CHDI FOUNDATIONCHDI

Accelerating Therapeutic Development for Huntington's Disease

2018 Conference

CHDI’s 13th Annual HD Therapeutics Conference took place February 26 – March 1, 2018, 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.

The motor network in Huntington’s disease: What brain rhythms mediate specific motor signs?

Philip A. Starr, MD, PhD, University of California, San Francisco

The hyperkinetic features of Huntington’s disease were originally modelled as an imbalance in the relative strength of the direct and indirect intrinsic basal ganglia pathways, but this model failed to explain critical clinical observations. More recently, hypokinetic and hyperkinetic disorders have been reconceptualized as arising from excessive synchronized oscillatory activity in the basal ganglia thalamocortical loop, in phenotype-specific frequency bands. Mouse models of HD suggest abnormalities in oscillatory synchrony in low gamma (30-40 Hz) bands. Human brain recordings that support the role of oscillatory activity in movement disorders have largely been performed in Parkinson’s disease and the isolated dystonias, but it is likely that changes in oscillatory synchronization are also relevant to other movement disorders including HD. This presentation will review the foundations of the “oscillation model” of movement disorders and suggest testable hypothesis about the circuit level origins of choreiform and bradykinetic manifestations of HD.

Top

Why the basal ganglia are essential for movement

Presentation not currently available

Henry Yin, PhD, Duke University

The basal ganglia have been implicated in movement for decades, yet there is no consensus on how these circuits contribute to motor control. I will discuss recent experiments my lab using wireless recording and motion capture in mice. Our results provide insights into the computational functions of the basal ganglia circuitry and challenge traditional assumptions in the field. I will also describe a new model of the basal ganglia proposed to explain the new findings. According to this model, the basal ganglia implement closed loop control the rate of change in various perceptual transitions. Perhaps the simplest example of this type of control is the control of movement velocity using proprioceptive inputs. I will describe results that demonstrate for the first time how the basal ganglia circuits enable velocity control and voluntary behavior, and how activity in these circuits are altered in zQ175 knock-in mice. These results shed light on a rich variety of clinical symptoms in movement disorders and suggest new therapeutic strategies.

Top

Cells that tile your brain: Astrocyte functions in the striatum

Presentation not currently available

Baljit S. Khakh, PhD, David Geffen School of Medicine, University of California, Los Angeles

Astrocytes tile the central nervous system, but their functions in neural microcircuits in vivo and their roles in mammalian behavior remain largely undefined. We will summarise our recent data regarding astrocyte dysfunctions in Huntington’s disease (HD) mouse models using physiological and molecular evaluations (imaging, electrophysiology and RNA-Seq). We will then focus on exploring how striatal astrocyte calcium signaling (and its reduction) contributes to striatal physiology. We will report data from the use of 2-photon laser scanning microscopy (2PLSM), electrophysiology, head mounted microscopes and a new genetic approach to reduce striatal astrocyte calcium signaling in vivo. In wild type mice, we found that reducing striatal astrocyte calcium signaling increased self-grooming behaviors by modulating medium spiny neuron (MSN) activity. The mechanism involved astrocyte-mediated neuromodulation mediated by ambient GABA and was corrected by normalizing astrocyte regulation of GABA. Furthermore, in a mouse model of Huntington’s disease, dysregulation of ambient GABA and astrocyte calcium signaling accompanied abnormal repetitive self-grooming, which was also relieved by normalizing astrocyte regulation of GABA. Thus, astrocyte-regulated neuromodulation contributes to innate behavior and to obsessive-compulsive-like phenotypes, revealing astrocytes as potential targets for early psychiatric phases of Huntington’s disease symptoms.

Top

PDE10i and HD: Learnings from the Amaryllis studies

Marielle Delnomdedieu, PhD, Pfizer Worldwide R&D

Degeneration of striatal medium spiny neurons underlies corticostriatal dysfunction in Huntington’s disease (HD). Inhibition of Phosphodiesterase-10A, an intracellular signaling regulator expressed in medium spiny neurons, increases corticostriatal function in preclinical models, and represented a promising therapeutic approach in HD.

The Amaryllis Clinical Trials were designed to assess safety, tolerability and the efficacy of 26-week of an oral daily dose with a PDE10A inhibitor (PF-02545920) on motor function in HD patients (A8241021, NCT02197130), as well as obtain safety and exploratory efficacy data in a 12-month open label extension study (A8241022, NCT02342548). The studies were initiated in 5 countries and 43 sites.

The 6-mth, double-blind, placebo-controlled, randomized trial was performed in patients with genetically confirmed symptomatic HD with BID dosing of 5mg or 20mg of PF-02545920, or placebo, with UHDRS-TMS as primary endpoint assessing motor function. Standard HD scales also assessed cognition and behavior. 216 (79%) out of 272 randomized subjects completed the trial with high dosing compliance. PDE10 was generally safe and sufficiently tolerated: adverse events were generally mild or moderate and occurred more frequently at 20mg PF-02545920 (90%) compared to 5mg PF-02545920 (86%) and placebo (72%). The most common observed AEs for the 20 mg group were somnolence, fatigue, weight decrease, anxiety, nausea, insomnia, dizziness and dyskinesia. Analysis of the primary endpoint (UHDRS-TMS) did not show a statistically significant change from baseline compared to placebo after 26 weeks of dosing with either PF02545920 5 mg or 20 mg BID. Based on the Amaryllis study not meeting pre-defined efficacy criteria for improvement of motor symptoms in HD as measured by TMS, decision was made to terminate the open-label extension study and the overall PDE10 HD program was terminated. In general, results from many of the secondary assessments were not significantly different from baseline and no significant changes were observed in standard motor clinical efficacy measures. Interestingly, the Q-Motor exploratory endpoints showed the most consistent and statistically significant changes (improvement) with PF-02545920 compared to placebo.
A summary of results from both studies and up to 18-mth of treatment with the PDE10i compound will be presented.

