Research progress on the role of mesenchymal stem cells in pyroptosis in sepsis

Abstract

Sepsis remains a highly critical and frequent condition in intensive care units, often leading to systemic inflammatory responses, immune suppression, and multiple organ dysfunction. Recent evidence has established pyroptosis, a type of pro-inflammatory programmed cell death, as a key mechanism in the development and progression of sepsis. This process is mediated by inflammasome activation, resulting in cell swelling, membrane rupture, and the release of pro-inflammatory cellular contents that amplify inflammatory cascades. Mesenchymal stem cells (MSCs), known for their anti-inflammatory, immunomodulatory, and tissue-regenerative capacities, have attracted attention as a promising therapeutic approach for sepsis. This review summarizes the mechanistic involvement of pyroptosis in septic pathogenesis and highlights recent advances in MSC-based therapies that target pyroptotic pathways, with the goal of elucidating the potential role of MSCs in the management of sepsis.

Introduction

Sepsis constitutes a major global public health challenge, characterized by life-threatening organ dysfunction resulting from a dysregulated host response to infection [1]. With an estimated 49 million cases annually worldwide, the disease claims nearly 20% of those affected. Epidemiological studies report an in-hospital incidence of approximately 189 cases per 100,000 person-years and a mortality rate of 26.7% [2, 3]. Data from intensive care units (ICUs) in China indicate an even higher 90-day mortality rate of 35.5% among septic patients [4]. Despite advancements in clinical diagnostics and therapeutic protocols, sepsis incidence and mortality remain alarmingly high. Current management strategies, including early antibiotic administration, source control, and supportive measures such as fluid resuscitation and vasopressor therapy, have yielded only modest improvements in patient outcomes. This stagnation is largely due to the highly complex and multifactorial pathogenesis of the disease [5].

Pyroptosis, an intensely inflammatory type of programmed cell death, contributes significantly to the pathogenesis of sepsis by exacerbating immune dysregulation, endothelial dysfunction, and systemic inflammation. As an important driver of disease progression, it has attracted considerable interest as a promising therapeutic target. Emerging interventions aimed at inhibiting or modulating pyroptotic pathways represent a novel and potentially effective direction for improving sepsis treatment outcomes [6, 7].

Mesenchymal stem cells (MSCs) are adult stem cells with self-renewal capacity and multipotent differentiation potential. In addition to their regenerative capabilities, MSCs possess a wide spectrum of biological functions, including anti-inflammatory, antimicrobial, anti-apoptotic, immunomodulatory, and tissue-reparative effects. Evidence indicates that MSCs can mitigate pyroptosis in sepsis through various mechanisms, such as inhibiting inflammasome activation, reducing oxidative stress, and transferring microRNAs (miRNAs) via extracellular vesicles [8–10].

This review aims to provide a comprehensive overview of the role of pyroptosis in sepsis and to evaluate the therapeutic potential of MSCs in mitigating sepsis-induced organ dysfunction through the regulation of pyroptotic pathways. While previous reviews have separately explored the pathogenesis of pyroptosis in sepsis or the broad therapeutic potential of MSCs, there is a lack of comprehensive synthesis that critically focuses on the mechanistic crosstalk between MSC-based interventions and pyroptotic signaling networks across multiple organ systems during sepsis. This review seeks to fill this knowledge gap by systematically delineating how MSCs, and particularly their secreted extracellular vesicles (MSC-EVs), target and modulate canonical, non-canonical, and alternative pyroptosis pathways to alleviate septic organ injury. We will synthesize recent advances to elucidate the precise molecular mechanisms involved, evaluate the evidence for MSC-EVs as a novel cell-free therapeutic strategy, and discuss the associated challenges and future directions for translating these findings into clinical practice.

Pyroptosis

Definition and characteristics of pyroptosis

Pyroptosis is a pro-inflammatory form of programmed cell death triggered by microbial infection or endogenous danger signals. It is mediated by the activation of caspases such as caspase-1, −4, −5, or −11. Upon cleavage, the pyroptosis-related protein gasdermin D (GSDMD) releases its N-terminal domain (GSDMD-N), which oligomerizes and integrates into the plasma membrane to form pores. This results in cell swelling, membrane rupture, and the release of proinflammatory cytokines such as IL-1β and IL-18 [11]. As a critical innate immune defense mechanism, pyroptosis helps protect the host against pathogenic infections. Based on the initiating caspases, pyroptosis can be categorized into the canonical pathway (mediated by caspase-1) and the non-canonical pathway (mediated by caspase-4/5/11) [12, 13].

Molecular mechanisms of pyroptosis

Inflammasome formation

Inflammasome assembly is a crucial mechanism in the host defense against microbial pathogens and endogenous danger signals, playing a central role in the initiation of pyroptosis. The inflammasome is a multiprotein complex typically consisting of a sensor molecule, the adaptor protein ASC (apoptosis-associated speck-like protein containing a CARD), and an inflammatory caspases [14]. Common sensor molecules include Nod-like receptors (NLRs) such as NLRP1, NLRP3, NLRP6, NLRP7, and NLRC4, as well as PYHIN family proteins like AIM2 (absent in melanoma 2) and IFI16 (interferon-γ-inducible protein 16) [15]. NLRs function as intracellular pattern recognition receptors (PRRs) and generally comprise three domains: leucine-rich repeats (LRRs), a central NACHT domain, and an N-terminal pyrin domain or caspase activation and recruitment domain (CARD). Based on the N-terminal domain, NLRs are classified into NLRP (pyrin domain-containing) or NLRC (CARD-containing) subtypes [16, 17]. For instance, NLRC4 contains one or more N-terminal CARD domains, while NLRP3 carries a pyrin domain. Upon activation, the sensor protein recruits pro-caspase-1 via ASC. ASC contains two essential domains: a pyrin domain and a CARD domain. It interacts through its pyrin domain with upstream sensors, promoting the oligomerization of ASC into speck-like complexes. Subsequently, the CARD domain of ASC facilitates the proximity and autoactivation of pro-caspase-1. Mature caspase-1, composed of noncovalently linked p10 and p20 subunits, then cleaves proinflammatory cytokines such as IL-1β and IL-18, as well as gasdermin D (GSDMD), thereby triggering pyroptosis [18–21]. NLRP3 inflammasome activation, one of the most extensively studied pathways, typically requires two signals: a priming signal (e.g., from TLR4 activation) that upregulates NLRP3 and pro-IL-1β transcription, and an activation signal (e.g., ATP, pore-forming toxins, or crystalline substances) that promotes NLRP3 oligomerization and inflammasome assembly [22].

