeat expansion disease onset and progression in HD (Genetic Modifiers of Huntington's Disease Consortium., 2019). The strongest signal comes from genetic variation in the DNA repair gene FAN1, a nuclease of the Fanconi anaemia (FA) pathway, while other prominent modifications have been identified in members of the mismatch repair (MMR) pathway, MSH3, MLH1 and PMS2. Similarly, transcriptome-wide association studies (TWAS) have revealed a signature in which reduced MSH3 but increased FAN1 expression are associated with later onset, slower progression and CAG repeat stability in HD. In cell and animal models, deficiency of MSH3, MSH2 and MLH1, or increased expression of FAN1, have been shown to prevent somatic expansion in HD. This is consistent with analyses linking FAN1 loss of function variants, such as p.R507H (Genetic Modifiers of Huntington's Disease Consortium.) 2019), which display earlier onset of HD.

DNA repair genes including FAN1 have been shown to underlie a common genetic mechanism modulating somatic expansion in various polyglutamine diseases, and FAN1 knockout has been shown to increase the rate of somatic expansion in the CGG repeat in FXS. Collectively, these results indicate that FAN1 expression has a dose-dependent protective effect on repeat expansion, providing a credible mechanism for its defensive influence in vivo and suggesting that DNA maintenance is a common driver of pathogenesis in repeat expansion disorders.

Despite its importance for both the onset and progression of repeat expansion disease, the molecular relationship between MM R proteins and FAN1 is not well understood. The mechanisms by which FAN1 protects against repeat instability remain unclear.

Current therapeutic strategies focus on mitigating symptoms associated with repeat expansion diseases after disease onset, and there are no clinically approved therapeutics that cure or even delay the onset of repeat expansion diseases. Thus, there exists an urgent and significant unmet need in the art for effective therapies that treat, prevent or delay the onset of repeat expansion diseases.

SUMMARY OF THE INVENTION

The inventors have overcome the above problems by identifying the molecular mechanisms underlying FAN1 mediated inhibition of somatic expansion. In more detail, the inventors discovered that FAN1 inhibits somatic expansion by two distinct functions: (1) by sequestering MLH1 thereby preventing MLH1 interacting with MSH3; and (2) by promoting accurate DNA repair via its nuclease activity. Advantageously, increasing or replicating these FAN1 functions significantly inhibits somatic expansion thereby providing a new and unexpected therapeutic strategy for treating, preventing or delaying the onset of repeat expansion diseases.

The invention provides a composition for use in treating, preventing or delaying the onset of a repeat expansion disease in a subject, wherein the composition comprises an MLH1 inhibitor and/or a FAN1 derived nuclease.

In one embodiment, the repeat expansion disease is selected from myotonic dystrophy (DM1 and DM2), amyotrophic lateral sclerosis and frontotemporal dementia caused by somatic expansion in the C90RF72 gene, Huntington's disease, spinocerebellar ataxias (SCAs 1, 2, 3, 6, 7 and 17), Friedreich's ataxia (FRDA), Fragile X Tremor Ataxia Syndrome (FXS/FXTAS), Fragile X Syndrome, dentatorubral-pallidoluysian atrophy (DRPLA), spinal and bulbar muscular atrophy (SBMA), and Unverricht-Lundborg myoclonic epilepsy (EPM1). In one embodiment, the repeat expansion disease is Huntington's Disease. In one embodiment, the repeat expansion disease is Fragile X Syndrome.

In one embodiment, the composition comprises an MLH1 inhibitor. In one embodiment, the MLH1 inhibitor is selected from a small molecule, a peptide, a cyclic peptide, an aptamer, or a peptidomimetic. In one embodiment, the MLH1 inhibitor is a cyclic peptide.

In one embodiment, the MLH1 inhibitor comprises an MLH1-binding fragment of FAN1. In one embodiment, the MLH1 inhibitor is a peptide comprising the amino acid sequence SPYF. In one embodiment, the MLH1 inhibitor comprises a peptide having at least 70% sequence identity to residues 120-140 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide having at least 70% sequence identity to residues 73-165, residues 73-190, residues 73-349, residues 1-165, residues 1-190, and/or residues 1-349 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor binds directly to MLH1, optionally wherein the MLH1 inhibitor binds directly to the 52 site of MLH1.

In one embodiment, the MLH1 inhibitor promotes MLH1 binding to FAN1. In one embodiment, the MLH1 inhibitor comprises a kinase, and wherein the kinase promotes MLH1 binding to FAN1 by phosphorylating the FAN1 SPYF motif. In one embodiment, the MLH1 inhibitor comprises a phosphatase, and wherein the phosphatase promotes MLH1 binding to FAN1 by dephosphorylating the FAN1 SPYF motif In one embodiment, the MLH1 inhibitor stabilises MLH1 binding to FAN1, or a fragment thereof. In one embodiment, the MLH1 inhibitor is an allosteric stabiliser of the FAN1-MLH1 interaction. In one embodiment, the MLH1 inhibitor is a direct stabiliser of the FAN1-MLH1 interaction.

In one embodiment, the composition comprises a FAN1 derived nuclease.

In one embodiment, the nuclease does not bind MLH1. In one embodiment, the nuclease comprises at least 70% sequence identity to residues 893-1008 of SEQ ID NO: 2. In one embodiment, the nuclease comprises at least 70% sequence identity to SEQ ID NO: 2. In one embodiment, the nuclease does not comprise an SPYF domain.

The invention also provides a vector for use in treating, preventing or delaying the onset of a repeat expansion disease in a subject, wherein the vector comprises a nucleic acid sequence encoding an MLH1 inhibitor and/or a nucleic acid sequence encoding a FAN1 derived nuclease. In one embodiment, the repeat expansion disease is selected from myotonic dystrophy (DM1 and DM2), amyotrophic lateral sclerosis and frontotemporal dementia caused by somatic expansion in the C9ORF72 gene, Huntington's disease, spinocerebellar ataxias (SCAs 1, 2, 3, 6, 7 and 17), Friedreich's ataxia (FRDA), Fragile X Tremor Ataxia Syndrome (FXS/FXTAS), Fragile X Syndrome, dentatorubralpallidoluysian atrophy (DRPLA), spinal and bulbar muscular atrophy (SBMA), and UnverrichtLundborg myoclonic epilepsy (EPM1). In one embodiment, the repeat expansion disease is Huntington's Disease. In one embodiment, the repeat expansion disease is Fragile X Syndrome.

In one embodiment, the vector comprises a nucleic acid sequence encoding an MLH1 inhibitor of the invention. In one embodiment, the vector comprises a nucleic acid sequence encoding a FAN1 derived nuclease of the invention. In one embodiment, the vector comprises a nucleic acid sequence encoding an MLH1 inhibitor of the invention and a nucleic acid sequence encoding a FAN1 derived nuclease of the invention.

In one embodiment, the vector is selected from an adeno-associated virus (AAV) vector, a HIV-based lentivirus vector, equine immunodeficiency virus (Ely) vector, a feline immunodeficiency virus (Fly) vector, and a herpes simplex virus vector. In one embodiment, the vector is an AAV vector.

In one embodiment, the composition or the vector is formulated for delivery to the striatum and/or the cortex of the subject. In one embodiment, the composition or the vector comprises a targeting ligand. In one embodiment, the targeting ligand facilitates uptake of the composition and/or the vector through the blood brain barrier. In one embodiment, the targeting ligand comprises a compound that facilitates delivery to and/or uptake by neurons.

The invention also provides a method of treating or preventing a repeat expansion disease in a subject comprising administering to the subject the composition of the invention or the vector of the invention.

The invention also provides a method of identifying MLH1 inhibitors comprising: (a) culturing cells expressing MLH1 and MSH3 in the presence of an agent; (b) purifying MLH1 and proteins bound thereto; (c) determining the level of MSH3 that is bound to MLH1; and (d) comparing the level of MLH1-bound MSH3 to a control level.

In one embodiment, the control level is the level of MLH1-bound MSH3 in the absence of the agent and a reduced level of MLH1-bound MSH3 in the presence of the agent indicates that the agent is an MLH1 inhibitor.

In one embodiment, the control level is the level of MLH1-bound MSH3 in the absence of the agent and the same or higher level of MLH1-bound MSH3 in the presence of the agent indicates that the agent is not an MLH1 inhibitor.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1. The FAN1 N-terminal region (p.73-349) mediates its interaction with MLH1 and its effect on CAG stabilisation activity (A) Co-immunoprecipitation (co-IP) extracts from human HD induced pluripotent stem cells (iPSCs) showing FAN1 interacts with MutLa components MLH1 and PMS2. Note MSH3 is absent from antiFAN1 IP fraction. (B) Co-IP extracts from human HD lymphoblasts confirming FAN1 interacts with MLH1. (C) Pull down assays using GFP-Trap beads in U2OS cells showing FAN1 interacts with MutL components (MLH1 and PMS2, PMS1 or MLH3) but not MutS components (MSH3 and MSH3; and MSH2 and MSH6) or proliferating cell nuclear antigen (PCNA). FAN11-cells act as a negative control, demonstrating specificity of the pulldown. (D) Crosslinks identified between FAN1, MLH1 and PMS2 in unstimulated HEK203T cells and HD lymphoblasts. Grey parts on the proteins are structurally unsolved (no PDB structure available). Grey lines = interprotein crosslinks; intraprotein crosslinks indicated by arrows; dashed black lines = crosslinks close to the SPYF motif (See also: Figure SA, Table 1). (E) Schematic illustrating FAN1 constructs cloned into U2OS system. Locations of UBZ-null (C44A/C47A) and nuclease-null (D960A) mutations are also outlined. UBZ = Ubiquitin-binding zinc-finger domain; SAP = SAF-A/B, Acinus and PIAS domain; TPR = tetratricopeptide repeat domain; VRR_NUC = virus-type replication-repair nuclease domain. (F) Pull down using GFP-Trap beads in U2OS cells expressing GFP-FAN1 deletion constructs. FAN1A73-349 did not interact with MLH1. Note that inactivation of UBZ or VRR_NUC domains (C44A/C47A and D960A mutants, respectively) does not affect FAN1-MLH1 interaction. (G) Co-IP extracts using cEMLH1 antibody in U205 cells showing N-terminal FAN11-349 is sufficient to interact with MLH1. (H) MMC viability curves in U2OS cells expressing FAN1 variants (mean ± SD) showing lower viability when FAN1 lacks an intact nuclease domain (See also: Figure 5E). (I) CAG expansion rates in U205 cells expressing truncated FAN1 constructs, including mutations within key functional domains (UBZ -C44A/C47A; VRR_NUC -D960A). Note that only WTA73-349 shows a higher expansion rate than cells expressing FAN1 wT but does not equate to FANli (mean ± SEM, n=2-5 independent experiments, F(5,97)=40.8, p<0.001 by one-way ANOVA with FDR correction of 5%). ***p<0.001, ns = non-significant. (1) Fragment analysis traces illustrating expansion of the exogenous HIT 118 CAG repeat in U2OS cells expressing FAN1 constructs over 6 weeks in culture with time courses plotted (K,L). Note that cells expressing FAN1 Wr73-349 (individual data points shown) expand at a rate between that of FAN1wr or FAN11-349 (individual data points shown) and [AMY-cells (mean ± SD, 95% Cl in shaded areas, n=2-5 independent experiments).

Figure 2. A conserved SPYF motif in FAN1 is required for MLH1-binding (A,B) Co-IP extracts using GFP-Trap beads in U2OS cells expressing truncated FAN1 constructs with quantification (C) showing progressively longer FAN1 N-terminal fragments bind more MLH1. Note residues 120-140 are important for MLH1-binding. (mean ± SEM, n=4-5 independent experiments, F(5,22)=88.31, p<0.001 by one-way ANOVA with FDR correction of 5%). *p<0.05, ***p<0.001, ns = non-significant. (D) Conservation analysis schematic showing SPYF motif is heavily conserved within common model species (residues with >80% consensus are shaded grey). (E) Schematic illustrating FAN1 constructs with mutations at conserved SPYF residues which were cloned into U2OS system. Nuclease-null mutation (D960A) is also outlined. UBZ = Ubiquitin-binding zinc-finger domain; SAP = SAF-A/B, Acinus and PIAS domain; TPR = tetratricopeptide repeat domain; VRR_NUC = virus-type replication-repair nuclease domain. (F) MMC viability curves in U205 cells expressing FAN1 SPYF mutants (mean ± SD). Note viability is only reduced in FAN11 line (See also: Figure 5F). (G) Input and GFP-Trap pull down fractions from U2OS cell extracts expressing FAN1 SPYF mutants with quantification (1-1) showing reduced MLH1-binding with mutation of SPYF motif relative to WT construct. 0123A is displayed as a control, having a mutation outside the conserved motif. (mean ± SEM, n=5 independent experiments; F(4,17)=744.6, p<0.001 by one-way ANOVA with FDR correction of 5%). ***p<0.001.

