Abstract The aggregation of α-synuclein (αSyn) is a molecular hallmark of Parkinson’s disease (PD) and other synucleinopathies. Understanding the molecular mechanisms that determine the aggregation of this protein may thus facilitate the development of disease-modifying therapies. While αSyn is most commonly expressed as a 140-residue protein (αSyn-140), recent evidence suggests an involvement of alternatively spliced αSyn isoforms in disease onset and progression. Here, we report and characterise the interaction between αSyn-140 and the aggregation-prone αSyn-112 variant, one of the most abundant αSyn splice isoforms. We found that amounts as low as 1% of αSyn-112 accelerate the nucleation and aggregation of αSyn-140. To further investigate this phenomenon, we employed MALDI-MS and NMR spectroscopy, confirming that αSyn-140 and αSyn-112 monomers interact strongly with one another. Furthermore, to assess the association of αSyn-112 with disease pathology, we performed immunohistochemical staining combined with confocal microscopy on PD brain samples. Thereby, we found an increase in the number as well as the area of αSyn-112 immunoreactive aggregates compared to healthy controls. These results illustrate how low-abundance αSyn splice isoforms can modulate the aggregation landscape of αSyn-140 and in turn contribute to the molecular heterogeneity of synucleinopathies.

The traditional approach of using double-blind, placebo controlled, parallel group trial designs has confirmed the efficacy of a large number of agents in relieving the symptoms of Parkinson's disease (PD) but has not, to date, led to the discovery of any disease-modifying treatments for PD. There are multiple potential reasons underlying the failure to find disease modifying approaches, which may in part relate to; inadequate understanding of PD pathophysiology and therefore inappropriate target selection; the possibility that even good candidate drugs may simply fail to reach and to ultimately engage with their putative targets at the required dose; and the significant heterogeneity of the disease both in terms of its pathophysiology and its motor and non-motor symptoms. PD also has some additional challenges that may be addressed by careful consideration of trial design. This includes; its generally slow rate of disease progression necessitating long follow-up times to identify evidence of disease slowing; lack of understanding regarding the optimal stage of disease that might be most amenable to intervention; as well as lack of consensus regarding which outcome measures best capture patient relevant disease progression, and which biomarkers might consistently and objectively provide the earliest indication of disease progression. In this review we will discuss these issues and potential approaches that may help in the evolution of clinical trial design and thus ultimately provide a pathway to increase the likelihood of successful identification of disease-modifying treatments for Parkinson's disease.

There are currently no validated peripheral biomarkers for the diagnosis, differentiation or progression of the neurodegenerative synucleinopathies, Parkinson's disease and multiple system atrophy. Diagnostic biomarkers that reflect the disease mechanisms or progression biomarkers that change with disease severity would be extremely valuable for assessing disease-modifying therapies. Our objective was to explore putative protein biomarkers of Parkinson's disease and multiple system atrophy, in relation to clinical disease severity, using the nucleic acid-linked immuno-sandwich assay central nervous system disease panel for biomarker quantification. We used the nucleic acid-linked immuno-sandwich assay CNS disease panel to test plasma from 161 Parkinson's disease patients collected at three time points (0, 48, 96 weeks) and serum from 43 multiple system atrophy patients at three time points (0, 24, 48 weeks) and compared results to paired plasma and serum samples collected from (n = 39) age-matched healthy control individuals at a single time point. We also tested paired CSF samples collected on two occasions, separated by 96 weeks from a subgroup of Parkinson's disease participants (n = 51) and after an interval of 48 weeks in a subgroup of multiple system atrophy participants (n = 23). All samples were taken contemporaneously with objective clinical assessments of disease severity. Biomarker comparisons were made across disease status and in relation to disease severity using linear modelling. Multiple proteins showed significantly different quantitative levels (false discovery rate-corrected P value < 0.05) between peripheral samples from Parkinson's disease and healthy controls and multiple system atrophy and healthy controls. For Parkinson's disease, we identified three key classes of proteins that showed significant differences between Parkinson's disease and controls: (i) amyloidogenic proteins, specifically, oligomeric alpha-synuclein was significantly higher in Parkinson's disease compared to controls. A number of other aggregating proteins also exhibited differences. (ii) Metabolic pathways, including the adipokine (chemokine-like protein TAFA-5), were associated with Parkinson's disease diagnosis, and (iii) inflammatory pathways (interleukin-7) were associated with Parkinson's disease diagnosis. Importantly, some of these same proteins were significantly associated with Parkinson's disease severity including oligomeric and phosphorylated forms of alpha-synuclein and insulin-like growth factor-1 receptor. We also confirmed as expected that neurofilament light levels strongly distinguish multiple system atrophy patients from healthy controls, while also demonstrating that serum inflammatory proteins (interleukin-6) as well as the phosphorylated alpha-synuclein ratio are strongly associated with multiple system atrophy severity. These results from the nucleic acid-linked immuno-sandwich assay multiplex platform provide additional insights into the complex pathogenetic mechanisms associated with alpha-synucleinopathy related neurodegeneration. Individual protein levels or the combination of multiple protein candidates may usefully serve as diagnostic biomarkers, or as biomarkers for disease progression in trials of potential disease-modifying interventions.

