Keywords

Introduction

Subarachnoid hemorrhage (SAH) is a severe and emergent cerebrovascular disease with high morbidity and mortality and always leads to poor outcome. SAH mainly affects middle-aged patients and counts a highest fatality in all stroke subtypes, which brings about a huge burden on economy and society [1] . Inflammation has been extensively studied in early brain injury (EBI) after SAH, which is considered to be the main cause of mortality and lead to poor outcome [2, 3] . NLRP3 inflammasome , as a part of pro-inflammation, has been well established to take part in the pathophysiology of EBI after SAH [4,5,6]. The inflammatory response and its effect on the occurrence of microthrombosis are frequently discussed as potential protagonists.

Microthrombosis , usually found to be dissociation between cerebral vasospasm (CVS) and delayed cerebral ischemia (DCI), regarded as the major cause of neurologic deterioration and consequent poor outcome [7] . Microthrombi were firstly described in patients who were thought to have died of DCI. The autopsy study showed significantly high amounts of microthrombi were found in ischemic regions [8] . Recently, the occurrence of microthrombi was verified after SAH in animal models [9,10,11] . Furthermore, endothelial cell apoptosis and coagulation induced by inflammation have been well established [7, 12] ,which were considered the main cause of microthrombosis [10] .

17-Allylamino-demethoxygeldanamycin (17-AAG) , a specific inhibitor of HSP90, has been proved to show its effect on preventing cell death especially necroptosis [11, 13, 14] . Moreover, necroptosis plays a potential role in inflammation [15] . Furthermore, HSP90 inhibition shows its protection effect by attenuating inflammation response [16] and maintaining blood-brain barrier after ischemia stroke [17] . RIP3 has been described as a key protein in NLRP3 inflammasome previously [4, 18] . More recently, HSP90 has been proven to be an upstream regulator of RIP3 in necroptosis [19, 20] .More recently, the HSP90 inhibitor can reduce cell death induced by oxygen-glucose deprivation model [19] . Therefore, we hypothesize that 17-AAG reduces microthrombosis via HSP90/RIP3/NLRP3 pathway after SAH.

Methods and Materials

Animals and SAH Model

Adult male Sprague-Dawley rats (250–280 g) were purchased from the Animal Center of Central South University (Changsha, China). Rats were housed in a room with constant temperature (25 °C), humidity control and with a 12/12 h light/dark cycle. Standard animal chow and water were freely available. All the experimental procedures were approved by the Institutional Animal Care and Use Committee of Central South University.

SAH model was conducted by modified endovascular perforation method as previously described. Briefly, rats were anesthetized with chloral hydrate (400 mg/kg, intraperitoneally). The left common carotid artery and external and internal carotid arteries were exposed, and a 4–0 monofilament nylon suture was inserted into the left internal carotid artery through the external carotid artery stump until feeling resistance and then advanced 3 mm to perforate the bifurcation of the anterior and middle cerebral artery. Sham rats underwent identical procedures except the perforation.

Experiment Design

Experiment 1: For the time course experiment, 70 animals were divided randomly into 5 groups (n = 12/group): sham and SAH (1, 2, 3, 5 days (d)) groups. The neurological function was measured by modified Gracia score, and beam balance test was measured. Rats in the sham group underwent a procedure similar to that of the SAH group except perforation; rats in the SAH group underwent perforation and then euthanized in different time point after SAH. At the same time, SAH grade was measured. Then every group was divided into two groups used for Western blot and immunohistochemistry, respectively (n = 6/group).

Experiment 2: For 17-AAG treatment, 36 rats were divided randomly into three groups: sham (n = 6), SAH + vehicle (n = 12), and SAH + 17-AAG (n = 12). Modified Gracia score and beam balance test were measured. Then every group was divided into two groups used for Western blot and immunohistochemistry, respectively. Brain samples were collected after perfusion by 4% PFA for immunostaining.

Measurement of SAH Grade

As a parameter to evaluate the severity of SAH, SAH grade was obtained according to a grading system that was described previously. Briefly, the system was based on the amount of subarachnoid blood clots distributed in the six segments of basal cistern: grade 1, no subarachnoid blood (score = 0); grade 2, minimal subarachnoid clots (score = 1); grade 3, moderate subarachnoid clots with recognizable arteries (score = 2); and grade 4, blood clots covering all arteries (score = 3). A total score ranging from 0 to 18 was obtained by adding the scores from all six segments. The grading of SAH was performed by a partner who was blinded to the experiment. Rats with the SAH grade lower than 9 were excluded from this study.

Assessment of Neurological Function

The neurological status of all rats was evaluated at 24 h after SAH induction using the previously described modified Garcia scoring system and beam balance test [21].

The assessment of neurological score was performed by a partner who was blind to the experiment.

