CHDI’s 14th Annual HD Therapeutics Conference took place February 25 – February 28, 2019, in Palm Springs. This unique conference series focuses on drug discovery and development for Huntington’s disease, and draws participants and speakers from the biotech and pharmaceutical sectors as well as academia and research institutions. 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.
- Epigenetic control of myelination and functional regeneration Qing Richard Lu, PhD, Cincinnati Children’s Hospital Medical Center
- Augmenting endogenous and transplant-mediated myelin repair by human oligodendrocyte progenitors Fraser Sim, PhD, SUNY University at Buffalo
- White matter loss in premanifest HD: From systems to cellsPeter McColgan, MB, BCh, BAO, PhD, University College London
- Modelling network spread via white matter to predict neural degeneration and disconnection in Huntington’s disease Govinda Poudel, PhD, Australian Catholic University
- Gene therapy for Huntington’s disease: Silencing the villain Pavlina Konstantinova, PhD, uniQure BV
- VY-HTT01, an AAV miRNA gene therapy targeting huntingtin for the treatment of Huntington’s disease Dinah Sah, PhD, Voyager Therapeutics
- Allele specific editing of HTT using CRISPR/Cas9 approaches Beverly L Davidson, PhD, University of Pennsylvania/Children’s Hospital of Philadelphia
- Identification and development of orally administered, CNS-penetrant small molecules that lower huntingtin protein levels by inducing a novel splicing event that alters the stability of huntingtin mRNA Anuradha Bhattacharyya, PhD, PTC Therapeutics
- Harnessing biofluid biomarkers for early detection of Huntington’s disease and enrichment of clinical trials Lauren Byrne, MRes, University College London
- Therapy-mediated reduction of cerebrospinal fluid mutant huntingtin: What does it mean? Amber Southwell, PhD, University of Central Florida
- Small RNA biomarkers for the early detection and monitoring of Huntington’s disease David W Salzman, PhD, sRNAlytics, Inc.
- Genetically engineered non-human primate models for brain disorder research Guoping Feng, PhD, Massachusetts Institute of Technology
- Disease modeling and brain mapping using genetically modified marmosets Hideyuki Okano, MD, PhD, Keio University School of Medicine
- Characterization of the gene regulatory network of HTT Richard M Myers, PhD, HudsonAlpha Institute for Biotechnology
- The incomplete splicing of huntingtin Gill Bates, PhD, Queen Square Institute of Neurology, UCL
- Huntingtin post-translational modifications and enzymes as potential therapeutic targets Christopher A Ross, MD, PhD, Johns Hopkins University School of Medicine
- Systematic interaction mapping to define the quantitative huntingtin interactome Erich E Wanker, PhD, Max Delbrueck Center for Molecular Medicine
- Genetic modifiers Marcy MacDonald, PhD, Massachusetts General Hospital
- Genetic overlap between neuropsychiatric disorders and psychiatric symptoms in HD Peter Holmans, PhD, Cardiff University
- Human genetic modifiers of HD age of onset reveal DNA repair and handling genes as candidate therapeutic target Hilary Wilkinson, PhD, CHDI
- Accelerating the path from discovery of therapeutic targets to effective and safe medications: Human genomics to the rescue Alan R Shuldiner, MD, Regeneron Genetics Center
Epigenetic control of myelination and functional regeneration
Qing Richard Lu, PhD, Cincinnati Children’s Hospital Medical Center
Myelination by oligodendrocytes in the central nervous system is critical for rapid action potential conduction and to provide trophic support for axonal as well as neuronal maintenance. Mutant huntingtin protein is known to inhibit myelin gene expression and impair oligodendrocyte myelination, suggesting dysfunction of mature oligodendrocytes is involved in Huntington’s disease pathogenesis. Upon functional screening for small-molecule epigenetic modifiers, we identify a set of epigenetic-modifying enzymes including histone deacetylase 3 (HDAC3) as a potent inhibitor for myelinating cell maturation. We show that attenuation of HDAC3-mediated deacetylase activity markedly enhances myelin sheath growth and regeneration and improves functional recovery after peripheral nerve injury. Moreover, in the central nervous system, our preliminary studies indicate HDAC3 inhibition promotes remyelination in lysolecithin-induced demyelination lesion and restores the motor function in experimental autoimmune encephalomyelitis (EAE) demyelinating animal models. Thus, our findings identify a cell-intrinsic epigenetic inhibitory machinery that counters myelinogenic signals and maintains myelin homeostasis, suggesting that HDAC3 may be a potential therapeutic target for enhancing myelin repair in Huntington’s disease.
Augmenting endogenous and transplant-mediated myelin repair by human oligodendrocyte progenitors
Presenter could not attend – presentation not available
Fraser Sim, PhD, SUNY University at Buffalo
The adult human CNS harbors an abundant population of endogenous stem/progenitor cells that have the potential for extensive oligodendrocyte replacement and myelin repair. The emerging role for oligodendrocyte dysfunction in a number of neurological and neuropsychiatric diseases suggests that novel strategies aimed at harnessing endogenous oligodendrocyte progenitor cells (OPCs) to both replace dysfunctional oligodendrocytes and modulate neural function may represent a successful approach to treat these disorders. Using human primary OPCs, we have characterized a number of signaling pathways that can be targeted to overcome pathological inhibitors of oligodendrocyte generation that limit myelin repair and contribute to disease progression. By combining pharmacological and genetic approaches to overcome the inhibitory microenvironment in models of demyelinating disease, we are able to improve the efficacy of endogenous murine and transplanted human OPCs to elicit new myelin synthesis and remyelination. These complimentary strategies will be largely discussed in the context of primary demyelination. Indeed, as grafted human glia have been shown to modulate animal models of Huntington’s disease (HD), we anticipate that targeted OPC pharmacotherapy may have a similar potential to ameliorate HD pathology in a translationally accessible manner.