Figure 3. FAN1 SPYF motif and nuclease activity stabilise the HTT CAG repeat (A) CAG expansion rates in U2OS cells expressing FAN1 constructs with mutations at conserved SPYF motif. Note that mutation of this domain results in hastened expansion of the HIT CAG repeat. 0123A is displayed as a control, having a mutation outside the conserved motif. (mean ± SEM, n=2-5 independent experiments, F(5,83)=28.64, p<0.001 by one-way ANOVA with FDR correction of 5%). **p<0.01, ***p<0.001, ns = non-significant. (B) CAG expansion rates in U2OS cells expressing truncated N-terminal constructs of FAN1, showing residues 120-140 contribute significantly to HIT CAG repeat stability. (mean ± SEM, n=2-5 independent experiments, F(5,86)=22.38, p<0.001 by one-way ANOVA with FDR correction of 5%). *p<0.05, ***p<0.001, ns = non-significant. (C) M MC viability curves in U2OS cells expressing FAN1F129A and FAN1F129A/D960A mutants (mean + SD, n=6-7 independent experiments). Note resistance to MMC toxicity is only maintained in the F129A line (See also: Figure 5G). (D) Input and GFP-Trap pull down fractions from U205 cell extracts expressing FAN1wT and FAN1F129A/D960A showing reduced MLH1-binding with mutation of SPYF motif relative to WT. Note equivalent FAN1wT and FAN1F129A/D960A expression (n=2 independent experiments). (E) Fragment analysis traces illustrating expansion of the exogenous HIT 118 CAG repeat in U205 cells expressing FAN 129A or FAN1F129A/D960A mutants over 6 weeks in culture with time courses plotted (F; mean + SD, 95% Cl in shaded areas) and quantified (G). Cells expressing FAN1F129A/D960A show equivalent expansion as FAN1 4-cells. (mean ± SEM, n=2-5 independent experiments, F(3,72)=39.27, p<0.001 by one-way ANOVA with FDR correction of 5%). **p<0.01, ***p<0.001, ns = non-significant.

Figure 4. FAN1 regulates mismatch repair (MMR) activity through MLH1-binding (A,B) Western blots showing MMR protein expression in U2OS MLH1 (A) and MSH3 (B) knockout lines. (C) CAG expansion rates in FAN1, MLH1-/-and MSH3-/-U205 cell lines. Note, knockout of MSH3 or MLH1 ablates CAG repeat expansion (mean ± SEM, n=2-5 independent experiments, F(2,72)=272.5, p<0.001 by one-way ANOVA with FDR correction of 5%). ***p<0.001, ns = nonsignificant. (D) 6TG viability curves in U205 cells expressing FAN1 constructs and MLH1, showing cells with an intact FAN1 SPYF motif have enhanced resistance to 6TG, indicating reduced mismatch repair activity. MLH11 cells serve as a control (mean ± SD, n=5 independent experiments) (See also: Figure 5H). (E) Co-IP of MLH1 and binding partners from FAN11" and FAN1wT cells. Note FAN1 expression reduces MSH3 levels in MLH1 IP fractions but does not affect PMS2, with quantification (F; mean ± SEM, n=4 independent experiments, t(6)=5.2, p=0.001 by independent-samples t-test). **p<0.01. (G) ChIP extracts from FAN1wT and FAN1'-U2OS cells, immunoprecipitated with aMLH1 antibodies and DNA amplified with primers targeting HIT CAG repeat region. Note the decreased levels of both long (exogenous HIT) and short (endogenous HIT) amplicons in WT ChIP fractions. (H) Fragment analysis traces from U205 FAN14-extracts show the presence of the CAG repeat from the endogenous HTT allele (20 CAG units) and the longer exogenous repeat (118 CAG) from the exon 1 construct in both input and ChIP fractions. The lack of signal in the control IP (-Ab) shows the specificity of the procedure. (I) Quantification of DNA levels in ChIP fractions from WT and FAN1i U20S cells. Primer pairs proximal to the CAG repeat (P1 and P2) and toward the 3' end of 1-I-F (HTT2) were used (mean ± SEM, n=3 independent experiments, P1: F(2,6)=20.76, p=0.002, P2: F(2,6)=17.84, p-0.003, HTT2: F(2,6)=23.56, p=0.001 by one-way ANOVA with FDR correction of 5%). **p<0.01, ***p<0.001, ns = non-significant) *p<0.05, ***p<0.001. (.1) Schematic summarising FAN Vs proposed role in restricting CAG expansion. FAN1 binds the MutLa (MLH1-PMS2) complex, preventing MMRmediated expansion of the CAG repeat. FAN1's nuclease activity also works independently to restrain expansion.

Figure 5.

(A) Identified crosslinks between FAN1, MLH1, PMS2, FANCD2 and FANCI. The crosslink map was generated using xiVIEW. Lines show interprotein crosslinks and intraprotein crosslinks are shown by arrows. (Related to Figure 1D, Table 1). (B) GFP live cell imaging of U205 cells expressing the indicated FAN1-GFP fusion constructs with quantification (C, D). Note the nuclear localization and formation of DNA repair foci in response to MMC in all lines but UBZ cells (mean ± SEM, n=3 independent experiments, [C] WT: t(20)=6.54, p<0.001, D960A: t(20)=6.54, p<0.001, UBZ: t(20)=0.27, p=0.789, 1-349: t(20)=2.76, p=0.12, WT813-349: t(20)=4.42, p<0.001, Q123A; [D] WT:t(20)=9.21, p<0.001, Q123A: t(20)=10, p<0.001, 5126A: t(20)=9.99, p<0.001, Y128A: t(20)=10.55, p<0.001, F129A: t(20)=9.83, p<0.001 by independent-samples t-tests) *p<0.05, ***p<0.001, ns = nonsignificant (Related to Figure 1). (E) Quantification of MMC viability curves in U2OS cells expressing FAN1, showing lower viability when FAN1 lacks an intact nuclease domain (mean ± SEM, n=5-8 independent experiments; F(5,204)=36.21, p<0.001 by one-way ANOVA with FDR correction of 5%). ***p<0.001, ns = non-significant (Related to Figure 1H). (F) Quantification of MMC viability curves in U205 cells expressing FAN1 SPYF mutants, showing no significant difference in viability to WT cells (mean ± SEM, n=6-8 independent experiments; F(5,257)=14.25, p<0.001 by one-way ANOVA with FDR correction of 5%). ***p<0.001, ns = non-significant (Related to Figure 2F). (G) Quantification of MMC viability curves in U205 cells expressing FAN 1F129AID960A mutant, showing no significant difference in viability to FANVF cells (mean ± SEM, n=6-7 independent experiments; F(2,112)=42.1, p<0.001 by one-way ANOVA with FDR correction of 5%). ***p<0.001, ns = non-significant (Related to Figure 3C). (H) Quantification of 6TG viability curves in U205 cells expressing FAN1 SPYF mutants (mean ± SD). Note that FAN1 SPYF mutants have decreased 6TG resistance, similar to FAN1, whereas FAN1wT 6TG resistance approaches but does not reach MLH1-/-levels. (mean ± SEM, n=5 independent experiments, F(4,106)=23.09, p<0.001 by one-way ANOVA with FDR correction of 5%) . * p<0.05, ***p<0.001, ns = non-significant (Related to Figure 4D). (I) 6TG viability curves in U205 cells expressing FAN1 deletion constructs (mean ± SD) with quantification (.1). Note that FAN1 lacking amino acids 73-349 has decreased 6TG resistance, similar to FAN1, whereas FAN11-349 maintains 6TG resistance but does not reach MLH11 levels. (mean ± SEM, n=5 independent experiments, F(3,104)=13.49, p<0.001 by one-way ANOVA with FDR correction of 5%). ** p<0.01, ***p<0.001, ns = non-significant (Related to Figure 4). (K) Co-IP extracts using aMSH3 antibody in U2OS cells showing reduced MLH1 binding when FAN1 is expressed. Note the absence of FAN1 in the MSH3 IP fraction and the stable levels of MSH2 between lines. (n=4 independent experiments) (Related to Figure 4).

Figure 6. MLH1 and MSH3 knockout causes microsatellite instability at tetra-and di-nucleotide repeats. (Related to Figure 4G-I) Repeat length for both alleles are plotted for U2OS cells of the given genotypes. Each point represents an allele, and genomic loci are labelled, including tetranucleotide (D8S321, D20582, D9S242, MYCL1, D20585), dinucleotide (D2S123, D5S346, D175250, D18S64, D18569) and stable control pentanucleotide (Penta C and Penta D; data not shown) loci. Those showing microsatellite instability (MSI) are circled. MLH1 and MSH3 knockout induces repeat contraction or expansion at tetranucleotide and dinucleotide loci, including D20S85, MYCL1, D20S82, D9S242 and D17S250.

DETAILED DESCRIPTION OF THE INVENTION

To investigate the mechanistic significance of FAN1 and MLH1, the inventors examined the interaction of these genetic modifiers, and their role in somatic expansion in the context of HD as a model repeat expansion disease. The inventors demonstrated that FAN1 directly interacts with MLH1, but not with MSH3. Surprisingly, the inventors discovered that the FAN1 N-terminal region, in particular a SPYF motif (p.126-129), mediates the binding of FAN1 to MLH1, and that this interaction protects against somatic expansion. Despite previous experiments demonstrating that inactivation of FAN1 nuclease activity had no impact on somatic expansion inhibition, the inventors also made the surprising discovery that the nuclease domain of FAN1 contributes to its protective effects.

The inventors have demonstrated that FAN1 N-terminal deletion constructs lacking the SPYF motif fail to stabilise the CAG repeat (i.e. prevent somatic expansion); FAN11-12° accelerates repeat expansion to the same rate as FAN1-/-, whereas longer constructs containing the SPYF motif, including FAN11465, FAN173-165, FAN173-190, FAN11-19°, FAN173-349, or FAN11-349, slow the expansion rate significantly. Consistent with this, deleting residues 120-140 (FAN1al20-140) from the FAN11-349 construct reduces stabilisation activity. SPYF mutations reduce FAN1-MLH1 binding and accelerate repeat expansion. FAN1-MLH1 binding and CAG stabilisation activity correlate closely, indicating they are mechanistically linked. The homology between the FAN1 SPYF and MSH3 M IP-box support the inventors' hypothesis that FAN1 competes with MSH3 for MLH1 binding. A M IP-box is found in several MLH1 interaction partners, including MSH3, EX01 and NTG2, and it has been shown to interact with the C-terminal S2 site of MLH1, a region comprising several conserved residues. The crosslinking results presented herein show that interactions between the FAN1 SPYF motif and MLH1 are clustered at the unstructured central domain of MLH1 and include crosslinks consistent with an interaction near the S2 site. FAN1 binding would therefore sterically inhibit MLH1's interaction with MSH3 and modulate MutSI3 (MSH2 and MSH3) driven MMR activity. The close association between FAN1, MLH1 and PMS2 demonstrates that FAN1 interacts functionally with the MutLa (MLH1 and PMS2) complex. Consistent with previous data from mouse models, the inventors have demonstrated that MLH1 or MSH3 knockout prevents CAG repeat expansion, showing the absolute requirement of MutS13-driven MMR for CAG repeat expansion. In summary, the inventors have discovered that FAN1 competes with MSH3 for MLH1 binding, thereby preventing MMR-driven somatic expansion.

Cells defective in MMR are resistant to 6-thioguanine (6TG) toxicity and display microsatellite instability. MLH1-/-U2OS cells are resistant to 6TG and show instability at an EMAST locus in the genome indicating these cells have dysregulated MMR. Interestingly, cells overexpressing FAN1 with an active SPYF domain showed significantly increased resistance to 6TG, as compared to FAN1-/-cells. These cells did not show alterations at EMAST loci which likely reflects the partial inhibition of MMR activity and the relatively short time course of the assay. Importantly, FAN1 constructs lacking an active SPYF motif did not protect against 6TG toxicity, showing that MLH1-binding underlies FANrs regulation of MMR activity. This is interesting because it suggests that FAN1 may be modulating both MutSa (MSH2 and MSH6) and MutS13-driven MMR activity. The inventors have demonstrated that FAN1 sequesters MLH1 and prevents interaction with MSH3. The lack of MSH2 and MSH6 in anti-FAN1 IP fractions confirms earlier reports that these proteins do not directly interact and suggests that a similar mechanism may operate to regulate MutSa-MLH1 interactions. MMR interactions with the FA-pathway and FAN1 itself have been reported previously but direct inhibition of MMR, mediated by MLH1 sequestration, has not. It is evident from experiments in mouse models that FAN1 and MLH1 interact genetically and play a crucial role in regulating somatic expansion, likely by modulating MMR activity.

A role of FAN1 in protecting against somatic instability has previously been shown to function independently of its nuclease activity. However, the inventors have made the surprising discovery that the FAN1 nuclease domain does contribute to FAN1 repeat stabilisation activity. The FAN1F12gAmgem SPYF and nuclease double mutant demonstrated that FAN1-MLH1 binding and nuclease activity have independent, but additive, repeat-stabilising effects. p.D960A nuclease inactivation has previously been shown not to affect repeat instability, however the inventors believe that overexpression of FAN1 mutants in U2OS cells might have masked the subtle contribution of the nuclease domain by sequestering most available MLH1 and shutting down error-prone MMR. In the absence of this dominant activity, for example following SPYF mutation, the stabilisation activity of the nuclease domain can be observed. In this scenario, FANVs nuclease activity could operate downstream of MSH3-mediated recruitment of MLH1, regulating the repair process to reduce errant CAG incorporation, possibly by acting directly on the DNA. This proposal is supported by data showing FAN1 binds directly to CAG repeat DNA.

Considering the significance of these findings and the shared influence of FAN1 across multiple trinucleotide disorders, therapeutics that sequester or inhibit MLH1 or increase FAN1 nuclease activity will have wide-ranging clinical relevance for repeat expansion diseases.

