Differential Effects of APOE Genotype on MicroRNA Cargo of Cerebrospinal Fluid Extracellular Vesicles in Females With Alzheimer’s Disease Compared to Males

Abstract

Multiple biological factors, including age, sex, and genetics, influence Alzheimer’s disease (AD) risk. Of the 6.2 million Americans living with Alzheimer’s dementia in 2021, 3.8 million are women and 2.4 million are men. The strongest genetic risk factor for sporadic AD is apolipoprotein E-e4 (APOE-e4). Female APOE-e4 carriers develop AD more frequently than age-matched males and have more brain atrophy and memory loss. Consequently, biomarkers that are sensitive to biological risk factors may improve AD diagnostics and may provide insight into underlying mechanistic changes that could drive disease progression. Here, we have assessed the effects of sex and APOE-e4 on the miRNA cargo of cerebrospinal fluid (CSF) extracellular vesicles (EVs) in AD. We used ultrafiltration (UF) combined with size exclusion chromatography (SEC) to enrich CSF EVs (e.g., Flotillin+). CSF EVs were isolated from female and male AD or controls (CTLs) that were either APOE-e3,4 or -e3,3 positive (n = 7/group, 56 total). MiRNA expression levels were quantified using a custom TaqMan™ array that assayed 190 miRNAs previously found in CSF, including 25 miRNAs that we previously validated as candidate AD biomarkers. We identified changes in the EV miRNA cargo that were affected by both AD and sex. In total, four miRNAs (miR-16-5p, -331-3p, -409-3p, and -454-3p) were significantly increased in AD vs. CTL, independent of sex and APOE-e4 status. Pathway analysis of the predicted gene targets of these four miRNAs with identified pathways was highly relevant to neurodegeneration (e.g., senescence and autophagy). There were also three miRNAs (miR-146b-5p, -150-5p, and -342-3p) that were significantly increased in females vs. males, independent of disease state and APOE-e4 status. We then performed a statistical analysis to assess the effect of APOE genotype in AD within each sex and found that APOE-e4 status affects different subsets of CSF EV miRNAs in females vs. males. Together, this study demonstrates the complexity of the biological factors associated with AD risk and the impact on EV miRNAs, which may contribute to AD pathophysiology.

Recommendations for reproducibility of cerebrospinal fluid extracellular vesicle studies

“…Separation methods aimed to increase EV purity have largely been assessed using higher input volumes of CSF (≥500 μL). In SEC, the resin type and length/volume ratio of the column affects the yield and purity of CSF EVs (Sandau et al., 2022; Ter‐Ovanesyan et al., 2021). The sample volume, flow rate, and particle concentration also impacts separation.…”

“…A limited number of studies have systematically compared the efficacy of various separation and concentration methods for use with CSF EVs (Akers et al., 2015; Hayashi et al., 2020; Krušić Alić et al., 2022; Muraoka, Lin, et al., 2020; Sandau et al., 2022; Saugstad et al., 2017; Sjoqvist et al., 2020; Soares Martins et al., 2018; Ter‐Ovanesyan et al., 2021; Thompson et al., 2018; Wang et al., 2019). Among them, UC, SEC and affinity isolation were successfully applied to profiling CSF EVs by liquid chromatography‐mass spectrometry (LC‐MS) proteomics or proximal extension assay protein panels (Muraoka, Lin, et al., 2020; Sjoqvist et al., 2020), while use of a membrane affinity‐based kit did not meet the quality assessment (Sjoqvist et al., 2020).…”

“…Importantly, measurement outcomes can be dependent on the sensitivity and specificity of the method as well as the CSF collection site and other biological factors (e.g., age) (Akers et al., 2016; Coumans et al., 2017; Saugstad et al., 2017; Tietje et al., 2014; van der Pol et al., 2014). It is also important to note that CSF contains a smaller number of nanoparticles/EVs compared with plasma and serum (Sandau et al., 2020, 2022) which presents a challenge for downstream analysis (Brennan et al., 2020; Muraoka, Lin, et al., 2020). Each of these factors should be taken into consideration when designing experiments, interpreting data, and reporting methods and results.…”

“…However, in vitro studies have shown that astrocyte‐ and neuron‐derived EVs include ApoE (Nikitidou et al., 2017; Zheng et al., 2018), which is increased in secreted EVs following Aβ 42 treatment (Nikitidou et al., 2017). ApoE is associated with human CSF EV preparations in normal donors (Sandau et al., 2022) and in relapsing remitting multiple sclerosis patients (Welton et al., 2017). However, studies to validate the localisation of proteins in CSF EVs (cargo, corona and/or contaminant) using techniques that resolve single EVs (e.g., immunoEM, Surface‐enhanced Raman spectroscopy (Koster et al., 2021), and super‐resolution microscopy (Frigerio et al., 2022; McNamara et al., 2022)), are needed.…”

“…But these miRNAs may not be recommended for studies focused on younger participants, or those with different CNS diseases. To identify miRNAs that are highly abundant in control CSF, we compared the expression profiles from CSF EV miRNAs assayed by qPCR (Sandau et al., 2022) and by RNA‐seq (Alsop et al., 2022) and found miR‐204‐5p, 30c‐5p and 150‐5p in the top 25 for both datasets. Thus, these miRNAs may serve as positive controls for CSF miRNA studies regardless of assay parameters.…”