Crosstalk between neutrophils and hepatic stellate cells in liver injury and repair: Mechanisms and implications

Abstract

The liver has a remarkable capacity for regeneration; however, persistent or dysregulated inflammation can impair tissue repair, thereby promoting chronic fibrosis and, ultimately, cirrhosis. Neutrophils and hepatic stellate cells (HSCs) are key mediators of this pathological transition. Beyond the traditional view of neutrophils as short-lived effectors of tissue injury, accumulating evidence indicates that they exhibit marked plasticity and can adopt either pro-inflammatory or pro-repair phenotypes depending on the microenvironment. Recent studies have revealed a dynamic and bidirectional communication network between neutrophils and HSCs that regulates injury severity, resolution of inflammation, tissue repair, and fibrotic progression. This review summarizes the mechanistic evidence for neutrophils-HSCs crosstalk, with a focus on the following aspects: (1) Initiating soluble mediators, including reactive oxygen species, proteases, and cytokines; (2) Signal amplification through extracellular matrix remodeling and mechanotransduction; and (3) The role of neutrophil extracellular traps as a signaling platform that sustains profibrotic programs. In addition, the review highlights how temporal factors and disease context determine whether this crosstalk facilitates regenerative repair or evolves into a self-amplifying fibrotic circuit. Finally, translational opportunities are discussed, and it is proposed that next-generation antifibrotic strategies should prioritize phase-specific and precision-based remodeling of key regulatory nodes.

Key Words: Liver injury; Liver repair; Neutrophils; Hepatic stellate cells; Cell crosstalk

Core Tip: Crosstalk between neutrophils and hepatic stellate cells is a crucial regulatory axis that determines whether liver injury resolves or progresses to fibrosis. This interaction is highly context- and stage-dependent. During acute injury, it triggers transient reparative signaling; however, in diseases such as metabolic dysfunction-associated steatohepatitis, liver cirrhosis, and hepatocellular carcinoma, it promotes persistent pathological activation. Targeting specific nodes within this crosstalk (e.g., neutrophil extracellular traps, NOD-like receptor family pyrin domain containing 3, transforming growth factor-β) enables a precision-remodeling strategy, offering stage-specific anti-fibrotic interventions while preserving essential immune repair functions.

INTRODUCTION

Liver injury is defined as dysfunction and morphological changes of the liver due to different pathogenic factors, such as virus infection, alcohol intake and drug administration. The clinical manifestations of liver injury include hepatocellular necrosis, hepatic dysfunction and liver fibrosis[1,2]. Across different etiologies, the disease commonly progresses along a steatosis-inflammation-fibrosis continuum and may ultimately lead to liver failure, cirrhosis, or hepatocellular carcinoma (HCC)[1]. Fibrosis represents a critical turning point in this process and is driven by hepatocyte injury, immune activation, and hepatic stellate cells (HSCs) transdifferentiation[3]. Although inflammation initially facilitates debris clearance and tissue regeneration, persistent or dysregulated signaling, such as that mediated by tumor necrosis factor-α (TNF-α), and interleukin-1β (IL-1β), redirects tissue repair toward excessive extracellular matrix (ECM) deposition, thereby promoting fibrosis and cirrhosis[4,5]. Therefore, identifying the determinants that distinguish regenerative repair from pathological fibrosis is essential for the development of targeted antifibrotic therapies[4].

Neutrophils are among the earliest and most abundant leukocyte populations that infiltrate the liver after hepatic injury[6,7]. Although neutrophils are recruited at an early stage of hepatic injury, their roles are not uniformly injurious. Accumulating evidence indicates that neutrophils exhibit marked plasticity and can adopt pro-resolution and pro-regenerative phenotypes in specific microenvironments, thereby contributing to the regulation of inflammation resolution, tissue repair, and regeneration[8-10]. At the same time, HSCs, the principal effector cells in liver fibrosis, are now recognized as critical regulatory hubs that sense inflammatory and mechanical signals, secrete chemokines, and control the localization and function of immune cells[11-13]. In recent years, increasing attention has been directed toward the interaction between neutrophils and HSCs, suggesting the existence of a dynamic and bidirectional crosstalk neutrophil extracellular traps (NETs) between these two cell types. The functional consequences of this interaction are highly dependent on the temporal stage and disease context[14,15]. This review integrates these mechanisms across the continuum of injury, resolution, repair, and fibrosis, and highlights translational precision-remodeling strategies centered on NETs, inflammasomes, and metabolic reprogramming.

THE BIOLOGICAL BASIS OF NEUTROPHILS AND HSCS IN LIVER INJURY

Neutrophils heterogeneity and plasticity in the liver injury microenvironment

Analogous to the M1/M2 paradigm in macrophages[16], neutrophil polarization in the tumor microenvironment was initially classified into N1 (antitumor) and N2 (protumor) phenotypes[17,18]. In recent years, this classification framework has been extended to the field of tissue injury and repair to describe the bidirectional regulatory role of neutrophils in the inflammation-fibrosis process[19]. However, this analogy has important limitations. First, the N1/N2 classification is derived primarily from the transcriptomic features of tumor models, whereas the corresponding molecular markers in liver injury and repair remain insufficiently defined. Second, neutrophil phenotypic transition is a continuous and dynamic process rather than a simple binary state. Recent single-cell transcriptomic studies have substantially refined this concept, demonstrating that neutrophils exist along a functional continuum rather than as discrete N1/N2 categories. Xie et al[20] mapped neutrophil maturation and tissue-imprinted transcriptional states across multiple organs and revealed graded shifts in metabolic, antimicrobial, and immunoregulatory programs that cannot be adequately captured by a binary classification. The consensus statement by Quail et al[21] further proposed that neutrophils functional states should be defined by integrated transcriptional, metabolic, and surface marker signatures, with phenotypes distributed along a multidimensional landscape shaped by tissue context, time after injury, and the local cytokine milieu. Accordingly, in this review, the terms “N1” and “N2” are used only as functional shorthand for the opposite ends of this continuum, namely pro-inflammatory/cytotoxic and pro-resolving/pro-reparative states, while acknowledging that intermediate, hybrid, and tissue-specific subsets, such as CD177+, SiglecF+, PLAUR+, and PD-L1+ populations, coexist and may dynamically interconvert in the injured liver[22].