Somatic expansion Somatic expansion, or repeat expansion, is the process by which short tandem repeats within a repetitive region of DNA are expanded thereby increasing the length of the repeat regions. Regions of DNA that are susceptible to somatic expansion are said to be somatically unstable. The rate of somatic expansion is tissue-specific and can also vary significantly between individuals.

Somatic expansion plays a crucial role in the pathogenesis of repeat expansion diseases because the length of repeat regions is the main determinant of age of disease onset as well as the rate and severity of disease progression. Therapeutically targeting somatic expansion provides the most promising method for not only treating repeat expansion diseases, but also delaying the onset of or preventing disease onset and progression.

DNA repair, particularly mismatch repair (MMR), is the major driving force of somatic expansion. MMR driven somatic expansion is thought to require the MutS(3 heterodimer (MSH3-MSH2) which recognises large loops in slipped DNA and recruits MutLa (MLH1-PMS2) which incises the DNA through its endonuclease activity. Thereafter, repair is conducted by a DNA polymerase and ligase 1 (LIG1), and additional repeat units are incorporated.

MLH1 is an important part of the MMR MutL endonuclease complex. Reduced expression of MLH1 has been associated with later onset of HD, and studies in mice have shown that MLH1 is required for somatic instability. MLH1 heterodimerizes with PMS2, PMS1 or MLH3 to form the MutLa, MutL13 or MutLy mismatch repair endonuclease complexes, respectively.

Human wild type MLH1 is represented by SEQ ID NO: 1: MSFVAGVIRRLDETVVNRIAAGEVIQRPANAIKEMIENCLDAKSTSIQVIVKEGGLKLIQIQDNGTGIRKEDLD IVCER FTTSKLQSFEDLASISTYGFRGEALASISHVAHVTITTKTADGKCAYRASYSDGKLKAPPKPCAGNQGTQITVE DLFYN IATRRKALKN PSEEYGKILEVVGRYSVHNAGISFSVK KQGETVADVRTLPNASTVDN I RSI FGNAVSRELIEIGCEDKTL AFK MNGYISNANYSVKKCIFLLF I NHRLVESTSLRKAIETVYAAYLP KNTHPFLYLSLEISPQNVDVNVHPTKHEVH FL HEESILERVQQHIESKLLGSNSSRMYFTQTLLPGLAGPSGEMVKSTTSLTSSSTSGSSDKVYAHQMVRTDSREQ KLD AFLOPLSKPLSSQPQAIVTEDKTDISSGRARQQDEEMLELPAPAEVAAKNQSLEGDTTKGTSEMSEKRGPTSSN PRK RHREDSDVEMVEDDSRKEMTAACTPRRRIINLTSVLSLQEEINEUGHEVLREMLHNHSFVGCVNPCMALAQHQT KLYLLNTTKLSEELFYQILIYDFANFGVLRLSEPAPLFDLAMLALDSPESGWTEEDGPKEGLAEYIVEFLKKKA EM LAD YFSLEIDEEGNLIGLPLLIDNYVPPLEGLPIFILRLATEVNWDEEKECFESLSKECAM FYSIRKQYISEESTLSGQQSEVPG SI PNSWKWTVEH IVYKALRSHILPPKHFTEDGNILQLANLPDLYKVFERC (SEQ ID NO: 1) MSH3 has been identified as a key driver of pathogenesis of various repeat expansion diseases, including HD, DM1, and FXS. MSH3 identifies mispaired bases or DNA loop-outs and initiates MMR. Decreased expression of MSH3 in the cortex has been associated with later onset of HD, while increased expression has been associated with increased somatic expansion and earlier disease onset. MSH3 knockout in HD mice prevents somatic expansion.

The rate of somatic expansion can be measured using methods known in the art and described herein. Typically, the rate of repeat expansion involves measuring the length of the repeat region over time. FAN1

FAN1 is an endonuclease and 5'-3' exonuclease which excises aberrant interstrand crosslinks (ICL) that impair transcription and ensures the recovery of stalled replication forks. Human wild type FAN1 is represented by SEQ ID NO: 2: MMSEGKPPDKKRPRRSLSISKNKKKASNSIISCFNNAPPAKLACPVCSKMVPRYDLNRHLDEMCANNDFVQVDP G QVGLI NSNVSM VDLTSVTLEDVTPKKSPPPKTN LTPGQSDSAKREVKQKISPYFKSNDVVCKNQDELRNRSVKVICL GSLASKLSRKYVKAKKSIDKDEEFAGSSPOSSKSTVVKSLIDNSSEIEDEDQILENSSQKENVFKCDSLKEECI PEHMVR GSKIM EAESQKATRECEKSALTPGFSDNAIMLFSPDFTLRNTLKSTSEDSLVKQECIKEVVEKREACHCEEVKMTVAS EAKIQLSDSEAKSHSSADDASAWSNIQEAPLQDDSCLNNDIPHSIPLEQGSSCNGPGQTTGHPYYLRSFLVVLK TVL EN EDDMLLFDEQEKGIVTKFYQLSATGQKLYVRLFQRK LSWIKMTKLEYEEIALDLTPVIEELTNAGFLQTESELQELS EVLELLSAPELKSLAKTFHLVNPNGQKQQLVDAFLKLAKQRSVCTWGKNKPGIGAVILKRAKALAGQSVRICKG PRA VFSRILLLFSLTDSMEDEDAACGGQGQLSTVLLVN LGRMEFPSYTINRKTHIFQDRDDLIRYAAATHM LSDISSAMA NGNWEEAKELAQCAKRDWNRLKNHPSLRCHEDLPLFLRCFTVGWIYTRILSRFVEILQRLH MYEEAVRELESLLSQR IYCPDSRGRWWDRLALNLHQHLKRLEPTIKCITEGLADPEVRTGHRLSLYQRAVRLRESPSCKKEKHLFQQLPE MAV QDVKHVTITGRLCPQRGMCKSVFVM EAGEAADPTIVLCSVEELALAHYRRSGFDQGIHGEGSTFSTLYGLLLWDIIF M DGI PDVERNACQAFPLDLCTDSFFTSRRPALEARLQL1H DAPEESLRAWVAATWHEQEGRVASLVSWDRFTSLQ QAQDLVSCLGGPVLSGVCRHLAADFRHCRGGLPDLVVWNSQSRHFKLVEVKGPNDRLSHKQMIWLAELQKLGAE VEVCHVVAVGAKSQSLS (SEQ ID NO: 2) FAN1 has been identified as the most significant genetic modifier of somatic expansion: increased FAN1 expression has been associated with delayed disease onset; and reduced FAN1 expression has been associated with increased rates of somatic expansion. Despite its significance, the functions by which FAN1 inhibits somatic expansion were previously unknown. The inventors have made the surprising discovery that FAN1 inhibits somatic expansion by: (1) inhibiting MLH1 interaction with MSH3; and (2) promoting accurate repair via its nuclease activity.

MLH1 inhibitors As used herein, 'MLH1 inhibitor' refers to an agent that reduces, inhibits or prevents interaction between MLH1 and MSH3. In one embodiment, the MLH1 inhibitor is a small molecule. In one embodiment, the MLH1 inhibitor is a peptide. In one embodiment, the MLH1 inhibitor is an aptamer. In one embodiment, the MLH1 inhibitor is a peptidomimetic. In one embodiment, the MLH1 inhibitor is a cyclic peptide.

In one embodiment, MLH1 inhibitors of the invention mimic or promote the MLH1 sequestering activity of FAN1. MLH1 inhibitors compete with MSH3 for interaction with MLH1 thereby reducing or inhibiting the formation of the MMR complex and subsequent somatic expansion. In one embodiment, MLH1 inhibitors interact directly with MLH1. In one embodiment, MLH1 inhibitors interact directly with the C-terminal 52 site of MLH1. The C-terminal 52 site of MLH1 is a highly conserved binding site of MLH1 encompassing several conserved amino acids within the C-terminal sequence of SEQ ID NO: 1. The 52 site is required for MLH1 interaction with the MLH1-interacting peptide box (MIP-box) and is known in the art (see e.g. Dherin et al. Mol. Cell. Biol. 2009;29(3):907918 and Gueneau et al. Nat Struct Mol Biol. 2013:20(4):461-8). In one embodiment, the MLH1 inhibitor interacts with at least one of residues L503, 5505, D567, F568, N570, M621, D624, Y625, Y626, E669 and L676 of SEQ ID NO: 1. In one embodiment, the MLH1 inhibitor interacts with at least one of residues 5505, F568, M621, D624, Y625, and E669 of SEQ ID NO: 1.

In one embodiment, the MLH1 inhibitor comprises FAN1 or an MLH1-binding fragment thereof. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 70% sequence identity to residues 120-140 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 75%, 80%, 85%, 90% or 95% sequence identity to residues 120-140 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising residues 120-140 of SEQ ID NO: 2.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 70% sequence identity to residues 73-165 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to residues 73-165 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising residues 73-165 SEQ ID NO: 2.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 70% sequence identity to residues 73-190 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to residues 73-190 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising residues 73-190 of SEQ ID NO: 2.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 70% sequence identity to residues 73-349 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to residues 73-349 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising residues 73-349 of SEQ ID NO: 2.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 70% sequence identity to residues 1-140 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and haying at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to residues 1-140 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising residues 1-140 of SEQ ID NO: 2.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 70% sequence identity to residues 1-165 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to residues 1-165 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising residues 1-165 of SEQ ID NO: 2.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 70% sequence identity to residues 1-190 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and haying at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to residues 1-190 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising residues 1-190 of SEQ ID NO: 2.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 70% sequence identity to residues 1-349 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising a SPYF domain and having at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to residues 1-349 of SEQ ID NO: 2. In one embodiment, the MLH1 inhibitor comprises a peptide comprising residues 1-349 of SEQ ID NO: 2.

In one embodiment, the MLH1 inhibitor comprises a peptide comprising the SPYF domain of SEQ ID NO: 2 and at least 3 contiguous amino acids located at the immediate N-terminus of the SPYF domain and/or the immediate C-terminus of the SPYF domain. For example, an MLH1 inhibitor comprising a peptide comprising the SPYF domain of SEQ ID NO: 2 and at least 3 contiguous amino acids located at the immediate N-terminus of the SPYF domain comprises the following sequence: QKISPYF. An MLH1 inhibitor comprising a peptide comprising the SPYF domain of SEQ ID NO: 2 and at least 3 contiguous amino acids located at the immediate C-terminus of the SPYF domain comprises the following sequence: SPYFKSN. An MLH1 inhibitor comprising a peptide comprising the SPYF domain of SEQ ID NO: 2 and at least 3 contiguous amino acids located at the immediate N-terminus of the SPYF domain and the immediate C-terminus of the SPYF domain comprises the following sequence: QKISPYFKSN. In one embodiment, the MLH1 inhibitor comprises a peptide comprising the SPYF domain of SEQ ID NO: 2 and at least 5, at least 10, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100 contiguous amino acids located N-terminus and/or C-terminus of the SPYF domain.

Herein, amino acid residue numbering corresponds to the position of the amino acid(s) within the relevant sequence, when read in the N-to C-direction.

In one embodiment, the MLH1 inhibitor of the invention competes with MSH3 and FAN1 for interaction with MLH1. The inventors discovered that FAN1 variants that are unable to bind MLH1 significantly inhibit somatic expansion via their nuclease activity. Without wishing to be bound by theory, the inventors believe that MLH1 inhibitors that sequester MLH1 independently of FAN1 advantageously increase the nuclease activity of FAN1.

In one embodiment, the MLH1 inhibitor comprises the chemical formula: In one embodiment, the MLH1 inhibitor comprises the chemical formula: In one embodiment, the MLH1 inhibitor of the invention promotes MLH1 interaction with FAN1 thereby increasing FAN1's ability to sequester MLH1. In turn, this reduces or prevents MLH1 interaction with MSH3, thereby reducing or preventing formation of the MutSp complex. In one embodiment, the MLH1 inhibitor is a FAN1 activator. As used herein, a "FAN1 activator" is any compound or agent capable of initiating and/or promoting interaction between FAN1 and MLH1. In one embodiment, the FAN1 activator upregulates the expression of FAN1.

Phosphorylation of the SPYF motif is believed to modulate the MLH1-binding capacity of FAN1. In one embodiment, the MLH1 inhibitor comprises a kinase wherein the kinase phosphorylates the FAN1 SPYF motif. In one embodiment, the kinase phosphorylates the serine residue of the FAN1 SPYF motif. In one embodiment, the kinase is a serine-threonine protein kinase. In one embodiment, the kinase is a cyclin-dependent kinase. In one embodiment, the kinase is the cyclin-dependent kinase 5. In one embodiment, the kinase phosphorylates the tyrosine residue of the FAN1 SPYF motif. In one embodiment, the kinase is a tyrosine protein kinase.

In one embodiment, the MLH1 inhibitor comprises a phosphatase wherein the phosphatase dephosphorylates the FAN1 SPYF motif. In one embodiment, the phosphatase dephosphorylates the serine residue of the FAN1 SPYF motif. In one embodiment, the phosphatase is a serine-threonine phosphatase. In one embodiment, the phosphatase dephosphorylates the tyrosine residue of the FAN1 SPYF motif. In one embodiment, the phosphatase is a tyrosine phosphatase. In one embodiment, the phosphatase is a protein phosphatase 1.

