Background Sex differences affect the occurrence, progression and regression of subarachnoid haemorrhage (SAH). Oestrogen plays a protective role in alleviating the vasospasm and neuronal apoptosis induced by SAH. However, whether oestrogen affects blood‒brain barrier (BBB) integrity has not been fully studied. Oestrogen has been found to regulate the sonic hedgehog (SHH) signalling pathway through the oestrogen receptor in gastric cancer and adrenal glands, and the SHH signalling pathway has an important role in maintaining the BBB by upregulating the expression of tight junction proteins. In this study, we investigated the relationship between oestrogen and the SHH signalling pathway using clinical data and established an experimental SAH model to explore whether oestrogen could ameliorate BBB damage after SAH through the SHH pathway.
Methods Correlations between oestrogen and the SHH pathway were analysed by patients’ cerebrospinal fluid (CSF) samples and the Genotype-Tissue Expression database (GTEx). Then, an experimental rat SAH model was established using the endovascular perforation method and treated with oestrogen, oestrogen inhibitors and SHH signalling pathway inhibitors. Then, the effects of oestrogen on BBB damage were analysed by western blot, immunofluorescence and neurobehavioural experiments.
Results ESLIA detection and correlation analysis showed that oestrogen levels in patients’ CSF were positively correlated with the SHH pathway, which was further verified by GTEx gene-correlation analysis. SHH was found to be mainly expressed in neurons and astrocytes in rats under physiological conditions and was upregulated by oestrogen pretreatment. In the SAH model, oestrogen pretreatment was found to reverse SAH-induced decreases in the SHH pathway, which were counteracted by oestrogen receptor inhibitors. Furthermore, oestrogen pretreatment reduced SAH-induced BBB damage, brain oedema and neurological dysfunction, which were eliminated by SHH pathway inhibitors.
Conclusion In conclusion, we demonstrate here that oestrogen pretreatment ameliorates brain injury after SAH, at least in part through SHH pathway-mediated BBB protection.
- Cerebrovascular Disorders
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
WHAT IS ALREADY KNOWN ON THIS TOPIC
Sex differences exist in the occurrence and prognosis of patients with aneurysmal subarachnoid haemorrhage (SAH), in which oestrogen plays an important role.
WHAT THIS STUDY ADDS
This study found that oestrogen attenuates blood‒brain barrier damage after experimental subarachnoid haemorrhage via the sonic hedgehog pathway.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
This study may help explain the clinical sex differences associated with perihaematomal brain oedema and the prognosis of neurological function after SAH and provide potential targets for future treatment.
Subarachnoid haemorrhage (SAH), caused mainly by ruptured aneurysms, has remarkably high mortality and disability rates, although it constitutes only 5%–7% of strokes.1 In previous investigations, sex differences have been found to be associated with the incidence and outcome of SAH. There was a significant difference in the incidence of SAH in men versus women premenopause and postmenopause, with a higher prevalence in males than in premenopausal females and a lower prevalence in males than in postmenopausal females.2 3 Oestrogen replacement therapy is considered to have protective effects against the occurrence of SAH.4 Although the overall mortality after SAH is higher in women than in men, whether gender differences have an impact on functional outcome is controversial in previous studies. Whether oestrogen, an important factor influencing the incidence of SAH, also has a significant prognostic impact has not been adequately studied in clinical trials.
Blood‒brain barrier (BBB) dysfunction, vasospasm, neuronal necrosis and apoptosis are important pathological processes of brain injury after SAH. Previous studies on the effect of oestrogen on the prognosis of SAH have mainly focused on its antivasospasm, reduction of necrosis and apoptosis effects. Oestrogen downregulates TNF-α expression via oestrogen receptors and reduces apoptosis in neurons after SAH.5 6 By reducing iNOS expression and maintaining eNOS expression in the basilar artery, oestrogen alleviates vasospasm after SAH.7
Oestrogen and its receptors are widely expressed in the brain and have regulatory functions on neurons, astrocytes and endothelial cells.8 It has been shown that oestrogen can upregulate the sonic hedgehog (SHH) pathway through its receptor in adrenal cortical cells9 and gastric cancer cells.10 Importantly, the SHH signalling pathway plays a key role in maintaining BBB integrity.11 In addition, studies have shown that upregulation of the SHH signalling pathway can improve the prognosis of SAH.12 However, it is not clear whether oestrogen can ameliorate BBB damage after SAH through the SHH pathway.
