The multifaceted roles of gasdermins in cancer biology and oncologic therapies

BBA - Reviews on Cancer 1876 (2021) 188635 Available online 14 October 2021 0304-419X/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Review Article The multifaceted roles of gasdermins in cancer biology and oncologic therapies David Sarri´ o a , b , * , Jeannette Martínez-Val c , ´ Angela Molina-Crespo a , b , Laura S´ anchez c , Gema Moreno-Bueno a , b , d , * a Biochemistry Department, UAM, & IIBm "Alberto Sols" CSIC-UAM, c/ Arzobispo Morcillo 4, 28029 Madrid, Spain. b Centro de Investigaci´ on Biom´ edica en Red, ´ area de C´ ancer (CIBERONC), c/ Melchor Fern´ andez Almagro 3, 28029 Madrid, Spain. c Zoology, Genetics and Physical Anthropology Department, Santiago de Compostela University, Avda/ Alfonso X O Sabio s/n, 27002 Lugo, Spain. d MD Anderson Cancer Center Foundation, c/ Arturo Soria 270, 28033 Madrid, Spain. ARTICLE INFO Keywords: Gasdermin Pyroptosis cancer cell death cancer treatment Inflammation ABSTRACT The involvement of the Gasdermin (GSDM) protein family in cancer and other pathologies is one of the hottest topics in biomedical research. There are six GSDMs in humans (GSDMA, B, C, D, GSDME/DFNA5 and PJVK/ DFNB59) and, except PJVK, they can trigger cell death mostly by pyroptosis (a form of lytic and proinflammatory cell death) but also other mechanisms. The exact role of GSDMs in cancer is intricate, since depending on the biological context, these proteins have diverse cell-death dependent and independent functions, exhibit either pro-tumor or anti-tumor functions, and promote either sensitization or resistance to oncologic treatments. In this review we provide a comprehensive overview on the multifaceted roles of the GSDMs in cancer, and we critically discuss the possibilities of exploiting GSDM functions as determinants of anti-cancer treatment and as novel therapeutic targets, with special emphasis on innovative GSDM-directed nano-therapies. Finally, we discuss the issues to be resolved before GSDM-mediated oncologic therapies became a reality at the clinical level. 1. The gasdermin gene family The gasdermin family (GSDM) comprises six genes in humans (GSDMA, B, C, D, GSDME/DFNA5, and PJVK/DFNB59) and ten in mice (Gsdma1–3, Gsdmc1–4, Gsdmd, Gsdme, and Dfnb59) [1–3]. GSDME (A.K.A DFNA5, deafness autosomal dominant 5) is the most ancient GSDM gene, being present in some invertebrates like corals [2,4], while GSDMB appears only in specific mammalian species, being the only GSDM member not present in the mouse and rat genomes [2]. GSDM usually cluster in specific chromosomal regions (17q21 contains GSDMA and GSDMB; 8q24 GSDMC and GSDMD; Table 1) indicating that gene duplications occurred during vertebrate evolution [2,4]. In fact, GSDMB likely originated from a local duplication and DNA-strand inversion of GSDMA in some mammals [2,4]. The name Gasdermin originates from the “Gastric and dermal” expression of GSDMA, the first identified GSDM gene [5]. GSDMs are generally expressed in the digestive tract, where they show specific expression patterns: in the gastro-esophageal epithelium GSDMA is mostly detected in differentiated cells, GSDMB in the basal layers, GSDMC in the suprabasal and differentiated regions and GSDMD in the differentiating cells [1,3]. In addition, GSDMs show specific expression patterns in multiple tissues/organs. For example, GSDMA is expressed in skin, lung, mammary glands; GSDMB in liver, lung, colon and immune cells; GSDMC in skin and spleen; GSDMD in the intestine and leukocytes; GSDME in reproductive organs and nervous system and PJVK in testes and the auditory nervous system, among others [1–4,6,7]. 2. GSDM at the crossroads between pyroptosis and other cell death mechanisms GSDMs are cytoplasmic proteins (around 50 KDa) with nine Abbreviations: CT, C-terminal domain; DAMPs, damaged-associated molecular patterns; DCT, Decitabine; GZM, Granzyme; KO, Knock-out; LDH, Lactate dehydrogenase; LPS, lipopolysaccharide; NLRP3, NOD-like receptor protein 3; NC, Nanocapsules; NT, N-terminal domain; NP, Nanoparticles; PLK1, Polo like kinase 1; ROS, reactive oxygen species; SCC, Squamous Cell Carcinoma; TGF-β, transforming growth factor-β; TNF, Tumor necrosis factor; TSG, Tumor suppressor gene; WB, Western blot; WT, Wild-type. * Corresponding authors at: Biochemistry Department UAM, & IIBm "Alberto Sols" CSIC-UAM, & CIBERONC E-mail addresses: [email protected] (D. Sarri´ o), [email protected] (G. Moreno-Bueno). Contents lists available at ScienceDirect BBA - Reviews on Cancer journal homepage: www.elsevier.com/locate/bbacan https://doi.org/10.1016/j.bbcan.2021.188635 Received 30 June 2021; Received in revised form 28 September 2021; Accepted 10 October 2021 BBA - Reviews on Cancer 1876 (2021) 188635 2 conserved motifs (containing leucine-rich regions) of still unknown function, distributed throughout the protein [3]. GSDMs, with the exception of PJVK, share a moderately conserved N-terminal (NT) and C-terminal (CT) domains that are interconnected by a central hinge region, which is specific for each family member [2]. For many years, GSDMs functions have been unclear until the discovery, in 2015, that GSDMD trigger inflammasome-dependent lytic cell death [8–10] by forming membrane pores [11–15]. Further studies revealed that other GSDMs exhibit a similar protein 3D structure [15–18] and share the procell death function that is activated by a common mechanism [15,19,20]. The common model shows that cytosolic GSDMs have an inactive conformation in which the NT pore-forming domain is autoinhibited by binding to specific residues in the hydrophobic core of the CT [11–15,19,20]. This folded conformation is maintained by the linker interdomain region, which is cleaved by specific proteases activated after particular stimuli (Fig. 1A). The released GSDM NT binds to specific acid lipids (e.g., phosphoinositides and phosphatidylserine) and inserts into the cell membrane internal layer, oligomerizing and forming large transmembrane pores [11–15,19,20] (Fig. 1B). The upstream pathways leading to GSDM NT release are specific of each GSDM [7,19,20]. Regarding GSDMD, multiple pathogen signals (e.g., LPS, flagellin) or cell damage molecular patterns “DAMPs” (e.g., reactive oxygen species “ROS”, dsDNA) activate the multiprotein complex inflammasome, which then provokes the release of GSDMD NT domain by the pro-inflammatory caspases1/4/5/11 [7,19,20]. GSDMD NT pores facilitate the secretion of mature inflammatory cytokines (IL1β, IL18) and other intracellular molecules (i.e., Lactate dehydrogenase; LDH), while the influx of water and ions might be responsible for cell swelling [7,19,20]. Later, the activation and oligomerization of Ninjurin-1 (NINJ1) produces the extensive membrane rupture and cell lysis (Fig. 1B), which results in the release of diverse intracellular content, including the pro-inflammatory DAMP HMGB1 [21]). This