In one embodiment, the MLH1 inhibitor of the invention stabilises the FAN1-MLH1 interaction. As used herein, stabilising the FAN1-MLH1 interaction means reducing or preventing dissociation of MLH1 from FAN1. Increased stabilisation of the FAN1-MLH1 interaction reduces the amount of MLH1 that is available for interaction with MSH3. Reduced interaction between MLH1 and MSH3 advantageously reduces or inhibits somatic expansion. In one embodiment, the MLH1 inhibitor is an allosteric stabiliser of the FAN1-MLH1 interaction. An allosteric stabiliser binds to either FAN1 or MLH1 and increases the interaction affinity of the FAN1-MLH1 interaction. In one embodiment, the MLH1 inhibitor is a direct stabiliser of the FAN1-MLH1 interaction. A direct stabiliser binds to both FAN1 and MLH1 to increase the interaction affinity of the interaction. In one embodiment, the direct stabiliser binds the interface between MLH1 and FAN1. The interaction affinity between MLH1 and FAN1 can be measured using methods known in the art. For example, surface plasmon resonance analysis may be used to measure the dissociation constant (Kd).

In one embodiment, the MLH1 inhibitor of the invention comprises FAN1 derived nuclease activity.

The invention also provides a composition for use in treating, preventing or delaying the onset of a repeat expansion disease in a subject, wherein the composition comprises one or more MLH1 inhibitors as defined above.

In one embodiment, the composition further comprises one or more FAN1 derived nucleases of the invention.

Small molecules As used herein, small molecules are low molecular weight compounds, typically organic compounds with a maximum molecular weight of 900 Da, allowing for rapid diffusion across cell membranes. In some embodiments, the maximum molecular weight of a small molecule is 500 Da. Methods of producing small molecules are known in the art. Libraries of small molecules can be tested for their ability to reduce, inhibit or prevent MLH1 interaction with MSH3 using methods described herein.

Cyclic peptides Cyclic peptides are polypeptide chains that form a cyclic ring structure. Cyclic peptides may by monocyclic, i.e. comprising a single ring structure, or polycyclic, i.e. comprising several ring structures. Cyclic peptides may be naturally occurring or synthetic. Advantageously, cyclic peptides are less susceptible to proteolysis than their linear counterparts. Cyclic peptides may comprise L or D amino acids, or a mix of L and D amino acids. Cyclic peptides may comprise N-methylated amino acids. Cyclic peptides may comprise I3-amino acids. Cyclic peptides may be, partially or fully, a peptidomimetic or peptoid. Cyclic peptides may be lipidated and/or PEG-ylated.

Aptamers Aptamers are generally nucleic acid molecules that bind a specific target molecule. Aptamers can be engineered in vitro, are readily produced by chemical synthesis, possess desirable storage properties, and elicit little or no immunogenicity in therapeutic applications. These characteristics make aptamers particularly useful in pharmaceutical and therapeutic utilities. As used herein, "aptamer" refers in general to a single or double stranded oligonucleotide or a mixture of such oligonucleotides, wherein the oligonucleotide or mixture is capable of binding specifically to a target. Other aptamers having equivalent binding characteristics can also be used, such as peptide aptamers.

In general, aptamers may comprise oligonucleotides that are at least 5, at least 10 or at least 15 nucleotides in length. Aptamers may comprise sequences that are up to 40, up to 60 or up to 100 or more nucleotides in length. For example, aptamers may be from 5 to 100 nucleotides, from 10 to 40 nucleotides, or from 15 to 40 nucleotides in length. Where possible, aptamers of shorter length are preferred as these will often lead to less interference by other molecules or materials. Aptamers may be generated using routine methods such as the Systematic Evolution of Ligands by Exponential enrichment (SELEX) procedure. SELEX is a method for the in vitro evolution of nucleic acid molecules with highly specific binding to target molecules.

The SELEX method involves the selection of nucleic acid aptamers and in particular single stranded nucleic acids capable of binding to a desired target, from a collection of oligonucleotides. A collection of single-stranded nucleic acids (e.g., DNA, RNA, or variants thereof) is contacted with a target, under conditions favourable for binding, those nucleic acids which are bound to targets in the mixture are separated from those which do not bind, the nucleic acid-target complexes are dissociated, those nucleic acids which had bound to the target are amplified to yield a collection or library which is enriched in nucleic acids having the desired binding activity, and then this series of steps is repeated as necessary to produce a library of nucleic acids (aptamers) having specific binding affinity for the relevant target.

Peptidomimetics Peptidomimetics are compounds which mimic a natural peptide or protein with the ability to interact with the biological target and produce the same biological effect. Peptidomimetics are designed to permit molecular interactions similar to the natural molecular, e.g. the interaction between MLH1 and FAN1. Peptidomimetics may have advantages over peptides in terms of stability and bioavailability associated with a natural peptide. Peptidomimetics can have main-or side-chain modifications of the parent peptide designed for biological function. Examples of classes of peptidomimetics include, but are not limited to, peptoids and 13-peptides, as well as peptides incorporating D-amino acids.

FAN1 derived nucleases As used herein, 'FAN1 derived nuclease' refers to a nuclease that possesses the nuclease activity of wild type FAN1. In one embodiment, the FAN1 derived nuclease of the invention comprises the nuclease domain of FAN1. In one embodiment, the FAN1 derived nuclease lacks MLH1 binding activity.

In one embodiment, the FAN1 derived nuclease comprises a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95% or 100% sequence identity to residues 893-1008 of SEQ ID NO: 2. In one embodiment, the FAN1 derived nuclease comprises a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity to SEQ ID NO: 2. In one embodiment, the FAN1 derived nuclease comprises a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 2 wherein the SPYF domain is mutated. In one embodiment, the FAN1 derived nuclease comprises a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 2 wherein the SPYF domain has been deleted.

The invention also provides a composition for use in treating, preventing or delaying the onset of a repeat expansion disease in a subject, wherein the composition comprises a FAN1 derived nuclease of the invention. In one embodiment, the composition further comprises one or more MLH1 inhibitors of the invention.

Vectors The invention also provides a vector for use in treating, preventing or delaying the onset of a repeat expansion disease in a subject, wherein the vector comprises a nucleic acid sequence encoding an MLH1 inhibitor of the invention and/or a FAN1 derived nuclease of the invention.

In one embodiment, the vector comprises a nucleic acid sequence encoding an MLH1 inhibitor of the invention. In one embodiment, the vector comprises a nucleic acid sequence encoding FAN1 or an MLH1-binding fragment thereof. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide comprising a SPYF domain.

In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or 100% sequence identity to residues 120-140 of SEQ ID NO: 2. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity to residues 73-165 of SEQ ID NO: 2. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity to residues 73-190 of SEQ ID NO: 2. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity to residues 73-349 of SEQ ID NO: 2. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity to residues 1-140 of SEQ ID NO: 2. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity to residues 1-165 of SEQ ID NO: 2. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity to residues 1-190 of SEQ ID NO: 2. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity to residues 1-349 of SEQ ID NO: 2.

In one embodiment, the vector comprises a nucleic acid sequence encoding a FAN1 derived nuclease of the invention. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95% or 100% sequence identity to residues 893-1008 of SEQ ID NO: 2. In one embodiment, the vector comprises a nucleic acid sequence encoding a peptide having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to SEQ ID NO: 2 wherein the SPYF domain is mutated.

In one embodiment, the vector comprises a nucleic acid sequence encoding an MLH1 inhibitor of the invention and a nucleic acid sequence encoding a FAN1 derived nuclease of the invention.

In one embodiment, the vector is formulated for delivery to the striatum and/or the cortex of the subject. In one embodiment, the vector comprises a targeting ligand. In one embodiment, the targeting ligand facilitates uptake of the vector through the blood brain barrier. In one embodiment, the targeting ligand comprises a compound that facilitates delivery to and/or uptake by neurons. In one embodiment, the vector comprises an artificial capsid amino acid sequence which enables the viral particle to cross the blood brain barrier. In one embodiment, the vector is able to transfect and be expressed in non-dividing cells, e.g. brain cells.

In one embodiment, the vector is a viral vector. Viral vectors are usually non-replicating or replication-impaired vectors, which means that the viral vector cannot replicate to any significant extent in normal cells (e.g. normal human cells), as measured by conventional means. Non-replicating or replication-impaired vectors may have become so naturally (i.e. they have been isolated as such from nature) or artificially (e.g. by breeding in vitro or by genetic manipulation).

There will generally be at least one cell-type in which the replication-impaired viral vector can be grown -for example, modified vaccinia Ankara (MVA) can be grown in CEF cells. Typically, the viral vector is incapable of causing a significant infection in an animal subject, typically in a mammalian subject such as a human patient. Examples of viral vectors that are useful in this context include attenuated vaccinia virus vectors such as modified vaccinia Ankara (MVA) and NYVAC, or strains derived therefrom. Other suitable viral vectors include poxvirus vectors, such as avipox vectors, for example attenuated fowlpox vectors or canarypox vectors (e.g. ALVAC and strains derived therefrom). Alternative viral vectors useful in the present invention include adenoviral vectors (e.g. non-human adenovirus vectors), alphavirus vectors, flavivirus vectors, herpes viral vectors, influenza virus vectors and retroviral vectors. In one embodiment, the vector is selected from an adeno-associated virus (AAV) vector, a HIV-based lentivirus, equine immunodeficiency virus (Ely), feline inununodeficiency virus (Fly), and herpes simplex virus. In one embodiment, the vector is AAV.

In one embodiment, the vector is an expression vector. Expression vectors are nucleic acid molecules (linear or circular) that comprise one or more polynucleotide sequences encoding a polypeptide(s) of interest, operably linked to additional regulatory elements required for its expression. In this regard, expression vectors generally include promoter and terminator sequences, and optionally one or more enhancer sequences, polyadenylation signals, and the like. Expression vectors may also include suitable translational regulatory elements, including ribosomal binding sites, and translation initiation and termination sequences. The transcriptional and translational regulatory elements employed in the expression vectors of the invention are functional in the host cell used for expression.

Identification of MLH1 inhibitors The invention provides a method of identifying agents that bind directly to MLH1. MLH1 inhibitors may be identified using suitable methods known in the art. In one embodiment, the method is used to identify agents that bind directly to the 52 site of MLH1. In one embodiment, the method comprises performing immunoprecipitation to identify agents that bind directly to MLH1. In one embodiment, the method comprises performing chromatin immunoprecipitation (ChIP) to identify agents that bind directly to MLH1.

In one embodiment, the method comprises screening a library of peptides for MLH1 binding activity, optionally wherein peptides are also screened for MLH1 52 site binding activity. In one embodiment, a library of cyclic peptides is screened for MLH1 binding activity. In one embodiment, the invention provides a high-throughput screening method for MLH1 binding activity. MLH1 binding activity may be measured by immunoprecipitation, Chip, or cell binding assays. Suitable immunoprecipitation assays will typically utilise anti-MLH1 antibodies. MLH1 binding activity may be measured by pull-down assays. MLH1 binding activity may also be measured using affinity chromatography and tagged MLH1, e.g. His-tagged MLH1, or Streptavidin-tagged MLH1. Suitable affinity chromatography, immunoprecipitation and pull-down assays and suitable antibodies and purification tags are known in the art. Cell binding assays may include mammalian two hybrid and protein complementation assays using FAN1 and MLH1 full length or deletion constructs.

The invention provides a method of identifying agents that reduce, inhibit or prevent interaction between MLH1 and MSH3. In one embodiment, the method comprises: (a) culturing cells expressing MLH1 and MSH3 in the presence of an agent; (b) purifying MLH1 and proteins bound thereto; (c) determining the level of MLH1-bound MSH3; and (d) comparing the level of MLH1-bound MSH3 to a control level. In one embodiment, the control level is the level of MLH1-bound MSH3 in the absence of the agent. In this embodiment, a reduced level of MLH1-bound MSH3 in the presence of the agent indicates that the agent reduces, inhibits or prevents MLH1 interaction with MSH3 -such an agent is an MLH1 inhibitor as defined herein. In one embodiment, the same or higher level of MLH1-bound MSH3 in the presence of the agent indicates that the agent is not an MLH1 inhibitor. The skilled person will readily understand that the method of the invention may alternatively comprise purifying M5H3 and proteins bound thereto.

In one embodiment, the cultured cells expressing MLH1 and MSH3 do not express wild type FAN1. In one embodiment, the cultured cells express FAN1 that is mutated to prevent MLH1 interacting with FAN1. In one embodiment, the method further comprises determining the level of MLH1-bound FAN1 and identifying whether the agent inhibits or promotes the FAN1-MLH1 interaction.

An MLH1 inhibitor typically reduces MSH3-MLH1 interaction by at least 10%, as compared to baseline interaction between MSH3 and MLH1. In some embodiments, the MLH1 inhibitor reduces interaction between MSH3 and MLH1 by at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100%. In one embodiment, reduced MSH3MLH1 interaction is determined by reduced formation of the MutSP complex.

The interaction between MLH1 and M5H3 can be measured using methods known in the art, e.g. by IP.

An MLH1 inhibitor that promotes the FAN1-MLH1 interaction typically increases the FAN1-MLH1 interaction by at least 10%, as compared to baseline interaction between FAN1 and MLH1. In some embodiments, the MLH1 inhibitor stimulates at least a 20% increase in interactions between FAN1 and MLH1, such as at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 100%.