In our study, we determined the correlation between oestrogen and the SHH signalling pathway and clarified the protective roles of oestrogen on BBB function after SAH. Furthermore, we explored the mechanisms by which oestrogen alleviates brain injury after SAH by using a variety of experimental methods.
Materials and methods
Cerebrospinal fluid (CSF) samples from 20 patients were obtained during ruptured aneurysm clamping surgery or during CSF examination. Each patient was informed of the purpose and procedure of the study, and written informed consent was obtained from each patient. Immediately after collection of all CSF samples, all samples were centrifuged at 1000 ×g for 20 min. The supernatant collected after centrifugation was used for the ELISA. A more detailed description of the materials and methods used is provided in online supplemental material and methods.
Genotype-Tissue Expression gene-correlation analysis
The Genotype-Tissue Expression database (GTEx) is an online database that provides expression data of different genes in different tissues of normal humans. All samples of the brain, excluding the cerebellum and spinal cord, were included. GEPIA (http://gepia2.cancer-pku.cn/) is a user-friendly analysis site that allows the use of data from GTEx to analyse correlations between genes.13 By calculating Pearson correlation coefficients between gene expression of oestrogen receptors, SHH signalling pathway and some tight junction proteins (TJs), we plotted heatmaps using Python. Correlation coefficients between the expression of oestrogen receptors (ESR1, ESR2) and SHH, SHH signalling pathways (SHH, PTCH1, GLI1), and TJs in brain tissues were calculated by using GEPIA as well.
Male Sprague-Dawley (SD) rats weighing between 280 and 330 g were purchased from the Animal Center of Chinese Academy of Sciences (Shanghai, China). All SD rats were raised in temperature-controlled and humidity-controlled animal quarters with a 12-hour light/dark cycle and unlimited access to food and water. The animal protocols used in this study were approved by the Animal Ethics Committee, and all experiments complied with the Animal Research Reporting of In Vivo Experiments guidelines.
Sacrifice of experimental animals
Experimental rats were anaesthetised by isoflurane gas at the experimentally set time points.14 After the disappearance of corneal and pain reflexes, the animals were sacrificed by transcardial perfusion with saline containing 0.1% EDTA after opening the thoracoabdominal cavity.
The experimental SAH model was generated by an endovascular perforation model in rats. The rats were intubated under isoflurane anaesthesia and mechanically ventilated until the corneal and pain reflexes disappeared.14 The external carotid artery was ligated and isolated, and a 4-0 nylon suture with the anterior end reshaped to a 60° angle was introduced through the dissected end of the external carotid artery to the left internal carotid artery. In the SAH group, after the nylon suture was inserted and resistance was felt, a further 3 mm was punctured, the bifurcation of the anterior and middle cerebral arteries was punctured, and the suture was immediately withdrawn. In the sham group, the wire embolus was introduced only into the internal carotid artery without perforation. At the time of sacrifice, the success of SAH was judged by assessing the SAH grade (≥8) as described previously. SAH models with insufficient scores were excluded from this study.15
The experimental design of the first experiment was as follows. All rats were randomly divided into three groups, including a normal group, a vehicle group and an oestrogen group. Each group included six rats that were euthanised after oestrogen intervention for 3 days, and brain tissue was harvested for immunofluorescence assays.
