The role of pyroptosis in ovarian cancer.

Pyroptosis is a distinct form of programmed cell death that involves key molecular components including inflammasomes, caspases, and gasdermins (GSDMs). It is characterized by the formation of pores in the plasma membrane, resulting in cellular swelling and subsequent osmotic lysis. The core molecular machinery of pyroptosis comprises three essential elements: inflammasomes, caspases, and gasdermins (GSDMs). In the subsequent sections, we will systematically elucidate the signaling pathways underlying pyroptosis from two main perspectives: (1) inflammasome-mediated activation cascades and (2) gasdermin-dependent pore formation. Inflammasomes are multiprotein complexes that are typically named after their sensor protein. They comprise two receptor families: the nucleotide-binding and oligomerization domain (NOD)-like receptor (NLR) family, which includes NLRP1, NLRP2, NLRP3, NLRP6, NLRC4, and NLRP12, and the PYHIN family, which encompasses absent in melanoma 2 (AIM2) and Pyrin. The apoptosis-associated speck-like protein containing a caspase recruitment domain (ASC) functions as an adaptor, bridging the inflammasome complex to procaspase precursors. ASC contains a pyrin domain (PYD) at its N-terminus and a caspase activation and recruitment domain (CARD) at its C-terminus. Consequently, ASC mediates homotypic interactions between PYD and CARD domains of NLRs or PYHIN proteins and procaspase-1 [ 20 ]. NLR proteins are composed of three domains: an N-terminal signaling domain that binds ligands, a central NACHT domain that mediates protein oligomerization, and a C-terminal leucine-rich repeat (LRR) domain responsible for signal sensing [ 21 ]. The N-terminal signaling domain of NLRP receptors is PYD, while that of NLRC receptors is CARD. ASC can interact with NLRs via PYD-PYD or CARD-CARD domain interactions [ 22 ]. All six NLR inflammasome receptors, as well as AIM2 and Pyrin, are capable of recruiting procaspase-1 in an ASC-dependent manner, whereas NLRP1 and NLRC4 can also recruit procaspase-1 through CARD-CARD interactions independent of ASC [ 21 ]. Upon detection of specific microbial or endogenous danger signals, the inflammasome cleaves procaspase-1 into its active form, caspase-1, which plays a pivotal role in innate immunity [ 5 ]. The human NLRP1 gene encodes a unique inflammasome sensor characterized by the presence of NACHT, LRR, PYD, function-to-find domain (FIIND), and CARD domains. In contrast, mice possess three NLRP1 paralogs (NLRP1a–c), all of which lack the PYD domain, indicating a significant structural divergence between species. Vance and Bachovchin’s team was the first to elucidate the precise mechanism of NLRP1b activation via ubiquitin-proteasome-mediated degradation. Following proteasomal processing, NLRP1b undergoes autoproteolysis at the F983-S984 site within the FIIND domain, resulting in the release of a C-terminal fragment (amino acids 984–1233) containing the UPA subdomain and CARD. Unlike the full-length protein, this truncated fragment evades proteasomal degradation, subsequently self-oligomerizing and recruiting caspase-1 through CARD-CARD interactions to initiate inflammasome assembly. In humans, NLRP1 is predominantly expressed in epithelial tissues, especially the skin and respiratory tract [ 23 ]. NLRP1 can be activated by anthrax lethal toxin, Toxoplasma gondii, Shigella flexneri, the small molecule Dpp8/9 inhibitor VbP, and UVB radiation [ 24 ]. Notably, these triggers do not activate rodent NLRP1, underscoring species-specific regulatory mechanisms. The NLRP3 inflammasome can be activated by diverse stimuli, including: pathogen-derived ligands: microbial cell wall components, nucleic acids, and pore-forming toxins; environmental crystalline agents: silica, asbestos, and alum; endogenous danger signals: ATP, serum amyloid A, and urate crystals [ 25 ]. Classical activation of the NLRP3 inflammasome involves ionic fluxes (including K⁺ efflux, Ca²⁺ mobilization, and Cl⁻ efflux), mitochondrial dysfunction and reactive oxygen species (ROS) production, phagolysosomal rupture, trans-Golgi network disassembly, and endoplasmic reticulum stress [ 25 ]. Non-canonical activation of NLRP3 occurs during Gram-negative bacterial infection, wherein human caspase-4/5 or murine caspase-11 directly bind cytoplasmic LPS. Recent studies have identified Nur77 (an NR4A orphan nuclear receptor) as a novel LPS sensor. Nur77 binds both LPS and mitochondrial DNA (mtDNA) released through GSDMD pores, thereby facilitating NLRP3 inflammasome assembly [ 26 , 27 ]. NLRP3 activation proceeds in two phases. Priming involves NF-κB-mediated transcriptional upregulation of NLRP3 via FADD/caspase-8 [ 28 ]. Additionally, ERK/c-Jun/AP-1 signaling enhances NLRP3 expression during influenza virus infection [ 29 ], while Fyn kinase promotes NLRP3 transcription via the PKCδ/NF-κB pathway in response to α-synuclein [ 30 ]. During activation, NLRP3 senses PAMPs or DAMPs (e.g., fungi, bacteria, viruses). NEK7, a cell cycle kinase, binds to the LRR domain