Angiotensin-(1–7) suppresses pyroptosis in cerebral endothelium to decrease blood–brain barrier permeability and cognitive impairments in sepsis

Abstract

Objective

This study aimed to explore the influence of the angiotensin-(1–7)/Mas receptor and angiotensin II/angiotensin II type 1 receptor pathways on pyroptosis during sepsis and their subsequent effects on cognitive function.

Methods

Adult C57BL/6 mice were subjected to cecal ligation and puncture to induce sepsis. Brain microvascular endothelial cells were treated with angiotensin II and angiotensin-(1–7) to evaluate their impact on pyroptotic processes. Cognitive performance was assessed using the Morris water maze method, and blood–brain barrier permeability was quantified using Evans blue staining.

Results

Compared with the sham group, sepsis induced sustained activation of the angiotensin II/angiotensin II type 1 receptor pathway, whereas the angiotensin-(1–7)/Mas receptor pathway was progressively suppressed. Genetic ablation of cysteine-dependent aspartate protease-1 significantly attenuated pyroptosis in brain endothelial cells, decreased blood–brain barrier permeability, and enhanced cognitive function in septic mice compared with that in the cecal ligation and puncture group. Angiotensin-(1–7) treatment improved cognitive function in septic mice and significantly suppressed angiotensin II-induced pyroptosis, with these effects reversed by the Mas receptor antagonist A-779.

Conclusions

This study identified a novel mechanism in which angiotensin-(1–7) selectively suppresses angiotensin II-induced pyroptosis in brain endothelial cells, consequently ameliorating cognitive deficits during sepsis.

Introduction

Sepsis, a life-threatening condition characterized by high mortality rates, remains a formidable challenge in critical care medicine despite adherence to the “Surviving Sepsis Campaign” guidelines, which provide comprehensive treatment frameworks. ¹ Despite advances, persistent issues, including low recovery rates, high incidence, and mortality, substantially compromise patient outcomes, underscoring the urgent need for novel therapeutic strategies. ² Sepsis-associated encephalopathy (SAE), characterized by cognitive deficits in sepsis survivors, not only elevates mortality but also complicates patient management and public health efforts. ³ However, the mechanisms underlying SAE remain unclear.

A well-recognized hallmark of SAE is increased blood–brain barrier (BBB) permeability, representing a critical pathological basis for brain dysfunction in sepsis. The BBB constitutes a highly specialized interface maintained by the synergistic interplay among cerebral endothelial cells, astrocytes, microglia, and pericytes, serving as a selective filter for molecular exchange. ⁴ Its dynamic responsiveness allows functional adjustment—through mechanisms such as tight junction regulation—to maintain brain homeostasis by preventing the entry of harmful substances and ensuring optimal nutrient supply. ⁵ BBB integrity preservation is essential for neuronal and brain health. During sepsis, elevated central nervous system (CNS) albumin levels are observed in both animal models and human studies, indicative of BBB disruption.^6,7 This disruption permits inflammatory mediator and reactive oxygen species (ROS) infiltration, triggering glial activation, neurotransmitter dysregulation, mitochondrial damage, and neuronal apoptosis, culminating in brain dysfunction.^8–10 Nevertheless, the specific molecular mechanisms driving BBB permeability during sepsis remain unclear. Critically, whether endothelial cells undergo pyroptosis post sepsis, contributing to BBB dysfunction, remains unknown.

Cysteine-dependent aspartate proteases (caspases) are central to programmed cell death and inflammation, orchestrating inflammatory responses, cell proliferation, and differentiation.^11–13 These enzymes critically regulate cell death pathways. Inflammasome activation drives caspase-1 activation in multiple immune cells, thereby mediating innate immunity. Caspase-1 processes pro- interleukin 1 beta (IL-1β) and pro-interleukin 18 (IL-18) into bioactive cytokines, promoting inflammation and pyroptosis via gasdermin D (GSDMD) cleavage—a process characterized by plasma membrane rupture with consequent release of proinflammatory mediators.^14,15 The role of pyroptosis in cerebral endothelial cells during sepsis and its regulatory mechanisms remain to be elucidated.

