DOI: https://doi.org/10.21203/rs.3.rs-2471501/v1
At present, there is no effective treatment for secondary brain injury caused by spontaneous intracerebral hemorrhage (ICH). This study aims to explore new therapeutic targets after ICH. Galectins-8 is a tandem repeat galectin with a unique preference for α2,3-sialylated glycans, and its expression is ubiquitous. Gal − 8 regulates cytokine production, cell adhesion, apoptosis, chemotaxis, endocytosis, differentiation and migration, including immune cells. We used wild-type(WT)C57BL/6J mice and the mice of Galectins-8 gene knockout to establish intracerebral hemorrhage model by collagenase injection and found that Galectins-8 was highly expressed around the hematoma and in the center site of the hematoma after intracerebral hemorrhage. We also found that inhibiting the expression of Galectins-8 or Galectins-8 gene knockout mice may attenuate secondary brain injury following intracerebral hemorrhage by reducing microglia-induced inflammatory responses. Galecectin-8 knockout mice had significantly reduced expression of inflammatory factors, such as TNF-α(P = 0.0353), MCP-1(P = 0.0469), and HMBG1(P = 0.0466). This is in contrast to previous studies that have suggested Galectins-8 as a neuroprotective factor. From this, we draw a conclusion that Galectins-8 played an crucial role in regulating the inflammatory response during intracerebral hemorrhage. Our study highlights Galectins-8 as a potential therapeutic target to protect the brain against secondary brain damage during intracerebral hemorrhage.
Stroke is the second leading cause of disability and death in the world after coronary artery disease(Cassidy and Cramer 2017).Cerebral hemorrhage (ICH), generally refers to bleeding in the brain parenchyma caused by the rupture of a cerebral blood vessel due to non-traumatic factors, has a rapid onset and is the most serious form of stroke, accounting for about 10–15% of strokes. It has a high morbidity and mortality rate, and places a heavy burden on society and families(Poon et al. 2014). And because of the high mortality and morbidity of ICH, the concern for effective treatment of ICH has been increasing. Brain injury after cerebral hemorrhage consists mainly of primary injury caused by hemorrhage and hematoma, with the main damage occurring within minutes to hours after the start of hemorrhage, mainly due to the mass effect caused by mechanical damage, and secondary brain injury (SBI) caused by the pathological response to the hematoma(Zhou et al. 2014). Initial hematoma volume, age, and ventricular volume may determine the severity of secondary injury(Aronowski and Zhao 2011). There is growing evidence that brain edema, neuroinflammation and neuronal damage are the main causes of secondary brain injury after cerebral hemorrhage, and the immune system is closely associated with them. Immune system is activated early after vessel rupture. After cerebral hemorrhage, the exuded blood components activate microglia, triggering an immune response involved in the process of cerebral hemorrhage injury (Shao et al. 2019).
Microglia are resident immune cells in the brain, and they have an active immune monitoring function to maintain the homeostasis of neuronal networks and are involved in central nervous system (CNS) development as well as homeostatic interactions(Umpierre and Wu 2021). Increasing evidence suggests that the development of neuroinflammation and secondary brain injury after cerebral hemorrhage is caused by activation of the innate immune system(Giordano et al. 2020; Sansing et al. 2011) Microglia of CNS macrophages are capable of eliciting innate immune responses when they sense danger signals and acquire different inflammatory phenotypes through activation and polarization(Parada et al. 2019).Activated microglia release inflammatory factors such as tumor necrosis factor-α (TNF-α), chemokines (chemokine ligand 2 (CXCL2)), interleukin-6 (IL-6) and interleukin-1β (IL-1β) that cause neurological damage and are referred to as the M1 phenotype (pro-inflammatory phenotype)(Ghosh et al. 2016). In addition, microglia promote high expression of IL-10, IL-4, insulin-like growth factors-1 (IGF-1), brain-derived neurotrophic factor (BDNF), and transforming growth factor β (TGF-β) to exert neuronal tissue recovery and anti-inflammatory effects, which are known as the M2 phenotype (anti-inflammatory phenotype)(Siddiqui et al. 2016). Whether pro-inflammatory phenotypes or anti-inflammatory phenotypes, these phenotypes play a considerable role in the ensuing neuroinflammation associated with cerebral hemorrhage.
