Accumulating evidence has established self-perpetuating inflammatory cascades as pivotal drivers of AE-IPF pathogenesis (Vázquez et al., 2011). Histopathological hallmarks of AE-IPF, including diffuse inflammatory infiltrates, intra-alveolar hemorrhage, proteinaceous edema, and hyaline membrane deposition, are consistently observed in clinical specimens (Tanaka et al., 2021) and experimental models (Chen et al., 2019). Studies investigating the molecular mechanisms revealed that the mRNA levels of nuclear factor-kappa B (NF-κB) were elevated, as were those of TNF-α, CXCL1, IL-1β, IL-6, and IL-8 (Chen et al., 2017; Yegen et al., 2022; Schupp et al., 2015; Papiris et al., 2018), indicative of progressive inflammatory milieu dysregulation. Acute inflammation represents an evolutionarily conserved defense mechanism involving the coordinated activity of the innate and adaptive immune responses to eliminate pathogens and initiate tissue repair (Hawiger and Zienkiewicz, 2019). However, in AE-IPF, this mechanism is dysregulated, primarily due to pattern recognition receptor (PRR) hyperactivation (Masumoto et al., 2021; Agarwal and Jindal, 2008). TLR engagement on AECs, endothelial cells, and macrophages initiates canonical NF-κB signaling via inhibitor of NF-κB (IκB) kinase-mediated phosphorylation (Le et al., 2023; Arora et al., 2019; McElroy et al., 2022). Activated NF-κB, acting as a dimer, is released from cytoplasmic inhibition and translocates to the nucleus, where it drives the transcription of inflammatory factors such as TNF-α, IL-1β, and IL-6 (Yu et al., 2020). Parallel inflammasome activation through the NOD-like receptor protein 3 (NLRP3) (Jäger et al., 2021) and absent in melanoma 2 (AIM2) (Cho et al., 2020) platforms in macrophages induces the caspase-1-dependent cleavage of pro-IL-1β and pro-IL-18, which, coupled with gasdermin D-mediated pyroptosis, amplifies the inflammatory response (Tseng et al., 2023). IL-1β, primarily secreted by monocyte-derived macrophages, initiates inflammatory cascades by activating CD4⁺ T cells and B cells, leading to the production of IL-6 and TNF-α (Dinarello, 2011). IL-6 rapidly responds to tissue damage, driving T-cell proliferation, antibody secretion by B cells, and inflammation amplification via STAT3 signaling (Tanaka et al., 2014). TNF-α exerts dual proinflammatory effects through TNF receptor-mediated, caspase-8-dependent apoptosis and endothelial vascular cell adhesion molecule-1 (VCAM-1)/ICAM-1 upregulation, thereby facilitating transendothelial neutrophil migration (Zelová and Hošek, 2013). Simultaneously, these cytokines collectively drive the polarization of alveolar macrophages toward a proinflammatory M1 phenotype (Cheng et al., 2021) and recruit circulating neutrophils, which accumulate in pulmonary parenchyma and undertake phagocytic clearance of pathogens and cellular debris (Piccinini and Midwood, 2010). Activated neutrophils and M1 macrophages further amplify the inflammatory response through the sustained secretion of TNF-α, IL-1β, IL-6, and CXCL8, thereby establishing chemotactic gradients that perpetuate monocyte-neutrophil infiltration (Meyer et al., 2021). This self-reinforcing inflammatory cascade propagates through feedforward mechanisms (Karki and Kanneganti, 2021; Marchioni et al., 2018), ultimately precipitating the acute exacerbation of stable IPF via pulmonary microenvironment destabilization (Savin et al., 2022) (Figure 3).