The second experimental design is as follows. All rats were randomly divided into seven groups, including one sham group and six SAH groups. Each group contained six rats. According to the experimental design at 6 hours, 12 hours, 24 hours, 48 hours, 72 hours and 168 hours after SAH, the rats were euthanised, and brain tissue was harvested for western blot assays.
The third experiment is designed as follows. All rats were randomly divided into six groups, including the sham group, SAH group, SAH+vehicle group, SAH+oestrogen (33 µg/kg) group, SAH+oestrogen+fulvestrant (ICI 182780, 2 mg/kg) group and SAH+oestrogen + cyclopamine (1 mg/kg) group.5 16 17 Each group contained six rats, and vehicle, oestrogen, fulvestrant and cyclopamine were administered for 3 days before and after SAH. Brain tissue was harvested for western blotting and ELISAs after all rats were euthanised 3 days after SAH. A more detailed description of the materials and methods used is provided in online supplemental materials and methods.
Human brain microvascular endothelial cells (hBMVEC) (hCMEC/D3, Shanghai Zhong Qiao Xin Zhou Biotechnology) were cultured in phenol red-free Dulbecco’s modified Eagle’s medium (DMEM) high-glucose complete medium with 10% fetal bovine serum and 1% penicillin/streptomycin to avoid the oestrogen-mimetic effects of phenol red.18 The intervention groups were given oxyhaemoglobin (20 µM), oxyhaemoglobin+human recombinant SHH (HrSHH, 0.1 µg/mL), and oxyhaemoglobin+HrSHH+cyclopamine (10 µM) interventions for 24 hours and were harvested along with the control group.11 16 19 These interventions were all dissolved in DMEM. All items required for in vitro experiments are shown in online supplemental table 1. A more detailed description of the materials and methods used is provided in online supplemental materials and methods.
Values are expressed as the mean and/or SD. Data were analysed using GraphPad Prism V.9.0.0 (GraphPad Software, La Jolla, California, USA), Python V.3.8. Data correlation was assessed using Pearson’s correlation coefficient. Statistical comparisons between two groups were performed using t tests, statistical comparisons between more than two groups were performed using one-way analysis of variance, and comparisons between more than two groups were performed using the Turkish post hoc test to determine the significance of pairwise differences. A probability of p<0.05 was considered statistically significant.
The SHH signalling pathway is correlated with oestrogen and its receptors
CSF samples were used to detect SHH concentrations and oestrogen concentrations by ESLIA kits. Correlation analysis suggested that SHH and oestrogen content were moderately positively correlated (R=0.6, p<0.05) (figure 1A, online supplemental table 2). To investigate the relationship between the transcriptional expression of oestrogen receptors, the SHH signalling pathway and TJs, we analysed the correlation between the expression of each of these genes in GTEx. The expression of each gene in the oestrogen receptors and SHH signalling pathway showed a positive correlation (R>0.2, p<0.05), except for ESR2, which was not correlated with SHH but had a p value>0.05 (figure 1B, online supplemental table 3). Then, we analysed the correlations between oestrogen receptor and SHH genes, between oestrogen receptor and SHH signalling pathway genes, between oestrogen receptor and TJs genes and between SHH signalling pathway and TJs genes based on the same data in GEPIA figure 1C-F. All showed a good positive correlation (R>0.2, p<0.05).
Tissue distribution of SHH and alterations after oestrogen intervention
Immunofluorescence was used to assess the expression of SHH in the brain. For immunofluorescence, we used GFAP as a marker for astrocytes and NeuN as a marker for neurons, and SHH was found to colocalise with cells positive for both markers. The expression of SHH was increased in temporal basal cortex brain tissue in the group treated with oestrogen for 3 days (figure 2A–B). Additionally, we used isolectin B4 as an endothelial cell marker and found that SHH failed to colocalise with endothelial cells (figure 2C). We then enumerated the proportion of cells costained with SHH as well as the total count of cells. Oestrogen intervention did not change the number of GFAP+astrocytes and NeuN+neurons, and the proportion of cells costained with SHH was not significantly altered (figure 2D–G). Additionally, we semiquantified SHH in temporal basal cortex brain tissue using western blotting, and SHH was significantly increased after 3 days of oestrogen intervention (figure 2H). These results suggest that oestrogen significantly increases the synthesis of SHH by neurons and astrocytes.