The renin–angiotensin–aldosterone system (RAAS) critically regulates fluid balance, inflammation, and vascular homeostasis and involves multiple interrelated axes of bioactive molecules.^16–19 In sepsis, RAAS undergoes marked alterations, with renin, angiotensin (Ang) II, and Ang-(1–7) expressions correlating with disease severity and patient outcomes. Ang II, via the action of Ang II type 1 receptor (AT1R), induces vasoconstriction, sympathetic activation, oxidative stress, and immune cell infiltration.^20–23 Conversely, Ang-(1–7) confers protection through the Mas receptor (MasR), antagonizing deleterious Ang II/AT1R effects.^24,25 The regulatory potential of these axes regarding cerebral endothelial pyroptosis and SAE pathogenesis remains poorly defined, representing a critical research gap.

Herein, we elucidated the dynamic interplay between Ang-(1–7)/MasR and Ang II/AT1R signaling pathways responsible for modulating cerebral endothelial pyroptosis post sepsis. Specifically, we examined their impact on BBB permeability and cognitive dysfunction, thereby offering novel insights into SAE pathophysiology while identifying potential therapeutic targets for this devastating condition.

Materials and methods

Animal allocation and experimental design

This study was conducted in accordance with the Declaration of Helsinki (1975, as reviewed in 2024). The Research Ethics Committee at Guangdong Provincial People’s Hospital approved all study protocols (approval no. KY2023-079-01). A cohort of 62 adult male C57BL/6 mice aged 6–8 weeks, with an average weight of 20 ± 2 g, was meticulously divided into five experimental groups: (a) the sham surgery (Sham) group; (b) 24-h post cecal ligation and puncture (CLP) group (CLP 24 h); (c) 48-h post CLP (CLP 48 h) group; (d) 72-h post CLP (CLP 72 h) group; and (e) CLP 72 h + Ang-(1–7) group. The CLP groups underwent the specified surgical procedures, while the Sham group underwent sham laparotomy and served as the control. Mice in the CLP 72 h + Ang-(1–7) group received daily injections of Ang-(1–7) (2 mg/kg) for 3 consecutive days following CLP. ²⁶ To elucidate the role of caspase-1 in SAE, we utilized both caspase-1-deficient mice and wild-type (WT) mice, totaling 57 mice across the Sham + WT, CLP 72 h + WT, and CLP 72 h + caspase-1^-/- groups.

Sepsis model establishment through CLP

Mice were anesthetized with 3% pentobarbital sodium; then, the surgical site was prepared aseptically. Via a midline incision, the cecum was exteriorized, ligated at its base to one-third of its length, and punctured using an 18-gauge needle. After minimal fecal extrusion, the cecum was replaced, and the incision closed layer wise. Postoperatively, the wound was disinfected again using povidone–iodine, and mice were monitored until complete anesthetic recovery.

Murine sepsis score (MSS)

To assess the severity of sepsis in the CLP model, mice were evaluated at 72 h post CLP. The evaluation focused on several key parameters indicative of sepsis, including the mice’s appearance, consciousness level, activity, eye condition, respiratory rate, and respiratory quality. These parameters are integral components of the MSS. Mice were scored using the previously reported scoring method. ²⁷

Enzyme-linked immunosorbent assay (ELISA)

ELISA was used to quantify serum Ang II (ab209883, Abacm), Ang-(1–7) (NBP2 - 69079, R&D Systems), tumor necrosis factor-alpha (TNF-α) (ab208348, Abacm), and interleukin 6 (IL-6) (ab222503, Abacm) levels. Standards and samples were prepared, and enzyme-labeled antibodies were added. After incubation and washing, substrates A and B were introduced, and optical density (OD) values were measured within 15 min following the addition of a stop solution.