An imbalance between T-helper 17 (Th17) cells and regulatory T cells (Tregs) is a hallmark of immune dysregulation in pulmonary pathologies, including IPF (Thomas et al., 2023). These T-cell types originate from the divergent differentiation of naive CD4⁺ T cells. Th17 cells predominantly secrete IL-17A, IL-22, and IL-23, thus promoting proinflammatory responses, whereas Tregs counterbalance immune activation through IL-10 and TGF-β-mediated immunosuppression (Lee, 2018). Clinical studies have identified elevated Th17 cell frequencies and IL-17A concentrations in BALF during acute exacerbations (Wei et al., 2016), with murine models supporting these findings, as evidenced by the observed reduction in the number of inflammatory foci and diminished fibrosis in AE-PF mice with IL-17A deficiency relative to their wild-type counterparts (Chen et al., 2022). Mechanistically, IL-17A/IL-17RA axis activation leads to TGF-β release and neutrophil-mediated fibrogenesis, thereby functionally coupling inflammatory and fibrotic pathways in IPF pathogenesis (Huangfu et al., 2023; Lorè et al., 2016). In acute exacerbation, IL-17A promotes the chemotaxis of neutrophils and eosinophils (Chen et al., 2022) while suppressing Treg-mediated immunosuppression (Xu et al., 2023). In contrast, Treg depletion exacerbates inflammatory infiltration and fibroproliferation in experimental models (Moyé et al., 2020). The IL-23/Th17 regulatory axis plays an essential role in disease progression, as evidenced by the attenuation of airway inflammation and the decrease in the number of fibroblast foci in IL-23-deficient mice, alongside the efficacy of IL-23 neutralization in reducing Th17 populations and mitigating acute exacerbations (Gaffen et al., 2014). Collectively, these findings underscore the potential of Th17/Treg rebalancing as a therapeutic paradigm for AE-IPF (Figure 4).

Oxidative stress occurs when the equilibrium between ROS/reactive nitrogen species (RNS) production and the antioxidant capacity is disrupted, leading to cellular dysfunction and tissue damage (Jomova et al., 2023). Pulmonary neutrophils and macrophages constitute primary ROS/RNS sources through NADPH oxidase and myeloperoxidase activities (Morris et al., 2022). During acute exacerbations, respiratory bursts lead to the release of superoxide (O₂ ⁻), hydrogen peroxide (H₂O₂), nitric oxide (NO), and hypochlorous acid (HOCl) by phagocytic cells into alveolar spaces, which overwhelm endogenous defenses (Glennon-Alty et al., 2018). In addition to direct macromolecular damage, these oxidants promote TGF-β secretion from alveolar epithelium (Cui et al., 2011), triggering myofibroblast differentiation via α-smooth muscle actin (α-SMA) overexpression and excessive ECM deposition (Estornut et al., 2021). ROS accelerate cellular senescence by inducing telomere erosion, DNA damage, and proteotoxic stress, consequently promoting cell cycle arrest and the development of a senescence-associated secretory phenotype (SASP) (Korfei et al., 2020; Ogrodnik, 2021). In patients with IPF, senescent cells in the lung microenvironment displaying this ROS-induced feature secrete a wide range of proinflammatory and profibrotic mediators, such as IL-1β, IL-6, plasminogen activator inhibitor-1 (PAI-1), VCAM-1, matrix metalloproteinase-2 (MMP-2), and TGF-β (Álvarez et al., 2017), which collectively function as a signaling hub that perpetuates tissue injury responses and drives fibrotic progression (Schafer et al., 2017; Waters et al., 2018). While oxidative stress drives stable IPF progression through well-characterized mechanisms such as fibroblast activation and cellular senescence induction, its contribution to acute exacerbation remains ambiguous. Emerging evidence has implicated oxidative stress in AE-IPF. Alveolar macrophages in AE-IPF demonstrate marked upregulation of inducible nitric oxide synthase (iNOS) expression at both protein and mRNA levels (Miyamoto et al., 2022). iNOS catalyzes the oxidation of arginine, generating nitric oxide (NO) and peroxynitrite, the excessive production of which can induce protein nitrosation, thus amplifying oxidative damage (Cinelli et al., 2020). The antioxidant N-acetylcysteine (NAC) reduces mortality in AE-PF murine models (Hu et al., 2017) and inhibits fibroblast proliferation and activation (Yang et al., 2018), implicating oxidative stress as a central driver of AE-IPF pathobiology.