Western Blot Analysis

Western blot was performed as previously described [22] . Briefly, tissues were extracted by RIPA buffer. The protein concentrations were detected by using a bicinchoninic acid (BCA) assay (Beyotime, Shanghai, China). The protein samples were separated by 10% SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to polyvinylidene fluoride membranes (Millipore, USA). After blocked, the membranes were incubated overnight at 4 °C with the following primary antibodies: anti-HSP90 (1:1000, Proteintech), anti-RIP3 (1:1000, Abcam, USA), anti-NLRP3 (1:500, Abcam, USA), anti-ASC (1:300, Santa Cruz, USA), anti-IL-1β (1:1000, Proteintech), and anti-GAPDH (1:1000, Proteintech). After incubation with a secondary antibody, the immunocomplexes were visualized by enhanced chemiluminescence (SuperSignal Pierce Biotechnology).

Immunohistochemistry

Immunohistochemistry was performed as previously described [23] . Thirty micrometer coronal brain sections were cut as previously described in a cryostat (Leica CM3050S, Buffalo Grove, IL). Sections from each animal were divided into several subsets for immunohistochemistry. Microthrombi were visualized by immunohistochemistry with fibrinogen staining. After antigen retrieval, the slides were incubated in anti-fibrinogen antibody (1:500) (LifeSpan, Seattle, WA, USA) for 2 h and then incubated by biotinylated secondary antibody (1:400, Abcam, Cambridge, UK) for 1 h and ABC reagents (1:400; Vector Laboratories, Burlingame, CA, USA) for 1 h, with the immunoreactivity visualized in 0.003% H2O2 and 0.05% 3, 3′-diaminobenzidine(DAB), and counterstaining was performed with hematoxylin.

Immunofluorescence

Double immunofluorescence staining was performed as previously described [24] . The primary antibodies were biotinylated lectin (1:500, B-1175, Vector), followed by incubation with appropriate fluorescence-conjugated secondary antibody AMCA streptavidin (1:200, SA-5008, Vector). TUNEL staining (In Situ Cell Death Detection Kit, Roche) was performed following the manufacturer’s instruction. The sections were visualized under a fluorescence microscope Leica DMi8.

Quantification and Statistical Analysis

Microthrombi counts were performed as previously described [23] . Predefined regions of interest (ROIs) were photographed at ×200 magnification. The cumulative number of microthrombi was counted in a blinded fashion. The data were expressed as means ± SD. One-way analysis of variance (ANOVA) was used in this study to compare means of different groups followed by a Tukey’s multiple comparison test. The statistical analysis was carried out using SPSS 16.0 software (SPSS Inc., Chicago, Illinois, USA). Statistically significance was accepted with p < 0.05.

Results

General Observation and SAH Severity and Localization of SAH

A total of 107 rats were used. Eighteen rats were sham group and 87 rats underwent SAH. Two rats were excluded from the study due to mild SAH. At 24 h after SAH induction, blood clots were mainly observed around the Willis circle and ventral brainstem (Fig. 1a). No statistical differences in the average of SAH grades were observed between SAH groups (Fig. 1b). About 17.2% rats (15 of 87) under SAH condition died within 24 h after SAH induction. No animals died in the sham group (0 of 18 rats). No statistical significance was observed for mortality between operated groups.

Fig. 1
figure 1

General observation and SAH grade and neurological score. (a) Representative images of the brains from each group at 24 h after operation. Subarachnoid blood clots were mainly around the Willis circle and ventral brainstem. No blood present in the sham group. (b) There were no significant difference of SAH grade between the SAH groups (p < 0.05). Similar SAH bleeding severity was observed in all SAH groups. (c) Modified Garcia score and (d) beam balance test between sham, SAH + vehicle, and SAH + 17-AAG group. n = 6 per group. *p < 0.05 vs. sham, #p < 0.05 vs. SAH + vehicle

Full size image

Time Course of HSP90 Detected in the Left Hemisphere and Microthrombosis in the Cerebellum Following SAH

Western blot was performed to determine the HSP90 expression in the left hemisphere at the 1, 2, 3, and 5 days after SAH. Results showed that HSP90 level increased as early as 1 day after SAH and peaked 2 days and then decreased (Fig. 2a). Quantitative analyses of HSP90 time course show nearly 2.8 times higher at 2 days when compared with sham group (Fig. 2a). Immunohistochemistry was performed to detect the microthrombosis at same time point (Fig. 2b). Results showed that microthrombi counts increased as early as 1 day after SAH and peaked 2 days and then decreased (Fig. 2c).