Neuropsychological assessment using the Morris water maze

The Morris water maze, comprising a 160-cm-diameter pool and a 12-cm-diameter submerged platform placed 1 cm below the surface, was interfaced with an automated image acquisition system. Four quadrants were delineated, and water temperature was maintained at 22°C ± 2°C. Beginning on postoperative day 7, the mice underwent adaptive training, starting from a fixed position facing the pool wall in each quadrant, with a maximum of 120 s to find the platform. Failure to locate the platform within this interval resulted in manual guidance to the target, followed by a 15-s dwell period. Escape latency was recorded on days 2–5, and platform crossing frequency in the previous platform zone was assessed on day 6.

BBB permeability assessment using Evans blue staining

To assess BBB permeability, mice received intravenous Evans blue (4 mL/kg of a 2% solution) via the tail vein. Subsequently, the animals were euthanized under pentobarbital anesthesia and transcardially perfused with phosphate-buffered saline followed by 4% paraformaldehyde for brain harvesting and imaging. Brain samples were incubated in formamide at 60°C for 24 h, and Evans blue concentration was quantified spectrophotometrically at 620 nm.

Cultivation of brain microvascular endothelial cells (MVECs)

The bEnd.3 cell line (ATCC, VA, USA; RRID: CVCL_0170), representative of brain MVECs, was cultured in DMEM supplemented with 10% fetal bovine serum at 37°C under a 95% air/5% CO₂ atmosphere. Upon reaching confluence, cultures were passaged, and adherent cells were selected for propagation. Cells were divided into four groups: (a) control group (Con); (b) Ang II treatment group (Ang II); (c) Ang II + Ang-(1–7) treatment group (Ang II + Ang-(1–7)); and (d) combined Ang II + Ang-(1–7) + A-779 treatment group (Ang II + Ang-(1–7) + A-779). The Ang II group was exposed to Ang II at 100 nM for 24 h, while the Ang II + Ang-(1–7) treatment group received pretreatment with Ang-(1–7) at 100 nM for 1 h prior to Ang II stimulation. The combined treatment group (fourth group) was first treated with A-779 (a MasR antagonist, 1 μmol/L, Bachem) for 30 min, followed by treatment with Ang II and Ang-(1–7).

Lactate dehydrogenase (LDH) release detection

Pyroptosis assessment was performed via measurement of LDH release, utilizing a commercial assay kit (ab65393, Abcam). Following the treatment, the culture supernatant was collected and subsequently centrifuged at 4°C. In accordance with the manufacturer’s instructions, the resultant supernatant was analyzed. The LDH release was quantified using the following formula: (test sample-low control)/(high control-low control).

Protein expression evaluation using western blot analyses

Western blot analysis was conducted to assess the expression levels of angiotensin-converting enzyme (ACE), AT1R, ACE2, MasR, caspase-1, GSDMD-N, IL-1β, and IL-18 in the hippocampal tissue and cerebral vascular endothelial cells. Protein lysates were prepared, and concentrations were determined using a Bicinchoninic Acid Protein (BCA) protein assay kit. Samples were mixed with loading buffer, denatured, and separated using sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) before transfer onto polyvinylidene fluoride membrane (PVDF) membranes. Membranes were blocked with 5% BSA and incubated with antibodies against ACE (1:1000; Clone ID, EPR22291-247; Catalog No., ab254222; RRID, AB_3073965; Abcam), AT1R (1:1000; Catalog No., ab124505; RRID, AB_10976053; Abcam), ACE2 (1:1000; Catalog No., ab15348; RRID, AB_301861; Abcam), MasR (1:1000; Catalog No., ab314026; Abcam), caspase-1 (1:1000; Catalog No., ab138483; RRID, AB_2888675; Abcam), GSDMD-N (1:1000; Catalog No., WL05411; RRID, AB_3677351; WanLeiBio), IL-1β (1:1000; Clone ID, EPR23851-127; Catalog No., ab254360; RRID, AB_2936299; Abcam), IL-18 (1:1000; Clone ID, 62-3G1; Catalog No., ab191860; RRID, AB_2750951; Abcam), zonula occludens (ZO)-1(1:1000; Catalog No., ab96587; RRID, AB_10680012; Abcam), occludin (1:1000; Clone ID, EPR20992; Catalog No., ab216327; RRID, AB_2737295; Abcam), and claudin-5 (1:1000; Clone ID, E8F3D; Catalog No., 49564; RRID, AB_3065250; Cell Signaling Technology) overnight at 4°C, followed by incubation with enzyme-labeled secondary antibodies. Band intensities were quantified using ImageJ 1.39u software (National Institutes of Health, Bethesda, MD, USA), with β-actin serving as an internal loading control.