As used herein, the interaction of two proteins, e.g. MLH1 and MSH3, encompasses binding of the two proteins Therapeutic indications Repeat expansion diseases are a class of genetic diseases caused by somatic expansion of short tandem repeats. Repeat expansion diseases are also known as repeat expansion disorders, trinucleotide repeat disorders, or microsatellite expansion diseases. A repeat expansion disease according to the invention includes, but is not limited to myotonic dystrophy (DM1 and DM2), amyotrophic lateral sclerosis and frontotemporal dementia caused by somatic expansion in the C90RF72 gene, Huntington's disease, spinocerebellar ataxias (SCAs 1, 2, 3, 6, 7 and 17), Friedreich's ataxia (FRDA), Fragile X Tremor Ataxia Syndrome (FXS/FXTAS), Fragile X Syndrome, dentatorubralpallidoluysian atrophy (DRPLA), spinal and bulbar muscular atrophy (SBMA), and UnverrichtLundborg myoclonic epilepsy (EPM1).

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of a repeat expansion disease in a subject. The invention also provides a method of treating, preventing or delaying the onset of a repeat expansion disease in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of a repeat expansion disease in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of a repeat expansion disease in a subject.

Huntington's disease is a monogenic neurodegenerative condition arising due to inheritance of CAG trinucleotide repeats in exon 1 of the huntingtin (HTT) gene. In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of Huntington's disease in a subject. The invention also provides a method of treating, preventing or delaying the onset of Huntington's disease in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of Huntington's disease in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of Huntington's disease in a subject.

Fragile X Syndrome is a neurodevelopmental disorder caused by repeat expansion of the CGG triplet repeat within the FMR1 (fragile X mental retardation 1) gene on the X chromosome. In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of Fragile X Syndrome in a subject. The invention also provides a method of treating, preventing or delaying the onset of Fragile X Syndrome in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of Fragile X Syndrome in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of Fragile X Syndrome in a subject.

Several repeat expansion diseases involve the somatic expansion of CAG trinucleotides. These disorders are collectively known as polyglutamine (polyQ) diseases and include Huntington's disease (HD), dentatorubral-pallidoluysian atrophy (DRPLA), spinal and bulbar muscular atrophy (SBMA) and the spinocerebellar ataxias (SCAs) 1, 2, 3, 6, 7 and 17. In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of a polyglutamine disease in a subject. The invention also provides a method of treating, preventing or delaying the onset of a polyglutamine disease in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of a polyglutamine disease in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of a polyglutamine disease in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of dentatorubral-pallidoluysian atrophy (DRPLA) in a subject. The invention also provides a method of treating, preventing or delaying the onset of DRPLA in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of DRPLA in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of DRPLA in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of spinal and bulbar muscular atrophy (SBMA) in a subject. The invention also provides a method of treating, preventing or delaying the onset of SBMA in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of SBMA in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of SBMA in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of a spinocerebellar ataxia (SCA) selected from SCA1, SCA2, SCA3, SCA6, SCA7 and SCA17 in a subject. The invention also provides a method of treating, preventing or delaying the onset of a spinocerebellar ataxia (SCA) selected from SCA1, SCA2, SCA3, SCA6, SCA7 and SCA17 in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of a spinocerebellar ataxia (SCA) selected from SCA1, SCA2, SCA3, SCA6, SCA7 and SCA17 in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of a spinocerebellar ataxia (SCA) selected from SCA1, SCA2, SCA3, SCA6, SCA7 and SCA17 in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of myotonic dystrophy (DM1 and DM2) in a subject. The invention also provides a method of treating, preventing or delaying the onset of myotonic dystrophy (DM1 and DM2) in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of myotonic dystrophy (DM1 and DM2) in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of myotonic dystrophy (DM1 and DM2) in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of amyotrophic lateral sclerosis and/or frontotemporal dementia caused by somatic expansion in the C90RF72 gene in a subject. The invention also provides a method of treating, preventing or delaying the onset of amyotrophic lateral sclerosis and/or frontotemporal dementia caused by somatic expansion in the C90RF72 gene in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of amyotrophic lateral sclerosis and/or frontotemporal dementia caused by somatic expansion in the C90RF72 gene in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of amyotrophic lateral sclerosis and/or frontotemporal dementia caused by somatic expansion in the C9ORF72 gene in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of Friedreich's ataxia (FRDA) in a subject. The invention also provides a method of treating, preventing or delaying the onset of FRDA in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of FRDA in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of FRDA in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of Fragile X Tremor Ataxia Syndrome (FXS/FXTAS) in a subject. The invention also provides a method of treating, preventing or delaying the onset of Fragile X Tremor Ataxia Syndrome (FXS/FXTAS) in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of Fragile X Tremor Ataxia Syndrome (FXS/FXTAS) in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of Fragile X Tremor Ataxia Syndrome (FXS/FXTAS) in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in treating, preventing or delaying the onset of Unverricht-Lundborg myoclonic epilepsy (EPM1) in a subject. The invention also provides a method of treating, preventing or delaying the onset of EPM1 in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the treatment or prevention of or to delay the onset of EPM1 in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for treating, preventing or delaying the onset of EPM1 in a subject.

In one embodiment, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is for use in a method of reducing or inhibiting somatic expansion in a subject. The invention also provides a method of reducing or inhibiting somatic expansion in a subject, the method comprising administering to the subject an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention in the reduction, inhibition or prevention of somatic expansion in a subject. The invention also provides use of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention for the manufacture of a medicament for reducing or inhibiting somatic expansion in a subject. In one embodiment, the rate of somatic expansion is reduced. In one embodiment, somatic expansion is prevented. The rate of somatic expansion can be measured using methods known in the art.

The therapeutic use or method of the invention may comprise administering a therapeutically effective amount of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention, either alone or in combination with other therapeutic agents, to a subject. When used in combination with one or more additional therapeutic agent(s) or treatment(s), an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention may be administered before, simultaneously with, or after the administration of the one or more additional therapeutic agent(s) or treatment(s).

The therapeutic use or method of the invention may comprise administering a therapeutically effective amount of an MLH1 inhibitor of the invention in combination with a different MLH1 inhibitor of the invention to a subject. The first MLH1 inhibitor may be administered before, simultaneously with, or after the administration of the second MLH1 inhibitor.

The therapeutic use or method of the invention may comprise administering a therapeutically effective amount of a FAN1 derived nuclease of the invention with a different FAN1 derived nuclease of the invention to a subject. The first FAN1 derived nuclease may be administered before, simultaneously with, or after the administration of the second FAN1 derived nuclease.

The therapeutic use or method of the invention may comprise administering a therapeutically effective amount of an MLH1 inhibitor of the invention in combination with a therapeutically effective amount of a FAN1 derived nuclease of the invention to a subject. The MLH1 inhibitor may be administered before, simultaneously with, or after the administration of the FAN1 derived nuclease of the invention.

As used herein, the term "treatment" or "treating" embraces therapeutic measures. Treatment of a repeat expansion disease can be characterised by a reduced rate of somatic expansion and/or delayed onset of disease and/or disease symptoms. The term "treating" may refer to inhibiting a repeat expansion disease, disorder or condition, i.e. arresting the development thereof; and/or relieving a repeat expansion disease, disorder or condition, i.e. causing regression of the disease, disorder and/or condition.

An MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention may also be used as a preventative therapy. As used herein, the term "preventing" includes preventing the onset and/or the progression of a repeat expansion disease, and/or symptoms associated with a repeat expansion disease. The term "preventing" includes preventing a repeat expansion disease, disorder or condition from occurring in a subject that may be predisposed to the repeat expansion disease, disorder and/or condition but has not yet been diagnosed as having the repeat expansion disease, disorder and/or condition.

As used herein, "delaying the onset of" means increasing the time to onset of the repeat expansion disease or of one or more symptoms of the repeat expansion disease. For example, onset can be said to be delayed when the time to manifestation of one or more symptoms of a repeat expansion disease takes at least 5% longer than would be expected in the absence of treatment with an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention, e.g. an increase in time of at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90% or at least 100%.

When used in therapy, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention can delay the onset, reduce the severity, or ameliorate one or more symptoms of a repeat expansion disease. When used in therapy, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention can prolong the life span of a subject beyond that expected in the absence of such treatment.

As used herein, the term "subject" refers to an animal or any living organism including, but not limited to, members of the human, primate, equine, porcine, bovine, murine, rattus, canine and feline specie. In one embodiment, the subject is a mammal. In one embodiment, the subject is a human. As used herein, the term "patient" may be used interchangeably with "subject" and "human". The subject may have been diagnosed with a repeat expansion disease using known clinical methods. The subject may be identified based on the presence of a genetic and/or biochemical marker of a repeat expansion disease. The subject may be identified as being at risk of a repeat expansion disease based on the presence of genetic markers, e.g. the length of inherited repeat regions. The subject may be asymptomatic.

Genetic and biochemical markers of repeat expansion diseases are known in the art. For example, a genetic marker may comprise a threshold number of repeats in a repeat region of a gene associated with the repeat expansion disease. HD is typically diagnosed by identifying the number of CAG repeats in the HTTgene. In one embodiment, the subject has >35 CAG repeats. In one embodiment, the subject has >39 CAG repeats.

Pharmaceutical compositions and formulations The invention also provides compositions comprising an MLH1 inhibitor, FAN1 derived nuclease and/or vector of the invention. In some embodiments, compositions comprising an MLH1 inhibitor, FAN1 derived nuclease and/or vector of the invention are for use as a medicament. In some embodiments, compositions comprising an MLH1 inhibitor, FAN1 derived nuclease and/or vector of the invention are for use in treating, preventing or delaying the onset of a somatic expansion disease. In some embodiments, compositions comprising an MLH1 inhibitor, FAN1 derived nuclease and/or vector of the invention are for use in delaying the onset of a somatic expansion disease.

An MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention may be formulated for delivery to the striatum and/or the cortex of the subject. An MLH1 inhibitor and/or FAN1 derived nuclease of the invention may be covalently or non-covalently linked to a targeting ligand specifically designed to facilitate the uptake into the cell, cytoplasm and/or nucleus. The targeting ligand may comprise a compound that recognises a cell, tissue or organ specific element facilitating cellular uptake and/or a compound that facilitates uptake into cells. The targeting ligand may facilitate intracellular release of the agent of the invention from vesicles, e.g. endosomes or lysosomes. The targeting ligand may comprise compounds that facilitate uptake of the agent of the invention into the brain through the blood brain barrier. The targeting ligand may comprise compounds that facilitate uptake of the agent of the invention into the striatum and/or cortex.

The invention provides a composition for use in treating, preventing or delaying the onset of a somatic expansion disease wherein the composition comprises a vector, MLH1 inhibitor and/or FAN1 derived nuclease of the invention. The composition of the invention may be combined or administered in addition to a carrier, diluent and/or excipient. Alternatively or in addition the composition of the invention may further be combined with one or more of a salt, excipient, diluent, and/or immunoregulatory agent.

Generally, the carrier is a pharmaceutically-acceptable carrier. Non-limiting examples of pharmaceutically acceptable carriers include water, saline, and phosphate-buffered saline. In some embodiments, the composition may be in lyophilized form, in which case it may include a stabilizer, such as BSA. In some embodiments, it may be desirable to formulate the composition with a preservative, such as thiomersal or sodium azide, to facilitate long term storage. Examples of buffering agents include, but are not limited to, sodium succinate (pH 6.5), and phosphate buffered saline (PBS; pH 6.5 and 7.5).

The composition may be formulated as a neutral or salt form. Pharmaceutically acceptable salts include acid addition salts formed with inorganic acids such as, for example, hydrochloric or phosphoric acids, or with organic acids such as acetic, oxalic, tartaric, maleic, and the like. Salts formed with the free carboxyl groups may also be derived from inorganic bases such as, for example, sodium, potassium, ammonium, calcium, or ferric hydroxides, and such organic bases as isopropylamine, trimethylamine, 2-ethylamino ethanol, histidine, procaine, and the like.

Administration of an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention is generally by conventional routes e.g. intravenous, subcutaneous, intraperitoneal, or mucosal routes. The administration may be by parenteral injection, for example, a subcutaneous, intradermal or intramuscular injection. For example, formulations comprising an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention may be particularly suited to administration intravenously, intramuscularly, intradermally, or subcutaneously. Administration of small molecule MLH1 inhibitors and/or FAN1 derived nucleases of the invention may be by injection, such as intravenously, intramuscularly, intradermally, or subcutaneously, or by oral administration (small molecules with molecular weight of less than 500 Da typically exhibiting oral bioavailability).

Accordingly, an MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention may be prepared as injectables, either as liquid solutions or suspensions. Solid forms suitable for solution in, or suspension in, liquid prior to injection may alternatively be prepared. An MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition of the invention may be encapsulated or embedded in a delivery vehicle. In various aspects, the delivery vehicle is a liposome, a lysosome, a microcapsule, or a nanoparticle.

Oral formulations may include normally employed excipients such as pharmaceutical grades of mannitol, lactose, starch, magnesium stearate, sodium saccharine, cellulose, magnesium carbonate, and the like. These compositions take the form of solutions, suspensions, tablets, pills, capsules, sustained release formulations or powders.

As used herein, a "therapeutically effective amount" means an amount of MLH1 inhibitor, FAN1 derived nuclease, vector and/or composition effective in treating, preventing or delaying the onset of a repeat expansion disease and thus producing the desired therapeutic or preventative effect in a subject.