of NLRP3 independently of its kinase activity. Potassium efflux triggers JNK-mediated phosphorylation of NEK7 (at Thr190/191), thereby strengthening the NEK7-NLRP3 interaction and promoting inflammasome activation [ 31 ]. The NLRC4 inflammasome is composed of three functional domains: LRR, CARD, and NACHT. While bacterial flagellin and components of type III secretion systems can activate the NLRC4 inflammasome [ 32 ], NLRC4 does not directly interact with these activators. Instead, members of the neuronal apoptosis inhibitory protein (NAIP) family act as ligand-sensing receptors that trigger NLRC4 inflammasome assembly. In mice, NAIP1 and NAIP2 recognize bacterial needle and inner rod proteins, respectively, whereas NAIP5 and NAIP6 detect flagellin [ 33 ]. Notably, the human genome encodes only one functional NAIP protein, which is capable of sensing both flagellin and needle proteins [ 34 ]. AIM2, a member of the AIM2-like receptor (ALR) family, is characterized by the presence of a pyrin domain (PYD) and a C-terminal hematopoietic interferon-inducible nuclear protein with a 200-amino-acid repeat (HIN200) domain. The ALR family is exclusively expressed in mammalian species. Among ALR family members, AIM2 is the only known protein capable of forming an inflammasome complex. The PYD domain of AIM2 interacts with ASC, while its C-terminal oligonucleotide/oligosaccharide-binding (OB) fold domain recognizes double-stranded DNA (dsDNA) [ 35 ]. Activation of AIM2 in response to bacterial stimuli (e.g., Francisella spp.) requires type I interferon (IFN) signaling [ 36 ]. The HIN200 domain binds to dsDNA (including viral DNA) through sequence-independent electrostatic interactions, with a minimum required length of approximately 80 base pairs [ 37 ]. In the resting state, the PYD domain forms an intramolecular complex with the HIN domain. Upon dsDNA binding to the HIN domain, AIM2 undergoes conformational changes that permit its PYD to interact with the PYD of ASC. This interaction facilitates the recruitment of caspase-1 through CARD-CARD interactions between ASC and caspase-1, ultimately leading to the activation of caspase-1 and subsequent maturation of IL-1β and IL-18. Pyrin, encoded by the MEFV gene on chromosome 16, forms an inflammasome complex in response to Rho-modifying toxins produced by various bacterial pathogens. These pathogens include Clostridioides difficile (TcdB), Vibrio parahaemolyticus (VopS), Histophilus somni (IbpA), Clostridium botulinum (C3), and Burkholderia cenocepacia [ 38 ]. These toxins induce covalent modifications within the Switch I region of Rho GTPases, resulting in RhoA inactivation. Notably, while Pyrin senses bacteria-mediated RhoA inhibition, it does not directly interact with RhoA. Activation of the Pyrin inflammasome requires two sequential steps: (1) dephosphorylation of Pyrin and subsequent dissociation from its chaperone protein 14-3-3, and (2) microtubule-dependent conformational changes that expose the Pyrin PYD domain. This structural rearrangement enables PYD-mediated interaction with the adaptor protein ASC, thereby facilitating inflammasome assembly [ 39 ]. The canonical pathway recognizes pathogen-associated molecular patterns (PAMPs) present on toxins, viruses, and bacteria, as well as damage-associated molecular patterns (DAMPs) that are generated following tissue or cellular injury [ 8 , 40 ]. Studies have also found that homeostasis-altering molecular processes (HAMPs) can be recognized [ 41 ]. Cells utilize pattern recognition receptors (PRRs) to detect these signals. PRRs can be classified into two types: those located on the plasma and endosomal membranes, which recognize DAMPs and PAMPs in the extracellular environment—including Toll-like receptors (TLRs) and C-type lectin receptors (CLRs)—and those situated within the intracellular environment, represented by NLRs, AIM2-like receptors (ALRs), and RIG-I-like receptors (RLRs) [ 42 ]. The canonical pathway involves five major types of inflammasomes: NLRP3, NLRP1, NLRC4, AIM2, and Pyrin. NLRs act as sensors for inflammasomes, recruiting procaspase-1 via ASC-dependent or self-assembly mechanisms upon recognition of their respective ligands. Procaspase-1 is cleaved to its active form, caspase-1, which in turn cleaves GSDMD [ 43 ]. The N-terminal fragment of GSDMD inserts into the plasma membrane, forming pores that permit water influx, resulting in cellular swelling and eventual rupture. Active IL-1β and IL-18, processed by caspase-1, are also released into the extracellular space through these pores, thereby exacerbating the inflammatory response [ 7 ]. The non-canonical pathway differs from the canonical pathway in that it does not require inflammasomes as intermediaries. In the non-canonical pathway, lipopolysaccharide (LPS) from Gram-negative bacteria directly binds to human caspase-4 and − 5 (or caspase-11 in mice), which subsequently cleave GSDMD, leading to pore formation and cell lysis [ 5 ]. Notably, caspase-4, −5, and − 11 cannot directly process IL-1β and IL-18, representing a key