BACKGROUND:Seasonal influenza causes significant morbidity and mortality annually. In 2025, the genetically divergent A/H3N2 K subclade (J.2.4.1) emerged with substantial haemagglutinin mutations. However, despite suggested antigenic escape, UK vaccine effectiveness estimates and epidemiological data demonstrated a relatively normal influenza season across 2025-26. We examined neutralising antibody responses in human cohorts to investigate existing and vaccine-induced immunity to K clade viruses. METHODS:We characterised the antigenic relationships of a selection of A/H3N2 viruses spanning recent evolution including a subclade K virus using antigenic cartography, followed by serological antibody profiling of four human cohorts from the United Kingdom and Norway using microneutralisation (MN) and haemagglutination inhibition (HAI) assays. FINDINGS:Antigenic cartography from single-infection ferret antisera suggests significant antigenic drift from the vaccine strains. MN and HAI titres from 243 individuals across 4 human cohorts (ages 1-105 years) were measured for comparison. The 2025/26 Northern Hemisphere seasonal inactivated egg-derived trivalent influenza vaccine (eTIV, with J.2 A/H3N2) significantly boosted MN and HAI titres against all A/H3N2 viruses tested, including a subclade K virus (p < 0.001). Furthermore, serological profiles of cohorts stratified by age groups (≤5, >5-≤15, >20-≤25, >25-<60, and ≥60) showed pre-existing reactivity against the emergent subclade K viruses, with minimal inter-age variation, suggesting there was not an immunity gap within particular age groups. INTERPRETATION:The 2025/26 seasonal inactivated eTIV vaccine effectively boosted neutralising titres despite substantial genetic and antigenic drift. Human serological profiling should be included in risk assessments and continued surveillance. FUNDING:The Francis Crick Institute with core funding from Cancer Research UK, UK Medical Research Council, and Wellcome Trust; UK Research and Innovation and UK Medical Research Council; National Institute for Health Research University College London Hospitals Biomedical Research Centre; UK Health Security Agency; Norwegian Institute of Public Health.