The dosage ranges for administration of the compounds of the present invention are those which produce the desired therapeutic effect. It will be appreciated that the dosage range required depends on the precise nature of the compound, the route of administration, the nature of the formulation, the age of the subject, the nature, extent or severity of the subject's condition, contraindications, if any, and the judgement of the attending physician. Variations in these dosage levels can be adjusted using standard empirical routines for optimisation. Similarly, the dose of a compound of the invention for use in a method of the invention can be readily determined by one of skill in the art, and is a dose that reduces the rate of somatic expansion.

EXAMPLES

The invention will be further clarified by the following examples, which are intended to be purely exemplary of the invention and are in no way limiting.

FAN1-MLH1 binding demonstrated in vitro in multiple human cell models The inventors hypothesised that FAN1 could directly interact with MMR factors at CAG repeats to modulate expansion. To test the functional significance of the FAN1-MLH1 interaction in an HD context the inventors first tested this hypothesis in induced pluripotent stem cells (iPSCs) derived from a juvenile HD patient originally carrying 125 CAG repeats. Immunoprecipitation (IP) using FAN1 antibodies showed MLH1 and PMS2 were present in FAN1 pull-down fractions, whereas MSH3 was absent, and conversely, FAN1 was present in MLH1 pull-down fractions alongside PMS2 and MSH3 (Figure 1A). To further confirm this interaction in independent cell lines, the inventors used HD lymphoblastoid (LB) cells carrying more typical, shorter, disease-associated CAG lengths (Figure 1B). To exclude antibody-specific artefacts, the inventors also validated this interaction in U205 cells expressing GFP-FAN1 and confirmed that MLH1, PMS2 and MLH3 were readily detected in GFP-trap pull-down fractions, whereas MSH2, MSH3 and MSH6 were absent (Figure 1C).

To further dissect the interaction between FAN1 and the MLH1-PMS2 heterodimer, the inventors performed crosslinking immunoprecipitation mass-spectrometry (xl P-MS) experiments using HEK293T cells, expressing Myc-tagged FAN1, and lymphoblastoid cells, expressing endogenous FAN1. The inventors observed interactions between FAN1 and its known FA-complex interactors, FANCD2 and FANCI (Figure SA, Table 1). Interestingly, analysis of the aggregated crosslinking data from both experiments showed multiple proximity areas between FAN1, MLH1 and PMS2 (Figure 1D, Table 1). Three crosslinks were observed between FAN1 and MLH1, two in the N-terminal part of FAN1 and one in the TPR (tetratricopeptide repeat) domain. The inventors identified four crosslinks between FAN1 and PMS2, three of which were in the N-terminal region of FAN1 and one in the TPR domain (Figure 1D). Intriguingly, a cluster of four crosslinks between N-terminal FAN1 (p.120-168) and both PMS2 and MLH1 was observed. One of the FAN1 intra-protein crosslinks (K539-S646) was in the structured region of the protein (4RID) at a distance of 27A, which is consistent with the maximal distance for the crosslinker used, while all other crosslinks involve unstructured regions with no atomic coordinates present in the Protein Data Bank (PDB). Together, these data show for the first time that MutLa, but not MSH3, directly interacts with FAN1, and point to specific contact areas that are believed to be critical for this interaction.

The FAN1 N-terminal region (p.73-349) mediates its interaction with MLH1 and its effect on CAG stabilisation activity To pinpoint the MLH1-binding region(s) of FAN1 the inventors expressed a series of GFP-tagged FAN1 deletion constructs in a well-characterised U2OS HD cell model (GooId et ol. Hum. Mol. Genet 2019;28:650-661). These constructs were designed to retain the major functional domains of FAN1 (Figure 1E). GFP pull-down fractions from cell extracts expressing a FAN1 construct comprising the first 349 residues (FAN1'-349) contained levels of MLH1 similar to those produced using full-length FAN1 (FAN1). In contrast, FAN1°73-349, a deletion construct missing most of this N-terminal region but retaining the nuclear localisation signal (NLS; p.11-25), the ubiquitin-binding zone (UBZ), SAP, TPR and nuclease domains, did not form a complex with MLH1 (Figure 1F). The interaction of the N-terminus of FAN1 with MLH1 was confirmed by reverse IP using MLH1 antibodies. This showed FAN1wT and FAN11-349 bind MLH1 with approximately equal efficiency (Figure 1G). It is also worth noting that PMS2 partitions with MLH1 in IP fractions derived from FAN1 knockout (FAN1'), FAN1wT and FAN11-349 cells, indicating FAN1 does not influence the MutLa complex interaction.

To exclude the possibility that deleting a large section of the FAN1 sequence creates an inactive form of the protein that is unable to bind MLH1 because it is misfolded or mis-localised, the inventors performed functional analyses. Live cell imaging using the FANI"3-349 GFP tag showed exclusively nuclear localisation (Figure 55,C). Mitomycin C (MMC) stimulates the formation of nuclear FAN1 repair foci in a manner mediated by the UBZ domain and requires FAN1 nuclease activity for ICL repair and survival. In M MC cell viability assays, as expected, FAN11-349 was present exclusively in the nucleus and formed DNA repair foci, though not as efficiently as the full-length protein as it lacks the DNA-binding SAP domain, and it provided no protection against M MC toxicity (Figure 1H, 5B,C,E). In turn, FAN1°73-349 formed repair foci and protected against M MC genotoxicity, indicating that this protein was functional in ICL repair and is therefore unlikely to be misfolded (Figure 1H, 5B,C,E). Thus, the MLH1-binding capacity of these constructs likely reflects the protein's biological activity rather than mis-localisation or misfolding. These data also suggest that the UBZ domain and nuclease activity are not required for the FAN1-MLH1 interaction. To confirm this independently the inventors expressed the p.C44A/C47A and p.D960A FAN1 mutants, deficient in ubiquitin-binding and nuclease activity respectively, in U205 cells and assessed their MLH1-binding capacity using GFP-trap pull-down assays. Both constructs bound to MLH1 as effectively as the full-length protein (Figure 1F). Cells expressing these constructs also displayed the expected response to M MC treatment with the p.D960A but not the p.C44A/C47A variant showing reduced viability (Figure 1H, 55,C,E).

To assess the effect of the FAN1-MLH1 interaction on CAG repeat instability, the inventors measured CAG repeat expansion over 40 days in isogenic U2OS cells expressing each construct. The nuclease-deficient p.D960A and p.C44A/C47A UBZ mutations did not protect against repeat expansion, relative to the full-length protein (Figure 11). FAN11-349 was also able to stabilise the CAG repeat, with the same expansion rate as FAN1wT (Figure 1I,J,K), but FAN 1A73-349, the inverse construct lacking most of the N-terminal region, did not slow CAG expansion as effectively (Figure 1I,J,L). Importantly, the expansion rate in FAN1°73-349 cells was not as fast as FAN1"-cells, suggesting that a FAN1 region outside of residues 73-349 also contributes to CAG repeat stabilisation activity.

Taken together these structure-function analyses show that the FAN173-34g N-terminal region is necessary and sufficient for interaction with MLH1 and protection against CAG expansion, independent of UBZ and nuclease activity.

The FANI 126SPYF129 domain mediates MLH1 interaction and confers CAG repeat stabilisation in conjunction with FANI nuclease activity The inventors observed that FAN11-155 and FAN11-19° constructs both bind MLH1 robustly, but FAN11-140 showed a reduced interaction (Figure 2A). Quantification of GFP-trap pull-down fractions suggested MLH1-binding increased as the FAN1 N-terminal constructs lengthen, whereas FAN11-12° and the deletion construct FAN1A120-140 showed little or no MLH1-binding (Figure 2A, B, C). Therefore, MLH1-binding absolutely requires FAN1 residues 120-140, but downstream sequences could contribute to complex stability. This data is consistent with on-bead crosslinking experiments that showed close associations between MLH1-PMS2 and the N-terminal region of FAN1 (Figure 1D).

The N-terminal region of FAN1 is largely unstructured and relatively non-conserved. It does, however, contain three highly conserved regions, the first of which consists of a SPYF motif (p.126129) similar to the MLH1-interacting peptide box (MIP-box) found in many of MLH1's interaction partners (Figure 2D). Considering the similarity to a known MLH1-binding sequence and the data from the inventors' structure-function analysis, the inventors explored the role of 125SPYr29 in the FAN1-MLH1 interaction. The inventors introduced a series of alanine substitutions into the SPYF motif using site-directed mutagenesis and expressed these mutants as GFP fusion proteins in U2OS cells (Figure 2E). Importantly, cells expressing these constructs were protected against MMC toxicity and formed nuclear repair foci normally, suggesting the SPYF mutations did not affect ICL repair activity (Figure 2F, Figure 5B,D,F). Instead, GFP-trap pull-down fractions showed residues within the SPYF motif, in particular the aromatic residues Y128 and F129, as critical for MLH1-binding whereas mutation of a well-conserved residue outside this sequence (Q123) did not affect binding (Figure 2G,H). These data agree closely with the inventors' structure-function analysis and demonstrate that FAN1 interacts with MLH1 through its conserved N-terminal SPYF motif.

The S126A, Y128A and F129A mutations accelerated repeat expansion, whilst Q123A had no effect (Figure 3A). FAN11-12° did not stabilise the CAG, whilst longer SPYF-containing constructs, including FAN11-155, significantly restrained CAG expansion (Figure 3B). Consistent with this, deleting residues 120-140 (FAN1n120-140) from the FAN11-349 construct accelerated repeat expansion (Figure 35). As for the SPYF mutants, CAG repeat stabilisation activity and MLH1-binding correlate closely, indicating they are mechanistically linked.

Mutations within the SPYF motif showed increased expansion rates relative to WT, significant because it shows nuclease function alone does not fully stabilise the CAG repeat, even with equivalent expression levels between FAN1 WT and SPYF mutants. Despite this, the inventors observed that the expansion rate in SPYF-deficient constructs was not as fast as in FAN11-cells (Figure 3A), suggesting there is residual stabilisation activity downstream of p.349, with the most likely candidate being the nuclease domain. To assess this, the inventors introduced the nuclease-deficient p.D960A mutation into a SPYF-deficient construct (FAN1F129A). Immunoblots demonstrated that FAN1F129A/D960A was expressed at equivalent levels to FAN1wT (Figure 3D) and it was able to form DNA repair foci in response to MMC, a response requiring a functional UBZ domain (Figure 5B,D). However, GFP-trap pull-down experiments demonstrated reduced MLH1 binding while decreased MMC viability showed deficient ICL repair (Figure 3C,D, 5G). Importantly, repeat expansion in FAN1F129A/1)960A cells was faster than the F129A single mutant and equivalent to FAN11 cells (Figure 3E,F,G).

Taken together these data show that the FAN1 SPYF motif mediates its MLH1 interaction, and that FAN l's protective stabilisation of the CAG repeat involves MLH1 binding and the nuclease domain.

FAN1 regulates MMR activity by competing with MSH3 for MLH1 binding Consistent with reduced MMR activity, MLH1 and MSH3 knockout abolishes repeat expansion (Figure 4A,B,C) and in the case of MLH1, increases resistance to 6-thioguanine (6TG) (Figure 4D, 5H). Surprisingly, the inventors observed that expression of FAN1' or FAN11-349 in a MLH1au background also increases 6TG resistance relative to [AM."-cells, whereas expression of FAN1°73-349 or SPYF mutants had no effect (Figure 4D and 51-1,I)J). This suggests the SPYF motif sequesters MLH1 away from its other binding partners, reducing MMR activity and ultimately preventing repeat expansion. To explore this possibility, the inventors tested the ability of MLH1 to associate with MSH3 in the presence or absence of FAN1. This is of particular significance, given the key role of MSH3 in somatic expansion (Figure 4B,C) and the similarity of the FAN1 SPYF motif to the MIP-box in MSH3, which mediates binding to MLH1. Consistent with this, MSH3 pulldowns showed that MLH1 levels were reduced in FAN1wI samples relative to FAN11-(Figure 5K). FAN1 was not observed in these II3s, confirming it does not interact directly with MSH3. This suggests that FAN1 controls MMR complex assembly by sequestering MLH1. In MLH1 pulldowns, FAN1, MSH3 and PMS2 were recovered. The presence of FAN1 did not affect PMS2 levels, suggesting it does not interfere with MutLa complexing, but MSH3 levels were reduced in FAN1wI relative to FAN1-/-samples (Figure 4E)F). This indicates that FAN1 competes with MSH3 for binding to MLH1.

One consequence of this may be a reduction of MSH3-dependent MLH1 recruitment to the CAG repeat. To assess this, the inventors performed a chromatin immunoprecipitation (ChIP) assay, involving anti-MLH1 IP from FAN]." and FAN1wT U205 cells. PCR across the HTT CAG repeat identified the endogenous 20 CAG and exogenous 118 CAG repeat in both samples (Figure 4G). The presence of long and short repeat sequences was confirmed by fragment analysis of the ChIP samples (Figure 4H). qRT-PCR analysis showed there was less HTT CAG DNA in anti-MLH1 ChIP fractions from FAN1wT cells, relative to RANI." (Figure 41). This is consistent with FAN1 reducing MLHVs interaction with the CAG repeat.

To further explore the role of DNA repair genes implicated in somatic instability the inventors analysed FAN11, FAN1wT, MLH11 and MSH31 U2OS cell lines for evidence of microsatellite instability (MSI) over the course of the inventors' CAG repeat expansion assays.