distinction between the non-canonical and canonical pathways. Upon LPS stimulation and increased K⁺ efflux, caspase-4, −5, and − 11 can activate NLRP3, which subsequently activates caspase-1 to cleave GSDMD and release active IL-1β and IL-18 [ 18 ]. The canonical and non-canonical pathways of pyroptosis are shown in Fig. 1 . Fig. 1 The canonical (left) and non-canonical (right) pathways of pyroptosis The canonical (left) and non-canonical (right) pathways of pyroptosis In the canonical pathway (left panel), PAMPs or DAMPs activate inflammasome sensors (NLRP3, NLRP1, NLRC4, AIM2, or Pyrin), leading to inflammasome assembly. Subsequently, the inflammasome activates caspase-1, which cleaves GSDMD to generate its pore-forming N-terminal fragment (GSDMD-NT). GSDMD-NT forms pores in the plasma membrane, facilitating the release of mature IL-1β and IL-18, which are processed by caspase-1 from their pro-forms (pro-IL-1β and pro-IL-18). In the non-canonical pathway (right panel), LPS directly binds and activates caspase-4, −5, or −11. These caspases cleave GSDMD, triggering pore formation. Additionally, LPS-induced K⁺ efflux promotes NLRP3 inflammasome activation, resulting in caspase-1-mediated GSDMD cleavage and subsequent IL-1β/IL-18 maturation and release. Created with BioGDP.com [ 44 ]. The gasdermin (GSDM) family comprises key effectors of pyroptosis and belongs to a class of pore-forming proteins. In humans, the GSDM family includes six homologous genes: gasdermin A, B, C, D, E (also known as DFNA), and DFNB (also known as PJVK) [ 45 ]. With the exception of DFNB59, GSDMs are composed of an N-terminal (GSDM-NT), a linker domain, and a C-terminal (GSDM-CT) domain, whereas PJVK lacks the linker domain. The C-terminal domain adopts a globular conformation that completely covers the N-terminal domain [ 46 ]. The minimal functional unit required to induce pyroptosis is the GSDM-NT. The GSDM-NT exhibits a conserved structural architecture comprising three modules: a β1-β2 loop region (“wrist” domain), a globular “palm” subdomain containing the α1-helical thumb that forms the pore apex, and four extended β-strands (15–22 residues) forming antiparallel β-hairpins that insert into the lipid bilayer as “fingers”. In the autoinhibited state, GSDMs maintain a compact conformation in which these structural elements are tightly packed. Upon activation, disordered regions undergo conformational changes to form membrane-inserting β-strands, whereas the globular palm subdomain remains relatively rigid. Comparative structural analyses reveal variations in the orientation angle between palm and finger domains among GSDM family members. Notably, bacterial GSDMs display a closer membrane proximity of the palm subdomain, which likely enhances lipid interactions and pore stability. This conserved yet adaptable architecture enables GSDM-NT to undergo conformational changes required for pore formation while maintaining structural integrity during membrane insertion [ 47 ]. GSDM oligomerization and pore formation can occur via two distinct mechanisms. The first mechanism involves initial insertion of GSDM monomers or small oligomers into the membrane, forming arc-shaped or slit-like structures that subsequently coalesce into larger annular pores. Alternatively, GSDM proteins may first assemble into complete ring-shaped oligomers in the cytosol, which then weakly associate with membranes prior to pore formation. While the first mechanism is well characterized, the process of membrane insertion and the underlying molecular details of the second pathway remain unclear [ 48 ]. Studies have revealed that mammalian GSDM pores typically comprise 24–34 symmetrically arranged protomers, forming functional pores with inner diameters ranging from 15 to 21.5 nm [ 47 ]. These structural features enable GSDM pores to accommodate the passage of mature cytokines while maintaining membrane integrity during pyroptosis. These pores serve two main functions in pyroptosis: (1) enabling the release of inflammatory cytokines such as IL-1β and IL-18, and (2) allowing water influx, which leads to cellular osmotic imbalance and subsequent rupture [ 7 ]. Figure 2 illustrates the mechanisms of GSDM-mediated pyroptosis. Fig. 2 Mechanisms of GSDM-mediated pyroptosis Mechanisms of GSDM-mediated pyroptosis NK cells and cytotoxic T lymphocytes (CTLs) deliver granzyme A (GZMA) and granzyme B (GZMB) into target cells through perforin-dependent mechanisms. GZMA directly cleaves GSDMB to generate functional pore-forming fragments, while GZMB can either directly cleave GSDME or activate the caspase-3/GSDME pathway. Activated caspase-3, stimulated by TNF-α or certain chemotherapeutic agents, promotes pyroptosis through the caspase-1/GSDMD pathway or direct cleavage of GSDME when abundantly expressed, resulting in plasma membrane pore formation. Upon UVC irradiation, post-translational modifications of GSDME relieve its autoinhibitory conformation, leading to pyroptosis. Yersinia infection induces pyroptosis by inhibiting TAK1, which leads to RIPK1 activation