Abstract

Background: Endothelial dysfunction plays a key role in intracerebral hemorrhage (ICH), with vascular endothelial cadherin (VE-cadherin) being essential for maintaining blood vessel integrity and the blood–brain barrier. Hypertension increases ICH risk by damaging blood vessel integrity due to inflammatory cascades. Lower VE-cadherin levels in hypertensive patients suggest its potential as an early predictor of ICH risk.

Methods: This 12-month study included 40 hypertensive control patients and 40 hypertensive patients with nonlobar ICH. Blood samples were analyzed using enzyme-linked immunosorbent assays to measure VEcadherin, interferon gamma, and interleukin-17 levels. Receiver operating characteristic analysis determined a VE-cadherin cutoff value, and a regression model assessed its association with ICH risk.

Results: A VE-cadherin cutoff value of 400.8 pg/dL was identified, with higher levels independently linked to lower odds of nonlobar ICH. VE-cadherin was the only biomarker that remained statistically significant in the regression model. These findings suggest that reduced VE-cadherin levels contribute to ICH development, and its measurement may help identify high-risk patients for early intervention.

Conclusion: VE-cadherin dysfunction in hypertension may serve as a predictor of nonlobar ICH risk. Its protective role highlights its potential as a biomarker for risk assessment and prevention strategies in hypertensive patients. These findings may pave the way for targeted interventions in hypertensive populations, warranting further research to confirm its clinical utility.

Keywords: Adherens junction, Gap junction, Hypertension, Intracerebral hemorrhage, Stroke, Vascular endothelial-cadherin

INTRODUCTION

Intracerebral hemorrhage (ICH) remains a significant clinical challenge, particularly among hypertensive patients, where endothelial dysfunction plays a critical role in disease progression. Nonlobar ICH, which primarily affects deep brain structures such as basal ganglia, thalamus, and brainstem, is strongly linked to chronic hypertension and small vessel diseases.[ 12 ] Nonlobar ICH, unlike lobar ICH that is often associated with cerebral amyloid angiopathy, is characterized by lipohyalinosis and fibrinoid necrosis that leads to vessel fragility and rupture.[ 17 ] The disruption of vascular endothelial cadherin (VE-cadherin) in hypertensive patients may contribute to blood–brain barrier (BBB) breakdown and vessel fragility, increasing susceptibility to hemorrhage.[ 13 ] Understanding the role of VE-cadherin in nonlobar ICH could provide valuable insights into its pathophysiology and help identify biomarkers for risk stratification and prevention in hypertensive populations given its high morbidity and mortality.

VE-cadherin is an endothelial-specific adhesion molecule crucial for maintaining cell junction integrity, influencing processes such as cell proliferation, apoptosis, and the functioning of VE growth factor receptors, making it vital for embryonic angiogenesis.[ 10 ] Interactions between VEcadherin and proteins such as α-catenin, β-catenin, and p120-catenin also play a role in controlling permeability as well as managing adherens junction (AJ) integrity through intracellular signaling.[ 12 , 15 ] Other interactions between VEcadherin and other molecules include platelet endothelial cell adhesion molecule-1 (PECAM-1) to regulate endothelial function and vascular homeostasis and VE growth factor receptor 2 (VEGFR2) with mitogen-activated protein kinase (MAPK) signaling pathways to maintain endothelial responses to hemodynamic changes.[ 7 ] Dysregulation of VEcadherin has been implicated in various vascular pathologies, including BBB disruption, a key factor in the pathogenesis of hypertensive ICH,[ 16 ] prompted by inflammatory mediators, including cytokines, leukocytes, and growth factors, which stimulate the formation of radial stress fibers and actomyosin contraction, ultimately leading to the retraction of junctions.[ 3 ]