Fig. 2
figure 2

Microthrombosis in cerebellum and expression HSP90 in left hemisphere after SAH. (a) Representative Western blot images and quantitative analyses of HSP90 time course from the left hemisphere after SAH. (b) Representative image and (c) quantitative analyses of microthrombosis in sham and SAH group with different time point. n = 6 per group. *p < 0.05 vs. sham

Full size image

HSP90 Inhibitor 17-AAG Treatment Reduces RIP3 and NLRP3 Inflammasome, Ameliorates Microthrombosis, and Improves Short-Term Neurobehavior

The modified Garcia and beam balance scores were significantly lower in the SAH + vehicle group than those in the sham group, and administration of 17-AAG improved the neurological scores in SAH + 17-AAG group significantly compared with SAH + vehicle group (P < 0.05; Fig. 1c, d). At 2 days after SAH, the expression of HSP90 was remarkably increased, and RIP3, NLRP3, ASC, and IL-1β were dramatically increased in SAH + vehicle group compared with the sham group. However, 17-AAG treatment reduced the level of HSP90 and inhibited the expression of RIP3, NLRP3, ASC, and IL-1β in SAH + 17-AAG group when compared with SAH + vehicle group (P < 0.05; Fig. 3a). When co-labeling with TUNEL and vessel marker lectin, the results show the endothelial cells on the inner surface vessels under apoptosis in SAH + vehicle when compared to sham; however the 17-AAG treatment can reverse this change (Fig. 3b, c). The microthrombosis was remarkably increased at 2 days in SAH + vehicle group compared with the sham group. However, 17-AAG treatment reduced microthrombosis when compared with SAH + vehicle group (P < 0.05; Fig. 3d, e).

Fig. 3
figure 3

The effects of 17-AAG in the protein expression of RIP3, NLRP3, ASC, IL-1β, and microthrombosis. (a) Representative Western blot images and quantitative analyses of HSP90, RIP3, NLRP3, ASC, and IL-1β. n = 6 per group. *P < 0.05 vs. sham group; #P < 0.05, vs. SAH + vehicle group. Bars represent mean ± SD. (b) Representative image and (c) quantitative analyses of double staining of TUNEL and lectin. (d) Immunostaining of fibrinogen and (e) quantitative analyses of microthrombi. n = 6 per group. *P < 0.05 vs. sham group; #P < 0.05, vs. SAH + vehicle group

Full size image

Discussion

In the present study, we found that both microthrombosis and the expression of HSP90 increased in the brain after SAH in rats. In addition, the 17-AAG treatment improved neurofunction after SAH, which were accompanied by a decrease in RIP3 and NLRP3 inflammasome expression. Furthermore, 17-AAG showed the neuroprotective effects after SAH, which were associated with the decreased microthrombosis 2 days after SAH.

The role of NLRP3 inflammasome in the pathophysiology of EBI after SAH has been proved previously [12] . The NLRP3 form a complex with ASC and pro-caspase-1 when activated and then maturate IL-1β and IL-18 by cleaved caspase-1 contributes to inflammation after SAH [12] . ROS generation, K+ efflux, and Cl− efflux were proved be three common upstream of NLRP3 activation [25] . Recently, RIP3 has been elucidated to have regulated NLRP3 inflammasome after SAH [4] . HSP90 , as a chaperone protein, has been proven to show its effect on ischemia stroke. HSP90 inhibitors such as 17-AAG show its protective effects in ischemia stroke by attenuating inflammatory responses, ameliorating neuronal autophagic death, and protecting neural progenitor cell death and blood-brain barrier from disruption [16, 17, 26, 27]. More recently, HSP90 as a direct upstream regulator of RIP3 has been proved in neuronal oxygen-glucose deprivation model, and inhibition of HSP90 protected neurons from necroptosis [20] . Interestingly, a clinical study show that the antibody level of heat shock proteins increased in stroke patients [28] . However, the HSP90 regulate the NLRP3 activation after SAH have not been elucidated. In our present study, we found that HSP90 increased 1 day and peaked 2 days after SAH, and administration of HSP90 inhibitor 17-AAG reduced inflammation via HSP90/RIP3/NLRP3 signaling pathway as well as improve neurofunction.

Microthrombosis , as well as pro-inflammatory cascades and blood-brain barrier disruption, consist the main cascade events of secondary injury after SAH. One of the main causes of microthrombi formation after SAH was endothelial cell injury [10] . What’s more, endothelial apoptosis after SAH has been shown to be associated with inflammatory mediators such as interleukin-1β (IL-1β) [7, 29] . In the present study, our results show that 17-AAG can reduce IL-1β by inhibiting NLRP3 inflammasome formation and then reduce microthrombosis after SAH. Thus, reduce NLRP3 inflammasome mediated inflammatory response by administration of 17-AAG may be a pential target to therapy microthrombosis after SAH.

These findings suggested that the administration of 17-AAG could attenuate NLRP3 inflammasome activation, inhibit microthrombosis, and improve neurofunction after SAH, at least, in part, through the HSP90/RIP3/NLRP3 signaling pathway.

Conclusion

Our study showed that 17-AAG can improve neurofunction and alleviate inflammation through the HSP90/RIP3/NLRP3 pathway as well as reduce microthrombosis after SAH. It may provide an optical method to treatment of SAH .