In this study, we investigated the protective effects of rGAS6 against stroke-induced HT by influencing microglial polarization to suppress BBB disruption and its underlying molecular regulation. Our main findings are as follows: (1) rGAS6 treatment alleviated neurological damage and BBB disruption and reduced cerebral hemorrhage by activating GAS6/Axl pathway in the acute phase of HT. (2) The improvement induced by rGAS6 was abolished by R428, which suppressed Axl expression. (3) In vivo and in vitro studies demonstrated that rGAS6 shifted microglial polarization toward an anti-inflammatory M2 phenotype, thereby reducing neuroinflammation through activation of the GAS6/Axl/SOCS1/3 pathway. (4) The protective effect of rGAS6 on BBB function was associated with its modulation of microglia and the associated inflammation, achieved through activation of the GAS6/Axl/SOCS pathway. These results indicate that rGAS6 could be a promising therapeutic agent for clinical application to reduce IS-induced HT via regulating GAS6/Axl/SOCS1/3 pathway.
Recently, HT has emerged as a major cause of poor prognosis in patients with IS, especially in individuals with severe comorbid diseases such as hyperglycemia and hypertension or those receiving delayed tPA administration beyond the standard 4.5 h treatment time window [2, 29]. BBB disruption is a major contributor to the occurrence of HT [3]. Several studies have shown that drugs blocking BBB disruption can protect the brain from HT and extend the therapeutic window for tPA administration [11, 30, 31]. Consequently, there is a need to urgently identify effective drugs to prevent BBB disruption for treating post-IS HT. The TAM system has been implicated in regulating innate immune response, promoting phagocytosis, inhibiting apoptosis, and stimulating cell proliferation [32]. Our previous study noted that the baseline serum Axl levels in patients with IS were negatively associated with the HT risk in patients with IS after tPA treatment, suggesting that Axl may serve as a promising therapeutic target for HT prevention following IS [28].
Axl is a member of the TAM family and exists in two forms: soluble Axl (s-Axl) and p-Axl. GAS6 is a ligand for TAMs receptors, with the highest affinity for the Axl receptor [23]. Previous studies have suggested that Axl activation is uniquely dependent on GAS6 which promote Axl phosphorylation [21]. However, s-Axl competes with total Axl for GAS6 binding, leading to the shedding and inactivation of GAS6. This process suppresses the Axl activation and its subsequent physiologic benefit [33–35]. Therefore, in our study, we aimed to supplement GAS6 levels and activate Axl signaling by exogenous rGAS6, and then assess its impact on BBB function and HT. As expected, our results showed that the administration of exogenous rGAS6 significantly augmented Axl phosphorylation, protected the BBB, and ameliorated HT. However, injection with Axl antagonist, R428, suppressed the phosphorylation of Axl, as well as abolished the beneficial effects of rGAS6 on HT. These findings strongly support the pivotal role of GAS6/Axl signaling in improving HT and highlight the clinical value of rGAS6 treatment.
Microglia are key initiators of brain immune responses following IS [36, 37] They can rapidly respond to ischemia by activating and migrating to the damaged cerebral vasculature, where they trigger a cascade of proinflammatory cytokine signaling that disrupts the integrity of the BBB, which begins before any detectable changes in BBB permeability. Since Axl is known as a pivotal innate immune regulator that dampens inflammation and maintains immune homeostasis, and is highly upregulated in microglia 24 h after HT, we explored whether the rGAS6-associated BBB protection in the initial phase of HT is caused by the regulation of microglial activation and subsequent neuroinflammation. We initially investigated the impact of rGAS6 on microglial polarization. As anticipated, rGAS6 induced a significant shift in microglial polarization toward the anti-inflammatory M2 phenotype. This shift was accompanied by a decrease in proinflammatory cytokines levels and an increase in anti-inflammatory cytokines levels. However, these effects were abolished by R428 injection. In vitro experiments using rGAS6 and Axl siRNA on microglial BV2 cells further confirmed these observations.