inflammasome/GSDMD-mediated mechanism is termed pyroptosis (“fiery death”) and can be divided into “canonical” (caspase 1) or “noncanonical” (caspase 4/5/11) depending on the stimuli and caspases involved [7,19,20]. Pyroptosis occurs mostly in leukocytes and epithelial cells and provokes an inflammatory reaction of the surrounding cells [7,19,20]. Pyroptosis is mechanistically and biologically different to other programmed cell death types, such as apoptosis, necroptosis or ferroptosis, and dying cells exhibit distinct morphologic features: extensive cell swelling and membrane ballooning in absence of cell detachment, intact nucleus with chromatic condensation, among others [22]. Except for PJVK, the NT of other GSDMs can produce lytic cell death [15,19,20] with the release of LDH and other molecules. Therefore, pyroptosis was renamed as “Gasdermin-mediated programmed cell death” [23], but accumulating evidences proved that GSDMs can orchestrate other cell death processes and additional death-independent functions. Indeed, each GSDM is activated by specific proteases in a cell context-dependent way (Fig. 1A) that could lead to different biological consequences: GSDMD NT is released by inflammatory caspases during canonical and non-canonical pyroptosis [9,10], whereas Yersinia infection [24,25] or RIPK3 signaling [26] provokes caspase-8-mediated GSDMD activation. Cathepsin-G activates GSDMD during NOD-like receptor protein 3 (NLRP3)-stimulated pyroptosis in macrophages and neutrophils [27], but neutrophil elastase (NE) cleavage in neutrophils could result in pyroptosis [28] or NETosis [29]. Of note, GSDMD pores can release cytokines in leukocytes in absence of cell death [30], and GSDMD-cell death can be inhibited by the Endosomal sorting complexes required for transport (ESCRT-III) mechanism [31]. GSDME is activated by apoptotic caspase-3, in response to multiple stimuli [32,33], leading to cell lysis considered by some authors as “necrosis secondary to apoptosis” [32,34], while GSDME processing via killer cell-derived Granzyme-B (GZMB) induces pyroptotic and immunogenic cell death in tumor cells [35]. Surprisingly, even caspase-8 (after Yersinia infection) and caspases 4/11 can initiate signaling pathways leading to GSDME cleavage and cell death [36]. Immunocyte-released Granzyme-A (GZMA) cleavage of GSDMB can produce either lytic cell death in cancer cells [37] or selective killing of Shigella intracellular bacteria but not the infected cells [38]. Moreover, unprocessed GSDMB enhances caspase-4/ GSDMD non-canonical pyroptosis [39]. Finally, both TNFα +hypoxia [40] and α -Ketoglutarate [41] induce Caspase-8 cleavage of GSDMC but in different residues (Fig. 1A). GSDMC NT provokes pyroptosis in cancer cells but it could result in chronic tumor necrosis [40]. The proteases activating human and mouse GSDMA and PJVK proteins are still unknown. Notably, cell-death function can be inhibited by protease processing within the GSDM NT domain, like GSDMD cleavage by caspases3/7 [42] or enterovirus 3C protease [43]. Likewise, diverse caspases (1/ 3/6/7) can cleave and inactivate the GSDMB NT [16,39], contradicting the work by Panganiban and collaborators, which suggested that caspase-1 cleavage of GSDMB linker induced pyroptosis during asthma [44]. Of note, GSDM-mediated cell death is more complex than merely forming cell membrane pores, since some GSDM NT can also target intracellular organelle (mitochondria, neutrophil granules and possibly the nucleus) [4,7] (Fig. 1C). In particular, activated GSDMs (GSDMA/ A3/D/E) trigger mitochondrial damage [45–48], and conversely, altered mitochondrial function induces GSDMD/E cleavage [49]. In fact, time lapse microscopy reveals that during pyroptosis mitochondrial damage precedes cell lysis [45,46]. Mitochondria dysfunction can occur in diverse ways. After caspase-3/GSDME cleavage, GSDME pores permeabilize the mitochondrial membrane leading to downstream apoptosome activation [49]. Thus, GSDME activation cause a positive feedback loop enhancing mitochondrial apoptosis and/or pyroptosis. By contrast, constitutively active mutant GSDMA3 proteins could cause mitochondrial damage by two mechanisms: binding to the mitochondrial chaperone Trap1, which promotes oxidative stress and loss of the mitochondria membrane potential triggering apoptosis-independent cell death [47], or through stimulation of pro cell-death autophagy [48]. Consistent with the GSDM mitochondrial targeting, GSDM NTs exhibit strong binding affinity for cardiolipin, a lipid enriched in the internal mitochondrial membranes [15,50]. Taken together, these evidences proved that GSDMs coordinate an extensive and complex cross-talk between diverse cell death pathways [22,34], but the precise functions of each GSDM in physiology and disease has only started to emerge. The molecular mechanisms regulating inflammasome-dependent and independent pyroptosis and the implication of GSDMs in physiological processes (e.g., response to infectious agents) and pathologies has been extensively revised elsewhere [7,19,20]. As examples, GSDM procell death functions are involved in multiple inflammatory pathologies such as sepsis, autoimmune encephalomyelitis (GSDMD), asthma and inflammatory bowel diseases (GSDMB), among others. Additionally, rare GSDM mutations (Fig. 1) provoke pathogenic phenotypes in mice (Gsdma3 mutations that impede CT protein inhibitory function trigger skin inflammation and hair loss, among other defects) and humans (GSDME and PJVK truncating mutations produce hereditary deafness disorders) [7,19,20]. In this review we provide a comprehensive overview on the multifaceted roles of the GSDM family in cancer, covering not only their procell death activities in particular tumor contexts, but also other functions that eventually could lead to tumor progression. Moreover, we will focus on the implication of GSDMs in mediating cancer response to treatment and the relevance of GSDMs as novel therapeutic targets for clinical oncology management, with special emphasis on novel GSDM-directed nano-therapies. 3. Anti-tumor and pro-tumor effects of GSDMs The pro-cell death activity of GSDMs could lead to the idea that these proteins play mainly an anticancer function. Yet, GSDM genes are often located in genomic regions frequently amplified in cancers (Table 1), D. