2024 Presentations | CHDI Foundation

CHDI’s 19th Annual HD Therapeutics Conference took place February 26 – 29, 2024 in Palm Springs, California. This unique conference series focuses on drug discovery and development for Huntington’s disease and draws participants from the biotech and pharmaceutical sectors as well as academia. The conference is intended as a forum where all participants can share ideas, learn about new disciplines, network with colleagues and build new collaborative partnerships. We are indebted to all of the conference speakers, and especially grateful to those who are able to make their presentations available here for a wider audience.

- - Featured Speaker
  - Humanizing drug discovery David Altshuler, MD, PhD, Vertex Pharmaceuticals

Accelerated epigenetic aging involves polycomb repressive complexes in Huntington’s disease

Karine Merienne, PhD, University of Strasbourg

With time, differentiated cells undergo epigenetic aging characterized by progressive erosion of epigenetic landscape, leading to a loss of cellular identity. It is thought that DNA damage drives epigenetic aging. The mechanism involves changes in histone modifications, including active H3K27 acetylation (H3K27ac) and repressive H3K27 trimethylation (H3K27me3) marks. Specifically, epigenetic aging associates with depleted H3K27ac at cellular identity genes, leading to their repression, together with reactivation of developmental genes—normally silenced in differentiated cells—as a result of increased H3K27ac and reduced H3K27me3. Since Huntington’s disease (HD) is age-related, epigenetic aging could be accelerated in HD vulnerable cells. In support to this hypothesis, we previously showed loss of H3K27ac at striatal identity genes in HD. However, epigenetic derepression of developmental genes was unclear due to lack of sensitivity of epigenomic data, which were generated on bulk striatal tissue. Here we assessed the epigenomic landscape of neuronal and non-neuronal striatal cells in two reference HD mouse models, using cell-type-specific multi-omics, including temporal analysis at three pathological stages. We show acceleration of derepression of neurodevelopmental genes in HD striatal neurons, which is mediated by general histone re-acetylation, including increased H3K27ac, H3K9ac and H3K18ac, together with reduction of H3K27me3 and H2AK119 ubiquitylation, catalyzed by polycomb repressive complexes PRC2 and PRC1, respectively. Our data indicate that the mechanism implicates paralog switch of PRC1 CBX proteins. Thus, we provide evidence for acceleration of epigenetic aging of HD vulnerable neurons, involving PRC1, a repressive complex linked to DNA damage response.

Top

The roles of RNA adenine methylation in Huntington’s disease and other neurological disorders arising from CAG repeat expansions

Yinsheng Wang, PhD, University of California, Riverside

Microsatellite repeat expansions within genes contribute to a number of neurological diseases. Accumulation of toxic proteins and RNAs with repetitive sequences, and/or sequestration of RNA-binding proteins by expanded repeat-containing RNAs, are thought to be important contributors to disease etiology. In this presentation, I will discuss our discovery about the implications of RNA methylation in CAG repeat expansion diseases. In particular, we found that the adenosine in CAG repeat RNA can be methylated to N1-methyladenosine (m1A) by TRMT61A, and the ensuing m1A can be demethylated by ALKBH3. We also observed that m1A/rA ratio in CAG repeat RNA increases with repeat length in a mouse model of Huntington’s disease and a Drosophila model of SCA3, which is attributed to diminished expression of ALKBH3 elicited by the repeat RNA. Additionally, TDP-43 binds directly with m1A in RNA at high affinity, which stimulates the cytoplasmic mis-localization and formation of gel-like aggregates of TDP-43, resembling the observations made for the protein in neurological diseases. We also found that m1A in CAG repeat RNA contributes to CAG repeat expansion-induced neurodegeneration in Caenorhabditis elegans. Moreover, we observed that N6-methyladenosine (m6A), which also forms in CAG repeat RNA, contributes to repeat-associated non-AUG (RAN) translation. Together, our study offers a new paradigm about the mechanism through which nucleotide repeat expansion contributes to neurological diseases and reveals a novel pathological function of m1A and m6A in RNA. These findings may provide an important mechanistic basis for the therapeutic interventions of neurodegenerative diseases emanating from nucleotide repeat expansions.

Top

Lessons for HTT-lowering therapies in Huntington’s disease from HTT silencing in mice

Jeff Carroll, PhD, University of Washington

We are interested in the impact of huntingtin (HTT) lowering treatments on specific components of HD pathogenesis, particularly transcriptional dysregulation and somatic instability (SI) in Htt’s CAG tract. In the first stream of work in our lab, we are pursuing a clearer understanding of HTT’s biological roles in vivo and the risks posed by HTT lowering. To investigate these questions, we have conducted a number of studies on the impact of wildtype HTT lowering in mice. Our results confirm previously proposed consequences of HTT loss, but also revealed completely novel molecular and physiological alterations. We discovered that global Cre-mediated loss of HTT induced cell fate changes, transcriptional alterations, and thalamic neurodegeneration associated with progressive vascular calcification in the brain. We also observed that this neurodegenerative phenotype is preceded by rapid increases in plasma neurofilament light (NfL) levels, suggesting that total HTT loss is poorly tolerated by specific neuronal populations, whose identification we are currently pursuing. In the second stream of work in the lab, we are studying the relationship between HTT lowering, transcriptional alterations, aggregate formation, and SI in mouse models of HD. We find that some, but not all, HTT-lowering approaches both lower HTT levels and reduce SI in Htt’s CAG repeat. Similarly, we find that some, but not all, antisense oligonucleotides (ASOs) targeting huntingtin are capable of reducing levels of the huntingtin exon-1 aberrantly spliced fragment (Htt1a) generated from long expanded CAG repeats. I will discuss our findings in the context of recent data emerging from experiments with ASOs and zinc finger transcriptional repressors (ZFPs), and suggest that different HTT-lowering modalities could be complementary as HD therapeutics.