Galectins are a family of soluble mammalian lectins in which a conserved carbohydrate recognition domain (CRD) is a unique property that exerts multiple biological effects by binding to β-galactoside (Elola et al. 2007). Currently, there is sufficient evidence that galactose lectin is a major mediator of the innate and acquired immune response(Liu and Rabinovich 2010). It has been proposed that Galectins modulate neuroinflammatory and neurodegenerative processes, promote microglia polarization, immune surveillance and neuroprotection through specific and dynamic patterns of extracellular and intracellular interactions with glycoglycans, suggesting potential therapeutic targets(Barake et al. 2020). Galectin-8(Gal-8) is derived from a tandem repeat sequence with two different CRDs covalently fused, and the N-terminal CRD of Gal-8 binds preferentially to α2,3-sialylated glycans, a unique specificity in galectins(Ideo et al. 2003). There are increasing reports supporting its involvement in immune responses. galectin-8 is highly expressed in the nervous system mainly in the hippocampus and choroid plexus and in the cerebrospinal fluid, and is mainly localized to microglia/macrophages(Stancic et al. 2011).
A growing number of reports support the involvement of Gal-8 in the immune response. The role of Gal-8 in autoimmune inflammation has been previously proposed (Tribulatti et al. 2009). Pardo E's study of patients with experimental autoimmune encephalomyelitis and relapsing-remitting multiple sclerosis (RRMS) shows that Gal-8 as an immunosuppressant protects against inflammatory damage in the CNS(Pardo et al. 2017). However, in the past few years, it has also been suggested that Gal-8 is involved in inflammatory processes. The activation of platelets(Schroeder et al. 2016) and endothelial cells (Cattaneo et al. 2014)by Gal-8 further supports this hypothesis. Amplification of inflammatory signals from neighboring cells via paracrine secretion may increase Gal-8 secretion. These dual roles of Gal-8 made me interested in studying Gal-8 expression after ICH and its effects on neuroinflammation.
We selected C57BL/6 wild-type(WT) and Gal-8 knockout(KO) mice and randomly divided t into four groups: (1) Sham: all mice were injected with a microsyringe without pumping collagenase; (2) ICH + Vehicle: Intraoperatively, 0.5 ul of sterile saline containing 0.075 U of type VII collagenase was injected into the striatum; (3) ICH + Thiodigalactoside(TDG): Intraoperatively, 0.5 ul of sterile saline containing 0.075 U of type VII collagenase was injected into the striatum, followed by intraperitoneal injection of TDG, 5 mg/kg, every other day for 10 days; (4) ICH + Gal-8KO: Gal-8KO mice were operated on as in the ICH group. All C57BL-6 WT mice were purchased from the Zhengzhou University Experimental Animal Center, and Gal-8KO mice were purchased from Jiangsu Ji Cui Yao Kang Science and Technology Co., Ltd. All mice were housed in SPF-grade animal rooms at the Second Affiliated Hospital of Zhengzhou University, with constant temperature, ad libitum diet and artificial circadian rhythm. All operations followed the rules and regulations established by the Animal Ethics Committee of Zhengzhou University School of Medicine.
Establishment of ICH model in mice by collagenase injection(Askenase and Sansing 2016). Mice were anesthetized with inhaled isoflurane and immobilized on an intracerebral hemorrhage localizer. After skin preparation and disinfection, the scalp is incised to ensure that the left and right skull bones are at the same level as the anterior and posterior fontanel. Taking Bregma as the origin (left side opening 2.0 mm, front 0.6 mm, depth 3.2 mm), 0.075 u (0.5 ul) collagenase was slowly injected with microinjection pump within 5 min, and stopped the injection for 10min after the injection to prevent collagenase reflux. After the surgery, slowly pull out the the microsyringe needle. Bone wax to close the bone holes and suture the scalp. After the operation, ensure that the body temperature of the mice is normal, and put them back into the cage after awakening.