Activation of HSCs and their immunoregulatory capacity

HSCs reside in the space of Disse and maintain a quiescent phenotype under physiological conditions, which is characterized by the storage of lipid droplets, particularly vitamin A[11]. In response to liver injury, HSCs undergo activation and transition into a myofibroblast-like phenotype, with increased proliferation, contraction and greatly increased ECM secretion. This conversion is key to the development of liver fibrosis[23]. Besides their role in liver fibrosis, HSCs also play significant immunoregulatory roles. They are equipped with pattern recognition receptors, are responsive to inflammatory cytokines and produce a range of chemokines, which affect the migration and positioning of immune cells[24,25]. Simultaneously, HSCs-driven ECM deposition changes the mechanical properties of the liver, which amplifies the activation status of HSCs through mechanotransduction mechanisms, such as Yes-associated protein. This process establishes a positive feedback loop that promotes fibrotic progression and subsequently affects immune cell behavior[26,27]. Collectively, these findings indicate that HSCs function as integrative hubs linking inflammatory and mechanical signals, rather than merely serving as passive fibrogenic cells. Recent single-cell RNA sequencing studies have further revealed substantial intrahepatic heterogeneity of HSCs along the lobular axis and during fibrogenesis. Dobie et al[28] identified zonated HSCs subsets in the murine liver, showing that central vein-associated HSCs constitute the dominant pathogenic collagen-producing population in centrilobular fibrosis, whereas portal vein-associated HSCs preferentially contribute to portal tract fibrosis. Filliol et al[29] further demonstrated that distinct HSCs subpopulations exert opposing effects on hepatocarcinogenesis: Cytokine-producing HSCs restrain HCC development, whereas myofibroblastic HSCs promote tumor progression. These findings indicate that the immunoregulatory and fibrogenic outputs of HSCs, including the chemokines, adhesion molecules, and matrix components involved in neutrophils crosstalk, are subset and zone-specific rather than uniform. This issue should be considered in future mechanistic and therapeutic studies of neutrophils-HSCs interactions.

MECHANISMS BY WHICH NEUTROPHILS DRIVE HSCS ACTIVATION

Inflammatory mediators and oxidative signals initiate HSCs activation

In the early phase of liver injury, neutrophils are the first major effector cells to be mobilized and recruited to the liver in large numbers[30,31]. The pro-inflammatory mediators released by neutrophils, including reactive oxygen species (ROS), myeloperoxidase (MPO), neutrophils elastase (NE), TNF-α, and IL-1β, constitute an inflammation-oxidative stress driver module that promotes HSCs activation[32]. Multiple studies support this mechanism. In a metabolic dysfunction-associated steatohepatitis (MASH) model, Pulli et al[33] showed that MPO aggravates hepatocyte injury and promotes HSCs activation and fibrogenesis, whereas MPO deficiency reduces transforming growth factor-β (TGF-β) levels, the number of activated HSCs, and fibrosis severity. In addition, hypochlorous acid generated by MPO further amplifies this effect through oxidative fragmentation of the ECM. Therefore, MPO is now recognized as a pivotal neutrophils-derived mediator linking hepatocyte injury to HSCs activation.

Beyond oxidative mediators, neutrophils-derived TNF-α and IL-1α/β promote HSCs activation by disrupting the balance between matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases, upregulating innate immune receptors such as toll-like receptors2 (TLR2) and cluster of differentiation 14, and inducing pro-inflammatory and ECM-remodeling gene programs. Together, these effects establish an inflammation-matrix metabolic imbalance circuit that sustains the myofibroblast phenotype[34,35]. Neutrophils-derived proteases, including NE and MMPs, further reinforce this circuit through ECM degradation and the exposure of bioactive matrix fragments, which alter local mechanical cues[27,35]. In vitro studies[36] have indicated that S100A8/A9 secreted by neutrophils promotes the migration of bone marrow-derived mesenchymal stem cells (BMSCs) and HSCs to the injured liver, where these cells contribute to fibrogenesis through transdifferentiation into myofibroblasts.

Neutrophils-derived extracellular vesicles (EVs) play an important regulatory role in HSCs activation. Relevant studies[37] have demonstrated that NE transported by EVs can directly activate HSCs through the extracellular signal-regulated kinase 1/2 signaling pathway, thereby promoting the progression of liver fibrosis. Treatment with the NE-specific inhibitor sivelestat effectively attenuates the activity of this signaling axis, underscoring the functional importance of EV-derived NE in driving the fibrotic phenotype of HSCs.

NETs as central amplifiers of fibrotic signals

NETs are composed of decondensed chromatin DNA, citrullinated histone H3 (Cit-H3), and granule proteins, including MPO and NE, and form relatively stable extracellular mesh-like structures within inflamed tissues[38]. These structures spatially anchor and amplify inflammatory signals[39]. NETs accumulation has been observed in various liver disease settings and is positively correlated with the severity of inflammation, tissue injury, and fibrosis progression[40,41]. Notably, the profibrotic effects of NETs involve synergistic interactions among multiple components, particularly in MASH-related fibrosis.