Top

Functional proteomics captures a link between Huntington’s disease pathobiology and protein interaction dynamics

Ileana M Cristea, PhD, Princeton University

The complex pathobiology of Huntington’s disease (HD) is driven by mutation of a single causative gene, huntingtin (HTT). Efforts to understand the underlying mechanisms linking HTT polyglutamine expansion to HD progression have revealed the enormous potential for HTT to form protein complexes. Yet the breadth of these interactions represents a barrier toward identifying those that are proximal to disease pathogenesis. How HTT interactions and derived functions impact cellular processes to drive pathologies, such as neuronal cell death in the striatum and metabolic dysfunction in the liver, remains unclear. We address these questions by employing state-of-the-art proteomics, multi-omic investigations via computational platform development, and genetic HD modifier assays. To characterize unique molecular signatures of HD progression, we constructed local and global HTT-dependent protein networks to define direct and consequential impacts of HTT mutation or lowering. We characterized the cell-wide impact of HTT loss or polyglutamine expansion in a liver model of HD pathobiology using thermal proximity coaggregation assay mass spectrometry, an approach that captures global protein interaction networks and dynamics. Our results point to divergent alterations in the assembly of critical protein complexes in KO versus mutant HTT, including chromatin remodeling and calcium signaling proteins. To place these findings in the broader context of HD pathobiology, we characterized determinants of striatal vulnerability by defining HTT protein interactions at different stages of disease progression. Using Flag-HTT mouse HD models and designing a method to profile the relative stability of protein interactions, we identified polyQ- dependent mHTT interactions that also modify HD progression in a fly model. Expanding to characterizing domain-localized interactions with endogenous HTT, we found a functional epitope-dependent dichotomy between nuclear-associated versus cell membrane interactions. We identified striatal enriched targets and pathways linking mHTT with disruption of transcriptional regulation by the Mediator complex and calcium ion homeostasis. To catalyze interrogation of HTT protein interactions and facilitate multi-omic analyses within the HD community, we developed a web-based HD protein network and multi-omics platform, HTT-OMNI. Collectively, functional proteomics offers a new perspective on HD, resolving pleiotropic roles for HTT mutation that link HD pathobiology to protein interaction dynamics in transcription and ion homeostasis.

Top

Antibody localization of mutant huntingtin in cerebral cortex and basal ganglia of human Huntington’s disease and Q175 mouse in relation to regional and cellular vulnerability

Anton Reiner, PhD, The University of Tennessee Health Science Center

We studied cerebral cortex and basal ganglia in heterozygous Q175 mice and in human HD, using antibodies putatively selective for different forms of mutant huntingtin. These antibodies and their targets are: 1) PHP1 – polyproline region of mutant exon1 huntingtin fibrils; 2) 2B4 – polyproline region of exon1 huntingtin; and 3) MW8 – C-terminus of exon1 huntingtin.
Diffuse nuclear PHP1 in striatal projection neurons (SPNs) was evident at 2 months in Q175 mice, and declined with age, as nuclear and neuropil aggregates came to predominate. PHP1 was sparse in cholinergic and NPY+ interneurons, consistent with their low vulnerability. PHP1 was, however, also sparse in parvalbuminergic striatal interneurons, inconsistent with their high vulnerability. In cortex, diffuse nuclear PHP1 was first evident in layer 4 neurons, and with age progressed to widespread neuropil and nuclear aggregates. Layer 5 neurons were PHP1-poor throughout, inconsistent with their higher HD vulnerability. In human, PHP1 aggregates were abundant in neuronal nuclei, neuronal cytoplasm, and neuropil of striatum and cerebral cortex, but sparser in globus pallidus, with abundance increasing until grade 3 and then declining. In mouse and human, PHP1 and 2B4 colocalized.
MW8 was first evident in Q175 cortex and striatum as diffuse nuclear immunolabeling at 6 months, with aggregate immunolabeling predominating later. MW8 aggregates were largely restricted to SPNs and sparse in parvalbuminergic and cholinergic interneurons, while layer 5 was the most lightly immunolabeled cortical layer. MW8 colocalized completely with PHP1 and 2B4 in NIIs and neuropil aggregates, but largely co-occurred with ubiquitin only in NIIs. MW8 was sparse in human cortex and basal ganglia. By contrast to mouse, MW8 in human detected an aggregated mutant huntingtin form that was largely the same as detected by anti-ubiquitin, but not as broad as detected by PHP1 and 2B4. In human, MW8 burden at grade 1 across cortical regions predicted grade 2 neuron loss, and MW8 burden at grade 1 across basal ganglia regions predicted grade 4 neuron loss. Our results indicate that the mutant huntingtin forms detected by PHP1, 2B4 and MW8 show some association with regional and cellular HD vulnerability, but do not entirely explain it.

Top

Unveiling new insights into the molecular mechanisms of DNA repair: MutSβ and FAN1

Maren Thomsen, PhD, Proteros Biostructures GmbH

Huntington’s disease (HD) is fatal neurodegenerative disorder caused by an expanded CAG repeat in the huntingtin (HTT) gene. In HD, somatic instability refers to the propensity of these expanded CAG repeats to undergo even further expansion or contraction during cell division in somatic tissues, particularly in the brain, contributing to the observed clinical variability. Somatic expansion of the CAG repeats correlates with an earlier onset and more severe manifestations of the disease. Both MutSβ and FAN1, DNA repair proteins that maintain genomic integrity, have been implicated by genetic studies as modifiers of age of onset of HD. Understanding the detailed molecular mechanisms of MutSβ and FAN1 in canonical DNA repair in comparison to how they process DNA containing CAG repeats might offer new therapeutic intervention points for HD.
MutSβ, a key member of the MMR pathway, undergoes numerous conformational changes from DNA mismatch recognition to its release. DNA-free MutSβ complex displays an open conformation compatible with DNA binding. In the presence of (CAG)2 loop DNA, we observe two distinct conformations of MutSβ: one weakly associated with both nucleotide binding sites occupied by ADP, and the canonical mismatch- bound conformation with only MSH2 occupied by ADP. Addition of ATP or a non-hydrolyzable analogue initiates the conformational change towards the sliding clamp with both nucleotide binding sites occupied by ATP.
Genome-wide association analyses of HD patients revealed a strong correlation of the FAN1 R507H mutation with early disease onset. By solving cryo-EM structures of the FAN1:PCNA:DNA complex, we identified a direct interaction between FAN1 R507 and PCNA D232 and confirmed that complex formation with PCNA is disrupted in FAN1 mutants carrying the disease-promoting R507H mutation. We propose that the limited capability of FAN1 R507H to form a complex with PCNA and DNA leads to less efficient FAN1-mediated repair of CAG loop-outs, explaining the HD-promoting role of this mutation.
These findings present novel insights into the molecular mechanisms of MutSβ and FAN1 and might provide a new foundation for structure-guided drug discovery, offering new strategies for treatment.

