Reactive astrocyte-derived exosomes enhance intracranial lymphatic drainage in mice after intracranial hemorrhage

Abstract

Background: After intracranial hemorrhage (ICH), the formation of primary hematoma foci leads to the development of secondary brain injury factors such as perihematomal edema (PHE) and accumulation of toxic metabolites, which severely affect the survival and prognosis of patients. The intracerebral lymphatic system, proposed by Jeffrey J. Iliff et al., plays an important role in central nervous system (CNS) fluid homeostasis and waste removal, while reactive astrocyte-derived exosomes have shown therapeutic potential in CNS disorders. Our study focuses on the effects of hemin-treated reactive astrocyte-derived exosomes on the functional integrity of the glymphatic system (GLS) after ICH and their potential mechanism of action in repairing brain injury.

Methods: Hemin, an iron-rich porphyrin compound, was used to construct the in vitro model of ICH. Primary astrocytes were treated with complete medium supplemented with different concentrations of hemin to obtain exosomes secreted by them, and mice with ICH induced by the collagenase method were intervened by intranasal administration. Solute clearance efficiency was assessed by intracranial injection of cerebrospinal fluid tracers and fluorescent magnetic beads. Immunofluorescence analysis of Aquaporin 4 (AQP4) polarization and astrocyte proliferation. Magnetic Resonance Imaging was used to visualize and quantify the volume of hematoma foci and PHE, and Western Blot was used to analyze the accumulation of toxic metabolites, while neuronal apoptosis was detected by a combination of TUNEL assay apoptosis detection kit and Nissl staining, and their functional status was analyzed. Gait analysis software was used to detect functional recovery of the affected limb in mice.

Results: Exosomes from hemin treated astrocytes facilitated the recovery of AQP4 polarization and attenuated astrocyte proliferation around hematoma foci in mice with ICH, thereby promoting the recovery of the GLS. Meanwhile, exosomes from hemin treated astrocytes reduced PHE and toxic protein accumulation, decreased apoptosis of cortical neurons on the affected side, and facilitated recovery of motor function of the affected limb, and these effects were blocked by TGN020, an AQP4-specific inhibitor.

Conclusions: Exosomes from hemin treated astrocytes attenuated secondary brain injury and neurological deficits in mice with ICH by promoting the repair of GLS injury.

Keywords: Astrocyte-derived exosomes; Glymphatic system; Intracranial hemorrhage.

Fig. 1

Activation of primary astrocytes with hemin and characterization of astrocyte-derived exosomes. (A) Morphology of primary astrocytes under bright field microscopy. (B, C) Western Blot quantification of GFAP relative expression after different concentrations of hemin treatment of primary astrocytes. (D) CCK-8 detection of the viability of primary astrocytes under different concentrations of hemin treatment. (E) The size distribution of exosomes secreted by astrocytes was detected by NTA, and the graph shows the average value of three experiments after 500-fold dilution of the stock solution. (F) Morphological features of exosomes under TEM. (G) Western Blot analysis of the expression of exosomes markers Alix, CD9, HSP70, and TSG101. (H) Distribution of exosomes around hematoma foci in mice after exosomes administration by nasal feeding after ICH. All data are presented as mean ± SD (n = 4 per group). p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 2

H-Exo promotes AQP4 polarization and reduces astrocyte proliferation in the acute phase of ICH. (A) Immunofluorescence staining of AQP4 with GFAP around hematoma foci. (B) Quantitative analysis of the percentage area of the GFAP-positive region in (A). (C) Immunofluorescence staining of AQP4 with CD31 around hematoma foci. (D) Quantitative analysis of the percentage area of double-positive areas of AQP4 versus CD31 in (B) to reflect AQP4 polarization. All data are presented as mean ± SD (n = 6 per group). p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 3

H-Exo improves intracranial lymphatic drainage in the acute phase of ICH. (A) Representative coronal fluorescence images after basal ganglia injection of RITC-Dextran. (B) Quantitative analysis of the percentage area of RITC-Dextran-positive area in the basal ganglia injected in (A). (C) Representative coronal fluorescence images after RITC-Dextran injection in the cisterna magna. (D) Quantitative analysis of the percentage area of RITC-Dextran-positive area in the cisterna magna injected in (C). (E) Immunofluorescence images of beads injected into the cisterna magna converging on the dorsal MLV. (F) Quantitative analysis of the percentage of positive area of beads within the transverse sinus in (E). (G) Immunofluorescence images of cisterna magna injected beads draining into the largest diameter section of the DCLN. (H) Quantitative analysis of the percentage area of the bead-positive region in (G). All data are presented as mean ± SD (n = 6 per group). p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 4

H-Exo reduces intracranial accumulation of toxic metabolites and attenuates cerebral edema. (A-E) Western Blot quantification of phosphorylated tau protein expression in the hemorrhagic side hemisphere. (F) MRI T2-weighted images showing intracranial hematoma foci and PHE in mice. (G) Quantitative analysis of hematoma foci volume in (F). (H) Quantitative analysis of water content of the cerebral hemisphere on the hemorrhagic side. (I-K) Quantitative analysis of BBB-related protein expression in the hemorrhage side cerebral hemisphere by Western Blot. All data are presented as mean ± SD (n = 6 per group). p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 5

H-Exo promotes functional recovery of the affected limb in ICH-recovery mice. (A) mNss scores of mice at 1, 3, 5, 7, and 14 days after ICH. (B) Representative images of crawling footprints of mice during ICH recovery. (C) Quantitative analysis of the single-support phase of the affected limb as a percentage of the walking cycle in mice during ICH recovery. (D-G) Quantitative analysis of maximum force and maximum contact area of the affected anterior and posterior limbs after contact with the treadmill in mice recovering from ICH. All data are presented as mean ± SD (n = 8 per group). p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 6

TGN020 reverses outcome in ICH-recovery mice by inhibiting GLS function. (A) Representative coronal fluorescence images after basal ganglia injection of RITC-Dextran. (B) Quantitative analysis of the percentage area of RITC-Dextran-positive area in the basal ganglia injected in (A). (C) Representative coronal fluorescence images after RITC-Dextran injection in the cisterna magna. (D) Quantitative analysis of the percentage area of RITC-Dextran-positive area in the cisterna magna injected in (C). (E) mNss scores of mice at 1, 3, 5, 7, and 14 days after ICH. (F) Representative images of crawling footprints of mice during ICH recovery. (G) Quantitative analysis of the percentage of walking cycles during the single-support phase of the affected limb in mice during ICH recovery. (H-K) Quantitative analysis of maximum force and maximum contact area of the affected anterior and posterior limbs after contact with the treadmill in mice during ICH recovery. All data are presented as mean ± SD (n = 6–8 per group). p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 7

GLS functional status affects neuronal survival in the acute phase of ICH. (A-H) Western blot quantification of phosphorylated tau protein with BAX and BCL expression in the affected cerebral hemisphere of ICH acute phase mice. (I) Immunofluorescence staining of Tunel and Neun in the cerebral cortex around the hematoma foci. (J) Quantification of the percentage of Neun-positive areas in the Tunel and Neun double-positive areas in (I) to detect the percentage of neuronal apoptosis. (K) Nissl staining of the cerebral cortex around the hematoma foci. (L) Quantitative analysis of Nissl body expression in the cerebral cortex under high magnification field of view in (K). All data are presented as mean ± SD (n = 6 per group). p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001