and subsequent caspase-8-mediated cleavage of GSDMD. In another pathway, α-ketoglutarate (α-KG) facilitates the recruitment of GSDMC to internalized death receptor 6 (DR6), which undergoes ROS-dependent oxidation in response to α-KG, ultimately activating GSDMC. Group A Streptococcus (GAS) promotes pyroptosis through secretion of the cysteine protease SpeB, which directly cleaves GSDMA. Under conditions of amino acid starvation and severe ATP depletion, the nutrient-sensing kinase ULK1 phosphorylates GSDMA at serine 353, relieving the autoinhibitory effect of the C-terminal domain and enabling GSDMA oligomerization and pore formation. Created with BioGDP.com [ 44 ]. The Gsdma gene was initially identified as a pathogenic and susceptibility gene for skin inflammation and alopecia in mutant mouse lines, encompassing three spontaneous and six mutagen-induced Gsdma3 variants [ 49 ]. These gain-of-function mutations cause either premature C-terminal truncation of GSDMA3 or disruption of the autoinhibitory interaction between its C-terminal and N-terminal domains, resulting in constitutive activation of GSDMA3 and subsequent initiation of pyroptosis [ 50 ]. During invasive infection by Streptococcus pyogenes (Group A Streptococcus, GAS), the cysteine protease exotoxin SpeB cleaves GSDMA after Gln246, generating an active N-terminal fragment (GSDMA-NT, amino acids 1–246) that binds to and disrupts acidic lipid membranes [ 51 ]. This proteolytic activation exemplifies a pathogen-mediated pyroptosis pathway. Under conditions of amino acid starvation and severe ATP depletion, cellular stress activates the nutrient-responsive kinase ULK1. ULK1-mediated phosphorylation of GSDMA at Ser353 relieves C-terminal autoinhibition, thereby enabling GSDMA oligomerization and the execution of pyroptosis [ 52 ]. This mechanism establishes GSDMA as a metabolic stress sensor in pyroptotic cell death. GSDMB exists in at least five distinct variants (Q8TAX9-1 to Q8TAX9-4 and Q8TAX9-6) generated through alternative splicing of exons 6 and 7 [ 53 ]. These isoforms exhibit differential functional properties within pyroptotic pathways. During cytotoxic lymphocyte-mediated target cell killing, granzyme A (GZMA) specifically cleaves GSDMB at two primary sites (Lys229 and Lys244), generating 30 kDa and 16 kDa fragments. Edman sequencing and site-directed mutagenesis have confirmed Lys244 as the predominant cleavage site, with this proteolytic event triggering target cell pyroptosis [ 54 ]. Comparative studies have shown that cells expressing the longest GSDMB isoform (Q8TAX9-4) display significantly increased susceptibility to cytotoxic T lymphocyte (CTL)- and natural killer (NK) cell-mediated pyroptosis compared to cells expressing shorter isoforms or lacking GSDMB expression [ 55 ]. This isoform-dependent vulnerability suggests a critical role for alternative splicing in regulating pyroptotic sensitivity. GSDMC was initially identified as an executor of pyroptosis activated by caspase-8 in cancer cells. Recent studies have revealed an additional metabolic activation pathway, whereby the TCA cycle intermediate α-ketoglutarate (α-KG) triggers GSDMC-dependent pyroptosis via death receptor 6 (DR6) signaling. α-KG-mediated activation occurs via a sequential mechanism: (1) α-KG induces reactive oxygen species (ROS) generation, which oxidizes DR6; (2) oxidized DR6 recruits and activates GSDMC on internalized membranes; (3) activated caspase-8 cleaves GSDMC at Asp365 (TNF-induced) or Asp240 (α-KG-induced) to generate the pyroptosis-inducing N-terminal fragment (GSDMC-NT). This dual activation pathway demonstrates how GSDMC integrates both immunological (TNF) and metabolic (α-KG) signals to execute pyroptosis in tumor cells [ 56 , 57 ]. During infection of macrophages by Yersinia, the bacterial effector protein YopJ inhibits transforming growth factor β-activated kinase 1 (TAK1/MAP3K7) activity, thereby promoting receptor-interacting serine/threonine-protein kinase 1 (RIPK1)-dependent, caspase-8-mediated cleavage of GSDMD and subsequent pyroptosis [ 58 ]. S-palmitoylation of full-length GSDMD, catalyzed by palmitoyl acyltransferases, can induce pore-forming activity in vitro [ 59 ]. Caspase-8 facilitates the formation of Fas-associated death domain protein (FADD)-RIPK1-caspase-8 complexes that cleave GSDMD following Yersinia or LPS stimulation [ 60 ]. During rotavirus infection of intestinal epithelial cells, GSDMD is cleaved by caspase-3/7 at Asp88, generating an N-terminal fragment (~ 13 kDa) and a C-terminal fragment (~ 42 kDa) [ 61 ]. Recent studies have revealed distinct mechanisms for GSDME activation. Under UVC irradiation, full-length GSDME undergoes poly(ADP-ribosyl)ation (PARylation), which enables its activation through removal of the autoinhibitory C-terminal domain without proteolytic cleavage [ 62 ]. During cytotoxic lymphocyte (CTL)-mediated killing, granzyme B (GZMB) and caspase-3 directly cleave GSDME at Asp270 to generate an N-terminal fragment (GSDME-NT) that exhibits membrane pore-forming activity [ 63 ].

