Aquaporin 4 and its isoforms regulation ameliorate AQP4 Mis-localization-induced glymphatic dysfunction in ischemic stroke

Abstract

Introduction: The glymphatic system, a brain waste clearance pathway, is impaired during ischemic stroke-induced edema, although the underlying mechanisms remain unclear.

Objectives: This study investigates the temporal dynamics of glymphatic dysfunction post-stroke and the roles of aquaporin 4 (AQP4), its isoforms, and syntrophin alpha 1 (SNTA1) in AQP4 polarization.

Methods: Using a transient middle cerebral artery occlusion (tMCAO) mouse model, glymphatic function was assessed via cisterna magna contrast injection and magnetic resonance imaging. The AQP4 antagonist TGN-020 was administered to elucidate edema's role in glymphatic dysfunction. AQP4 isoforms viral vectors and SNTA1 modulation were used to study AQP4 polarization and glymphatic function. Techniques included western blotting, q-PCR, immunofluorescence, TEM and behavioral tests. Transcriptomic and metabolomic analyses were performed to assess gene expression and metabolic changes.

Results: Cerebrospinal fluid (CSF) flow decreased during the hyperacute phase, recovering with edema resolution. By administering TGN-020 to reduce edema, distinct alterations in the localization of AQP4 were observed. Specifically, there was a notable increase in AQP4 localization within the astrocyte end-feet. Consequently, CSF inflow and interstitial fluid (ISF) drainage were restored. Transcriptomic sequencing was used to analyze ubiquitination-related channels in tMCAO mice. Metabolic sequencing showed that TGN-020 therapy protected the metabolic stability. Our findings highlight the critical role of AQP4 isoforms in the polarized distribution of AQP4. The upregulation of the AQP4-M1 isoform exacerbated edema and motor dysfunction, whereas the AQP4-M23 isoform corrected the mis-localization of AQP4. Inhibition of AQP4 not only restored the polarized integrity of AQP4 in astrocyte end-feet but also alleviated the metabolic disruptions caused by tMCAO. Furthermore, overexpression of SNTA1 enhanced AQP4 polarity by modulating the expression of AQP4 isoforms.

Conclusion: Cerebral edema disrupts AQP4 localization and glymphatic function following stroke. TGN-020 modulates AQP4 polarization through regulation of AQP4 isoforms and restores glymphatic dysfunction. AQP4-M23 isoform emerges as a key regulator of AQP4 polarization, providing new insights into ischemic stroke pathophysiology.

Keywords: AQP4 isoforms; Aquaporin 4; Glymphatic system; Ischemic stroke; TGN-020.

Graphical abstract

Fig. 1

Dynamic Evaluation of tMCAO Brain Injury and Glymphatic System Function in Mice Over Time. A Overview of the experimental process and group allocations. B Representative T2WI and DWI scans of the sham group and ischemic stroke group at various time points. C Ischemic lesion and brain swelling volumes in sham and tMCAO group at indicated time points (n = 8 mice per group). D Representative Western blots of AQP4 protein expression. E Quantification of relative AQP4 protein expression normalized (n = 8 mice per group). F Representative sagittal MRI images showing the flow of the contrast agent Gd-DTPA. G-I Quantitative analysis of signal-to-noise ratio variations and AUC in three ROIs across a 210-minute DCE Sequence (n = 4 mice per group). All data are shown as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. sham;^P < 0.05,^^P < 0.05,^^^P < 0.001,^^^^P < 0.0001 vs. tMCAO 2 h; #P < 0.05, ##P < 0.01 vs. tMCAO 1 d; &&&P < 0.001 vs. tMCAO 3 d). All data were compared by one-way ANOVA with Tukey’s post hoc test.

Fig. 2

TGN-020 Improves Brain Injury and Glymphatic System Dysfunction in tMCAO Mice. A Overview of the second part of the experimental process and group allocations. B Representative T2WI and DWI images of three distinct mouse groups. C Ischemic lesion volumes and brain swelling volumes in each group (n = 5 mice per group). D Schematic of intracerebral injection procedure in mice. Representative fluorescence microscopy images show Ovalbumin (45-kDa, red) inflow into CSF at 30- and 60-minute. The white arrow indicates the residual tracer deposition. Scale bar: 1000 μm. E Quantification of Ovalbumin inflow at 30- and 60-minute in brain sections (n = 4 mice per group). F Brain slices from each group show dextran (3-kDa, green) and ovalbumin (45-kDa, yellow) inflow into the brain cistern. Scale bar: 1000 μm. G Quantification of Glucan and Ovalbumin inflow in brain sections from each group (n = 4 mice per group). H Visual depiction of interstitial drainage (highlighted in red) following the introduction of the Ovalbumin tracer into the infarcted side. Scale bar: 1000 μm. I Quantification of the residual Ovalbumin tracer-covered area fraction in each group (n = 5 mice per group). All data are shown as mean ± SD. *P < 0.05, ****P < 0.0001 vs. sham; #P < 0.05, ###P < 0.001, ####P < 0.0001 vs. tMCAO + vehicle. One-way ANOVA with Tukey’s multiple comparisons test was performed in (C) and (I). Two-way repeated-measures ANOVA with Tukey’s post hoc test in (E) and (G).