Immunofluorescence staining

Immunofluorescence staining was performed to detect the expressions of ACE, AT1R, ACE2, MasR, caspase-1, ZO-1, occludin, claudin-5, and IL-1β in cerebral endothelial cells. Mice were euthanized under pentobarbital sodium anesthesia, and brains were extracted after cardiac perfusion with phosphate-buffered saline and fixation with 4% paraformaldehyde. Brains were fixed ex vivo, dehydrated in a sucrose gradient, and sectioned into 4-μm slices. Slices were blocked and incubated with antibodies against ACE (1:100; Clone ID, EPR22291-247; Catalog No., ab254222; RRID, AB_3073965; Abcam), AT1R (1:100; Catalog No., ab124505; RRID, AB_10976053; Abcam), ACE2 (1:100; Catalog No., ab15348; RRID, AB_301861; Abcam), MasR (1:100; Catalog No., ab314026; Abcam), caspase-1 (1:100; Catalog No., ab138483; RRID, AB_2888675; Abcam), and IL-1β (1:100; Clone ID, EPR23851-127; Catalog No., ab254360; RRID, AB_2936299; Abcam) overnight at 4°C, followed by incubation with secondary antibodies (Alexa Fluor® 488-conjugated goat anti-mouse IgG and Alexa Fluor® 555-conjugated donkey anti-rabbit IgG) at room temperature. Slices were mounted with a fluorescence medium and examined under a fluorescence microscope. Corresponding in vitro cerebral endothelial cells were processed identically for immunofluorescence analysis.

Statistical analyses

Data were analyzed using Statistical Package for Social Sciences (SPSS) software (version 22.0, IBM Corp., Armonk, NY, USA) and expressed as mean ± SD. T-test was used to compare two groups, and one-way analysis of variance (ANOVA) was used for comparisons between three or four groups. Repeated measures data were analyzed using repeated measures ANOVA. Post-hoc multiple comparisons were performed using Tukey’s honest significant difference (HSD) test. A p-value <0.05 was considered statistically significant.

Results

Evaluation of the CLP model

The MSS as well as plasma TNF-α and IL-6 levels, were evaluated to confirm successful induction of sepsis in the CLP model. The MSS and levels of TNF-α and IL-6 were significantly elevated after CLP than after sham treatment (p < 0.05) (Figure 1).

Figure 1.

Evaluation of the CLP model. (a) Murine Sepsis Score. (b, c) Plasma TNF-α and IL-6 levels evaluated using ELISA. Data are presented as mean ± SD. *p < 0.05, n = 6 per group. CLP: cecal ligation and puncture; Sham: sham group; TNF-α: tumor necrosis factor-alpha; IL-6: interleukin-6; ELISA: enzyme-linked immunosorbent assay.

Enduring activation of the Ang II/AT1R pathway following sepsis

Following sepsis induction via CLP, the hippocampal concentrations of Ang II and protein expressions of ACE and AT1R were notably higher in the CLP group at the 24-h mark than in the Sham group (p < 0.05). This trend persisted, with 72-h values showing a pronounced increase compared with the 24-h values (p < 0.05) (Figure 2(a) to (d)). The expressions of ACE and AT1R in cerebral endothelial cells, as evaluated using double immunofluorescence, were upregulated in the CLP group at 24 h, with a further increase noted at 72 h post CLP (Figure 2(e) and (f)).

Figure 2.

