James Shorter - US grants

NIH Roadmap Initiative tag

DESCRIPTION (provided by applicant): Exploring TDP-43 aggregation and disaggregation A spectrum of seemingly diverse neurodegenerative disorders is now unified by a common underlying theme: the accumulation of non-amyloid, ubiquitylated TDP-43 inclusions in the central nervous system. These lethal disorders range from amyotrophic lateral sclerosis (ALS) to frontal temporal dementia lobar degeneration with ubiquitin positive inclusions (FTLD-U). A causative role for TDP-43 in ALS pathogenesis has been validated by the isolation of mutations in the TDP-43 gene, which are associated with familial and sporadic forms of ALS. Moreover, a yeast model of TDP-43 proteinopathies has established a direct connection between TDP-43 aggregation and toxicity. However, the mechanistic basis of TDP-43 aggregation and how ALS-associated mutations affect the aggregation process directly remain unclear. Further, whether there are cellular factors that can antagonize or reverse the aggregation process remains unknown. We are particularly interested in how two AAA+ ATPases, Hsp104 and p97, might antagonize or reverse TDP-43 aggregation. Hsp104 solubilizes and reactivates proteins from denatured aggregates. p97 prevents aggregation of model substrates, and mutations in p97 are linked with conditions where TDP-43 forms intranuclear aggregates. We hypothesize that understanding these issues will greatly enhance our understanding of ALS and related disorders. Hence, we aim to: (1) Define the mechanisms of TDP-43 aggregation using pure components in vitro. (2) Antagonize and reverse TDP-43 aggregation with Hsp104 and p97 in vitro. (3) Prevent or reverse TDP- 43 aggregation and toxicity in vivo. These studies will provide important new mechanistic insights into TDP-43 aggregation and toxicity and how this might be antagonized. Realization of our objectives will empower the development of therapies for ALS and other TDP-43 proteinopathies. PUBLIC HEALTH RELEVANCE: Several devastating diseases that are caused by nerve degeneration, including amyotrophic lateral sclerosis (ALS), have been unified by a common underlying theme: the clumping of a protein, TDP-43, in nerve cells. Our proposed studies aim to provide new mechanistic insights into how TDP-43 clumps and is toxic, and how this process might be antagonized or reversed. Realization of our objectives will empower the development of new therapies for ALS and related disorders, which are currently untreatable.

DESCRIPTION (provided by applicant): Human immunodeficiency virus type-1 (HIV-1), the causative agent of acquired immune deficiency syndrome (AIDS), has infected ~60 million people worldwide and caused over 25 million deaths. Sexual transmission is the major route of HIV-1 infection and factors that promote this infectious route have recently been identified in semen. Fragments of prostatic acid phosphatase are a major component of semen and form amyloid fibrils that bind HIV virions and can promote infection by several orders of magnitude up to 105-fold. Therefore, a potential preventative strategy is to deploy agents that eliminate these amyloid forms, which are termed Semen-derived Enhancer of Virus Infection (SEVI) fibrils. Unfortunately, amyloid fibrils are notoriously stable and difficult to eradicate. In other settings, they are connected with various fatal neurodegenerative disorders. However, various microbes have harnessed the amyloid form for beneficial purposes, and systems have evolved that can rapidly reverse amyloid formation. One natural protein has emerged that resolves amyloid fibrils with unprecedented alacrity: the protein disaggregase, Hsp104. Hsp104 rapidly solubilizes amyloid forms of several proteins, including yeast prion proteins Sup35 and Ure2, as well as ?-synuclein, which forms amyloid fibrils in Parkinson's disease. We hypothesize that Hsp104 or SEVI-optimized variants can be generated to rapidly dissolve or remodel SEVI fibrils and thereby diminish SEVI-enhanced HIV infection. Thus, we aim to: (1) Develop Hsp104 variants that rapidly disassemble SEVI fibrils~ and (2) Determine whether disassembled products have diminished ability to promote HIV infection. These studies will provide the foundations for developing SEVI disaggregases as preventative agents with the ultimate goal of incorporating them into a gel or solution that dissolves SEVI fibrils in semen and reduces sexual transmission of HIV. The ability to reverse fibril formation (rather than simply inhibit it) and blck sexual transmission of HIV will provide a powerful and much needed weapon against the global HIV/AIDS pandemic. Our approach of targeting a host protein conformer (SEVI fibrils) is fundamentally different from traditional microbicidal approaches that target the virus, and this strategy is anticipated to synergize with direct antiviral strategies.