The canonical pyroptosis pathway

The canonical pyroptosis pathway is caspase-1-dependent and initiates upon recognition of pathogen-associated molecular patterns (PAMPs) or damage-associated molecular patterns (DAMPs) by pattern recognition receptors (PRRs). In most cases, the adaptor protein ASC (apoptosis-associated speck-like protein containing a CARD) binds to the precursor of caspase-1, facilitating the assembly of inflammasome complexes and subsequent activation of caspase-1 [23]. Alternatively, NLRC4 can directly recruit and activate caspase-1 independently of ASC [24]. Following activation, caspase-1 cleaves gasdermin D (GSDMD), releasing its N-terminal fragment (GSDMD-NT), which oligomerizes and perforates the plasma membrane by forming transmembrane pores [25]. GSDMD-NT pores facilitate the release of mature IL-1β and IL-18, and also induce osmotic cell swelling, leading to lytic cell death [26]. Concurrently, caspase-1 processes the pro-inflammatory cytokines pro-IL-1β and pro-IL-18 into their mature forms, IL-1β and IL-18, which are released through these GSDMD pores. Beyond initiating pore formation, activated GSDMD further drives progressive membrane destabilization and rupture. This terminal step of pyroptotic cell lysis is additionally facilitated by the membrane-disrupting protein NINJ1 (nerve injury-induced protein 1), which cooperates with GSDMD-induced structural damage to ensure effective execution of lytic cell death [27] (see Fig. 1).

Fig. 1.

Molecular mechanisms of pyroptosis

The non-canonical pyroptosis pathway

The non-canonical pyroptosis pathway operates independently of caspase-1 and is primarily mediated by caspase-4 and caspase-5 in humans, and caspase-11 in mice. Unlike the canonical pathway, it does not rely on traditional inflammasome complex formation. Upon cytosolic entry, lipopolysaccharide (LPS)—a key component of the cell wall in Gram-negative bacteria such as Escherichia coli and Salmonella Typhi—directly binds to and activates caspase-4/5/11. These caspases then cleave gasdermin D (GSDMD), generating its active N-terminal fragment (GSDMD-NT), which forms pores in the plasma membrane and initiates pyroptotic cell death [28, 29]. Although caspase-4/5/11 cannot directly process pro-IL-1β or pro-IL-18, GSDMD pore-induced potassium efflux stimulates NLRP3 inflammasome assembly and caspase-1 activation. This in turn leads to the maturation and release of IL-1β and IL-18, illustrating a key crosstalk mechanism between the non-canonical and canonical pyroptosis pathways [30–33].

LPS gains access to the cytosol through multiple mechanisms to engage its intracellular receptors. For instance, high-mobility group box 1 (HMGB1)—secreted by immune or endothelial cells or released from damaged cells—can form complexes with LPS that are internalized via the receptor for advanced glycation end-products (RAGE) into macrophage and endothelial lysosomes. HMGB1 promotes lysosomal membrane permeabilization, enabling LPS leakage into the cytosol and subsequent caspase-11 activation [34]. Additionally, outer-membrane vesicles (OMVs) derived from Gram-negative bacteria deliver LPS into the cytosol through endocytosis, leading to caspase-11 activation [35]. In humans, guanylate-binding protein 1 (hGBP1) binds LPS with high affinity via electrostatic interactions and recruits caspase-4, facilitating its activation [36, 37]. Pannexin-1 also serves as a key mediator in caspase-11-driven non-canonical pyroptosis: upon intracellular LPS detection, activated caspase-11 cleaves pannexin-1, resulting in ATP release that activates P2 × 7 receptors and amplifies pyroptotic signaling. Notably, pannexin-1-deficient mice are protected from endotoxin-induced shock, underscoring the role of potassium efflux as a selective regulator of non-canonical NLRP3 inflammasome activation [38] (see Fig. 1).

Pyroptotic pathways mediated by apoptotic caspases-3/8

Emerging evidence indicates that gasdermin E (GSDME) can shift caspase-3–mediated apoptosis toward pyroptosis in the presence of tumor necrosis factor-α (TNF-α) or specific chemotherapeutic agents. As one of the most evolutionarily conserved members of the gasdermin family, GSDME contains a pore-forming N-terminal domain common to all gasdermins [39]. During apoptosis, caspase-3 activated by chemotherapeutics or TNF-α cleaves GSDME, enabling its N-terminal fragment to perforate the plasma membrane. Electron microscopy reveals that permeabilized cells display characteristic pyroptotic morphology—notably large membrane blebs resulting from osmotic imbalance—suggesting that although apoptotic signaling is initiated, pyroptotic kinetics prevail and ultimately determine the mode of cell death [40–42].

Caspase-8, another apoptosis-related cysteine protease, has also been implicated in pyroptosis. Studies by Sarhan, Schwarzer, and others have shown that infection with Yersinia pestis or Yersinia pseudotuberculosis induces caspase-8–dependent cleavage of GSDMD, leading to pyroptosis and IL-1α release [43, 44]. In breast cancer cells, programmed death-ligand 1 (PD-L1) promotes a shift from TNF-α–induced apoptosis to pyroptosis: under hypoxic conditions, STAT3 and PD-L1 co-translocate to the nucleus and enhance GSDMC transcription. TNF-α then stimulates caspase-8–mediated cleavage of GSDMC, generating a pore-forming N-terminal fragment that initiates pyroptosis. Certain chemotherapeutic agents similarly induce pyroptosis through caspase-8 and GSDMC [45]. Given its ability to regulate apoptosis, necroptosis, and pyroptosis, caspase-8 is often described as a molecular rheostat that fine-tunes cell death pathways [46].

Moreover, the NLRP3 inflammasome can regulate gasdermin E (GSDME) expression in human Th17 cells. Following T-cell receptor (TCR) stimulation, the NLRP3 inflammasome detects elevated calpain activity driven by calcium signaling, which promotes the maturation of pro-IL-1α and initiates a proteolytic cascade. This cascade sequentially activates caspase-8, caspase-3, and ultimately cleaves GSDME. The resulting pyroptosis not only promotes the long-term survival of T cells but also facilitates the release of the alarmin interleukin-1α (IL-1α), which plays an essential role in mediating antifungal immune responses [47] (see Fig. 1).

Granzyme-mediated pathways

Liu et al. demonstrated that chimeric antigen receptor T (CAR-T) cells induce pyroptosis in target cells by rapidly activating intracellular caspase-3 through the release of granzyme B, thereby triggering a caspase-3/GSDME-dependent cell death pathway [48]. In addition, granzyme B can directly cleave GSDME to initiate pyroptosis, which potentiates a systemic antitumor immune response and inhibits tumor metastasis [49]. Other studies indicate that granzyme A (GzmA) cleaves GSDMB, resulting in pyroptosis in target cells. This mechanism enables natural killer (NK) cells and cytotoxic T lymphocytes to eliminate GSDMB-expressing tumor cells, suggesting a potential therapeutic approach leveraging immune-mediated pyroptosis [50]. It is noteworthy that different splice isoforms of GSDMB exhibit distinct functions; only the N-terminal fragments of specific isoforms are capable of inducing pyroptosis [51]. Tumors may evade such immune attack by expressing non-cytotoxic GSDMB variants. Therefore, modulating alternative splicing to promote the expression of cytotoxic GSDMB isoforms represents a promising strategy for enhancing antitumor immunity.