Immunofluorescence staining of frozen sections: Three days mice with intracerebral hemorrhage were selected to make frozen section. The steps were as follows: 1 ×phosphate buffer saline(PBS)for 5 min × 3 rinses; 200 µL 0.1% phosphate-buffered saline with Tween 20 (PBST) for 15 min; 0.3% PBST or 5 min × 3 rinses; incubate in 5% donkey serum solution for 1h at room temperature; 0.3% PBST for 5 min × 2 rinses; the first antibody (GFAP, Iba-1, MPO) was added at 4℃ overnight; the second antibody (Cy3 or 488) was added at room temperature for 1h without light, and then 0.3% PBST for 10 min × 3 rinses .Finally, put the brain slice on the slide and dry it away from light, put 1–2 drops of 5% DAPI solution on the slide, cover the coverslip, and dry it away from light at room temperature.. The cells were observed and photographed using a confocal microscope.
Immunofluorescence staining of paraffin sections: After placing the tissue sections in the oven at 60℃ for 30 min, then put the paraffin sections into xylene Ⅰ, xylene Ⅱ, anhydrous ethanol Ⅰ, anhydrous ethanol Ⅱ, 95% ethanol, 80% alcohol, 75% alcohol in turn, and DDH2O for dewaxing; then place in citric acid antigen repair solution for antigen repair, draw water-blocking circles, add 5% BSA to the circles for closure, followed by primary antibody (Galectin 8/LGALS8 Rabbit mAb ABclonal) and incubate overnight at 4°C. Fluorescent secondary antibody (FITC) was added after PBS washing, and the tissue sections were incubated with the same amount of BSA. Next, tissue sections were incubated in the dark for 1 hour, then repeatedly incubated with NEUN Mouse Monoclonal Antibody and CY3 Antibody, followed by DAPI staining and fluorescence microscopy photography.
The mice were anesthetized intraperitoneally with 0.3% pentobarbital sodium. After the mice were rapidly cervically dislocated and executed, the brains were rapidly separated to obtain bilateral cerebral hemispheres and cerebellum, and the cerebral hemispheres were cut into the left and right hemispheres along the longitudinal fissure and placed on a weighed tin foil. The brain tissues were immediately weighed using an electronic molecular balance, which was the Wet Weight (WW), and then quickly baked in an oven at 100°C for at least 24 h. When the weight no longer changed, they were removed and weighed again to obtain the Dry Weight (DW), and then the brain tissue water content was calculated according to the Eliot formula: Brain Water Content (BWC) = (WW-DW)/WW × 100%.
Frozen sections of 50um from each group of mice were stained with LFB/CV, and then the volume of cerebral hematoma and the degree of cerebral swelling were counted. The specific calculation formula was as follows: brain hemorrhage volume/residual lesion volume = the sum of hemorrhage area at each level × section thickness; brain swelling degree = (hemorrhage side cerebral hemisphere volume - contralateral cerebral hemisphere volume)/hemorrhage side volume × 100%.