Babuta et al[14] provided evidence that NETs may activate NOD-like receptor family pyrin domain containing 3 (NLRP3), thereby contributing to the induction of a profibrotic phenotype in HSCs in vitro. In vivo studies have demonstrated that DNase-mediated disruption of NETs attenuates the activation of monocytes and HSCs, thereby reducing liver injury and fibrosis. These findings underscore the importance of the NETs-NLRP3/IL-1β pathway in the exacerbation of fibrosis. In a metabolic model of MASH fibrosis, Xia et al[39] further demonstrated that NETs not only facilitate HSCs activation but also induce metabolic reprogramming, which is characterized by enhanced mitochondrial respiration, increased glycolysis, and altered metabolic profiles. This process promotes fibrosis through pathways such as toll-like receptor 3 (TLR3)/cyclooxygenase-2 (COX-2)/prostaglandin E₂ (PGE₂), suggesting that NETs can mediate the transition from inflammation to fibrotic remodeling. In addition, Wang et al[42] demonstrated that extracellular histones stimulate collagen expression in LX-2 cells through toll-like receptor 4 (TLR4) -MyD88 signaling and promote liver fibrosis in vivo. These findings indicate that the extracellular histone-TLR4 axis may serve as a potent profibrotic stimulus. Given that histones are key components of NETs, these results provide mechanistic support for the direct activation of HSCs by NETs-associated damage-associated molecular patterns (DAMPs). This multi-component synergistic mechanism makes NETs an effective platform for amplifying profibrotic signals.

NEUTROPHILS REGULATE HSCs FATE TRANSITION

Shifting towards promoting the restoration of neutrophils function

Neutrophils are not simply removed from the liver by apoptosis as DAMPs and pro-inflammatory cytokines disappear; instead, they can be re-programmed from an early pro-damage to a pro-resolution/pro-regenerative state, thereby contributing to regenerative inflammation. This is seen in models of acute injury such as acetaminophen (APAP) liver injury where neutrophils exacerbate early tissue injury but then drive inflammation resolution and tissue repair through, in part, monocyte/macrophage fate[43]. In the repair phase, neutrophils produce collagenolytic MMPs, such as MMP-8 and MMP-9, to remove ECM, decrease matrix density and stiffness, and potentially mediate HSCs deactivation or reprogramming through mechanotransduction signaling[44,45]. In line with this, in cholestatic injury, neutrophils depletion decreases collagenase activity and impairs early type I collagen turnover, further suggesting a role for neutrophils in the initiation of fibrosis repair[46]. On the other hand, neutrophils can undergo a pro-regenerative program by releasing hepatocyte growth factor (HGF) and modulating the sinusoidal niche. Specifically, CD177+ neutrophils increase transendothelial migration and HGF production by MMP-9 mediated ECM remodeling to support regeneration[47]. Thus, the timely switch to pro-repair neutrophils programs may be a key factor in determining the fate of liver injury and/or progression to fibrosis.

Facilitating the clearance of activated HSCs and fibrosis regression

Fibrosis regression is not simply a passive decrease in collagen, but a step-wise process that involves several key steps: (1) Remodeling and degradation of the ECM; (2) Inactivation or removal of myofibroblasts, mainly activated HSCs; and (3) Shift of the inflammatory environment from pro-fibrotic to pro-regressive[48,49]. In this process, neutrophils may help the removal of the fibrotic effector cells (activated HSCs) by indirectly regulating the function of macrophages to remove the activated HSCs[29].

Achieving myofibroblast clearance by driving repair-type macrophages

Research indicates[50] that neutrophils can induce monocytes/macrophages to adopt a pro-repair phenotype through signals such as ROS, thereby enhancing their capacity to clear necrotic debris and apoptotic cells and ultimately promoting inflammation resolution and tissue reconstruction. In the context of fibrosis, pro-regressive macrophages are considered to contribute to fibrosis regression by phagocytosing and eliminating apoptotic or senescent myofibroblasts while secreting regulatory factors that promote ECM degradation.

Promoting the activation of the inactivation-apoptosis-senescence clearance pathway for HSCs

Neutrophils further influence fibrosis regression by engaging three programmed HSCs exit routes: Phenotypic deactivation to a near-quiescent state, apoptotic clearance, and induction of senescence followed by immune-mediated removal[49,51,52]. Within this tripartite framework of reprogramming, matrix softening, and clearance induction, neutrophils function as indirect determinants of HSCs fate.

REVERSE REGULATION: THE ROLE OF HSCS IN MODULATING NEUTROPHILS FUNCTION

The active role of HSCs in neutrophils recruitment, retention, and Functional polarization

Hepatocytes can secrete CXC chemokines under various conditions of liver injury and inflammation. In certain models, hepatocytes have been identified as a major source of the chemotactic factor CXC chemokine ligand 1 (CXCL1)[53,54]. The C-X-C chemokine receptor 2 (CXCR2) signaling axis has been directly linked to hepatic neutrophils infiltration, underscoring its importance in regulating inflammatory responses in the liver[53]. In human systems, primary HSCs and HSCs-derived cell lines can produce chemokines such as interleukin-8 in response to pathogen-associated molecular patterns or inflammatory cytokines, thereby potentially driving neutrophils recruitment and activation[55]. In addition, activated HSCs can upregulate intercellular adhesion molecule 1 and may also regulate the expression of adhesion molecule families such as vascular cell adhesion molecule. This regulation increases the likelihood of neutrophils retention at the site, thereby prolonging tissue residence, a process relevant to inflammatory cell recruitment and adhesion[56,57]. Beyond mobilization, activated HSCs can produce granulocyte-macrophage colony stimulating factor and interleukin 15, which prolong neutrophils survival, suggesting that HSCs can influence neutrophils longevity by modulating the cytokine milieu[15]. Furthermore, Liu et al[58] reported that activated HSCs stimulate neutrophils to release NETs, which in turn further activate HSCs, thereby perpetuating the fibrotic process. This mechanism indicates that HSCs can establish a pro-inflammatory and profibrotic positive feedback loop.

