2.1 Inflammation and immune

The improper presence of microorganisms (bacteria, etc.) and their products can trigger a host immune response that is critical for maintaining and restoring homeostasis but can lead to tissue damage if excessive.³¹ In practice, the greater the inflammatory response is, the stronger the cellular damage and thus the higher the risk of organ dysfunction.¹⁸ At its onset, sepsis manifests as an overwhelming release of inflammatory mediators (cytokine storms) in response to infection.^{32, 33} The immune response to infection can work with innate immunization defense, components of the innate immune system at the site of pathogen exposure that activate and recruit circulating immune cells.^{34, 16} Immune cells have pathogen-recognition receptors (PRRs) on their surface that are activated by binding to pathogen-associated molecular patterns (PAMPs) or damage-associated molecular patterns (DAMPs), such as bacteria, burn, trauma, and so on. Receptor binding triggers intracellular signaling cascades, which in turn leads to the production of associated inflammatory mediators, thus initiating an immunoinflammatory cascade.^{35, 36} The ensuing chain reaction will eventually lead to the activation of an even stronger “armed force,” which is the adaptive immune response.³⁷ This explosive activation and the resulting immune “cytokine storm” have been identified as the pathogenic pathway for sepsis.^{33, 38}

In the sepsis-induced inflammatory response, both external factors derived from pathogens and internal factors released by damaged cells (such as lipopolysaccharide [LPS] and high-mobility group box-1 [HMGB-1] protein) can recognize PRRs, including Toll-like receptors (TLRs), C-type lectin receptors (CLRs), retinoic acid inducible gene I (RIG-I)-like receptors, and NOD-like receptors (NLRs), leading to the activation of downstream pathways.³⁹ The activation of different receptors results in the translocation of transcription factors, such as nuclear factor-κB (NF-κB), into the nucleus and subsequent activation of target genes, which encode proinflammatory cytokines such as tumor necrosis factor (TNF), interleukin (IL)−6, IL-12, and interferons (IFNs) (Figure 1).⁴⁰ The secretion of proinflammatory factors progresses to a life-threatening cytokine storm. Moreover, soluble cytosolic NLRs also contribute to the immune imbalance associated with sepsis.^{41, 42} The activation of specific NLRs is regulated by adaptor receptor-interacting protein kinase 2 (RIP2) (also known as RICK), leading to NF-κB activation and adaptor protein 1 activation. However, certain other NLRs (e.g., NLRP and NLRC4) participate in the assembly of distinct inflammasome protein complexes.⁴³ The inflammasome cleaves procaspase-1 into active caspase-1, which then cleaves pro-IL-1β and pro-IL-18, resulting in the release of the highly inflammatory cytokines IL-1β and IL-18 and triggering a form of programmed cell death (known as pyroptosis) characterized by intense inflammation.⁴⁴ When these mechanisms are driven by sepsis, they can become catastrophic and lead to a life-threatening inflammatory cascade.^{45, 46} Dectin is one of the main CLRs that plays a significant role in various biological processes. This particular Dectin has the ability to induce the production of reactive oxygen species (ROS) and activate inflammatory responses through the action of SRC and SYK kinases.⁴⁷ Interestingly, in addition to the PPRs mentioned above, some double-stranded RNA receptors have been found, mainly including RIG-I, melanoma differentiation-associated gene 5, and laboratory of genetics and physiology 2, and are also associated with sepsis induced immune dysfunction.^48-50 It has been noted that PRR can be activated by both exogenous PAMPs and endogenous DAMPs.^{36, 51} In relation to the induction of sepsis from within the body, it has been found that hepatocytes release a significant amount of HMGB-1, which binds to bacterial endotoxin known as LPS (Figure 2).^{52, 53} This bacterial endotoxin is then transported to the cytoplasm through receptors called receptor for advanced glycation end products (RAGE), which are expressed in vascular endothelial cells (ECs) and macrophages.⁵⁴ As a result of this process, caspase-11-mediated cell death, or apoptosis, occurs, ultimately leading to sepsis, multiple organ failure, and death.^{55, 56}

2.2 Inflammation and coagulopathy

Disseminated intravascular coagulation (DIC) is a life-threatening syndrome that is commonly considered an organ dysfunction that is often present in sepsis, with severity positively correlated with mortality.^{57, 58} DIC overactivation is characterized by an intravenous coagulation cascade, exhaustion of anticoagulants, and inhibition of fibrinolysis.⁵⁹ The dysregulation of inflammation has been found to trigger the coagulation response in sepsis, while the activation of the coagulation reaction exacerbates the inflammatory response.²⁹ This highlights the crucial interaction between inflammation and coagulation, which is seen as the primary driver of sepsis pathogenesis.⁶⁰ Throughout the onset and progression of sepsis, the generation of cytokines not only facilitates the activation of coagulation factors and platelets but also initiates the mechanism of anticoagulation (such as the system of antithrombin, the system of activated protein C (APC), and the inhibitor of tissue factor [TF] pathway). This can result in the leakage of blood vessels, DIC, and extensive accumulation of fibrin in either the blood vessels or tissues (Figure 3).^{61, 62} In general, coagulation activation is controlled by three vital physiological anticoagulant pathway systems, including the TF pathway inhibitor system (TFPI), APC system, and antithrombotic system, which complement each other to modulate coagulation activation. In the early stages of sepsis, insufficient balance of TFPI in TF-dependent coagulation events supports impaired physiological functions of anticoagulants identified during sepsis.⁶³ In addition, hyperinflammation and innate immunity work together in coordination to amplify the coagulation cascade, in which bacterial-derived products and DAMPs (such as neutrophil extracellular traps [NETs], HMGB1 and cell-free DNA [cfDNAs]) contribute to the development of DIC through multiple pathways.^{25, 64} In the advanced phases of sepsis, immune cell apoptosis (such as dendritic cells [DCs] and lymphocytes) results in immune stagnation by diminishing the host's capacity to eliminate intruding microorganisms and enhancing susceptibility to subsequent infections.^{65, 66} DIC, a key factor in multiple organ failure during sepsis, manifests as the formation of numerous microvascular blood clots, subsequently depleting platelets and coagulation factors, ultimately resulting in impaired hemostasis.⁶⁷ In contrast, these thrombin and coagulation factors exhibit proinflammatory activity by cleaving protease-activated receptors (PARs) to form a vicious cycle in which increased inflammation and coagulation further enhance the mortality rate of sepsis.^{25, 57, 68}