After the mice were decapitated, the brain tissue around the hematoma was taken, weighed, scissors were used to cut up the brain tissue, and then a grinding rod was used to grind the brain tissue, after which the pre-prepared cell lysis solution was added to the EP tube and the ice bath was applied for 10 minutes. Then Ultrasound 1.5s, interval 2.9s, time 1min. Cryogenic Ultracentrifuge at 4,000 rpm for 20 minutes. The proteins in the supernatant were collected for protein blotting analysis. The same amount of protein is transferred to PDVF membrane by electrophoresis. Seal PDVF membranes with 5% skim milk for 1 hour at room temperature, then incubate overnight at 4°C in the refrigerator with antibodies (Raf-1 monoclonal antibody (1:1000), MCP-1 mouse monoclonal antibody (1:1000), ERK1/2 rabbit Polyclonal antibody (1:1000), phospho-ERK1/2antibody affinity (1:1000), mouse anti TNF-α monoclonal antibody, IL-10 mouse antibody, HMGB-1 Rabbit Polyclonal antibody, TGF-β1 rabbit polyclonal antibody). Then incubate the secondary antibody (HRP-labeled goat anti-rabbit, HRP-labeled goat anti-mouse) for 2h, Tris-Buffered Saline with Tween 20(TBST) rinse and ECL ultrasensitive luminescent solution development exposure.
GraphPad Prism 7 software was used to analyze and plot all experimental data, and data are presented as mean ± standard error (mean ± SEM). Comparisons between groups were made using Student's t-test or one-way ANOVA for normally distributed continuous variables. When ANOVA showed significant differences, post hoc Bonferroni test was used for pairwise comparisons. A statistically significant difference is indicated at P < 0.05.
Gal-8 is now known to be localized mainly in the cytoplasm of brain microglia. For example, in central nervous system diseases, microglia are activated in patients with active multiple sclerosis, and the expression of Gal-8 is increased(Stancic et al. 2011). However, it is unclear how Gal-8 is expressed in the acute phase of cerebral hemorrhage. We first examined Gal-8 expression after ICH. The expression of Gal-8 was found to be increased around the hematoma compared to the sham group (Fig. 1a). But there was an explosive increase in Gal-8 expression in the central part of the hematoma, especially in the perivascular area (Fig. 1b). Therefore, the high expression of Gal-8 after intracerebral hemorrhage is partly related to endothelium.
Groups of mice (n = 5) were tested at different time points (6 h, 12 h and 1,3,7 and 14 days) after ICH. Western blot analysis showed that Gal-8 expression increased at 6 hours after ICH (p = 0.0014), peaked at 12 hours and lasted for 24 hours (p = 0.0119), and then gradually decreased (Fig. 1c, d).
To investigate the role of Gal-8 in secondary brain injury induced by ICH, we chose TDG as a broad-spectrum inhibitor of Gal-8, which has anti-inflammatory effects (Chen et al. 2016). We found that TDG can significantly reduced the degree of brain swelling after intracerebral hemorrhage(), however, there was no significant difference in cerebral hematoma volume and brain water content compared with the two groups (Fig. 2a–d).
Since microglia inflammatory response plays an important role in ICH secondary injury, we also explored the effect of Gal-8 on immune cells in ICH mice during the acute phase. We used Iba-1 and GFAP as surface markers for microglia and astrocytes, and MPO as a surface marker for neutrophils, positive cells were found in the microglia, astrocytes, and neutrophils surrounding the hematoma after ICH, and they were dramatically increased around the perihematomal tissue after intracerebral hemorrhage; in contrast they were significantly down regulated by TDG treatment (GFAP: p = 0.0393, t = 2.625; Iba1: p = 0.0272, t = 2.904; MPO: p = 0.0128, t = 3.503) (Fig. 3a–e). FJB fluorescence staining was used to observe necrotic neurons around the hematoma, However, we did not find that TDG significantly reduced the number of necrotic neurons around the hematoma (Fig. 3g, h).
To further understand the pro-inflammatory or anti-inflammatory effect of Gal-8 in the acute stage of ICH, we detected the expression of TNF-α, TGF-β, HMGB-1, IL-10, MCP-1 around the hematoma by western blotting analysis. The expression of TNF-α (P = 0.0158), HMGB-1 (P = 0.0032), and MCP-1 (P = 0.0057) was increased after ICH, and the expression of these inflammatory factors was significantly reduced in Gal-8 knockout mice (TNF-α:P = 0.0353, HMGB-1:P = 0.0466, MCP-1:P = 0.0469) (Fig. 4). However, the expression of anti-inflammatory cytokines TGF-β and IL-10 in WT and Gal-8 knockout mice was not statistically significant (P > 0.05) (Fig. 4). Therefore, we suggest that Gal-8 promotes the expression of pro-inflammatory factors in the acute phase of cerebral hemorrhage, thereby exacerbating secondary brain injury after cerebral hemorrhage.