Activation vs senescence: Context-dependent 'inflammatory brakes'

During the resolution phase of inflammation and tissue repair, changes in the activation state of HSCs are pivotal for disrupting the feedback loop described above. Studies on fibrosis regression suggest that a substantial proportion of activated HSCs do not undergo complete apoptosis; instead, they undergo deactivation or phenotypic regression, accompanied by downregulation of inflammation- and fibrosis-related genes[59,60]. This process weakens chemotactic gradients and facilitates neutrophils clearance. In addition, senescence of activated HSCs has been shown to restrict liver fibrosis progression and enhance immune-mediated clearance of pathological stromal cells[61]. It should be noted that the senescence-associated secretory phenotype is highly context-dependent and may involve interleukin-6 (IL-6) and CXCR2 ligands, which at certain stages may paradoxically contribute to immune cell recruitment and regulation of liver regeneration[62]. Therefore, HSCs senescence does not necessarily indicate complete termination of inflammation; its overall effect depends on the stage of injury and the microenvironmental context.

In summary, HSCs need to be seen as active regulators of feedback inflammatory pathways, not just victims of neutrophils-driven inflammation. By producing chemokines, regulating adhesion, survival and ECM, HSCs play a crucial role in regulating the duration of the inflammatory response and the consequent pathology. Such regulation significantly affects the outcomes of liver injury and determines if the injury is repaired or progresses to chronic inflammatory conditions and liver fibrosis. Neutrophils-HSCs interaction is shown in Figure 1.

Figure 1 Central amplifier of fibrosis via neutrophil-hepatic stellate cells crosstalk. Neutrophils mediators (neutrophil extracellular traps, extracellular vesicles, S100A8/A9, cytokines) activate hepatic stellate cells (HSCs), driving extracellular matrix production and chemokine release. Activated HSCs, promote neutrophil recruitment and activation, forming a self-sustaining positive-feedback loop. ROS: Reactive oxygen species; NE: Neutrophils elastase; MPO: Myeloperoxidase; NET: Neutrophils extracellular trap; EVs: Extracellular vesicles; TNF-α: Tumor necrosis factor-α; IL: Interleukin; HGF: Hepatocyte growth factor; TLR: Toll-like receptors; HSC: Hepatic stellate cell; ECM: Extracellular matrix; NLRP3: NOD-like receptor family pyrin domain containing 3.

Integrating neutrophils-HSCs crosstalk into the broader hepatic immune network

Although this review focuses on the neutrophils-HSCs axis, hepatic inflammation and fibrosis are inherently multicellular processes, and neutrophils-HSCs interactions occur within a network that includes Kupffer cells (KCs), monocyte-derived macrophages (MoMFs), dendritic cells, and adaptive immune populations. Resident KCs sense damage-associated molecular patterns and microbial signals and serve as the principal initial source of CXCL1, CXCL2, IL-1β, and TNF-α, thereby recruiting neutrophils. In chronic injury, the loss or functional exhaustion of KCs triggers replacement by Ly6C^hi MoMFs, which adopt scar-associated profibrotic transcriptional programs and partly drive HSCs activation[63]. Single-cell mapping of human cirrhosis by Ramachandran et al[64] identified a TREM2+CD9+ scar-associated macrophage subset that interacts with HSCs and endothelial cells within the fibrotic niche, and further demonstrated that neutrophils are recruited and licensed within this niche rather than acting in isolation. On the adaptive immune side, Th17 cells, IL-17/IL-22 signaling, MAIT cells, and exhausted CD8+ T cells have been shown to amplify or restrain neutrophils-HSCs circuits in MASH, alcohol-related liver disease, and HCC[65]. Therefore, the neutrophils-HSCs interactions discussed in this review should be regarded as one critical, but not exclusive, module within a multicellular inflammation-fibrosis network, and effective therapeutic targeting will likely require simultaneous consideration of macrophage polarization and adaptive immune tone.

DISEASE CONTEXT DEPENDENCY: VARIED OUTCOMES OF NEUTROPHILS-HSCS INTERACTION

Neutrophils-HSCs crosstalk is a stage and context-dependent module. It is transient and reparative in acute injury, progressively self-amplifying and profibrotic in chronic disease, and becomes integrated into the cirrhosis/HCC immune niche to promote angiogenesis, immunosuppression, and therapy resistance.

An operational stage-specific framework for neutrophils-HSCs crosstalk

To translate the conceptual model of context-dependent crosstalk into actionable terms, three operationally defined stages are proposed, each with characteristic molecular signatures, dominant functional outputs, and a distinct therapeutic posture toward neutrophils (Table 1). Stage I (acute injury, hours to days) is characterized by high serum alanine transaminase (ALT)/aspartate aminotransferase, peak hepatic Ly6G^hi neutrophils infiltration, transient elevation of plasma MPO-DNA and Cit-H3, and early α-SMA induction without bridging fibrosis. At this stage, neutrophils serve dual roles, including initial cytotoxic input followed by reparative phagocytosis and miR-223 transfer to macrophages. Therefore, indiscriminate suppression should be avoided. Instead, NETs formation via peptidylarginine deiminase 4 (PAD4) and downstream NLRP3 signaling should be selectively attenuated during the early phase while preserving the later reparative phenotype[14,50,66]. Stage II (chronic active fibrosis, months to years) is characterized by persistent moderate ALT elevation, elevated neutrophil-to-lymphocyte ratio (NLR), sustained plasma MPO-DNA/Cit-H3, FibroScan ≥ 8.5 kPa or enhanced liver fibrosis (ELF) ≥ 9.8, histological stage F2-F3, and increasing profibrotic chemokines such as CCL2 and CXCL10. At this stage, neutrophils-HSCs crosstalk has become self-amplifying. Targeting the NETs-TLR3/COX-2/PGE2 bridge, NLRP3-mediated amplification, and TGFβR1 Locking layers is therefore justified, with the aim of disrupting, rather than abolishing, neutrophils function. Stage III (cirrhosis/HCC, F4 with tumor) is characterized by high NLR, enrichment of intratumoral PD-L1+/PLAUR+ neutrophils, a NETs-rich immunosuppressive niche, and elevated PIVKA-II/AFP. At this stage, neutrophils-HSCs crosstalk sustains an immune-excluded and protumor microenvironment. Combination strategies, such as PAD4 or NLRP3 inhibition combined with TGF-β receptor 1 (TGFβR1) blockade and immune checkpoint inhibitor (ICI), may therefore be warranted, with strict attention to infection risk in cirrhotic patients with immune paresis[67,68].