Next, we investigated the mechanisms underlying the involvement of the GAS6/Axl system in the modulation of microglia and inflammation. Previous studies have demonstrated the broad involvement of the TAM family in inhibiting TLR-induced inflammation. Specifically, TLR upregulates TAMs via the induction of the IFNAR-STAT1 pathway [38, 39], which subsequently activates the downstream expression of SOCS1 and SOCS3, resulting in the inhibitation of TLR signaling [20, 24, 40]. Remarkably, the SOCS family, known as classical feedback inhibitors of cytokine signal transduction, is also crucial in regulating macrophage polarization and inflammatory responses [41–45]. In studies conducted on APPswe/PS1dE9 mice, scientists discovered that SOCS3 suppressed microglial polarization toward the M1 phenotype by blocking IL-6 production [46]. In a rat model of subarachnoid hemorrhage, downregulation of SOCS1 contributed to early brain injury following subarachnoid hemorrhage by inducing inflammatory responses [45]. On this basis, we presumed that SOCS signaling might be critically involved in the GAS6/Axl-mediated microglial polarization. Among the SOCS family members, SOCS1 and SOCS3 are the most studied. Therefore, we analyzed the expression of SOCS1 and SOCS3 proteins in the brains of HT rats using western blotting. The results shown that rGAS6 significantly reversed the HT-induced downregulation of SOCS1 and SOCS3 at 24 h after reperfusion, whereas R428 abolished this effect. Consistent with the in vivo findings, experiments conducted on microglial BV2 cells also revealed a significant upregulation of SOCS1 and SOCS3 expression following rGAS6 treatment compared to the OGD/R group, whereas transfection with Axl siRNA transfection reversed their upregulations. Furthermore, additional knockdown of either SOCS1 or SOCS3 combined with rGAS6 treatment aggravated M1 polarization of microglia and the release of proinflammatory cytokines. Accordingly, it is plausible that the GAS6/Axl/SOCS1/3 signaling cascades play a critical role in rGAS6-mediated microglial polarization and inflammatory inhibition following HT.
Since Axl is widely expressed in different cell types, it may affect BBB function through various unknown mechanisms beyond its effects on microglia. To further clarify the crosstalk between the effects of rGAS6 on microglial polarization and BBB integrity, we established an OGD/R and tPA-stimulated Bend.3 endothelial cell model to simulate IS-induced HT in vitro, and subsequently co-cultured them with BV2 cells following reoxygenation. As anticipated, co-cultured with OGD/R-stimulated BV2 cells significantly aggravated endothelial cell function impairment, whereas rGAS6-treated BV2 cells improved endothelial cell function. The additional transfection of Axl siRNA in BV2 cells abolished the rGAS6-induced protection. Accordingly, we conclude that rGAS6 affects BBB function, at least partially, by modulating microglia polarization and inflammation via the activation of GAS6/Axl signaling.
In animal models, acute hyperglycemia-enhanced HT was reported years ago and is supported as a classical HT model [47, 48] Clinical studies have also demonstrated that hyperglycemia or diabetes was independently associated with HT in patients with IS [49–51]. Our study observed that the hyperglycemia-enhanced HT model consistently induced parenchymal hematoma 24 h after reperfusion with a stable and more severe BBB disruption. In contrast, the thromboembolic stroke model with tPA treatment tended to produce punctate or petechial hemorrhage with variable BBB [52]. Therefore, we chose the acute hyperglycemia-induced HT model for the animal studies. Since post-IS HT is critically dependent on the disruption of BBB,, which is significantly influenced by microglia-mediated inflammation, regardless of treatment, we hypothesize that the postulated effect of rGAS6 on the modulation of microglia and preservation of BBB integrity in hyperglycemia-enhanced HT can also alleviate tPA-induced HT. Our in vitro study supports this hypothesis.
Our study had some limitations. First, it is important to acknowledge that the GAS6/TAM system exerts multiple actions in the CNS [21, 53, 54]. As previously reported, GAS6 efficiently recognized phosphatidylserine on the surface of apoptotic cells, leading to enhanced engulfment by macrophages and reduced inflammation [20, 21]. In ICH mice, Axl is significantly upregulated in macrophages during the recovery phase, mediating the efferocytosis of erythrocytes and hematoma resolution [54]. Moreover, Axl upregulation occurs in astrocytes 72 h after TBI, promoting the transformation of astrocytes into a phagocytic phenotype, thereby ameliorating neuroinflammation [53]. In addition to its effects on immune cells, Axl is widely expressed in different endothelial cells, participating in reducing endothelial cell apoptosis and inhibiting the chemotaxis and adhesion affinity between monocytes and endothelial cells [55–57]. Further investigation is needed to explore the association between these actions, and the protective effect of rGAS6 against HT. Second, our current data primarily investigated the neuroprotective effects of rGAS6 during the acute phase of HT (within 24 h). However, it is important to explore whether continuous injection of rGAS6 or injection during the recovery phase of HT can influence the long-term prognosis of HT through different mechanisms.