Project summary: Our research objective is to define the mechanistic basis of Hsp104, a protein disaggregase and hexameric AAA+ (ATPases Associated with diverse Activities) protein from yeast, which remains poorly understood. Hsp104 couples ATP hydrolysis to the dissolution and reactivation of diverse proteins trapped in disordered aggregates, toxic preamyloid oligomers, amyloids, and prions. Hsp104 is the only factor known to dissociate ?-synuclein (?-syn) oligomers and amyloids connected with Parkinson's disease (PD) and rescue ?-syn-induced neurodegeneration in the substantia nigra of a rat PD model. However, Hsp104 activity is limited against ?-syn and very high Hsp104 concentrations are needed for optimal effects. Thus, we engineered potentiated Hsp104 variants, which dissolve fibrils formed by neurodegenerative disease proteins such as TDP-43, FUS, and ?-syn, and mitigate neurodegeneration in the metazoan nervous system at concentrations where Hsp104 is inactive. Curiously, Hsp104 is absent from metazoa. Thus, Hsp104 and potentiated variants could represent a disruptive technology to enhance proteostasis to counter neurodegenerative disease and enable purification of irksome, aggregation-prone proteins for valuable basic or pharmaceutical purposes. However, these endeavors are frustrated by a limited mechanistic understanding of Hsp104, which despite intense investigation remains stalled at a low level of resolution. Three critical barriers impede our understanding of Hsp104. First, we do not understand how Hsp104 selects clients for disaggregation, which limits our ability to tailor Hsp104 activity for specific substrates. This issue is pernicious because potentiated Hsp104 variants can have damaging, off-target effects due to promiscuous activity, which could restrict therapeutic or biotechnological applications. Second, Hsp104 sequence space remains largely unexplored. It is unclear whether natural Hsp104 orthologues exist with divergent enhanced or selective activity against neurodegenerative disease substrates. Third, there is no atomic structure of the Hsp104 hexamer and conflicting cryo-electron microscopy reconstructions have confused the field. Based on our preliminary data, we hypothesize that: (1) potentiated Hsp104 variants can be engineered to be more substrate specific to avoid damaging off-target effects; (2) natural Hsp104 orthologues exist with enhanced activity against neurodegenerative disease substrates and minimal off-target effects; and (3) large structural changes in Hsp104 hexamers upon ATP hydrolysis drive protein disaggregation. Thus, we will meet three aims: (1) Define potentiated Hsp104 variants with enhanced substrate selectivity; (2) Define conserved and divergent activities of natural Hsp104 orthologues; (3) Define high-resolution structural changes in Hsp104 and potentiated variants that drive protein disaggregation. In this way, we will secure a high- resolution mechanistic view of Hsp104, which will empower the engineering of new Hsp104 nanomachines with selective potentiated activity for key applications in biotechnology and medicine.