Glucagon-like peptide-1 receptor agonists (GLP-1RAs), established therapies for type 2 diabetes and obesity, are increasingly recognized fortheir potential in neurodegenerative diseases. Preclinical studies across diverse neurodegenerative conditions consistently demonstrate neuroprotective effects of GLP-1RAs, including reduced protein aggregation, enhanced autophagy, improved mitochondrial function, suppression of neuroinflammation, and preservation of synaptic integrity. Epidemiological analyses further suggest reduced incidence of dementia, Parkinson disease, and multiple sclerosis among long-term GLP-1RA users. Early human trials provide signals of target engagement, such as preserved cerebral glucose metabolism, altered inflammatory biomarkers, and slowed brain atrophy, although clinical outcomes to date remain mixed and trials in rarer disorders are sparse. Translation is constrained by uncertainty around optimal molecule choice, CNS penetrance, tolerability, adherence, and heterogeneity of response. Furthermore, next-generation dual and triple agonists may offer enhanced efficacy but remain untested in neurodegeneration. Conceptually, GLP-1RAs share pleiotropic effects with exercise - one of the few interventions with proven disease-modifying potential - by enhancing insulin signaling, stabilizing mitochondria, reducing inflammation, and promoting synaptic plasticity. This overlap highlights their promise as "pharmacological analogues of exercise," and underscores the need for biomarker-driven, disease-specific trials to establish whether GLP-1RAs can deliver durable disease modification across the spectrum of neurodegenerative diseases.

Parkinson's disease-associated proteins PINK1 and Parkin collaboratively regulate stress-induced mitophagy. While in vitro human neuronal cultures are valuable for studying the roles of PINK1 and Parkin in a disease-relevant context, the impact of culture conditions on these processes remains largely underexplored. Here, it is shown that human induced neurons (iNeurons) cultured in N2B27 and BrainPhys medium exhibit distinct PINK1-Parkin-dependent mitophagy phenotypes. Specifically, BrainPhys-cultured iNeurons show greater resistance to PINK1-dependent mitophagy initiation, linked to a reduction in glucose availability and reduced PINK1 protein availabilities, leading to decreases in stress-induced and basal mitophagy fluxes. These findings highlight the critical impact of culture conditions on mitophagy dynamics and emphasize the need to account for media-specific differences when using in vitro models to investigate mitophagy mechanisms in human neurons.

Dissecting biological pathways highlighted by Mendelian gene discovery has provided critical insights into the pathogenesis of Parkinson's disease (PD) and neurodegeneration. This approach ultimately catalyzes the identification of potential biomarkers and therapeutic targets. Here we identify PSMF1 as a gene implicated in parkinsonism and childhood neurodegeneration. We find that biallelic PSMF1 missense and loss-of-function variants co-segregate with phenotypes from early-onset PD to perinatal lethality with neurological manifestations across 18 pedigrees with 25 affected subjects, showing clear genotype-phenotype correlation. PSMF1 encodes the proteasome regulator PSMF1/hPI31, a highly conserved, ubiquitously expressed partner of the 20S proteasome and neurodegeneration-associated F-box-O 7 and valosin-containing proteins. We demonstrate that PSMF1 variants may affect proteasomal abundance and assembly, and are associated with alterations of mitochondrial membrane potential, respiration, dynamics and mitophagy in patient-derived fibroblasts. Furthermore, Drosophila and mouse models of PI31 loss of function exhibit age-dependent motor impairment, as well as brain-wide mitochondrial membrane depolarization and dopaminergic neurodegeneration in aged flies, and diffuse gliosis in mice. Collectively, our findings unequivocally link defective PSMF1/hPI31 to early-onset parkinsonism and neurodegeneration, and suggest proteasomal and mitochondrial dysfunction as pathogenic contributors.

Artificial intelligence is transforming our capability to solve biological challenges. In dimensionality bottleneck regimes exacerbated by high-dimensional biological data, Neural networks force distinct concepts into the lower dimensions known as superposition. Although this superposition is widely known to hinder interpretability, its impact on corrupting the geometry of latent spaces remains critically overlooked. Here, we utilized sparse autoencoders (SAEs) trained on over 100,000 multiplexed images of patient-derived Parkinson's disease and healthy neurons to resolve superposition. This approach bypasses the mathematical non-uniqueness of feature attribution by shifting to interpretable latent representation analysis. We theoretically and empirically demonstrate that superposition contaminates representational metric spaces, and thereby SAEs successfully recover geometric fidelity. By treating these geometrically purified representations as single-cell state vectors, we adapted single-cell RNA sequencing (scRNA-seq) data analysis methodologies directly to the image domain. Finally, we introduce GW-map, utilizing Gromov-Wasserstein optimal transport to align these image representations with authentic scRNA-seq data de novo. This coupling reconstructs hierarchical neuronal pathology pathways such as Calcium-AIS scaffold, without reference spatial transcriptomics, establishing a scalable foundation for spatial biology. Code is available at https://github.com/jijihihi/Bio_superposition