Pyroptosis and sepsis

Sepsis is characterized by a highly complex pathogenesis centered on dysregulated inflammatory and immune responses. During the early hyperinflammatory phase, invading pathogens or severe tissue injury trigger a potent systemic inflammatory reaction. As the condition progresses, depletion of inflammatory mediators, lysosomal enzymes, and immune cells may lead to a state of immunosuppression, increasing susceptibility to secondary infections and multiple organ dysfunction [52, 53]. Maintaining a balanced immune response is therefore essential both for effective pathogen clearance and for minimizing collateral tissue damage; conversely, either excessive inflammation or profound immunosuppression can exacerbate organ failure and secondary infections [54].

Pyroptosis, a highly inflammatory type of programmed cell death, contributes significantly to both early and late stages of sepsis through canonical (caspase-1-mediated) and non-canonical (caspase-4/5/11-mediated) pathways. In the early phase, canonical pyroptosis facilitates rapid host defense by releasing IL-1β and IL-18. Unlike apoptosis, this process does not require DNA fragmentation and proceeds through an autocatalytic caspase-1 cascade that is considerably faster than the caspase-3/7-driven apoptotic pathway. This results in swift cell lysis, providing an efficient first-line mechanism against infection [55]. Meanwhile, non-canonical pyroptosis enhances cytokine release and reduces the available cytosolic space for pathogens, thereby restricting their replication and promoting clearance—strengthening innate immunity and antimicrobial defense [56–58].

However, when pyroptosis becomes dysregulated, it can lead to an excessive release of pro-inflammatory mediators, amplifying both local and systemic inflammation and exacerbating tissue injury. Moreover, the hyperinflammatory state in sepsis is frequently accompanied by concurrent immunosuppression, further aggravating the clinical outcome and contributing to lethal septic shock and multi-organ failure [59, 60]. As a result, uncontrolled pyroptosis is considered a major factor underlying the high mortality associated with sepsis [61, 62]. In summary, while pyroptosis plays a protective role in early sepsis by promoting immune activation and pathogen clearance, its dysregulation in later stages exacerbates inflammatory damage and wors prognosis [63, 64]. Thus, pyroptosis represents a ‘double-edged sword’ in the pathogenesis of sepsis. Thus, pyroptosis represents a ‘double-edged sword’ in the pathogenesis of sepsis. Given its central role in driving immune dysregulation and organ injury, targeting pyroptosis has emerged as a promising therapeutic strategy. Mesenchymal stem cells, through their paracrine actions, particularly via extracellular vesicles, have shown the ability to suppress excessive pyroptosis by interfering with inflammasome assembly and cytokine maturation [65].

Pyroptosis and septic lung injury

Alveolar macrophages (AMs) are essential for maintaining pulmonary homeostasis and defending against respiratory pathogens [66]. Pyroptosis of AMs is closely associated with septic lung injury [67, 68]. During sepsis, while pyroptosis contributes to pathogen clearance, excessive release of inflammatory mediators and dysregulated immune responses resulting from AM pyroptosis can exacerbate acute lung injury (ALI) [69]. Studies indicate that Gram-negative bacteria induce NLRC4- and caspase-11-mediated pyroptosis in AMs via flagellin or LPS, whereas coronaviruses and fungi primarily activate NLRP3- and AIM2-dependent pathways [67, 68, 70]. Extracellular histones further promote AM pyroptosis through NLRP3 inflammasome activation, worsening pulmonary inflammation in acute respiratory distress syndrome (ARDS) [71]. Clinical and experimental evidence strongly supports the involvement of AM pyroptosis in ALI. Li et al. showed that conditional knockout of NLRP3 in AMs reduced GSDMD expression and alleviated ALI in septic mice [72]. Similarly, Wu et al. used the caspase-1 inhibitor AC-YVAD-CMK to suppress AM pyroptosis and mitigate septic lung injury [73]. In an LPS-induced macrophage injury model, Ying et al. found that miR-495 overexpression inhibited NLRP3 activation and downregulated GSDMD, thereby attenuating lung damage [74]. Furthermore, Pu et al. reported that deletion of autophagy-related protein 7 (ATG7) enhanced NLRC4 inflammasome activation and pyroptosis, impairing host defense against Pseudomonas aeruginosa and worsening lung injury [75].

Recent studies have increasingly highlighted the role of neutrophil pyroptosis in sepsis-associated lung injury. During sepsis, pyroptotic neutrophils release large quantities of cytokines and tissue-damaging factors, exacerbating acute lung injury (ALI) [76]. Furthermore, pyroptosis upregulates the expression of multiple chemokines, promoting further neutrophil recruitment into the lungs and amplifying local inflammatory responses, which intensifies tissue damage [77]. Neutrophil extracellular traps (NETs)—web-like structures composed of DNA, histones, and antimicrobial proteins released by activated neutrophils—function to capture and eliminate extracellular pathogens [78]. However, NETs can also induce pyroptosis in pulmonary epithelial and endothelial cells, contributing to the progression of ALI [79]. Li et al. reported that NETs may trigger pyroptosis in alveolar macrophages via AIM2 inflammasome activation and caspase-1 signaling [80], establishing a vicious cycle between NET formation and macrophage pyroptosis that aggravates lung tissue injury. Notably, unlike endothelial cells and macrophages, neutrophil pyroptosis does not primarily depend on the canonical caspase-1 pathway. Chen et al. observed that although neutrophils from Salmonella-infected mice showed significant NLRC4 inflammasome and caspase-1 activation, pyroptosis did not occur [81]. Kovacs et al. further demonstrated that while both caspase-1 and caspase-11 cleave GSDMD in neutrophils, only caspase-11 activation is sufficient to induce pyroptosis, highlighting the central role of caspase-11 in this process [82].

The alveolar–capillary barrier, composed of monolayers of pulmonary endothelial and epithelial cells, is frequently compromised during sepsis. PAMPs and DAMPs can induce pyroptosis in these structural cells, representing a key mechanism in barrier disruption [83]. Yang et al. showed that LPS triggers pyroptosis in pulmonary vascular endothelial cells through TLR4-mediated NLRP3 inflammasome activation and caspase-1 cleavage [84]. Lai et al. found that group 2 innate lymphoid cells (ILC2s) protect against endothelial pyroptosis by secreting IL-9, which inhibits caspase-1 activation in septic mice [85]. Beyond the canonical pathway, caspase-11 has emerged as an important mediator of endothelial pyroptosis. Burkholderia infection can induce caspase-11-dependent non-canonical pyroptosis in pulmonary epithelial cells, and genetic deletion of caspase-11 attenuates sepsis-induced lung injury [86]. Cheng et al. reported that LPS promotes endothelial pyroptosis via the TLR4/caspase-11 axis: activated caspase-11 cleaves GSDMD, generating GSDMD-N fragments that bind mitochondrial membranes, induce mitochondrial damage, and form oligomeric pores on the cell surface [87]. These pores facilitate mitochondrial DNA release, inhibit endothelial proliferation, and impair vascular repair [88]. Targeting key molecules in pyroptotic pathways within epithelial and endothelial cells may thus represent a promising therapeutic strategy for septic lung injury (see Fig. 2).