White matter loss in premanifest HD: From systems to cells
Peter McColgan, MB, BCh, BAO, PhD, University College London
White matter loss can be seen many years before disease onset in the brains of premanifest HD gene expansion carriers. The earliest white matter changes are seen around the striatum, within the corpus callosum and in the posterior white matter tracts.
By constructing white matter networks, using diffusion MRI, I reveal how specific white matter connections are affected in preHD. I demonstrate the selective vulnerability of cortical-striatal white matter connections to highly connected hub regions in the cerebral cortex, showing loss of these white matter connections correlate with symptoms in the premanifest stage.
Over time a hierarchy of white matter vulnerability is observed in preHD, where cortico-striatal connections are affected first followed by inter-hemispheric and intra-hemispheric connections. Change over 2 years is predominantly seen in posterior cortico-striatal connections.
I link these system level white matter changes to HD related abnormalities at the cellular level by investigating the relationship between regional gene expression and regional loss of white matter connections in preHD. I demonstrate how cortico-striatal and inter-hemispheric white matter loss is associated with the expression of synaptic genes, particularly those showing abnormal transcription in HD.
Having established cortico-striatal selective vulnerability I show how we can define sub-regions of the striatum based on white matter connectivity in healthy brains. This enables investigation of white matter loss to specific striatal sub-regions in preHD. Using longitudinal data from the Track-On HD study I show wide spread changes across the white matter of striatal sub-regions at baseline, with only motor-striatal sub- regions showing white matter loss over 2 years.
Finally, I outline the CLEAR-HD study where I will use ultra-high field 7T MRI to link white matter loss in preHD with markers of neuronal loss in the deep and superficial layers of the cerebral cortex. In a cohort of healthy participants, I demonstrate strong relationships between quantitative 7T MRI contrasts, layer specific cell count and gene expression. These contrasts are particularly sensitive to cortical genes showing abnormal transcription in HD.
Modelling network spread via white matter to predict neural degeneration and disconnection in Huntington’s disease
Govinda Poudel, PhD, Australian Catholic University
Emerging evidence suggests that both cell-autonomous and trans-axonal mechanisms contribute to the structured pattern of cortico-striatal degeneration and disconnection in Huntington’s disease (HD). We implemented a novel method, which models spread of neuropathology as a passive diffusion via white matter, to determine whether network spread determines pattern of degeneration and disconnection in Huntington’s disease. We used T1-weighted structural magnetic resonance imaging (MRI) and high-angular resolution diffusion weighted imaging (HARDI) data from symptomatic HD (N=26) and age-matched healthy control individuals (N=26) who participated in IMAGE-HD. The structural MRI data was analysed using freesurfer software to determine the pattern of volume changes in 82 cortical and sub-cortical regions (as per Desikan-killiany atlas) in HD compared to controls. The HARDI data was analysed using MRTrix and probabilistic tractography to develop a 82 x 82 canonical map of healthy white matter connectome and to identify the pattern of white matter disconnection in HD. The network diffusion model was tested on the healthy connectome to identify whether disease spread, from a particular sub-cortical or cortical region, can explain volumetric changes and white matter disconnection in HD. We found that diffusion of pathology, especially from sub-cortical brain regions, generates a spatial pattern that best recapitulates the typical neurodegenerative characteristics of HD. The best prediction was achieved by seeding diffusion from accumbens (r=0.52, p<0.001). Furthermore, white matter connections most vulnerable to diffusion were also found to be most susceptible to disconnection in HD, particularly when diffusion was initiated from the striatum. Application of network diffusion on null networks showed that the findings using the canonical connectome are significantly different from null models. These findings are the first to suggest that passive diffusion may facilitate trans-axonal spread of pathology in HD.
Disentangling the power of baseline white matter status for predicting future gray matter atrophy in premanifest Huntington’s disease
Dorian Pustina, PhD, CHDI
Early premanifest stages of Huntington’s disease are characterized not only by abnormally low striatal volumes but also by white matter (WM) damage. Differently from gray matter atrophy, however, the pattern of WM damage has not been consistently found, and ultimately the value of WM integrity as a potential biomarker remains unclear. In this presentation I will provide results from two recent studies that investigated WM properties in a multivariate, multimodal context with robust cross-validation.