Top

A novel target in the MutSβ-DNA trinucleotide repeats association for treating expansion diseases

Wei Yang, PhD, National Institutes of Health

Expansion of trinucleotide repeats causes Huntington’s disease, Fragile X syndrome and over twenty other monogenic disorders. How mismatch repair protein MutSβ and large repeats of CNG (N=A, T, C or G) cooperate to drive the expansion is poorly understood. Contrary to expectations, we find that MutSβ prefers to bind the stem of an extruded (CNG) hairpin rather than the hairpin end or hairpin-duplex junction. Structural analyses reveal that in the presence of MutSβ, CNG repeats with N:N mismatches adopt a B form-like pseudo-duplex, with one or two CNG repeats slipped out forming uneven bubbles that partly mimic insertion-deletion loops of mismatched DNA. When the extruded hairpin exceeds 40-45 repeats, it can be bound by three or more MutSβ molecules, which are resistant to ATP-dependent dissociation. We envision that such MutSβ-CNG complexes recruit MutLγ endonuclease to nick DNA and initiate the repeat expansion process. To develop drugs against the expansion diseases, we have identified lead compounds that prevent MutSβ binding to CNG repeats but not to mismatched DNA.

Top

Targeting somatic CAG repeat expansion: From genetics to therapeutics

Sarah Tabrizi, MD, FRCP, PhD, FMedSci, University College London

The pathological huntingtin (HTT) trinucleotide repeat underlying Huntington’s disease (HD) continues to expand throughout life. Repeat length correlates both with earlier age at onset (AaO) and faster progression, making slowing its expansion an attractive therapeutic approach. Genome-wide association studies have identified candidate variants associated with altered AaO and progression, with many found in DNA mismatch repair (MMR) associated genes. We have generated a stable CRISPR interference HD iPSC line in which we can specifically and efficiently lower gene expression from a donor carrying over 125 CAG repeats. Lowering expression of each member of the MMR complexes MutS (MSH2, MSH3 & MSH6), MutL (MLH1, PMS1, PMS2 & MLH3) and LIG1 resulted in characteristic MMR deficiencies. Reduced MSH2, MSH3 and MLH1 slowed repeat expansion to the largest degree, while lowering either PMS1, PMS2 and MLH3 slowed it to a lesser degree. These effects were recapitulated in iPSC derived striatal cultures where MutL factor expression was lowered. Here, reducing the expression of MMR factors by CRISPRi to levels typically reached by current therapeutics effectively slows the pathogenic expansion of the HTT CAG repeat tract.
MSH3 is relatively tolerant of loss-of-function variation in humans, making it an ideal therapeutic target. Here, we show that an MSH3-targeting antisense oligonucleotide (ASO) effectively enters HD patient iPSC-derived striatal neurons. ASO treatment led to a dose-dependent reduction in MSH3 levels and CAG repeat expansion. Maximal knock-down also effectively slowed CAG repeat expansion in FAN1–/– neurons, which otherwise have an accelerated expansion rate. Bulk and single nucleus RNA-Sequencing identified significant transcriptional changes associated with MSH3 lowering, which were not cell-type specific and revealed no disruption of the DNA damage response or cancer pathways. Additionally, we validate the efficacy of the ASO in vivo with a novel humanised MSH3 knock-in mouse model, demonstrating target engagement and excellent in vivo pharmacology. Our study validates the use of ASO-mediated MSH3 lowering to delay onset and slow disease progression in HD patients.

Top

Genetic modifiers of somatic CAG repeat expansions as a therapeutic target for Huntington’s disease

Ricardo Mouro Pinto, PhD, Massachusetts General Hospital & Harvard Medical School

Huntington’s disease (HD) is a rare neurodegenerative disease caused by an expanded CAG repeat in the HTT gene, with larger alleles being associated with earlier disease onset and more severe clinical phenotypes. Despite this being a single gene disorder, where the underlying genetic mutation has been known for just over 30 years now, there remains no cure or disease-modifying therapy, indicating that novel approaches are critical. A hallmark of HD, and most repeat expansion disorders, is that the repeats are highly unstable, both intergenerationally (parent to child) and in somatic tissues, where the repeat expands progressively over time in a cell-/tissue-specific manner. Notably, in HD, medium-spiny neurons of the striatum, which succumb most severely to the effects of the HTT mutation, exhibit the most dramatic CAG expansions. These observations, together with growing genetic evidence from genome-wide association (GWA) studies in HD patients, support the hypothesis that progressive repeat length increases in somatic tissues contribute to the pathogenic process. Consequently, the HD disease process can be viewed as two components: 1) somatic CAG expansions are a rate driver mechanism that determines the timing of disease onset, and 2) a toxicity mechanism that triggers the damage caused by a CAG repeat expanded above a critical threshold. To date, among the broader DNA damage response, we have primarily implicated the DNA mismatch repair pathway as the critical driver of somatic repeat expansions. We are developing CRISPR-based therapeutic strategies targeting this process as a therapeutic for HD and potentially many other repeat expansion disorders.