Parkinson’s disease (PD) is characterised by insoluble α-synuclein (αSyn) aggregates in Lewy bodies (LBs) within the substantia nigra, with cortical pathology appearing as the disease progresses. Late-stage LB deposition, cellular stress, and neuronal loss obscure disease-driving events, we therefore performed multi-regional transcriptomic and aggregate profiling in early-midstage PD brains (Braak 3–4), where cortical regions are pathologically unaffected. We report neuroimmune activation as an early PD feature, characterised by the expansion of a high- SNCA -expressing microglial state. This robust immune signature occurs prior to LB formation, but is associated with oligomeric αSyn within cortical microglia. In hiPSC-derived microglia, both endogenous αSyn oligomerisation, and exogenous oligomer uptake, trigger transcriptional reprogramming, characterised by interferon-driven inflammation, antigen presentation, and mitochondrial suppression, closely mirroring the early PD brain. These findings describe mechanisms by which αSyn oligomerisation potently initiates early neuroinflammation, highlighting a critical interplay between proteinopathy and immune activation at the earliest stages of disease. ### Competing Interest Statement The authors have declared no competing interest. Aligning Science Across Parkinson's, ASAP-000478, ASAP-000509

Hereditary ataxias are a heterogeneous group of neurogenetic conditions characterised by the clinical syndrome of progressive loss of coordination from neurodegeneration of the cerebellum. A commonality across the most prevalent ataxias is the underlying disease mechanism secondary to expansions of short tandem DNA repeats. There is currently an incomplete understanding of the pathogenic mechanisms of these repeat expansion disorders, a core feature of which revolves around RNA-dysregulation. In this study, we used both bulk and single nuclear RNA-sequencing to study post-mortem brain tissue of human donors with a range of repeat-expansion ataxias to reveal further mechanistic insights. We compared post-mortem paired cerebellar and frontal cortex tissue bulk RNA-sequencing data from 23 ataxia patients and 22 sex-, age-matched controls from two brain banks (spinocerebellar ataxia (SCA)1, SCA2, SCA6, SCA7, SCA17, Friedreich's ataxia (FRDA), and 7 cases with unknown molecular diagnoses). We analysed bulk RNA-sequencing data for transcript usage, differential and cell-type-specific expression to transcriptomically profile these diseases. We also generated single nuclear RNA-sequencing data of the cerebellum from donors with SCA1, SCA2, SCA6 and FRDA to decipher changes in cell type proportions in the disease state. Using this approach, we found that: (i) despite the commonalities in the genetics of ataxia, there were components of their transcriptional signatures which were distinct; (ii) there were extensive transcriptional changes evident not only in the cerebellum but also the frontal cortex in ataxia cases; (iii) activation of immune and inflammatory pathways, as well as involvement of non-neuronal cell types was a feature of all ataxias to a lesser or greater extent. This study provides a novel resource to understand the mechanisms of disease in ataxia. Furthermore, taken together, these results highlight immune pathways and the role of non-neuronal cell types as early and potentially important therapeutic targets. These findings provide a map of transcriptomic changes in ataxia to further understanding of the underlying pathogenesis. ### Competing Interest Statement The authors have declared no competing interest.