Fig. 2.

Pyroptosis in sepsis-associated organ injury

Emerging evidence also indicates that granzyme A, released by cytotoxic lymphocytes including cytotoxic T cells (CTLs) and natural killer (NK) cells, can enter target cells via perforin and induce pyroptosis by cleaving GSDMB at residues Lys229 and Lys244 [50]. However, the involvement of lymphocyte-mediated pyroptosis in ALI remains poorly understood and merits further investigation.

Pyroptosis and septic-induced cardiomyopathy

During sepsis, activation of the NLRP3 inflammasome has been detected in cardiomyocytes, cardiac fibroblasts and cardiac macrophages, and accumulating evidence indicates that pyroptosis actively participates in the initiation and progression of septic-induced myocardial injury (SIMI) [89, 90].

Busch et al. reported that in murine sepsis models NLRP3 inflammasome activation promotes IL-1β maturation, resulting in cardiomyocyte atrophy and impaired systolic–diastolic performance. Genetic deletion of NLRP3 [91, 92] or pharmacological suppression of the inflammasome with heat-shock protein 70 (HSP70), sodium tanshinone IIA sulfonate or emodin [93–95] significantly improves both survival and cardiac function. Subsequent studies confirmed that sepsis markedly up-regulates the expression of NLRP3-inflammasome components (NLRP3, ASC, caspase-1) and their downstream effectors (GSDMD, IL-1β and IL-18) in cardiomyocytes, concomitant with elevated levels of the myocardial injury markers troponin I, CK-MB and LDH and with depressed cardiac performance [96, 97]. In addition, activation of the caspase-11/GSDMD axis has been documented in septic myocardium, indicating that the non-canonical pyroptotic pathway also contributes to the pathogenesis of SIMI [98].

Cardiac fibroblasts not only provide structural support to cardiomyocytes but also communicate with them through paracrine signalling and microtubule-mediated cross-talk; pyroptosis of these fibroblasts is therefore another mechanism underlying sepsis-associated cardiac depression [99]. Zhang et al. observed that NLRP3, caspase-1 and IL-1β expression is markedly increased in cardiac fibroblasts of septic mice, and that NLRP3 knockdown ameliorates cardiac dysfunction [100]. Other work demonstrated that hepatocyte-derived angiotensinogen (AGT) can be internalized by cardiac fibroblasts via low-density lipoprotein receptor-related protein 1 (LRP1). Inside the fibroblast, AGT activates the NLRP3 inflammasome through an angiotensin II-independent pathway, enhancing IL-1β release. The subsequent paracrine signaling suppresses the expression of sarcoplasmic/endoplasmic reticulum Ca2+-ATPase (SERCA2a) in adjacent cardiomyocytes, thereby contributing to impaired contractile function [101]. Compounds such as intermedin1-53 and the carbon-monoxide-releasing molecule-3 (CORM-3) suppress the NLRP3/caspase-1/IL-1β pyroptotic cascade in cardiac fibroblasts and consequently attenuate septic myocardial depression, further corroborating the pathogenic role of fibroblast pyroptosis [102–104].

Pyroptosis of cardiac macrophages also exacerbates septic cardiac dysfunction [105]. Wang et al. reported that monocyte-derived exosomes can deliver the TXNIP–NLRP3 complex to cardiac macrophages, thereby promoting the maturation of IL-1β and IL-18 and amplifying cardiovascular inflammation [106]. The small-molecule PSSM1443 disrupts TXNIP–NLRP3 interaction, lowering caspase-1, IL-1β and IL-18 levels in septic mice. Zhao et al. found that IL-30 derived from cardiac macrophages is up-regulated in septic hearts; IL-30 deficiency skews monocyte-derived macrophages toward a Ly6C-high phenotype and increases pyroptosis, aggravating cardiac dysfunction. Conversely, NLRP3 inflammasome inhibition with MCC950 dampens inflammation and improves cardiac performance in sepsis [107] (see Fig. 2).

Pyroptosis and septic liver injury

The liver is one of the organs most frequently affected during sepsis, and septic liver injury constitutes an independent risk factor influencing patient prognosis [108, 109]. In murine models of sepsis-associated liver injury, pyroptosis-related proteins—including NLRP3 and caspase-1—are markedly elevated, and the frequency of pyroptotic events correlates positively with biochemical indices of hepatic damage [110]. Pharmacologic activation of peroxisome proliferator-activated receptor-γ (PPAR-γ) blunts the TXNIP/NLRP3 axis, thereby attenuating hepatocyte pyroptosis and mitigating sepsis-induced liver injury [111].

Kupffer cells (KCs), the resident macrophages of the liver, are pivotal in controlling bacterial infection and maintaining systemic immune homeostasis [112]. During sepsis, LPS can trigger canonical NLRP3–caspase-1-mediated pyroptosis in KCs, while KCs themselves release cathepsin B to activate caspase-11-dependent non-canonical pyroptosis, collectively exacerbating septic liver injury [113, 114]. Following infection with septic pathogens, the NLRP3 inflammasome promotes activation and collagen deposition in hepatic stellate cells (HSCs) and subsequently induces HSC pyroptosis, amplifying hepatic inflammation and increasing the risk of chronic fibrosis and even cirrhosis [115–117] (see Fig. 2).

Collectively, these findings indicate that the mechanisms by which pyroptosis drives septic liver injury are intricate and warrant further detailed investigation.

Pyroptosis and sepsis-induced acute kidney injury

In sepsis-associated acute kidney injury (AKI), renal inflammatory responses are markedly amplified, accompanied by a significant up-regulation of NLRP3 within kidney tissue [118]. NLRP3-inflammasome–driven pyroptosis is now recognized as a pivotal pathogenic mechanism underlying this renal damage. Pyroptosis in renal cells is orchestrated through multiple signaling cascades—including the caspase-11/GSDMD, miR-18a-3p/GSDMD, and RIPK3/GSDMD pathways—culminating in the massive release of pro-inflammatory cytokines such as IL-1β and IL-18. In addition, HMGB1 liberated during pyroptosis further activates and chemoattracts leukocytes, perpetuating an inflammatory loop that aggravates renal tissue injury [119–123].