Project summary There are no effective treatments for various fatal neurodegenerative disorders, including amyotrophic lateral sclerosis (ALS), frontotemporal lobar degeneration (FTLD), or multisystem proteinopathy (MSP) in which specific RNA-binding proteins (RBPs) with prion-like domains mislocalize and aggregate in the cytoplasm of degenerating neurons. For example, wild-type FUS, TAF15, and EWSR1 accumulate in cytoplasmic aggregates and are depleted from the nucleus in degenerating neurons in some forms of FTLD, whereas wild- type or mutant hnRNPA1 and hnRNPA2 exhibit this phenotype in degenerating neurons and other tissues in MSP. For all of these RBPs, which bear a PY-nuclear localization signal (NLS), as well as TDP-43, which bears a distinct canonical NLS, a key pathological event is their mislocalization to cytoplasmic aggregates. Indeed, from this perspective ALS, FTD, and MSP can be viewed fundamentally as nuclear-transport disorders. We hypothesize that agents able to reverse RBP mislocalization and aggregation and thereby restore the RBPs to native form, function, and nuclear localization would mitigate toxicity by simultaneously eliminating: (1) any toxic gain of function of the misfolded form; and (2) any loss of function due to sequestration in cytoplasmic aggregates. Remarkably, our preliminary findings suggest that the nuclear import factor, Karyopherin-?2 (Kap?2, also known as transportin), is such an agent. Thus, Kap?2 can prevent and reverse the aggregation of various RBPs bearing a PY-NLS, and subsequently transport them back to the nucleus. A role for Kap?2 as a nuclear import factor is well established. However, our discovery that Kap?2 has disaggregase activity is unprecedented. Mutations in the PY-NLS of FUS are linked with ALS, and these mutations directly weaken the interaction between Kap?2 and FUS. Here, we propose a series of multidisciplinary studies that employ pure protein biochemistry, yeast, mammalian neuronal culture, and Drosophila models of RBP-opathies to meet two aims: (1) Define Kap?2 activity in preventing and reversing aggregation, mislocalization, and toxicity of specific disease-linked RBPs in vitro and in vivo; (2) Engineer enhanced Kap?2 variants to recognize and disaggregate ALS-linked FUS variants bearing mutations in their PY-NLS. Thus, we will exploit Kap?2 as a bifunctional disaggregase and nuclear import factor to combat pathogenesis associated with cytoplasmic mislocalization and aggregation of FUS, TAF15, EWSR1, hnRNPA1, and hnRNPA2. Our proposed studies will elucidate how a nuclear import factor, Kap?2, can be harnessed and engineered to prevent and reverse these deleterious RBP mislocalization and misfolding events, which will empower the development of therapies for specific forms of ALS, FTD, and MSP.

Project summary In Parkinson's Disease (PD), the proteostasis network fails to counter the misfolding of ?-synuclein (?-syn). ?- Syn forms soluble toxic oligomers and self-templating amyloid fibrils that can initiate and propagate disease de novo. ?-Syn fibrils cluster into large cytoplasmic inclusions termed Lewy Bodies, a pathological hallmark of PD. Human molecular chaperones, Hsp110, Hsp70, and Hsp40 can disaggregate ?-syn fibrils and reduce their toxicity. These findings together with our advances with Hsp104 and potentiated variants suggest that ?-syn oligomers and fibrils are not intractable but can be rapidly disassembled into non-toxic forms. More recently, a second human protein disaggregase, HtrA1, has been discovered that disassembles and degrades tau and amyloid-beta fibrils. HtrA1 is a chaperone and homo-oligomeric PDZ serine protease found in the cytoplasm and extracellular space, which selectively degrades misfolded substrates while leaving folded substrates alone. Our unpublished data suggest that HtrA1 rescues ?-syn toxicity in yeast, and, disassembles and degrades ?- syn fibrils in vitro. We hypothesize that small-molecule enhancers of HtrA1 disaggregase activity could stimulate the elimination of deleterious ?-syn accumulations in the degenerating neurons of PD patients. Thus, we propose that the unanticipated protein disaggregase activity of HtrA1 represents a promising, novel PD- relevant target. We hypothesize that enhancing the activity of HtrA1 disaggregase activity with specific small molecules will enable dissolution and degradation of toxic oligomeric and amyloid forms of ?- syn, and confer therapeutic benefits in PD. Thus, we will pursue two specific aims: (1) Isolate small molecules that enhance the ?-syn disaggregase activity of HtrA1; and (2) Assess the ability of small-molecule enhancers of HtrA1 disaggregase activity to rescue ?-syn aggregation and toxicity in mammalian primary neuron models of PD. By the end of this project, there will be a clear ?go/no go? decision for moving a small- molecule enhancer of HtrA1 into rodent models and ultimately PD patients. Small-molecule stimulation of HtrA1 disaggregase activity could eliminate deleterious ?-syn misfolding in degenerating dopaminergic neurons and provide a game-changing solution for PD. Importantly, small-molecule enhancers of HtrA1 disaggregase activity may also have important applications in other neurodegenerative disorders caused by deleterious protein misfolding, including Alzheimer's disease and frontotemporal dementia.