In Study 1 we used regularized canonical correlations to search common patterns of variance from four types of neuroimaging features: 1) volumetric scores of putamen, caudate, and lateral ventricles, (2) longitudinal slopes of volumetric atrophy, (3) whole-brain functional connectivity, and (4) whole-brain WM diffusion anisotropy measures. Forty-seven premanifest HD and 55 healthy participants from the TRACK-ON study were included, and the stability of canonical correlations was investigated with multiple 4-fold cross- validations. WM anisotropy emerged as the most predictive modality overall; it predicted both baseline volumetric scores (canonical correlation r=0.67) and, more importantly, future atrophy slopes (r=0.39). These predictions were achieved by placing maximal importance to the integrity of the corpus callosum. In Study 2 we went one step further to train a support vector machine model on baseline data from one study (e.g., TRACK-ON, 47 premanifest HD) and explicitly predict future volumetric atrophy in another study (e.g. IMAGE-HD, 32 premanifest HD). Predictors were either baseline WM properties or baseline volumetric scores. Once again, future striatal atrophy was best predicted by baseline WM properties, this time across datasets and study populations, whereas baseline volumes showed no predictive power.
These results suggest that the structural integrity of white matter carries important information on the future progression of striatal atrophy, which has been the classical hallmark of pathology in HD. I will discuss these results in light of the emergence of neurofilament light as a marker of HD progression, the link found in the literature between WM integrity and neurofilament light levels, and other evidence implicating WM integrity as a predictor of future clinical scores. This research is part of a collaboration between IBM Research and CHDI.
Gene therapy for Huntington’s disease: Silencing the villain
Pavlina Konstantinova, PhD, uniQure BV
Extensive preclinical data in transgenic animal models supports that lowering aberrant mRNA and protein species ameliorates the molecular and clinical phenotype in several repeat expansion disorders, including Huntington’s disease (HD), spinocerebellar ataxias (SCA), and C9orf-related amyotrophic lateral sclerosis/ frontotemporal lobar dementia (ALS/FTD). uniQure has developed a therapeutic approach using the RNA interference mechanism that has high fidelity to avoid off-target effects, and efficiently lowers mutant mRNA and protein species by silencing targets in both the nucleus and cytoplasm. This platform, called miQure, uses a recombinant adeno-associated viral vector (rAAV) that carries a small microRNA against a human disease preRNA target. The AAV vector and transgene (rAAV-miRNA) is delivered to the relevant brain regions that are involved in the neuropathogenesis of specific diseases. For example, proof-of-concept studies using rAAV5-miHTT (AMT-130) in several HD animal transgenic models and Investigational New Drug (IND)-enabling GLP toxicology and safety studies in non-human primates, have shown excellent tolerability, efficacy, and safety using convection enhanced intra-striatal, neurosurgical delivery. Preclinical pharmacokinetic, pharmacodynamic, and biomarker translational studies have shown the applicability of clinical, neuroimaging and laboratory tests as potential efficacy endpoints in human HD clinical trials. A human equivalent dose to lower mutant huntingtin protein by 75% in the striatum and 50% in the cerebral cortex was extrapolated using a regression plot based on small and large HD transgenic animal model studies. The results of this model were used to design a blinded, imitation surgical controlled Phase 1/2 human clinical study using AMT-130 to establish the dose, safety, tolerability and explore efficacy signals in HD subjects. This clinical trial and IND-enabling studies in other human repeat expansion disorders using the miQure platform are planned for 2019.
VY-HTT01, an AAV miRNA gene therapy targeting huntingtin for the treatment of Huntington’s disease
This presentation not available at request of presenter
Dinah Sah, PhD, Voyager Therapeutics
VY-HTT01 is a potent AAV gene therapy encoding a primary miRNA targeting human HTT mRNA selectively for knockdown. Here, we describe studies in the nonhuman primate on characterizing the pharmacology, biodistribution and tolerability of HTT knockdown with VY-HTT01 in the cortex, putamen and caudate. MRI- guided convection-enhanced delivery into the putamen and thalamus resulted in distribution of VY-HTT01 to the cortex, caudate, putamen and thalamus, and robust suppression of HTT in these regions. Distribution and HTT knockdown in primary motor and somatosensory cortical neurons was demonstrated in laser captured cortical neurons and supported by in situ hybridization for vector genomes and HTT mRNA. Furthermore, good tolerability was supported by in-life observations, clinical pathology and histopathological analysis of the brain 5 weeks after dosing. Taken together, these results demonstrate the potential of VY-HTT01, an AAV gene therapy targeting HTT with RNAi, administered with combined infusions into the putamen and thalamus, for the treatment of Huntington’s disease.
This work was done in partnership with Sanofi.
Allele specific editing of HTT using CRISPR/Cas9 approaches
Beverly L Davidson, PhD, University of Pennsylvania/Children’s Hospital of Philadelphia
Huntington disease (HD) is a fatal dominantly inherited neurodegenerative disorder caused by CAG repeat expansion within the first exon of the huntingtin gene. Although mutant huntingtin (mHTT) is ubiquitously expressed, the brain shows robust and early degeneration. Current gene silencing approaches for lowering mHTT expression, including RNA interference or antisense oligonucleotides, have been efficacious in mouse models, but basal mutant protein levels are still detected. To fully mitigate expression from the mutant allele, we hypothesize that allele specific genome editing can occur via prevalent promoter-resident single nucleotide polymorphisms (SNPs) in heterozygosity with the mutant allele. This approach would also avoid reducing normal HTT protein levels. Here, we identified SNPs in HTT that either cause or destroy protospacer associate motifs critical for CRISPR targeting, using either fully active Cas9 or a nuclease dead Cas9 to induce epigenetic silencing. Importantly, in both cells from HD patients and a transgenic HD model harboring the human allele, we show selective targeting of the mutant allele, resulting in mHTT lowering while avoiding silencing of normal HTT.
