Ischemic stroke is a leading cause of death and disability in humans, with few pathophysiological therapies other than recanalizing occluded blood vessels (1, 2). Acutely inhibiting aquaporin-4 (AQP4) was proposed recently as a promising new pathophysiological therapy targeting central nervous system (CNS) edema post-injury (3, 4). Because water transportation through AQP4 is a passive process, depending on osmotic gradients. AQP4 contributes to the formation of cellular toxic edema at first, but it is also essential for the resolution of vasogenic edema in CNS injury. And studies revealed that AQP4 deficient animals displayed higher levels of CNS water content than control animals at a later phase of CNS injury (5). AQP4 is the most abundant aquaporins in the brain, it has a polarized distribution tendency on the astrocyte endfeet facing vessels under physiological conditions, this distribution tendency is critical for the formation and resolution of edema, and clearance of interstitial solutes in the brain (6). Commonly, methods of inhibiting AQP4 mainly include gene knockout, small interfering RNA, heavy metal ions, and small molecule inhibitors (7). Small molecule inhibitors have the potential to be applied in clinical for their security. N-(1,3,4-thiadiazol-2-yl) pyridine-3-carboxamide dihydrochloride (TGN-020) is one of them and has been proven to inhibit AQP4 in vitro and in vivo via the intracellular ubiquitin-proteasome system (8, 9).

AQP4 is implicated not only in edema formation and resolution, but also in astrocyte migration and astrogliosis (10, 11). However, the changes in peri-infarct astrocyte reactivity related to acute inhibition of AQP4 have not been clarified, which are crucial for peri-infarct tissue repair and neurological function recovery. After stroke, reactive astrogliosis and loss of perivascular AQP4 polarization occur and persist for long time in the peri-infarct area (12–14). Reactive astrogliosis is beneficial for limiting the infarct territory initially; however, its increasing dysregulation at the recovery stage accentuates inflammation and inhibits axon regeneration, thus interfering with long-term sensorimotor functional recovery (15, 16). Besides, loss of AQP4 polarization impairs the glymphatic system, a newly-discovered waste clearance system in the brain (17), causing toxic protein deposition and cognitive deficits (18, 19). Modulating reactive astrogliosis and the loss of AQP4 polarization in the peri-infarct area might be beneficial therapeutic strategies during later stages to promote neurological function recovery.

In this study, we acutely inhibited AQP4 using TGN-020 in transient middle cerebral artery occlusion (MCAO) rats, evaluated the brain edema and infarct volume at 1 and 14 days, and the peri-infarct astrogliosis extent, AQP4 expression patterns, and neurological function at 14 days after MCAO. In addition, we analyzed correlations of the AQP4 expression patterns and the ultra-high b-values apparent diffusion coefficient (ADCuh). We aimed to investigate the effect of acutely inhibiting AQP4 on peri-infarct astrocyte reactivity and subacute outcome and the feasibility of ADC to reflect the expression patterns of AQP4.

In the present study, acute inhibition of AQP4 using TGN-020 decreased the edema and infarct lesion volume 1day post-stroke, attenuated peri-infarct astrogliosis, AQP4 depolarization, and infarct lesion volume, promoting neurological recovery at 14 days post-stroke. Additionally, we found that AQP4 polarization correlated negatively with astrogliosis, and ADCuh could reflect the AQP4 polarization.

Our results showed that acute inhibition of AQP4 by TGN-020 reduced brain edema 1day post-stroke, which is consistent with previous research (24, 25). Traditionally, it is thought that post-stroke edema comprises cytotoxic edema and vasogenic edema, in which AQP4 plays inductive and counteractive roles, respectively, with edematous fluid mainly coming from blood plasma (26, 27). However, recently, researchers found that cerebrospinal fluid immediately flowed towards the brain parenchyma through the influx pathway of the glymphatic system after ischemic stroke, and an absence of AQP4 reduced the cerebrospinal fluid influx significantly (4, 28). Regardless, the traditional or newly-found mechanism of edema both suggest that acute inhibition of AQP4 could reduce post-stroke edema at the early stage (29, 30). However, perivascular AQP4 is essential for the dissipation of vasogenic edema and the glymphatic clearance of Aβ and tau (8, 9). The deficiency of AQP4 would cause the accumulation of water and neurotoxic protein in the recovery stage of CNS injury (3, 31). In this study, we further investigated the peri-infarct expression patterns of AQP4 14 days after acute inhibition of AQP4. No differences in AQP4 expression intensity were found between the TGN-020 group and the control group, but AQP4 polarization of the TGN-020 group was higher than that of the control group, in other words, the perivascular AQP4 was increased in the TGN-020 group when compared with the control group. As is shown in our study, swelling extent of the ipsilateral hemisphere has turned to normal in both groups 14 days post stroke. These AQP4 might not contribute to water transmembrane diffusivity, but play roles in neurotoxic waste elimination. Researchers found that toxic molecules present in the area of liquefactive necrosis can leak across the glial scar and were removed by the glymphatic system in peri-infarct tissue (32). So, it can be inferred that the higher AQP4 polarization of the TGN-020 group is beneficial for glymphatic clearance and neurological recovery.

For ischemic stroke and other multiple CNS diseases, peri-infarct reactive astrogliosis is usually accompanied by loss of AQP4 polarization in the same area (23, 33–35). Our experiment showed that the polarization of AQP4 correlated negatively with the astrogliosis area, indicating that the astrogliosis extent might affect the polarization of AQP4. The close relationship between astrogliosis and AQP4 polarization was also discovered in rodent models of traumatic brain injury and multiple microinfarcts, though needing further investigations to determine the underlying mechanisms. Some researchers regard the loss of AQP4 polarization as an important feature of reactive astrocytes rather than a pathological consequence of endfeet damage (34, 36). We consider that the decreased astrogliosis after acute inhibition of AQP4 might contribute to the preservation of AQP4 polarization.