2.3 Complement system

In the late 19th century, scientists made an interesting discovery—a plasma element known as the complement system. This component, which was found to be thermally unstable, served as a complement to antibodies in the process of eliminating microorganisms. This is why it was named “complement.”^{69, 70} However, it is important to recognize that the complement system can be both beneficial and detrimental. When dysregulated or excessively activated, it can turn against its host, causing harm to tissues, resulting in organ failure, and, in severe cases, leading to mortality.^{71, 72} The function of complement is beyond infectious defense, which promotes tissue regeneration and repair, facilitates organ development, and can be transformed and coordinated with plentiful processes, including hemostasis leading to inflammation that causes thrombosis.⁷¹ During the treatment phase of patients with sepsis, chemical reperfusion injuries (myocardial infarction, stroke, and transplant dysfunction) trigger (DAMPs) that trigger complement activation in chronic nervous system and rheumatic diseases, thereby contributing to inflammatory and immune disorders.^{73, 74} Dysregulation of the complement system marks a disruption in host recognition of defense signaling pathways associated with inflammation, coagulation, and bacterial cell lysis, and is a prelude to multiorgan dysfunction in sepsis.⁷⁵

The activation of complement cascades is a widely acknowledged phenomenon that occurs via three distinct pathways. These pathways include the classical pathway (triggered by the complex formed by antigens and antibodies), the alternative pathway (initiated by factors B and D), and the lectin pathway (detected via the recognition of mannose derived from pathogens).⁷⁶ After the complement system was activated, these three pathways gathered at the C3 level to produce cleavage products C3a and C5a, as well as the terminal membrane-attacking complex C5b, which forms pores in the cell membranes of bacteria and cells, eventually leading to lysis (Figure 3).⁷⁷ C3a, C5a, and C5b mediate antibacterial responses and proinflammatory effects by interacting with cell membrane receptors and triggering crosstalk among multiple signaling pathways. In sepsis, the innate immune functions of neutrophils are impaired due to excessive C5a generation, for example, chemotaxis, phagocytosis, and H₂O₂ production are blocked.^{78, 79} Thus, complement-mediated activation of neutrophils is not only responsible for diverse significant impacts in sepsis but also assumes a contradictory function in innate defense.

Furthermore, sepsis is characterized by an early excessive production of C5a, which leads to an unregulated inflammatory response, ultimately resulting in tissue damage and the failure of multiple organs.⁸⁰ Sustained complement activation can cause a massive release of C5a, which may impair the innate immune response of neutrophils, resulting in reduced inflammation and impaired ability to kill bacteria.^{81, 82} Moreover, elevated levels of C5a can impede the targeted migration of neutrophils. Notably, the secretion of TNF by neutrophils can inhibit the transcription of NF-κB, as induced by the C5a-mediated increase in IκBα levels. Furthermore, C5a disrupts the C3a–C3aR axis and promotes the premature expulsion of granulocytes and hematopoietic stem cells from the bone marrow, leading to a less targeted but more progressive inflammatory response.⁸³ Through C5a signaling, CXCR4 levels on granulocytes are concurrently reduced, while proteases are released, resulting in the degradation of matrix proteins and inhibition of the effect of stromal cell-derived factor 1, which causes phenotypic changes in neutrophils.⁸⁴ Moreover, activation of the complement system triggers the cross-activation of the NOD-like receptor protein 3 (NLRP3) inflammasome and prethrombotic pathways (Figure 2).⁸⁵ Pentraxins (such as CRP, SAP, PTX3) released in response can initiate the classical pathway by interacting with C1q.⁸⁶ Apart from pentraxins, the complement system can be directly activated by coagulation proteases through the extrinsic protease pathway (Figure 2).⁸⁷ The uncontrolled activation of the complement system during sepsis contributes to tissue damage and dysfunction in organs.

2.4 Mitochondrial dysfunction

Mitochondria play a vital role in energy generation, protein synthesis, regulation of cell growth and cycle, and catabolism.^{88, 89} Nevertheless, in cases of sepsis, the mitochondrial damage or malfunction can disrupt cellular metabolism,^{41, 90-92} impede energy production, and trigger oxidative stress.^{93, 94} Consequently, this may provoke apoptosis in both organ and immune cells, ultimately leading to immune disorders, multiple organ failure, and potentially fatal outcomes.^{95, 96} It has been proven that the involvement of specific NLRP3 inflammasome activators could expand mitochondrial instability, resulting in pyroptotic cell death and enhancing the formation of pores in the plasma membrane, which then release inflammatory factors through caspase-1 dependent mechanisms. However, excessive ROS due to electron transport chain damage, Ca²⁺ overload, or reduced endogenous antioxidants can also trigger patterns of cell death, such as apoptosis and autophagy.^97-99 In the pathogenesis of sepsis, mitochondrial function is disorganized, mainly manifested by a decrease in oxidative phosphorylation, an increase in ROS production, the effects of hormonal changes, downregulation of the genes encoding mitochondrial proteins and an increase in apoptosis, which eventually leads to altered mitochondrial biogenesis.^100-102 Damaged mitochondria release a cascade of DAMPs, further activating and enhancing the immune response.^{103, 104} The changes in mitochondrial function that occur early in sepsis are an adaptive mechanism by which mitochondria protect cells. For example, the production of harmful ROS is regulated by oxidative phosphorylation.^{105, 106} In addition, in the early stages of sepsis, mitochondrial autophagy and mitochondrial biogenesis are increased to limit mitochondrial dysfunction.^{107, 108} Therefore, the development of sepsis is accompanied by changes in different mitochondrial functions.

Furthermore, molecular events resulting from mitochondrial dysfunction in sepsis trigger the activation of caspase-1, which intensifies the response to inflammatory responses, such as perturbations of membrane permeability and damage to the mitochondrial network.^{109, 110} Although serum proinflammatory factors and antiapoptotic proteins have been shown to play important roles in the activation and inhibition of the NLRP3 inflammasome, the pathogenesis of sepsis-induced acute respiratory distress syndrome is primarily an elevated intracellular oxidative stress-mediated apoptotic event.^111-113 The alterations in the stage of organ function and how persistent organ dysfunction further affects mitochondrial performance remain puzzling, but the confusion is still the subject of translational medicine research because these periodic responses illustrate the potential of targeted therapy to alleviate the organ dysfunction induced by sepsis.