Identification and development of orally administered, CNS-penetrant small molecules that lower huntingtin protein levels by inducing a novel splicing event that alters the stability of huntingtin mRNA
This presentation not currently available at request of presenter
Anuradha Bhattacharyya, PhD, PTC Therapeutics
Huntingtin (HTT) lowering has significant therapeutic potential for Huntington’s disease (HD), which is caused by an autosomal dominant polyglutamine repeat expansion mutation in the HTT gene. No disease-modifying treatments are currently available. Multiple huntingtin-lowering strategies are being employed in clinical and pre-clinical studies for the treatment of HD, including antisense oligonucleotides, RNA interference, zinc-finger transcriptional repressors and CRISPR/Cas9 gene editing. Our objective is to address the unmet medical need for patients with HD by developing orally bioavailable HTT-lowering molecules that would delay the onset or slow the progression of their disease. The HTT-lowering program is part of PTC’s proprietary pre-mRNA splicing platform, where we leverage our knowledge of splicing regulation as a novel platform to discover and develop small-molecule splicing modifiers. Utilizing this cutting-edge technology, we have identified small molecule splicing modifiers that cross the blood brain barrier upon oral delivery, and uniformly lower HTT levels in the key affected areas of the HD mouse brain – striatum and cortex. We have shown that these molecules modify splicing of the HTT pre-mRNA, resulting in the inclusion of a poison exon that leads to the degradation of HTT mRNA.
The presentation will describe how these small molecules reduce the level of pathogenic HTT in a dose- dependent manner in HD affected cells and animals, and thus have the potential to be utilized as therapeutics to alter the course of HD. We continue to optimize these molecules for improved oral bioavailability, penetration of the blood-brain-barrier and efficacy.
Harnessing biofluid biomarkers for early detection of Huntington’s disease and enrichment of clinical trials
Lauren Byrne, MRes, University College London
Huntington’s disease (HD) is a progressive neurodegenerative disorder where there is a pressing need for sensitive biomarkers. Quantification of mutant huntingtin (mHTT) in CSF (Wild et al, 2015, JCI) and neurofilament light (NfL) in CSF and blood (Byrne et al, 2018, Lancet Neuro) have emerged as promising potential biofluid biomarkers of HD progression.
We assessed mHTT and NfL in parallel to compare their relative performance as HD biomarkers using ultrasensitive immunoassays in 80 participants (20 healthy controls, 20 premanifest HD and 40 manifest HD). All underwent clinical assessments and standardized CSF and blood collections, with voluntary MR imaging.
CSF mHTT, CSF NfL and plasma NfL were significantly higher as disease progressed but NfL had stronger associations with all clinical measures than mHTT. Both CSF and plasma NfL were associated with brain volume measures, but CSF mHTT was not. CSF mHTT, CSF NfL and plasma NfL were closely correlated and highly stable within individuals. CSF mHTT had perfect accuracy for distinguishing between controls and HD mutation carriers, and both CSF and plasma NfL had excellent accuracy for distinguishing between premanifest and manifest HD. Sample size calculations suggest low participant numbers needed to incorporate these measures into clinical trials. Unbiased event-based computational modelling revealed the biofluid biomarkers to be the earliest detectable alterations in HD, preceding changes in caudate volume, motor, global brain volume and cognitive measures.
This cross-sectional study provides further evidence to support mHTT and NfL as biofluid biomarkers for HD and suggests that they are some of the earliest detectable changes in HD.
As well as presenting these recently-published data (Byrne et al, 2018, Sci Trans Med), I will discuss unpublished work on these and other biofluid biomarkers, as well as next steps for their application in upcoming HTT-lowering trials; preliminary data from the recently completed 24-month longitudinal follow- up of HD-CSF; an update on HDClarity – the multisite CSF collection initiative; and a possible future in which biofluid measures are incorporated into clinical assessment, counselling and decision-making.
Therapy-mediated reduction of cerebrospinal fluid mutant huntingtin: What does it mean?
Amber Southwell, PhD, University of Central Florida
The first huntingtin (HTT) lowering safety trial recently completed, demonstrating dose-dependent reduction of cerebrospinal fluid (CSF) mutant (mt) HTT. However, we don’t know where CSF mtHTT comes from or how it enters CSF, which confounds interpretation of treatment-induced changes. We are using our ultrasensitive immunoprecipitation and flow cytometry (IP-FCM) mtHTT detection assay to query what this exciting new data tells us about therapeutic activity in the brain.
The source of CSF mtHTT is unknown. Some therapeutics, such as ASOs, may be most active in regions in contact with the CSF. If these regions are the major sources of CSF mtHTT protein, then CSF-based predictions of HTT lowering in deeper structures may not be straightforward. Conversely, if regional contributions to CSF mtHTT are similar, then changes in CSF mtHTT level could be used to more accurately infer changes in basal ganglia mtHTT. Thus, we are interrogating the source(s) of CSF mtHTT protein using BACHD mice (floxed mtHTT exon 1) crossed to brain region and cell type-specific cre mice as well as ectopic delivery of mtHTT to restricted brain regions.