Experimental studies have shown that pharmacological or genetic targeting of inflammasome-related pathways can attenuate pyroptosis and ameliorate sepsis-AKI. Dexmedetomidine, for instance, suppresses the TLR4/NOX4/NF-κB axis, while the caspase-1–specific inhibitor AC-YVAD-CMK effectively blocks canonical inflammasome activation [124, 125]. Likewise, inhibiting the non-canonical inflammasome confers renoprotection: genetic knockdown of IRF2 or the use of caspy2—a zebrafish caspase-4/5/11 ortholog—significantly attenuates pyroptosis and mitigates LPS-induced acute kidney injury in animal models [126–128] (see Fig. 2).

Pyroptosis and septic encephalopathy

The pathogenesis of septic encephalopathy is complex and remains incompletely understood [129]. Recent studies have identified NLRP3 inflammasome-mediated pyroptosis as a critical driver of this disorder. During sepsis, activation of the NLRP3 inflammasome in microglia, astrocytes, and neurons triggers canonical caspase-1–dependent pyroptosis, releasing IL-1β, IL-18, and other pro-inflammatory cytokines that precipitate the neuroinflammatory lesions characteristic of septic-associated encephalopathy (SAE) [130–132]. In addition, NLRP3 inflammasome signaling amplifies systemic inflammation, disrupts endothelial tight junctions, and compromises the viability of cerebral microvascular endothelial cells, collectively impairing blood–brain barrier integrity [133–135]. Matthias and colleagues first demonstrated that Streptococcus pneumoniae infection induces up-regulation of both caspase-1 and caspase-11 in the murine hippocampus, highlighting the relevance of pyroptosis in SAE [136]. Sun et al. reported that the purinergic ligand-gated ion channel P2 × 7 receptor confers neuroprotection by engaging the extracellular signal-regulated kinase (ERK) pathway to modulate NLRP3/caspase-1-mediated pyroptosis [137].

Emerging evidence indicates that several therapeutic interventions—including hydrogen gas, recombinant club cell secretory protein-16 (rCC16), the transcriptional regulator Maf1, the bromodomain protein 4 (BRD4) inhibitor JQ1, dexipramexole (DPX), and dexmedetomidine (DEX)—can attenuate NLRP3 inflammasome activity by distinct mechanisms, thereby reducing IL-1β and IL-18 secretion, alleviating cerebral inflammation, and improving the pathological manifestations of septic encephalopathy [134, 138–143] (see Fig. 2).

Pyroptosis and intestinal dysfunction

The gut is one of the organs most vulnerable to sepsis and other critical illnesses, and it often serves as the trigger for multi-organ failure. Pyroptosis contributes to the initiation and progression of sepsis-related intestinal dysfunction [144, 145]. Studies have shown that the BRD4 inhibitor JQ1 attenuates sepsis-induced colonic inflammatory injury by modulating NLRP3-mediated pyroptosis via the NF-κB pathway [146]. In parallel, antagonism of the P2 × 7 receptor suppresses LPS-induced, caspase-11/P2 × 7R-dependent pyroptosis of intestinal epithelial cells, thereby reducing gut inflammation [147] (see Fig. 2).

Pyroptosis and sepsis-associated coagulopathy

Sepsis-induced coagulopathy (SIC) is a disorder characterized by endothelial injury and deranged coagulation in sepsis and is a major cause of death [148–150]. Recent findings indicate that platelet-endothelial cell adhesion molecule-1 (PECAM-1) can accelerate the onset of sepsis-associated disseminated intravascular coagulation (DIC) by inhibiting macrophage pyroptosis [151]. Berberine derivatives prevent and treat sepsis-related coagulopathy by blocking the non-canonical pyroptotic pathway via suppression of macrophage scavenger receptor-1 (MSR1) [152]. Caspase-11-mediated non-canonical pyroptosis amplifies sepsis-associated coagulation cascades through GSDMD pore formation and increased tissue factor activity [153]. Yuan et al. demonstrated that inhibiting AP2-associated protein kinase 1 (AAK1)-mediated LPS internalization effectively prevents caspase-11 activation and mitigates the pathological course of DIC [152].

Mesenchymal stem/stromal cells (mscs): physiological characteristics and therapeutic potential

MSCs are multipotent stem cells of mesodermal origin. First isolated from bone marrow in 1968 by Friedenstein et al., they were described as adherent, fibroblast-like, non-hematopoietic precursor cells and are therefore often referred to as bone-marrow-derived mesenchymal stromal cells (BMSCs) [154, 155]. Subsequent studies revealed that MSCs also reside in adipose tissue, muscle, tendon, umbilical cord, placenta, spleen, peripheral blood, and dental pulp [156–158]. Vascular pericytes have been identified as another source; these mural cells exhibit chemotactic activity during inflammation and injury and display an MSC-like phenotype [159]. In 2006, the International Society for Cellular Therapy proposed three minimal criteria for defining MSCs: (1) adherence to plastic under standard culture conditions; (2) expression of CD105, CD90, and CD73, and absence of HLA-DR, CD34, CD45, CD19, and CD11b; (3) trilineage differentiation into osteoblasts, chondrocytes, and adipocytes under appropriate in-vitro conditions [160]. Owing to their broad availability, ease of isolation, expansion, and culture, MSCs have become a feasible therapeutic option for a variety of clinical disorders [161, 162].

Self-renewal and multilineage differentiation

Often termed “seed cells”, MSCs can, under optimal conditions, differentiate into a spectrum of specified cell types—including epithelial, muscular, neuronal, and connective-tissue cells such as bone, cartilage, and fat. They are capable of lifelong self-renewal, continuously providing fresh cells to replace aged or dead cells. This is attributed to their unique mode of renewal, “ndogenous asymmetric division”, which preserves the stem-cell pool and ensures its stability and flexibility [163, 164] (see Fig. 3).

Fig. 3.

Physiological properties of mesenchymal stem cells

Low immunogenicity

Immunogenicity denotes the capacity of an antigen to elicit an immune response by activating, proliferating, and differentiating specific immune cells to produce antibodies and sensitized lymphocytes. MSCs lack major histocompatibility complex class II molecules and co-stimulatory factors, endowing them with inherent low immunogenicity. Consequently, allogeneic MSCs rarely provoke host immune rejection [165, 166] (see Fig. 3).

Anti-inflammatory and immunomodulatory properties

MSCs play a crucial role in modulating immune responses. Upon tissue damage or inflammation, they home to the injured site, interact with immune cells, secrete bioactive factors, eliminate pathogens, suppress inflammatory cascades, and reduce oxidative stress and apoptosis [167–170]. Additional studies have shown that MSCs can systemically attenuate multi-organ injury and prolong survival in animal models by inhibiting reactive oxygen species (ROS) production, promoting macrophage M2 polarization, and down-regulating NLRP3 signaling [171, 172] (see Fig. 3).