2.5 Pyroptosis

Pyroptosis is a lytic type of programmed necrosis that is linked to the secretion of proinflammatory cytokines. This process has pivotal functions in facilitating the defensive response of the innate immune system against microbial infections and invading pathogens.^114-117 These signature genes in pyroptosis are NLRP3, apoptotic speck-like protein containing caspase activation and recruitment domains (CARD) (ASC), cleaved Caspase-1, Gasdermin-D (GSDMD) p30, IL-1β, and IL-18, which play a prominent role in the response to infection, ultimately fueling inflammation (Figure 2).^{21, 109, 110, 118, 119} The expression levels of inflammatory genes show complex correlations to a certain extent, ultimately leading to the rampant proliferation of inflammatory gene profiles (such as NLRP3, NLRC4, NOD, IL-1β, and IL-18), while in sepsis studies, the same gene regulation pattern was observed with a higher amplitude of change, revealing its clinical correlation with the severity of sepsis.^{39, 120, 121} It has been established that NLRP3 inflammasome activation requires “prime” and “activation.”^{122, 123} The NF-κB signaling pathway is an essential priming event that ensures accurate response of the inflammasome to LPS stimulation and prevents inappropriate activation of NLRP3.^124-126 Recent research has suggested that the initiation of NLRP3 deubiquitination may occur through a nontranscriptional mechanism. Specifically, these studies have proposed that mitochondrial ROS are responsible for inducing priming by causing the deubiquitination of the NLRP3 inflammasome. This finding challenges the traditional belief that NLRP3 deubiquitination is primarily regulated by transcriptional processes. Instead, it suggests that a direct interaction between mitochondrial ROS and the NLRP3 inflammasome is involved in this important regulatory step. Further investigation is needed to fully understand the exact molecular mechanisms underlying this nontranscriptional initiation of NLRP3 deubiquitination.^126-128 In addition, LPS-induced cytidine/uridine monophosphate kinase 2 (CMPK2), which is rate limiting for mtDNA synthesis, reduced NLRP3 inflammasome assembly (Figure 2).^{129, 130} In the absence of NLRP3 activators, prime-signaling provided by NF-κB or CMPK2 is insufficient for activating inflammasomes, suggesting that NLRP3 inflammasome activators play a pivotal role in triggering inflammation.^131-133 NLRP3 inflammasome activators include several physically and chemically diverse extracellular stimuli, such as extracellular adenosine triphosphate (ATP) or uric acid crystals, exogenous particulates such as aluminum salts and titanium dioxide, pathogens and pore-forming toxins.^134-138 During sepsis, ATP activates the P2 × 7 receptor and NLRP3 signal transduction in an autocrine manner after activation of the inflammasome, leading to changes in metabolic status and convergence of inflammasome signals and is associated with increased lethality.^{139, 140} Mechanistically, these inflammasome activators function to increase downstream binding of NEK7 with NLRP3^141-143 and regulate NLRP3 oligomerization, inflammasome assembly, and catalytic cleavage of procaspases.

Recently, a novel TLR4-independent mechanism for triggering pyroptosis was discovered.^{144, 145} Intracellular LPS directly binds to the procaspase-11 recruitment CARD domain, which then oligomerizes and activates procaspase-11.^{146, 147} The induction of pyroptosis, which is a critical event in septic mice, is dependent on Caspase-11 activity.^{44, 148-150} Additionally, pyroptosis in human mononuclear cell lines is mediated by homologous caspase-4/5. The activation of NLRP3 is regulated by Caspase-11, and it is important to note that the cleavage of GSDMD by active caspase-1/4/5/11 leads to the release of the functional gasdermin-N domain. This released domain subsequently forms pores in the membrane, facilitating the active release of inflammatory cytokines and intracellular components (Figure 2).^151-156 The ability of the gasdermin family to form pores plays a significant role in the unique molecular and structural mechanisms underlying pyroptosis.^{157, 158} Interestingly, the pore formed by GSDMD-N enables nonselective ion diffusion without increasing osmotic pressure, unlike membrane blebbing and cell swelling.^{159, 160} Mitochondrial dysfunction may trigger the release of oxidized mtDNA and cardiolipin into the cytosol, consequently activating the NLRP3 inflammasome and leading to the production of IL-1β. Furthermore, the activation of inflammasomes can be driven by the significant efflux of potassium induced by toxins such as α-toxin, which form potassium channels.^161-164

2.6 Neuroinflammatory reflex

Homeostasis of neuroinflammatory reflex system interactions is also considered to be a crucial part of the host response during sepsis.¹⁶⁵ The neuroinflammatory response to sepsis is not due to infection of the brain but rather systemic inflammation associated with innate immunity to the central nervous system (CNS) through various pathways.¹⁶⁶ LPS can activate TLR2 and TLR4 surrounding the cerebrum, choroid plexus and leptomeninges, thus profoundly inducing the transcription of inflammatory mediators throughout the brain parenchyma.¹⁶⁷ However, the neuroinflammatory reflex immune system can suppress inflammation by activating the splenic nerve in the celiac nerve plexus, which causes the spleen to release norepinephrine (NA) and CD4⁺ T cell subsets to secrete acetylcholine. The characteristics of cerebral dysfunction in sepsis are altered mental states, including disorientation and confusion.^{168, 169} Patients with severe formal sepsis related to cerebral dysfunction may develop coma.^{170, 171} Importantly, sepsis-associated cerebral dysfunction can be present without other organ dysfunctions (such as liver failure) that might be associated with encephalopathy.⁴⁰ As a result, patients experiencing cerebral dysfunction associated with sepsis face a significant mortality risk and commonly endure prolonged cognitive and functional aftereffects. The activation/dysfunction of the vagus nerve and endothelium triggers the release of cytokines and neurotoxic mediators. This mechanism exemplifies the brain's immune system operations by transmitting distinct signals along various pathways to the brain. These incoming signals subsequently prompt the CNS's outgoing response, activating both the sympathetic and parasympathetic branches of the autonomic nervous system. Consequently, the activation of the adrenal axis is regulated, which counteracts the inflammatory process and restores cardiovascular homeostasis.^{172, 173} Additionally, the production of cytokines is primarily regulated by a specific division of the autonomic nervous system, particularly the sympathetic branch.¹⁷⁴ Existing data suggest that NA can react to LPS, and elevated levels of NA were observed in close proximity to immune cells, inhibiting the expression of TNF-α and IL-12 (proinflammatory cytokines), while promoting the expression of IL-10 (an anti-inflammatory cytokine).¹⁷⁵ Furthermore, acetylcholine signaling, as a neurotransmitter, also plays a crucial role in regulating the inflammatory response of sepsis, thereby introducing novel avenues for sepsis treatment.^{176, 177}