Additionally, we do not know how mtHTT enters CSF. CSF mtHTT is not detected in all premanifest HD mutation carriers and acute brain injury causes a transient increase in CSF mtHTT, suggesting that mtHTT is released from dying neurons. Thus, any neuroprotective therapy would be expected to reduce CSF mtHTT, and HTT lowering treatment-induced changes may represent a combination of target engagement and neuroprotection. However, we have observed mtHTT in the CSF of mice lacking neurodegeneration and reduced CSF mtHTT in BACHD mice lacking astrocytic mtHTT, suggesting that there may be both passive and active clearance mechanisms involved. If active clearance is the primary mechanism of mtHTT release, then treatment-induced changes may mostly represent HTT lowering. To investigate the mechanism(s) of mtHTT release to CSF, we are ectopically delivering intra and extracellular mtHTT with or without inhibitors of secretion and glymphatic clearance in the presence or absence of neuronal insult.
Delineating the source and mechanism(s) of entry of CSF mtHTT protein will greatly enhance interpretation of this promising HD biomarker.
Small RNA biomarkers for the early detection and monitoring of Huntington’s disease
David W Salzman, PhD, sRNAlytics, Inc.
Small RNAs (sRNAs) are a class 17-36 nucleotide, non-coding RNAs that regulate gene expression at the posttranscriptional level. The current library of human sRNAs consists of 60,606 annotated genes. However, sRNAs are subject to variability during and after biogenesis resulting in an array of sequence variants for each annotated gene called isoforms. Each isoform can have templated and/or non-templated variants at the 5’ and/or 3’ end that impacts the biological function and localization of these genes. sRNAlytics developed
a computational platform that combines next generation sequencing with big data analytics and machine learning to identify biomarkers of disease. Using our platform and data from Hoss et al. (2015), Reed et al. (2017) and other sources, we identified isomiRs that are uniquely expressed in the frontal cortex of Huntington’s patients, but not in >2,500 non-Huntington’s controls. IsomiRs were validated using targeted RT-qPCR in 119 independently collected samples from (64) brains, and (55) cerebrospinal fluid samples. Our results showed that these biomarkers are (i) only expressed in individuals carrying the huntingtin gene mutation, (ii) can be detected up to 20 years prior to symptomatic onset, and (iii) increase with abundance over time. sRNAlytics filed these biomarkers as a Context of Use for early-detection and disease monitoring through the FDA Biomarker Qualification Program. We believe these Huntington’s disease specific isomiRs will enable the discovery of new druggable targets and provide an efficacious biomarker to facilitate ongoing drug development programs.
Genetically engineered non-human primate models for brain disorder research
Guoping Feng, PhD, Massachusetts Institute of Technology
The advancement of genomic technologies has brought unprecedented insight into the genetics of various brain disorders. A pressing challenge now is to translate these genetic findings into neurobiological mechanisms for developing effective treatments. A key step towards meeting this challenge is to develop animal models that reflect human disease mechanisms. In the past, precise genetic manipulations were restricted to rodents. Although the ability to genetically modify the mouse genome has revolutionized biomedical research, its impact on our understanding of higher brain function and brain disorders is limited due to the inherent differences in the structure and physiology of the brain between rodents and humans. Non-human primates provide an attractive model to study higher brain function and brain disorders. They are much more closely related to humans than are rodents, and this is reflected in their brain development, structure and physiology. However, until recently genetic manipulations have been limited to rodents. The recent development of highly efficient CRISPR genome-editing technology made it feasible to directly manipulate the genome in zygotes, thus expanding genetic manipulations to many species. I will present recent advances in developing efficient methods for genetic engineering in non-human primates as well as novel primate models for studying brain disorders.
Disease modeling and brain mapping using genetically modified marmosets
Hideyuki Okano, MD, PhD, Keio University School of Medicine
The common marmoset (Callithrix jacchus) is a small New World primate that has been extensively used as biomedical research models. There is also an increasing interest in common marmoset in brain science due to the availability of genetic modification (GM) technology. We first developed GM technology of marmoset by lentiviral mediated transgenesis (Sasaki et al., Nature, 2009), followed by generation of knock-out marmoset using genome editing (Sato et al., Cell Stem Cell, 2016). In the present talk, I wish to mention our recent data of generation of transgenic marmoset models of neurodegenerative diseases, including Parkinson disease (PD). By using this technology, we generated transgenic marmoset lines harboring aggregation-prone α-synuclein (A30P) mutation, which is found in human familial PD. To evaluate the transgenic marmoset as a disease model, we analyzed phenotypes found at multiple stages in human PD progression in the transgenic marmoset. As a result, α-synuclein (A30P) transgenic marmoset exhibited early to middle stage of PD phenotypes, including sleep deficit and motor symptoms, respectively. Furthermore, we found that the fiber number of nigra-striatal pathway was reduced in the transgenic marmoset brain. It is notable that basal ganglia circuits change in transgenic marmoset like in PD patients. Furthermore, the pathological analysis showed that phosphorylated α-synuclein deposit and Lewy-body formation in the brain of the transgenic marmoset. Therefore, this animal model would be a powerful tool for investigating the pathogenesis of PD. In addition, I will mention a model marmoset of a neurodevelopmental disorder, Rett syndrome, obtained by genome editing of MECP2 gene. Abnormalities in brain structure and function in these marmoset models may accelerate discovery of disease biomarkers and mechanisms toward translation (Okano et al., Neuron, 2016).