Chronic hypertension induces endothelial remodeling, leading to altered tight junction composition, reduced VEcadherin expression, and increased BBB permeability.[ 2 ] Cells exposed to elevated pressure exhibit an elongated and convoluted morphology, along with stratified structures, and demonstrate diminished expression of VE-cadherin, leading to tissue injury, exacerbated by oxidative stress from vasoactive peptides such as angiotensin-II and endothelin-1. These factors create favorable conditions for the development of damage-associated molecular patterns (DAMPs) and neoantigens, including isosketal protein adducts. DAMPs activate innate immunity through toll-like receptors on type 1 macrophages, dendritic cells, and natural killer-β cells, stimulating interleukin (IL)-23 production, T cell proliferation, and the subsequent release of pro-inflammatory cytokines such as IL-17A, interferon gamma (IFN-γ), and tumor necrosis factor-alpha.[ 2 ] IFN-γ specifically disrupts tight junctions and contributes to endothelial dysfunction in the BBB by impairing pericyte function and increasing Rho kinase activity.[ 11 ] Simultaneously, IL-17 compromises tight junction integrity by increasing the permeability of brain endothelial cells and reducing the expression of tight junction molecules, including occludin and zonula occludens-1 (ZO-1).[ 25 ] The combined effects of IFN-γ and IL-17 lead to significant damage to tight junctions in the BBB, resulting in endothelial cell dysfunction and increased barrier permeability. This dysfunction facilitates the leakage of leukocytes and pro-inflammatory cytokines into the walls of small blood vessels in the brain, disrupting autoregulation. Consequently, this cascade produces fragile blood vessel walls and increased cerebral blood flow, ultimately heightening the risk of blood vessel rupture.[ 23 ]

This emerging evidence suggests that serum VE-cadherin levels may serve as a biomarker for endothelial dysfunction in hypertensive individuals.[ 9 ] Variability in VE-cadherin expression raises important questions about its role in mediating ICH risk. While some hypertensive patients develop ICH, others remain unaffected despite similar clinical profiles.[ 18 ] This suggests that differences in endothelial resilience, potentially reflected in VE-cadherin levels, could influence disease susceptibility. Dysfunctional VE-cadherin signaling driven by inflammatory mediators and mechanical stress may lead to junctional retraction and increased vascular permeability, thereby facilitating hemorrhagic events.[ 18 ]

This review aims to consolidate current knowledge on VEcadherin’s role in endothelial health, ultimately shedding light on its role in the pathophysiology of nonlobar ICH. By elucidating its mechanistic contributions, we seek to establish VE-cadherin as a potential biomarker for hemorrhagic stroke risk stratification, paving the way for targeted interventions in hypertensive populations.

MATERIALS AND METHODS

Study population

Since there were no previous clinical studies analyzing the effect of VE-cadherin on hypertension or stroke, we calculated statistical power from an in vitro study by Ohashi et al.[ 21 ] In their study, application of pressure resulted in 51.1% of VE-cadherin expression. Based on their result, the required sample size to reach 80% power was 20. This study was then conducted over 12 months, collecting a sample of 80 participants through consecutive sampling. The sample included 40 hypertensive patients with spontaneous nonlobar ICH and 40 hypertensive patients without such hemorrhage. We excluded patients with other vascular abnormalities, coagulation factor disorders, polycystic kidney disease, malignancy, trauma-related ICH, patients with a history of using antiplatelet, anticoagulant, or traditional medicines before sample retrievals, prolonged use of corticosteroids, or other hereditary conditions that involved collagen (i.e., Marfan syndrome), and patients under the age of 18. Patients with manifestations of intracranial hemorrhage were diagnosed with a head computed tomography (CT) scan. Samples were obtained from the emergency departments of two Indonesian tertiary referral hospitals and one Indonesian public health center. This study was approved by the Local Ethics Committee at the Local Institutional Research Board. Informed consent was obtained from each patient or their guardian before participation.

Data collection

All included patients were subjected to a detailed medical history and thorough clinical examination. These included demographic data, medical history focused on hypertension (i.e., signs and symptoms, hypertension history, blood pressure examination, routine use of medications, past history or familial history of stroke, smoking history), and laboratory parameters. The location of nonlobar intracranial hemorrhage was determined using a head CT scan. We collected blood samples to determine IL-17, IF-γ, and VE-cadherin levels. We collected admission laboratory parameters, which include blood glucose Hemoglobin A1C level, renal functions, and lipid profile.