2.7 Necroptosis

In contrast to caspase-dependent programmed cell death, necroptosis is an unprogrammed and autonomous form of cell death that can be activated by various virulence factors discharged by the pathogen. However, sepsis-induced changes in organ perfusion and hypovolemia can also contribute to necrosis.¹⁷⁸ This process, in turn, can bolster the release of extracellular “alarm proteins” such as HMGB1, thereby exacerbating local inflammation. Therefore, necroptosis plays an essential role in cell death during sepsis. Additionally, necroptosis triggers inflammation and immunogenic events in different diseases by interacting with death receptors and diverse cytoplasmic protein kinases. Unlike necrosis, which usually results in irreversible pathological damage, necroptosis is employed by organisms to acquire appropriate levels of energy supply and promote cell regeneration.^{179, 180} Initially, defined as a type of TNF-induced necrosis, necroptosis was subsequently recognized as a specific kinase-dependent process after being effectively subdued by RIPK1 inhibitors.^181-183 Necroptosis is specifically regulated by signal transduction proteins known as RIPK1 and RIPK3, which operate in complexes of membrane-associated proteins. These complexes are activated by TLR (TLR3, TLR4) via rhomboid-like (RHIM) domain-containing proteins (TRIF, DAI).^184-186 Successive stimuli induce interactions between coupling proteins and kinases and then initiate necroptosis cytoplasmic complexes, thereby responding by inducing signal transduction associated with cell death/survival outcomes.^187-189 Necroptosis can be induced by various stimuli other than TNF/TNFR1, including immune death signals (FASL/TRAIL), bacterial and cellular stress signals (LPS/TLR4, poly(I:C)/TLR3), and type I/II IFNs.^190-192 These stimuli trigger the formation of necrosomes by activating death receptors. Furthermore, the recruitment of RIPK1 to FAS without an inhibitor of apoptosis leads to the assembly of necrosomes and necroptosis.¹⁹³ On the other hand, activation of the cytosolic complex promotes cell survival and proinflammatory signaling through NF-κB and MAPK activation.^{194, 195} However, the mechanisms that determine whether a cell undergoes apoptosis, necroptosis, or cell survival are still not fully understood. Genetic studies have shown that deficiencies in FADD or Caspase-8 can result in RIPK3-mediated necroptosis, leading to embryonic lethality and inflammation.^196-198 In clinical sepsis, levels of RIPK3 were found to be significantly elevated at different time points.^{199, 200} Moreover, higher levels of RIPK3 were associated with an increased incidence of organ dysfunction and septic shock, highlighting the pathological effects of RIPK3-mediated pathways.

Collaboration between necroptosis and pyroptosis can expand the transmission of inflammatory signals, cause tissue injury, and exacerbate organ failure, both of which play pivotal roles in the progression of sepsis.^{201, 202} Although both processes can be triggered by similar stimuli, different pathways are involved in intracellular signaling, targeting distinct death regulatory proteins, specifically RIPK3 or GSDMD.^203-205 The release of intracellular components differs in the active stage of sepsis development; necroptosis leads to the release of HMGB1, while pyroptosis results in the release of the proinflammatory cytokines IL-1β and IL-18.^{26, 206} To explore potential synergistic effects in sepsis-induced injury, studies have investigated the protective effects of dual deletions of RIPK3 and GSDMD. Knockout (KO) of RIPK3/GSDMD or mixed lineage kinase domain-like/GSDMD has demonstrated protective effects against septic shock, systemic coagulation, and multiple organ injury caused by necroptosis and pyroptosis double obstruction.^{207, 208} Both RIPK3 and GSDMD contribute to lytic cell death, which collaborates to amplify necrotic inflammation and the release of TF in macrophages and ECs, resulting in significant tissue damage (Figure 3). Hence, RIPK inhibition is proposed to play a protective role in the sepsis model. A recent in vivo study using LPS-stimulated human kidney tubular epithelial cells showed increased RIPK3 expression, with enhanced insights into the unique role of RIPK3 in regulating mitochondrial function during sepsis, as evidenced by observations of mitochondrial depolarization in vitro.

2.8 Endothelial barrier dysfunction

The endothelium forms the inner layer of cells from blood vessels and lymphatics and plays a vital role in controlling blood flow and vascular tone as well as participating in immune responses.²⁰⁹ The destruction of the endothelial barrier is the most prominent phenomenon in the pathogenesis of sepsis. The integrity of the endothelial barrier is maintained by intracellular junction molecules, which can regulate the anticoagulant and anti-inflammatory immune properties of the endothelial intima in the process of sepsis.²¹⁰ Oxidative stress and inflammation lead to decreased vascular permeability, hypotension, and inadequate organ perfusion and eventually result in organ failure.⁶⁸ These signature features play a key role in the pathogenesis of sepsis-induced multiple organ failure.

Altered endothelial function was activated by disruption of intracellular junction molecule expression and dynamics of cytoskeleton contraction.^{17, 211, 212} In sepsis, a large number of novel therapeutic targets have been discovered, such as sphingosine 1-phosphate 1, PAR1, vascular endothelial growth factor (VEGF), and angiopoietin 1 (Ang-1), which have been found to play a regulatory role in endothelial dysfunction (Figure 3).^213-215 The anti-inflammatory properties of Ang-1 derive from its ability to reduce leukocyte transmigration and cytokine production via Tie receptor 2 (Tie2) receptor signaling at the endothelial barrier.^{47, 216, 217} Inflammation-induced release of Ang-2 disrupts constitutive Ang1–Tie2 signaling by preventing Ang-1 from binding to its receptor in sepsis, thereby reducing physiologic efforts to suppress ongoing inflammation.^218-220 Therefore, the Ang/Tie2 pathway has broad impacts on vascular remodeling, inflammation, and EC survival.^{221, 222} In addition, Ang/Tie2 activation induces the expression of KLF2 via the phosphoinositide-3 kinase/Akt pathway and counteracts VEGF-mediated vascular permeability (Figure 3).²²³

2.9 ER stress

The ER is an intracellular organelle that is involved in protein translocation, folding, posttranslational modification, and further transport to the Golgi apparatus.²²⁴ Moreover, the ER provides a place for calcium storage, lipid synthesis and carbohydrate metabolism.^{225, 226} Under stress and inflammation conditions, the ER loses homeostasis in its function, which is termed ER stress. Unfolded or misfolded proteins accumulate in the ER during sepsis, altering its homeostasis and leading to oxidative stress and severe calcium disorders that result in ER stress.^{227, 228} Recent studies on ER stress signaling have revealed intriguing interactions between ER stress and sepsis-related cell death.^229-231 Recently, many studies have demonstrated that inhibition of ER stress can stabilize protein conformation, promote the transport of mutant proteins and improve ER folding ability.²³² Therefore, ER stress is suggested as a potential therapeutic target for a variety of diseases, including diabetes, cystic fibrosis, sepsis and ischemic brain injury.²³³ In sepsis animal models, markers of increased ER stress (such as glucose-regulated protein 94 [GRP94], CEBP homologous protein [CHOP], and caspase-12) are detected in several organs, including the heart and liver, and these markers are directly connected with the extent of organ dysfunction, which may be a major cause of sepsis-induced multiple organ failure.^234-236 Studies have demonstrated that caspase-12 deficiency confers protection in septic mice and that the presence of caspase-12 leads to enhanced vulnerability to bacterial infection and septic mortality, suggesting a detrimental role of caspase-12 in sepsis.²³⁷ These effects have been demonstrated in cecal ligation and puncture (CLP) models of sepsis, where ER stress leads to abnormal lymphocyte apoptosis in mice during sepsis, suggesting that the ER stress-mediated apoptosis pathway may be a new target for the clinical prevention and treatment of sepsis-induced lymphocyte apoptosis.¹⁵ During sepsis, the transcription factor CHOP is a major inducer of apoptosis in response to ER stress; however, recent evidence suggests an inflammatory role of CHOP as a mediator of the inflammatory response in sepsis.²³⁸ A major role for CHOP, which acts as an amplifier of the inflammatory response in the pathogenesis of sepsis, and the ability of H₂S treatment to counter CHOP signaling via upregulation of NF-E2-related factor 2 are highlighted.²³⁹ ER stress causes abnormal apoptosis in septic animals, suggesting that ER stress-mediated apoptosis represents a potential new target for the clinical prevention and treatment of sepsis.^{240, 241} However, these observations increase our knowledge of the biological mechanisms in the context of ER stress and sepsis and simultaneously shed light on new targets and suggest novel strategies for the treatment of this condition. Further research is warranted to elucidate the exact mechanism of how ER stress contributes to sepsis-associated cell death.

