Search bioRxiv⌕ Search

Biology subjects

Spillantini, M. G.

Publications and source records attributed to Spillantini, M. G..

13 recordsLinked to original sources

Templated seeding induces the formation of Alzheimer paired helical tau filaments in human brain organoids

The prion concept postulates that seeded aggregation results in the recruitment of soluble protein monomers and their assembly into amyloid filaments with the same structures as those of the seeds. Here we show that this concept applies to tau filaments formed in cortical organoids obtained from human iPSCs of control subjects. Tau seeds were extracted from the frontal cortex of individuals with Alzheimer disease. When added to human cortical organoids expressing endogenous levels of three-repeat tau, they gave rise to filaments of three-repeat tau whose formation required endogenous tau expression. By cryo-EM, these seeded tau filaments were identical to the paired helical filaments of the seeds. This novel finding demonstrates the faithful templated seeding of tau in human nerve cells expressing endogenous levels of tau and opens the way to studying the mechanisms of paired helical filament formation and their downstream effects in a human system amenable to genetic modification.

neuroscience↗

Early synaptic pathology is associated with small tau aggregates in Alzheimer's disease

Alzheimers disease (AD) is phenotypically characterised by progressive memory loss, which involves tau aggregation and synaptic pathology. Here we characterised the tau aggregates in individual synaptosomes from AD cases and controls measuring their number and size using single molecule fluorescence microscopy. A total of 7,888 synaptosomes from pre-frontal cortex samples were studied, showing the presence of AT8-positive tau aggregates in a small fraction of synaptosomes ([~]3%) from control brains, reaching [~]20% by Braak stage 6 with more larger aggregates. We then investigated the multi-phosphorylation of synaptic tau aggregates for AT8 and T181 and quantified the co-localisation of phosphatidylserine and CD47, synaptic "eat me" and "dont eat me" signals respectively, along with synaptogyrin-3, which contributes to tau mediated synaptic dysfunction. T181, phosphatidylserine, and synaptogyrin-3 co-localisation with AT8-positive tau were increased during stage 3 and CD47 was decreased, indicating early synaptic pathology is associated with the formation of small tau aggregates, contributing to microglia-driven synaptic loss.

neuroscience↗

Proteomizer: Leveraging the Transcriptome-Proteome Mismatch to Infer Novel Gene Regulatory Relations

The correlation between transcriptomic (Tx) and proteomic (Px) profiles remains modest, typically around r = 0.5 across genes and r = 0.3 across samples, limiting the utility of transcriptomic data as a proxy for protein abundance. To address this, we introduce Proteomizer, a deep learning platform designed to infer a samples Px landscape from its Tx and miRNomic (Mx) profiles. Trained on 8,613 matched Tx-Mx-Px samples from TCGA and CPTAC, Proteomizer achieved a Tx-Px correlation of r = 0.68, representing the highest performance reported to date for this task. We further developed a Monte Carlo simulation framework to evaluate the impact of proteomization on differential expression analysis. Proteomizer substantially improved the accuracy of differential gene expression detection, with p-value precision increasing by up to 62-fold, and by as much as six orders of magnitude for a subset of genes enriched in mitochondrial and ribosomal functions. However, performance gains did not generalize to unseen tissue types or datasets generated using different protocols. Finally, we applied explainable AI (XAI) techniques to identify regulatory relations contributing to Tx-Px discrepancies. Our predictions from 100 highly annotated genes were cross-compared against by a literature-based biological knowledge graph of 322 million annotations: our explainers achieved a ROC-AUC of 0.74 in predicting miRNA-gene downregulation interactions. To our knowledge, this is the first study to systematically evaluate the biological relevance, limitations, and interpretability of proteomization models, establishing Proteomizer as a state-of-the-art tool for multiomic integration and hypothesis generation. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=199 HEIGHT=200 SRC="FIGDIR/small/660946v1_ufig1.gif" ALT="Figure 1"> View larger version (54K): org.highwire.dtl.DTLVardef@59b64dorg.highwire.dtl.DTLVardef@1cf82a4org.highwire.dtl.DTLVardef@89fdb6org.highwire.dtl.DTLVardef@16fe109_HPS_FORMAT_FIGEXP M_FIG C_FIG miRNA: microRNA; Mx: miRNome; MS: mass spectrometry; PTM: post-translational modifications; Px: proteome; Px: predicted proteome; r: Pearson correlation coefficient; RBP: RNA binding proteins; ROC-AUC: area under the receiver operating characteristic curve; RPPA: reverse-phase protein array; TF: transcription factors; Tx: transcriptome; XAI: explainable artificial intelligence.