Top

Using stem cells to model and treat Huntington’s disease

Clive Svendsen, PhD, Cedars-Sinai Regenerative Medicine Institute

Regenerative Medicine provides a platform to both model and treat HD. For modeling we have been making induced pluripotent stem cells from patient fibroblast and blood samples and differentiating them into striatal neurons. This system allows us to model the earliest stages of HD develop in the dish and we have discovered significant developmental phenotypes that point to an early start in the pathology of HD. We have also used stem cells either alone or modified to release a powerful growth factor GDNF in a number of pre-clinical models of HD. The transplant produces mainly astrocytes which migrate through the striatum and will secrete GDNF for long time periods. This results in amelioration of pathology and behavioral changes in models of HD. The same cells are now being used in a Phase 2a clinical trial for ALS where 10 patients have been treated and are currently being followed for safety and efficacy. Whilegenetic treatments are obviously the number one interest right now for patients, stem cell and gene therapy approaches may still hold promise for patients at later stages of the disease where cells have already been lost.

Top

Unbiased phenotypic characterization of CAG-length variation in ESC-derived cortical neurons

Bruno Chilian, PhD, Evotec AG

Human embryonic stem cells (hESC) are uniquely suited for modeling genetically-driven neurodegenerative diseases like HD in a cellular context. The ability of hESC to give rise to essentially unlimited numbers of genetically-matched neurons carrying the mutation associated with HD makes possible a wide range of experimental approaches based on cell phenotype. However, as HD is primarily an adult-onset disease, the embryonic nature of hESC-derived neurons makes it difficult to detect a phenotypic signature in these cells. To address this limitation, we have adapted automated (high-content) microscopy for the analysis of hESC-derived HD neuronal models. This approach produces large, multiparametric data sets which can be analyzed in an unbiased manner to detect the relatively subtle cellular phenotypes resulting from the HD mutation. This strategy extends the value of HD stem cell models by enabling a variety of high-throughput phenotypic screening approaches that are central to therapeutic discovery.

For our study, we adapted and optimized stem cell differentiation protocols to produce cortical neurons from a series of isogenic hESC clones based on the RUES2 parental line (NIHhESC-09-0013).The parental line was modified by CRISPR/Cas9 to increase the CAG repeat size at a single allele of the HTT locus from the parental lengths of 20 and 22 (Ruzo et al, 2018; in press). We produced neurons from three clones, one with CAG20 representing the healthy state, and two with CAG56 for the diseased state. Due to the short cultivation timelines, our model represents immature, embryonic neurons, well before the onset of any clinically relevant pathology.

In order to detect CAG-length dependent phenotypic differences in an unbiased fashion, we applied the Cell Painting methodology (Gustafsdottir et al, 2013): A novel technique that uses generic organelle dyes, automated image analysis and machine learning to distill a set of phenotypic features which describes the changes associated to the perturbagen – in this case CAG-length variation. The method is sensitive enough to detect subtle phenotypic patterns, which is expected to be necessary in our early-stage HD model.

Initial results show clear phenotypic signature differences between the investigated clones. This holds true across different supervised and unsupervised analysis methodologies.

The next step will be the extension of the study to encompass additional independent clones covering a wider range of CAG-lengths. The goal is to determine, whether a consistent phenotypic pattern exists, which can be linked to CAG length variation.

This would have a twofold benefit: (i) Deepening of our understanding of HD pathogenesis with the potential for novel treatment paradigms. (ii) The possibility of a “phenotypic reversal” screen for molecules that revert the high-CAG signature back to the normal-CAG phenotype.

Top

Transplantation of human pluripotent stem cells-derived telencephalic progenitors to treat and model HD

Josep M. Canals, PhD, IDIBAPS-University of Barcelona

Human pluripotent stem cells (hPSCs) offer great advantages for modeling and treating Huntington’s disease (HD). In recent years, we have focused our research on hPSC differentiation and transplantation to develop therapeutic strategies for HD.

Our goal is for hPSC differentiation to recapitulate striatal development. We have characterized the role of selected striatal specific transcription factors (TFs) during development and use this information to enhance in vitro striatal neuron differentiation. We have recently contributed to the development of a hPSC differentiation protocol which produces functional neurons in 37 days in vitro (DIV). At 16DIV telencephalic progenitors are generated including a large population that express a range of striatal specific TFs such as Dlx, Gsh1/2, Ascl1 and EBF1. At 37DIV almost all cells express MAP2. Of these more than 50% express Ctip2, with a small subpopulation co-expressing DARPP-32. Mature neurons show spine-like structures by electronic microscopy, and express synaptic proteins and neuronal ion channels. These observations correlate with the finding that more than 80% of neurons display spontaneous firing activity.