Anti-fibrotic properties

MSCs exert potent anti-fibrotic effects by (i) modulating immune responses and oxidative stress, (ii) remodeling the extracellular matrix (ECM), (iii) reducing myofibroblast activation, and (iv) inhibiting transforming growth factor-β (TGF-β)-driven epithelial-to-mesenchymal transition (EMT), thereby preventing the differentiation of multiple cell types into ECM-secreting myofibroblasts [173] (see Fig. 3).

Pro-angiogenic capacity

Under the influence of VEGF, IGF-1, FGF and S1P, MSCs can differentiate into vascular endothelial cells [174, 175]. They also enhance neovascularization via paracrine activation of TLR receptors, which in turn up-regulates VEGF and other angiogenic cytokines [176, 177] (see Fig. 3).

Tissue repair

MSCs promote tissue repair through direct differentiation or cell fusion, paracrine release of cytokines, exosomes or micro-vesicles, and intercellular transfer of exosomes, cytokines or organelles via tunneling nanotubes (TNTs) to injured cells [178] (see Fig. 3).

Anti-microbial activity

Following inflammatory stimulation, MSCs recruit monocytes, macrophages and neutrophils to infection sites and enhance their phagocytic capacity, facilitating pathogen clearance [179]. MSCs also secrete antimicrobial peptides (AMPs) such as cathelicidin, β-defensins and lipocalin-2, which directly kill bacteria, fungi and viruses [180]. These AMPs disrupt microbial membranes and prevent immune evasion, thereby assisting immune cells in eliminating invading pathogens [181]. Through these mechanisms, MSCs mitigate infection-related tissue damage and reduce microbial invasion of healthy tissues (see Fig. 3).

Regulation of programmed cell death (PCD)

PCD—genetically regulated, active and orderly cellular self-elimination—includes pyroptosis, apoptosis, necroptosis, autophagy and ferroptosis [182, 183]. MSCs modulate PCD by secreting extracellular vesicles and soluble factors (e.g., growth factors) that reshape the injured microenvironment via immunomodulation and pro-angiogenic actions [184]. Additionally, MSCs form gap-junction proteins (Cx) and TNTs with injured cells, enabling selective transport of ions, proteins, peptides and genetic material to supply energy and metabolic support, thereby fine-tuning PCD pathways [185, 186].

Notably, the immunomodulatory properties of MSCs establish their role as potent regulators of pyroptosis in the context of sepsis. Beyond simple cell replacement, They can suppress the priming step of NLRP3 inflammasome activation by secreting anti-inflammatory factors (e.g., PGE2, TSG-6) that inhibit the NF-κB pathway, thereby reducing the expression of NLRP3 and pro-IL-1β. MSCs mediate therapeutic effects through the secretion of bioactive factors that directly inhibit inflammasome priming and activation, suppress caspase-1/4/11 cleavage, prevent GSDMD-mediated pore formation, and attenuate the release of pyroptosis-associated proinflammatory cytokines, including interleukin-1β (IL-1β) and interleukin-18 (IL-18), across multiple organ systems [187, 188].

Mesenchymal stem cell-derived extracellular vesicles (MSC-evs) in the regulation of pyroptosis during sepsis

During the past decade, MSCs have shown remarkable promise in sepsis research; however, divergent mechanistic interpretations across studies reflect the multi-target and multi-pathway nature of MSC action. Early dogma assumed that the therapeutic value of MSCs stemmed primarily from their capacity to home to injured tissues, differentiate, and replace damaged cells. Yet subsequent work has demonstrated that >99% of intravenously infused MSCs are trapped in filter organs such as liver, spleen and lung, with < 1% reaching the site of injury [189, 190]. Moreover, MSCs that do engraft typically vanish within a few days, resulting in minimal long-term chimerism. Notably, functional improvements are observed within 24 to 48 h, well before differentiation could reasonably occur, and lineage-tracing studies demonstrate that the newly formed reparative cells are predominantly derived from the host rather than from the transplanted MSCs [191]. Collectively, these findings argue that direct differentiation is neither the sole nor the dominant mechanism of MSC-mediated repair.

Against this backdrop, investigators observed that MSC-conditioned medium could recapitulate the salutary effects of intact cells, implicating soluble secreted factors as key executors of tissue repair and underscoring the primacy of endocrine/paracrine signaling in MSC biology [192]. In 2010, Lai et al. isolated exosomes from MSCs and demonstrated their reparative capacity, formally establishing EVs as essential conveyors of MSC function [193]. Subsequent studies have converged on the conclusion that the therapeutic effects of MSCs are mediated primarily, if not entirely, by extracellular vesicles, including exosomes and apoptotic bodies, which deliver bioactive cargo through paracrine signaling. MSC-derived extracellular vesicles are now recognized as the main effector component of MSC therapy [194–196].

MSC-EVs are membrane-bound vesicles 30–150 nm in diameter. Biogenesis proceeds sequentially through early endosomes (EE), intraluminal vesicles (ILV), late endosomes (LE), and multivesicular bodies (MVB), culminating in fusion with the plasma membrane and release into the extracellular milieu [197]. These vesicles can act locally within the microenvironment or circulate systemically to reach distant organs. Compared with whole MSCs, which are susceptible to pulmonary entrapment and potential embolism, extracellular vesicles, due to their smaller size, can traverse the pulmonary capillary bed more efficiently, cross the blood-brain barrier, reduce material loss, and lower procedural risks [198]. MSC-derived extracellular vesicles retain the diverse molecular repertoire of their parent cells, including proteins, RNAs, DNAs, and lipids, thereby recapitulating many of the functions attributed to MSCs. Target cells internalize these vesicles through ligand-receptor interactions, phagocytosis, or direct membrane fusion, resulting in functional modulation and the propagation of downstream signaling [199]. Even after MSCs are cleared from the body, their released EVs can persist and exert biological effects for extended periods [200–202]. Furthermore, whereas MSC viability declines during transport and storage, MSC-EVs remain stable at −80 °C in PBS-HAT buffer for 7–251 days [203]. Producing therapeutic doses of MSCs requires billions of cells and about 10 weeks of culture, entailing substantial time and cost; in contrast, 3-D culture systems and other scalable bioprocesses allow more efficient EV isolation [204, 205]. Because MSC-EVs are non-replicating, they circumvent stem-cell-associated risks while functioning as natural nanocarriers for targeted delivery and precise modulation of effector molecules [206, 207]. These unique biological and logistical advantages provide a robust foundation for their application in sepsis-induced multi-organ injury, and encouraging pre-clinical data already demonstrate their efficacy in modulating pyroptotic pathways across diverse organ systems.​.