In the present talk, I will also introduce new methods so called “Quick Marmoset Projects” to generate neuroscience models (e.g. subsets of neurons-specific reporter-expressing marmosets) and disease models quickly through the in vivo genome editing in the post-natal marmoset brains, using homology-independent targeted integration (HITI) which can be applied for post-mitotic cells. The in vivo genome editing by HITI method can be used for generating Huntington disease model marmosets.
Turning genes into medicines: Challenges in gene therapy for genetic disease
Katherine A High, MD, Spark Therapeutics
The delivery of therapeutic nucleic acid sequences by viral or non-viral vectors has been a goal of twenty- first century medicine, and the pace of regulatory approvals of gene therapies for genetic disease has begun to accelerate. This presentation will review challenges and lessons in the development programs for gene therapies for two different genetic diseases, one a rare form of congenital blindness, and the other hemophilia. For the first, challenges in the clinical development program included an ultra-rare disease population, a dearth of natural history data, and the absence of any licensed products for the disease. A consequence of this latter point was that there were not well-established endpoints for determining efficacy of an investigational agent. A central challenge in clinical trial design for the pivotal study was choice of controls. For a paired organ like the eye, the contralateral eye is, from a scientific standpoint, the ideal control, since it has the same mutation at the same stage of disease progression, but regulatory agencies prefer a trial design that evaluates use of the investigational agent in the same way that it will be used clinically, i.e. bilateral injection. Advantages and disadvantages of the open-label, randomized, controlled, crossover design that was selected will be discussed. Assessment of safety in the setting of a complex delivery procedure will also be discussed.
Hemophilia gene therapy presents a different problem set compared to an inherited retinal dystrophy. The delivery procedure (intravenous infusion of vector targeting the liver) is straightforward and relatively risk-free, and endpoints are well-understood and easily measured. However, in contrast to the relatively immunoprivileged subretinal space, delivery of AAV vectors to the liver via the circulation presents multiple challenges from the human immune system. Many adults carry circulating antibodies to AAV based on prior exposure to wild-type AAV, and these can reduce or eliminate target organ transduction for vector delivered through the circulation. In addition, a dose-dependent, delayed cellular immune response to vector capsid, not well-predicted by animal studies, can lead to elimination of transduced cells if not appropriately managed. The implications of these observations for Huntington’s disease will be discussed.
Characterization of the gene regulatory network of HTT
Richard M Myers, PhD, HudsonAlpha Institute for Biotechnology
A gene regulatory network comprises all of the cis-acting regulatory elements (CREs) such as enhancers and promoters, and the DNA binding proteins, transcription factors (TFs) and others, which in concert control the expression of a gene. We have sought to characterize thoroughly the gene regulatory network of HTT for several reasons. An understanding of the regulation of HTT expression would facilitate not only the interpretation of the results in the field to lower HTT expression through various means, but also the direction of future efforts. Furthermore, by establishing the TFs responsible for the regulation of HTT expression, we hope to find clues to its normal function.
To identify HTT CREs we have used multiple genomic assays in our primary model system, human cultured neurons derived from induced pluripotent stem cells. These include ATAC-seq to measure open chromatin, ChIP-seq to measure TF occupancy and histone post-translational modifications, Next-generation Capture-C to find 3-dimensional chromatin interactions with the HTT promoter, and STARR-seq to assay enhancer activity of millions of DNA test elements. From the analysis of these data we nominated several regions as enhancers of HTT and tested the effects of their loss-of-function on HTT expression through targeted dCas9- KRAB constructs, a technique established to be capable of CRE activity ablation. For several of the putative HTT CREs, ablation resulted in a significant decrease in HTT expression, further establishing their role in the HTT gene regulatory network.
From these new HTT enhancers and the well-established promoter, we looked for known TF motifs within “TF footprints” revealed by ATAC-seq data. By comparing these motifs with existing ChIP-seq data (non-neuronal sources) and filtering for mRNA expression in cultured neurons, we identified a set of TFs potentially binding to HTT CREs and thus perhaps regulating HTT expression. By using ChIP-seq with these TFs in cultured neurons, we confirmed occupancy at the HTT promoter and at a subset of CREs for several candidates. Although knockdown of mRNA of these TFs through multiple means did not generate significant reductions in HTT expression, we remain confident that many of these TFs will eventually confirm to be regulators of HTT.