To investigate if hPSC-derived telencephalic progenitors are useful for regenerative medicine, we transplanted 16DIV progenitors into mouse brains. Transplanted healthy and HD-derived progenitors integrate into host brains, differentiate into neuronal subpopulations, send projections to the afferent areas and establish synaptic connections. Interestingly, HD-derived progenitor transplantation reproduces many HD-related phenotypes from 3 months post-transplantation onwards. Cell death and autophagy induction are observed in both the transplanted HD cells and the host brain, which in turn alters behavior.

Many transplanted cells express Ctip2, including a subpopulation that co-expresses DARPP32 which indicates a medium spiny neuron (MSN) identity. However, other neuronal populations are also present. To increase the MSN percentage we are analyzing the different progenitor populations present at 16DIV by single-cell RNA-seq. We have identified different subpopulations and specific cell surface markers in some cases. We are currently analyzing whether these markers can be used to select specific subpopulations for transplantation to increase MSN percentage post-transplantation.

In conclusion, we show that hPSC differentiation and transplantation is a viable strategy for regenerative medicine approaches, and also a useful tool for modeling human HD in vivo.

Top

Human neural stem cell transplantation rescues functional deficits in HD mice

Leslie M. Thompson, PhD, University of California, Irvine

Effective neurorestorative or neuroregenerative strategies based on human stem cells has growing support and offers a possible therapeutic approach that could modulate pathology in a complex tissue such as the brain. Initial evidence of potential efficacy in humans arose from human fetal-cell transplants that produced some cognitive and motor benefit in a subset of HD patients. Progress has also been made using stem cell transplants to treat HD animal models, most recently with human MSCs genetically engineered to overexpress BDNF in YAC128 and R6/2 mice, as well as use of astrocyte progenitors in R6/2 mice. We have evaluated the potential efficacy of intrastriatal injection of human embryonic stem cell derived neural stem cells (hNSCs) in R6/2, Q140 and zQ175 HD mouse models. Specifically, we focused our studies on outcomes using GMP grade hNSC cells, ESI-017 hNSCs. In HD fragment model R6/2 mice, transplants improve motor deficits, rescue synaptic alterations and appear to make synaptic contacts with mouse cells. Further, implanted hNSCs are electrophysiologically active. hNSCs also improved motor and late stage cognitive impairment in a second HD model, Q140 knock-in mice. Disease modifying activity is suggested by reduction of aberrant accumulation of mutant HTT protein and expression of BDNF in both models. Finally, we have carried out a long term study in zQ175 mice and observe long term cell survival and differentiation into neuronal cell types, improvement in the cell membrane properties of zQ175 MSNs, improvements in running wheel and evidence of synaptic contact with host mouse cells by electron microscopy. These findings hold promise for future development of stem-cell based therapies.

Top

Safety and efficacy of human embryonic stem cell-derived oligodendrocyte progenitor cells in subjects with subacute cervical spinal cord injury

Presentation not currently available

Jane S. Lebkowski, PhD, Asterias Biotherapeutics, Inc

AST-OPC1 is a cryopreserved population of cells derived from human embryonic stem cells which contains oligodendrocyte progenitor cells and other characterized cell types. The initial safety of AST-OPC1 was evaluated in a phase 1 clinical trial that enrolled 5 patients with neurologically complete T3-T11 thoracic spinal cord injury (SCI). Based on favorable 5 year safety data, an open-label phase 1/2a clinical trial (SCiStar) was executed to evaluate the safety and activity of escalating doses of AST-OPC1 in patients with severe (AIS-A or AIS-B) C4-C7 SCI. Doses of 2×106, 1×107 and 2×107 AST-OPC1 are being tested in 5 cohorts of patients. Subjects are followed by the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) neurological exam and other assessments including MRI to determine safety and activity. Enrollment is completed and patients are being followed. Cohorts 1 (AIS-A, 2×106 cells) and 2 (AIS-A, 1×107 cells) have been followed for 1 year, while 1 year follow-up is pending for subjects in cohorts 3 (AIS-A, 2×107 cells), 4 (AIS-B, 1×107 cells) and 5 (AIS-B, 2×107 cells). The results to date from patients in cohorts 1 and 2 indicate no safety issues associated with the delivery of AST-OPC1 or the cells themselves. All patients in cohorts 1 and 2 have experienced improved upper extremity motor function as assessed by upper extremity motor scores (UEMS) and neurological level of motor function. The average UEMS improvement for the 3 patients in cohort 1 was 6.3 points at 12 months, with each patient demonstrating one neurological level improvement in motor function. The 6 patients in cohort 2 have completed 12 months of follow-up, recovered an average 12.3 motor points. Four of 6 subjects in cohort 2 achieved a ≥2 motor level neurological improvement by one year of follow-up. The motor function improvement in patients in cohort 2 compares favorably with a closely matched historical group of 62 traumatic SCI patients derived from the EMSCI (European Multicenter Study about Spinal Cord Injury) database of over 3300 patients. The data to date demonstrate the safety of AST-OPC1 and provide encouraging early evidence of improvements in arm, hand and finger function.