Top

Identifying good targets for reducing somatic instability in multiple repeat expansion diseases

Karen Usdin, PhD, National Institutes of Health

The identification of a single target whose depletion would safely reduce expansions in multiple repeat expansion diseases (REDs) could expedite the development of effective therapeutics. To better understand the commonalities and differences between the factors impacting expansion in different REDs, we have generated mice carrying both an expanded CAG-repeat at the Htt locus and a CGG-repeat in the Fmr1 locus that is associated with the Fragile X related disorders (FXDs). These mice show similar levels of expansion of both repeats in many tissues. However, there are some important differences in some cell types that provide clues to cis-acting modifiers of expansion risk. We have also previously shown that PMS2 and MLH3 are required for expansion in a mouse embryonic stem cell (mESC) FXD model, while PMS1 is also important, perhaps essential. This finding is surprising since these 3 MLH1-binding proteins have not previously been reported to interact. Furthermore, there is evidence that in some systems PMS2 is protective. HD/FXD- mutant mice that lack PMS2 provide interesting insights into this effect. The role of PMS1 is also interesting since no clear somatic role for this relatively abundant protein has been described. Using patient derived iPSCs, we have now shown that a similar role for all 3 MLH1-binding partners is also seen in a CAG-repeat expansion disease that results in a glutaminase deficiency. PMS1 is also important for expansion in iPSCs from a patient with Friedreich’s ataxia, a GAA-repeat expansion disease. These data, together with the fact that the loss of PMS1 is not associated with any clear deleterious effect in somatic cells, makes PMS1 an appealing therapeutic target.

Top

Developing oral small molecule splice modulators targeting PMS1 to treat repeat expansion diseases: Huntington’s disease and beyond

Travis Wager, PhD, Rgenta Therapeutics

The PMS1 gene is a key component of the DNA mismatch repair pathway, implicated in the pathological somatic trinucleotide repeat expansion observed in Huntington’s disease (HD) and other trinucleotide repeat expansion disorders such as DM1, FXS, and FRDA. Genome-wide association studies (GWAS) have identified PMS1, along with other genes in the mismatch repair pathway (e.g., MLH1, MSH3, PMS2, and FAN1), as genetic modifiers impacting the age of onset in HD. Functional studies, including the genetic knockout of PMS1 in HD mouse models, which stalled repeat expansion, further reinforce the gene’s important role in the progression of somatic repeat expansions.
To date, there has been a lack of successful strategies for directly targeting the PMS1 protein with small molecule inhibitors. Leveraging the integrative RNA-targeting small molecule discovery platform from Rgenta Therapeutics, we have pioneered the development of first-in-class oral small molecules. These molecules are designed to modulate PMS1 pre-mRNA splicing, curtailing the production of the aberrant PMS1 protein, thereby potentially preventing or decelerating the onset of somatic repeat expansion diseases. Our leading compounds have demonstrated the ability to decrease PMS1 RNA and protein levels in human cell lines and in animal models. Significantly, these molecules exhibit promising pharmacological efficacy in inhibiting repeat expansions in disease models and possess favorable pharmacokinetic profiles that facilitate crossing the blood-brain barrier. We will detail our discovery process, highlight the progress of our oral small molecule splice modulators, and discuss their potential in treating the underlying causes of repeat expansion diseases.

Top

One brain, many genomes: Somatic mutation in human brain during development and disease

Christopher A Walsh, MD, PhD, Boston Children’s Hospital, Howard Hughes Medical Institute, & Harvard Medical School

Although it had long been assumed that the genomes of all neurons are identical, recent work shows that every cell division causes mutations even during normal development, and that postmitotic neurons continue to accumulate mutations throughout life even in the absence of cell division. Recent studies implicate clonal somatic mutations in some brain malformations, other pediatric focal epilepsy and adult temporal lobe epilepsy, and some cases of autism spectrum disorders and schizophrenia. These mosaic causes of neuropsychiatric disease illuminate mutations that are different than those that create risk for these diseases when present in all cells, suggesting they may be important for understanding the mechanisms of these conditions. Reading out functionally silent developmental mutations reveals a post-mortem, forensic cell lineage map that records the cell divisions that generate each person, and differences in the ways that different humans develop.
Sequencing the DNA genome from a single neuron reveals a universe of genomic diversity, with transposon insertion, copy number variants, and hundreds of single nucleotide variants (SNV) distinguishing the genome of one neuron from another. Surprisingly, neurons accumulate about 20 mutations per year in human brain, with more than a thousand such mutations present in old age, even though neurons do not undergo cell division. SNV accumulate faster in rare genetic disorders associated with precocious neurodegeneration, and in more common forms of degeneration as well such as Alzheimer’s disease, suggesting that this age- related mutation is likely relevant to many forms of age-related neurodegeneration, potentially including Huntington’s disease.

Top

Dysregulation and rescue of subthalamic nucleus locomotor function in Huntington’s disease mice

Mark D Bevan, PhD, Northwestern University

As a key component of the basal ganglia indirect and hyperdirect pathways, the subthalamic nucleus (STN) is thought to restrict movement. Consistent with this view, lesion or prolonged inhibition of the STN increases movement vigor and propensity, while optogenetic excitation has opposite effects. Subthalamic and motor activity are also inversely correlated in some movement disorders. However, most STN neurons increase their activity during normal movement. Furthermore, in hypokinetic Huntington’s disease (HD) mice, the firing of a subset of STN neurons is impaired through NMDA receptor-dependent mitochondrial oxidant stress and KATP channel activation. To address the role of STN pathology in HD, STN activity was recorded and manipulated in head-fixed 6-9 month wild-type (WT) and Q175 HD mice at rest and during self-initiated/-paced treadmill locomotion. Most STN neurons (type 1) exhibited locomotion-dependent increases in activity, with half also encoding the locomotor cycle. A minority of STN neurons exhibited reductions in activity or were uncorrelated with movement. Brief optogenetic inhibition of the dorsolateral STN, where type 1 neurons are concentrated, slowed, dysregulated, and prematurely terminated locomotion in WT mice. In Q175 HD mice, abnormally brief, low-velocity locomotion, analogous to that generated by optogenetic inhibition, was specifically associated with hypoactivity of type 1, motor cortical-recipient STN neurons.
To determine whether loss of STN activity is cell autonomous and contributes to locomotor deficits in Q175 mice, STN mHTT expression was lowered through viral expression of a zinc finger protein-linked transcriptional repressor targeting the expanded CAG repeat. This treatment alleviated STN mitochondrial oxidant stress and rescued firing ex vivo and in vivo. Although the duration and velocity of self-initiated locomotion were significantly increased, behavioral rescue was partial, presumably due to aberrant activity in untreated, upstream nuclei. Consistent with this hypothesis, we found that iSPNs and prototypic external globus pallidus neurons are hyperactive in Q175 mice, possibly due to impaired astrocytic regulation of extracellular K+. Together these and other data argue that 1) mHTT dysregulates the basal ganglia indirect and hyperdirect pathways through cell class-specific, cell-autonomous and non-cell-autonomous mechanisms, and 2) widespread mHTT lowering throughout the corticobasal ganglia thalamocortical circuit will be required to effectively treat HD.