MSC-evs in the regulation of sepsis-associated organ injury

In sepsis-related acute lung injury MSCs blunt the TLR4/NF-κB/MAPK and TGF-β/Smad cascades to restrain pyroptosis and apoptosis in lung tissue, attenuate neutrophil infiltration, T-cell proliferation and M1 macrophage polarization, and mitigate endothelial damage [208, 209]. MSC-derived exosomes deliver miR-7704 to silence MyD88/STAT1 signaling in pulmonary macrophages, driving M2 polarization, restoring lung function and improving survival [210]. Furthermore, miR-125b-5p–loaded MSC-Exos ameliorate ALI by targeting the pyroptotic machinery in alveolar macrophages, where they inhibit caspase-1–mediated GSDMD processing to prevent pyroptotic cell death and concomitantly reduce pro-inflammatory cytokine release, thereby attenuating pulmonary inflammation and improving disease outcomes [211, 212]. And BMSC-derived exosomes enriched in miR-384-5p down-regulate Beclin-1, rescue autophagy impairment in alveolar macrophages, decrease macrophage death and vascular permeability, and elevate survival in septic ALI rats [213].

In sepsis-induced cardiac injury MSC-EVs armed with miR-34a-5p and miR-223-3p modulate the HMGB1/AMPK and FOXO3/NLRP3 axes respectively to inhibit cardiomyocyte pyroptosis and improve myocardial dysfunction [214, 215]: miR-34a-5p blocks HMGB1 translocation and dampens HMGB1-driven inflammation and pyroptosis, miR-223-3p silences FOXO3 to prevent NLRP3 inflammasome assembly and IL-1β/IL-18 release, and miR-146a-5p carried by MSC-EVs represses MYBL1, curbing cardiomyocyte inflammation and apoptosis while promoting survival [216].

In sepsis-associated hepatic injury MSCs suppress NLRP3 and caspase-1 expression, preventing GSDMD-mediated hepatocyte pyroptosis, limiting IL-1β/IL-18 secretion and attenuating LPS-induced liver damage [217]. As evidenced by existing studies, MSC-EVs inhibit hypoxia-inducible factor 1α (HIF-1α) to promote the polarization of hepatic macrophages towards the M2 phenotype, thereby contributing to inflammation resolution. Alternatively, MSC-EVs loaded with miR-26a-5p inhibit oxidative stress, a key activator of the NLRP3 inflammasome, thus interrupting a critical upstream signal for pyroptosis initiation [218, 219].

In sepsis-related neurological injury, MSC-EVs loaded with distinct RNA cargo exert multifaceted neuroprotective effects through converging anti-pyroptotic mechanisms: miR-140-3p-enriched MSC-EVs directly target and suppress HMGB1 expression, thereby inhibiting the NF-κB/NLRP3 inflammasome signaling axis and attenuating caspase-1-mediated GSDMD cleavage in activated microglia, which alleviates neuroinflammation and improves cognitive dysfunction in sepsis-associated encephalopathy; miR-146a-5p-loaded MSC-EVs act on the TRAF6 axis to disrupt TLR4/MyD88/NF-κB signaling, concurrently suppressing proinflammatory cytokine release and directly inhibiting caspase-1-mediated GSDMD cleavage, thereby attenuating both inflammation and microglial pyroptosis; while lncRNA RMRP–containing MSC-EVs modulate the EIF4A3/SIRT1 pathway, enhancing SIRT1-mediated deacetylation of NLRP3 and promoting antioxidant defenses, thereby quenching ROS production and further restraining the pyroptotic cascade [220–222].

In sepsis-induced acute kidney injury, MSC-EVs loaded with distinct RNA cargo exert multifaceted renoprotective effects through converging anti-apoptotic and anti-pyroptotic mechanisms: miR-223-3p-enriched MSC-EVs directly silence HDAC2 expression, thereby derepressing SNRK transcription to inhibit the NLRP3 inflammasome signaling axis, which attenuates caspase-1-mediated GSDMD cleavage and IL-1β/IL-18 release in tubular epithelial cells; concurrently, these EVs activate SIRT1/Parkin-mediated mitophagy to eliminate damaged mitochondria, reduce ROS production, and prevent mitochondrial dysfunction–driven apoptosis, collectively mitigating sepsis-AKI [223, 224]. miR-125b-5p–rich MSC-EVs specifically target and inhibit p53 signaling, rescuing tubular epithelial cells from G2/M cell cycle arrest, suppressing Bax-mediated apoptosis while upregulating Bcl-2, and promoting renal tubular epithelial proliferation and repair [225].

In sepsis-associated intestinal injury, BMSC-derived exosomes containing miR-539-5p directly target and silence NLRP3 expression, thereby inhibiting the NLRP3/caspase-1 signaling axis and preventing caspase-1-mediated GSDMD cleavage in intestinal epithelial cells; this dual suppression reduces GSDMD-mediated pore formation and curtails the release of pyroptotic cytokines IL-1β and IL-18, while also attenuating ROS production and protecting tight junction proteins, ultimately inhibiting pyroptosis, restoring intestinal barrier integrity, and alleviating inflammatory bowel disease progression [226].

These studies primarily represent work published within the past three to five years and collectively suggest that MSC-EVs exert robust anti-pyroptotic and organ-protective effects across multiple sepsis models. MSC-EVs suppress NLRP3 inflammasome activation and inhibit pyroptosis mediated by caspase-1 and GSDMD. They reduce pro-inflammatory cytokine release, restore epithelial and endothelial barrier function, and shift macrophage and microglial phenotypes toward inflammation-resolving states. These protective effects involve diverse RNA cargos and signaling pathways across pulmonary, cardiac, hepatic, neurological, renal, and intestinal systems. Mechanistically, these studies converge on a shared anti-pyroptotic axis that mitigates multi-organ injury. The emerging concept of “pyroptotic exosomes (pyroEVs)” further reframes pyroptosis as an immunomodulatory hub rather than mere cell death [227]. Building on this, prompting efforts to optimize MSC-EVs by pre-conditioning strategies such as LPS + ATP stimulation that substantially increase EV yield and functional potency, endowing MSC-EVs with superior anti-inflammatory and reparative capacities [228]. However, whether these strategies consistently translate into effective organ protection across diverse sepsis models remains uncertain, and their reproducibility has yet to be systematically evaluated.

Despite these encouraging findings, several critical uncertainties persist. A systematic review and meta-analysis of preclinical sepsis models demonstrate that, although MSC-EVs can reduce mortality, substantial heterogeneity in EV source, isolation methods, dosing, timing, and characterization limits reproducibility and standardization [229]. Moreover, more recent work using murine models reveals that small extracellular vesicles from MSCs accumulate rapidly in the liver and lungs after intravenous injection, raising concerns about biodistribution, off-target sequestration, and the influence of pharmacokinetics on therapeutic efficacy [230].