Table 1 Stage-specific operational framework for neutrophil-hepatic stellate cells crosstalk and therapeutic posture.

Stage	Clinical/histological window	Dominant cellular-molecular signature	Neutrophil functional output	Therapeutic posture
I Acute injury (hours-days)	ALT/AST ↑↑, no fibrosis	Ly6Ghi infiltration peak; transient MPO-DNA/cit-H3↑; early αSMA induction	Dual: Early cytotoxic; late reparative	Selective: PAD4 inhibition + NLRP3 dampening in early window; preserve late reparative phenotype
II Chronic active fibrosis (months-years)	NLR ↑, FibroScan ≥ 8.5 kPa or ELF ≥ 9.8, F2-F3	Sustained MPO-DNA/cit-H3 ↑; CCL2, CXCL10 ↑; persistent NETs-HSCs loop	Self-amplifying pro-fibrotic	Disrupt loop: COX-2 + NLRP3 + TGFβR1; avoid prolonged NETs ablation
III Cirrhosis/HCC (F4+ tumor)	High NLR, PIVKA-II/AFP ↑	PD-L1+/PLAUR+ TANs; NETs-rich immune-excluded niche	Pro-tumor, immunosuppressive	Combination: PAD4 or NLRP3 + TGFβR1 + ICI; close infection monitoring

ALT: Alanine transaminase; AST: Aspartate aminotransferase; MPO: Myeloperoxidase; PAD4: Peptidylarginine deiminase 4; NLRP3: NOD-like receptor family pyrin domain containing 3; NLR: Neutrophil-to-lymphocyte ratio; NETs: Neutrophil extracellular traps; HSCs: Hepatic stellate cells; COX-2: Cyclooxygenase-2; HCC: Hepatocellular carcinoma; TGFβR1: Transforming growth factor-β receptor 1; ICI: Immune checkpoint inhibitor.

Acute liver injury: Time-limited crosstalk and reparative outcomes

In acute liver injury, such as APAP-induced hepatotoxicity, neutrophils infiltration is an early event in the sterile inflammatory response. A time-course study of APAP-induced liver injury defined the early phase as 6-24 hours after injury, corresponding to peak neutrophils infiltration, whereas the late or resolution phase occurred at 48-72 hours after injury and was characterized by necrotic tissue clearance and the initiation of regeneration. It should be noted that this temporal framework is specific to the APAP model and may differ in other types of liver injury, such as ischemia-reperfusion injury (IRI) or acetylsalicylic acid-induced injury[31]. Within specific models and temporal windows, neutrophils are not the primary determinants of hepatocyte death. Increasing evidence indicates a temporal dichotomy in their mode of action: During the early stage of liver injury, tissue damage is driven mainly by intrinsic hepatocyte toxic processes and DAMP-mediated innate immune signaling. In contrast, during the resolution and repair phase after the peak of injury, neutrophils exert reparative functions by phagocytosing necrotic cell debris, regulating inflammation resolution, and promoting the generation of regeneration-associated signals[43,69]. Chauhan et al[70] demonstrated that, in a model of APAP-induced acute liver failure, inhibition of platelet CLEC-2 signaling enhanced neutrophils recruitment and phagocytic function, accelerated necrotic tissue clearance, and facilitated recovery of liver function. These findings suggest that the platelet-neutrophils signaling axis is crucial for regulating post-injury repair and provide further direct evidence for the reparative role of neutrophils in the late phase of acute liver injury. It is also important to emphasize that this reparative pattern depends strongly on the type of injury. In IRI, neutrophils mainly function as amplifiers of inflammation and tissue damage. Their excessive recruitment and activation are closely associated with increased hepatocyte necrosis and microcirculatory dysfunction. In this setting, inhibition of neutrophils infiltration or blockade of their pro-inflammatory effects generally alleviates tissue injury and improves hepatic functional outcomes[71,72]. Therefore, analysis of neutrophils-HSCs crosstalk in acute liver injury should be conducted within the framework of specific pathological subtypes and temporal windows to avoid oversimplified generalization of its biological functions. The temporal relationship between neutrophil phenotypic transition and HSCs activation in acute liver injury is illustrated in Figure 2.

Figure 2 Time-stratified neutrophil polarization and hepatic stellate cells activation after liver injury. Neutrophils shift from N1 (0-24 hours; pro-inflammatory) to N2 (48-72 hours; reparative). Hepatic stellate cells activate during this transition and peak in repair, driving extracellular matrix remodeling and collagen synthesis; persistent activation may sustain fibrosis. ROS: Reactive oxygen species; HSC: Hepatic stellate cell; IL: Interleukin; TGF-β: Transforming growth factor-β; DAMPs: Damage-associated molecular patterns.

Chronic liver disease: Persistent crosstalk and pro-fibrotic positive feedback loops

In chronic liver diseases such as MASH and alcohol-related liver disease, sustained hepatocyte stress, gut-derived inflammatory stimuli, and repeated cycles of cell death and regeneration transform the normally reversible neutrophils-HSCs interaction into a persistently escalating pathological circuit. NETs serve as a principal amplifier of this circuit, and the NETs-NLRP3 and NETs-TLR3/COX-2/PGE2 axes jointly convert chronic neutrophils input into sustained HSCs activation, metabolic reprogramming, and ECM deposition[14,39]. Therefore, neutrophils-HSCs crosstalk in the chronic phase is best regarded not as an accessory inflammatory response, but as a central driver of fibrotic progression.