Parkinson's disease (PD) is an increasingly prevalent neurodegenerative disorder, largely sporadic in origin, with limited understanding of age- and region-specific lipid alterations in the human brain. Dysregulation of glycosphingolipid catabolism has been implicated in PD, yet comprehensive spatiotemporal profiling remains sparse. Here, we performed targeted lipidomics across eight anatomically distinct brain regions in post-mortem controls, mid-stage, and late-stage PD cases using high-precision tissue dissection. Each region displayed distinct lipid signatures, with several age-associated alterations-most notably in hexosylceramides, including glucosylceramide. In PD, glycosphingolipids were reduced in subcortical regions but elevated in cortical regions, particularly gangliosides, HexCer, and Hex2Cer, accompanied by increased sphingolipids and decreased phospholipids. The most pronounced mid-stage changes occurred in the putamen, where very long chain ceramide species and plasmalogen PE decreased, then normalising in late-stage disease. Lyso-phosphatidylcholine increased progressively throughout PD progression. Integrating proteomic data, we observed sphingomyelin levels associated with PD-related proteins, while dysregulated mitochondrial function correlated with antioxidant plasmalogens, long-chain ceramides, lyso-phosphatidylcholine, and HexCer in the putamen. These findings highlight region- and stage-specific lipid alterations in PD and their potential convergence with mitochondrial dysfunction.

We tested cross-neutralization against highly pathogenic avian influenza A(H5N1) virus in adults vaccinated with 2021-2023 seasonal quadrivalent influenza vaccine in the United Kingdom. Seasonal quadrivalent influenza vaccines are unlikely to protect vulnerable persons against severe H5N1 disease during widespread transmission. Enhanced measures are needed to protect vulnerable people from H5N1 virus infection.

BACKGROUND:There are currently no disease-modifying therapies (DMTs) registered for Parkinson's disease (PD). The Edmond J. Safra Accelerating Clinical Trials in Parkinson Disease (EJS ACT-PD) initiative will expedite clinical assessment of putative DMTs through a multi-arm multistage (MAMS) trial, testing several treatments against a common placebo arm and replacing unsuccessful therapies early. OBJECTIVE:The objective of this study was to describe the treatment selection process for the EJS ACT-PD clinical trial platform. METHODS:A Treatment Selection Working Group (TSWG) identified compounds using complementary strategies, such as literature search, related initiatives (Cure Parkinson's International Linked Clinical Trials [iLCT] initiative), and expert suggestions. Compounds were classified into five mechanistic subgroups (mitochondrial, lysosomal, protein, inflammation, "other"). "Go/No-Go" criteria and a scoring system covering preclinical, pharmacological, and clinical evidence were devised. Experts scored the candidates for quantitative rankings. Dossiers adapted from iLCT documents were produced for the top-ranked compounds and in turn prioritized by the TSWG. Practical and logistical considerations from the Steering Committee (SC) guided the final decision. Patient and Public Involvement and Engagement representatives provided feedback throughout the process. RESULTS:A total of 293 interventions were identified, 52 of which passed the "Go/No-Go" criteria and were scored. Dossiers of the 14 top-ranked compounds were submitted to the SC. Telmisartan, terazosin, and ursodeoxycholic acid were selected as the initial interventions. CONCLUSIONS:Drug selection in DMT PD MAMS trials requires consideration of scientific and practical issues. We present a robust system that can inform similar initiatives. © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

OBJECTIVE:Exenatide, a glucagon-like peptide-1 (GLP-1) receptor agonist, has neuroprotective effects in preclinical models of multiple system atrophy (MSA). We investigated these effects in a proof-of-concept clinical trial. METHODS:In this single-center, randomized, open label trial, participants with MSA were randomly assigned (1:1) to receive subcutaneous injections of exenatide 2 mg weekly for 48 weeks, or as controls, followed by a 48-week washout period. The primary outcome was the Unified Multiple System Atrophy Rating Scale (UMSARS) parts I + II combined score at 48 weeks. Objective secondary outcome measures included the numbers of participants losing ambulation; scoring ≥ 3 on UMSARS part I items for falls, speech, swallowing, as well as timed walking and measures of quality of life and cognition. RESULTS:Between September 23, 2020, and May 6, 2022, 50 participants were recruited (25 in each group). At 48 weeks, UMSARS parts I + II scores had worsened by 6.1 points (95% confidence interval [CI] = 3.0 to 9.3, SD = 6.9) in the exenatide group and by 13 3 points (95% CI = 9.2 to 17.3, SD = 9.4) in the control group, an adjusted mean difference of -7.4 points (-11.3 to -3.6, p = 0.0003). There were no statistically significant differences at either 48 or 96 weeks in the secondary outcome measures. Biomarker analysis of neurofilament light chain and cerebral spinal fluid (CSF) alpha-synuclein oligomer load, sensor-derived gait measures, and imaging findings were also similar between groups. INTERPRETATION:Exenatide was associated with positive effects on participant-reported symptoms and clinician-rated MSA severity. In contrast, none of the objective comparisons differed according to randomization. Given the open label trial design, the discrepancy between the primary outcome and the objective measures may be explicable as placebo effects/observer bias. ANN NEUROL 2025;98:991-1003.