The reactive astrogliosis that occurs after ischemic stroke is extremely complex and incompletely understood, playing both detrimental and beneficial roles on neurological recovery (37–39). Some studies found that reactive astrocyte was beneficial for vascular repair and axonal regrowth after CNS injury (40, 41), while other studies revealed that reactive astrocyte could restrict neural repair by expressing growth inhibitory factors and forming glial scars (42). These contradictory roles of reactive astrocytes may be due to different reactive phenotypes induced by injury (43, 44). The reactive astrocytes in neuroinflammation of ischemia could be classified into A1s and A2s, which exert different functions (45, 46). The A1s exert the neurotoxic role with classical complement cascade gene upregulation, while the A2s upregulate many neurotrophic factors to promote neuronal recovery (47, 48). Therapies aimed at enhancing pro-reparative functions and reducing harmful functions in reactive astrocytes may benefit the outcome of ischemic stroke (49).

In this study, acute inhibition of AQP4 reduced peri-infarct astrogliosis and preserved AQP4 polarization, accompanied by a decreased lesion volume and improved neurological function. AQP4 is implicated in astrocyte migration and astrogliosis after brain insult (50, 51), which was supported by the reduced peri-infarct astrogliosis after inhibition of AQP4 observed in our study. Moreover, we inferred that the reduced astrogliosis might attenuate inflammation and promote neural rejuvenation by reducing the number of neurotoxic A1s astrocytes, contributing to peri-infarct tissue repair and functional outcomes. Reactive astrocytes of different phenotypes exhibit double-edged effects on pathological progression (49, 52, 53), our experiments and substantive studies that demonstrated inhibiting reactive astrogliosis facilitated neural rejuvenation and the long-term functional outcome might be attributed to a decrease of the neurotoxic A1s astrocytes (54–56). Besides, we speculated that the preserved AQP4 polarization benefits the cognitive recovery of TGN-020-treated rats by increasing the drainage of toxic extracellular fluid in the core of the infarct. Perivascular AQP4 is a critical component of the brain glymphatic system (57, 58). The loss of AQP4 polarization would impair the clearance efficiency of the glymphatic system, resulting in toxic protein deposition and the induction of cognitive deficits after stroke (59, 60). Therapeutic strategies that improved the AQP4 polarization might be effective to enhance the glymphatic function and contribute to the neurological recovery (61).

Deciphering changes in AQP4 are helpful to understand its roles in the pathology of ischemic stroke; however, most analytical methods remain highly invasive or destructive. According to the literature, aquaporin overexpression produces contrast in DWI by increasing tissue water diffusivity (62). ADCuh (b values > 2000 s/mm²) could reflect the expression level of aquaporin by estimating water transmembrane diffusivity (63). However, the relationship between ADCuh and aquaporin expression patterns in different studies are controversial. Some studies found that ADCuh correlated positively with the aquaporins expression intensity in tumors (64, 65), however, studies on ischemic stroke showed that ADCuh correlated negatively with aquaporin expression intensity (66–68). In our study, ADCuh correlated positively with the polarization of AQP4 rather than its expression intensity. This was probably the result of no adequate deviations among the AQP4 expression intensity of rats in this study, which did not allow us to infer a statistically significant correlation with ADCuh. Besides, the polarization of AQP4 might be more consequential for the functions of the protein than its expression intensity, as implied by other studies (69, 70).

The present study had certain limitations. Firstly, the inconsistency of lesion volumes before intervention between groups was avoided to the greatest extent, however, it could still not be excluded from the analysis. Longitudinal studies including data before inhibiting AQP4 might be more conclusive. Secondly, because higher b-value images lead to more imaging artifacts, the correlations between ADCuh and AQP4 expression patterns were only analyzed in the peri-infarct striatum, thus further studies should be carried out using MRI machines with a higher performance. Thirdly, we didn’t use the gene transcriptome analysis or key molecular markers immunostaining to differentiate the specific changes of two groups of reactive astrocytes after acute inhibition of AQP4, which will be carried out in our further studies. Besides, we only evaluate the role of AQP4 inhibition after ischemia onset, without investigating the effect of AQP4 inhibition on astrocyte and venules after recirculation, giving the TGN-020 along with the removal of filament in establishing the artery occlusion stroke animal model may be helpful to answer that.

In conclusion, we found acutely inhibiting AQP4 with TGN-020 not only decreased the edema at the early stage of ischemic stroke but also reduced peri-infarct astrogliosis and AQP4 depolarization, promoting sensorimotor and cognitive recovery at the subacute stage. This study extends the evaluation timepoint of previous studies investigating the effect of TGN-020 on ischemic stroke, providing further supportive evidence that acute inhibition of AQP4 after stroke is a viable therapeutic strategy. Furthermore, we revealed that AQP4 polarization correlated negatively with astrogliosis in the peri-infarct area, indicating therapies targeting astrogliosis might be effective to preserve AQP4 polarization and promote neurological recovery in ischemic stroke. And our results showed that ADCuh could reflect the AQP4 expression patterns, it might be a useful tool to decipher the AQP4 expression noninvasively.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Fudan University Institutional Animal Care and Use Committee.

Study conception and design: YY and CS; experiment implementation, statistical analysis and figure preparation: CS, LL, XH, JT, and XZ; manuscript writing: CS and LY; paper reviewing: CL and YY. All authors read and approved the final manuscript.

This research was supported by the National Natural Science Foundation of China, Nos. 81771788. The funding sources had no role in study conception and design, data analysis or interpretation, paper writing or deciding to submit this paper for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.