Fig. 3

TGN-020 Improves the Structure of Intracerebral Transport in the Glymphatic and Meningeal Lymphatic Systems. A Comparison of GFAP-positive areas (green) and AQP4 polarization (red) in the peri-infarct cortex. Scale bar, 50 μm. B Quantification of AQP4 polarization across the three groups (n = 6 mice per group). C-E Representative Western blot bands and densitometric quantification of AQP4-M1 and AQP4-M23 in the peri-infarction area (n = 6 per group). F Perivascular space structure observed by HE staining. TEM images showing perivascular astrocyte end-feet (green) in each group. H Representative images of meningeal lymphatic vessels stained with LYVE1. The enlarged view shows meningeal lymphatic vessels in the cortical subarachnoid space (COS) region. H-I Quantification of LYVE1 distribution and vessel diameter in the three groups (n = 6 mice per group). All data are shown as mean ± SD. *P < 0.05, ****P < 0.0001 vs. sham; ##P < 0.01 vs. tMCAO + vehicle. ALL data were compared by one-way ANOVA with Tukey’s post hoc test. AC, astrocytes; drAC, detached retracted astrocytes; EC, endothelial cell; RBC, red blood cell.

Fig. 4

Effects of TGN-020 on cognitive dysfunction and metabolic pattern in tMCAO mice. A Representative image of swim traces from the Morris water maze test. B-D Data from the MWM test analyzing spatial learning and memory abilities of tMCAO mice (n = 10 mice per group). E Representative thermal imaging from the probe trial in the novel object recognition test. F Relative cognitive index with the novel object analyzed to evaluate memory ability following tMCAO (n = 10 mice per group). G Duration and speed in the rotarod test used to assess motor function in each group (n = 10 mice per group). H Transcriptomic changes in tMCAO mice. Red indicates down-regulated genes, blue dots indicate up-regulated genes, and gray dots indicate no significant changes. There was no significant change in the expression of AQP4 and SNTA1. I GSEA shows the ubiquitin-activity pathway. J Workflow for metabolomics analysis. K Volcano plot comparing differential metabolites between the tMCAO + vehicle and tMCAO + TGN-020 groups. L Pie chart depicting the proportions of different metabolite categories between the tMCAO + vehicle and tMCAO + TGN-020 groups. M Heatmap of the top 20 differentially expressed metabolites. N Analysis of affected metabolic pathways in comparisons between the tMCAO + vehicle and tMCAO + TGN-020 groups. All data are shown as mean ± SD. ****P < 0.0001 vs. sham; #P < 0.05, ##P < 0.01, ####P < 0.0001 vs. tMCAO + vehicle. One-way ANOVA with Tukey’s multiple comparisons test was performed in (B), (C) and (F). Two-way repeated-measures ANOVA with Tukey’s post hoc test in (D).

Fig. 5

Different AQP4 Isoforms Affect Brain Injury and Glymphatic Function in tMCAO Mice. A Overview of the third part of the experimental process and group allocations. B Relative mRNA expression levels of two isoforms of AQP4 in the peri-infarction area after tMCAO (n = 6 mice per group). C Representative T2 and DWI images of the three distinct groups of mice. D Ischemic lesion volumes and brain swelling volumes in each group (n = 6 mice per group). E Description of brain injection techniques and representative brain slice staining from three groups showing dextran (3-kDa, green) inflow into the brain cistern (above) and parenchyma (down) at 30 min. Scale bar: 1000 μm. F Quantification of Glucan inflow into the cistern (n = 3 mice per group). G Quantification of the residual Glucan in each group (n = 3 mice per group). H Comparison of GFAP-positive areas (green), AQP4 (red), and CD31 (grey) in the peri-infarct cortex. Scale bar: 100 μm. I Quantification of AQP4 polarization across the three groups (n = 4 mice per group). J Representative thermal imaging from the probe trial in the NOR test. K Relative cognitive index with the novel object analyzed to evaluate memory ability (n = 10 mice per group). L Duration and speed in the rotarod test used to assess motor function in each group (n = 10 mice per group). All data are shown as mean ± SD. *P < 0.05, **P < 0.01, ****P < 0.01 vs. tMCAO + AAV-CTRL; #P < 0.05, ##P < 0.01, ####P < 0.0001 vs. tMCAO + AAV-AQP4-M1. All data were compared by one-way ANOVA with Tukey’s post hoc test.

Fig. 6

SNTA1 overexpression improves glymphatic clearance in tMCAO mice. A Overview of the last part of the experimental process and group allocations. B-F Representative Western blot bands and densitometric quantification of AQP4-M23, AQP4-M1, SNTA1 and AQP4-M23/AQP4-M1 in the peri-infarction area (n = 8 per group). G Co-IP of SNTA1 with AQP4 from tMCAO mice brain precipitated by the SNTA1 antibody. H Description of brain injection techniques and representative brain slice staining showing ovalbumin (45-kDa, green) inflow into the brain cistern (left) and parenchyma (right) at 30 min. Scale bar:1000 μm. I Quantification of Ovalbumin inflow into the cistern (n = 3 mice per group). J Quantification of residual Ovalbumin-covered area fraction (n = 3 mice per group). K Comparison of SNTA1(green), AQP4 polarization (red), and GFAP (grey) in the peri-infarct cortex. Scale bar: 200 μm. L Quantification of AQP4 polarization across the four groups (n = 3 mice per group). All data are shown as mean ± SD. **P < 0.01, vs. tMCAO + shCTRL; #P < 0.05, ##P < 0.01 vs tMCAO + AAV-CTRL. All data were compared by one-way ANOVA with Tukey’s post hoc test.