GFAP as Astrocyte-Derived Extracellular Vesicle Cargo in Acute Ischemic Stroke Patients-A Pilot Study

Abstract

The utility of serum glial fibrillary acidic protein (GFAP) in acute ischemic stroke (AIS) has been extensively studied in recent years. Here, we aimed to assess its potential role as a cargo protein of extracellular vesicles (EVs) secreted by astrocytes (ADEVs) in response to brain ischemia. Plasma samples from eighteen AIS patients at 24 h (D1), 7 days (D7), and one month (M1) post-symptoms onset, and nine age, sex, and cardiovascular risk factor-matched healthy controls were obtained to isolate EVs using the Exoquick ULTRA EV kit. Subsets of presumed ADEVs were identified further by the expression of the glutamate aspartate transporter (GLAST) as a specific marker of astrocytes with the Basic Exo-Flow Capture kit. Western blotting has tested the presence of GFAP in ADEV cargo. Post-stroke ADEV GFAP levels were elevated at D1 and D7 but not M1 compared to controls (p = 0.007, p = 0.019, and p = 0.344, respectively). Significant differences were highlighted in ADEV GFAP content at the three time points studied (n = 12, p = 0.027) and between D1 and M1 (z = 2.65, p = 0.023). A positive correlation was observed between the modified Rankin Scale (mRS) at D7 and ADEV GFAP at D1 (r = 0.58, p = 0.010) and D7 (r = 0.57, p = 0.013), respectively. ADEV GFAP may dynamically reflect changes during the first month post-ischemia. Profiling ADEVs from peripheral blood could provide a new way to assess the central nervous system pathology.

Keywords: acute ischemic stroke; astrocyte-derived extracellular vesicles; glial fibrillary acidic protein; western blotting.

Figure 1

Bead flow separation data for the tetraspanin captured antibodies coupled with Exo-FITC staining. The first column (I) depicts beads with no captured extracellular vesicles (EVs), while the second column (II) depicts beads with captured EVs. Only singlets were considered for analyses (Ia,IIa). Plots of forward scatter (FSC) versus (vs.) fluorescein isothiocyanate (FITC) intensity showed that 100% of the particles were FITC-negative in the no-EVs control (Ib). In comparison, 100% of the particles were FITC-positive in the EVs-containing sample (IIb). A histogram of the fluorescence distribution is also shown for both types of beads (Ic,IIc).

Figure 2

Bead flow separation data for the glutamate aspartate transporter (GLAST) captured antibodies coupled with Exo-FITC staining. The first column (I) presents beads with no captured EVs, while the second column (II) presents beads with captured EVs. Only singlets were considered for analyses (Ia,IIa). FSC vs. FITC intensity plots showed that 100% of the particles were FITC-negative in the no-EVs control (Ib), while 100% were FITC-positive in the EVs-containing sample (IIb). A histogram of the fluorescence distribution is also shown for both beads-types (Ic,IIc).

Figure 3

Representative blots of total-(TEV) and astrocyte-derived EV (ADEV) aliquots of three AIS patients (a–c) and one healthy control (a) probed with the anti-GFAP antibody are shown. Approximate molecular weight markers in kilodaltons (kDa) are labeled adjacently on the left. GFAP protein bands were observed with the indicated antibody: full-length GFAP and additional GFAP bands near 25 kDa molecular weight were detected (a–c). Bands near 37 kDa molecular weight detected in TEVs of a few patients (c) are also shown.

Figure 4

Full-length GFAP band intensities in TEVs (a) and ADEVs (b) of AIS patients and healthy controls (HC). Data are represented as individual value boxplots with median and interquartile range (IQR) (Mann–Whitney U test, ns—not significant).

Figure 5

Full-length GFAP band intensities in TEVs (a) and ADEVs (b) during the patients’ follow-up: D1, D7, and M1. Data are represented as individual value boxplots with median and IQR (Friedman’s ANOVA, Dunn’s post hoc).

Figure 6

Flowchart of patient enrollment and follow-up (MCA—middle cerebral artery, CT—computer tomography).