While mRNA splicing dysregulation is a well-established contributor to neurodegeneration in disorders such as amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), its role in Parkinson’s disease (PD) remains underexplored. Here, we analyse transcriptomic data from >500 post-mortem human brain samples from individuals with and without PD to show that splicing alterations are frequently detected. Differentially spliced genes were significantly more enriched for those causally-implicated in both PD and ALS than genes that were differentially expressed. Furthermore, we observed a strong association between these splicing alterations and dysfunction of the RNA-binding protein (RBP), TAR DNA-binding protein 43 (TDP-43). Strikingly, genes and exon junctions affected by TDP-43 knockdown overlapped significantly with those dysregulated across brain regions in PD. In brains from individuals with the LRRK2 c.6055G>A (p.G2019S) mutation, the most common genetic cause of PD, we also observed significant enrichment of TDP-43-dependent splicing changes. This finding was corroborated in human pluripotent stem cell-derived midbrain dopaminergic neurons and a LRRK2 p.G2019S knock-in mouse model, where reduced nuclear TDP-43 levels evidenced the well-recognised loss-of-function mechanism contributing to splicing dysregulation. By leveraging our RNA-based analyses we predicted TDP-43-dependent novel peptide sequences and validated their existence within human LRRK2 mutation mDNs, while also demonstrating an overall loss of protein and mRNA expression in mis-spliced genes. Collectively, our findings reveal that PD is marked by extensive splicing dysregulation dependent on TDP-43, making TDP-43 a promising new therapeutic target in PD. ### Competing Interest Statement The authors have declared no competing interest. Aligning Science Across Parkinson's, https://ror.org/03zj4c476, ASAP-000478, ASAP-000509, ASAP-000486

Dissecting biological pathways highlighted by Mendelian gene discovery has provided critical insights into the pathogenesis of Parkinson's disease (PD) and neurodegeneration. This approach ultimately catalyzes the identification of potential biomarkers and therapeutic targets. Here, we identify PSMF1 as a novel gene implicated in parkinsonism and childhood neurodegeneration. We find that biallelic PSMF1 missense and loss-of-function variants co-segregate with phenotypes from early-onset PD to perinatal lethality with neurological manifestations across 17 pedigrees with 24 affected subjects, showing clear genotype-phenotype correlation. PSMF1 encodes the proteasome regulator PSMF1/PI31, a highly conserved, ubiquitously expressed partner of the 20S proteasome and neurodegeneration-associated F-box-O 7 and valosin-containing proteins. We demonstrate that PSMF1 variants impair mitochondrial membrane potential, dynamics and mitophagy, and may affect proteasomal abundance and assembly in patient-derived fibroblasts. Furthermore, Drosophila and mouse models of PSMF1 loss of function exhibit age-dependent motor impairment, as well as brain-wide mitochondrial membrane depolarization and dopaminergic neurodegeneration in aged flies, and diffuse gliosis in mice. Collectively, our findings unequivocally link defective PSMF1 to early-onset parkinsonism and neurodegeneration, and suggest proteasomal and mitochondrial dysfunction as mechanistic contributors.