The ER maintains proteostasis through the coordination of protein biosynthesis, folding, and quality control. When biosynthetic demands exceed ER processing capacity, misfolded proteins accumulate, triggering ER stress—a state characterized by functional impairment and arrested protein synthesis (Oakes and Papa, 2015). This initiates the unfolded protein response (UPR) via three transmembrane sensors, namely, inositol-requiring enzyme one alpha (IRE1α), protein kinase R-like ER kinase (PERK), and activating transcription factor 6 (ATF6) (Read and Schröder, 2021; Hwang and Qi, 2018). The endoribonuclease domain of IRE1α is rapidly activated, leading to the splicing of X-box binding protein 1 (XBP1) mRNA, and, consequently, the generation of the transcriptionally active XBP1s isoform, which enhances the expression of genes encoding ER chaperones and components of the protein degradation machinery (Langlais et al., 2021; Grandjean et al., 2020). ATF6 undergoes regulated intramembrane proteolysis, following which its cytosolic fragment translocates to the nucleus and upregulates the expression of protein folding machinery-related genes (Lei et al., 2024). PERK demonstrates delayed but sustained activation during chronic stress, phosphorylating eukaryotic initiation factor 2 alpha (eIF2α), thereby reducing global protein synthesis, while simultaneously selectively inducing C/EBP homologous protein (CHOP) expression (Li et al., 2014). Persistent UPR failure increases CHOP-mediated transcription of Bcl-2-associated X protein (BAX), BCL2-antagonist/killer (BAK), and caspase-3, thus committing cells to apoptosis (Shore et al., 2011). In IPF, ER stress can arise through viral infections, oxidative damage, hypoxia, or impaired autophagy (Aghaei et al., 2020). ER stress elicits distinct consequences include the apoptosis or EMT of AECs, defective repair, and the differentiation of fibroblasts into α-SMA-positive myofibroblasts that drive collagen overproduction and fibrotic lesion expansion (Burman et al., 2018). Notably, viral-induced ER stress in IPF promotes the E3 ubiquitin ligase RNF5-mediated degradation of stimulator of interferon genes (STING), compromising antiviral defenses and predisposing to acute exacerbations (Qiu et al., 2017; Zhang et al., 2022). Concurrent CHOP/ATF6 overexpression activates NF-κB-dependent inflammation (Chen et al., 2019), establishing the disruption of ER proteostasis as a key mechanism linking chronic fibrosis progression and acute deterioration in IPF pathogenesis.