Cirrhosis and HCC: Immunosuppression and tumor-promoting interactions

As the disease progresses to cirrhosis and HCC, the persistent imbalance between neutrophils and HSCs becomes further embedded within the tumor microenvironment, leading to immunosuppressive and tumor-promoting interactions. Tumor-associated neutrophils adopt a protumor phenotype and promote tumor progression by enhancing angiogenesis, suppressing CD8+ T-cell function, and facilitating invasion and metastasis[73]. At the same time, activated HSCs enhance the protumor activity of neutrophils through secretion of TGFβR1[74]. Intrahepatic activated HSCs can transdifferentiate into cancer-associated fibroblasts (CAFs). Through ECM deposition and tissue remodeling, these cells establish a dense matrix barrier that impedes immune cell infiltration and limits drug penetration into the tumor core, while simultaneously fostering an immune-excluded microenvironment[75,76]. Importantly, CAFs can induce immunosuppressive neutrophils phenotypes, such as PD-L1+ neutrophils, through signaling pathways including IL-6-STAT3, thereby further strengthening the myeloid-dominated immunosuppressive network[77]. Increasing evidence indicates that neutrophils enrichment and the associated immunosuppressive programs are closely associated with poor responses to ICI therapy. Mechanistically, specific neutrophils subpopulations, such as PLAUR+ neutrophils, contribute to the formation of an immunosuppressive microenvironment and drive resistance to anti-programmed cell death protein 1 therapy[78]. Clinically, early dynamic changes in neutrophils-related inflammatory markers, such as the neutrophils-to-lymphocyte ratio, may help predict ICI response and stratify patient prognosis[79].

IMPLICATIONS FOR LIVER INJURY TREATMENT: FROM BLANKET SUPPRESSION TO PRECISION REMODELING

Neutrophils-HSCs crosstalk is phase-dependent: It supports tissue repair during the late stage of acute injury but becomes self-amplifying and profibrotic in chronic disease. Therefore, therapeutic strategies should be tailored to specific time windows and molecular nodes, while avoiding indiscriminate neutrophils depletion.

Acute liver injury

Neutrophils are not always the main effectors of hepatocyte death in the case of APAP hepatotoxicity or some sterile inflammatory disorders. They may have positive functions in the recovery phase following injury. Thus, rather than indiscriminately blocking neutrophils, it is more effective to target excessive NETs and downstream amplification processes. PAD4 enzymatic activity plays a role in histone citrullination and the formation of NETs, and is an upstream, targetable node in the NETs-fibrogenic platform. The PAD4 inhibitor GSK484 has been demonstrated to effectively alleviate NETs-induced damage and improve survival in mice with acute liver failure, suggesting its potential for translation to block excessive NETs during the acute phase[66]. Additionally, the stimulator of interferon genes (STING) pathway contributes to inflammatory amplification and may also be directly involved in fibrosis. Arumugam et al[80] demonstrated that the mitochondrial DNA-STING axis is a critical pathway for HSCs activation and is coupled with TGF-β-induced HSCs transformation. Furthermore, a systematic review of liver disease contexts suggests that STING deficiency or inhibition reduces cell death, inflammation, and fibrosis severity in chemically induced fibrotic models[81]. Therefore, therapeutic strategies may consider the concurrent application of PAD4 inhibition alongside STING inhibition or modulation to disrupt the acute amplification loop. However, this strategy necessitates careful evaluation of potential infection risks and immune defense requirements.

Chronic liver disease

In the chronic stage, the most well-validated therapeutic targets can be grouped into three layers of hierarchical importance: The NETs-HSCs bridging layer, the inflammation amplification layer and the HSCs activation-locking layer.

Firstly, in the “NETs - HSCs bridging layer”, the TLR3/COX-2/PGE2 axis offers key mechanistic support for the conversion of the neutrophils effectors (input) into persistent HSCs activation (output). Xia et al[39] showed that the COX-2 inhibitor celecoxib alleviates MASH-related fibrosis. This finding indicates that the TLR3/COX-2/PGE2 axis could play a potentially important role during the transition from chronic inflammation to fibrosis and during sustained fibrosis amplification. This is best illustrated in MASH, as persistent NETs are prominent. PGE2 (and its precursor COX-2) are critical for the physiological maintenance of several organs including mucosal protection and renal perfusion. As such, the use of this drug in the clinic should focus on a staggered approach, short-term use, and strict patient selection in order to preserve the antifibrotic effect while limiting the potential side effects. Moreover, the novel non-steroidal vitamin D receptor agonist 16i has been demonstrated to suppress the expression of fibroblast activation protein alpha, a crucial protein in the fibrogenesis process, thus blocking the positive feedback loop between NETs and HSCs activation[58].

Secondly, in the inflammation amplification layer, the NLRP3 inflammasome plays a key role in linking the detection of danger signals to the production and release of IL-1β and IL-18. Besides its ability to amplify inflammation, the NLRP3 inflammasome can be directly linked to the HSCs activation NETs axis, thereby triggering liver fibrosis[82]. In vivo studies[14] demonstrated that administering anti-Ly6G antibodies or disrupting NETs, with deoxyribonuclease treatment, reduces HSCs activation, as well as liver injury and fibrosis. NLRP3 suppression by MCC950 or NLRP3 deficiency downregulates NETs, liver injury and fibrosis in mice with MASH and alcohol-induced liver injury. Pharmacology studies have shown that the NLRP3 inhibitor MCC950 reduces inflammatory cell infiltration and ameliorates fibrosis in mice models of MASH, and thus represents a useful tool for mechanistic studies and preclinical development[83]. Additionally, the orally active selective NLRP3 inhibitor dapansutrile (OLT1177) has been found to be safe and effective in inflammatory diseases such as gout, suggesting clinical feasibility[84]. It's worth noting that NLRP3 plays a role in tissue repair. Consequently, over-inhibition may adversely affect tissue regeneration and lead to infections.

