Parkinson's disease (PD) is the second most common neurodegenerative disease, affecting over 8.5 million people worldwide, and its global burden is projected to reach over 17 million cases by 2040 [3]. PD manifests through a combination of motor and non-motor symptoms, including tremors, postural instability, cognitive impairment, dementia, and pain [4]. These symptoms progressively worsen over time and strongly reduce the patients' quality of life. PD pathology is characterized by the progressive degeneration of dopaminergic neurons in the substantia nigra pars compacta caused by the accumulation of aggregated forms of the pre-synaptic protein alpha-synuclein (alpha-Syn). The deposition of aggregated alpha-Syn is now recognized as a central feature of a group of disorders collectively called synucleinopathies, that include Dementia with Lewy Bodies (DLB) and Multiple System Atrophy (MSA). Despite considerable progress on the structural biology of alpha-Syn aggregates, molecular mechanisms mediating their cell-to-cell transmission, propagation and neurotoxicity remain only partially understood. Hence, treatment options are limited to symptomatic therapies that do not halt nor slow down the progression of the disease.
alpha-synuclein aggregation
alpha-Syn is a natively unfolded protein that predominantly exists in a monomeric form under physiological conditions [5]. Upon binding to cellular membranes, it can adopt an alpha-helical multimeric conformation [6]. Due to its structural flexibility, alpha-Syn is capable of assuming various conformations, including stable small oligomeric aggregates, which may exist in equilibrium with monomers under physiological conditions [5]. alpha-Syn cellular levels are regulated by a tight modulation of production, aggregation, and clearance [7]. Any disruption in these processes can result in abnormal alpha-Syn levels, potentially leading to misfolding and aggregation. Aggregation of alpha-Syn is thought to initiate in the cytoplasm or on cellular membranes [8]. Here, the monomeric protein adopts a β-sheets-enriched conformation that leads to the formation of insoluble aggregates that can recruit endogenous alpha-synuclein, promoting the formation of bigger oligomers and fibrils. Over time, smaller aggregates coalesce into larger inclusions, ultimately forming LBs or Lewy neurites [8]. A debate persists regarding whether fibrils or oligomers represent the toxic species primarily responsible for PD pathology. While several studies support both hypotheses, current evidence suggests that toxicity is more closely associated with the process of fibril formation and its intermediate products rather than with oligomers or fibrils themselves [5].
Onset and progression of synucleinopathies
Experimental evidence from various cellular and animal models indicates that alpha-Syn transfer occurs in two defined steps: the release of pathogenic alpha-Syn species from infected cells, and their uptake via passive or active endocytic pathways [9, 10]. Once alpha-Syn aggregates have been internalized, they recruit endogenous alpha-Syn and are thought to seed fibril growth, but little is known about what drives their toxicity or how they interact with endogenous alpha-Syn and other factors to promote endogenous alpha-Syn aggregation [11, 12]. Similarly, additional factors are likely to modulate different cellular responses to the accumulation of pathogenic alpha-Syn species. For instance, cellular quality control (QC) systems have been implicated in clearing alpha-Syn aggregates from neurons. However, the activity of some of these QC components (e.g., Hsp40-Hsp70 chaperones) may either disaggregate fibrils to monomeric species or in fact help fibrils to propagate by creating new smaller seeds [13, 14]. A comprehensive understanding of the cellular response to alpha-Syn is still lacking.
Forward genetics in the study of synucleinopathies
Targeted whole-genome perturbation is revolutionizing our understanding of human health and disease, particularly in the field of neurodegeneration. RNA-interference and mutagen-mediated screenings yielded a plethora of previously unknown targets modulating alpha-Syn expression, aggregation and clearance [15-17]. Our lab performed a genome-wide RNAi-based screen for genes regulating the intercellular transfer of alpha-Syn and identified 38 genes, one of which had been associated to PD through genome-wide association studies (GWAS) [1]. However, siRNA screens suffer from off-target effects and the knockdown efficiency is typically 60-80%. Consequently, siRNA-based screens may miss important modifiers including catalytic proteins which rarely display pathological haploinsufficiency. CRISPR-based perturbations are much more efficacious and target-specific than RNA interference and allow for both genetic activation and knockout/epigenetic silencing [2]. Gene functions may not consistently manifest under both loss-of-function and gain-of-function conditions and often exhibit a discernible phenotype only when perturbed in one specific direction [18] This implies that activation and silencing screens are complementary and non-overlapping. Combining these two approaches is a powerful method to define the complete landscape of modifiers of alpha-Syn aggregation.Recently, CRISPR screens in human cells confirmed heparan sulfate proteoglycans as the major receptors for alpha-Syn PFFs and revealed that loss of C3orf58 (DIPK2A) or SLC39A9, which are involved in Golgi-associated functions, significantly reduced PFF uptake by impairing PFF binding to the cell surface [19]. Additionally, another genome-wide pooled CRISPR/Cas9 KO screen in HEK293 cells identified tetraspanin 3 (TSPAN3) as a potential regulator of alpha-Syn oligomerization, highlighting its potential as a therapeutic target for PD prevention and treatment [20].