Apoptosis constitutes a pivotal mechanism in IPF pathogenesis, serving as the convergent endpoint of oxidative stress, inflammatory cascades, and ER stress, while also amplifying cytokine storms and fibrogenesis (Phan et al., 2021). This caspase-dependent programmed cell death occurs through both extrinsic and intrinsic pathways (Xu et al., 2019). The extrinsic pathway, activated through death receptors or TNF receptors, initiates caspase-8 activation via death domain adaptors and represents the principal apoptotic mechanism in inflammatory settings (Amaral and Bortoluci, 2020). Meanwhile, the intrinsic pathway involves mitochondrial permeability transition accompanied by cytochrome c and apoptotic protease-activating factor 1 (Apaf-1) release, culminating in caspase-9-mediated caspase-3 activation. The intrinsic pathway is predominantly activated by hypoxia (Suresh et al., 2023), oxidative stress (Liu et al., 2023), and ER stress (Wang et al., 2022). In patients with AE-IPF, the expression of cyclin A2 and α-defensin is upregulated, whereas advanced glycation end-product receptor levels are reduced, leading to marked alveolar epithelial injury and apoptosis (Konishi et al., 2009). Type II AECs, functioning as adult stem cells, activate repair mechanisms post-injury by proliferating and differentiating into Type I AECs (Wang et al., 2024). However, excessive Type II AEC apoptosis in IPF not only impairs epithelial regeneration but also promotes TGF-β secretion by macrophages via apoptotic bodies, thus accelerating fibrosis (Kim KK. et al., 2018). Failed repair induces EMT and fibroblast hyperactivation, which drives pathological ECM deposition (Chambers and Mercer, 2015; Chan and Liu, 2022). Notably, apoptotic susceptibility is cell-type-specific in IPF, with Type II AECs demonstrating enhanced vulnerability while fibroblasts and endothelial cells develop apoptosis resistance (Corteselli et al., 2022), creating a microenvironment favoring epithelial denudation and aberrant repair (Ptasinski et al., 2021). In infection-triggered exacerbations, bacterial proapoptotic peptides (e.g., corisin) promote mitochondrial ROS accumulation and membrane destabilization in AECs, suppressing the levels of the antiapoptotic factors baculoviral IAP repeat-containing (BIRC) 5, BIRC7, and B-cell lymphoma 2 (BCL2), while upregulating those of caspase-3/Apaf-1 (D'Alessandro-Gabazza et al., 2020). HIF-1α-mediated BNIP3/SERPINE1 upregulation is observed in non-infectious AE-PF models (Yegen et al., 2022), with both proteins promoting AECs apoptosis and post-injury fibrosis (Bhandary et al., 2012; Bernard et al., 2018). This suggests that hypoxia-driven apoptosis contributes to spontaneous exacerbations.

A foundational study by McMillan et al. (2008) established an AE-PF model through the sequential intratracheal administration of 28 mg/mL fluorescein isothiocyanate (FITC) followed by intranasal inoculation with 5 × 10⁴ plaque-forming units (PFUs) of murine γ-herpes virus-68 (γHV-68) at 14-day intervals. Although FITC-induced fibrosis displays fibrotic histopathological hallmarks, including chronic inflammation and extracellular matrix deposition, its pathogenesis is lymphocyte-independent (Moore et al., 2001), which substantially diverges from human IPF pathophysiology, leading to its progressive replacement by BLM-based models. Ashley et al. (2014) employed a similar γHV-68 challenge (5 × 10⁴ PFUs intranasal) for AE induction but diverged in primary fibrosis induction, using 0.025 U BLM, administered intratracheally, instead of FITC. Viral challenge on day 14 post-BLM elicited characteristic AE-IPF features within 7 days, corroborating McMillan’s findings. Notably, these exacerbation phenotypes were absent following inoculation with Pseudomonas aeruginosa or the influenza virus in BLM-primed mice.

Chen et al. (2019) established an AE-PF model through the sequential intratracheal administration of BLM (4 mg/kg, equivalent to 60 U/kg) followed by intranasal challenge with HSV-1 (5 × 10⁵ PFUs) on day 21. This HSV-1-based model, which employed human-pathogenic herpesvirus rather than the species-specific γHV-68, better approximates AE-IPF pathophysiology, although its 28-day survival rate of 21.4% raises translational concerns. Qiu et al. (2017) employed a comparable protocol with modified BLM dosing (5 U/kg intratracheal) and earlier HSV-1 administration (day 14 post-BLM). Although the use of this variant replicated Chen’s histopathological and functional characteristics, it resulted in similarly prohibitive mortality (∼25% survival), further underscoring the need for model refinement.