The incomplete splicing of huntingtin
Gill Bates, PhD, Queen Square Institute of Neurology, UCL
We have shown that exon 1 of the huntingtin (HTT) gene does not always splice to exon 2, generating a small polyadenylated transcript (Httexon1) that is translated to produce the highly pathogenic exon 1 huntingtin (HTT) protein. The cryptic polyadenylation sites in intron 1 that are activated are situated at 680 bp and 1.2 kb into mouse intron 1, and 2.8 and 7.3 kb into human intron 1. The level of incomplete splicing increases with increasing CAG repeat length and occurs in all knock-in mouse models of HD, YAC128 mice and in HD postmortem brains and fibroblast lines. We have interrogated the mechanism causing incomplete splicing using HEK293 cells expressing a mouse mini-gene Htt construct. We defined the intron 1 sequences required for incomplete splicing, showed that this could be modulated by altering the levels of the splicing factor SRSF6, that a precise spatial relationship of the CAG repeat to the 5’ splice site and intron is required, and that slower PolII transcription rates favor the production of the small Httexon1 transcript. We have converted the minigene system into an assay that is being used to screen for small molecules to enhance splicing.
We are using a combination of genetics and Htt lowering agents to determine the extent to which the exon 1 HTT protein contributes to the pathogenic process. We have found that incomplete splicing can be readily detected in mouse embryonic fibroblasts (MEFs) from zQ175 and YAC128 mice and are using these cell lines to screen HTT lowering agents that target both the Httexon1 and full-length Htt transcripts before assessment in vivo. For mouse Htt, we have established a 14-plex quantigene assay that probes all of the well-defined Htt transcripts and can be used both in vivo, and for the zQ175 MEFs in a 96-well format. We have previously shown that MW8 is a neo-epitope antibody for the C-terminus of the exon 1 HTT protein, have used this to establish a TR-FRET assay that is specific to exon 1 HTT, and can be used to quantify changes in exon 1 HTT protein levels.
Huntingtin post-translational modifications and enzymes as potential therapeutic targets
Christopher A Ross, MD, PhD, Johns Hopkins University School of Medicine
Now, over 25 years since the identification of the HD gene, potential disease-modifying therapies are coming to the clinic. While this represents a triumph of rational drug design, the challenges of delivering large-molecule HTT-lowering agents for a lifetime suggest the value of continued search for small-molecule alternatives. The HTT protein itself has generally not been thought to be a good therapeutic target. However post-translational modifications (PTMs) are usually catalyzed by enzymes, which can be modulated, and thus may be therapeutic targets. Most previously identified PTMs appear to be protective. We have conducted a systematic screen of the HTT protein to find additional sites of post-translational modification, using human postmortem brains and human iPSC-derived HD cells, for quantitative proteomics. We have also conducted functional experiments in cell culture, to identify modifications which enhance toxicity – so that enzyme inhibition would reduce toxicity. We have developed cell toxicity assays in primary neurons and in immortalized striatal precursors to determine the effect of altering the relevant amino acids (with the proviso that serine to glutamate or aspartate changes may not be effective phosphomimetics) or using small- molecule inhibitors. We have identified five clusters of PTMs (phosphorylation, acetylation or methylation so far) within predicted unstructured regions of HTT, consistent with the cryo-EM structure of HTT, in which the PTMs are in unresolved, presumably unstructured loops. Initial cell data suggest the PTMs facilitate toxicity. PTMs may also modify HTT localization and response to cell stress. The modifications which cluster together appear to be capable of influencing each other with a form of local crosstalk. Knowledge of the structure of HTT may also enable the identification of long-distance crosstalk. We have used in vitro screens of synthetic peptides in order to identify candidate kinases or other modifying enzymes, focusing initially on serine 116/120. Inhibitors of these candidate kinases can reduce mutant HTT cell toxicity. Additional experiments will be required to determine whether the effects are dependent on modulation of direct modification of HTT, or on other cellular actions, and to determine specificity of the enzymes and their inhibitors. However, these early and preliminary results appear consistent with the possibility of using HTT PTMs and modifying enzymes as targets for developing small-molecule therapeutics for HD.
Systematic interaction mapping to define the quantitative huntingtin interactome
Erich E Wanker, PhD, Max Delbrueck Center for Molecular Medicine
Proteins in cells do not function in isolation. Rather, they associate with varying numbers of partner proteins to perform their specific cellular tasks. Huntingtin (HTT) is a large protein with many functional associations and is often considered a hub protein. Using genetic and biochemical methods, several hundred putative HTT-associated proteins have been identified. How many of them indeed interact with HTT in neuronal cells under physiological conditions remains largely unknown. On a more detailed level, it needs to be clarified to which domains in HTT partner proteins bind, how strong the association is and whether it is influenced by the pathogenic polyglutamine tract in HTT. We have systematically reviewed the interactions of 20 HTT fragments and 37 previous HTT interactors using an automated Y2H system. In total, 115,000 potential PPIs were tested. They were validated using our recent mammalian cell-based Luminescence Two Hybrid technology – LuTHy (Trepte 2018, Mol Syst Biol), which provides two quantitative readouts in one experiment to capture high- and low-affinity binders and information on interaction strength. We mapped interactions to N- and C-terminal domains in HTT and validated 90% of the Y2H HTT interactions. The results not only confirm functional links of HTT to autophagy and vesicle transport processes and their disturbance in disease, for example, but also reveal many new associations among the HTT partner proteins. This suggests that large multi-protein complexes with functional relevance are formed around full-length HTT in vivo.
We provide a comprehensive resource that can be consulted for specific structure-function investigations with experimental and in-silico tools.