Persistent activation of the Ang II/AT1R pathway following sepsis. (a) Plasma levels of Ang II measured using ELISA. (b, c, d) western blot analysis performed to determine the ACE and AT1R levels across groups. (e, f) Immunofluorescence reveals CD31-positive endothelial cells, ACE and AT1R expressions, and their co-localization with endothelial cell markers. Scale bars, 20 μm. Data are presented as mean ± SD. *p < 0.05, n = 6 per group. CLP: cecal ligation and puncture; Sham: sham group; Ang II: angiotensin II; AT1R: angiotensin II type 1 receptor; ACE: angiotensin-converting enzyme; ELISA: enzyme-linked immunosorbent assay; CD31: cluster of differentiation 31.

Temporal dampening of the Ang-(1–7)/MasR pathway after sepsis

Following sepsis induction via CLP, serum Ang-(1–7) levels and hippocampal protein levels of ACE2 and MasR were transiently elevated at 24 h post CLP compared with the Sham group (p < 0.05), followed by a progressive decrease, with levels at 72 h significantly lower than those at 24 h (p < 0.05) (Figure 3(a) to (d)). Double immunofluorescence analysis of cerebral endothelial cells indicated an initial increase in ACE2 and MasR expressions post CLP, which markedly diminished by 72 h (Figure 3(e) and (f)).

Figure 3.

Temporal downregulation of the Ang (1–7)/MasR pathway after sepsis. (a) Plasma levels of Ang (1–7) measured using ELISA. (b, c, d) western blot analysis performed to assess ACE2 and MasR levels across groups. (e, f) Immunofluorescence reveals CD31-positive endothelial cells, ACE2 and MasR expressions, and their co-localization with endothelial cell markers. Scale bars, 20 μm. Data are presented as mean ± SD. *p < 0.05, n = 6 per group. CLP: cecal ligation and puncture; Sham: sham group; Ang (1–7): angiotensin (1–7); ACE2: angiotensin-converting enzyme 2; MasR: Mas receptor; ELISA: enzyme-linked immunosorbent assay; CD31: cluster of differentiation 31.

Caspase-1-driven pyroptosis in cerebral endothelial cells in the context of sepsis

At 72 h post CLP sepsis induction, hippocampal expression levels of caspase-1, GSDMD-N, IL-1β, and IL-18 were significantly higher in septic mice than in sham-operated controls (p < 0.05), with these levels substantially reduced upon caspase-1 deletion (p < 0.05) (Figure 4(a) and (b)). Double immunofluorescence staining confirmed elevated cerebral endothelial cell expressions of caspase-1 and IL-1β in septic mice, with a significant reduction following caspase-1 knockout (Figure 4(c) and (d)).

Figure 4.

Caspase-1-driven pyroptosis in cerebral endothelial cells during sepsis. (a, b) western Blot analysis performed to assess levels of caspase-1, GSDMD-N, IL-1β, and IL-18 across groups. (c, d) Immunofluorescence reveals CD31-positive endothelial cells, caspase-1 and IL-1β expressions, and their co-localization with endothelial cell markers. Scale bars, 20 μm. Data are presented as mean ± SD. *p < 0.05, n = 6 per group. CLP: cecal ligation and puncture; Sham: sham group; caspase-1: cysteine-dependent aspartate protease 1; GSDMD-N: gasdermin D N-terminal fragment; IL-1β: interleukin 1 beta; IL-18: interleukin 18; CD31: cluster of differentiation 31.

Abolition of caspase-1 diminishes BBB permeability and slows down cognitive decline in sepsis

To investigate the impact of pyroptosis on BBB permeability following sepsis, Evans blue staining was performed, which revealed significant leakage in septic mice at 72 h after CLP compared with that in sham-operated mice (p < 0.05). However, this leakage was notably reduced in caspase-1 knockout mice (p < 0.05) (Figure 5(a) and (b)). Consistent with these findings, the expression levels of tight junction proteins in brain tissues, including claudin-5, occludin, and ZO-1, were significantly decreased in septic mice (p < 0.05), whereas these levels were significantly increased in caspase-1 knockout mice (p < 0.05) (Figure 5(c) and (d)). Cognitive performance, assessed using the Morris water maze, revealed extended escape latency and reduced platform crossings in septic mice, with caspase-1 knockout resulting in significant amelioration of these cognitive deficits (p < 0.05) (Figure 5(e) to (g)).