Among bacterial infection models, the Streptococcus pneumoniae (Spn) challenge paradigm demonstrates relative methodological maturity. Cho et al. (Cho et al., 2020) established AE-PF through the sequential oropharyngeal administration of BLM (0.01 mg/mouse, equivalent to 0.15 U/mouse), followed 14 days later by intranasal Spn serotype 3 instillation (1 × 10⁵ colony-forming units [CFUs]/mouse). In contrast, Knippenberg et al. (2015), Bormann et al. (2021), and Moyé et al. (2020) employed adenovirus-mediated TGF-β1 overexpression combined with Spn infection. Their experimental protocol involved the oropharyngeal administration of TGF-β1-encoding adenoviral vectors (1 × 10⁸ PFUs/mouse) to induce progressive fibrosis, followed by Spn serotype 19 challenge (1 × 10⁷ CFUs/mouse) via the same route during disease stabilization. TGF-β1, recognized as the principal fibrogenic mediator in IPF pathogenesis (Caja et al., 2018), induces spontaneous extracellular matrix remodeling in these adenoviral overexpression models. Histopathological evaluation revealed characteristic IPF features, including diffuse collagen deposition, honeycombing, and the development of fibroblastic foci (D'Alessandro-Gabazza et al., 2018). Unlike the self-limiting fibrosis observed in BLM-induced models post-treatment cessation, TGF-β1-overexpressing systems demonstrate progressive fibrotic deterioration that better mirrors the relentless disease progression observed in human IPF.

D’Alessandro-Gabazza et al. (D'Alessandro-Gabazza et al., 2020) induced acute exacerbation in TGF-β1 transgenic PF models through the intratracheal administration of Staphylococcus nepalensis (1 × 10⁸ CFUs/mouse). This bacterial challenge occasioned pronounced macrophage/neutrophil infiltration, exacerbated collagen deposition, and enhanced alveolar epithelial cell apoptosis, a central pathogenic mechanism in IPF progression (Sauler et al., 2019). Mechanistic studies revealed that S. nepalensis secretes the proapoptotic peptide corisin under hypertonic microenvironmental conditions, directly driving pan-alveolar epithelial apoptosis. Building on these findings, the same research group (D'Alessandro-Gabazza et al., 2022) developed an optimized AE-PF model combining BLM exposure with corisin challenge.

Chen et al. (2022) established a novel Gram-negative bacteria-induced AE-PF model through the sequential intranasal administration of BLM (30 μg/mouse, equivalent to 0.45 U/mouse), followed by H. influenzae administration (1 × 10⁷ CFUs/mouse) at 7-day intervals. This is the sole reported Gram-negative bacterial model of AE-PF. This study systematically compared BLM and Haemophilus influenzae dosing regimens (BLM: 15, 30, and 60 μg/mouse, equivalent to 0.225, 0.45, and 0.9 U/mouse; H. influenzae: 1 × 10⁶–1 × 10⁸ CFUs/mouse), and reported that the 30 μg BLM dose optimally induced pulmonary fibrosis without elevating mortality, while a bacterial load of 1 × 10⁷ CFUs induced significant weight loss with sustained pulmonary colonization. Mechanistically, chronic IL-6/IL-8-mediated neutrophilic airway inflammation drives immunopathology and functional decline in chronic lung diseases (Saliu et al., 2021). Clinically, 89% of bacterial isolates from the airways of patients with AE-IPF are Gram-negative (Weng et al., 2019). Compared with existing viral and Gram-positive bacterial models, this Gram-negative model demonstrated superior survival rates while maintaining pathophysiological fidelity.

Contemporary histopathological analyses of lung biopsies and autopsies have confirmed that DAD is the cardinal feature of AE-IPF (Dotan et al., 2020; Kawaguchi et al., 2022; Kishaba et al., 2020). LPS, a potent experimental inflammogen, reliably induces DAD pathology (Domscheit et al., 2020) and acute exacerbation of lung disease (Mohamed et al., 2019), providing a translational platform bridging preclinical models and human pathophysiology.

Kimura et al. (2015) and Senoo et al. (2021) established a model employing sequential BLM-induced fibrosis (1 mg/kg equivalent to 15 U/kg; intratracheal) followed by LPS challenge (0.5 mg/kg) on day 7, eliciting hallmark clinical features, including acute hypoxemia (PaO₂ <60 mmHg within 24 h) and diffuse ground-glass opacities on CT. However, single-dose LPS models exhibit transient inflammation with spontaneous resolution within 7 days. To circumvent this limitation, Jia et al. (2023) implemented fractionated LPS dosing (1 mg/kg on days 5, 7, and 9 post-BLM), achieving sustained exacerbation concomitant with hepatic inflammation and collagen deposition, which mirrored the subclinical liver fibrosis observed in >30% of patients with IPF (Cocconcelli et al., 2021).