In recent years, the molecular pathways underlying pyroptosis have been progressively elucidated. In addition to the classical and non-classical pathways, caspase-3 and caspase-8—traditionally regarded as apoptotic caspases—have also been implicated in the activation of pyroptosis, thereby establishing intricate links between pyroptosis and other forms of regulated cell death (RCD). Autophagy negatively regulates pyroptosis by promoting NLRP3 degradation, whereas ferroptosis has been reported to facilitate pyroptosis. Activation of autophagy attenuates NLRP3 overactivation. Iron (Fe²⁺) and reactive oxygen species (ROS)-inducing agents can trigger pyroptosis via the Tom20–Bax–caspase–GSDME pathway, thereby modulating tumor progression [ 121 ]. These findings add a new dimension to our understanding of the complex mechanisms governing pyroptosis. Both in the context of direct tumor treatment and in modulating anti-tumor immunity, therapies promoting pyroptosis represent a promising strategy for ovarian cancer management. Nevertheless, several challenges remain. Clinical studies evaluating small molecule inhibitors, especially those promoting pyroptosis, are currently limited. The development of targeted agents remains in its infancy. In addition, drugs such as paclitaxel or cisplatin, despite their ability to induce pyroptosis, are often limited by the development of drug resistance. Novel small molecule inhibitors, including α-NETA, BI2536, and simvastatin, have demonstrated encouraging results in vitro and in animal models. However, further in-depth studies are required to clarify the molecular mechanisms and advance the clinical application of these agents.

Although the role of pyroptosis in OC progression still requires further study, it is generally believed that pyroptosis can trigger interactions between non-specific immunity and specific immunity, regulating the tumor microenvironment, inducing immune-stimulating responses, and activating anti-tumor immunity. It is regarded as a promising new strategy for cancer treatment. Table  2 presents the mechanisms through which compounds can induce pyroptosis in ovarian cancer cells. Table 2 Compounds inducing pyroptosis signal pathways in OC Classification Compounds Molecular Pathway Reference Chemotherapy drugs Paclitaxel Caspase-3/GSDME [ 91 ] Cisplatin MEG3/NLRP3/Caspase-1/GSDMD [ 92 ] Reagents α-NETA Caspase-4/GSDMD [ 93 ] Besharotin Caspase-4/GSDME [ 94 ] CBL0137 ROS/BAX/Cyt c/caspase-9/caspase-3/GSDME [ 95 ] BI2536 Caspase-3/GSDME [ 96 ] Olaparib/Niraparib NF-κB/TNF/caspase 8/GSDMD/E [ 97 ] Natural products Osthole GSDME [ 98 ] Nobiletin ROS/GSDMD/GSDME [ 99 ] Gambogic acid ROS/p53/mitochondrial/caspase-3/GSDME [ 100 ] P4 IL-6/ROS/NLRP3/Caspase-1/GSDMD [ 101 ] CA CASP4/TXNIP/NLRP3/GSDMD [ 102 ] DHA Caspase-1 [ 103 ] Target CRLF1 ASK1/JNK/Caspase-3/GSDME [ 104 ] RAC1 PAK4/MAPK/Caspase-1/GSDMD [ 105 ] PFKFB3 Caspase-3/9/NLRP3 [ 106 ] IRF3 RIPK1/FADD/caspase-8 [ 107 ] FTO NLRP3/Caspase-1/GSDMD [ 108 ] Compounds inducing pyroptosis signal pathways in OC OC is highly sensitive to platinum-based drugs, exhibiting an overall response rate as high as 75.7% [ 109 ]. Paclitaxel combined with platinum-based chemotherapy constitutes the standard first-line regimen in clinical practice [ 14 ]. Cisplatin and paclitaxel exert their anti-tumor effects by regulating apoptosis and inducing pyroptosis, with this dual mechanism potentially enhancing their efficacy against OC [ 91 ]. Recent studies have demonstrated that, during paclitaxel-induced mitotic arrest, cGAS-dependent phosphorylation of IRF3 at Ser386 and Ser396 facilitates assembly of the RIPK1/FADD/caspase-8 complex, leading to caspase-8 activation and subsequent GSDME-mediated pyroptosis [ 107 ]. Cisplatin upregulates the lncRNA maternal expressed gene 3 (MEG3), which activates the NLRP3/caspase-1/GSDMD pyroptosis pathway, thereby inhibiting tumor growth and metastasis [ 92 ]. Compared with adjacent normal tissues, OC tissues exhibit significantly downregulated expression of fat mass and obesity-associated protein (FTO). Functional studies have shown that FTO overexpression enhances cisplatin (DDP) sensitivity and upregulates NLRP3 expression, thereby suppressing tumorigenesis in xenograft mouse models. Mechanistically, FTO promotes NLRP3/caspase-1/GSDMD-dependent pyroptosis, which contributes to increased DDP chemosensitivity and inhibition of OC progression [ 108 ]. With the emergence of OC resistance, numerous studies have focused on identifying novel agents or pathways to overcome chemoresistance. Qiao et al. reported that α-NETA significantly increases the expression of pyroptosis-related molecules, including caspase-4 and GSDMD, and its administration reduces tumor size in mouse models