Thirdly, within the “HSCs activation-locking layer”, the TGF-β/TGFβR1 (ALK5) axis serves as a key regulator for sustaining the myofibroblastic phenotype and ECM deposition, thereby determining whether the fibrotic program progresses to a stable, self-perpetuating state. The TGFβR1 inhibitor galunisertib (LY2157299) has advanced to clinical trials for conditions such as HCC, highlighting its therapeutic potential and translational applicability[85]. Concurrently, stress pathways, including p38 MAPK, can perpetuate pro-inflammatory and stress signals in chronic inflammatory settings and interact synergistically with TGF-β signaling. Consequently, although the efficacy of the p38 MAPK inhibitor losmapimod is often indication- and endpoint-dependent, its extensive evaluation in human trials supports its use in combination therapy as an anti-inflammatory module rather than as a primary monotherapy[86].

Based on this hierarchical model, we propose a staged intervention strategy for progressive fibrosis targeting three key nodes: COX-2 (bridging), NLRP3 (amplification), and TGFβR1 (locking). This approach aims to disrupt the NETs-HSCs activation-inflammation loop, thereby enhancing antifibrotic efficacy while avoiding excessive suppression of physiological immune repair (Table 2).

Table 2 Intervention targets for staged management of chronic liver diseases.

Intervention layer	Key way	Clinical evidence	Drugs under development	Transformation challenges
NETs-HSCs signal bridging (fibrosis initiation phase)	TLR3/COX-2/PGE2 axis, TLR4-MyD88 axis	The COX-2 inhibitor celecoxib demonstrated anti-fibrotic effects in the MASH model	PAD4 inhibitor GSK484 (preclinical) VDR agonist 16i (preclinical)	Prostaglandin E2 plays a vital physiological role in gastrointestinal mucosal protection and renal blood flow regulation; prolonged systemic inhibition may induce adverse reactions
Inflammation magnification layer (fibrosis amplification phase)	NLRP3 inflammasome	The NLRP3 inhibitor MCC950 alleviates fibrosis in MASH models	Dapansutrile (OLT1177) has demonstrated favourable safety in the treatment of gout	NLRP3 is coupled with tissue repair and antimicrobial immunity; prolonged or excessive suppression may increase infection risk and limit regenerative capacity
HSCs activation locking layer (fibrotic lock-in period)	TGF-β/TGFβR1 (ALK5)	The TGFβR1 inhibitor galunisertib has entered Phase II clinical trials in HCC	Galunisertib (LY2157299), the p38 MAPK inhibitor losmapimod	TGF-β possesses extensive immunomodulatory functions, suppressing potentially unpredictable immune-related adverse events

NETs: Neutrophil extracellular traps; HSCs: Hepatic stellate cells; TLR: Toll-like receptors; COX-2: Cyclooxygenase-2; PGE2: Prostaglandin E2; MASH: Metabolic dysfunction-associated steatohepatitis; PAD4: Peptidylarginine deiminase 4; NLRP3: NOD-like receptor family pyrin domain containing 3; TGF-β: Transforming growth factor-β; TGFβR1: Transforming growth factor-β receptor 1.

Cirrhosis and HCC: Integrating anti-fibrotic and immunotherapy response design

In the setting of MASH-associated HCC, persistent NETs formation is likewise associated with tumor initiation and progression, as well as with an imbalance in inflammatory and immune responses[87,88]. At the core of this process, the cGAS-STING and NLRP3 pathways together form a hub driven by DNA and danger signals, thereby amplifying inflammatory responses and regulating immune activity.

The role of NETs in the STING pathway in liver disease and tumors is highly dynamic. It can either help to activate the immune response against tumors, or, in the context of chronic inflammation, can lead to immune dysfunction and organ damage. Thus, blocking or modulating this pathway needs to take into account the therapeutic window and monitoring. The NLRP3 inflammasome activates not only immune cells to enhance inflammation but can also be activated in HSCs, thus directly driving fibrosis and linking inflammation to matrix remodeling. This offers a rationale for the parallel development of inflammation-barrier dysfunction[89]. Downstream, the TGFβR1 pathway is a critical component of matrix stabilization and immune exclusion, which is relevant for both antifibrotic activity as well as potential immune modulatory effects. Clinical trials of the TGFβR1 inhibitor galunisertib (LY2157299) are underway for HCC, highlighting the clinical possibility of this pathway. Moreover, TGFβ is a major pathway of immune exclusion and resistance to immunotherapy, which suggests a strong potential for combination with ICIs[85,90].

In immunotherapy, a logical combination treatment should involve barrier breach, matrix re-arrangement, release from immunosuppression and immune stimulation. Blocking TGFβR1 may help to disrupt the ECM and immune exclusion. Meanwhile, inhibition of PAD4/NETs or NLRP3-related pathways may help curtail adenosine axis activity and inflammasome-induced immunosuppression of myeloid cells. Overall, such strategies may work in combination with ICIs to convert the immune-exclusion environment to an immune-inflamed environment[90,91]. However, this framework requires rigorous clinical studies to validate its safety and efficacy. Patients with liver cirrhosis or HCC often exhibit impaired immunity and substantial heterogeneity in hepatic reserve capacity. Inhibition of pathways such as PAD4, STING, and NLRP3 may increase the risk of infection and compromise immune surveillance. Therefore, precision-based stratification, dynamic monitoring, and safety-oriented measures, such as short treatment courses or localized delivery, are essential to improve translational feasibility in real-world settings. Figure 3 illustrates a tiered model of liver pathology and the corresponding pharmacological intervention strategies.