Top

Genetic modification of HD

Jong-Min Lee, PhD, Massachusetts General Hospital & Gem-HD Consortium

Modifiers of disease change the timing and/or severity of pathogenesis, leading to an alteration of various disease phenotypes. In HD, there is evidence for modification, resulting from a genetic interaction between the presence of an expanded HTT CAG repeat and variations in other genes. Identification of these genetic modifiers can provide both insight into disease cascades and guidance for development and testing of HD therapeutics. Although age at HD onset is a phenotype highly correlated with the size of CAG repeat expansion, there is substantial variance in age at onset that remains unexplained by the mutation. The remaining variance provides a phenotype that can be used to capture genetic modifiers altering the phase of HD pathogenesis leading to disease onset. Using age at onset corrected for CAG repeat size as the phenotype, we performed genome-wide association study to identify genetic modifiers of HD. Single SNP analysis of ~9,000 HD subjects confirmed previously discovered onset modifier signals on chromosome 15, 8, and 3. Additional genome-wide significant modification signals were also detected in other genomic locations involving 1) HTT with a special repeat structure, 2) DNA repair genes, and 3) genes involved in other pathways. Together, our genetic analyses based on observations in humans support modification of HD through genetic interaction between the expanded CAG repeat and DNA repair, as well as other pathways, providing potential avenues for therapeutic development and a clear demonstration of the power of the genetic approach that is now being applied to other aspects of the disease.

Top

Trinucleotide repeat instability: The dark side of DNA mismatch repair

Guo-Min Li, PhD, University of Texas Southwestern Medical Center

DNA mismatch repair (MMR) is a critical genome maintenance system whose primary role is to ensure replication fidelity by correcting both base-base mismatches and insertion-deletion mispairs generated during DNA replication. Loss of the MMR function leads to genome instability and cancer development. Key MMR proteins in human cells have been identified, which include the base-base mismatch recognition protein MutSβ(MSH2-MSH6 heterodimer), the insertion-deletion mispair recognition protein MutSβ(MSH2-MSH3 heterodimer), endonuclease MutLα(MLH1-PMS2 heterodimer), exonuclease 1, single-strand DNA-binding protein RPA, proliferating cellular nuclear antigen (PCNA), replication factor C (RFC), and DNA polymerase δ. While MMR is well known for its role in maintaining genome stability, its components, particularly MutSβ, have been implicated in promoting the instability of trinucleotide repeats, including (CAG) repeat expansions associated with Huntington’s disease. However, the mechanism by which MutSβ promotes (CAG) repeat expansions remains unknown. In this presentation, we show our recent data supporting a model for error-prone processing of (CAG)n or (CTG)n sequences by MutSβ and its interacting proteins. During DNA replication and/or repair, a DNA hairpin structure can be readily formed within (CAG) or (CTG) repeats at the 3’ end of the primer strand via strand slippage, and MutSβ recognizes the (CAG)n or (CTG)n hairpins formed in the nascent DNA strand. The binding of MutSβ to the hairpin structure recruits DNA polymerase β to the complex through physical interactions between these two proteins. Polymerase β is then capable of using the hairpin as a primer to add several nucleotides to the 3′-end of the hairpin, which stabilizes/fixes the hairpin structure in the nascent DNA strand. DNA polymerase δ is then recruited to the site for the high-fidelity and highly processive DNA synthesis via polymerase switch. Our studies therefore provide a novel mechanism for trinucleotide repeat expansions in both dividing and non-dividing cells.

Top

Structure and mechanism of human DNA mismatch repair machines: Potential targets for therapeutic intervention

Lorena S. Beese, PhD, Duke University School of Medicine

The DNA mismatch repair (MMR) pathway recognizes and corrects defects that arise during DNA replication and genetic recombination. However, in addition to maintaining genomic stability, several MMR proteins are associated with triplet repeat expansions underlying Huntington’s and several other neurodegenerative diseases. Here we will discuss the complex molecular mechanisms of the MMR machinery with a particular focus on the assemblies involved in the lesion recognition and excision stages. MutSa (MSH2-MSH6) and MutSb (MSH2-MSH3) heterodimers are the main MMR lesion sensors that recognize mispairs, insertion- deletion loops (IDLs) or chemical damage. These two proteins share the MSH2 subunit but exhibit distinct, partially overlapping substrate specificities, encoded by the MSH3 and MSH6 subunits. Of note, MutSβ recognizes triplet repeats and is directly implicated in pathogenic expansions underlying Huntington’s disease. Upon lesion recognition, Exonuclease I is recruited to excise the lesion. A detailed understanding of these processes may provide a foundation for the development of therapeutic strategies that disrupt triplet repeat expansions.

Top

Genome stability and brain health

Peter J. McKinnon, PhD, St Jude Children’s Research Hospital

Genome stability is critically important for human health. This is apparent from the myriad of inherited human syndromes characterized by defective DNA damage responses. The nervous system is particularly prone to the consequences of genome damage, which can lead to neurodegeneration, neurodevelopmental disorders or cancer. Defects in genome maintenance are also increasingly being linked to broader neurologic health issues, including age-related neurodegenerative events that mar cognitive ability and quality of life. Therefore, understanding the mechanistic connections between faulty DNA damage signaling and human disease is of fundamental biomedical importance. However, the underlying basis for pathology in these syndromes remains mostly unknown. Despite the compelling links between compromised genome stability and neurologic disease, scant data are available that adequately illuminate the disease-specific etiology and in vivo processes that maintain the neural genome. DNA repair is important during neurogenesis when the rapid proliferation of progenitors requires a stringent genomic maintenance program to ensure transmission of genetic fidelity. Postmitotic neural populations also require an efficient DNA damage response to address the frequent DNA lesions arising from events such as oxidative stress or transcription-associated DNA damage. I will discuss data from novel mouse models that connect DNA damage signaling to neural homeostasis. Understanding how DNA damage signaling pathways function in the nervous system is essential for illuminating fundamental processes that regulate brain function and prevent disease.