of epithelial ovarian cancer (EOC), demonstrating potential as a novel chemotherapeutic agent [ 93 ]. All-trans retinoic acid (ATRA) is a well-established retinoid with in vitro efficacy against OC; however, many OC cell lines are resistant to ATRA [ 110 ]. Studies have found that RXR is highly expressed in EOC, and Besharotin—a selective retinoid X receptor (RXR) agonist—demonstrates greater safety and potency at lower concentrations compared to ATRA. It induces both apoptosis and pyroptosis via the caspase-4/GSDME pathway [ 94 ]. High expression of cytokine receptor-like factor 1 (CRLF1) in OC is associated with poor survival and chemoresistance. CRLF1 inhibits pyroptosis via the ASK1/JNK/caspase-3/GSDME pathway and promotes AKT activation, which in turn suppresses the ASK1/JNK/caspase-3/GSDME cascade. Owing to tumor heterogeneity, genetic alterations, limited specificity, and substantial side effects, direct targeting of AKT in clinical practice remains challenging. Therefore, targeting CRLF1 may promote pyroptosis and enhance chemosensitivity [ 104 ]. Aberrant activation or inhibition of the ras-related C3 botulinum toxin substrate 1 (RAC1) signaling pathway plays a critical role in various diseases. RAC1 is highly expressed in OC and inhibits the ubiquitination and degradation of PAK4. Studies indicate that RAC1 suppresses classical pyroptosis mediated by caspase-1/GSDMD via the PAK4/MAPK pathway, thereby promoting paclitaxel resistance in OC. Consequently, combining RAC1 inhibitors with paclitaxel may reverse paclitaxel resistance [ 105 ]. The interplay between apoptosis and pyroptosis in OC treatment is pivotal for the development of novel therapeutic strategies. Continued research on innovative compounds and molecular pathways holds promise for improving clinical outcomes and overcoming drug resistance. Specific small molecule inhibitors can induce pyroptosis in ovarian cancer cells, thereby influencing tumor cell death and immune responses to promote anti-tumor effects. CBL0137 modulates both p53 and nuclear factor kappa-B (NF-κB), exerting anti-tumor effects across multiple cancer types [ 111 ]. The FACT complex, a transcription elongation factor involved in most chromatin-related processes, can suppress ROS-induced cell death. Yang et al. reported that CBL0137 promotes the production of reactive oxygen species (ROS) by inhibiting FACT, which induces BAX binding to the mitochondrial outer membrane, increases mitochondrial membrane permeability, and releases cytochrome c (Cyt c). Cytochrome c then activates caspase-9, which subsequently activates caspase-3, leading to the cleavage of GSDME and the induction of pyroptosis [ 95 ]. Huo et al. demonstrated that BI2536, a cell cycle regulator and PIK1 inhibitor, activates caspase-3, thereby inducing both apoptosis and GSDME-mediated pyroptosis in OC. Additionally, BI2536 enhances CD8 + T cell infiltration at tumor sites, further augmenting anti-tumor immunity [ 96 ]. ARID1A mutations are present in approximately 50% of ovarian clear cell carcinoma (OCCC) cases [ 112 , 113 ]. Zhou et al. found that inhibition of the mevalonate pathway in ARID1A-deficient OCCC can drive inflammasome-mediated immunoregulatory pyroptosis, synergizing with immune checkpoint blockade (ICB). Therefore, inhibitors of the mevalonate pathway, such as simvastatin, may inhibit tumor cell growth and enhance anti-tumor immunity by promoting pyroptosis; however, this effect is specific to ARID1A-mutant OCCC and is not observed in wild-type OCCC [ 114 ]. SF3B1, the most frequently mutated gene encoding the splicing factor 3b subunit, is highly expressed in ovarian cancer and is associated with poor prognosis and decreased cytotoxic immune cell infiltration. Pladienolide B (PB), a potent inhibitor of cancer cell growth that targets SF3B1, induces pyroptosis in ovarian cancer cells and promotes cytotoxic lymphocyte infiltration. Furthermore, PB increases programmed death ligand 1 (PD-L1) expression and acts synergistically with ICB therapy in OC [ 115 ]. PARP inhibitors olaparib and niraparib induce pyroptosis in ovarian cancer through the NF-κB–TNF–caspase-8–GSDMD/E axis. Mechanistically, activated caspase-8 directly cleaves GSDMD or indirectly activates GSDME via caspase-3, and this pyroptotic cell death subsequently enhances anti-tumor immune responses [ 97 ]. The effectiveness of small molecule inhibitors is strongly influenced by the tumor’s genetic background, particularly mutations in genes such as ARID1A and SF3B1, underscoring the importance of molecular profiling for individualized ovarian cancer therapy. Traditional Chinese medicine extracts and metabolic modulators can inhibit the proliferation of OC cells by inducing pyroptosis, thus providing a strategic