bioinformatics↗

GhostBuster: A Deep-Learning-based, Literature-Unbiased Gene Prioritization Tool for Gene Annotation Prediction

All genes are not equal before literature. Despite the explosion of genomic data, a significant proportion of human protein-coding genes remain poorly characterized ("ghost genes"). Due to sociological dynamics in research, scientific literature disproportionately focuses on already well-annotated genes, reinforcing existing biases (bandwagon effect). This literature bias often permeates machine learning (ML) models trained on gene annotation tasks, leading to predictions that favor well-studied genes. Consequently, standard ML performance metrics may overestimate biological relevance by overfitting literature-derived patterns. To address this challenge, we developed GhostBuster, an encoder-decoder ML platform designed to predict gene functions, disease associations and interactions while minimizing literature bias. We first compared the impact of biased (Gene Ontology) versus unbiased training datasets (LINCS, TCGA, STRING). While literature-biased sources yielded higher ML metrics, they also amplified bias by prioritizing well-characterized genes. In contrast, models trained on unbiased datasets were 2-3x more effective at identifying recently discovered gene annotations. Notably, one of the unbiased channels (TCGA), combined minimal amounts of literature bias with robust performance, at a test ROC-AUC of 0.8-0.95. We demonstrate that GhostBuster can be applied to predict novel gene functions, refine pathway memberships, and prioritize intergenic GWAS hits. As the first ML framework explicitly designed to counteract literature bias, GhostBuster offers a powerful tool for uncovering the roles of understudied genes in cellular function, disease, and molecular networks. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=197 SRC="FIGDIR/small/660948v1_ufig1.gif" ALT="Figure 1"> View larger version (59K): org.highwire.dtl.DTLVardef@5318eforg.highwire.dtl.DTLVardef@18a2f99org.highwire.dtl.DTLVardef@13b0b87org.highwire.dtl.DTLVardef@171d3e0_HPS_FORMAT_FIGEXP M_FIG C_FIG GCN: graph convolutional neural network; GeneOntol.: Gene Ontology; LINCS: Library of Integrated Network-Based Cellular Signatures; MLP: multi-layer perceptron (deep learning); PPI: (physical) protein-protein interactions; SVM: support vector machine; TCGA: The Cancer Genome Atlas; y/n: yes or no (binary classification).

bioinformatics↗

Inflamed Microglia like Macrophages in the Central Nervous System of Prodromal Parkinson's Disease

We hypothesized that prodromal Parkinsons Disease (PD), characterized by REM Sleep Behavior Disorder and hyposmia, is an inflammatory disorder initiated in the gut that evolves into a neurodegenerative process manifest as classical PD. To test this hypothesis, we performed single-cell RNAseq analysis of cerebrospinal fluid (CSF) and blood from 118 individuals, comparing healthy subjects with prodromal PD and manifest PD to patients with the central nervous system (CNS) autoimmune disease, multiple sclerosis (MS). Surprisingly, we identified increased numbers of immune cells in the CSF of patients with prodromal PD. Single-cell RNA sequencing revealed increases in CSF-specific microglia-like macrophages with enhanced JAK-STAT and TNF signaling signatures in prodromal PD. CSF macrophages exhibited similar transcriptional profiles to dural macrophages and gut muscularis macrophages from -synuclein-expressing PD model mice. Finally, shared memory T cell clones were found in humans between gut and CSF T cells. These findings uncover a myeloid-mediated TNF inflammatory process in the CNS of patients with prodromal PD, with an immunological linkage between the gut and the CNS.

immunology↗

Alpha-synuclein aggregation and dopaminergic neuron death in a new mouse model of Parkinson disease expressing human full-length and C-terminally truncated 1-120 alpha-synuclein.