2.10 Selective autophagy

Autophagy is a vital innate immune defense and self-protective mechanism against microbial challenges. It is a self-degradation process of basic metabolic mechanisms that degrades intracellular proteins and dysfunctional organelles to maintain cellular homeostasis and survival.^{242, 243} Increasing evidence shows that autophagy malfunction is associated with many human diseases, such as cancer, neurodegenerative diseases, liver diseases, and inflammatory diseases.²⁴⁴ It is noteworthy that autophagy can be mobilized in various organs in early sepsis, manifested by increased accumulation of autophagic vesicles and enhanced expression of autophagy-related proteins.^{245, 246} Autophagy is likely to have a protective effect in sepsis through the mechanisms described below: elimination of pathogens, neutralization of microbial toxins, regulation of cytokine release, reduction in apoptosis, and promotion of antigen expression.^247-249 Importantly, mitochondria-produced ROS then oxidize HMGB1, which is released by apoptotic cells, resulting in the elimination of its immunogenicity and the inability to activate innate immune cells.^{233, 250, 251} However, during sepsis, the reduction of oxidized HMGB1 by thioredoxin is not efficient.²⁵² HMGB1 directly interacts with the autophagy protein Beclin-1, displacing Bcl-2, which leads to the formation of autophagy initiation complexes and the elimination of harmful oxidative stress triggers.^{243, 252-255} When binding to AIM2, all forms of intracellular HMGB1 that have been reduced by thiol display the highest affinity and subsequently initiate inflammasome signaling during redox stress (Figure 1).²⁵⁶ Therefore, activation of the inflammasome pathway is a critical prerequisite for promoting protective autophagy and enhancing cell survival.

Sepsis-induced autophagy is initiated by binding PAMPs within the microbial structure to PRRs, such as TLRs, which further trigger different intracellular events and lead to increased autophagic activity by promoting the conversion of LC3-I (microtubule-associated protein 1A/1B-light chain 3-I) to LC3-II.^{257, 258} The binding of LPS from gram-negative pathogens to TLR4 activates autophagy through the MAPK/P38 (mitogen-activated protein kinase) signaling axis, while the binding of lipoteichoic acid to TLR2 induces autophagy via the MAPK1/ERK2–MAPK3/ERK1 pathway.^259-261 Thus, different TLR ligands exhibit variable autophagy-inducing abilities.^{262, 263} Although autophagy levels were regulated by the severity of sepsis, mitochondrial biogenesis was successfully restored by injection of cell-permeable TATI-beclin-1 through the PINK1/Parkin and AMP-activated protein kinase (AMPK)/Unc-51-like protein kinase 1 signaling pathways and maintained sepsis cardiac function.²⁶⁴ Beclin-1 plays a vital role in supporting PINK1/Parkin-mediated mitophagy by interacting with Parkin. This interaction helps localize mitochondrial-related membranes, leading to the tethering of ER mitochondria and the formation of mitochondrial autophagosome precursors.^{233, 265} The recruitment and activation of Parkin by the PINK1 protein on the mitochondrial outer membrane is the initial step in this process. Parkin then constructs ubiquitin chains on damaged mitochondria, facilitating their binding with LC3 on autophagosomes to induce mitophagy (Figure 1).^{227, 266, 267} Furthermore, crosstalk can occur between autophagy and apoptosis.^{268, 269} In a CLP mouse model of sepsis, T cell apoptosis was induced, while autophagy in CD4⁺ spleen cells and CD8⁺ spleen cells was downregulated.^{3, 270} In addition, mice with lymphocyte-specific deletion of the Atg5 or Atg7 genes, which are needed for autophagy, had increased mortality, immune dysfunction, and T-cell apoptosis compared with control mice. For example, T cell-specific deletion of Atg5 in mice leads to increased Il-10 production by T cells after CLP.^{271, 272} These findings suggest that inadequate autophagy results in sepsis-induced immunosuppression.

3.1 GSDMD: promoters of inflammation and pyroptosis

Recent studies have found that gasdermin (GSDMs) can induce the transition from apoptosis to pyroptosis, and induce inflammatory and antitumor immunity. Activation of GSDMs can occur through caspases and granzymes, resulting in apoptosis in various scenarios.^{150, 151, 291, 292} For instance, when the level of GSDM expression is insufficient to induce pyroptosis, caspase/granzyme-driven apoptosis can be converted into pyroptosis by expressing GSDMs.^293-295 The occurrence of pyroptosis is primarily associated with inflammation, which arises from the assembly of a multiprotein complex in response to PAMPs, DAMPs, or environmental stress. Notably, the regulatory network of GSDMD, which plays a significant role in inflammatory diseases, has demonstrated its ability to safeguard animals against septic shock or lethal endotoxemia.^{296, 297} Hence, GSDMD is regarded as a promising target for sepsis treatment.

The pore-forming properties of the gasdermin family play a crucial role in pyroptosis, which is a form of cell death with inflammation.²⁹⁸ When active caspase-1/4/5/11 cleaves GSDMD (gasdermin protein), it releases the functional gasdermin-N domains. These domains have the ability to induce liposome permeation and form intrinsic pores in the cell membrane. This process ultimately leads to the formation of membrane pores and the release of inflammatory cytokines, specifically IL-18 and IL-1β. Studies have shown that deleting GSDMD can block pyroptosis and provide protection against sepsis, systemic blood coagulation, and multiorgan damage.^{279, 299, 300} GSDMD not only induces cell death but also contributes to necro-inflammation by releasing TF in macrophages and ECs, resulting in significant tissue damage.^{121, 301} Moreover, evidence suggests that GSDMD is involved in coagulation responses during sepsis. It promotes the release of coagulation factor III (F3) into the bloodstream, leading to systemic coagulation. Activation of GSDMD through caspase 11 (CASP11) is needed for this process.^{60, 302} Interestingly, GSDMD-independent pyroptosis relies on the activation of the purinergic receptor P2 × 7 (P2RX7), but disrupting P2RX7 does not affect the coagulation cascade in septic mice.^{302, 303} This emphasizes the unique role of GSDMD in promoting coagulation.^{304, 305} Additionally, experiments with intraperitoneal injections of nuclear sulfonamides, such as pseudokinase, have shown that they can directly inhibit GSDMD. This inhibition offers protection against sepsis in mice compared with control groups.^{273, 304} Based on these findings, the GSDMD system is considered a potential therapeutic target for treating sepsis and warrants further investigation.