Figure 3 Hierarchical framework of stage-specific therapeutic strategies for liver disease progression. Fibrosis progression is organized into bridging, amplification, and locking layers, targeting neutrophil extracellular traps-hepatic stellate cells crosstalk, NOD-like receptor family pyrin domain containing 3 signaling, and transforming growth factor-βR1-driven extracellular matrix/immune exclusion, respectively, enabling stage-specific disruption of inflammation-fibrosis coupling. HSC: Hepatic stellate cell; IL: Interleukin; NET: Neutrophils extracellular trap; TGFβR1: Transforming growth factor-β receptor 1; NLRP3: NOD-like receptor family pyrin domain containing 3; PAD4: Peptidylarginine deiminase 4; PGE2: Prostaglandin E₂; COX-2: Cyclooxygenase-2; HCC: Hepatocellular carcinoma; ECM: Extracellular matrix; ICI: Immune checkpoint inhibitor; MASH: Metabolic dysfunction-associated steatohepatitis; ALD: Alcohol-related liver disease.

Clinical monitoring and biomarker-guided implementation of precision remodeling

Translation of stage-specific neutrophils-HSCs remodeling into clinical practice requires accessible and validated biomarkers, as well as clear decision rules. For the assessment of neutrophils activation and NETs burden, plasma MPO-DNA complexes, Cit-H3, cell-free DNA, and NE have been proposed as quantitative indicators and have shown correlations with disease activity in MASH, alcohol-related liver disease, and HCC[14,15,39,92]. SERPINE1 and THBS1 have recently been validated as noninvasive serum biomarkers of NETs-driven early fibrosis in metabolic dysfunction-associated steatotic liver disease (MASLD) and may facilitate the operational definition of the Stage I-to-II transition[40]. NLR and its longitudinal stability predict clinical outcomes and ICI response in advanced HCC, supporting its use as an inexpensive tool for prognosis and treatment stratification[79]. Fibrosis stage and progression should be assessed using validated noninvasive methods, including fibrosis-4 index (FIB-4), ELF, vibration-controlled transient elastography (FibroScan), and magnetic resonance elastography, in accordance with current EASL guidelines[93,94]. Two recent practice-changing developments further support the translational relevance of this framework: The regulatory approval of resmetirom for MASH with significant fibrosis (F2-F3), based on phase 3 evidence showing that fibrosis regression is achievable in a stage-restricted population[95]; and the finding that liver-related events in MASLD increase with baseline fibrosis stage[96]. Together, these findings support the view that the timing of intervention and patient selection are critical determinants of therapeutic success. Therefore, a biomarker-guided algorithm is proposed: (1) Confirm fibrosis stage using FIB-4 plus transient elastography (ELF); (2) Quantify NETs burden using plasma MPO-DNA/Cit-H3 and SERPINE1/THBS1 to identify NETs-driven patients; (3) Match the intervention to disease stage; and (4) Monitor on-treatment NLR, MPO-DNA, and elastography findings to guide treatment titration or interruption. Prospective biomarker-stratified trials are needed to validate this algorithm.

CONTROVERSIES AND UNRESOLVED ISSUES

Although the interaction between neutrophils and HSCs in liver fibrosis has attracted considerable attention, many controversies and unresolved questions remain and warrant further investigation.

Controversy regarding the applicability of N1/N2 classification

The relevance of the N1/N2 classification, originally developed for the tumor microenvironment, to liver injury and repair remains under debate. First, the molecular markers that define the N1 and N2 phenotypes have not been adequately validated in liver injury models. Second, neutrophil phenotypic transitions may occur along a continuum rather than within a simple binary framework. Future studies should therefore focus on developing neutrophils classification systems tailored to liver disease models by using advanced approaches such as single-cell transcriptomics. In addition, single-cell multi-omics should be applied in clinically relevant liver injury cohorts to construct a continuum-based, organ- and stage-specific neutrophils atlas, define the molecular signatures of functional subsets such as CD177+, SiglecF+, PLAUR+, and PD-L1+ neutrophils in the liver, and link these signatures to clinical outcomes and therapeutic responses.

Species-specific considerations

The current use of murine models in research presents significant issues due to the differences between human and mouse neutrophils. For instance, human neutrophils live longer, display different granule protein profiles and respond differently to inflammatory cues than murine neutrophils. As such, results from murine models need to be carefully validated before being translated into humans. To overcome these differences, it is necessary to develop preclinical models such as organoids and humanized mice to translate findings from animal studies to humans.

Safety considerations for NETs inhibition

NETs are critical for the immune response to infection. So full inhibition may lead to an elevated risk of infection, especially in high-risk patients, such as cirrhosis patients. Therefore, selective inhibition or targeted delivery to desired tissues is important, but this presents significant challenges for the development of NETs-targeted drugs. There is a need for well-designed prospective clinical studies to test the safety and efficacy of stratified combination therapies.

Defining the therapeutic time window

While this article calls for a stratified approach to treatment, there is still a need for clear criteria to determine the therapeutic window at different stages. Reliable stratification according to clinical parameters, including fibrosis, inflammation and NETs, is important for the development of individualised treatment approaches.

CONCLUSION

Neutrophils and HSCs interact to form an important axis that defines the fate of the liver injury response, including reparative healing and fibrogenesis. This review explores the molecular ontology, temporal orchestration, disease stage and translational therapeutic implications of this axis. Neutrophils/HSCs interaction is a context- and stage-dependent “module” that evolves from a reparative role in acute injury to a self-amplifying, pro-fibrotic role in chronic liver disease, and to immunosuppressive and tumour-permissive role in cirrhosis and HCC. Hence, the translational focus is not to dampen neutrophil activity or inflammation in general, but to strategically re-program the crosstalk network to turn off the pathological feed forward loop and retain the function for repair. Future advances will rely on single-cell atlases capturing diversity of hepatic neutrophils and HSCs, validated and non-invasive biomarkers of NETs burden and disease stage, and biomarker-based trials that delineate the therapeutic window of each intervention.

ACKNOWLEDGEMENTS

I would like to sincerely thank teachers for their invaluable contributions to the development of this manuscript.