Top

Polyglutamine expansion in huntingtin impairs transcription-coupled DNA repair by inactivating DNA strand break repair enzyme PNKP

Partha S. Sarkar, PhD, University of Texas Medical Branch

The pathological mutation in Huntington’s disease (HD) is the expansion of a polyglutamine (polyQ) tract at the N-terminus of huntingtin due to the expansion of a CAG triplet repeat in the mutant huntingtin (mHTT) gene, which leads to progressive deterioration of cognitive and motor functions. However, the mechanism by which expanded polyQ sequences in mHTT cause neuronal dysfunction and neurodegeneration remains unknown. We report that wild-type (WT) HTT forms a multiprotein transcription-coupled DNA repair (TCR) complex with DNA strand break repair enzyme polynucleotide kinase (PNKP), deubiquitinating enzyme ataxin 3 (ATXN3), RNA polymerase II subunit A (POLR2A), cyclic AMP response element-binding (CREB) protein (CBP), and several transcription factors. This novel TCR complex identifies lesions in the template DNA strand and mediates their repair during transcriptional elongation. By contrast, the polyQ expansion in mHTT depletes PNKP and ATXN3 activities, disrupting the functional integrity of the TCR complex, adversely impacting both TCR and transcription. Low PNKP activity leads to impaired DNA repair and persistent accumulation of DNA lesions, predominantly in actively transcribing genes, resulting in aberrant activation of ATM-dependent pro-apoptotic p53 signaling. Increasing PNKP activity in the mutant cells substantially decreased DNA strand breaks, blocked aberrant ATM→p53 pathway activation, and dramatically improved cell survival. Moreover, our findings demonstrate that mHTT-mediated abrogation of ATXN3 activity increases ubiquitination and degradation of CBP, negatively influencing CREB-dependent transcription. Collectively, these findings provide new insights into how mHTT-mediated abrogation of PNKP and ATXN3 activities and impaired TCR disrupt transcription and simultaneously induce DNA strand breaks to activate the pro- apoptotic DDR-ATM→p53 signaling to trigger neurotoxicity. Our results also suggest potential targets for developing therapeutic modalities to combat neurodegeneration and neuronal dysfunction in HD.

Top

Cell types, circuits and CNS complexity

Nathaniel Heintz, PhD, Howard Hughes Medical Institute & The Rockefeller University

The mammalian brain is composed of over five hundred discrete cell types. One important property of individual cells of a type is a shared molecular and epigenetic “ground state” that dictates their fine-tuned functions, including their responses to external events and internal physiological cues. One hypothesis that arises from these findings is that the impacts of aging, the penetrance and effects of genetic lesions, and the consequences of external influences and behaviors are experienced by each cell type differently. This presentation will focus on the characterization of cell types and circuits that are impacted in human psychiatric and neurodegenerative disease. It will include discussion of the rational design of cell type specific symptomatic and disease modifying therapies for these devastating disorders.

Top

Huntingtin regulates cortical development: Consequences for Huntington’s disease

Sandrine Humbert, PhD, Université Grenoble Alpes

The bulk of interest in the HTT protein has centered on the fact that, when mutated, HTT causes Huntington’s disease (HD), a devastating neurological disorder. The mutation causing HD is an abnormal polyglutamine stretch in HTT. Given the adult onset and dysfunction and death of adult neurons characterizing HD, most studies have focused on the toxic effects elicited by mutant HTT in post-mitotic neurons. However, the protein is ubiquitous and expressed in the developing embryo where it plays an essential role as revealed by the early embryonic lethality at day 7.5 of the complete knockout of the HTT gene in mouse. Anyway, the roles of the wild-type protein during development have been overlooked. We will discuss how HTT regulates several steps of mouse embryonic corticogenesis. HTT is important to maintain the pool of cycling progenitors and also regulates the polarization of newly generated neurons. We will describe the underlying molecular mechanisms by which HTT mediates its effects. Finally, we will also show the consequences of the presence of an abnormal polyglutamine expansion in HTT during cortical neurogenesis and consider the viewing of HD as a developmental disorder.

Top

Post-translational modifications of huntingtin: Approaches towards understanding their therapeutic potential

Andrea Caricasole, PhD, IRBM Science Park

Post translational modifications (PTMs) are diverse and important modifiers of protein function used by organisms to reversibly and homeostatically mediate responses to change. In the case of the huntingtin protein, several PTMs (of which S/T phosphorylations are notable examples) have been identified and a few have been studied in biological systems, mostly through the introduction of mutations which either abolish or mimick their presence at specific positions in the protein. In particular, genetic analyses in mice have highlighted the potential of phosphorylation in the N-terminal part of the protein (pS13/pS16, pS421) for the modulation of mutant huntingtin in vivo toxicity, and suggest their modulation may have therapeutic potential. However, the capacity to quantitatively measure huntingtin PTMs to examine their potential to ameliorate the consequences of the HD mutation and the possibility to modulate their levels through pharmacological means, in order to examine their potential to ameliorate the consequences of the HD mutation in cells and animals, remain challenging. An overview of some of the current efforts to address these challenges will be presented, with specific reference to phosphorylation of residues T3, S13 and S421.