Top

Insights into the regional vulnerability of Huntington disease from an astrocyte-centric point of view

Osama Al-Dalahmah, MD, DPhil, Columbia University Irving Medical Center, New York Presbyterian Hospital

Huntington disease (HD) is caused by a mutation in the HTT gene which is present in all brain cells. However, the neuropathology of HD exhibits distinct regional, and cell-type specific, vulnerability. This is most pronounced in the striatum which shows selective degeneration of medium spiny projection neurons. Other brain regions including the cingulate cortex are also affected to varying degrees, and the mechanisms underlying HD’s selective regional and cell-type specific vulnerability is under investigation. From the point of view of a glial biologist and a neuropathologist, astrocytes are prime candidates to contribute to HD’s regional vulnerability because they are neuroanatomically varied, show regional heterogeneity, and exhibit “reactive states” in HD. To date, it is unknown which HD astrocytic states are compensatory or detrimental, and how these states correlate with neurodegeneration.
In this talk, I will present a multi-omic approach to map the states of astrocyte reactivity in HD across different brain regions. I will describe the results of bulk-level and single nucleus RNAseq from HD and control post-mortem brains, as well as the results of lipidomic analysis. I will then describe how we integrated the results and focused on two broad astrocytic states: one is a disease-associated state and is enriched in genes correlated with CAG repeat length, and the other is a compensatory state and is depleted in regions that exhibit severe neurodegeneration. I will provide support for the neuroprotective role of the compensatory astrocytic state from genome-wide association studies and in vitro co-culture models. Together, I hope to convince you that HD is a brain disease where astrocytes play a major role.

Top

Cell-type specific somatic CAG expansion and epigenetic regulation in Huntington’s disease

Matthew T Baffuto, BS, The Rockefeller University

To understand molecular mechanisms accompanying progression in Huntington’s disease (HD), our laboratory has developed FANS-seq, an innovative methodology for deep profiling of transcription, chromatin accessibility, somatic CAG expansion, DNA methylation and hydroxymethylation, and histone modifications in specific cell types in the human brain. This presentation will include data from analysis of the striatum, cerebral cortex, thalamus, hippocampus, and amygdala. Our data demonstrate that extensive somatic expansion occurs in specific cell types in each of these structures.
To understand mechanisms regulating transcription of genes known to play important roles in somatic expansion, we will present data identifying enhancers present in key MMR genes that are both species and cell type specific. We will present also epigenetic findings that correlate with altered transcription in cell types undergoing somatic CAG expansion, and discuss these data in light of mechanisms previously implicated in transcriptional dysregulation in HD. Together, these discoveries emphasize the crucial role of cell-type specificity in unraveling both the molecular mechanisms underlying somatic CAG expansion and pathogenesis in HD.

Top

Huntington’s disease neuropathology is asynchronous and results from long somatic CAG-repeat expansion: Evidence from diverse sources

Bob Handsaker, BS, Harvard Medical School & Broad Institute

Analyses of Huntington’s disease modifier genes have suggested a potential role for somatic CAG-repeat instability. I will describe experiments that, in multiple biological contexts (striatum and cortex; human and mouse), are converging on a shared conclusion: somatic expansion beyond a high threshold of about 150 CAG-repeat units drives HD neuropathology.
Our results are based on an approach we developed for measuring, at single-cell resolution, the HTT CAG- repeat expansions and genome-wide RNA expression. Analyzing brain tissue from the striatum (caudate nucleus) of persons with HD reveals that rates of CAG-repeat expansion vary by cell type; long somatic expansions (beyond 100 CAG-repeat units) are seen in the neuronal cell populations that are vulnerable in HD, while other cell populations exhibit more-modest somatic expansion. Our analyses suggest that somatic expansion progresses asynchronously in individual neurons and is highly nonlinear, with a small fraction of neurons having reached repeat lengths of 100-1000 CAGs at any moment in time (even as most have 40-100 repeats). Neurons with expansions beyond 150 CAGs exhibit distorted gene expression, which escalates with further somatic expansion and culminates in the expression of a senescence pathway.
Analyses of the cerebral cortex of persons with HD are revealing an analogous trajectory in cortical projection neurons. As in the striatum, cortical projection neurons begin to exhibit a large constellation of gene-expression changes as the CAG repeat crosses a threshold of about 150 repeat units; these transcriptional distortions escalate with further expansion beyond this threshold. We observe this relationship (and the same high repeat-length threshold) in cortical projection neurons in all cortical layers.
Analogous experiments on the striatum and cortex of a mouse HD model (Q111) further support the idea that transcriptional distortions occur as the CAG repeat expands beyond about 150 repeat units; however, the higher starting repeat length in mice greatly reduces the asynchronicity with which individual neurons cross this threshold.
These and other lines of evidence are converging in support of our model that long somatic repeat expansions are the necessary and sufficient driver of neuropathology in HD, and that individual neurons reach this threshold at very different times.