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 3 Table 1 Expression and functional roles of GSDMs in human cancers. Cancer type (suggested function*) Expression in tumors (method) & effects in untreated cancer cells. Ref Human GSDMA (GSDM, GSDM1) [17q21.1] Mouse Gsdma1/2/3 [11D] Breast, gastric & ovarian (anti-tumor) No expression was detected in any of the 24 cancer cell lines (Northern blot), even though GSDMA gene was amplified in 4/4 (100%) breast and 2/8 (25%) gastric HER2 cell lines. [52] Esophageal & Gastric (antitumor) Undetectable expression in 58/60 (97%) tumors and 17/21 (81%) cell lines (RT-PCR). Overexpression decreases colony formation in MKN28 cells. [1] Gastric (anti-tumor) Undetectable expression in 11/18 (61%) tumors and 8/11 (73%) cell lines (RT-PCR). GSDMA is restored by inhibition of promoter hypermethylation in 6/8 (75%) lines. Overexpression in cell lines decreases cell growth and induces cell death. [51] Ovarian (pro-tumor) Greater mean expression in tumors (n =379) compared to normal tissue (n =88) (mRNA DBs). Higher GSDMA cancer expression associates with poor survival. [53] Human GSDMB (GSDML, PRO2521) [17q21.1] Not present in mouse Bladder (pro-tumor) Larger mean expression in tumors compared to normal tissue (n =19) (mRNA DBs). GSDMB and USP24 protein expression are positively correlated in bladder cancers (n =80; IHC). GSDMB silencing in T24 and 5637 cells dampens proliferation, migration and invasion and reduces in vivo tumor growth of T24 xenografts. GSDMB controls glycolysis and in vivo cancer growth via STAT3 signaling. [59] Breast (pro-tumor) Greater mean expression in tumors compared to normal tissue (RT-PCR, n =18). Higher mRNA tumor expression associates with poor prognosis (n =1628; mRNA DBs). Its over-expression in MCF7 cells induces cell motility, invasion, gelatin degradation in vitro & in vivo tumor growth and metastasis (mostly isoform 2). GSDMB silencing reduces migration and invasion of HCC1954 cells. [56] HER2 Breast (pro-tumor) Overexpressed (80/212; 66%) (IHC) and gene-amplified (FISH) (73/123; 59%) in HER2-positive tumors. Higher levels (n =212; IHC & n =2096; mRNA DBs) associate with metastasis, poor prognosis and treatment relapse. [55] HER2 Breast (pro-tumor) GSDMB expression promotes migration in HCC1954, SK-BR-3, and BT474 cells. Enhances lung metastasis in MDA-MB-231 cells. [61] Cervical (pro-tumor) More frequently expressed in tumors (16/21; 76%) than corresponding non-neoplastic areas (8/21; 38%) (IHC). Its silencing reduces cell growth in Hela cells. [57] Cervical (uncertain) Equal expression frequency in tumors (23/27; 85%) than corresponding non-neoplastic areas (24/27; 89%) (IHC). [37] Colorectal (uncertain) Similar expression frequency in tumors (166/230; 72%) than corresponding non-neoplastic areas (154/230; 67%) (IHC). [37] Esophageal (anti-tumor) Less frequent expression in tumors (44/80; 55%) than corresponding non-neoplastic areas (72/80; 90%) (IHC). [37] Esophageal & Gastric (pro-tumor) Expressed in 47/60 (78%) tumors and 21/21 (100%) cell lines (RT-PCR). Co-amplified with HER2 in 2/8 (25%) gastric cancers. Its overexpression in MKN28 cells does not affect colony formation. [1] Gastric (anti-tumor) Less frequent expression in tumors (34/75; 45%) than corresponding non-neoplastic areas (66/75; 88%) (IHC). [37] Gastric (pro-tumor) More frequently expressed in tumors (44/52; 85%) than normal tissues (29/82; 35%) (RT-PCR). Stronger expression in tumor cells compared to normal cells (ISH, n =5). [58] Gastric, CRC & HCC (uncertain) Similar mean expression between tumors (gastric n =21; HCC n =15; colon n =9) than corresponding non-neoplastic areas (RT-PCR). [6] HCC (uncertain) Similar expression frequency in tumors (17/21; 81%) than corresponding non-neoplastic areas (21/21; 100%) (IHC). Its silencing does not affect cell growth in HepG2 cells. [57] Lung (pro-tumor) Higher mean expression in lung adenocarcinomas than normal tissue (n =515; mRNA DBs). [54] Multiple cell line types GSDMB protein endogenous expression detected only in 18/54 (33%) cell lines (WB). Its expression could be induced by IFN-Ɣ in 11 cell lines (WB). GZMA induces pyroptosis killing of GSDMB-expressing cell lines. [37] OSCC (pro-tumor) Higher mean expression in lymph-node metastatic cancers vs non-metastatic (n =53; microarrays and RT-PCR) [60] Pancreatic (uncertain) Similar expression frequency in tumors (41/77; 53%) than corresponding non-neoplastic areas (45/77; 45%) (IHC). [37] Human GSDMC (MLZE) [8q24.21] Mouse Gsdmc1–4 [15D1] Breast (pro-tumor) Higher levels associate with worse overall survival (n =626; IHC). [40] Colorectal (pro-tumor) Higher mean expression in tumors vs adjacent normal tissue (n =44; RT-PCR). GSDMC protein detected only in tumors, not in normal tissue (n =44; IHC). GSDMC was silenced (DLD-1 and LoVo cells) or over-expressed (SW480 and WiDr cells). GSDMC upregulation promoted in vitro proliferation, anchorage-independent growth and in vivo tumorigenesis. [68] Esophageal & Gastric (antitumor) Expressed in 34/60 (57%) of tumors and 16/21 (76%) cell lines (RT-PCR). Its overexpression reduces slightly colony formation in MKN28 cells. [1] (continued on next page) D. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 4 Table 1 (continued) Cancer type (suggested function*) Expression in tumors (method) & effects in untreated cancer cells. Ref Lung (pro-tumor) Greater mean expression in lung adenocarcinomas vs normal tissue (n =515; mRNA DBs). Higher GSDMC expression associates with poor prognosis, metastasis and radio-resistance (mRNA DBs). Hypermethylation associates with gene silencing. [54] Melanoma (pro-tumor) Expressed in 8/26 (30%) melanomas but not in nevi (n =5). More frequently expressed in metastatic (6/11; 55%) than non-metastatic (2/15; 13%) melanomas (IHC). [67] Human GSDMD (GSDMDC1, DFNA5L, DF5L, FKSG10) [8q24.3] Mouse Gsdmd [15D3-E1] ESCC (uncertain) Higher mean expression in tumors than in normal tissues (n =148; mRNA DBs and n =30; IHC). [133] Esophageal & Gastric (antitumor) Expressed in 38/60 (63%) of tumors and 21/21 (100%) cell lines (RT-PCR). Its overexpression strongly reduces colony formation in MKN28 cells. [1] Gastric (anti-tumor) Strong expression more frequent in normal tissue (41/61; 67%) than in tumors (27/102; 27%) (IHC). mRNA levels higher in normal vs matched tumor tissue (n =39; RT-PCR). GSDMD expression in BGC823 cells reduces in vitro proliferation and in vivo cancer growth. [69] Lung (pro-tumor) Higher expression (IHC score; n =168) in adenocarcinomas and SCCs than normal tissue. Overexpression associates with increased tumor size, stage and lower survival in adenocarcinomas. GSDMD silencing in PC9, H1703 and H1975 cells reduces proliferation, promotes apoptosis and attenuates in vivo tumor growth. [70] Osteo-sarcoma (pro-tumor) Expressed in 20/41 (49%) tumors and none of non-neoplastic areas (IHC). Greater mean protein expression in tumors than matched normal tissues (n =61; WB). Higher levels associate with metastasis, resistance to chemotherapy, and poor survival. [134] Salivary ACC (pro-tumor) More frequently expressed in ACC (33/33; 100%), than adenomas (23/29; 79%) & normal tissues (24/33; 73%) (IHC). Strong IHC expression only seen in ACC. Over-expression in ACC-LM and ACC-83 cells enhances in vitro invasion. [71] Human GSDME (DFNA5) [7p15.3] Mouse Dfna5 [6B2.3] Breast (anti-tumor) Lower mean expression in tumors vs normal samples (n =1142; mRNA DBs). Reduced GSDME in tumors does not associate with survival. Greater GSDME levels in ER-negative