3.2 HMGB1: crucial intermediators between inflammation and coagulation

HMGB1 was initially identified as a protein that binds to the nucleus and enhances the transcription of genes by ensuring the stability of nucleosome formation. Follow-up investigations have validated the involvement of HMGB1 in various cellular processes, including DNA recombination, repair, replication, and gene transcription. These activities are facilitated by the presence of repeated positively charged domains (HMG boxes) within the N-terminal region of HMGB1.³⁰⁶ Thus, human or mouse HMGB1 protein not only functions as a nuclear factor but also functions as a vital cytokine mediating responses to infection, injury, and inflammation.³⁰⁷ Once it is actively secreted by immune cells or the epithelium or released by damaged cells, extracellular HMGB1 can serve as a DAMP and lethal mediator in critical diseases such as sepsis and COVID-19 in both mice and humans.^{308, 309} Recent research has indicated that HMGB1 may potentially trigger inflammasome activation through cell interactions in mouse sepsis models.^{55, 310} As a late-phase alarmin in sepsis, HMGB1 interacts with RAGE to deliver cytosolic LPS, consequently activating the NLRP3 inflammasome in murine macrophages and lung ECs, leading to caspase-11-dependent pyroptosis.^{274, 310, 311} HMGB1 also binds to cellular receptors for RAGE that enter the lysosomes of macrophages and ECs and trigger activation of the transcription factors NF-κB and MAPK.³¹² In addition, HMGB1 destabilizes lysosomal membranes and promotes LPS release into the cytoplasm, leading to pyroptosis, and thus promoting lethal coagulation.^313-315 These results establish a pathway dependent on DAMP for the transportation of LPS into the cytoplasm, triggering the activation of CASP11 and coagulation. This pathway operates independently from the non-DAMP pathways previously documented, which involve the release of outer membrane vesicles by bacteria or the involvement of human guanylate binding protein 1 (GBP1).^316-318

Note that, although the expression of CASP11 induced by LPS requires TLR4, it is not necessary for sepsis death induced by poly(I:C), which may help explain the unsatisfactory results of clinical trials attempting to treat sepsis with TLR4 inhibitors.^{121, 319, 320} These findings suggest that in mice, the activation and formation of pores by HMGB1 are necessary for F3 activation through the exposure of phosphatidylserine (PS).^{306, 313} As a result, PS-binding proteins such as lactadherin and MFG-E8 can restrict cytoplasmic F3 activation induced by LPS in murine macrophages.^{321, 322} An unexpected discovery was that glycine, an osmoprotectant, inhibits the release of F3 from infected BMDMs but does not affect cytoplasmic F3 activation induced by LPS in murine macrophages.^{302, 303, 323-325} Several possible reasons exist for these seemingly contradictory outcomes: the disruption of the plasma membrane by Ninjurin 1 during lytic cell death, including pyroptosis, might also contribute to HMGB1 release and F3 activation during lethal infection in mice.^{278, 324, 326, 327} In vitro studies have concluded that GSDMD-mediated influx of Ca²⁺ promotes F3 activation, partly through anoctamin 6 (TMEM16F).^{328, 329} In contrast to pyroptosis mediated by GSDMD-induced release of HMGB1 in macrophages, GSDMD-deficient mice typically release HMGB1 during endotoxemia.^{243, 299, 330, 331} Consequently, these findings suggest the possibility of overlapping and diverse functions of GSDMD and inflammasomes in regulating HMGB1 release, which may play a role in feedback mechanisms controlling inflammation and coagulation.

3.3 STING: multiple regulatory functions of STING in sepsis

Stimulator of interferon response, cGAMP interactor 1 (STING1 or TMEM173), is a transmembrane adaptor protein that is commonly found in the ER of both human and mouse cells.³³² Recently, the activation of the STING1 pathway in myeloid cells, such as macrophages and monocytes, has been associated with inflammation, coagulation, and tissue damage in mouse models of sepsis. These models include polymicrobial infections or bacteremia induced by Escherichia coli or Streptococcus pneumoniae.³³³ To better understand the role of STING in regulating the innate immune response to sepsis, extensive research has been conducted to identify key regulators of the STING pathway.^{276, 334, 335} Among the potential targets, anapestic lymphoma receptor tyrosine kinase (ALK) has been proposed as a novel regulatory factor for the STING-mediated innate response. ALK, a receptor tyrosine kinase commonly associated with tumors, exhibits relatively low expression levels in healthy individuals.³³⁶

The binding of ALK/EGFR has been theoretically demonstrated to activate STING in innate immune cells dependent on Akt. This finding provides supporting evidence for a novel signaling pathway implicated in the development of sepsis pathogenesis and septic shock.³³⁷ Therefore, the ALK–EGFR–Akt pathway plays a critical regulatory role in the STING-mediated innate immune response, which has been proven to be another important pathogenesis of fatal sepsis. In previous studies, STING mutations were shown to perturb ER calcium homeostasis and drive T cells to overreact to ER stress induced by TCR signaling.³³⁸ This chronic increase in STING-mediated ER stress can effectively trigger the death of T cells due to apoptosis.³³⁹ Additionally, the severity of DIC and mortality in sepsis patients is linked to the expression of STING and GSDMD.^{279, 292, 303} Systemic coagulation mediated by STING is not dependent on the classical STING-induced pathway. It is noteworthy that during endotoxemia in mice, STING1-induced apoptosis of CD4⁺ T cells promotes inflammation-induced immunosuppression, indicating the potential of STING as a target for treating T cell-mediated diseases.^{278, 324} Moreover, when cytosolic DNA is recognized by cyclic GMP–AMP synthase (CGAS), activated STING1 can be moved to lysosomes, causing lysosomal rupture and subsequent K⁺ efflux, ultimately leading to the activation of the NLRP3 inflammasome.^{340, 341} GSDMD has been shown to either limit or induce CGAS-mediated STING1 activation by measuring the production of type I interferon genes in murine macrophages and ECs, respectively. This suggests that GSDMD has a dual function in the feedback control of STING1 activation.^{277, 342} In summary, the activity of STING1 is regulated by various binding proteins or posttranslational modifications associated with immunity, autophagy, and cell death, highlighting the complex contribution of STING1 to the development of diseases, particularly sepsis and coagulation-related conditions.^{275, 343} Recently, the STING pathway has been proposed as another viable therapeutic target for sepsis due to its multiple effects on immune homeostasis, coagulation, and inflammation.