avenue for novel therapies. Osthole has demonstrated anti-tumor effects, including the inhibition of OC cell migration and invasion [ 116 ]. Liang et al. found that osthole induces GSDME-dependent pyroptosis and suppresses OC cell proliferation [ 98 ]. Nobiletin, a polymethoxyflavonoid extracted from citrus fruits, was shown by Zhang et al. to induce ROS production and autophagy in OC, thereby promoting GSDMD/GSDME-mediated pyroptosis [ 99 ]. Tumor metabolism can be targeted at three distinct levels: (1) tumor cells, (2) the tumor microenvironment, and (3) systemic metabolic regulation [ 117 ]. Pyroptosis can also modulate cellular metabolic processes. Wang et al. investigated the tricarboxylic acid (TCA) cycle and found that citric acid (CA) levels were significantly reduced in ovarian cancer compared to benign ovarian tissues, indicating that CA may serve as a potential biomarker for ovarian cancer. CA promotes pyroptosis in OC cells via the CASP4/TXNIP/NLRP3/GSDMD pathway, thereby linking metabolic alterations to regulated cell death (RCD) mechanisms in cancer [ 102 ]. Glycolysis is essential for cancer stem cells (CSCs), which are intimately associated with tumor chemoresistance, metastasis, and recurrence. PFKFB3, a glycolytic enzyme, plays a pivotal role in cellular metabolism. PFKFB3 mRNA is upregulated in OC patients and cell lines, and its expression is associated with OC cell proliferation, metastasis, and poor prognosis. Studies have shown that PFKFB3 expression suppresses caspase-3/9 activity in OC, inhibits NLRP3 expression, and reduces NLRP3-induced pyroptosis [ 106 ]. Docosahexaenoic acid (DHA), an essential polyunsaturated fatty acid, can induce pyroptosis in ovarian cancer by increasing caspase-1 protein levels. OC, particularly HGSOC, is often refractory to immunotherapy; however, DHA induces pyroptosis rather than apoptosis in OC cells, thereby offering promise for OC immunotherapy [ 103 ]. Gambogic acid (GA), a natural compound derived from the resin of Garcinia hanburyi (the gamboge tree), has demonstrated potent antitumor activity. Recent studies have revealed that GA induces pyroptosis in ovarian cancer cells through the ROS/p53/mitochondrial/caspase-3/GSDME signaling pathway [ 100 ]. The interplay between metabolism and pyroptosis provides valuable insights into combination therapies targeting OC vulnerabilities. Further investigations are required to elucidate the underlying molecular mechanisms and assess the therapeutic potential of these compounds in comprehensive cancer treatment. Progesterone (P4) has been shown to induce pyroptosis in high-grade serous ovarian cancer (HGSOC) precursor cells, suggesting a novel approach for pyroptosis-based intervention in cancer prevention. TP53 mutations are prevalent in serous tubal intraepithelial carcinoma (STIC), a precursor lesion that can rapidly progress to HGSOC [ 11 ]. P4 activates intrinsic cell death pathways in HGSOC precursors by stimulating IL-6 secretion from fibroblasts, increasing ROS in TP53-deficient cells, and inducing pyroptosis via the NLRP3/caspase-1/GSDMD pathway [ 101 ]. Targeting pyroptosis to eliminate HGSOC precursor cells represents a promising therapeutic strategy, although further studies are required to facilitate clinical translation. Chronic inflammation creates a microenvironment that promotes cell proliferation, survival, and genetic alterations, thereby contributing to cancer development. Studies have shown that pelvic inflammatory disease (PID), a chronic inflammatory condition of the lower reproductive tract, is closely associated with OC [ 118 ]. Causative pathogens include Mycoplasma, Chlamydia, human papillomavirus (HPV), and human cytomegalovirus (HCMV). Recent studies have shown that Chlamydia trachomatis can induce macrophage pyroptosis via the NF-κB/caspase-1/caspase-11/GSDMD pathway, resulting in the release of LDH and IL-1β. This pyroptotic cell death leads to membrane rupture, decreased reticulate body replication, and reduced elementary body maturation, thereby lowering the infectivity of Chlamydia [ 119 ]. HCMV, whose surface-exposed glycoproteins act as viral PAMPs, activates macrophages via NF-κB and the NLRP3 inflammasome, which in turn promotes pyroptosis and IL-1β secretion through the caspase-1/GSDMD pathway. However, the precise balance between the benefits and detriments of HCMV-induced pyroptosis—specifically, whether it primarily promotes viral clearance or contributes to tissue damage—remains to be fully elucidated [ 120 ]. These insights into infection-induced inflammatory responses highlight the complex interplay between chronic inflammation, infectious agents, and carcinogenesis, thereby informing the development of strategies for cancer prevention and treatment.