Protein expression of SHH is reduced and oestrogen reverses alterations in the SHH signalling pathway after experimental SAH via oestrogen receptors
Western blotting was used to assess the protein expression of SHH in temporal basal cortex brain tissue. Results showed that SHH was significantly reduced at 3 days and slightly recovered at 7 days compared with the sham group (figure 3A–B).
By performing oestrogen intervention in rats for 3 days each before and after SAH, we applied western blotting to determine the altered protein expression in the SHH signalling pathway. Expressions of SHH, PTCH1, SMO and GLI1 in the SHH signalling pathway were significantly reduced at 72 hours of SAH compared with the sham group. These decreased expression levels could be reversed by 6 days of oestrogen intervention. In contrast to the oestrogen alone administration group, no reversal was observed in the group given both oestrogen and the oestrogen receptor inhibitor fulvestrant (figure 3C–G).
After that, we observed the alteration of SHH expression in astrocytes and neurons after SAH by immunofluorescence. We found that the number of astrocytes increased after SAH but not neurons. SHH expression was reduced in both astrocytes and neurons after SAH, and some astrocytes were missing SHH colocalisation. The decrease in SHH expression was partially reversed by oestrogen (figure 3H–M). Additionally, we did not find SHH expression on the endothelium after SAH (online supplemental fig S1).
The protein expression of TJs is downregulated after SAH and reversed by ostrogen through the SHH signalling pathway
In vivo, western blotting was used to measure the protein expression of TJs. TJs were decreased at 72 hours after SAH compared with the sham group. Oestrogen intervention upregulated the expression of TJs at 72 hours after SAH. This intervention outcome could be reversed by cyclopamine, an inhibitor of the SHH signalling pathway (figure 4A–C). This alteration was subsequently verified by immunofluorescence costaining of TJs and endothelial marker isolectin B4 (figure 4D, online supplemental fig S2).
In vitro, we used western blotting to determine the expression of TJs in hBMVEC . The expression of TJs was decreased in the group with oxyhaemoglobin-mimicking SAH compared with the control group. Endothelial cells administered HrSHH protein intervention for 12 hours had significantly elevated TJs compared with the oxyhaemoglobin group. Cyclopamine inhibited the upregulation of TJs by HrSHH protein (figure 4E–G).
Oestrogen reverses BBB permeability and brain water content after experimental SAH
We used ELISA to determine the albumin concentration and Evans blue absorbance measurements to indirectly estimate changes in BBB permeability. BBB permeability was significantly higher at 72 hours after SAH than in the sham group. In contrast, in the oestrogen intervention group, there was no significant difference in BBB permeability at 72 hours after SAH, indicating that oestrogen significantly reversed the increase in BBB permeability after SAH (figure 5A–C).
Similarly, brain water content measurements revealed a significant increase in brain water content 72 hours after SAH, which was reversed by oestrogen. When the SHH signalling pathway was blocked by cyclopamine, this reversal by oestrogen could not be detected (figure 5D).
Oestrogen alleviates neurological dysfunction after experimental SAH
We finally performed behavioural tests after the administration of oestrogen or oestrogen+cyclopamine in rats with experimental SAH. The modified Garcia score and the rotarod test showed similar results, with the oestrogen intervention group having significantly better scores in the modified Garcia score experiment again and longer dwell time in the rotarod experiment compared with the SAH group compared with the control group, suggesting that oestrogen can improve neurological dysfunction after experimental SAH (figure 6A–C). In contrast, this improvement was not observed in the group given cyclopamine concomitantly. In addition, in the Morris water maze, we chose escape latency and lane length as evaluation indexes. Rats in the oestrogen intervention group showed the shortest latency and swim path length compared with the other groups, which also implies better locomotor ability (figure 6D–G). The results suggest that oestrogen attenuates neurological dysfunction through the SHH signalling pathway.