3.4 C3a/C5a: the double-edged sword of the innate immune system

In the complement system, the most crucial molecule to inhibit is C3 because it obstructs all connections of the entire system to C3 and downstream.^{71, 78, 344, 345} Hence, there is a significant potential to decelerate all complement-mediated functions by obstructing C3.^{73, 346} It was verified that a synthetic inhibitor of C3 convertase (Compstatin) not only impeded complement activation during E. coli-induced sepsis in baboons but also mitigated other inflammatory responses, activation of coagulation, and multiple organ failure.^{71, 347} It is noteworthy that while blocking C3, the cascade is not completely intercepted by the other constituents, since activation of the classical and lectin pathways produces some C5 convertase, and some activation of the terminal pathway transpires.^{130, 348} Nevertheless, blocking C3a or C3aR might also serve as an alternative to inhibiting the anaphylactic toxin activity of C3 cleavage, although this approach has not been tested in clinical trials. C5a demonstrates the highest potency as an inflammatory mediator of the complement system.^{69, 79, 349} Blocking C5a signaling improved outcomes in multiple E. coli-induced sepsis animal models, including monkeys, mice, and rats with a variety of types of microbial sepsis.^{350, 351} According to a recent investigation, baboon survival was enhanced, coagulopathy was decreased, and endothelial and barrier function were maintained in an E. coli-induced sepsis model through the utilization of RA101295, a C5 peptide inhibitor. This study observed a noteworthy reduction in consecutive identical words, thereby demonstrating the effective role of RA101295 in combating sepsis-related challenges.^{71, 352} Therefore, C3a and C5a receptors may have different regulatory functions in sepsis. Nevertheless, recent research has illustrated that the utilization of a C5a receptor blocker resulted in improved survival rates across various forms of microbial sepsis. Conversely, the administration of a C3a receptor blocker led to diminished survival rates.^353-355 Consequently, it can be concluded that complement activation plays a significant role in the development of multiorgan dysfunction syndrome during sepsis. Consequently, the inhibition of complement fragments and/or their receptors could be a valuable approach in sepsis treatment. Despite the successful development of agents specifically designed to inhibit complement activation, their complexity and, in some instances, unfavorable effects hinder their therapeutic applicability. Further investigation into the alterations in complement activation throughout sepsis may aid in identifying patients and interventions that are most likely to yield positive outcomes.

3.5 Park 7: park7 is a unique antioxidant that blocks inflammation and ROS

Macrophages, as crucial cells in the innate immune system, have a significant impact on inflammation and immune processes.^356-358 In the initial phases of sepsis, macrophages commonly exhibit proinflammatory characteristics. However, an excessive inflammatory response by macrophages can trigger macrophage apoptosis, disrupt macrophage polarization, and ultimately lead to immunosuppression.^{359, 360} ROS can activate various TLR signaling pathways, subsequently governing macrophage function. These ROS are generated through the activation of NADPH oxidase.^{361, 362} Currently, the indispensable role of Parkinson's disease protein 7 (Park 7) in regulating the production of ROS by interacting with P47PHOx, a subunit of NADPH oxidase, is well known.^362-364 This interaction activates NADPH oxidase, leading to an increase in ROS levels in macrophages, thereby initiating TRL signaling and enhancing macrophage function to counteract sepsis-induced immune suppression.^{365, 366} Consequently, mice with Park7 KO can serve as an ideal model for studying advanced sepsis.^{367, 368} More specifically, during the later stages of sepsis, the impaired activation of macrophages can be attributed to the dampened TLR/NF-κB and/or TLR/MARKs signaling pathways induced by LPS/LTA/PGN/proinflammatory cytokines.^{285, 286, 369} P47phox, which serves as the proenzyme subunit of NADPH oxidase, plays a critical role in the assembly of NADPH oxidase.^{370, 371} The interaction between Park 7 and p47phox facilitates the phosphorylation and membrane translocation of p47phox, ultimately leading to the formation of the holoenzyme complex.^{372, 373} Subsequent activation of NADPH oxidase results in the generation of ROS, which, in turn, activate the MAPKs and NF-κB signaling pathways downstream of TLR signaling, thereby promoting macrophage activation.^374-376 Activated macrophages play a crucial role in preventing sepsis-induced immune suppression by releasing proinflammatory cytokines, eliminating pathogens, polarizing toward the M1 phenotype, and enhancing autophagy capacity.^377-379 Based on this knowledge, targeting the Park 7/P47PHOX/ROS axis could prove to be an effective therapeutic strategy against sepsis-induced immunosuppression.

3.6 SQSTM1: ‘Jack of all trades’ in health and septic shock

Sequestosome 1 (SQSTM1/P62) serves not only as a receptor for macroautophagy/autophagy but also as a versatile protein engaged in signaling cascades during inflammation and oxidative stress.^{380, 381} In vitro and in vivo experiments have emphasized the role of extracellular SQSTM1 as a deadly inflammatory mediator leading to sepsis and septic shock mortality, according to recent research.^{280, 382} In brief, SQSTM1 release occurs during tissue damage or microbial invasion, mainly in two ways: passively and actively. Passive release of SQSTM1 from human or mouse macrophages and monocytes is induced by the activation of TLR4-mediated transactivation of the Sqstm1 gene and STING1-mediated phosphorylation of the SQSTM1 protein at Ser403 in the active mode.^{383, 384} In vitro, cytoplasmic LPS also triggers GSDMD-dependent pyroptosis, which further facilitates the passive release of SQSTM1 from macrophages and monocytes.^{279, 280} Following its release, extracellular SQSTM1 binds to the INSR located on the membrane, resulting in the activation of glycolysis. This activation subsequently leads to the production of proinflammatory cytokines in a manner dependent on the transcription factor NF-κB.^{385, 386} Functionally, the pathway involving SQSTM1 and INSR can have a significant impact on tissue damage, systemic inflammation, organ failure, and mortality in experimental sepsis mouse models when either genetically deleted or pharmaceutically inhibited.^{387, 388} Moreover, the severity of sepsis in patients has been found to be associated with the activation of the SQSTM1–INSR pathway.

The SQSTM1–INSR axis activates NFKB and triggers inflammation, which occurs in macrophages via a pathway involving phospholipase C gamma 1 (PLCG1)-dependent lipid peroxidation. This process leads to elevated pyroptosis, a form of cell death.^{389, 390} Blocking SQSTM1 by administering neutralizing monoclonal antibodies or deleting the INSR or SQSTM1 genes in bone marrow cells (BMCs) effectively safeguards mice from fatal sepsis.^{279, 280} Patients with bacterial sepsis show increased levels of SQSTM1 and INSR mRNA in peripheral blood mononuclear cells, along with elevated serum SQSTM1 concentrations. This finding further supports the idea that the SQSTM1–INSR pathway may have a detrimental role in human sepsis.^{391, 392} Additionally, higher serum SQSTM1 concentrations have been found to be an independent risk factor for patients with steatosis and liver inflammation.^36,398 Therefore, revealing its proinflammatory potential would help shed light on the regulatory role of extracellular SQSTM1 in the pathogenesis of various inflammatory diseases.