and lobular carcinomas. Higher promoter hypermethylation in tumors. GSDME gene body methylation associates with reduced overall survival. [79] Breast (anti-tumor) Reduced mean expression in cancer vs normal samples (n =2509; mRNA DBs). Higher tumor GSDME does not associate with survival but correlate with immune-related genes. No association between promoter methylation and gene expression. [96] Breast (anti-tumor) Decreased mean expression in cancers vs normal samples (n =10; RT-PCR). Higher methylation in tumors. GSDME expression can be activated with demethylating agents. GSDME silencing increases colony formation, proliferation and invasion of MDA-MB-231 cells. [75] Breast and colon (anti-tumor) Lower mean expression in breast (n =1097) and colon (n =286) cancers vs normal samples (mRNA DBs). GSDME suppresses in vivo tumor growth and promotes anti-tumor immunity of murine cells (EMT6, CT26, 4 T1 & B16–10) in immunocompetent mice. [35] Colorectal (anti-tumor) Lower expression in cancers vs normal tissue (n =5) (RT-PCR). Higher hypermethylation in tumors (65%) than normal tissue (3%). GSDME in HCT116 cells reduces cell growth and colony formation. [73] Colorectal (uncertain) Using Gsdme KO mice, no clear effects were seen in two experimental models of intestinal cancer: the chemical induction by azoxymethane “AOM” or crossing with the Apc1638N/+strain. [82] ESCC (anti-tumor) Greater expression (IHC score) in cancers compared to normal tissue (IHC). Increased GSDME in SCCs (n =104; IHC) associates with enhanced therapy response and better prognosis. [116] Gastric (anti-tumor) Frequent promoter hypermethylation in tumors (53%). GSDME expression upregulated by demethylating agents in gastric cell lines. GSDME over-expression reduces colony formation and augments apoptosis in NUGC3 cells. [74] GBM (uncertain) Higher mean expression in tumors than in normal brain (mRNA DBs), but GSDME levels do not associate with patients’ survival. [89] Lung (uncertain) Expressed in 59% of cancers, similar levels than normal tissues (IHC). Expressed in 95% of cell lines (WB). GSDME does not affect tumor growth of human NCI-H3122 and HCC827 xenografts. [78] Multiple cell lines (anti-tumor) Protein detected in 30/57 (53%) cell lines, strong expression in 11/57 (19%) (WB). Gsdme KO mice do not exhibit developmental defects. [33] Melanoma (anti-tumor) GSDME silencing in B16-Ova cells increases in vivo tumor growth. [49] Human PJVK (Pejvakin, DFNB59) [2q31.2] Mouse Dfnb59 [2C3] Ovarian (anti-tumor) Lower mean expression in tumors (n =379) compared to normal tissue (mRNA DBs). Lower PJVK cancer expression associated with poor survival. [53] Table rows are ordered alphabetically by GSDM gene name (aliases indicated in parentheses and gene locus in square brackets) and then by tumor types. * The overall effect (proven or inferred) on tumor biology (excluding effects on cancer treatment) according to study results. Abbreviations: ACC: Adenoid Cystic Carcinoma; CRC: Colorectal; ESCC: Esophageal Squamous Cell Carcinoma; GBM: Glioblastoma Multiforme; IHC: Immunohistochemistry; ISH: in situ hybridization; mRNA DBs: Data Bases of mRNA expression (microarrays or RNAseq); SCC: Squamous Cell Carcinoma; WB: Western Blot. D. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 5 such as the 17q12–21 region (GSDMA and GSDMB), that contains the HER2/ERBB2 oncogene, or the 8q24 region (GSDMC and GSDMD), close to MYC oncogene. Moreover, GSDMs are either upregulated or downregulated (mRNA and/or protein) in a wide range of cancer types compared to their respective normal tissues (Table 1). In fact, accumulating evidences indicate that GSDMs play complex roles in cancer biology, possibly having either pro-tumor or anti-tumor functions depending on the cellular context (summarized in Table 1). The involvement of GSDMs in oncologic treatment response will be presented in section 4. GSDMA was initially described as frequently lost in gastroesophageal cancers [1,51] and breast cancer cells [52], even in HER2positive cell lines with GSDMA gene co-amplification [52]. GSDMA expression could be restored by methylation inhibitors [51]. In gastric cancer cells, GSDMA upregulation, mediated by TGF-β (transforming growth factor-β) and the transcription factor LMO1 (LIM domain only 1), promoted an apoptotic signal leading to reduced cell growth [51]. These data point to a potential tumor-suppressor gene (TSG) role for GSDMA. Contrarily, in expression databases GSDMA mRNA was upregulated in ovarian carcinomas versus normal tissue, and higher GSDMA levels associated with worse survival [53]. Likewise, GSDMA mRNA is upregulated in some lung cancer datasets [54] and breast tumors [55,56], but the relevance of these observations is uncertain. In fact, despite being the first GSDM gene identified [5], the mechanism of cleavage/activation and the precise functional effects of GSDMA in cancer remains unclear. Unlike other GSDMs, GSDMB does not consistently reduce cell growth of gastric cancer cells or other cell models [1,56,57]. Moreover, contrary to its neighbor gene GSDMA, GSDMB is frequently expressed (mRNA and/or protein) in several human tumor types and cancer cell lines, including gastric, hepatic, and breast, among others (Table 1). Nonetheless, compared to normal tissues, GSDMB levels are either upregulated [1,54–59], downregulated [37] or unchanged [6,37,57] depending on the tumor type and the methods used to measure its expression (Table 1). In terms of mRNA, GSDMB over-expression associates with advanced/metastatic disease in oral SCC [60] and gastric tumors [58], shorter disease free and metastasis free survival in breast cancer [56], and overall survival in only 3 of 33 cancer types tested by Zhou and colleagues [37]. However, since GSDMB mRNA is detected in various normal cell types [7] the true GSDMB expression in tumors should be assessed by immunohistochemistry. Using a validated GSDMB antibody and a specific FISH probe, Hergueta-Redondo and colleagues [55] demonstrated that GSDMB is over-expressed/gene-amplified in >60% of HER2 breast cancers. Similarly, in gastro-esophageal cancers GSDMB and HER2 co-amplification is a common finding [1]. Importantly, in HER2 breast carcinomas, GSDMB protein upregulation associated significantly with various adverse clinical parameters: disease progression, relapse and response after neoadjuvant therapy, as well as distant metastases and lymph node positivity in the adjuvant setting, independently of hormonal receptors status [55]. Moreover, GSDMB expression mediates multiple pro-tumor functions in breast neoplasias: increases in vitro cell migration, possibly by modulating the Rac GTPases, invasion and gelatin degradation [56], stimulates in vivo metastatic dissemination [56,61], and reduces sensitivity to anti-HER2 therapy. Actually, all these pro-tumor activities could be decreased with a novel anti-GSDMB nanotherapy [61], thus demonstrating that