Top

Uncovering the biophysical complexities of Httex1 structure, aggregation, and phase behavior

Rohit V. Pappu, PhD, Washington University in St. Louis

The reigning hypothesis is that monomeric forms of Httex1-encoded protein region undergo a toxic, gain- of-function conformational change as the polyQ length crosses the pathological threshold. We have tested this hypothesis in collaboration with the laboratories of Hilal Lashuel, Edward Lemke, and Danny Hatters. Our investigations, which combine data from single molecule FRET measurements, hydrogen-deuterium exchange mass spectrometry and NMR investigations, and atomistic computer simulations yield an unprecedented atomic-level structural description of monomeric Httex1 for a series of polyQ lengths that lie below and above the pathological threshold. The overall topology resembles that of a tadpole, with a globular head comprising of N17 adsorbed on a compact polyQ domain, and a semi-flexible proline-rich C-terminal tail. This tadpole-like topology is preserved for all polyQ lengths, although the surface area of the head increases continuously due increased prominence of the globular polyQ domain. Contrary to a sharp conformational transition at the monomer, we show that saturation concentrations that define distinct phase boundaries on the aggregation landscape of Httex1 decrease non-linearly as polyQ length increases. Similarly, proteomics analysis shows clear evidence for a gain-of-function increase in protein-protein interactions, whereby new interactions with proteins like FUS are engendered by the growing prominence of the polyQ globule within Httex1. Using the framework of phase separation, we show how and why specific protein- protein interactions, such as those with profilin, help modulate Httex1 phase behavior. Preferential binding of profilin to monomers and soluble oligomers weakens the driving forces for aggregation and phase separation through a mechanism known as polyphasic linkage. Our findings suggest that routes to therapeutic intervention are likely to involve targeting monomers and oligomers in the soluble pool and enhancing or mimicking the abilities of proteins like profilin by shifting phase boundaries within cells thus sub-saturating the concentrations of aggregation-prone Httex1.

Top

Age-dependent function of huntingtin with therapeutic implication

Xiao-Jiang Li, PhD, Emory University

Huntingtin (HTT) is important during animal early development, as germline deletion of HTT leads to early embryonic death of mice. However, it remains unknown whether depletion of HTT in adult animals is lethal or harmful to animal survival. Addressing this issue is important, as considerable efforts have gone into developing allele-specific suppression of mHTT expression. Using inducible HTT knockout mice, we found that HTT depletion does not lead to adult neurodegeneration or animal death at >4 months of age. This finding suggests that depletion of HTT in the adult brain could be a safe therapy. To test this idea, we used CRISPR/Cas9 to permanently deplete the endogenous expression of mutant HTT in the striatum of HD140Q knock-in mice. We found that stereotaxic injection of AAV vector expressing HTT gRNA and CRISPR in the striatum of HD140Q KI mice can effectively deplete HTT aggregates and early neuropathology. The reduction of HTT expression in striatal neuronal cells in the adult HD140Q KI mice does not affect their viability, but alleviates their motor deficits. Our studies show that HTT’s function is age-dependent and removing HTT is non-deleterious so that CRISPR/Cas9-mediated gene editing in a non-allele-specific manner can be used to efficiently and permanently eliminate polyQ expansion-mediated neuronal toxicity in the adult brain.

Top

RNA aggregation and repeat expansion disorders

Ankur Jain, PhD, University of California, San Francisco

Expansions of short GC-rich repeats underlie several neurological and neuromuscular disorders including Huntington disease, muscular dystrophy, and amyotrophic lateral sclerosis. The RNA transcript with repeats accumulates in the nuclei as aberrant “RNA foci” that sequester essential RNA binding proteins. Although RNA foci have been commonly observed in nearly all repeat expansion diseases, the mechanism of foci formation was not known. We identified that RNA foci result from sol-to-gel phase transition of the repeat- containing RNA. The disease-associated repeats form templates for multivalent intermolecular base-pairing interactions between RNA. An increase in the number of repeats corresponds to an increased valency for inter-molecular hybridization. Sol-gel phase transitions exhibit sharp valency dependence, and thus, beyond a critical number of repeats, the RNAs become excessively prone to gelation or aggregation. Compounds that disrupt base-pairing interactions disrupt RNA foci in patient-derived cells and offer a potential route to therapeutics. Analogous to protein aggregation disorders, we propose that sequence-specific gelation of RNAs could be a contributing factor to neurological disease.