Figure 5.

Caspase-1 inhibition reduces BBB permeability and attenuates cognitive decline in sepsis. (a) Gross observation of Evans blue (EB) leakage in the brain. (b) Quantification of EB leakage in the brain. (c, d) western blot analysis performed to assess claudin-5, occludin, and ZO-1 levels across groups. (e) Escape latency in mice. (f) Frequency of platform crossings by mice. (g) Swim path of mice. *p < 0.05. For Figure 5(a), (b), (c), and (d), the sample size was 6 per group; for Figure 5(e), (f), and (g), the sample size was 14 per group. CLP: cecal ligation and perforation group; Sham: sham group; ZO-1: zonula occludens-1; caspase-1: cysteine-dependent aspartate protease 1; BBB: blood–brain barrier.

Ang-(1–7) inhibition of Ang II-triggered pyroptosis in cerebral endothelial cells

Stimulation of cerebral endothelial cells with Ang II resulted in the elevation of LDH, caspase-1, GSDMD-N, IL-1β, and IL-18 levels compared with control cells. Conversely, treatment with Ang-(1–7) significantly attenuated these increases (p < 0.05). When compared with the Ang II + Ang-(1–7) treatment group, pretreatment with A-779, a MasR antagonist, effectively reversed the effects of Ang-(1–7) (p < 0.05) (Figure 6(a) to (d)). Consistent with these findings, double immunofluorescence staining was employed to detect the co-localization of caspase-1 and IL-1β expressions with endothelial cells (Figure 6(e) and (f)).

Figure 6.

Ang (1–7) inhibits Ang II-induced pyroptosis in cerebral endothelial cells. (a–c) western blot analysis performed to assess levels of caspase-1, GSDMD-N, IL-1β, and IL-18 across groups. (d) LDH release measured to assess pyroptosis. (e, f) Immunofluorescence reveals CD31-positive endothelial cells, caspase-1 and IL-1β expressions, and their co-localization with endothelial cell markers. Scale bars, 20 μm. Data are presented as mean ± SD. *p < 0.05, n = 6 per group. CON: control group; caspase-1: cysteine-dependent aspartate protease 1; GSDMD-N: gasdermin D N-terminal fragment; IL-1β: interleukin 1 beta; IL-18: interleukin 18; A-779: a MasR antagonist; LDH: lactate dehydrogenase; Ang II: angiotensin II; Ang (1–7): angiotensin (1–7); CD31: cluster of differentiation 31.

Ang-(1–7) attenuates cognitive impairment in sepsis

In the Morris water maze, Ang-(1–7) significantly decreased escape latency and increased platform crossings in septic mice post CLP (p < 0.05) (Figure 7(b) and (c)). However, it did not improve the MSS at 72 h post CLP (p > 0.05) (Figure 7(a)).

Figure 7.

Ang (1–7) mitigates cognitive impairment in septic mice. (a) Murine Sepsis Score. (b) Escape latency in mice. (c) Frequency of platform crossings by mice. Data are presented as mean ± SD. *p < 0.05, ^nsp > 0.05. For Figure 7(a), the sample size was 6 per group; for Figure 7 (c) and (d), the sample size was 14 per group. CLP: cecal ligation and perforation group; Ang (1–7): angiotensin (1–7); ns: nonsignificant.

Discussion

In this study, we identified sustained Ang II/AT1R activation concurrent with progressive Ang-(1–7)/MasR attenuation during sepsis progression. This imbalance was characterized by persistent upregulation of Ang II, ACE, and AT1R, versus transient induction followed by reduction in Ang-(1–7), ACE2, and MasR. Post-sepsis pyroptosis activation in cerebral endothelial cells manifested as elevated caspase-1 and GSDMD-N expressions, together with IL-1β and IL-18 release. Genetic caspase-1 ablation attenuated pyroptosis, thereby ameliorating BBB disruption and cognitive impairment in septic mice, evidenced by reduced Evans blue extravasation, decreased escape latency, and enhanced platform crossing. Furthermore, ex vivo Ang-(1–7) pretreatment suppressed Ang II-induced pyroptosis, manifested by reduced caspase-1, GSDMD-N, IL-1β, and IL-18 levels.