Clinical evidence remains extremely limited. A small RCT in COVID-19 induced ARDS combining MSCs and MSC-EVs reported reductions in inflammatory markers and acceptable safety, but the very small sample size, lack of sepsis-specific endpoints, and absence of long-term follow-up limit the generalizability of these findings to broader sepsis populations [231]. In addition, a detailed systems-biology review discussed the heterogeneity in miRNA cargo among MSC-EVs, the stochastic (low-copy) nature of miRNA loading, and the lack of consensus on potency biomarkers or optimal miRNA combinations for clinical translation [232].

Overall, while MSC-EVs represent a promising platform for modulating pyroptosis in sepsis, their reproducibility, pharmacokinetics, and clinical applicability remain uncertain. Future research should focus on rigorous mechanistic mapping, standardized EV production and characterization, optimization of pre-conditioning strategies, careful pharmacokinetic profiling, and well-powered clinical studies to comprehensively evaluate therapeutic potential and safety.

Conclusion

This review systematically delineates the central regulatory role of pyroptosis in the pathogenesis of sepsis. Via canonical (caspase-1-mediated), non-canonical (caspase-4/5/11-mediated), and alternative routes involving caspase-3/8 and granzymes, pyroptosis contributes to early pathogen clearance yet, when over-activated, ignites an inflammatory storm and immune imbalance that constitutes a pivotal node for multiple-organ dysfunction. This process underpins sepsis-related injury in the lung, heart, liver, kidney, nervous system, and intestine.

MSC-EVs have emerged as a novel therapeutic modality that modulates the pyroptotic signaling network and demonstrates remarkable efficacy in alleviating sepsis-associated organ damage. By loading specific miRNAs, MSC-EVs can target multiple pathways, suppress pyroptosis, and precisely regulate pro-inflammatory cytokine release, thereby mitigating organ injury. These attributes underscore the unique advantage of MSC-EVs as targeted therapeutic vectors.

Nevertheless, the clinical translation of MSC-EVs faces three major challenges: (1) the absence of standardized manufacturing protocols and quality-control systems; (2) suboptimal biocompatibility and tissue-penetration capacity of current delivery systems, limiting organ-specific enrichment; and (3) insufficient evidence-based support for combination regimens.

Future directions must address four critical imperatives. First, leveraging single-cell sequencing to delineate stage-specific pyroptotic signatures in sepsis will identify precise intervention targets. Second, developing biomimetic carrier systems is essential for efficient, organ-directed delivery. Third, validating combination strategies based on a “pyroptosis-modulation → immune-reprogramming → organ-protection” triad is required. Fourth, resolution of key translational challenges is paramount: (i) systematic comparison of bone marrow-, adipose-, and umbilical cord-derived MSCs to establish the optimal cell source based on immunomodulatory potency, scalability, and safety profiles; (ii) direct evaluation of various delivery routes, including intravenous, intratracheal, and organ-targeted administration, to optimize the therapeutic index while minimizing off-target effects; (iii) delineation of precise therapeutic windows across the sepsis continuum to enhance clinical efficacy; and (iv) comprehensive assessment of safety parameters encompassing immunogenicity, pro-coagulant activity, and long-term biodistribution. By systematically addressing these technical and translational imperatives, MSC-EV-based therapy may establish itself as a clinically viable strategy to attenuate sepsis-associated mortality.

Abbreviations

AAK1

AP2-associated protein kinase1

AGT

Angiotensinogen

AIM2

Absent in melanoma 2

AKI

Acute kidney injury

ALI

Acute lung injury

AMPs

Antimicrobial peptides

AMs

Alveolar macrophages

ARDS

Acute respiratory distress syndrome

ATG7

Autophagy-related protein 7

BMSCs

Bone-marrow-derived mesenchymal stromal cells

BRD4

Bromodomain protein 4

CARD

Caspase activation and recruitment domain

CAR-T

Chimeric antigen receptor T

CORM-3

Carbon-monoxide-releasing molecule-3

CTLs

Cytotoxic T lymphocytes

Cx

Gap-junction proteins

DAMPs

Damage-associated molecular patterns

DEX

Dexmedetomidine

DIC

Disseminated intravascular coagulation

DPX

Dexipramexole

ECM

Extracellular matrix

EE

Early endosomes

EMT

Epithelial-to-mesenchymal transition

ERK

Extracellular signal-regulated kinase

GSDMD

Gasdermin-D

GSDMD-NT

N-terminal fragment

GSDME

Gasdermin E

GzmA

Granzyme A

HIF-1α

Hypoxia-inducible factor 1α

hGBP1

Human guanylate-binding protein 1

HMGB1

High mobility group protein B1

HSCs

Hepatic stellate cells

HSP70

Heat-shock protein 70

ICUs

Intensive care units

IFI16

Interferon-γ-inducible protein 16

ILC2

Group 2 innate lymphoid cells

ILV

Intraluminal vesicles

KCs

Kupffer cells

LE

Late endosomes

LPS

Lipopolysaccharide

LRP1

Lipoprotein receptor-related protein 1

LRRs

Leucine-rich repeat domains

MSC-EVs

Mesenchymal stem cell-derived extracellular vesicles

MSCs

Mesenchymal stem cells

MSR1

Macrophage scavenger receptor-1

MVB

Multivesicular bodies

NACHT

Central nucleotide-binding domains

NETs

Neutrophil extracellular traps

NINJ1

Nerve injury-induced protein 1

NKs

Natural killer cells

NLR

Nod-like receptor

OMVs

Outer-membrane vesicles

PAMPs

Pathogen-associated molecular patterns

PCD

Programmed cell death

PECAM-1

Platelet-endothelial cell adhesion molecule-1

PPAR-γ

Peroxisome proliferator-activated receptor-γ

PRRs

Pattern recognition receptors

RAGE

Receptor for advanced glycation end-products

rCC16

recombinant club cell secretory protein-16

ROS

Reactive oxygen species

SAE

Septic-associated encephalopathy

SERCA2a

Sarcoplasmic/endoplasmic reticulum Ca2+-ATPase

SIMI

Septic-induced myocardial injury

TCR

T-cell receptor

TGF-β

Transforming growth factor-β

Th17 cells

T-helper 17 cells

TLR4

Toll-like receptor 4

TNF-α

Tumor necrosis factor-α

TNTs

Tunneling nanotubes

Author contributions

All authors contributed to the conception and design of this review. Xinqi Xu: Literature search, Writing-original draft, Data curation, Writing–review & editing. Jiapan An: Writing–review & editing. Tingyu Yang: Writing–review & editing. Bin Li: Conceptualization, Writing-review & editing. Zhimin Dou: Funding acquisition, Project administration, Writing–review & editing. The manuscript was drafted by Xinqi Xu. All authors provided critical revisions and approved the final manuscript.

Funding

This work was supported by Gansu Provincial Health Industry Research Program (GSWSKY2023-24), Gansu Provincial Youth Science and Technology Foundation (20JR10RA710), the Foundation of the First Hospital of Lanzhou University (ldyyyn2019-13).

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.