Project summary In Parkinson's Disease (PD), the most common neurodegenerative movement disorder that afflicts millions of people worldwide, the proteostasis network breaks down and fails to counter the misfolding of the small presynaptic protein ?-synuclein (?-syn). ?-Syn populates a range of misfolded structures ranging from soluble toxic oligomers to self-templating amyloid fibrils capable of initiating and propagating disease de novo. ?-Syn fibrils cluster into large cytoplasmic inclusions termed Lewy Bodies, a pathological hallmark of PD. Recently, we and others have discovered a series of human molecular chaperones, Hsp110, Hsp70, Hsp40, and HspB5 which can disaggregate ?-syn fibrils and reduce their toxicity. Whether this system can also disassemble toxic soluble ?-syn oligomers remains unclear. This endogenous disaggregase system likely becomes overwhelmed and fails to counter ?-syn misfolding in PD and related ?-synucleinopathies. Indeed, Hsp70 chaperones are often sequestered and depleted by excessive accumulation of misfolded proteins. Methods to stimulate the Hsp110, Hsp70, Hsp40, and HspB5 disaggregase machinery in the degenerating neurons of PD patients could reverse deleterious accumulation of ?-syn and provide a game-changing solution for PD. Thus, we propose that the Hsp110, Hsp70, Hsp40, and HspB5 disaggregase machinery represents a promising, novel PD- relevant target. We hypothesize that enhancing the activity of the Hsp110, Hsp70, Hsp40, and HspB5 disaggregase system with specific brain-penetrant small molecules will enable dissolution of toxic oligomeric and amyloid forms of ?-syn, and confer therapeutic benefits in PD. In the proposed studies, we will pursue two specific aims: (1) isolate brain-penetrant small molecules that enhance the ability of Hsp110, Hsp70, Hsp40, and HspB5 to disaggregate ?-syn oligomers and fibrils; and (2) Determine the ability of brain- penetrant, small-molecule enhancers of Hsp110, Hsp70, Hsp40, and HspB5 disaggregase activity to mitigate ?-syn misfolding and toxicity in primary neurons. This project makes an important first step toward exploring the feasibility of developing brain-penetrant small-molecule therapeutics that enhance the activity of the human ?-syn disaggregase machinery (Hsp110, Hsp70, Hsp40, and HspB5) as an alternative treatment strategy for PD. By the end of our studies, there will be a clear ?go/no go? decision for moving a brain-penetrant small- molecule enhancer of Hsp110, Hsp70, Hsp40, and HspB5 disaggregase activity into rodent models and ultimately PD patients. Small-molecule stimulation of the human protein-disaggregase machinery could reverse deleterious ?-syn misfolding in degenerating dopaminergic neurons and provide a transformative solution for PD and related ?-synucleinopathies including dementia with Lewy Bodies and multisystem atrophy. Importantly, brain-penetrant small-molecule enhancers of Hsp110, Hsp70, Hsp40, and HspB5 may also have important applications in other neurodegenerative disorders, including Alzheimer's disease.

Stress and aging can promote protein misfolding and aggregation, leading to cytotoxicity and disease. Indeed, protein misfolding and aggregation are linked with several intractable neurodegenerative diseases, including frontotemporal dementia (FTD). In ~45% of FTD cases, a nuclear RNA-binding protein with a prion-like domain, TDP-43, mislocalizes to cytoplasmic aggregates in degenerating neurons. We propose that a key therapeutic innovation for FTD (and other TDP-43 proteinopathies) will be to develop therapeutic TDP- 43 disaggregases that reverse the aberrant cytoplasmic aggregation of TDP-43 and return functional TDP-43 to the nucleus. We identified the first known human disaggregase system, which consists of three classes of human chaperones Hsp110 (e.g. Apg-2), Hsp70 (e.g. Hsc70), and Hsp40 (e.g. Hdj1) that act in concert. However, humans have multiple versions of Hsp110s (11 variants), Hsp70s (11 variants), and Hsp40s (52 variants) and the precise combinatorial interactions between Hsp110/Hsp70/Hsp40 variants are hypothesized to dictate substrate specificity. The three-gene nature of the human disaggregase system is a major challenge for engineering potentiated variants since most genetic engineering techniques target mutations to only a single gene. In addition, an exhaustive study of the additional Hsp110/Hsp70/Hsp40 complexes has been prohibitive since there are 6,292 possible Hsp110/Hsp70/Hsp40 combinations. This proposal will solve these challenges by leveraging the power of yeast genetics along with major advancements in synthetic DNA assembly and genome engineering technologies to explore and engineer human disaggregase systems in models of FTD. We hypothesize that a specific combination of Hsp110/Hsp70/Hsp40 most potently disaggregates TDP-43. Moreover, we hypothesize that it is possible to engineer and evolve potentiated variants of Hsp110, Hsp70, and Hsp40 to more effectively reverse deleterious TDP-43 misfolding in FTD. Thus, we will pursue four specific aims: (1) Define natural human Hsp110/Hsp70/Hsp40 combinations that rescue TDP-43 toxicity in yeast; (2) Engineer Hsp110/Hsp70/Hsp40 to rescue TDP-43 toxicity in yeast; (3) Define optimal TDP-43 disaggregases in vitro; and (4) Define optimal TDP-43 disaggregases that rescue FTD- linked TDP-43 toxicity in neuronal models. This experimental pipeline leverages the scale and power of yeast genetics to identify Hsp110/Hsp70/Hsp40 combinations and mutants that exhibit rescue of TDP-43 toxicity, which are then experimentally validated in bona fide human neurons. This project will greatly enhance our understanding of human disaggregase mechanisms by exhaustively screening the combinatorial space of three-gene disaggregase interactions and will likely identify new mechanisms to treat FTD.