The galectin family is considered a potential modulator of brain microglia polarization, immune surveillance, neuroinflammation, and neuroprotection, and has become a key regulator of immune responses. Gal-8 was first identified in a rat liver cDNA library as the third member of the mouse tandem repeat Galectin family(Hadari et al. 1995). Gal-8 is also expressed in normal tissues and can be found in many tissues, including lung, liver, kidney, spleen and others(Bidon et al. 2001). Recent studies have found that Gal-8 is also expressed in brain tissues. β-gal staining revealed that Gal-8 is expressed in the choroid plexus, lateral ventricles, and third ventricle, and immunoprotein blotting methods also revealed increased expression of Gal-8 in cerebrospinal fluid(Pardo et al. 2017). Our study provides the first evidence for the role of Gal-8 in a collagenase-induced mouse ICH model. We demonstrated a significant increase in Gal-8 expression in mouse brain after cerebral hemorrhage. We found the increased expression of Gal-8 around the hematoma and at the central site of the hematoma. In particular, the expression of Gal-8 in the vascular endothelium of blood vessels increased greatly. This finding is consistent with M. Stancic's study(Stancic et al. 2011). This gives us reason to believe that high expression of Gal-8 plays a role in the pathophysiology of ICH.
Increasing evidence that activation of the innate immune system during ICH can lead to neuroinflammation and secondary brain damage(Giordano et al. 2020). As central nervous system resident cells, microglia can quickly evoke an innate immune response after cerebral hemorrhage(Parada et al. 2019).The microglia involved in the immune and inflammatory response to ICH. Microglia activation drives pro-inflammatory, oxidative stress and cytotoxic cascade responses leading to cell death and dysfunction(Wang 2010); therefore, the activation of microglia macrophages is essential for secondary brain damage after ICH. Recently, a growing body of data has demonstrated that Gal-8 is involved in both acquired and innate immune responses and is highly expressed in inflammatory diseases with dual behaviors; it acts mainly as a pro-inflammatory-like molecule in different quiescent cells of the immune system, but also shows anti-inflammatory properties when these cells are activated(Tribulatti et al. 2020). There's a connection between the microglia and the Gal-8. In studies of active and remitting brain tissue from patients with multiple sclerosis, microglia are the major Gal-8 positive cells. Microglia are activated in active brain tissue, while Gal-8 expression is increased in active brain tissue compared to the chronic phase(Stancic et al. 2011). Therefore, we believe that the high expression of Gal-8 is related to the polarization of microglia. Our experiments revealed that inhibition of Gal-8 expression significantly reduced the number of activated microglia around the hematoma. Therefore, we speculate that Gal-8 promotes the activation of microglia.