As a common and important factor, gender difference was shown to have an important impact on the onset and progression of a variety of brain disorders. Oestrogen, a sex cortisol hormone that has an important effect on the development and maintenance of brain structures, plays an important role in this difference.20 21 Experimentally and clinically, oestrogen has been shown to have a role in reducing the development of aneurysmal SAH and an effect that explains the difference in the incidence of SAH in women before and after menopause.22 However, due to the lack of oestrogen testing in current clinical studies or subgroup analysis before and after menopause, the effect of oestrogen on brain injury after SAH has not been uniformly concluded. On the other hand, laboratory evidence has all agreed that oestrogen alleviates vasospasm, inflammation and necrotic effects via NF-κB, NOS and other signalling pathways after SAH.5–7 Previous studies have found that oestrogen improves BBB permeability.23 In our study, we investigated whether oestrogen improved BBB permeability or neurological impairment after SAH and the underlying mechanism.
Under physiological conditions, oestrogen and classical oestrogen receptors are extensively distributed in the brain. In previous research, oestrogen in blood could interfere with neuronal and astrocyte regulation through the BBB. We found a positive correlation between SHH protein levels and oestrogen levels in the CSF of SAH patients by collecting real-world human CSF samples. Oestrogen has been shown to upregulate the expression of the SHH signalling pathway through oestrogen receptors in adrenal cells and gastric cancer models. In our study, this correlation between oestrogen receptors and the SHH signalling pathway was analysed through GEPIA in transcriptional data from GTEx. The transcriptional expression of two oestrogen receptor classical receptors, ESR1/2, positively correlated with the transcript levels of SHH signalling pathway components.
In previous studies, hBMVEs were regulated by extracellular sources of SHH, which in turn expressed more TJs and maintained the stability of the BBB.11 However, a few related studies have shown that neurons are the main source of SHH in the brain rather than astrocytes.24 The BBB is regulated not only primarily by astrocyte-endothelial communication but also by neuron–endothelial communication. The distribution of SHH and which types of cells had elevated SHH expression after oestrogen stimulation remain worthy of investigation. In our experiments, oestrogen was expressed in both astrocytes and neurons, and an increase in SHH content was found in both types of cells without any change in cell count after oestrogen intervention, again confirming the relationship between SHH and oestrogen from the previous section.
Alternations in SHH after SAH have not been uniformly identified in previous studies. Hu et al found that SHH expression was downregulated 48 hours after SAH, and Zuo et al also found that expression was downregulated 24 hours after SAH.12 25 In contrast, Tao et al found that SHH was elevated 48 hours after SAH.26 Meanwhile, Hu et al and Li et al had different results for the changes in the SHH signalling pathway component GLI1 after SAH. In our study, we added the post-SAH time point setting compared with previous works and obtained similar results to Hu et al 72 hours after SAH, SHH protein in the temporal cortex was significantly decreased. Interestingly, this result was consistent with the decrease in TJs and brain oedema to a trough at 72 hours after SAH. For changes in downstream components after SAH, GLI1 and SMO were also both downregulated by changes in SHH after SAH. This process was reversed through the oestrogen receptor signalling pathway when oestrogen was given.
Tight junctions between hBMVEs are important proteins that maintain BBB stability.27 There are two peaks of TJs reduction after SAH, 3 hours and 72 hours.28 The mechanism of reduction may be related to the activation of apoptotic pathways in endothelial cells after SAH. SHH from extraendothelial cells increases the release of GLI1 into the nucleus by binding to PTCH1 on the endothelial cell membrane. The entry of GLI1 into the nucleus promotes the transcription of SOX18 and increases the expression of proteins such as claudin-3, claudin-5 and occludin.11 Seventy-two hours after SAH, our experiments revealed that the expression level of TJs was downregulated. In vivo, we found that TJs on endothelial cells were upregulated after oestrogen intervention through the SHH pathway. In vitro, by mimicking SAH for 24 hours, we found a decrease in the amount of TJs in endothelial cells, and SHH alleviated this decrease.