Sepsis, a life-threatening organ dysfunction caused by dysregulated host responses to infection, commonly causes SAE, which is associated with substantial morbidity and mortality. ³ SAE manifests as acute mortality and chronic cognitive deficits, with poorly elucidated pathophysiology and no targeted therapeutics. Although an optimal SAE model remains elusive, the CLP paradigm is widely accepted as the preclinical standard.^28–30 Using this model, we evaluated sepsis-induced cognitive dysfunction using the Morris water maze, demonstrating prolonged escape latency and diminished platform crossings post CLP, thereby validating this model for SAE pathogenesis research.

Pyroptosis, a recently characterized inflammatory cell death mode, is executed by inflammatory caspases and implicated in diverse immune and inflammatory processes. Response magnitude depends on GSDMD activation mode, cellular context, pyroptotic cell type, and membrane repair capacity post pore formation.^14,15 This intricacy positions pyroptosis as a critical mediator in both homeostatic and disease states. During microbial infection, pyroptosis may protect against pathogens yet potentiate systemic inflammation via mediator release, precipitating septic shock.^31,32 The specific contribution and molecular mechanisms of pyroptosis in sepsis, particularly SAE, remain obscure. In this study, we found that post sepsis, cerebral endothelial cells exhibit elevated caspase-1 and GSDMD-N expressions, concomitant with IL-1β and IL-18 release, indicative of pyroptosis. Genetic caspase-1 ablation attenuates BBB disruption and cognitive dysfunction in septic mice, highlighting the pivotal role of caspase-1-mediated pyroptosis in sepsis-induced CNS injury. Nevertheless, the precise regulatory mechanisms governing cerebral endothelial pyroptosis post-sepsis require further investigation.

The Ang II/AT1R and Ang-(1–7)/MasR axes constitute core RAAS components, critically regulating vascular tone, sympathetic activity, oxidative stress, and immune inflammation. During sepsis progression, ACE converts Ang I to Ang II, which signals through AT1R, whereas ACE2-mediated Ang-(1–7) generation from Ang I and Ang II counteracts Ang II/AT1R signaling through MasR engagement.^20–25 We demonstrated that sustained Ang II/AT1R activation during sepsis coincides with cerebral endothelial pyroptosis, evidenced by elevated Ang II, ACE, AT1R, caspase-1, GSDMD-N, IL-1β, and IL-18 levels. Conversely, the Ang-(1–7)/MasR axis, transiently activated to counteract Ang II-induced pyroptosis, is progressively attenuated as sepsis advances, thereby exacerbating pyroptosis. Notably, ex vivo Ang-(1–7) pretreatment ameliorates Ang II-induced pyroptosis, suggesting that pharmacological activation of the suppressed Ang-(1–7)/MasR axis represents a therapeutic strategy to attenuate pyroptosis, preserve BBB integrity, and improve cognitive outcomes post-sepsis.

Our study has certain limitations. First, although we used whole-genome caspase-1 knockout mice and observed a reduction in endothelial cell pyroptosis and improvement in cognitive function, it has been reported that Caspase-1 knockout can directly alleviate neuronal apoptosis, ³³ which is one of the reasons for improved cognitive function in whole-genome caspase-1 knockout mice after sepsis. The use of endothelium-specific knockout mice would have enabled us to confirm that the pyroptosis of cerebral vascular endothelial cells is the direct driving force of SAE. Second, both ACE and ACE2 are expressed in endothelial cells and nonvascular cells. Among nonvascular cells, neurons are the primary cell type in which these enzymes are distributed.^34–36 However, our study only focused on the expressions of ACE and ACE2 in sepsis endothelial cells and did not pay attention to the in-situ expression changes of ACE/ACE2 in the brain tissue.

Our findings suggest that Ang-(1–7) represents a potential therapeutic agent for attenuating BBB disruption and ameliorating cognitive dysfunction, potentially via the suppression of Ang II-induced cerebral endothelial pyroptosis post-sepsis.