Top

Humanizing drug discovery

David Altschuler, MD, PhD, Vertex Pharmaceuticals

The discovery of a new medicine starts with a therapeutic hypothesis: that modulating a target with a specific approach may provide a positive benefit:risk profile to a defined set of patients. Unfortunately, despite great progress in science and medicine, the success rate of candidate therapeutics is low — less than 10% of candidate medicines that enter clinical trials emerge as approved medicines. This low rate of success is particularly troubling because discovering, developing, and launching a new medicine takes many years and substantial financial investment.
While there are many reasons that a candidate medicine may fail, perhaps the most critical is the validity of the therapeutic hypothesis. If the therapeutic hypothesis is invalid, then however innovative and well- executed the trial may be, it is certain to fail. Initially, therapeutic hypotheses came from serendipity and observation of the human condition, and later from human physiology, endocrinology, and infectious diseases. During this period, relatively few therapeutic hypotheses could be tested, and yet success rates were higher than today, and resulted in many medications that are mainstays of clinical medicine. By the later part of the 20th century, most therapeutic hypotheses came from laboratory model systems, and industrialization was used to enable testing of many “shots on goal” as an approach to overcome low success rates. Over the past 50 years the rates of success in drug discovery have fallen substantially, and the costs to discover and develop each successful medicine have risen.
One approach to increase rates of success is to focus on hypotheses derived from causal human biology. Human genetics has identified the underlying molecular causes of many serious diseases, and there are now many examples where addressing the underlying cause results in patient benefit. Moreover, post-hoc analyses have demonstrated that candidate medicines that address targets with human genetic evidence have higher rates of success than those that lack human genetic evidence.
However, the challenges of basing therapeutics on human genetics are many and significant. There is a long path from discovery of a gene and variant to understanding the causal mechanism responsible for disease pathophysiology. Even if the causal mechanism is known, we often lack a therapeutic approach needed to address each biological mechanism. As Tolstoy wrote “happy families are all alike; each unhappy family is unhappy in its own way.” In other words, normal biology may be the same in many models, and studied with shared approaches. But solving each disease represents a unique challenge requiring a willingness to develop and apply whatever the patients require.
I will discuss learnings from multiples programs undertaken by Vertex Pharmaceuticals, including cystic fibrosis, sickle cell disease, APOL1-mediated kidney disease and other serious diseases.

Top

Genetic modifiers and somatic repeat instability

Vanessa Wheeler, PhD, Massachusetts General Hospital & Harvard Medical School

Genome-wide association studies (GWAS) in Huntington’s disease (HD) individuals support a two-step model of HD pathogenesis in which (1) the timing of phenotypic onset is determined by the rate of somatic CAG expansion, and (2) somatically expanded repeats trigger a toxic process(es) in vulnerable cells, resulting ultimately in clinical disease. Routes to slowing somatic CAG expansion therefore hold great promise as disease-modifying therapies that can act prior to subsequent toxicity. Somatic expansion is characteristic of many other microsatellite repeat disorders and therapies targeting this mechanism are likely to have broad application. We are taking a number of approaches to gain further insight into human genetic modifiers and somatic expansion. To identify novel modifiers of somatic CAG expansion we are using CRISPR-Cas9 editing in HD knock-in mice, using AAV-mediated delivery of guide RNAs to tissues of interest. This has enabled candidate in vivo HD modifier gene screening at scale, including testing of human GWAS candidates and interactions between modifier genes, providing new insight into pathways underlying CAG expansion and potential therapeutic targets. In parallel, we are leveraging human genetic variants to gain insights into the underlying mechanisms associated with disease modification. Finally, we are testing the impact of human modifier genes on phenotypic expression in mice with the goal of establishing genetically validated endpoints for therapeutic testing.

Top

Peripheral biomarkers of somatic expansion as a therapeutic target in Huntington disease and related disorders

Darren G Monckton, PhD, University of Glascow

Somatic expansion of the HTT CAG repeat is a key step in the pathogenesis of Huntington disease (HD), and a novel target for therapeutic intervention. To evaluate peripheral biomarkers for the in vivo efficacy of any such intervention, we have performed longitudinal analysis of the dynamics of somatic expansion of the HTT repeat in blood DNA using high-throughput ultra-deep sequencing. In the majority of HD individuals, who inherit relatively small expansions <46 CAG repeats, the magnitude of such changes are relatively modest. Thus, we have sought to define in more detail the longitudinal dynamics of the HTT repeat in individuals inheriting disease-associated alleles >46 CAG repeats. These data confirm that larger alleles are expanding very rapidly, and inclusion of such individuals in clinical trials should dramatically increase power. Improvements in the assay such as technical and biological replicates can also increase the sensitivity of the assay. We have also explored the potential utility of the ERDA1 locus that has a high frequency (~15%) of expanded alleles (50 to 100 CAG•CTG repeats) in the general population. Surprisingly, expanded ERDA1 alleles are remarkably stable with no evidence for any detectable somatic expansion in blood cell lineages. These data confirm that even for large-expanded alleles >70 repeats, cell-lineage dependent factors, including cell division, are insufficient to promote somatic expansion, and confirm a major role for additional cis-acting modifiers. In further support of a key role for cis-acting modifiers, we also show that the length- dependent rate of expansion of the ATXN3 CAG in the blood DNA of individuals with spinocerebellar ataxia type 3 (SCA3) is much lower than at the HTT locus. However, the average ATXN3 repeat length in SCA3 (~72 repeats) is much higher than the average HTT repeat length in HD (~43 repeats), and average somatic expansion rates are higher in SCA3. Moreover, ATXN3 expansion rates are highly age dependent, but only minimally length dependent. Thus, although SCA3 is much rarer than HD, the proportion of the patient population that might be suitable for a peripheral biomarker study might be higher.