GSDMB plays a key role in the biology and clinical behavior of HER2 breast carcinomas. Similarly, in bladder cancer cells GSDMB promotes proliferation, Fig. 1. GSDM pore-forming functions. A: Regulation of GSDM function by cleavage or mutation. Schematic representation of GSDM regions (NT, N-terminal domain; CT, C-terminal; and linker interdomain) showing the known proteases and respective cleavage sites (in parentheses, the cleavage sites of murine proteins are indicated in grey letters). The cleavage events inducing pore-forming activity are shown in red lines, and inhibitory processing in black. Gain-of–function mutations in the CT of GSDMA3 and GSDME are indicated by asterisks. B: Main steps in the formation of GSDM membrane pores. There are differences in the structure and size of GSDM pores. GSDMA3 form pores containing around 27 monomers and with an internal pore diameter of 16–18 nm; GSDMD, 33 monomers and up to 21 nm internal diameter. C: Summary of mechanisms of GSDM activation and biological effects. Casp, Caspase; CatG Cathepsin G; GZM, Granzyme; NE, Neutrophil Elastase. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) D. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 6 migration and invasion [59]. These evidences and the previous observations in gastric carcinomas [1,58] indicate that GSDMB could act like an oncogene. However, the in vivo effect of GSDMB on tumorigenesis and progression might depend on the cancer models, since it promotes tumor growth in MCF7 breast [56] and T24 bladder cancer xenografts [59] but no other human [61] and murine [37] cancer models (Table 1). At this point, it is important to highlight the existence of multiple GSDMB transcriptional variants [6,57] and at least four distinct translated protein isoforms (ENSEMBL:ENSG00000073605). Though, for unknown reasons the 411 aa protein sequence -Q8TAX9-, which was partially crystalized [16], and is regarded as “canonical” in all databases, does not match any of the transcriptional variants. The four translated isoforms, which differ on the alternative usage of exons 6–7, can play different functions in cancer [56] and inflammatory diseases [44,62]. The isoform differential expression was reported in normal and cancer tissues [6,57,58,63–65] and could be regulated by specific genomic elements [58,64,65] and SNPs [44,63,66]. However, there is scarce information on the precise functions of each translated variant in normal and pathological contexts. In this regard, Hergueta-Redondo et al. reported that in MCF7 breast cancer cells both GSDMB isoform 1 (GSDMB-1; lacks exon 6) and 2 (GSDMB-2; the shortest isoform, lacking exons 6–7) enhanced migration and invasion in vitro, while only GSDMB2 promoted in vivo tumor growth and metastatic dissemination in immunodeficient mice [56]. Opposing its pro-tumor functions, GSDMB can, under specific stimuli, exhibit pyroptotic activity in cancer cells [37,61], and thus an antitumor effect. Specifically, in a context of an activated antitumor immune response, the lymphocyte-derived Granzyme A (GZMA) can cleave GSDMB within cancer cells (Fig. 1A,C), activating caspaseindependent pyroptosis and promoting tumor clearance [37]. Contrasting with other studies (Table 1), Zhou and cols [37] reported a lower GSDMB immunohistochemical staining in gastro-esophageal tumors compared to normal tissues, but not other cancer types (Table 1), suggesting that GSDMB could act as tumor suppressor in these neoplasias. Unfortunately, this study did not evaluate the association of GSDMB staining with clinicopathological variables or disease survival, thus the clinical relevance of these observations remains unclear. GSDMC was initially identified as a marker of melanoma progression, being particularly over-expressed in metastatic melanoma cells [67]. Further studies reported opposing results regarding its expression and functional effects in tumors. On one hand, Saeki and cols found GSDMC expression in 50% of gastro-esophageal tumors and cell lines but was mostly lost in diffuse-type gastric carcinomas. In gastric cancer cells, GSDMC expression reduced cell proliferation, suggesting a possible tumor suppressor function [1]. On the other hand, GSDMC upregulation in lung [54] and breast [40] tumors associates with adverse clinical parameters (Table 1), and increases proliferation along with in vivo tumor growth in colorectal cancer cells [68], suggesting a pro-tumor role. Activation of GSDMC cleavage by drugs or metabolites ( α -KG) [41] induces pyroptosis in cancer cells (Fig. 1; Table 2), but can also provoke chronic necrosis [40]. In the latter case, under hypoxia nuclear PD-L1 activates GSDMC transcription in breast cancer cells. In parallel, activated macrophages secrete TNFα resulting in caspase-8 activation within cancer cells. The coordinated upregulation of GSDMC and caspase-8 induces GSMDC cleavage, causes pyroptosis and subsequent necrosis in breast cancer cells [40]. Paradoxically, chronic tumor necrosis fuels cancer progression, and therefore patients with GSDMC-overexpressing breast cancers show poor survival [40]. This study challenges the idea that GSDM-mediated cancer cell death can have beneficial effects on cancer progression. Whereas GSDMD role in leukocyte pyroptosis is well known, its implication in cancer is debatable. Depending on the study, both GSDMD upregulation and downregulation (mRNA and/or protein) have been described in tumors, where GSDMD overexpression associates with either good or unfavorable prognosis, as well as antitumor or pro-tumor effects (Table 1). For instance, GSDMD expression diminishes proliferation in gastric cancer cells, either via cell death induction [1] or by regulating cell cycle-related proteins [69], but it increases lung cancer growth in vivo [70] and invasion of salivary gland adenocarcinomas [71]. Among the GSDMs, GSDME exhibits the most consistent anti-tumor functions in the literature and is commonly considered as a potential TSG for several reasons. First, GSDME expression in cancer models (untreated cells) inhibits cell growth and/or provokes cell death [72,73]. Second, diverse anti-oncologic treatments induce GSDME cleavage and enhance cancer death (Table 2; discussed in section 4). Third, GSDME is usually expressed in normal tissues but is frequently downregulated in diverse cancer types, being this alteration sometimes associated with increased malignancy or poor prognosis (Table 1). Fourth, promoter hypermethylation (a common mechanism of silencing TSGs) of GSDME is commonly observed in cancers [72–77]. Besides, GSDME loss-of-function mutations occur in a small proportion of tumors, as an alternative mechanism to methylation for inactivating GSDMEmediated cancer cell death [35]. Finally, the tumor suppressor activity of GSDME is mediated by its cleavage by killer-cell GZMB, resulting in caspase-independent pyroptosis of cancer cells and enhancement of antitumor immunity [35]. Despite these evidences, and the consistent effect of GSDME on tumor sensitization to cancer therapies (Table 2), it is still questionable if GSDME per se generally acts as a bona fide TSG, and whether the evaluation of tumor GSDME protein levels have a reliable prognostic utility (Table 1). For instance, GSDME protein upregulation was observed in the majority in lung tumors and cell lines, and GSDME cleavage (pyroptotically active) was detected even in untreated specimens [78]. Besides, GSDME promoter hypermethylation does not universally correlate with gene silencing in diverse cancer datasets (Table 1) [77]. Interestingly, the gene body of GSDME is frequently hypomethylated in cancer compared to normal tissues, and in fact, the particular patterns of methylation in the GSDME gene promoter and body regions, rather than GSDME expression, can be used as a powerful biomarker for pan-cancer detection [79–81]. Apart from human samples, studies with in vivo animal models reported seemingly contradictory results regarding GSDME role in tumor development and progression. In xenograft cancer models, GSDME does not affect tumor growth of human lung cancer cells [78], but reduces tumor growth of murine melanoma [49], breast, and colon cell lines [35]. Moreover, comparing Gsdme Knock-out (KO) mice with wildtype (WT) animals in two experimental models of intestinal cancer, no clear effects on carcinogenesis, tumor differentiation and progression were evidenced, though an increased tumoral inflammation was observed in WT mice [82]. By contrast, Gsdme KO mice exhibit reduced tumorigenesis in inflammation–mediated colitis-associated colorectal cancer models [83], suggesting that GSDME-mediated pyroptosis and inflammation play a role in cancer initiation. Nonetheless, these conflicting data could be partly explained considering new findings demonstrating that GSDME tumor suppressive effect depends mostly on the subsequent tumor inflammation and immunogenic activation (see section 6). Finally, hitherto, the implication of PVJK is largely unknown. Recently, using ovarian carcinoma expression databases, PJVK was found downregulated in tumors compared with normal samples, and lower PJVK mRNA levels in cancers associated with poor survival rates [53]. 4. GSDMs modulate tumor response to anti-cancer treatments Evading apoptosis is a cancer hallmark that can provide resistance to anti-cancer therapies. In this scenario, activating pyroptosis or other cell death mechanisms could lead to tumor regression. GSDM-mediated cell death, which can proceed in caspase-dependent and independent ways (Fig. 1), occurs in response to multiple cell-damaging stimuli and anticancer treatments (chemotherapy, targeted drugs and immunotherapy) (Table 2). These data would bring to the idea that GSDMD. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 7 Table 2 Effect of GSDMs on anti-cancer therapy response. Tumors /cell lines Treatment Effect on cancer cells Ref GSDMB HER2 Breast tumors, PDXs & cell lines Trastuzumab Higher levels associate with poor treatment response (adjuvant or neoadjuvant contexts) in human tumors. Trastuzumab resistance associates with GSDMB upregulation in human PDXs. GSDMB expression promotes drug survival in SK-BR-3 and HCC1954 cells. [55] HER2 Breast cell lines Trastuzumab GSDMB upregulation increases trastuzumab survival in HCC1954, SK-BR-3, and BT474 cells. This effect can be partially blocked with anti-GSDMB nanotherapy. [61] Colon & Melanoma cell lines Anti-PD1 Exogenous GSDMB over-expression in murine CT26 and B16-F10 cell line xenografts sensitizes to anti-PD1 treatment in vivo. [37] GSDMC Breast cell line Multiple chemo drugs Most drugs upregulate GSDMC but only the antibiotic-type induce caspase-8/GSDMC pyroptosis in MDA-MB-231 cells. [40] Multiple cell lines DMα -KG Induces ROS and activation of DR6/caspase-8 axis leading to GSDMC pyroptosis in multiple human and mouse cancer cell lines in vitro. DMα -KG reduces in vivo tumor growth (Hela and B16 xenografts) and metastasis (B16) and this effect depends on the presence of GSDMC and DR6. [41] GSDMD AML primary cultures and lines Val-boroPro (DPP8/9 inhibitor) Activates CARD8/caspase-1/GSDMD-mediated pyroptosis and efficiently kills AML primary cells and 12/17 AML cell lines in vitro. Halts tumor progression of one PDX and MV4;11 cell line xenografts in vivo. [86] NPC cell lines Taxol Caspase-1 cleaves GSDMD and activates pyroptosis in vitro in HNE-2, 5-8F cell lines. GSDMD silencing does not affect tumor growth in vivo in untreated 5-8F xenografts. GSDMD sensitizes 5-8F xenografts to taxol treatment. [135] Esophageal SCC cell lines Metformin Metformin upregulates miR-497, which in turn downregulates PELP1. PELP1 reduction increases GSDMD pyroptosis in KYSE510 and KYSE140 cells. [133] Ovarian cell lines Alpha-NETA Induces caspase-4 and GSDMD upregulation and cell death in Ho8910PM cells. GSDMD silencing partly increases cell survival to the compound. [136] GSDME (TN) Breast cancer cell lines Tetra‑arsenic hexoxide Induces mitochondrial ROS-mediated caspase-3/GSDME cleavage and pyroptosis in mouse EO771, 4T1 and human Hs578T, MDA-MB-231 cancer cells but not in nontumorigenic cell lines (MCF10A, NMuMG). GSDME silencing reduces drug-induced cell death. [125] (TN) Breast cell lines Cetuximab miR-155-5p antagomir upregulates GSDME and switches apoptosis to pyroptosis after cetuximab treatment in EGF-overexpressing MDA-MB-231 and MDA-MB-468 cells. [112] Colon cell lines Lobaplatin After lobaplatin, caspase-3 cleaves GSDME and promotes pyroptosis via ROS/JNK/bax mitochondrial signaling pathway in HT-29 and HCT116 cells. GSDME silencing switched from pyroptosis to apoptosis but did not affect in vivo tumor growth upon lobaplatin treatment. [90] Colon cell lines TNF +CHX or navitoclax These drugs induce BAK/BAX/Caspase-3 activation and GSDME cleavage and pyroptosis in HCT116 cells. GSDME silencing reduces cell death to treatment. [121] Colon cell lines Ionizing radiation (IR) IR induces GSDME-pyroptosis via the LncRNA NEAT1 and miR488 in HCT116 cells. [113] Gastric cell lines 5-FU Stimulates Caspase-3 and GSDME cleavage and pyroptosis in SGC-7901 and MKN-45 cells. GSDME knock out switches pyroptosis to apoptosis. [137] GBM cell lines Galangin (natural flavonoid) Simultaneously induces autophagy plus caspase-3/GSDME cleavage and pyroptosis in U87MG and U251 cells. GSDME silencing switches pyroptosis to apoptosis. Inhibition of autophagy plus Galangin increases cell death in vivo in U87MG xenografts. [89] Esophageal SCC cell lines Cisplatin and BI2536 The Plk1 inhibitor BI2536 combined with cisplatin provokes caspase-3/GSDME cleavage and pyroptosis in nine SCC cell lines in vitro. The drug combination severely reduces in vivo tumor growth of KYSE150 tumor xenografts. [116] Head & Neck cell lines Triptolide Triptolide activates BAD/BAX-caspase 3-GSDME pyroptosis by repressing mitochondrial associated