Top

Ubiquitin-dependent quality control pathways in Huntington’s disease

Michael Rape, PhD, Howard Hughes Medical Institute & University of California, Berkeley

Ubiquitin-dependent protein degradation plays an essential role in eliminating misfolded and aggregation- prone proteins and thus ensuring cellular homeostasis. Mutations that result in protein misfolding or aggregation, such as polyglutamine extensions in the gene encoding Huntingtin (HTT), or mutations that inactivate ubiquitylation enzymes and effectors have been shown to cause neurodegenerative diseases. Given these observations, improving the ubiquitin-dependent degradation of mutant HTT is thought to be an attractive strategy for treating Huntington’s disease (HD), yet the implementation of this approach has been hampered by our incomplete knowledge of ubiquitylation enzymes and effectors that target mutant HTT for degradation. We recently discovered K11/K48-branched ubiquitin chains as a proteasomal priority signal that allows cells to degrade difficult or highly abundant proteins (Meyer and Rape, Cell 2014). By developing a bispecific antithat can detect endogenous K11/K48-branched ubiquitin chains, we now found that cells employ K11/K48-branched ubiquitin chains to degrade nascent, but misfolded polypeptides to prevent their deleterious aggregation (Yau et al., Cell 2017). We identified UBR4 and UBR5, a candidate modifier for the age of onset of HD, as responsible E3 ligases for K11/K48-branched chain synthesis, and p97/VCP, a disaggregase with known links to HD, as the major effector of these ubiquitin conjugates. Notably, mutant HTT, but not benign HTT variants, is modified with K11/K48-branched chains, recruits all known effectors of K11/K48-specific quality control, and competes with nascent misfolded polypeptides for recognition by K11/ K48-specific quality control. The accumulation of mutant HTT can therefore prevent the elimination of a large class of nascent, misfolded proteins and thus negatively impact multiple cellular pathways. Our results shed light on cellular defense mechanisms against accumulation of misfolded proteins, including HTT, and activation of K11/K48-specific quality control provides an exciting strategy to ameliorate the consequences of mutant HTT expression.

Top

Modeling the effects of regulating mutant or total huntingtin levels in HD mouse models using LacO-modified Htt knock-in alleles

Scott Zeitlin, PhD, University of Virginia

We are using Lac-regulatable HD knock-in mice, Htt(LacO-140Q/+); b-actin-LacIR tg and Htt(LacO-140Q/ LacO-20Q); b -actin-LacIR tg, to study the consequences of reducing either mutant or total Huntingtin expression at different postnatal ages (postnatal day 21, 3, 6, and 9 months of age), and to evaluate the effect of de-repressing mutant Huntingtin expression in older mice. The lactose analog isopropyl b-D-1 thiogalactopyranoside (IPTG) was added to the drinking water (10 mM) of pregnant females beginning at embryonic day 5 to de-repress Huntingtin expression, and then the IPTG was withdrawn at different postnatal ages to moderately repress Huntingtin expression. To study the effects of de-repressing mutant Huntingtin expression in older mice, IPTG was administered to 12-month-old Htt(LacO-140Q/+); b-actin-LacIR tg that had not received prior IPTG treatment.

Behavioral testing of the Htt(LacO-140Q/+); b-actin-LacIR tg mice from 3-18 months of age, revealed that repression of mutant Huntingtin expression at younger ages improved rotarod performance at 6 months of age, improved forelimb grip strength as they aged, and improved Barnes maze performance and prevented weight loss at 17-18 months of age. Repression of mutant Huntingtin expression at younger ages also delayed the accumulation of mutant Huntingtin aggregates. In contrast, de-repressing mutant Huntingtin expression starting at 12 months of age resulted in significantly reduced rotarod performance and forelimb grip strength by 18-22 months of age. Repression of total Huntingtin expression in Htt(LacO-140Q/LacO-20Q); b -actin- LacIR tg mice at younger ages also resulted in improved rotarod performance, improved Barnes maze performance at 17 months of age, and prevented weight loss in comparison to Htt(LacO-140Q/LacO-20Q) controls.

To identify gene expression changes that occurred following either repression or de-repression of Huntingtin expression at 6 months of age, we performed RNAseq using cortex and striatum isolated at 10 days or 30 days following IPTG withdrawal, and at 30 days following IPTG administration. The MAPK pathway in the cortex was the only pathway significantly affected 10 days after IPTG withdrawal but at 30 days, gene expression in multiple pathways was altered in both brain regions. Complementary changes in gene expression were observed in several pathways 30 days following Huntingtin de-repression.

Top

AAV-miRNA mediated HTT lowering as a potential treatment for Huntington’s disease: Dosing and biodistribution studies in mice and rhesus macaques

Presentation not currently available

Jodi L. McBride, PhD, Oregon Health and Science University & Oregon National Primate Research Center