Top

CTG18.1-mediated Fuchs endothelial corneal dystrophy: Molecular insights into a common trinucleotide repeat-associated disease

Alice E Davidson, PhD, University College London Institute of Ophthalmology

Fuchs endothelial corneal dystrophy (FECD) is an age-related corneal disease that represents the leading indication for corneal transplantation in the Western world. Remarkably, up to 81% of European FECD patients harbor at least one expanded copy of an intronic triplet repeat within the TCF4 gene (termed CTG18.1; OMIM #613267), making it the most common trinucleotide repeat expansion disease reported to date. FECD therefore signifies an important paradigm for much rarer, and currently incurable, devastating repeat expansion diseases, including Huntington’s disease. To date, several non-mutually exclusive mechanisms have been proposed to drive and/or exacerbate the onset of disease. These mechanisms include (i) TCF4 dysregulation; (ii) toxic gain-of-function from TCF4 repeat-containing RNA; (iii) toxic gain- of-function from repeat-associated non-AUG dependent (RAN) translation; and (iv) somatic instability of CTG18.1. The presentation will provide an up-to-date report on associations among genotype, ethnicity, sex, and age at first corneal transplant surgery (as a surrogate metric for disease severity) in a large, multicenter, and extensively genetically characterized FECD patient cohort. Furthermore, a current overview of molecular mechanisms hypothesised to underpin the disease will be provided, including new insights gained from delineating RNA toxicity and TCF4 isoform-specific dysregulation events in primary patient-derived corneal endothelial cell systems. A novel data series from patient-matched blood and corneal cells will also be presented to highlight our current understanding of somatic instability events that may, in part, explain the tissue-specific nature of this common and sight threatening disease.

Top

Enabling precision health at scale for all

Presentation not made available by presenter

Carlos D Bustamante, PhD, Galatea Bio Inc

The last 20 years have seen an explosion of genetic information and data. New technological advances have made it faster and less expensive to understand human genomes but most of such resources have gone to further understanding predominantly northern European communities. There are tremendous discoveries to be made from looking at large populations. This thesis has been supported by global biobanking efforts but so far these have focused primarily on northern European communities. We believe that expanding the diversity of the dataset can dramatically accelerate the discovery rate in the field of genomics.
I will discuss the work we are doing at Galatea Bio to build the Biobank of the Americas and deliver Precision Health at scale to all. I’ll discuss our 500,000 person cohort and plans to reach 10M participants with support from biopharma and health care partners, including Stanford Health.

Top

Population-scale analysis of the trinucleotide repeat expansion in the huntingtin gene from 854,251 human exomes

Presentation not made available by presenter

Sahar Gelfman, PhD, Regeneron Genetics Center

Population-scale characterization of normal and expanded HTT repeat length is lacking. Here, we leveraged whole-exome sequencing (WES) data to call HTT repeat expansions in six large EHR cohorts including the UK Biobank. Using these calls, we estimated prerisk/risk prevalence in various ancestries, and validated the results both experimentally and through association analyses. We used gangSTR to call repeat expansions across 854,251 samples, including 762,895 samples of European ancestry (89%), 41,130 samples of African ancestry (5%), and 31,784 samples of South Asian ancestry (4%). To estimate the accuracy of this calling approach, we experimentally validated 27 HD cases and 27 randomly chosen controls using a PCR-based approach (Asuragen AmplideX PCR kit). We assessed various sequence derived metrics and applied strict quality control criteria to the full cohort, and compared repeat lengths across cohorts and ancestries. Last, we binarized repeat length based on either prerisk/pathogenic (36 & 40) or percentile (Top 1%/0.1%/0.01%) cut-offs and associated them with case-control status using 49 HD cases and 558,736 controls. We find the sensitivity and specificity of the predictions to be 74% and 100%, respectively. Four HD cases predicted with non-pathogenic expansions were validated to carry normal repeat length. The prevalence of pre-risk (0.23%; n=1,916) and pathogenic expansions (0.053%; n=437) were higher than the prevalence of HD in general population (0.003-0.006%), yet consistent with literature. We also found the prevalence of pre-risk expansions to be higher in Europeans compared to south Asians (0.24% vs. 0.16%; Fisher’s exact p=7.4e-3). A meta-analysis showed that the top 0.1% repeat length of 38 associated most strongly with HD (OR=213.4; p=8.9e-37), compared with the risk threshold cutoff of 40 that had the largest effect size (OR=331.4, p=1.9e-32). This is the largest population-scale characterization of HTT repeat expansions called using whole exome sequencing data. Our results demonstrate that normal and pathogenic repeat expansions can be identified with high specificity and adequate sensitivity and help refine population prevalence and penetrance estimates.

Top

Update on phase I/II clinical trial of AMT-130 gene therapy for the treatment of Huntington’s disease

David H Margolin, MD, uniQure

AMT-130 is an investigational AAV5 viral vector containing an exon-1 HTT-targeting miRNA, thereby lowering production of HTT protein, administered by intra-striatal injection bilaterally. This presentation will describe interim results up to 24 and 30 months from the combined US and European high-dose and low-dose cohort participants, respectively (NCT04120493 and EudraCT 2020-001461-36).
Data from 39 patients will be presented. Baseline demographic and disease characteristics were generally balanced across all groups. The most common adverse events (AEs) for treated cohorts were related to lumbar punctures or the drug administration surgery. Serious AEs for treated cohorts were related to lumbar punctures, the drug administration surgery, or drug-related CNS inflammation.
Relative to baseline, patients treated with both doses of AMT-130 showed evidence of preserved neurological function on some measures. The high-dose cohort preserved diverse functions relative to baseline, with cUHDRS at or above baseline throughout year 2. The low-dose cohort TMS and TFC scores recovered to baseline during year 2. CSF NfL was acutely high post-operatively and then steadily declined towards baseline. CSF mHTT levels were close to the lower-limit of quantitation at baseline.
AMT-130 was generally well tolerated with a manageable safety profile. Clinical measures and CSF NfL showed promising trends and follow-up is ongoing.