MutSD deficiency results in MSI at tetra-and dinucleotide repeats, whereas MutSa deficiency causes MSI at mono-and dinucleotide repeats. Although MLH11 cells did not demonstrate CAG repeat expansion (Figure 4C), there was instability at tetranucleotide marker D20S85 (otherwise known as EMAST, or elevated microsatellite alterations at selected tetranucleotide repeats), indicating MMR deficiency (Figure 6). Similarly, MSH3 i cells showed MSI at several tetranucleotide (MYCL1, D9S242, D20582, D20585) and dinucleotide loci (D85321), indicating MutSP deficiency, but CAG repeat remained stable (Figure 4B,C, 6). Manipulation of FAN1 did not affect MSI in the time course of the assay (Figure 6). Collectively, these data suggest that FAN1 suppresses MMR activity by sequestering MLH1 away from MSH3, thus preventing error-prone repair and CAG repeat expansion.

Summary

The inventors demonstrated that FAN1 directly interacts with MLH1, but not with MSH3 and discovered for the first time that the N-terminal region of FAN1, in particular the 126SPYF'29 motif, mediates this interaction. The inventors also demonstrated that the N-terminal region of FAN1 protects against repeat expansion by sequestering MLH1 and preventing formation of the MutSI3 complex. Thus, the inventors have identified that sequestration of MLH1 inhibits repeat expansion, and that this advantageous effect can be achieved by a peptide comprising the SPFY motif present in the N-terminal region of FAN1. Unexpectedly, the inventors also discovered that FAN1 nuclease activity contributes to somatic stabilisation.

These data demonstrate that FAN1 inhibits somatic expansion by: (1) sequestering MLH1; and (2) promoting accurate repair via its nuclease activity. Therapeutically increasing or replicating these FAN1 functions significantly inhibits somatic expansion thereby providing a new and unexpected therapeutic strategy for treating, preventing, or delaying the onset of repeat expansion diseases.

Peptide sequence 1 Peptide sequence 2 Protein 1 Uniprot entry name Protein 2 Uniprot entry name xQuest score Crosslink position in Crosslink Crosslink position in protein 1 Crosslink position in protein 2 peptide 1 position in peptide 2 NIGFSHLQQR TRKQPLSK FACD2_HUMAN FACD2_HUMAN 34.37 5 3 1002 25 KTLELLVCR TPYPRPR FACD2_HUMAN PMS2 HUMAN 31.56 2 3 1363 149 LQEFLQTLR MVSKRR FANCI HUMAN FACD2_HUMAN 30.99 7 3 23 3 TNLTPGQSD S AKR KSPPPK FAN1 HUMAN - FAN1 HUMAN - 30.8 10 2 116 102 RLDETVVNR RSPLGQK MLH1_HUMAN PMS2_HUMAN 30.18 5 2 14 445 NPSEEYGK LDYELGR MLH1_HUMAN FANCI HUMAN 29.55 3 3 170 283 GGLPDLVVVV N SQSR KSPPPK FAN1 HUMAN - FAN1 HUMAN - 29.51 13 2 968 102 HVTITGRLCP OR KDVSISR FAN1 HUMAN - PMS2 HUMAN - 29.09 3 4 781 416 RHPSYPK TGEEKK FACD2_HUMAN PMS2_HUMAN 29.00 4 5 88 412 YSVHNAGISF S LESMSAK MLH1_HUMAN FACD2_HUMAN 28.07 14 3 196 773

VKKQGETVA

NSECDPTPS HR QPLSKK FACD2 HUMAN FACD2 HUMAN 27.77 7 5 896 30

- -

QGETVADVR IASLAR MLH1 _ HUMAN FACD2_HUMAN 27.74 4 3 200 1157 LSLYQRAVR KQKTDGSK FAN1HUMAN FACD2_HUMAN 27.24 4 7 747 877 RAKALAGQS V RICK VSMSMR FAN1 HUMAN - FANCI HUMAN - 26.88 3 4 528 512 YEKTISEAVVIK VPFLKNK FACD2_HUMAN FACD2_HUMAN 26.86 3 5 358 992 CLSQQADVR ASYSDGK FANCI HUMAN MLH1 HUMAN 26.51 3 4 596 131 TNLTPGQSD S AKR TSVVKSFK FAN1 HUMAN - FANCI HUMAN - 26.02 4 2 110 333 LLYVVNMAVR LREAFSLR FACD2_HUMAN PMS2_HUMAN 25.04 3 6 1267 425 VGTRLMFDH N GKIIQK QLINTLCSGR PMS2 HUMAN - FANCI HUMAN - 24.13 3 5 133 163 ALAGQSVRI C KGPR NHPSLR FAN1 HUMAN - FAN1 HUMAN - 23.68 11 4 539 646 LQEFLQTLR MVSKRR FANCI HUMAN FACD2_HUMAN 23.05 7 4 23 4 TGEEKKDVSI SR KYVKAKK PMS2 HUMAN - FAN1 HUMAN - 22.95 9 4 416 165 SFKDLQLLQ G SK LLGSNSSR FANG! HUMAN MLH1 HUMAN - 21.6 11 7 347 340 ALKNPSEEY OK GTSEMSEKR MLH1 HUMAN - MLH1 HUMAN - 21.33 3 8 167 461 LQASQVKLK S KGR KENLAYGK FACD2 HUMAN FANCI HUMAN - 20.98 7 6 322 146

-

KQGETVADVR LRESPSCKK MLH1_HUMAN FAN1 HUMAN 20.96 5 8 200 760 ALKNPSEEY OK GTSEMSEKR MLH1 HUMAN - MLH1 HUMAN - 20.61 3 3 167 456 ALAGQSVRI C KGPR LFQTLRR FAN1 HUMAN - FACD2_HUMAN 31.67 11 4 539 82

-

LLMVILEKST A SAQNK LEPTIK FACD2 HUMAN FAN1 HUMAN - 29.87 12 4 1148 725

-

SNDVVCK TDISSGR FAM_HUMAN MLH1_HUMAN 29.12 1 4 131 420 TLPNASTVD NI R VSMSMR MLH1 HU MAN _ FANCI 28.58 1 4 206 512

_

EVKQKISPYFK SPLGQK FAN1 HUMAN PMS2_HUMAN 28.53 9 1 128 445 KDVSISRLR NPSEEYGK PMS2THUMAN MLH1_HUMAN 27.93 6 6 418 173 SLMNLLFSLH V SYK KTHIFQDR FANCI HUMAN - FAN1 HUMAN - 27.8 1 2 1027 593 SVMIGTALNT S EMK VICLGSLAS KL SR PMS2 HUMAN - FAN1 HUMAN - 27.55 10 10 824 158 NIKKEYAK RSLSISK PMS2_HUMAN FAN1 HUMAN 27.29 6 6 181 20 KSKVNLMQH LYVRLFQR FANCI HUMAN - FAN1 HUMAN - 26.91 11 2 1280 418

M KLSTSR

VVTVEHIVYK AL R LSDILNEK MLH1 HUMAN - FANCI HUMAN - 26.42 2 2 715 782 LFQTLRRHP S YPK EGSLVNGK FACD2 HUMAN FANCI HUMAN - 25.88 10 3 88 120

-

KASNSIISCF GTSEMSEK FAN1 HUMAN - MLH1 HUMAN - 25.12 8 2 32 455

N NAPPAK

LLYVVNMAVR TDISSGR FACD2_HUMAN MLH1_HUMAN 25.09 3 5 1267 421 TSVVKSFK KLFQTLR FANCI_HUMAN FACD2_HUMAN 24.38 5 5 336 82 KTHIFQDR MVSKRR FAN1 _ HUMAN FACD2_HUMAN 23.79 2 3 593 3 KSAVAGFLL LL K LPEYFFENK FANCI HUMAN - FACD2 HUMAN 23.65 2 4 535 160

-

KTHIFQDR MVSKRR FANLHUMAN FACD2_HUMAN 22.83 2 4 593 4 SESPSLTQER KTHIFQDR FACD2_HUMAN FAN1 HUMAN 22.57 3 2 592 593 FVEILQRLHM Y EEAVR NIKKEYAK FAN1 HUMAN - PMS2 HUMAN - 22.51 11 4 683 179 QMFASRACR EPAKKK PMS2_HUMAN FANCI HUMAN _ 22.31 5 5 809 1324 LSWIKMTK ATRECEK FAN1 HUMAN FAN1 HUMAN 21.11 7 2 432 245 LESMSAK NMEKLVK FACD2_HUMAN FANC IHUMAN 20.69 5 4 775 1196 QM FASRAC R K SLNYTGEK PMS2_HUMAN FANC IHUMAN 20.05 5 1 809 1222 Table 1: List of the crosslinks identified between FAN1, MLH1, PMS2, FANCD2 and FANCI.

(Related to Figure 1D and Figure SA). Peptide sequences pairs involved in crosslinks are listed with the xQuest score and the position of the crosslink in the peptide and the protein.

METHODS

Cell culture and manipulation U2OS FAN1-1-cells were generated as previously described, featuring FRT sites introduced into the genome, enabling complementation with tetracycline-inducible FAN1 variants when co-transfected with Flp recombinase. This line was kindly gifted by Prof. John Rouse (University of Dundee, Scotland). Introducing a lentiviral HTT exon 1 construct harbouring 118 CAG repeats allows examination of the effects of different FAN1 activities/regions on repeat stability (Goold etal. Hum.

Mol. Genet 2019;28:650-661). U20S cells were maintained in DM EM with Gluta MAX, supplemented with 10% FBS and pen-strep. ICL repair assays were performed as described previously (Goold etal. Hum. Mol. Genet 2019;28:650-661). For quantifying GFP-FAN1 foci, cells were imaged using a fluorescent microscope and were considered positive with foci per nucleus.

Induced pluripotent stem cells (iPSC) were cultured in Essential 8 medium and grown on GeltrexTM basement membrane matrix. Lymphoblastoid cells derived from the TRACK-HD cohort were cultured in RPM! medium supplemented with 15% fetal bovine serum (FBS), 100 U/ml penicillin and 100 pg/ml streptomycin.

Immunoprecipitation, ChIP, cloning, SEIM and CRISPR ChIP analysis was performed with the EZ-Magna ChIP'" A Chromatin Immunoprecipitation Kit according to the manufacturer's instructions. Chromatin was fragmented by 15 cycles of 30 s sonication in a Bioruptor apparatus at 4°C. Immunoprecipitation was done overnight at 4°C using anti-MLH1 antibodies (BD Biosciences). DNA from ChIP and input fractions was quantified by SYBR (Thermo, #A25741) qRT-PCR using primers targeting two regions proximal to the CAG repeat (pair 1 forward CCGCTCAGGTTCTGCTTTTA, reverse GCCTTCATCAGCTTTTCCAG; pair 2 forward CCAGAGCCCCATTCATTG, reverse GCCTTCATCAGCTTTTCCAG), and one distal, at the 3' end of HTT (forward TGCCTTTCGAAGTTGATGCA, reverse TGCCACCACGAATTTCACAA). DNA levels were quantified relative to a genomic DNA standard. Results were expressed as percentage of the DNA levels in U205 FAN1 F ChIP fractions.

Cell extracts were prepared for SDS-polyacrylamide gel electrophoresis (PAGE) as described previously (Goold et al. Nat. Commun 2011;2:281). The antibodies used were a FAN1 sheep polyclonal antibody (Goold et al. Hum. Mol. Genet 2019;28:650-661); MSH3 or MLH1 monoclonal antibodies (BD Biosciences, UK); PCNA and MSH2 (Cell Signaling Technology, Danvers, MA, USA); and PMS2, GAPDH and GFP rabbit polyclonal antibodies (Santa Cruz Biotechnology, Dallas, TX, USA). Immunoblots were quantified with the Odyssey CLx Imaging System, (Lincoln, NE, USA) using Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and 13-actin as loading controls. For Immunoprecipitation (IP) analysis, washed cells were resuspended in IP buffer (20 mm Tris, pH 7.4, 150 mm NaCI, 1 mm EDTA, 1% Triton X-100 supplemented with Benzonase 2 U/ml and protease inhibitors) and incubated on ice for 20 min. The cell extracts were centrifuged at 10,000 g for 2 min and the supernatant fraction was used as input. GFP-Trap beads or the FAN1 sheep polyclonal and MSH3 or MLH1 monoclonal antibodies and protein G magnetic beads were used to capture protein complexes. Beads were washed 3 times in IP buffer and eluted by heating in SDS sample buffer.

FAN1 point mutations were generated by site-directed mutagenesis using the QuickChange XL kit according to the manufacturer's instructions (Agilent, CA, USA). The presence of the DNA base changes was confirmed by sequencing of the genomic DNA isolated from reconstituted cells. Deletion constructs were synthesised by GeneArt (Thermo Fisher) and subcloned into pcDNA5.1 FRT/TO GFP FAN1 using Bam H1, EcoRV and Not1 restriction sites. CRISPR guide sequences encoded in pX458 vector were used to inactivate the MSH3 and MLH1 genes in U2OS cells. Knockout was confirmed by Western blot, sequencing and functional assays.