hexokinase-II and provoking ROS in HK1 and FaDu cells. Combining Triptolide with erastin (SLC7A11 inhibitor) halts in vivo tumor growth in HK1 xenografts. [88] Liver cell lines Miltirone Induces ROS and mitochondrial damage leading to caspase-3/GSDME cleavage and pyroptosis in HepG2 and Hepa1–6 cells in vitro and cell death in vivo using Hepa1–6 tumor xenografts. GSDME silencing switches pyroptosis to apoptosis. [126] Lung cell lines and primary human tumors Trametinib, Erlotinib or Ceritinib GSDME sensitizes to Trametinib, erlotinib and ceritinib targeted therapies in vitro and in vivo using multiple cell lines. GSDME knock out switches pyroptosis to apoptosis but does not affect tumor growth when untreated. GSDME pyroptosis was observed in patients treated primary tumors. [78] Lung cell line Cisplatin or paclitaxel Cisplatin provokes a stronger activation of caspase-3/GSDME cleavage and pyroptosis than paclitaxel in A549 cells. GSDME knockdown inhibits cisplatinbut not paclitaxel-induced pyoptosis in vitro. [138] Lung & neuroblastoma cell lines Dasatinib Dasatinib upregulates GSDMD and GSDME protein levels and provokes caspase-3/GSDME cleavage and pyroptosis in A549 and SH-SY5Y cells in vitro. [139] Melanoma Cell lines Doxorubicin Stimulates autophagy plus caspase-3/GSDME cleavage and pyroptosis in SK-MEL-5, SK-MEL-28, and A-375 cells. Inhibition of eEF-2K decreases autophagy and upregulates sensitivity to doxorubicin and pyroptosis in vitro. [140] Melanoma cell line Iron plus ROS activators Iron upregulates ROS which in turn activates Tom20/bax/caspase-3 and GSDME pyroptosis in A375 cells. GSDME silencing but not GSDMD prevents Iron/CCCP pyroptosis in vitro. Iron supplementation plus sulfasalazine inhibits xenograft tumor growth and metastasis through GSDME-pyroptosis induction. [124] Melanoma cell lines BRAF and MEK inhibitors BRAFi+MEKi activate caspase-3 / GSDME cleavage and pyroptosis in vitro and in vivo in diverse melanoma models. GSDME-pyroptosis stimulates immune anti-tumor response. [85] Multiple cell line types Multiple chemo drugs Chemotherapy drugs provoke caspase-3/GSDME cleavage and pyroptosis in SH-SY5Y, Mewo, Hela, NCI-H522 and EMT6 cancer cells but also in normal keratinocytes, placental epithelial cells and smooth muscle cells. Gsdme KO mice display reduced in vivo toxicity of normal tissues to Cisplatin, 5-FU or bleomycin treatment. [33] AML: Acute myeloid leukemia; DMα -KG: Dimethylα -Ketoglutarate; GBM: Glioblastoma; NPC: nasopharyngeal carcinoma; SCC: Squamous Cell Carcinoma; TN: Triple Negative Breast Cancer. D. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 8 expressing cancer cells may be generally more sensitive to oncologic therapeutic challenge, but the reality is much more complex. First, the precise effect of GSDMs on cell death depends on the balance of different stimuli, signaling pathways, molecular alterations, and cellular contexts. In fact, the co-activation of specific signaling pathways might be required for GSDM pyroptosis. For instance, in HepG2 cells GSDME enhances etoposide cell death only in the presence of WT p53 [84] and GSDMC is upregulated by diverse chemotherapy drugs in MDAMB-231 cells, but only the antibiotic-type (doxorubicin/epirubicin) induce caspase-8/GSDMC-dependent pyroptosis [40]. Second, the GSDMs mediate an extensive crosstalk among diverse cell death mechanisms, and these could cooperate or compete in a biological contextdependent way. For example, inflammasome-triggered caspases (1/4/ 5/11) activate GSDMD pyroptosis but the apoptotic caspases 3/7 inhibits pyroptosis in monocytes [42]. Third, tumor cells could exhibit intrinsic or acquired resistance to pyroptosis. While loss of function mutations in human GSDM genes have been reported only for GSDME so far [35], alterations in upstream signaling pathways or caspases could also occur. Accordingly, GSDME-positive BRAF V600E/K melanomas resistant to BRAF+MEK inhibitors exhibit weak induction of the proapoptotic proteins BIM-EL and BMF, and thus reduced caspase-3 activity [85]. Similarly, 5/17 AML lines were intrinsically resistant to ValboroPro agent due to the lack of pro-caspase-1, which was required for GSDMD-mediated pyroptosis [86]. Apart from these considerations, the general picture shows that the GSDMs are mainly involved in switching apoptosis to lytic and inflammatory cell death processes (pyroptosis and necrosis, but not necroptosis or ferroptosis) [87] after cancer treatment (Table 2). GSDME is the main determinant for this apoptosis-to-pyroptosis switch in multiple in vitro and in vivo models treated not only with a myriad of anti-cancer agents (Table 2) but also with specific natural antitumor products, such as the epoxide triptolide [88] or the flavonol galangin [89]. In clinical specimens, GSDME-pyroptosis was also evidenced (measuring serum LDH concentrations) in lung cancer patients after chemoor EGFR inhibitorbased treatments [78]. Nonetheless, in many studies it is unclear if this apoptosis-pyroptosis switch translates into a bigger cancer killing. For example, in lobaplatin-treated colorectal cancer cells GSDME silencing did not affect tumor response (growth rate) in vivo and in vitro [90], and TRAIL treatment resulted in similar Hela cell death quantities, irrespective of GSDME levels [35]. Like GSDME, GSDMD acts as a therapy sensitizer or mediates apoptosis-pyroptosis switch in particular cancer cell types and oncologic treatments (Table 2), thus, these GSDMs could be important modulators of cancer therapy response. Unfortunately, many of the studies listed in Table 2 were not performed in clinical human specimens, so the impact of GSDM-mediated pyroptosis on cancer patient survival is still to be verified. Contrary to other GSDMs, GSDMB upregulation has been associated with therapy resistance. Specifically, in HER2 breast carcinomas GSDMB renders cancer cells more resistant to the anti-HER2 agents trastuzumab and lapatinib, but no to taxol chemotherapy [55,61], likely in a pyroptosis independent mechanism. For this reason, GSDMB significantly associates with worse breast cancer patients’ prognosis in both adjuvant and neoadjuvant anti-HER2 treatments [55]. Summarizing, the final effect of GSDMs on tumor treatment response depends on multiple factors, including the co-activation of specific signaling pathways, the drug used, cellular context and, as discussed later, the effect on the tumor microenvironment. 