Parkinsons disease (PD) is neuropathologically characterized by the presence of Lewy bodies and Lewy neurites made of aggregated alpha-synuclein (aSyn). Several studies have shown that Lewy bodies contain C-terminally truncated aSyn which is more prone to fibrillation and enhances the aggregation of human full-length aSyn in vitro. Nevertheless, in vivo studies addressing whether and how human C-terminally truncated aSyn may exacerbate the pathological aggregation and toxicity of human full-length aSyn are currently lacking. In the present study we aimed to determine whether the co-expression of human full-length and 1-120 truncated aSyn would enhance human full-length aSyn aggregation and toxicity. To this end we generated and characterized two novel alpha- synuclein mouse models. Specifically, we first produced a mouse line expressing full-length human aSyn under its own promoter (BACo mice) in the absence of endogenous mouse aSyn. Then, the BACo mice were crossed with the previously described MI2 mice, which express human 1-120 truncated aSyn under the control of the tyrosine hydroxylase (TH) promoter on a mouse aSyn null backgroundin order to obtain the MI2BACo mice, which express both human 1-120 truncated and full-length aSyn in a mouse aSyn null background. We found that aSyn aggregation was present and increased with age in both mouse lines but was significantly higher in the MI2BACo mice. These mice exhibited progressive nigrostriatal accumulation of aSyn aggregates which were often composed of full-length and 1-120haSyn, the latter being specifically recognized by using a novel antibody raised against the 1-120haSyn. In addition, we observed that the presence of both truncated and full-length aSyn in MI2BACo mice led to a faster and more pronounced aSyn aggregation which associated with more severe dopaminergic striatal fiber degeneration and nigral neuron loss. Collectively, our results indicate that the expression of human C-terminally truncated aSyn exacerbates human full-length aSyn pathology and support that the novel MI2BACo transgenic mouse line represents an innovative experimental model to study the biological basis of PD and test novel therapeutic approaches.

neuroscience↗

SynPull: a novel method for studying neurodegeneration-related aggregates in synaptosomes using super-resolution microscopy

Synaptic dysfunction is one of the primary hallmarks of both Alzheimers and Parkinsons disease, leading to cognitive and behavioural decline. While alpha-synuclein, beta-amyloid, and tau are involved in the physiological functioning of synapses, their pathological aggregation has been linked to synaptic dysfunction. However, the methodology for studying the small (sub-diffraction limit) and soluble aggregates -often called oligomers, formed by these proteins is limited. Here we describe SynPull, a novel method combining single-molecule pulldown, super-resolution microscopy, and advanced computational analyses, in order to reliably study the quantity and morphology of the oligomeric alpha-synuclein, beta-amyloid, and AT8-positive tau aggregates in synaptosomes harvested from post-mortem human brain samples and mouse models. Using SynPull, we show that AT8-positive tau is the predominant aggregate type in AD, with significantly more aggregates compared to the control samples, yet the aggregate size does not differ between disease and control samples. Meanwhile, the relatively smaller amount of alpha-synuclein and beta-amyloid aggregates found in the synapses are larger than the extra-synaptic ones. Collectively, these results show the utility of SynPull to study pathological aggregates in dementia, which can help further understand the disease mechanisms causing synaptic dysfunction. O_FIG O_LINKSMALLFIG WIDTH=140 HEIGHT=200 SRC="FIGDIR/small/609517v1_ufig1.gif" ALT="Figure 1"> View larger version (23K): org.highwire.dtl.DTLVardef@d53d06org.highwire.dtl.DTLVardef@1995971org.highwire.dtl.DTLVardef@bf3edaorg.highwire.dtl.DTLVardef@1b75ca5_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOGraphical abstract.C_FLOATNO Human post-mortem orbitofrontal cortex samples from subjects with neuropathological diagnosis of Alzheimers and Parkinsons disease, as well as age-matched controls cut into [~]300 mg sections, and MI2, APPNL-G-F, P301S, and C57Bl/6J mouse brains were first homogenised in synaptosome buffer using a Dounce homogeniser and then filtered and centrifuged to separate nuclei and organelles from the synaptic fragments. Then, the isolated synaptosomes were incubated on the SiMPull surface with anti-neurexin antibody overnight, followed by fixation and permeabilisation. Imaging antibodies against beta-amyloid, alpha-synuclein, and AT8-positive tau were added to the samples and dSTORM imaging was performed to super-resolve the aggregates. C_FIG

neuroscience↗

New SNCA mutation and structures of α-synuclein filaments from juvenile-onset synucleinopathy