Sample collection

Blood samples for measuring IFN-γ, IL-17, and VEcadherin were collected from all subjects using aseptic techniques, drawn into serum separator and three different ethylenediamine tetraacetic acid tubes, processed to separate serum and plasma, and stored at −40°C until analysis. Laboratory examinations were conducted directly at the clinical pathology laboratories of the local hospital using enzyme-linked immunosorbent assays (ELISA) method. The ELISA test procedure starts with preparing reagents, standards, and samples at room temperature and storing any unused wells at 2-8°C. 50 μL of standard solution is then added to standard wells, while 40 μL of the sample, 10 μL of biotinylated antibody, and 50 μL of streptavidin-horseradish peroxidase are added to each sample well. The mixture is thoroughly combined, and the plate is sealed and incubated at 37°C for 60 min.

After incubation, the plate is washed 5 times with a wash buffer. Next, 50 μL each of substrate solutions A and B are added to every well, and the plate is incubated in the dark for 10 min at 37°C. A final addition of 50 μL of stop solution causes a color change from blue to yellow. The absorbance (optical density [OD] value) is measured immediately at 450 nm using a microplate reader. This reading is proportional to the concentration of each analyte in the samples. Then, we plot the standard curve by plotting OD values on the y-axis against known concentrations on the x-axis. Sample concentrations were determined by comparing sample OD values to the curve to calculate unknown concentrations based on interpolation. All experiments were performed by a professional laboratory technician in accordance with the Helsinki Declaration.

Glucose analyses, as well as other important biochemical parameters, were conducted using the ARCHITECTplus 4000 chemical analyzer. In addition to glucose testing, the study also included assessments of HbA1c, urea, creatinine, and cholesterol levels. To ensure the accuracy and sensitivity of cholesterol analysis, high-performance liquid chromatography was employed. This advanced technique allows for the effective separation and quantification of different cholesterol fractions, thereby facilitating a comprehensive assessment of lipid profiles.

Radiological examination

Screening for spontaneous ICH was conducted through noninvasive imaging using CT scans. A standardized protocol was followed to evaluate the size and location of the hemorrhage, periventricular white matter changes, and the presence of cerebral microbleeds using a 128-slice CT scanner (Siemens SOMATOM, Germany). CT scan imaging was performed with patients in the supine position, ensuring the head was aligned in the neutral position with a gantry tilt to minimize beam-hardening artifacts from the skull base. The scans were standardized to capture the entire brain from the vertex to the foramen magnum. An experienced neuroradiologist, blinded to the patients’ clinical data, analyzed the CT images to ensure consistency in the measurement of hematoma size and distribution.

Blood pressure levels were measured using an electronic sphygmomanometer (Omron, Japan) throughout hospitalization. Hypertension was assessed based on systolic and diastolic pressures, which were monitored regularly to determine the severity of hypertensive disease and its potential contribution to the ICH.

Additional patient characteristics and family medical history were obtained through structured interviews and standardized questionnaires.

Statistical analysis

Patient’s demographic characteristics and laboratory values were categorized according to the presence of ICH. Categorical variables were compared using Chi-square test. When the expected frequency of one or more cells is <5, Fisher’s exact test was employed. Distribution of continuous variables was assessed and a comparison between the two groups was done using Student’s t-test or Mann-Whitney U-test according to their distribution. Any variables showing statistical significance were then analyzed in a multivariate logistic regression model. For continuous variables, we first determined their respective cutoff values using receiver operating characteristic (ROC) analysis before categorizing them into two and analyzing them in the regression model. Cut-off values were determined using the Youden’s index (sensitivity + specificity −1) to find the best balance between sensitivity and specificity. All statistical significance was set at P < 0.05. Statistical analysis was performed using the Statistical Package for the Social Sciences (version 25.0.0).

RESULTS

We included 40 patients in each group, for a total of 80 patients. From this sample, we found that patients with ICH were younger (53.1 vs. 61.5; P < 0.001), more likely to smoke (35.0% vs. 7.5%; P = 0.003), and more likely to have a history of previous stroke (15.0% vs. 0.0%; p 0.026). For the laboratory results, patients with ICH had a significantly higher value of blood urea nitrogen (31 vs. 21; P = 0.002) and IFN-γ (71.6 vs. 59.9; P = 0.044) while having a lower value of VE-cadherin (347.0 vs. 769.2; P = 0.007) [ Table 1 ].