Project summary. Our research objective is to define the mechanistic underpinnings of the protein disaggregases, Hsp104, and its partial human homolog, Skd3 (human ClpB), which are poorly understood. In non-metazoan eukaryotes, Hsp104 couples ATP hydrolysis to the disaggregation of diverse proteins trapped in disordered aggregates, preamyloid oligomers, and amyloids. Hsp104 is the only factor known to dissociate ?- synuclein (?-syn) oligomers and amyloids linked to Parkinson's Disease (PD) and rescue neurodegeneration in a rat PD model. However, Hsp104 activity is limited against ?-syn and high Hsp104 concentrations are required for optimal effects. Thus, we engineered potentiated Hsp104 variants, which dissolve fibrils formed by ?-syn as well as TDP-43 and FUS (which are linked to amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), an Alzheimer's Disease-Related Dementia (ADRD), which mitigate neurodegeneration in the metazoan nervous system more effectively than Hsp104. Though potent disaggregases, these potentiated Hsp104 variants lack substrate specificity and are prone to toxic off-target effects. To address this issue, we engineered new potentiated Hsp104 variants with minimal off-target effects and ?-syn-specific Hsp104 variants, which exhibited enhanced therapeutic utility. These engineered disaggregases could provide a disruptive technology to combat neurodegenerative disease and enable purification of aggregation-prone proteins for basic or pharmaceutical purposes. Curiously, Hsp104 does not have an exact metazoan ortholog. Remarkably, we have found that a partial homolog of Hsp104 found in human mitochondria, an AAA+ protein called Skd3 (human ClpB), has powerful protein disaggregase activity comparable to potentiated Hsp104 variants. Despite these important advances, our mechanistic understanding of Hsp104 and Skd3 is limited by three critical barriers. First, we do not understand how Hsp104 selects substrates for disaggregation. Thus, we have not yet developed TDP-43- or FUS-specific variants for ALS/FTD. Second, we do not understand how Hsp104 is regulated. Thus, the mechanism by which specific mutations in nucleotide-binding domain 2 (NBD2) potentiate Hsp104 remain unclear. Third, Skd3 is poorly characterized in terms of its disaggregase capabilities, structure, and mechanism. Based on our preliminary data, we hypothesize that: (1) potentiated Hsp104 variants can be engineered to be more selective for ALS/FTD-linked TDP-43 and FUS; (2) specific NBD2 mutations potentiate Hsp104 via a novel mechanism; and (3) Skd3 is a powerful human protein disaggregase with broad capabilities and mechanistic similarities to Hsp104. Thus, we will meet three aims: (1) Define Hsp104 variants with enhanced TDP-43 and FUS selectivity; (2) Define how specific NBD2 mutations potentiate Hsp104 activity; (3) Define the capabilities, mechanism, and structure of the human Skd3 AAA+ disaggregase. In this way, we will secure an enhanced mechanistic understanding of Hsp104 and Skd3, which will empower their development for important applications in biotechnology and medicine.