Kurniawan et al. (2023) established an AE-PF model that combined intratracheal BLM administration (5 mg/kg, equivalent to 75 U/kg) with intraperitoneal LPS challenge (7.5 mg/kg), and identified distinct pathophysiological phases—a neutrophilic BALF peak on day 1 (inflammatory phase), maximal fibrosis expansion on day 7 (proliferative phase), and partial resolution by day 14 (remodeling phase)—thus establishing a critical therapeutic window within 7 days post-exacerbation.

Miyamoto et al. (2022) demonstrated that ultralow-dose LPS (0.05–0.15 mg/kg) triggers AE-PF in BLM-primed Wistar rats (3 mg/kg [45 U/kg] daily for 7 consecutive days), replicating clinical exacerbation patterns after minor insults while preserving survival rates (83%–100% on day 7). This paradigm employs LPS doses 10-fold lower than conventional acute lung injury models, thus enhancing clinical relevance while maintaining experimental feasibility.

Chen et al. (2017), Wei et al. (2016) developed AE-PF models through sequential intratracheal BLM challenge. This dual-administration protocol demonstrates operational simplicity and high reproducibility while effectively mimicking acute lung injury dynamics during exacerbations, although limited by spontaneous fibrotic regression. In contrast, Yegen et al. (2022) implemented a chronic fibrotic priming strategy with biweekly intratracheal BLM (0.8 U/g twice weekly for 4 weeks) followed by double-dose exacerbation triggers (1.6 U/g twice weekly for 2 weeks). This extended protocol better recapitulates the progressive fibrotic trajectory of IPF, circumventing the issue of self-limiting fibrosis observed in acute models. However, the prolonged experimental timeline (6-week duration) and technical complexity present practical implementation challenges. Notably, the applied dose of 0.8 U/g (equivalent to 800 U/kg) substantially exceeds reported lethal dose ranges for BLM in murine models (typically 150–200 U/kg), raising critical concerns regarding experimental mortality and translational relevance.

Hu et al. (2017) investigated acute exacerbation mechanisms using a model combining BLM and exposure to straw combustion-derived particulate matter with an aerodynamic diameter ≤5 µm (PM2.5). Following 7-day intratracheal BLM priming (0.2 mg/mouse, equivalent to 3 U/mouse), mice underwent twice-daily PM2.5 inhalation (30 min/session), eliciting progressive macrophage infiltration with characteristic black particulate deposition and time-dependent collagen accumulation. This experimental paradigm aligns with clinical evidence linking air pollution exposure to accelerated IPF progression, including functional decline and radiological deterioration (Ko and Kyung, 2022; Mariscal-Aguilar et al., 2023). PM2.5, which predominantly comprises secondary sulfates, nitrates, and biomass combustion byproducts (Yao et al., 2016), demonstrates dose-dependent associations with fibrotic interstitial lung disease mortality and progression (Goobie et al., 2022), highlighting its emerging role in AE-IPF pathogenesis.

Kim MS. et al. (2018) developed a polyhexamethylene guanidine (PHMG)-cadmium co-exposure model, leveraging cadmium’s established pulmonary toxicity from cigarette smoke-mediated deposition and its mechanistic links to COPD/IPF overlap syndromes (Ganguly et al., 2018). Complementary studies by Yang et al. (2018) demonstrated that nickel chloride (NiCl₂, 5 mg/kg) can exacerbate BLM-induced fibrosis (2.5 mg/kg, equivalent to 37.5 U/kg) when co-administered intratracheally. These findings emphasize the underappreciated impact of occupational metallurgical exposures, which not only elevate IPF incidence among industrial workers but also exhibit dose-responsive mortality correlations in established IPF cohorts (Gandhi et al., 2023; Trethewey and Walters, 2018).