Table 1:

Characteristics of study participants.

We performed ROC analysis for continuous variables that showed significant statistical differences between the two groups [ Figure 1 ]. To be useful clinically, a specific cutoff point is required for continuous variables. With a cutoff point, patients can then be categorized into either a high or low risk group. Using the Youden’s index, for age, we determined 55 years old as the best cut-off value (Youden’s index: 1.48; Area Under the Curve [AUC]: 0.73 [95% confidence interval [CI]: 0.62−0.85]; P < 0.001). For VE-cadherin, it was 400.8 (Youden’s index: 1.43; AUC 0.71 [95% CI 0.60−0.83]; P = 0.001); for IFN-γ, it was 62.5 (Youden’s index: 1.35; AUC: 0.65 [95% CI 0.53−0.78]; P = 0.018).

Figure 1:

Receiver operating characteristic (ROC) (a) curve of age, (b) vascular endothelial cadherin, (c) interferon-gamma.

We then dichotomized age, VE-cadherin, and IFN-γ according to those cut-off values and performed logistic regression with other variables, which showed statistically significant differences between the two groups [ Table 2 ].

Table 2:

Cut-off value outcome in study participants.

From the regression model, only VE-cadherin retained its statistical significance. VE-cadherin higher than 400.8 pg/dL was independently associated with lower odds of having ICH (odds ratio 0.219 [95% CI 0.064−0.753]; P = 0.016). Figure 2 shows the difference in VE-cadherin between the two groups [ Figure 2 ].

Figure 2:

(a) Median error bar chart and (b) count cadherin bar chart showring the difference between hypertension only and hypertension + intracerebral hemorrhage (ICH) groups.

DISCUSSION

To the extent of our knowledge, this study is the first to report serum VE-cadherin levels in patients with ICH. VEcadherin, a key component of endothelial AJs, plays a critical role in vascular integrity by facilitating homotypic adhesion between endothelial cells and linking to intracellular signaling pathways that regulate cytoskeletal remodeling. This adhesion is mediated through a pericellular zipper-like structure along the lateral endothelial membrane, which connects adjacent cells.[ 6 ] Aside from physically connecting endothelial cells together, VE-cadherin is also linked to multiple intracellular components. The VE-cadherin and β-catenin complex interacts with α-catenin, binding it with other proteins such as vinculin and eplin that have important roles in remodeling the cytoskeleton and modulating vascular strength and integrity, tailoring it to the different needs of the vasculature.[ 9 ] Loss of VE-cadherin function has been demonstrated in vivo to cause leakage and hemorrhage.[ 12 , 17 ]

Our findings demonstrate a statistically significant difference in serum VE-cadherin levels between hypertensive patients with and without ICH. Although causality cannot remain undetermined, this disparity suggests a potential role for VE-cadherin in differentiating hypertensive individuals at higher risk of ICH. Given its mechanosensory function, VE-cadherin in conjunction with PECAM-1 responds to hemodynamic forces.[ 4 ] Shear stress and pressure-induced conformational changes in PECAM-1 activate Src family kinase Fyn, leading to the phosphorylation of VEGFR2. This initiates downstream signaling through phosphoinositide 3-kinase and MAPK, pathways involved in vascular tone regulation, angiogenesis, adhesion, and leukocyte recruitment.[ 3 , 16 ] MAPK also mediates vasoconstriction and inflammation, further linking endothelial dysfunction to ICH pathogenesis.[ 7 ]

Despite limited studies on hypertension’s impact on VEcadherin, Ohashi et al. described bovine aortic endothelial cells (BAEC) exposed to hydrostatic pressure as exhibiting a decreased amount of VE-cadherin than control on their peripheries.[ 21 ] They also found that those BAECs developed a tortuous shape, random orientation, and thick stress fibers, which occurred in BAECs in which the VE-cadherin expression was inhibited by an antibody, suggesting that the downregulation of VE-cadherin was the cause of such morphological changes.[ 22 ] Similarly, Nikitopoulou et al. studied the effect of pulmonary hypertension on the endothelial cells of rats. They found that pulmonary hypertension downregulated the expression of VE-cadherin on the surface of the endothelial cells. Interestingly, they also found that after the initial decline, VE-cadherin expression was eventually restored by day 30.[ 20 ] These findings align with our results, suggesting that hypertension reduced the expression of VE-cadherin on the endothelial surface, although the exact mechanism remains to be elucidated.

