Biology subjects

Lyketsos, C. G.

Publications and source records attributed to Lyketsos, C. G..

3 recordsLinked to original sources

bioRxiv2026

Proteomic identification and validation of novel neuronal EV-based markers for Alzheimers disease biomarker discovery

Extracellular vesicles (EVs) circulate in biofluids and carry tissue-specific molecular cargo, offering significant potential for the discovery of minimally invasive biomarkers. However, translation in neurodegenerative diseases has been hindered by the lack of validated neuronal EV surface markers that enable selective isolation from plasma. We hypothesized that proteomic profiling of EVs released from human induced pluripotent stem cell (hiPSC)-derived neurons would identify 1. robust Alzheimers disease (AD)-associated signatures that reflect disease pathogenesis, and 2. surface-accessible neuronal markers capable of enriching disease-relevant cargo. Neurons differentiated from AD patients and age-matched cognitively normal (CN) individuals were used to isolate EVs, which were characterized and analyzed by LC-MS proteomics in both total and membrane-enriched fractions. Proteomic profiling identified numerous dysregulated proteins, with a subset validated across independent AD datasets. We identified CNTNAP2 and STX1B as neuronal, brain-enriched EV surface proteins accessible for selective capture and confirmed their presence in EVs from post-mortem human brain, supporting them as bona fide brain-derived EV markers. Immuno-isolation of plasma EVs showed that CNTNAP2-positive EVs had a robust AD-associated increase in phosphorylated tau, identifying CNTNAP2 as a highly discriminative brain-derived EV marker and supporting its potential for blood-based AD diagnostics. Graphical AbstractPrevious studies indicate that extracellular vesicles (EVs) released from neurons carry disease-relevant cargo, yet the search for the optimal neuronal surface markers for the selective isolation of EVs pertinent to Alzheimers disease (AD) from blood is ongoing. To address this need, we performed proteomic profiling of EVs derived from hiPSC-neurons (iNEVs) of AD patients and cognitively normal individuals (CN). LC-MS analysis of whole and membrane-enriched EV fractions revealed robust protein dysregulation and identified CNTNAP2 and STX1B as surface-exposed neuronal EV proteins, confirmed in human brain-derived EVs (BDEVs). Immuno-isolation of plasma EVs demonstrated that CNTNAP2-positive EVs are enriched for phosphorylated tau in AD, whereas the previously used markers; NrCAM and ATP1A3-positive EVs showed limited discrimination. These findings establish a robust framework for proteomic discovery and nominate CNTNAP2 as a promising EV selection marker for blood-based AD diagnostics. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=146 SRC="FIGDIR/small/720870v1_ufig1.gif" ALT="Figure 1"> View larger version (48K): org.highwire.dtl.DTLVardef@754b45org.highwire.dtl.DTLVardef@a0a566org.highwire.dtl.DTLVardef@cb0134org.highwire.dtl.DTLVardef@1bc0275_HPS_FORMAT_FIGEXP M_FIG C_FIG

neuroscience↗

bioRxiv2023

Excitatory neurons derived from human induced Pluripotent Stem Cells show transcriptomic differences in Alzheimers patients from controls

The recent advances in creating pluripotent stem cells from somatic cells and differentiating them into a variety of cell types is allowing us to study them without the caveats associated with disease related changes. We have generated induced Pluripotent Stem Cells (iPSCs) from eight Alzheimers disease (AD) patients and six controls and used lentiviral delivery to differentiate them into excitatory glutamatergic neurons. We have performed RNA sequencing on these neurons and compared the Alzheimers and control transcriptomes. We find that 621 genes show differences in expression levels at adjusted p<0.05 between the case and control derived neurons. These genes show significant overlap and direction concordance with genes reported from a Single cell transcriptome study of Alzheimers patients, they contain 5 genes implicated with AD from genome wide association studies and they appear to be part of a larger functional network as indicated by an excess of interactions between them observed in the protein-protein interaction database STRING. Exploratory analysis with Uniform Manifold Approximation and Projection (UMAP) suggests distinct clusters of patients, based on gene expression, who maybe clinically different. If confirmed this finding will to contribute to precision medicine approaches to subgroup Alzheimers disease.

genetics↗

bioRxiv2018

First in-human PET study of 3 novel tau radiopharmaceuticals: RO6958948

Background[11C]RO-963, [11C]RO-643 and [18F]RO-948 (previously referred as [11C]RO6924963, [11C]RO6931643, and [18F]RO6958948, respectively) have been reported as promising PET tracers for tau imaging based on in vitro and preclinical PET data (1,2). Here we describe the first human evaluation of these novel radiotracers.\n\nMethodsAmyloid PET positive Alzheimers disease (AD) patients and young healthy subjects (YC) each received two different tau tracers. Dynamic 90 min scans were obtained after bolus injection of [11C]RO-963, [11C]RO-643 or [18F]RO-948. Arterial blood sampling was performed in 11 healthy controls (HC) and 11 AD. Regions were defined on MRI, and PET data were quantified by plasma reference graphical analysis (for VT) and target cerebellum ratio (SUVR60-90). SUVR images were also analyzed voxelwise. Five older healthy subjects (OC) each received two scans with [18F]RO-948 for evaluation of test-retest variability. Four AD subjects received a repeat [18F]RO-948 scan over about 1 year. Six additional HC (3M: 3F; 41-67y) each received one whole body dosimetry scan with [18F]RO-948.\n\nResultsIn YC, peak SUV values were observed in the temporal lobe with values of approximately 3.0 for [11C]RO-963, 1.5 for [11C]RO-643 and 3.5 for [18F]RO-948. Over all brain regions and subjects, the trend was that [18F]RO-948 had the highest peak SUV value, followed by [11C]RO-963, and then [11C]RO-643. Regional analysis of SUVR and VT for [11C]RO-643 and [18F]RO-948 clearly discriminated AD and HC groups. Compartmental modeling confirmed that [11C]RO-643 had lower brain entry than both [18F]RO-963 and [18F]RO-948, and [18F]RO-948 showed a better contrast between (predicted) areas of high vs low tau accumulation. Thus, our subsequent analysis focused on [18F]RO-948. Both voxelwise and region-based analysis of [18F]RO-948 binding in HC vs AD revealed multiple areas where AD and HC significantly differed. Of 22 high-binding regions, 13 showed significant group difference (following ANOVA, F=45, p<10-5). Voxelwise analysis also revealed a set of symmetrical clusters where AD>HC (threshold of p<0.001, cluster size k>50).\n\nConclusions[18F]RO-948 demonstrates superior characteristics to [11C]RO-643 and [18F]RO-963 for characterization of tau pathology in AD. Regional binding data and kinetic properties of RO-948 compare favorably with existing other tau PET tracers.

neuroscience↗