Genome-wide CRISPRa screen for modifiers of alpha-Syn hyperphosphorylation and aggregation
We performed a whole genome CRISPR activation screen to identify modifiers of the accumulation of phosphorylated alpha-Syn at residue serine 129 (pS129-alpha-Syn). Buildup of pS129-alpha-Syn occurs after the initial alpha-Syn aggregation [21, 22] and represents a reliable biomarker of the presence of alpha-Syn pathology [23, 24]. HEK cells were engineered to stably overexpress human alpha-Syn and the transactivator dCas9-VPR. Arrayed library plasmids were delivered through lentiviral transduction. alpha-Syn-dCas9-HEK cells were treated with sonicated alpha-Syn pre-formed fibrils (PFFs) complexed with the TransIT-LT1 Transfection Reagent. Exogenous PFFs induced intracellular deposition of phosphorylated alpha-Syn puncta which were detected by immunostaining with the monoclonal antibody P-syn/81A that only binds to hyperphosphorylated Ser129. Nuclei were stained with DAPI and cell bodies were visualized with HCS CellMask and Deep Red Stain. pS129-alpha-Syn positive cells were counted and normalized to the total number of cells. The development and validity of the assay were carried out using negative (non-targeting) and positive controls (Rab13). Overexpression of Rab13 was shown to decrease alpha-Syn aggregates [16].
For quality control, we used the strictly standardized mean difference (SSMD) [25]. The SSMD for all the plates ranged from good to excellent. The heatmaps for individual plates were inspected as well, and no plate gradients nor edge effects were detected. As anticipated, most of the screened genes had no impact on alpha-Syn hyperphosphorylation and aggregation. However, we identified several promising candidates that we aim to characterize in greater detail. Individual retesting of the candidates recapitulated what observed in the primary screen and confirmed the robustness of the approach. Interestingly, one of the top downregulators of pS129-alpha-Syn aggregation has been associated to PD through GWAS studies [26]. We also detected a member of the tetraspanin protein family, which has been recently linked to synucleinopathies through CRISPR screens [20]. A more detailed characterization of the hit genes and their mechanism will constitute part of the present application (see below).
High-throughput image analysis pipeline for segmentation of alpha-Syn aggregates
Over the past year, the Aguzzi lab has developed a python-based software for automated processing of high content screen image data, for the detection of alpha-synuclein inclusions in cells. In particular, the software performs image segmentation of images containing nuclei, cells, and alpha-synuclein aggregates as separate channels. The nuclei and cell segmentation are done using state-of-the-art, publicly available deep learning models (StarDist [27] and Cellpose [28]). The aggregates are segmented using an in-house trained deep learning model, particularly a Convolutional Neural Network. The segmented structures are the colocalized to quantify the percentage of cells containing alpha-synuclein inclusions. Using AggreQuant, we have analyzed a genome-wide screen containing in total 1.2 million images to identify the genes that influence the spread of alpha-Syn inclusions in cells. The software is publicly available on GitHub (AggreQuant: https://github.com/aecon/AggreQuant)
Objectives
We posit that most factors and mechanisms contributing to alpha-Syn pathologies are hitherto unknown, and that clarifying the molecular pathways driving neurodegeneration is a necessary step towards the development of efficient therapies. We have taken advantage of the CRISPR-Cas technologies that we have developed in the past 4 years, and of new genome-wide arrayed libraries, to address the molecular basis for alpha-Syn aggregation in an unbiased, hypothesis-free manner. Our thorough investigation has led to the discovery of a plethora of interesting and totally unexpected candidates. We plan to characterize their mode of action, and which molecular pathways are involved in the formation and spread of alpha-Syn aggregates. We will then investigate the specificity of the hit-induced modulation on multiple alpha-Syn fibril strains that are responsible for distinct synucleinopathies. Finally, we plan to validate our findings in more complex model systems, including samples derived from patients suffering from disease. Specifically, we shall address the following aims:
Uncover molecular pathways that regulate alpha-Syn aggregation
- Do candidate genes affect monomeric alpha-Syn?
- Which stage of the aggregation cascade is affected by putative hit genes?
- Are there any shared pathways among the identified hits?
investigate the importance of strain specificity in the context of alpha-Syn aggregation
- Do hit genes affect propagation only of specific alpha-Syn fibrillar strains?
- Are these effects confirmed in patient-derived samples?
identify genetic drivers of alpha-Syn induced cytotoxicity
- Can we identify synthetic lethal genes for alpha-Syn PFF-treated cells?
Relevance
Synucleinopathies constitute a substantial human and financial burden on society worldwide, and remain largely untreatable, due at least in part to a lack of understanding of disease mechanisms. We propose a unique combination of unbiased and mechanistic approaches to shed light on important and hitherto unanswered questions of etiology and cellular/molecular mechanisms of synucleinopathies. Our study will characterize structure-function relationships of alpha-Syn and identify commonalities and strain-specificity in alpha-Syn modulators and cellular responses. Further, these studies will result in cell model systems that may recapitulate the diseases at the molecular level. As we will interrogate the entire genome, we have reasons to believe that at least some of the identified candidates will be druggable and might therefore lead to the development of new therapeutics.