TJs are key proteins that maintain BBB stability. However, it is not known whether TJs can properly assemble into cell junctions after oestrogen stimulation to play a role in maintaining the stability of the BBB.29 There is no suitable method to detect this assembly process. In our experiments, we assessed BBB stability by performing a western blot assay of albumin and measurements of Evans blue absorbance and brain water content. Increased TJs after oestrogen stimulation reduced the increased BBB permeability due to SAH, indicating that these TJs are correctly assembled to the cell junctions.
Neurological injury after SAH is affected not only by oedema due to increased BBB permeability but also by decreased cerebral blood flow due to vasospasm and increased intracranial pressure. Previous studies have also found that oestrogen can reduce vasospasm after SAH.7 Whether oestrogen can improve neurological function after SAH by reducing acute brain injury after SAH through an improvement effect on BBB function remains unclear. In previous studies, there were no relevant behavioural tests on the role of SHH and oestrogen post-SAH.12 25 In our experiments, oestrogen improved neurological function in rats after SAH, and the reversal occurred when the SHH signalling pathway was blocked by cyclopamine. However, whether this conclusion can be generalised to clinical patients still needs to be confirmed by more clinical trials in the future.
Our experiments suggest that oestrogen is a protective factor for neurological function after SAH, which is consistent with the results of previous experiments. However, this conclusion was verified in male rats. We used only male rats to avoid endogenous oestrogen fluctuations in female rats and pathological alterations of sex hormones such as progesterone brought about by ovariectomy in female rats. Whether female rats differ from male rats after experimental SAH remains to be analysed in future studies. Meanwhile, we still note that clinical data suggest a higher mortality rate in women than in men after SAH, which is inconsistent with the results of many previous experiments. This may be related to differences in androgen, progesterone, and some nongender-related factors, such as smoking.30 The effect of a single-factor intervention of oestrogen in experiments may be masked by the effect of other factors in clinical patients.
Overall, our experiments explain the role of oestrogen in improving BBB stability and ameliorating neurological impairment after SAH through upregulation of the SHH signalling pathway by oestrogen receptors (figure 7). This study may help to explain the clinical sex differences in relation to perihaematomal brain oedema and the prognosis of neurological function after SAH. However, how to specifically agonise oestrogen receptors in the brain and thus reduce brain damage caused by SAH while avoiding the systemic side effects of oestrogen application remains to be studied in the future.
Data availability statement
Data are available on reasonable request.
Patient consent for publication
This study involves human participants and was approved by the Ethics Committee of the First Affiliated Hospital of Soochow University (2022-294). Participants gave informed consent to participate in the study before taking part.
JZ and HL are joint first authors.
JZ and HL contributed equally.
Contributors GC is responsible for the overall content as the guarantor. JZ and HL contributed equally to this work. GC and WY were the cocorresponding authors involved in the conception, design and execution of the study. All authors made significant contributions to the manuscript.JZ and HL contributed to the manuscript writing, ELISA, Western blot assay, immunofluorescence assay and data analysis. JZ contributed to the GTEx data analysis. ZX and JL were involved in the neurobehavioural experiments. CC and HS were involved in animal experiments, rat brain tissue collection and human CSF collection. XL performed the data analysis. GC and WY completed the study design and manuscript review and revision.
Funding This work was supported by the National Natural Science Foundation of China (81830036, 82220108012, 82071307 and 82271362), Natural Science Foundation of Jiangsu Province (BK20211552) and Gusu Health Personnel Training Project (GSWS2020022 and GSWS2019030).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.