Polarization of microglia after brain hemorrhage is a very complex process. Upon activation, microglia polarize and form different phenotypes, namely the classically activated M1 and alternatively activated M2 phenotypes, which may have different effects on neuroinflammation in models of brain disease(Xiong et al. 2016). Activation of microglia usually leads to the production of pro-inflammatory cytokines, which act as pro-inflammatory agents. Alterations in pro-inflammatory cytokine profiles and levels after ICH are thought to explain the altered microglia macrophage function. Some pro-inflammatory factors (TNF-α, IL-1β, IL-6) and chemokines (CXCL2) produced by microglia can promote neuroinflammation(Wang 2010; Zhou et al. 2014) after ICH. Indeed, when the signal is inhibited by TNF-α (Lin et al. 2012), IL-1β(Yang et al. 2014), thrombin(Wu et al. 2009), and HMGB1(Li et al. 2015), reduced nuclear factor -k- gene binding (NF-κB) expression and microglia macrophage activation after ICH can be observed. This suggests that the activation of microglia macrophages enhances the inflammatory and destructive effects of brain tissue after cerebral hemorrhage. The anti-inflammatory effects of microglia macrophages are mainly due to the neuroprotective effects of type M2 microglia macrophages by phagocytosis of lysates, including neurorepair and anti-inflammatory effects, which are usually associated with high expression of IL-10, IL-4, insulin-like growth factor 1(IGF-1), brain derived neurotrophic factor (BDNF), and TGF-β (Liu et al. 2021). In the experiments, by observing the changes in the cellular markers of M1 phenotype and M2 phenotype, there seems to be a process of interconversion between M1 phenotype and M2 phenotype(Taylor et al. 2017; Lan et al. 2017). In studies of brain injury, inhibition of gal-8 expression reduces the extent of brain swelling after intracerebral hemorrhage. Western blot analysis showed that HMGB-1, MCP-1 and TNF-α expression were increased in wild-type mice after ICH, while the expression of these factors was significantly reduced in Gal-8 knockout mice. Notably, the expression of IL-10 and TGF-β did not change significantly after the establishment of ICH models in wild-type and Gal-8 knockout mice. Based on the experimental results, we suggest that high Gal-8 expression drives microglia to polarize toward the M1 phenotype and release pro-inflammatory factors, thus acting as a pro-inflammatory agent.
This conclusion above contradicts previous findings that Gal-8 acts as a neuroprotective factor. Previous studies have suggested that Gal-8 has a neuroprotective effect and suggested that the immune response to Gal-8 may act as a suppressor, but this protective effect is mostly found in central nervous system autoimmune disorders, such as multiple sclerosis(Massardo et al. 2009), EAE(Pardo et al. 2017), and amyotrophic lateral sclerosis. This phenomenon may be due to the different cause of the pathogenesis of neuroinflammation between ICH and autoimmune diseases. The immunosuppressive effect of Gal-8 may be to suppress the autoimmune response, whereas neuroinflammation after ICH followed a cascade of inflammatory processes triggered by mechanical destruction of hematoma(Bai et al. 2020). Gal-8 does not act as an immunosuppressant against neuroinflammation following ICH.
Our study leaves much to be desired. On the one hand, we only observed that high expression of Gal-8 after ICH promoted the release of inflammatory factors, but the exact mechanism is not clear to us and further studies are needed. Second, we have only performed in vivo studies, and in vitro studies are still needed to verify the pro-inflammatory effect of Gal-8 after intracerebral hemorrhage. We still have a lot of work to fill these gaps.
Overall, we found that inhibition of Gal-8 expression or Gal-8 knockdown protected against secondary brain injury after ICH, which in turn demonstrated the neurotoxic effect of Gal-8 on secondary injury after ICH. gal-8 offers great therapeutic promise against ICH
Competing interests:
The authors have no relevant financial or non-financial interests to disclose.
Acknowledgements:
I am very grateful to Mr. Yansong Zhang and Dr. Bo Zhou from the Central Laboratory of the Second Affiliated Hospital of Zhengzhou University for their support and help during my experiments.
Funding:
Funding for this study came from a PhD research start-up grant funded by the Second Affiliated Zhengzhou University Hospital.
Authors’ contributions:
Professors Hongying Bai and Chao Jiang contributed to the concept and design of the study. Si Chen and Yuanyuan Xing were involved in the preparation of experimental materials, data collection and analysis. The thesis was mainly written by Jingjing Song. All authors evaluated previous versions of the manuscript and negotiated the final draft together.
Ethics approval:
The study was approved by the Ethics Committee of the second affiliated Hospital of Zhengzhou University.
Consent to participate: Not applicable.
Consent to publish: Not applicable.
Data, material, and/or code availability:
Data sets generated and/or analyzed during this study are available upon reasonable request to the corresponding author.