A 21-nucleotide duplication in one allele of SNCA was identified in a previously described disease with abundant -synuclein inclusions that we now call juvenile-onset synucleinopathy (JOS). Both wild-type -synuclein and its insertion mutant containing seven additional residues (MAAAEKT) after residue 22 were present in sarkosyl-insoluble material that was extracted from frontal cortex of the individual with JOS and examined by electron cryo-microscopy. The structures of JOS filaments, comprising either a single protofilament, or a pair of protofilaments, revealed a new -synuclein fold that differs from the folds of Lewy body diseases and multiple system atrophy (MSA). The JOS fold consists of a compact core, the sequence of which (residues 36-100 of wild-type -synuclein) is unaffected by the mutation, and two disconnected density islands (A and B) of mixed sequences. There is a non-proteinaceous cofactor bound between the core and island A. The JOS fold resembles the common substructure of MSA type I and type II dimeric filaments, with its core segment approximating the C-terminal body of MSA protofilaments B and its islands mimicking the N-terminal arm of MSA protofilaments A. The partial similarity of JOS and MSA folds extends to the locations of their cofactor-binding sites. Our findings provide insight into a likely mechanism of JOS fibrillation in which mutant -synuclein of 147 amino acids forms a nucleus with the JOS fold, around which wild-type and mutant proteins assemble during elongation.

neuroscience↗

Microglia become hypofunctional and release metalloproteases and tau seeds after phagocytosing live neurons with P301S tau aggregates

The microtubule-associated protein tau aggregates in multiple neurodegenerative diseases, causing inflammation and changing the inflammatory signature of microglia by unknown mechanisms. We have shown that microglia phagocytose live neurons containing tau aggregates cultured from P301S tau transgenic mice due to neuronal tau aggregate-induced exposure of the eat me signal phosphatidylserine. Here we show that after phagocytosis, microglia become hypophagocytic while releasing seed-competent insoluble tau aggregates. These microglia activate acidic {beta}-galactosidase, and release senescence-associated cytokines and matrix remodeling enzymes alongside tau, indicating a senescent phenotype. In particular, the marked NF{kappa}B-induced activation of matrix metalloprotease 3 (MMP3/stromelysinl) was replicated in the brains of P301S mutant tau transgenic mice, and in human brains from tauopathy patients. These data show that microglia that have been activated to ingest live neurons with tau aggregates behave hormetically, becoming hypofunctional while acting as vectors of tau aggregate spreading.

neuroscience↗

Tracking tau fibrillogenesis and consequent primary phagocytosis of neurons mediated by microglia in a living tauopathy model

Fibrillary tau pathologies have been implicated in Alzheimers and allied neurodegenerative diseases, while mechanisms by which neurons bearing tau tangles die remain enigmatic. To address this issue, we pursued tau and related key pathologies macroscopically by PET and MRI and microscopically by intravital two-photon laser optics. Time-course macroscopic assays of tau transgenic mice demonstrated intimate associations of tau deposition and increase of an inflammatory microglial marker, translocator protein (TSPO), with regional brain atrophy. Longitudinal microscopy of these mice revealed a rapid turnover of tau lesions resulting from continuous generation of new tau aggregates followed by loss of neurons and their fibrillar contents. This technology also allowed the capturing of the disappearance of tangle-bearing neurons several days after being engulfed by activated microglia. Notably, a therapeutic TSPO ligand profoundly suppressed the mobility and phagocytotic activity of microglia and improved neuronal survival in this model, supporting the involvement of primary phagocytosis of viable neurons by microglia in tau-primed neuronal death. Finally, partial depletion of microglia revealed roles of immune factors, MFG-E8 and C1q, as eat-me signals for an immediate attraction of phagocytic microglia towards the elimination of tangle-loaded neurons.

neuroscience↗

CSPα reduces aggregates and rescues striatal dopamine release in αsynuclein transgenic mice

Synuclein aggregation at the synapse is an early event in Parkinsons disease and is associated with impaired striatal synaptic function and dopaminergic neuronal death. The cysteine string protein (CSP) and synuclein have partially overlapping roles in maintaining synaptic function and mutations in each cause neurodegenerative diseases. CSP is a member of the DNAJ/HSP40 family of co-chaperones and like synuclein, chaperones the SNARE complex assembly and neurotransmitter release. Synuclein can rescue neurodegeneration in CSPKO mice. However, whether synuclein aggregation alters CSP expression and function is unknown. Here we show that synuclein aggregation at the synapse induces a decrease in synaptic CSP and a reduction in the complexes that CSP forms with HSC70 and STGa. We further show that viral delivery of CSP rescues in vitro the impaired vesicle recycling in PC12 cells with synuclein aggregates and in vivo reduces synaptic synuclein aggregates restoring normal dopamine release in 1-120hsyn mice. These novel findings reveal a mechanism by which synuclein aggregation alters CSP at the synapse, and show that CSP rescues synuclein aggregation-related phenotype in 1-120hsyn mice similar to the effect of synuclein in CSPKO mice. These results implicate CSP as a potential therapeutic target for the treatment of early-stage PD.