Top

Twelve-week interim data from PIVOT-HD: a phase 2, randomized, placebo-controlled study to evaluate the safety and efficacy of PTC518 in participants with Huntington’s disease

Amy-Lee Bredlau, MD, PTC Therapeutics

Background: Huntington’s disease (HD) is caused by expansion of CAG trinucleotide repeats in the huntingtin (HTT) gene. The mutant protein is ubiquitously expressed and causes HD pathogenesis through a toxic gain- of-function mechanism. Animal models demonstrate that reducing mutant huntingtin protein (mHTT) levels alleviates HD symptoms. PTC518 is a splicing modifier that promotes the inclusion of a pseudoexon (psiExon) containing a premature stop codon, leading to HTT mRNA degradation and lowering of HTT levels.
Objective: The objectives of this planned i12-week interim data readout are to evaluate the safety and biomarker effects of PTC518 in participants with HD from PIVOT-HD, an ongoing phase 2, placebo-controlled study.
Methods: Participants with HD were randomized to receive PTC518 5 or 10 mg or placebo once daily for 12 months. Interim safety and biomarker data from the 12-week portion of PIVOT-HD will be reported.
Results: Thirty-three participants were included in this planned interim readout. PTC518 dose-dependently lowered blood mHTT protein 27% and 29% from baseline levels for the 5 and 10 mg doses, respectively, with similar ratios of mRNA decreases seen at week 12. PTC518 exposure ratios in the cerebrospinal fluid (CSF) were consistent with or higher than the plasma unbound drug levels. Finally, PTC518 treatment was well tolerated, with no treatment-related serious adverse events and no reports of peripheral neuropathy or dose-limiting toxicities. There were no treatment-related spikes in CSF and plasma neurofilament light chain protein.
Conclusions: These interim results demonstrate that PTC518 is a potential therapeutic for the treatment of HD. (NCT05358717/2021-003852-18).

Top

New fluid biomarker data from GENERATION HD1 and what it means for GENERATION HD2

Peter McColgan, MD, PhD, F. Hoffmann-La Roche Ltd

In this presentation we will share new fluid biomarker data on CSF YKL-40, total Tau and GFAP from the GENERATION HD1 clinical trial with the huntingtin-lowering drug tominersen. We will focus on these markers as they have previously been explored in HD observational studies. We will first review what these biomarkers may mean biologically and how they change over time in HD, with respect to both the literature and the GENERATION HD1 longitudinal placebo data.
We will then discuss how these measures are affected by tominersen across a range of drug exposures (pooled across 120 mg every-8-weeks and every-16-weeks dosing regimens), demonstrating the avoidance of unfavourable biomarker changes at lower tominersen exposures. These data support the investigation of lower doses of tominersen in keeping with the 100 mg and 60 mg every-16-weeks dosing regimens in the ongoing GENERATION HD2 trial.
We will also provide a study update on GENERATION HD2 including participating trial sites and recruitment of participants.
Background: Tominersen is an antisense oligonucleotide that lowers huntingtin (HTT) protein. GENERATION HD2 is an ongoing phase II dose finding study of tominersen, evaluating safety, biomarker profiles and efficacy in individuals with prodromal and early manifest HD.
The design of GENERATION HD2 was informed by GENERATION HD1, a previous phase III study of tominersen where post-hoc analyses identified potential benefit in younger individuals with less disease burden at lower tominersen exposure. GENERATION HD1 was stopped early based on an overall benefit-risk assessment and these data have been presented previously; the wealth of data from GENERATION HD1 provide a unique and valuable resource that can be used to inform therapeutic development in HD.

Top

Innovation in Huntington’s disease clinical trial design

Jonas Dorn, PhD & Marcelo Boareto, PhD,F. Hoffmann-La Roche Ltd

A key challenge for drug development in Huntington’s disease (HD) is the limited sensitivity of the current clinical measures when it comes to progression over time, which leads to large and lengthy clinical trials. This not only delays the availability of treatments, it also makes it challenging for the HD field to run multiple trials in parallel as there are a limited number of individuals with HD that can participate in clinical trials.
We will present two complementary approaches toward optimization of study design: (1) Using the Huntington’s Disease Digital Motor Score (HDDMS) to measure disease progression with higher sensitivity, and (2) Utilizing disease progression modeling to increase statistical power.
(1) The HDDMS has been developed on data from >1000 people with HD. It captures motor performance of upper limbs, balance, and gait using smartphone-based assessments at home, combining them into a single score. The HDDMS has good psychometric properties and a higher sensitivity to change than current clinical measures such as cUHDRS and TFC. We will share initial analyses that suggest clinical trials powered for the HDDMS may allow for a reduction in sample size compared to trials powered for cUHDRS.
(2) We will present our efforts to quantify disease progression and placebo response using data from Enroll-HD and GENERATION HD1. We will show that patient characteristics at baseline are strong predictors of progression rate for different clinical scores and that placebo and learning/practice effects may affect motor and cognitive measurements. By correcting for these effects, disease progression in GENERATION HD1 appears to align with the expected progression inferred from Enroll-HD. By understanding disease progression and placebo response we will show how we can design more effective trials and potentially reduce sample sizes by 20% without mistakenly rejecting the null hypothesis (i.e. type I statistical error).
Together, these approaches may lead to smaller and shorter trials, enabling efficient development of novel therapies for HD.

Top

SHIELD-HD: First look at the data from the natural history study in people with Huntington’s disease, HD-ISS Stages 0 through 3

Swati Sathe, MD, CHDI

SHIELD-HD was initiated to obtain longitudinal information related to somatic instability and DNA damage repair genes in PwHD at various stages of the disease and correlate clinical assessments with genetic information. The study was expected to guide the design of a future interventional treatment trial in HD. SHIELD-HD was conducted for 96 weeks in 70 participants and continued to 120 weeks in 23 participants. CHDI is conducting an analysis on the data collected in SHIELD-HD with modified goals of studying various markers of disease progression both cross-sectionally and longitudinally, including those derived from brain MRI, biofluids, and clinical assessments. This presentation will provide data analysis conducted till end of January 2024.