3.7 PD-1/PD-L1: regulatory immune checkpoints in sepsis

PD-1/PD-L1 signaling pathways play a crucial role in the immunosuppression caused by sepsis. These pathways act through both the innate and adaptive immune systems, leading to various detrimental effects. These effects include the depletion of T cells, lymphopenia, apoptosis, impaired proliferation, decreased production of proinflammatory and anti-inflammatory factors, compromised secretion of antigen-presenting cells, and reduced functionality of BMCs.^{282, 393} Specifically, PD-1/PD-L not only play critical roles in tumor-infiltrating T lymphocytes but are also highly expressed in nonimmune cells, and the expression level is correlated with organ injury.^{27, 394} In addition to inhibiting the number and functional activity of T cells, PD-1 can also bind to its receptor PD-1 to deliver a coinhibitory signal to negatively regulate T cell activation and mediate apoptosis.^395-398 In short, PD-1 binding to PD-L inhibited T cell activation and cytokine production, which may provide new guidance for sepsis defense.^{399, 400}

Enhanced PD-L1 expression has been documented in various cell types in sepsis.^{283, 400-403} During sepsis, PD-L1 expression is increased on both stromal cells and DCs.^{404, 405} Splenic ECs obtained from sepsis patients exhibited higher levels of PD-L1 expression compared with the spleens obtained from patients who experienced brain death or needed emergent splenectomy due to trauma.^{401, 402, 406-409} However, multiple other studies have demonstrated a higher expression of PD-1 in memory subsets of B cells and CD4⁺ T cells among individuals suffering from sepsis.^{410, 411} Experimental investigations conducted on septic mice have confirmed the reprogramming of monocytes or macrophages during sepsis, resulting in decreased expression of human leukocyte antigen-DR (HLA-DR) and the release of proinflammatory cytokines. This is then followed by a decline in antigen presentation capability and phagocytosis.^{33, 412-414} Elevated levels of PD-1/PD-L1 observed on the surface of peripheral blood monocytes or macrophages have been associated with reduced phagocytosis, a decrease in the secretion of proinflammatory cytokines such as IL-6 and TNF-ɑ, and an increase in the release of anti-inflammatory cytokines such as IL-10.^{398, 415} Furthermore, clinical trials have revealed a decrease in the expression of HLA-DR and CD86, which are costimulatory molecules, on monocytes/macrophages in sepsis patients. Conversely, the expression of PD-1 and PD-L1 is elevated, indicating a correlation with cellular dysfunction. These biomarkers may serve as indicators of monocyte or macrophage dysfunction in patients with sepsis.⁴⁰⁸ In vitro studies have shown that PD-1-mediated immune dysfunction of macrophages during sepsis occurs through the alteration of macrophage migratory ability. However, the administration of an anti-PD-1/PD-L1 antibody can reverse these dysfunctions in sepsis patients.^{416, 417} Similarly, mice deficient in PD-1 have demonstrated an enhanced phagocytic capacity in macrophages. Kupffer cells, which act as macrophages in the liver and remove bacteria from the bloodstream, also exhibit increased cytokine production during sepsis.^{403, 417-420} The results presented here reinforce the concept that the abnormal activation of the PD-1/PD-L1 pathway is the main reason behind immunotherapy in individuals suffering from sepsis.

3.8 JAK/STAT: JAK–STAT signaling blocks sepsis-induced immunosuppression

Janus kinases and signal transducers and activators of transcription (JAKs–STATs) are pivotal signaling components downstream of cytokine receptors.²⁸⁹ During the onset of sepsis, the two main causes of death are SIRS-induced multiple organ failure and sepsis-induced immune rejection responsible for late infections.^{421, 422} The signal conduction of these two lethal processes is involved in the regulation of JAK–STAT pathways.⁴²³ SIRS is mainly mediated by STAT1 and STAT4, whereas compensatory anti-inflammatory response syndrome is predominantly mediated by STAT3 and STAT6, with the involvement of JAK1-2 and tyrosine kinase (TYK) 2 for both cases.^{424, 425} Moreover, the JAK–STAT pathway also plays an essential role in cell proliferation and apoptosis, mainly mediating sepsis-triggered emergency hematopoiesis and sepsis-induced organ dysfunction.^{290, 426} Thus, targeting the JAK–STAT signaling pathway in sepsis may reduce sepsis-induced MOF and sepsis-induced immunosuppression. The recent successful development of JAK–STAT targeted therapy in oncology and hematology has provided new insights for the treatment of sepsis patients.

3.9 SESN2: the protector of the immune system

Sestrin2 (SESN2) is a highly evolutionarily conserved protein associated with cellular responses to various stresses; it protects against oxidative stress, DNA damage, hypoxia, nutritional stress, ER stress, autophagy, metabolism and inflammation.^{110, 427, 428} SESN2 functions as an antioxidant, activating AMPK, and suppressing mTORC1 signaling. Initially, discovered as a downstream mediator of p53, it is induced by various detrimental environmental strains, including oxidative stress, ER stress, energy stress, and age- and obesity-related metabolic disorders.^{17, 429} SESN2, regulates autophagy, ER stress, and inflammasome activity under various conditions, exhibiting pleiotropic biological functions in cell homeostasis and metabolic homeostasis.⁴³⁰ Particularly, when facing ER stress, SESN2 is upregulated and exerts a protective effect against ER stress-induced cell death. In situations of high stress, the activation of ER stress occurs as a result of the build-up of proteins that are unfolded or improperly folded within the lumen of the ER. This activation is then followed by the initiation of three branches within the pathway for ER stress. SESN2 expression is upregulated by PKR-like ER kinase (via eukaryotic translation initiation factor 2α and activating transcription factor 4), inositol-requiring enzyme 1 (via Xbox binding protein 1, TRAF2 and c-Jun N-terminal kinase [JNK]) and activating transcription factor 6. SESN2 functions as a regulatory mechanism to mitigate the cellular response to ER stress or ER stress-induced apoptosis.²⁴ The results showed that hypoxia and NO strongly induced the expression of SESN2 in hypoxia-induced factor-1α, and the activation of SESN2 could prevent the peroxidation of peroxiredoxin, thus also acting as a cell protector.²⁸⁷ SESN2 expression has been detected in various immune cells, including monocytes, macrophages, natural killer (NK) cells, and T lymphocytes.²⁸⁸ Overexpression of SESN2 reduced ER stress-related cell death, and SESN2 knockdown exacerbated the extent of ER stress, resulting in enhanced ER stress-mediated apoptosis.²⁴ SESN2 expression may exert beneficial effects on immune cell function by activating AMPK, inhibiting mTORC1 signaling, inhibiting JNK pathway activation, reducing the degree of ER stress, activating autophagy, or attenuating the NLRP3 inflammasome.²⁸⁸ Therefore, SESN2 has multiple regulatory effects and could be a promising therapeutic target and play a protective role in various inflammatory diseases, such as sepsis.