Pyroptosis is a form of regulated cell death (RCD) mediated by the gasdermin (GSDM) family of proteins, which is characterized by a distinct morphological pattern of cell death that differs from other types of RCDs. The abnormal death phenotype of disrupted cell monolayers was first observed in 1986 when Arthur Friedlander exposed mouse macrophages to anthrax lethal toxin [ 1 ]. In 2001, Cookson and Brennan first coined the term “pyroptosis” to describe an inflammatory RCD induced by caspase − 1 activation in potentially endangered cells [ 2 ]. Subsequent studies proposed that pyroptosis is induced by the activation of caspase-1 via various pathways [ 3 ]. In 2015, Jianjin Shi et al. found that the activation of inflammatory caspases, such as GSDMD, resulted in pyroptosis, and in 2017, they reported that GSDMs are the executors of pyroptosis [ 4 , 5 ]. In 2018, the Nomenclature Committee on Cell Death (NCCD) revised the definition of pyroptosis as a form of RCD primarily activated by inflammatory caspases and dependent on GSDM - mediated pore formation in the plasma membrane [ 6 ]. To this day, the definition and understanding of pyroptosis are still evolving. The unique pattern of cell death associated with pyroptosis is attributed to the pores formed by GSDMs in the plasma membrane [ 5 ]. These pores act as non - selective membrane channels, disrupting the ion balance inside and outside the cell [ 5 ], which act as non-selective membrane channels, disrupting the ion balance inside and outside the cell. This results in a massive influx of water, leading to cell swelling and rupture [ 7 ]. Additionally, these GSDM pores release large amounts of inflammatory factors via membrane translocation, vesicle release, and membrane rupture. As a novel type of RCD, pyroptosis has been extensively studied in numerous diseases, including cancer, and has been verified to play a significant role in various cancers [ 8 ]. However, pyroptosis is a double - edged sword because its role in cancer is context - dependent. It can inhibit cancer, but it may also cause tissue damage and promote cancer progression [ 9 ]. Ovarian cancer (OC) is a malignancy that originates in the ovaries. According to the 2020 global cancer incidence and mortality estimates released by the International Agency for Research on Cancer (IARC), there were 19.3 million new cancer cases worldwide in 2020, with 313,959 cases of OC, accounting for 1.6% of all new cancer cases, and 207,252 deaths, accounting for 2.1% of all cancer deaths [ 10 ]. OC can be further subdivided into more than 15 molecular and pathological subtypes. Based on tissue origin, it can be classified into epithelial tumors, germ cell tumors, and sex cord - stromal tumors. Epithelial ovarian cancer (EOC) accounts for approximately 90% of all OC cases, and includes subtypes such as serous, endometrioid, clear cell, and mucinous [ 11 ]. The early symptoms of OC are often insidious, while in advanced stages, clinical manifestations such as pelvic masses and ascites may emerge [ 12 ]. According to the FIGO 2018 guidelines, for early - stage OC, the primary treatment consists of surgical resection of the primary tumor and metastases. For advanced - stage OC, neoadjuvant chemotherapy (NACT) combined with interval debulking surgery (IDS) is recommended [ 13 ]. However, owing to the lack of effective screening methods and the insidiousness of early symptoms, approximately 80% of OC patients are diagnosed at advanced stages [ 14 ]. The 5 - year overall survival rate is less than 50% [ 15 ], and the survival rate for advanced - stage patients can drop to as low as 25% [ 16 ]. Over 70% of advanced - stage EOC patients suffer from relapse after treatment [ 17 ]. Moreover, the chemoresistance of OC is a major factor contributing to its poor prognosis [ 15 ]. OC is a global health challenge with a low survival rate and high recurrence rate, which presents a significant challenge to treatment. Traditional therapies are insufficient to address the resistance of OC. Therefore, finding new intervention strategies is crucial. In recent years, there has been increasing evidence of a close relationship between pyroptosis and OC. In 2022, Liu Tianyi et al. carried out the first comprehensive review of the relationship between pyroptosis and ovarian cancer, systematically delineating the four major pyroptosis pathways. Their study provided an in - depth analysis of key regulators of pyroptosis in ovarian cancer, including GSDM family proteins, LncRNAs, osthole, nobiletin, and α - NETA [ 18 ]. Subsequently, Wang Aihong et al. built on this foundation by focusing on the interaction between pyroptosis and the tumor microenvironment (TME), with particular emphasis on translating basic research into clinical applications. Their study clarified the mechanisms by which PD − 1/PD - L1 inhibitors and CAR - T cells enhance anti - tumor immunity through pyroptosis induction [ 19 ]. As mechanistic research on pyroptosis progresses, bioinformatics models exploring its association with OC have emerged, along with the identification of novel pyroptosis - inducing agents. Notably, even paclitaxel, a first - line chemotherapeutic agent for ovarian cancer, has been associated with pyroptotic cell death. This paper builds upon existing literature by supplementing recent bioinformatics analyses of pyroptosis-related genes in OC and summarizing newly identified pyroptosis regulators. Furthermore, in alignment with current research trends, we delineate the pyroptosis signaling pathway with a focus on the pivotal role of GSDM proteins and systematically examine the functions of key pyroptosis-related molecules in OC pathogenesis.

Supplementary Material 1. Supplementary Material 1.