Nanoscale flow cytometry-based quantification of blood-based extracellular vesicle biomarkers distinguishes MCI and Alzheimer's disease

Abstract

Introduction: Accurate testing for Alzheimer's disease (AD) represents a crucial step for therapeutic advancement. Currently, tests are expensive and require invasive sampling or radiation exposure.

Methods: We developed a nanoscale flow cytometry (nFC)-based assay of extracellular vesicles (EVs) to screen biomarkers in plasma from mild cognitive impairment (MCI), AD, or controls.

Results: Circulating amyloid beta (Aβ), tau, phosphorylated tau (p-tau)181, p-tau231, p-tau217, p-tauS235, ubiquitin, and lysosomal-associated membrane protein 1-positive EVs distinguished AD samples. p-tau181, p-tau217, p-tauS235, and ubiquitin-positive EVs distinguished MCI samples. The most sensitive marker for AD distinction was p-tau231, with an area under the receiver operating characteristic curve (AUC) of 0.96 (sensitivity 0.95/specificity 1.0) improving to an AUC of 0.989 when combined with p-tauS235.

Discussion: This nFC-based assay accurately distinguishes MCI and AD plasma without EV isolation, offering a rapid approach requiring minute sample volumes. Incorporating nFC-based measurements in larger populations and comparison to "gold standard" biomarkers is an exciting next step for developing AD diagnostic tools.

Highlights: Extracellular vesicles represent promising biomarkers of Alzheimer's disease (AD) that can be measured in the peripheral circulation. This study demonstrates the utility of nanoscale flow cytometry for the measurement of circulating extracellular vesicles (EVs) in AD blood samples. Multiple markers including amyloid beta, tau, phosphorylated tau (p-tau)181, p-tau231, p-tau217, and p-tauS235 accurately distinguished AD samples from healthy controls. Future studies should expand blood and cerebrospinal fluid-based EV biomarker development using nanoflow cytometry approaches.

Keywords: Alzheimer's disease; biomarker; dementia; extracellular vesicle; mild cognitive impairment; nanoflow cytometry.

FIGURE 1

Nanoflow cytometry labeling of plasma EVs. A, Datagram of total plasma EV distribution by light scatter. B, Datagram of EVs labeled with a 488‐conjugated antibody against fibrillar amyloid. Red box indicates ROI of positively labeled particles. C, Particle count according to light scatter of standardized sizing beads (110 and 500 nm silicon beads, 280, 300, 500, 590, and 880 nm polystyrene beads). EVs, extracellular vesicles; LALS, long angle light scatter; ROI, region of interest; SALS, small angle light scatter

FIGURE 2

Individual markers that label significantly more EVs in Alzheimer's disease plasma samples compared to controls. Fluorescently labeled antibodies against the indicated proteins were incubated with plasma samples from controls, individuals with MCI or Alzheimer's disease (separated into mild, moderate, and severe categories). Each data point represents the average of triplicate incubations. Statistical comparisons were performed using Prism 9 (GraphPad) with Kruskal–Wallis, Dunn post hoc test for multiple comparisons, and a significance value set at P < 0.05. Aβ, amyloid beta; EVs, extracellular vesicles; LAMP1, lysosomal‐associated membrane protein 1; MCI, mild cognitive impairment; ptau, phosphorylated tau; SEM, standard error of the mean

FIGURE 3

Individual markers that label significantly more EVs in MCI and Alzheimer's disease plasma samples compared to controls. Fluorescently labeled antibodies against the indicated proteins were incubated with plasma samples from controls, individuals with MCI or Alzheimer's disease (separated into mild, moderate, and severe categories). Each data point represents the average of triplicate incubations. Statistical comparisons performed using Prism 9 (GraphPad) with Kruskal–Wallis, Dunn post hoc test for multiple comparisons, and a significance value set at P < 0.05. EVs, extracellular vesicles; MCI, mild cognitive impairment; ptau, phosphorylated tau; SEM, standard error of the mean

FIGURE 4

Discrimination of AD samples (mild and moderate) from HC samples using marker combinations. ROC curves of marker combinations using logistic regression to determine coefficients (yellow) or summation (green). ROC curves of individual markers (red and blue). Aβ, amyloid beta; AD, Alzheimer's disease; AUC, area under the receiver operating characteristic curve; HC, healthy control; ptau, phosphorylated tau; ROC, receiver operating characteristic

FIGURE 5

Discrimination of MCI from HC samples using marker combinations. ROC curves of marker combinations using logistic regression to determine coefficients (yellow) or summation (green). ROC curves of individual markers (red and blue). AUC, area under the receiver operating characteristic curve; HC, healthy control; MCI, mild cognitive impairment; ptau, phosphorylated tau; ROC, receiver operating characteristic

FIGURE 6

TEM of target markers. Transmission electron microscopy showing immunogold labeling of Aβ, ptau231, ptauS235, and tau on EVs. EVs were isolated from plasma using immunoprecipitation with beta‐III tubulin followed by size exclusion chromatography. Scale bar = 200 nm. Control was generated by omitting the primary antibody. Aβ, amyloid beta; EVs, extracellular vesicles; ptau, phosphorylated tau; TEM, transmission electron microscopy