The pathogenesis of nonlobar (deep) ICH is multifactorial, with chronic hypertension being the predominant risk factor, increasing the likelihood of ICH by 2–4-fold. Several theories have been suggested to explain the link between hypertension and lipohyalinosis, such as the formation of Charcot–Bouchard aneurysms, which weaken the affected vessels, making them more prone to rupture.[ 19 ] Hypertension is thought to exert abnormal mechanical forces on the vessel and cause fibrinoid necrosis, which would eventually lead to lipohyalinosis. However, as not all hypertensive patients suffer from ICH, other factors including diabetes mellitus, aging, oxidative stress, dyslipidemia, and familial history have all been found to be involved in the process of lipohyalinosis and ICH.[ 9 , 23 ] One interesting study by Zhong et al. reported a possible role of VE-cadherin in atherosclerosis and lipohyalinosis. In their study, they used an oxidized low-density lipoprotein (oxLDL) to create a model of endothelial dysfunction found in atherosclerotic vessels. Exposure to oxLDL internalized VEcadherin and disrupted endothelial integrity. Moreover, they also found that high-shear stress can reverse this effect.[ 25 ]

Our study found that hypertensive patients with ICH exhibited a lower serum level of VE-cadherin, and the difference remained significant even after adjusting for other confounding factors. Based on these results, it was possible that VE-cadherin might be one of the differences that caused hypertensive patients to have ICH. Lower serum VE-cadherin likely reflects reduced cleavage and release of membrane-bound VE-cadherin, indicative of more severe endothelial downregulation, therefore reducing the amount of its availability on cells’ peripheries,[ 6 ] impairing vascular integrity, predisposing affected vessels to rupture. Given its integral role in maintaining endothelial stability, VE-cadherin may be a key differentiating factor between hypertensive individuals with and without ICH.[ 5 ] The study by Zhong et al. also showed that the pathophysiologic processes of atherosclerosis and lipohyalinosis played a role in the internalization of VE-cadherin.[ 25 ] Therefore, a lower serum level might reflect a more advanced lipohyalinosis process and weaker blood vessels. In addition to VE-cadherin, other biomarkers such as PECAM-1/cluster of differentiation (CD)-31 and ZO-1 play distinct roles in endothelial function and may further contribute to the understanding of endothelial dysfunction in ICH. A recent study by Arif et al. shows that PECAM-1 is involved in leukocyte transmigration and endothelial response to shear stress, with its expression linked to inflammatory processes and vascular remodeling.[ 1 ] However, unlike VE-cadherin, PECAM-1 is not endothelial-specific, as it is also expressed in leukocytes, epithelial cells, fibroblasts, and other cell types, potentially limiting its specificity in assessing endothelial integrity.[ 14 ] A study on endothelial cell angiogenesis demonstrated that while PECAM-1 supports cell elongation, migration, and intercellular network formation, VE-cadherin is more directly involved in cell-to-cell adhesion and lumen formation, reinforcing its role in vascular stability.[ 24 ]

Similarly, ZO-1, a key tight junction protein, contributes to maintaining BBB integrity, and its expression is influenced by VE-cadherin.[ 5 ] Disruptions in ZO-1 have been associated with increased endothelial permeability in conditions such as ICH and stroke.[ 8 ] Recent findings suggest that ZO-1 interacts with YB-1 to regulate stress granule formation during angiogenesis, further linking its role to endothelial responses under pathological conditions. However, its presence in various cell types and involvement in multiple cellular processes may also affect its specificity as a biomarker of endothelial dysfunction.[ 8 ] Given its’ specific expression for assessing endothelial function and its direct role in maintaining vascular integrity, VE-cadherin may offer higher specificity in detecting endothelial dysfunctions, particularly in conditions such as ICH, compared to the other two endothelial dysfunction biomarkers. However, integrating VE-cadherin into diagnostic panels when combined with established markers such as PECAM-1 and ZO-1 could enhance risk stratification for hypertensive patients. In addition, therapeutic strategies aimed at stabilizing VEcadherin expression may mitigate endothelial damage and reduce ICH risk.