Marcy MacDonald, PhD, Massachusetts General Hospital
Genetic modifiers identified in humans can: provide insights into disease mechanisms; suggest therapeutic strategies; and generate guidelines for pre-clinical and clinical studies. Genome-wide association analysis of 9,000 Huntington’s disease (HD) individuals for age at motor onset has now revealed 23 genome-wide significant signals at 14 chromosome locations. A major modifier class (DNA maintenance genes) and unexpected ostensible signals at HTT reveal that it is the size of the uninterrupted CAG repeat, not the size of the polyglutamine tract in huntingtin, that is the primary determinant of the rate of the pathogenic process that precedes overt motor signs. Moreover, the observation of a non-DNA handling disease modifier class spurs the ongoing search for genetic modifiers of additional HD stages of interest for clinical studies, particularly the interval that spans the emergence of overt disease symptoms.
Genetic overlap between neuropsychiatric disorders and psychiatric symptoms in HD
Peter Holmans, PhD, Cardiff University
Background/motivation: Recent genome-wide association studies (GWAS) have identified genetic modifiers of age at motor onset (AMO) and disease progression in HD, along with a plausible biological hypothesis (DNA repair). Such promising results motivate the genetic study of other phenotypes in HD. Psychiatric symptoms are known to be more frequent in HD compared to the general population, and many psychiatric disorders have a genetic component. We hypothesise that psychiatric symptoms in HD also have a genetic component, and that this is shared with psychiatric disorders.
Materials and Methods: Psychiatric phenotypes in HD were defined using the HD clinical characteristics questionnaire (HDCC). This provides binary diagnoses of psychosis, depression, irritability, violence/ aggression, perseverative and obsessive behaviour, apathy and cognitive impairment. GWAS summary statistics for schizophrenia, bipolar disorder, ADHD, major depression (MDD), obsessive-compulsive disorder (OCD) and autism were obtained from the Psychiatric Genomics Consortium and used to construct measures of genetic risk via the polygenic risk score (PRS) method in 5160 unrelated HD individuals from REGISTRY and Enroll-HD. Additionally, risk scores were constructed for the neurological disorders Alzheimer’s disease, Parkinson’s disease and amyotrophic lateral sclerosis (ALS). Association between PRS and phenotype was carried out via logistic regression, correcting for CAG length and ethnicity.
Results: Schizophrenia polygenic risk score was associated with increased risk of psychosis, and also increased risk of irritability, depression and violence/aggression. Further significant associations were observed for depression with MDD and bipolar PRS and between violence/aggression and ADHD. A nominally significant association was observed between OCD and perseverative and obsessive behaviour. Notably, there were no significant associations between any of the neurological disorders and the HD symptoms. Interestingly, apathy and cognitive decline showed no association with any of the psychiatric PRS but were associated with PRS for HD progression.
Conclusions: Psychiatric symptoms in HD show genetic overlap with neuropsychiatric disorders. This may give insights into the biological processes governing the variability of phenotypic expression in HD. Further genetic study of such symptoms is therefore warranted, with more refined definitions.
Human genetic modifiers of HD age of onset reveal DNA repair and handling genes as candidate therapeutic targets
Hilary Wilkinson, PhD, CHDI
The convergence of human genetic data implicating DNA repair and handling genes with experimental model research describing the influence of DNA repair genes on the incidence of CAG repeat expansion has prompted the formation of the DNA Repair & Handling Major Focus Area at CHDI. DNA repair pathways play an important role in the maintenance of both nuclear and mitochondrial DNA integrity in neurons and other cell types. Our current goal is to investigate whether interventions in the DNA repair pathways are integral to arresting HTT somatic instability and provide therapeutic benefit. We have selected MutSβ as an initial therapeutic target. We are advancing reagents and assays to perform drug screens while in parallel pursuing mechanistic studies on mismatch repair protein machines and leveraging the Enroll-HD platform to provide information to support prosecution of clinical trials.
Accelerating the path from discovery of therapeutic targets to effective and safe medications: Human genomics to the rescue
Alan R Shuldiner, MD, Regeneron Genetics Center
Advances in high-throughput DNA sequencing technologies now provide an opportunity to leverage human genetic “experiments of nature” to discover, validate and derisk novel drug targets, identify new indications for existing drug targets, and develop biomarkers to select patient subpopulations more likely to respond or less likely to experience adverse effects. The Regeneron Genetics Center works with more than 60 academic and health care system collaborators world-wide and has sequenced >500,000 samples from large patient populations linked to rich electronic health record (EHR) data, disease-enriched cohorts, founder populations, undiagnosed Mendelian diseases, and clinical trial participants. Exome sequencing and genome-wide approaches at such scale provide the opportunity to agnostically query the genome for rare loss-of-function or other variants that enable truly novel genetic discoveries that inform human biology. Recent examples include discovery of HSD17B13 and B4GALT1 as novel drug targets for chronic liver disease and cardiovascular disease, respectively; validation of angiopoietin-line proteins (ANGPTLs) for hypertriglyceridemia, and identification of type 2 diabetes as a potential new indication for ANGPTL4 antagonism. Founder populations enriched for the same pathogenic variant provide opportunities to identify genetic modifiers of disease onset or severity as potential novel targets. This lecture will review some of the most fruitful approaches, providing specific examples of human genome-informed target discovery and validation with the ultimate goal to improve health and quality of life for patients with diseases of unmet clinical need.