Reducing expression of mutant huntingtin (HTT) in the brain is a promising treatment strategy for the neurodegenerative disorder, Huntington’s disease (HD). Our laboratory, in partnership with colleagues at the Children’s Hospital of Philadelphia, has previously demonstrated that partially reducing HTT expression in the transgenic HD mouse striatum, using a non-allele selective approach, is well tolerated and prevents many of the motor and neuropathological correlates of disease manifestation. In a follow-up study, we showed that a 45% reduction of rhesus macaque HTT in the posterior putamen is well tolerated and does not induce motor dysfunction, cell loss or gliosis. To continue moving our AAV-miRNA platform towards clinical application, we performed a large dosing and biodistribution study using Clearpoint, an FDA-approved, intra-MRI neurosurgical delivery platform. Twelve adult rhesus macaques received vehicle or one of three escalating doses of our lead clinical candidate, AAV2/1-miHDS1, administered over two sites in the putamen. Treated animals showed no aberrant neurological signs nor weight loss throughout the duration of the study. At necropsy brains were collected, numerous tissue punches were collected throughout each putaminal hemisphere and HTT mRNA expression was analyzed by qPCR to create a comprehensive putamen-wide distribution map of knockdown for each group. We found a dose-dependent, significant decrease in rhesus HTT mRNA expression throughout the putamen in all groups, ranging from 22-45%, relative to vehicle- injected controls, with HTT knockdown correlating with miHDS1 miRNA expression. Additionally, we identified doses of AAV2/1-miHDS1 that significantly reduce HTT expression throughout the entire rostral to caudal extent of the putamen. Together, this data further corroborates the tolerability of a partial HTT suppression in the putamen and demonstrates the utility of an intra-MRI neurosurgical platform to accurately target our brain structure of interest as we move AAV2/1-miHDS1 closer to clinical application as a potential treatment for HD.

Top

PRECISION-HD: Phase 1b/2a clinical trials of investigational stereopure antisense oligonucleotides WVE-120101 and WVE-120102 for the treatment of Huntington’s disease

Michael Panzara, MD, MPH, Wave Life Sciences

Previous research has shown that certain single nucleotide polymorphisms (SNPs) are associated with the mutant allele of the Huntingtin (HTT) gene in many patients with Huntington’s disease (HD). The unique association of certain SNPs with the mutant HTT allele opens the possibility of targeting these SNPs with specific treatments that are intended to selectively lower mHTT protein and potentially treat HD. WVE-120101 and WVE-120102 are investigational stereopure antisense oligonucleotides that are intended to target SNPs rs362307 and rs362331, respectively; combined, these two SNPs occur in approximately 70% of patients with HD. Results of in vitro studies show that WVE-120101 and WVE-120102 selectively knockdown mutant HTT mRNA and protein, and leave wild-type HTT mRNA and protein relatively intact. A drug that can lower mutant HTT protein, while leaving the wild-type HTT protein intact, may be able to slow down or stop the progression of HD.

PRECISION-HD1 and PRECISION-HD2 are ongoing Phase 1b/2a clinical trials of WVE-120101 and WVE- 120102, respectively, for the treatment of HD. The primary objective for each of the clinical trials is to evaluate the safety and tolerability of single and multiple doses of WVE-120101 and WVE-120102. Secondary objectives include evaluating plasma pharmacokinetics (PK) and CSF concentration of study drug, the pharmacodynamic (PD) effects as measured by level of mutant HTT protein in CSF, and clinical effects using the Total Functional Capacity (TFC) of the Unified Huntington’s Disease Rating Scale (UHDRS). The trials include ambulatory male and female HD patients ages ≥25 to ≥65 years of age. Patients must have early manifest HD, which is classified as Stage I or Stage II based on UHDRS Total Functional Capacity Scores 7 and 13 and clinical diagnostic motor features of HD, defined as UHDRS Diagnostic Confidence Score = 4. Patients are prescreened with an investigational clinical assay to determine if they have one or both of the targeted SNPs on the mutant HTT allele. Design of the PRECISION-HD trials and methods of the SNP prescreening assay will be presented.

Top

Development of IONIS-HTTRx: From first principles to the first successful HTT-lowering drug trial

Anne Smith, PhD, Ionis Pharmaceuticals, Inc.
Sarah J. Tabrizi, FRCP, PhD, FMedSci, UCL Institute of Neurology

Huntington’s disease (HD) is an autosomal dominant neurodegenerative disease caused by CAG repeat expansion in the HTT gene resulting in polyglutamine expansion in huntingtin protein (HTT). No disease- modifying treatments are currently available. In transgenic rodent models of HD, suppressing HTT production with an antisense oligonucleotide (ASO) delays disease progression and reverses disease phenotype. IONIS-HTTRx is the result of a comprehensive drug discovery effort to design a well-tolerated, potent ASO with high specificity to human HTT mRNA. Extensive preclinical studies were completed to demonstrate proof-of-mechanism and to select the most promising ASO for clinical testing.

Clinical study ISIS 443139-CS1 (NCT02519036) was designed to characterize the safety, tolerability, pharmacokinetics and pharmacodynamics of IONIS-HTTRx in patients with early manifest HD. In this first-in- human, multi-center, double-blind, placebo-controlled clinical trial, 46 HD patients were randomized (3:1) to receive four doses of IONIS-HTTRx or placebo by monthly bolus intrathecal injection followed by a 4-month untreated period. Five ascending-dose cohorts were enrolled with independent Data Safety Monitoring Board review of study data prior to dose escalation.

IONIS-HTTRx administration was safe and well-tolerated at all doses tested in this study. Adverse events were mostly mild and unrelated to study drug. There were no clinically-relevant adverse changes in laboratory parameters. No patients prematurely discontinued from the study. Significant, dose-dependent reductions in CSF mutant HTT (mHTT) were observed.

ASO technology has the potential to provide disease-modifying benefits to people with neurodegenerative diseases. In this study, IONIS-HTTRx was well-tolerated in individuals with early manifest HD, and no adverse safety signals emerged during the treatment or follow-up periods. Significant reductions in CSF mHTT were observed, indicating that IONIS-HTTRx is a promising therapeutic for the treatment of HD.

Top