Several limitations must be acknowledged. First, our sample population exhibited baseline differences, though logistic regression analysis controlled for confounders while maintaining statistical significance. Second, we measured serum VE-cadherin rather than endothelial membrane-bound VE-cadherin, limiting direct conclusions about junctional integrity. Direct assessment would require postmortem analysis, which presents its own challenges. Third, the study was conducted in a single population, Indonesia, where dietary, smoking habits, and genetic factors may influence vascular health raising concerns about generalizability, particularly in nonhypertensive individuals and populations with different ICH risk factors.

Future studies should explore VE-cadherin’s role in diverse populations and investigate its potential integration into clinical practice.

Our findings highlight VE-cadherin’s role in endothelial dysfunction and its potential as a biomarker for ICH risk in hypertensive individuals. Further research is needed to elucidate its mechanistic involvement and therapeutic implications, potentially paving the way for targeted interventions to reduce ICH incidence.

CONCLUSION

This study demonstrates that hypertensive patients with spontaneous nonlobar ICH exhibit significantly lower serum VE-cadherin levels compared to those without ICH. These findings suggest a potential link between VEcadherin downregulation – possibly due to hypertension-induced endothelial dysfunction – and the pathogenesis of nonlobar ICH. By highlighting VE-cadherin as a potential early biomarker, this study contributes to the limited body of research on endothelial injury in hypertensive cerebrovascular disease. However, as this was a cross-sectional study, causality cannot be established. Further longitudinal and interventional studies are warranted to validate these findings and assess the therapeutic potential of targeting VE-cadherin to prevent ICH in high-risk populations.

Ethical approval:

The research/study was approved by the Institutional Review Board at the University of Riau Ethical Committee, number B/159/UN19.5.1.1.8/8/UEPKK/2023, dated October 06, 2023.

Declaration of patient consent:

The authors certify that they have obtained all appropriate patient consent.

Financial support and sponsorship:

Nil.

Conflicts of interest:

There are no conflicts of interest.

Use of artificial intelligence (AI)-assisted technology for manuscript preparation:

The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.

Disclaimer

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Journal or its management. The information contained in this article should not be considered to be medical advice; patients should consult their own physicians for advice as to their specific medical needs.

References

1. Arif N, Zinnhardt M, Nyamay’Antu A, Teber D, Brückner R, Schaefer K. PECAM-1 supports leukocyte diapedesis by tension-dependent dephosphorylation of VE-cadherin. EMBO J. 2021. 40: e106113

2. Caillon A, Paradis P, Schiffrin EL. Role of immune cells in hypertension. Br J Pharmacol. 2019. 176: 1818-28

3. Cong X, Kong W. Endothelial tight junctions and their regulatory signaling pathways in vascular homeostasis and disease. Cell Signal. 2020. 66: 109485

4. Conway D, Schwartz MA. Lessons from the endothelial junctional mechanosensory complex. F1000 Biol Rep. 2012. 4: 1

5. Dejana E, Tournier-Lasserve E, Weinstein BM. The control of vascular integrity by endothelial cell junctions: Molecular basis and pathological implications. Dev Cell. 2009. 16: 209-21

6. Dejana E, Giampietro C. Vascular endothelial-cadherin and vascular stability. Curr Opin Hematol. 2012. 19: 218-23

7. Edvinsson L, Krause DN. Switching off vascular MAPK signaling: A novel strategy to prevent delayed cerebral ischemia following subarachnoid hemorrhage. Transl Stroke Res. 2024. 16: 952-61

8. El Bakkouri Y, Chidiac R, Delisle C, Corriveau J, Cagnone G, Gaonac’h-Lovejoy V. ZO-1 interacts with YB-1 in endothelial cells to regulate stress granule formation during angiogenesis. Nat Commun. 2024. 15: 4405