neuroscience↗

Neuropathological validation of the MDS-PSP criteria with PSP and other frontotemporal lobar degeneration

BackgroundProgressive supranuclear palsy (PSP) is clinically heterogeneous. Clinical diagnostic criteria were revised in 2017, to increase sensitivity and operationalize the diagnosis of PSP Richardsons syndrome (PSP-RS) and \"variant\" syndromes (vPSP).\n\nObjectivesTo determine the (1) sensitivity and specificity of the 1996 NINDS-SPSP and 2017 MDS-PSP criteria; (2) false positive rates in frontotemporal dementia with frontotemporal lobar degeneration (FTLD); and (3) clinical evolution of variant PSP syndromes (vPSP).\n\nMethodsRetrospective multicenter review of 108 neuropathologically-confirmed PSP patients and 81 patients with other forms of FTLD: 38 behavioral variant frontotemporal dementia (bvFTD), 14 non-fluent/agrammatic variant primary progressive aphasia (nfvPPA), and 29 corticobasal degeneration (CBD), Sensitivity and specificity of the MDS-PSP criteria were compared to the NINDS-SPSP criteria at baseline. In a subset of cases, the timing and frequency of clinical features were compared across groups over six years.\n\nResultsSensitivity for recognition of probable and possible PSP pathology was higher by MDS-PSP criteria (72.2-100%) than NINDS-SPSP criteria (48.1-61.1%). Specificity was higher by NINDS-SPSP criteria (97.5-100%) than MDS-PSP criteria (53.1-95.1%). False positives by MDS-PSP criteria were few for bvFTD (10.5-18.4%) but common for CBD and nfvPPA (fulfilling \"suggestive of PSP). Most vPSP cases developed PSP-RS-like features within six years, including falls and supranuclear gaze palsy, distinguishing frontal presentations of PSP from bvFTD, and speech/language presentations of PSP from nfvPPA.\n\nConclusionsThe 2017 MDS-PSP criteria successfully identify PSP, including variant phenotypes. This independent validation of the revised clinical diagnostic criteria strengthens the case for novel therapeutic strategies against PSP to include variant presentations.

neuroscience↗

Depopulation of α-synuclein aggregates is associated with rescue of dopamine neuron dysfunction and death in a new Parkinson disease model

Parkinsons Disease (PD) is characterized by the presence of -synuclein aggregates known as Lewy bodies and Lewy neurites, whose formation is linked to disease development. The causal relation between -synuclein aggregates and PD is not well understood. We generated a new transgenic mouse line (MI2) expressing human, aggregation-prone truncated 1-120 -synuclein under the control of the tyrosine hydroxylase promoter. MI2 mice exhibit progressive aggregation of -synuclein in dopaminergic neurons of the substantia nigra pars compacta and their striatal terminals. This is associated with a progressive reduction of striatal dopamine release, reduced striatal innervation and significant nigral dopaminergic nerve cell death starting from 6 and 12 months of age, respectively. Overt impairment in motor behavior was found in MI2 mice at 20 months of age, when 50% of dopaminergic neurons are lost. These changes were associated with an increase in the number and density of 20-500nm -synuclein species as shown by dSTORM. Treatment with the oligomer modulator anle138b, from 9-12 months of age, restored striatal dopamine release and prevented dopaminergic cell death. These effects were associated with a reduction of the inner density of -synuclein aggregates and an increase in dispersed small -synuclein species as revealed by dSTORM. The MI2 mouse model recapitulates the progressive dopaminergic deficit observed in PD, showing that early synaptic dysfunction precedes dopaminergic axonal loss and neuronal death that become associated with a motor deficit upon reaching a certain threshold. Our data also provide new mechanistic insight for the effect of anle138bs function in vivo supporting that targeting -synuclein aggregation is a promising therapeutic approach for PD.

neuroscience↗