Somatic instability assay DNA was extracted from samples by the QIAamp DNA Mini kit (Qiagen, #51306) and the HTT locus amplified by PCR (6-FAM-labelled F. primer: AAGGCCTTCGAGTCCCTCAAGTCCTT; R. primer: CGGCTGAGGCAGCAGCGGCTGT). The PCR product was denatured and analysed by capillary electrophoresis, on an Applied Bioscience" 3730XL DNA Analyzer (Thermo). Chromatographs were aligned in GeneMapper" v6. software (Thermo). To calculate modal CAG repeat length and instability index, GeneMapper data was exported and analysed with a custom R script, available at https://caginstability.m1 with an inclusion threshold of 20% of modal peak height and manually confirmed.

Mkrosatellite instability (MSI) analysis DNA from ChIP samples was amplified in parallel by fluorescently labelled PCR at unstable tetranucleotide (085321, D20582, 095242, MYCL1, D20S85), dinucleotide (D2S123, D5S346, D17S250, D18564, D18S69), mononucleotide (NR-21, NR-24, BAT-25, BAT-26,MON0-27, NR-27) and stable control pentanucleotide (Penta C and Penta D) loci. Fluorescently labelled fragments were separated by capillary electrophoresis and the repeat length of each allele determined with a custom R script, as above.

Mass-spectrometry Lymphoblastoid cells, expressing endogenous levels of FAN1, and HEK293T cells transiently overexpressing Myc-FAN1, were lysed 10 min on ice using PBS, 1% NP-40, Benzonase and protease inhibitors and centrifuged 5 min at 20,000 g to remove cell debris. Anti-c-Myc magnetic beads were incubated 2 h with HEK cell lysates. A sheep FAN1 antibody (GooId et Hum. Mol. Genet 2019;28:650-661) was incubated for 1 h with LB cell lysate and protein G magnetic beads were then added to the mix and incubated for an additional 1 h. Four washing steps were performed using lysis buffer. Crosslinking was done using 1 mM 553 dO/d12 for 30 min at 37°C. The reaction was quenched for 20 min at 37°C using ammonium bicarbonate at a final concentration of 100 mM. Prior to digestion, beads were resuspended in a buffer containing 2 M Urea, 100 mM ammonium bicarbonate, 10 mM DTT and denatured for 20 min at room temperature under agitation (1000 rpms) (Makowski et al. Mol. Cell Proteomics 2016;15:854-65). Samples were then alkylated, at room temperature and in the dark, using a final concentration of 50 mM iodoacetamide for 20 min, and diluted with 50 mM ammonium bicarbonate solution to obtain a final concentration of urea below 1 M. Digestion was performed using sequencing grade trypsin overnight at 37°C. Samples were fractionated in 3 fractions using C18-SCX StageTips prepared in-house as previously described (Rappsilber et al. Nat. Protoc 2007:2;1896-1906) with the following concentrations of ammonium acetate: 200 mM, 1 M and 1.5 M. Prior to mass spectrometry analysis, samples were further processed using C18 StageTips.

Crosslinked peptide mixtures were resuspended in 3% acetonitrile, 0.1% formic acid and were analysed by nano-LC-MS/MS using an Acquity M-Class system coupled to a Synapt G2Si mass spectrometer (Waters Corporation). Samples were loaded on the system and desalted by a reversed-phase Symmetry C18 trap column (180 p.m internal diameter) 20 mm length, 5 pm particle size, Waters Corporation) at a flow rate of 8 ilL/min for 3 min in 99% solvent A (Solvent A: MS-grade water, 0.1% formic acid -solvent B: Acetonitrile, 0.1% formic acid). Peptides were then separated using a linear gradient (0.3 pL/min, 35°C; 3-60% solvent B over 90 min) using a BEH130 C18 nanocolumn (75 1.1rn internal diameter, 400 mm length, 1.7 p.m particle size, Waters Corporation). The mass spectrometer was operated in data-dependent acquisition mode using a mass range of 502000 Th for both MS and MS/MS scans and scan times of 0.2 s and 0.3s respectively. The ten most intense precursor ions with a charge state between 3+ and 6+ were selected for fragmentation using the 'mid' collision energy ramp as described in James et al. Anal Chem 2019;91:1808-1814. Dynamic exclusion was used with a 30 second window to prevent repeated selection of peptides.

Raw mass spectrometry files were converted to MGF (Mascot Generic Format) using PLGS (v3.0.2) using slow deisotoping algorithm and automatic denoising for both MS and MS/MS data. MGF files were further converted to mzXML with MSConvert (Chambers et at Nat. Biotechnol 2012;30:918-20) using 32-bit binary encryption.

Crosslinking identification was performed using xQuest/xProphet (Leitner et al. Nat. Protoc 2014;9:120-137). Searches were performed using a database containing the sequences of FAN1, M LH1, PM52, FANCD2 and FANC1 using a search tolerance of 20 ppm. The amino acids involved in crosslinking reactions parameter was set to K, 5, T, Y and N-terminal amino acid. Up to three missed cleavages were allowed, carbamidomethylation of cysteine was set as a fixed modification and oxidation of methionine was set as a variable modification. Results were validated using xProphet with a 5% FOR.

Further validation of the crosslinks was performed by extracting the highest-ranking identification from the xProphet xml output, using a modified version of Validate XL (James et al. Anal Chem 2019;91:1808-1814), and only considering crosslinks scoring higher than 20. For these crosslinks, the presence of light and heavy crosslinked doublets in the RAW MS files was confirmed. Automated generation of tables and MGF files was done using an in-house Python script to allow crosslinking map representation using xiVIEW (Mendes et al. Mol. Syst. Biol 2019;15:e8994).

Statistical analysis CAG expansion time courses were analysed by linear regression in GraphPad Prism and slopes statistically compared by one-way ANOVA. Multiple comparisons were corrected for with a False Discovery Rate (FOR) of 5%. Area under curve (AUC) data were compared by a one-way ANOVA with an FDR correction of 5%. Data between two groups were analysed by one-tailed independent-samples t-tests. *p<0.05, ** p<0.01, ***p<0.001 ns = non-significant. For conservation analysis, the human FAN1 sequence was aligned in HomoloGene (NCB° with common model species and visualised with SnapGene software.

Claims

CLAIMS1. A composition for use in treating, preventing or delaying the onset of a repeat expansion disease in a subject, wherein the composition comprises an MLH1 inhibitor and/or a FAN1 derived nuclease.
2. The composition for use according to claim 1, wherein the repeat expansion disease is selected from Huntington's disease, Fragile X Syndrome, myotonic dystrophy (DM 1 and DM2), amyotrophic lateral sclerosis and/or frontotemporal dementia caused by somatic expansion in the C90RF72 gene, spinocerebellar ataxias (SCAs 1, 2, 3, 6, 7 and 17), Friedreich's ataxia (FRDA), Fragile X Tremor Ataxia Syndrome (FXS/FXTAS), dentatorubral-pallidoluysian atrophy (DRPLA), spinal and bulbar muscular atrophy (SBMA), and Unverricht-Lundborg myoclonic epilepsy (EPM1).
3. The composition for use according to claim 2, wherein the repeat expansion disease is Huntington's Disease.
4. The composition for use according to claim 2, wherein the repeat expansion disease is Fragile X Syndrome.
S. The composition for use according to any preceding claim, wherein the composition comprises an MLH1 inhibitor.
6. The composition for use according to claim 5, wherein the MLH1 inhibitor is selected from a small molecule, a peptide, a cyclic peptide, an aptamer, or a peptidomimetic.
7. The composition for use according to claim 6, wherein the MLH1 inhibitor is a cyclic peptide.
8. The composition for use according to any one of claims 5-7, wherein the MLH1 inhibitor comprises an MLH1-binding fragment of FAN1.
9. The composition for use according to any one of claims 5-8, wherein the MLH1 inhibitor is a peptide comprising the amino acid sequence SPYF.
10. The composition for use according to claim 9, wherein the MLH1 inhibitor comprises a peptide having at least 70% sequence identity to residues 120-140 of SEQ ID NO: 2.
11. The composition for use according to claim 9 or claim 10, wherein the MLH1 inhibitor comprises a peptide having at least 70% sequence identity to residues 73-165, residues 73190, residues 73-349, residues 1-165, residues 1-190, and/or residues 1-349 of SEQ ID NO: 2.
12. The composition for use according to any one of claims 5-11, wherein the MLH1 inhibitor binds directly to MLH1, optionally wherein the MLH1 inhibitor binds directly to the S2 site of MLH1.
13. The composition for use according to any one of claims 5-7, wherein the MLH1 inhibitor promotes MLH1 binding to FAN1.
14. The composition for use according to claim 13, wherein the MLH1 inhibitor is a FAN1 activator, optionally wherein the FAN1 activator increases the expression of FAN1.
15. The composition for use according to claim 13, wherein the MLH1 inhibitor comprises a kinase, and wherein the kinase promotes MLH1 binding to FAN1 by phosphorylating the FAN1 SPYF motif.
16. The composition for use according to claim 13, wherein the MLH1 inhibitor comprises a phosphatase, and wherein the phosphatase promotes MLH1 binding to FAN1 by dephosphorylating the FAN1 SPYF motif.
17. The composition for use according to any one of claims 5-7, wherein the MLH1 inhibitor stabilises MLH1 binding to FAN1, or a fragment thereof.
18. The composition for use according to claim 17, wherein the MLH1 inhibitor is an allosteric stabiliser of the FAN1-MLH1 interaction.
19. The composition for use according to claim 17, wherein the MLH1 inhibitor is a direct stabiliser of the FAN1-MLH1 interaction.
20. The composition for use according to any preceding claim, wherein the composition comprises a FAN1 derived nuclease.
21. The composition for use according to claim 20, wherein the nuclease does not bind MLH1.
22. The composition for use according to claim 20 or claim 21, wherein the nuclease comprises at least 70% sequence identity to residues 893-1008 of SEQ ID NO: 2.
23. The composition for use according to any one of claims 20-22, wherein the nuclease comprises at least 70% sequence identity to SEQ ID NO: 2.
24. The composition for use according to claim 23, wherein the nuclease does not comprise an SPYF domain.
25. A vector for use in treating, preventing or delaying the onset of a repeat expansion disease in a subject, wherein the vector comprises a nucleic acid sequence encoding an MLH1 inhibitor and/or a nucleic acid sequence encoding a FAN1 derived nuclease.
26. The vector for use according to claim 25, wherein the repeat expansion disease is selected from Huntington's disease, Fragile X Syndrome, myotonic dystrophy (DM1 and DM2), amyotrophic lateral sclerosis and/or frontotemporal dementia caused by somatic expansion in the C90RF72 gene, spinocerebellar ataxias (SCAs 1, 2, 3, 6, 7 and 17), Friedreich's ataxia (FRDA), Fragile X Tremor Ataxia Syndrome (FXS/FXTAS), dentatorubral-pallidoluysian atrophy (DRPLA), spinal and bulbar muscular atrophy (SBMA), and Unverricht-Lundborg myoclonic epilepsy (EPM1).
27. The vector for use according to claim 26, wherein the repeat expansion disease is Huntington's Disease.
28. The vector for use according to claim 26, wherein the repeat expansion disease is Fragile X Syndrome.
29. The vector for use according to any one of claims 25-28, wherein the vector comprises a nucleic acid sequence encoding the MLH1 inhibitor according to any one of claims 5-19.
30. The vector for use according to any one of claims 25-29, wherein the vector comprises a nucleic acid sequence encoding the FAN1 derived nuclease according to any one of claims 20-24.
31. The vector for use according to any one of claims 25-30, wherein the vector comprises a nucleic acid sequence encoding the MLH1 inhibitor according to any one of claims 5-19 and a nucleic acid sequence encoding the FAN1 derived nuclease according to any one of claims 20-24.
32. The vector for use according to any one of claims 25-31, wherein the vector is selected from an adeno-associated virus (AAV) vector, a HIV-based lentivirus vector, equine immunodeficiency virus (Ely) vector, a feline immunodeficiency virus (Fly) vector, and a herpes simplex virus vector.
33. The vector for use according to claim 32, wherein the vector is an MV vector.
34. The composition for use according to any one of claims 1-24 or the vector for use according to any one of claims 25-33, wherein the composition or the vector is formulated for delivery to the striatum and/or the cortex of the subject.
35. The composition for use or the vector for use according to claim 34, wherein the composition or the vector comprises a targeting ligand.
36. The composition for use or the vector for use according to claim 35, wherein the targeting ligand facilitates uptake of the composition and/or the vector through the blood brain barrier.
37. The composition for use or the vector for use according to claim 35 or claim 36, wherein the targeting ligand comprises a compound that facilitates delivery to and/or uptake by neurons.
38. A method of treating or preventing a repeat expansion disease in a subject comprising administering to the subject the composition according to any one of claims 1-24 or 34-37 or the vector according to any one of claims 25-37.
39. A method of identifying MLH1 inhibitors comprising: (a) culturing cells expressing MLH1 and MSH3 in the presence of an agent; (b) purifying MLH1 and proteins bound thereto; (c) determining the level of MSH3 that is bound to MLH1; and (d) comparing the level of MLH1-bound MSH3 to a control level.
40. The method of claim 39, wherein the control level is the level of MLH1-bound MSH3 in the absence of the agent and a reduced level of MLH1-bound MSH3 in the presence of the agent indicates that the agent is an MLH1 inhibitor.
41. The method of claim 39, wherein the control level is the level of MLH1-bound MSH3 in the absence of the agent and the same or higher level of MLH1-bound MSH3 in the presence of the agent indicates that the agent is not an MLH1 inhibitor.