9. Giannotta M, Trani M, Dejana E. VE-cadherin and endothelial adherens junctions: Active guardians of vascular integrity. Dev Cell. 2013. 26: 441-54

10. Harris ES, Nelson WJ. VE-cadherin: At the front, center, and sides of endothelial cell organization and function. Curr Opin Cell Biol. 2010. 22: 651-8

11. Jansson D, Scotter EL, Rustenhoven J, Coppieters N, Smyth LC, Oldfield RL. Interferon-γ blocks signalling through PDGFRβ in human brain pericytes. J Neuroinflammation. 2016. 13: 249

12. Keep RF, Andjelkovic AV, Xiang J, Stamatovic SM, Antonetti DA, Hua Y. Brain endothelial cell junctions after cerebral hemorrhage: Changes, mechanisms and therapeutic targets. J Cereb Blood Flow Metab. 2018. 38: 1255-75

13. Lampugnani MG, Dejana E, Giampietro C. Vascular endothelial (VE)-cadherin, endothelial adherens junctions, and vascular disease. Cold Spring Harb Perspect Biol. 2018. 10: a029322

14. Lertkiatmongkol P, Liao D, Mei H, Hu Y, Newman PJ. Endothelial functions of platelet/endothelial cell adhesion molecule-1 (CD31). Curr Opin Hematol. 2016. 23: 253-9

15. Li S, Ai N, Shen M, Dang Y, Chong CM, Pan P. Discovery of a ROCK inhibitor, FPND, which prevents cerebral hemorrhage through maintaining vascular integrity by interference with VE-cadherin. Cell Death Discov. 2017. 3: 17051

16. Li W, Chen Z, Chin I, Chen Z, Dai H. The role of VE-cadherin in blood-brain barrier integrity under central nervous system pathological conditions. Curr Neuropharmacol. 2018. 16: 1375-84

17. Magid-Bernstein J, Girard R, Polster S, Srinath A, Romanos S, Awad IA. Cerebral hemorrhage: Pathophysiology, treatment, and future directions. Circ Res. 2022. 130: 1204-29

18. Maltabe V, Kouklis P. Vascular endothelial (VE)-cadherin-mediated adherens junctions involvement in cardiovascular progenitor cell specification. Int J Dev Biol. 2022. 66: 77-83

19. Morello F, Perino A, Hirsch E. Phosphoinositide 3-kinase signalling in the vascular system. Cardiovasc Res. 2009. 82: 261-71

20. Nikitopoulou I, Orfanos SE, Kotanidou A, Maltabe V, Manitsopoulos N, Karras P. Vascular endothelial-cadherin downregulation as a feature of endothelial transdifferentiation in monocrotaline-induced pulmonary hypertension. Am J Physiol Lung Cell Mol Physiol. 2016. 311: L352-63

21. Ohashi T, Sugaya Y, Sakamoto N, Sato M. Hydrostatic pressure influences morphology and expression of VE-cadherin of vascular endothelial cells. J Biomech. 2007. 40: 2399-405

22. Regenhardt RW, Das AS, Ohtomo R, Lo EH, Ayata C, Gurol ME. Pathophysiology of lacunar stroke: History’s mysteries and modern interpretations. J Stroke Cerebrovasc Dis. 2019. 28: 2079-97

23. Sierra C, Coca A, Schiffrin EL. Vascular mechanisms in the pathogenesis of stroke. Curr Hypertens Rep. 2011. 13: 200-7

24. Suya Y, Graham J, Kahn JW, Schwartz EA, Gerritsen ME. Functional roles for PECAM-1 (CD31) and VE-cadherin (CD144) in tube assembly and lumen formation in three-dimensional collagen gels. Am J Pathol. 1999. 155: 887-95

25. Zhong T, Li Y, He X, Liu Y, Dong Y, Ma H. Adaptation of endothelial cells to shear stress under atheroprone conditions by modulating internalization of vascular endothelial cadherin and vinculin. Ann Transl Med. 2020. 8: 1423