Plasma Extracellular Vesicle Size and Concentration Are Altered in Alzheimer's Disease, Dementia With Lewy Bodies, and Frontotemporal Dementia

Abstract

Alzheimer's disease (AD), frontotemporal dementia (FTD), and dementia with Lewy bodies (DLB) are the three major neurodegenerative dementias. In this study, we provide evidence that an alteration in extracellular vesicles (EVs) release is common across the three most common neurodegenerative dementias, AD, DLB, and FTD. Specifically, we analyzed plasma EVs in three groups of patients affected by AD, DLB, and FTD, and we found a significant reduction in EVs concentration and larger EVs size in all patient groups. We then investigated whether the loss of neurotrophic factors is also a common pathogenic mechanism among FTD, DLB, and AD, and if levels of neurotrophic factors might affect EVs release. Plasma levels of progranulin and cystatin C (CysC) were partially altered; however, taking together all variables significantly associated with the diagnostic groups only EVs size and concentration were able to distinguish patients from controls. The diagnostic performance of these two EVs parameters together (ratio) was high, with a sensitivity of 83.3% and a specificity of 86.7%, able to distinguish patients from controls but not to differentiate the different forms of dementias. Among the candidate neurotrophic factors, only CysC levels were associated with EVs concentration. Our study suggests that an alteration in the intercellular communication mediated by EVs might be a common molecular pathway underlying neurodegenerative dementias. The identification of shared disease mechanisms is of pivotal importance to develop treatments to delay disease progression. To this aim, further studies investigating plasma EVs size and concentration as early biomarkers of dementia are required.

Keywords: Alzheimer’s disease; biomarkers; dementia with Lewy bodies; extracellular vesicles; frontotemporal dementia; nanoparticle tracking analysis; plasma.

FIGURE 1

EVs concentration and size in patient and control groups. (A) Representative spectra from NTA of CTRL (yellow), AD (green), DLB (orange), and FTD (red) plasma EVs. (B) Quantification of EVs concentration with NTA in CTRL, AD, DLB, and FTD plasma samples. A statistically significant decrease in EVs concentration was observed in the three pathological groups compared to controls. (C) Representation of EVs size measured with NTA in CTRL, AD, DLB, and FTD plasma samples. An increase in size was observed in AD, DLB, and FTD compared to CTRL samples. Average ± SEM; **p < 0.01, ***p < 0.001, one-way ANOVA with Bonferroni’s post test.

FIGURE 2

Classification tree. Subjects are classified on the basis of the most predictive variables, EVs size and concentration, among all the ones which resulted to be significantly associated with subject groups. CTRL, controls; PTS, patients; AverageSize, EVs size; AverageConcentration, EVs/ml.

FIGURE 3

(A) ROC curves for EVs concentration/size. The ratio of EVs concentration/size was used to evaluate the discrimination of PTS from CTRL; AUC CTRL vs. PTS 0.86 (black); CTRL vs. AD 0.84 (red); CTRL vs. DLB 0.89 (green); CTRL vs. FTD 0.84 (orange); AUC comparison with DeLong test, p > 0.436. (B) CysC levels and EVs concentration are negatively/positively associated with CTRL/PTS, respectively (std beta = –0.28; std beta = 0.31 respectively). CTRL, controls (green); PTS, patients (red).