5. Targeted activation of GSDM cytotoxicity in tumor cells as novel therapeutic options Diverse oncologic treatments induce GSDM-mediated cell death, thus suggesting that targeted approaches specifically aimed at activating of GSDM cytotoxicity in tumors can be exploited as novel therapeutic options (Fig. 2). To this end, promising therapeutic effects has been recently obtained using two general types of approaches: a) Target activation of the intrinsic GSDM pro-cell death function in cancer cells through nanomedicines; b) Delivering GSDM cytotoxic peptides/ expression constructs into tumors. Within the first type of approaches, three studies targeted GSDMEpyroptosis with nanocarriers that combined agents to induce GSDME transcription (such as the DNA methylation inhibitor decitabine, DCT) with anti-cancer drugs (Fig. 2A). Fan and colleagues [91] used tumortargeting nanoliposomes loaded with cisplatin (LipoDDP) and DCT to trigger GSDME-pyroptosis in 4T1 breast xenografts. The intravenous administration of LipoDDP-DCT upregulated GSDME expression (by inhibiting its gene promoter hypermethylation), induced caspase-3/ GSDME cleavage and the subsequent pyroptosis led to a reduced tumor growth and metastatic spread in vivo. Similarly, Zhao and collaborators [92] designed biomimetic nanoparticles (BNP), by fusing breast cancer membranes onto a poly-lactic-co-glycolic acid (PLGA) core, that were loaded with the photosensitizer indocyanine green (ICG) and DCT. In 4T1 tumor xenografts the “BNP” nanomedicine was activated by low-dose photo-activation (local hyperthermia) and induced GSDME-dependent pyroptosis by the combination of two effects: On one hand, ICG-mediated puncture of cancer cell membranes prompted a sharp increase in cytoplasmic Ca 2+ , provoking cytochrome c release and caspase-3 activation. On the other hand, DCT release up-regulated GSDME transcription. Importantly, this “activatable” therapeutic agent reduced significantly the size of primary tumor and metastasis by activating a strong systemic antitumor immune response. Finally, Hu et al. [93] designed a nano-drug delivery system based on mPEG-PLGA-PLL copolymer that was loaded with arsenic trioxide (As 2 O 3 ) as therapeutic agent for hepatocellular carcinoma (HCC) treatment. Intratumor administration of these nanoparticles (As 2 O 3 -NPs) resulted in caspase3/GSDME cleavage and pyroptosis of Huh7 xenografted tumor cells. Moreover, As 2 O 3 reduced the expression of DNA methyltransferases (Dnmt1, 3a and 3b) which modulate the transcriptional levels of GSDMD or GSDME in a cell line-dependent way. In terms of pre-clinical efficacy, compared to free As 2 O 3 , As 2 O 3 -NPs greatly inhibited tumor growth and showed no obvious systemic toxicity. Instead of using drugs/compounds affecting GSDM expression and/ or activation, Molina-Crespo and collaborators [61] created the first nanotherapy targeting directly a GSDM protein with a therapeutic antibody (Fig. 2B). Specifically, the nanomedicine (termed AbGB-NC) combined a therapeutic anti-GSDMB monoclonal antibody (AbGB) with biocompatible nanocapsules (NCs) functionalized with hyaluronic acid (HA). Since GSDMB upregulation promotes multiple pro-tumor effects in HER2 breast carcinomas [55,56], the authors next validated the therapeutic effect of this nanomedicine in different HER2/GSDMB+ breast cancer models. The results demonstrated that the intracellular released AbGB, but not an irrelevant antibody, was able to impact significantly on tumor behavior in diverse ways: a) decreasing in vitro cell migration; b) increasing the sensitivity to trastuzumab anti-HER2 therapy; c) reducing tumor growth in vivo of orthotopic breast cancer xenografts; and d) limiting lung metastasis colonization in vivo, with no obvious unwanted cytotoxic effects. At the mechanistic level, in vitro tests indicated that the binding of the AbGB to GSDMB (within the CT region and close to the hinge region) could affect the protein configuration thus enhancing the interaction of GSDMB protein with sulfatides. This interaction would subsequently decrease cell migration, and importantly may release the intrinsic pro-cell death activity of GSDMB. In agreement with this, the AbGB-NCs, but no the control treatments, increased the cell death rate (TUNEL assay, not specific test for pyroptosis) of cancer cells in vivo, and specifically of GSDMB-overexpressing tumors (Fig. 2B). Overall, this study proved that GSDMB is a novel therapeutic target for aggressive HER2/GSDMB+breast cancers, and indicated, for the first time, that GSDMB cytotoxic effect and other functions in cancer can be modulated directly with intracellular antibodies. Whereas the above nanotherapies rely on the endogenous expression D. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 9 of the GSDMs by the tumor cells, other therapeutic approaches have been designed for extrinsically delivering GSDM cytotoxic constructs into cancer cells. As a potential treatment for schwannoma tumors Ahmed and collaborators [94] generated an adeno-associated serotype1 virus (AAV1)-based vector that, under the control of a Schwann cellspecific promoter (P0), expresses the pyroptotic NT region of GSDMD (residues 1–276). The intratumoral injection into intra-sciatic nerve of the AAV1-P0-GSDMD-NT in human and syngeneic mouse cells xenograft schwannoma models reduced tumor burden by increasing cell death and reducing cell proliferation. Moreover, this treatment alleviated tumorassociated pain while causing no evident neurologic toxicity. Recently, Wang et al. [95] devised a very clever approach by which the release of the pyroptotic GSDMA3 protein occurs specifically within tumor cells (Fig. 2C). Specifically, the authors designed a complex nanobioorthogonal system based on two interacting components. As first component, the mouse GSDMA3 cleaved protein (NT plus CT domain) was conjugated via a triethylsilyl (TES) ether linker to gold nanoparticles (NP) to generate the NP-GSDMA3, a biocompatible nanosystem that accumulated mostly into tumor tissue of 4T1 cells xenografts. As second component, the authors used the tumor-imaging probe phenylalanine trifuoroborate (Phe-BF 3 ), which was taken up specifically cancer cells. Phe-BF3 desilyated the silyl ether bond of NP-GSDMA3 resulting in the intracellular release of a pytoptotic active GSDMA3 protein and the killing of 4T1 tumor cells in vivo. After three rounds of injection with Phe-BF3 plus NP-GSDMA3 the tumor burden reduced enormously, while the administration of Phe-BF 3 or NP-GSDMA3 alone, or a loss-offunction GSDMA3-NP did not have any effect. No obvious cytotoxic effects were observed in other organs in mice treated Phe-BF 3 plus NPGSDMA3 [95]. Remarkably, some of these GSDM-targeted nanotherapies achieved tumor regression through the activation of a potent anticancer inflammatory reaction (Fig. 2) (discussed below), thus indicating that pyroptosis-inducing nanotherapies are promising approaches that could halt tumor progression simultaneously by attacking tumor cells and enhancing immune response. 6. GSDM-mediated cell death: effects on the tumor microenvironment and the cancer response to immunotherapy GSDM pyroptosis and the immune response are engaged in a complex bidirectional crosstalk. On one hand, cancer cell lysis and the subsequent release of DAMPs and cytokines promotes an immunostimulatory tumor microenvironment that could lead to cancer rejection [87]. On the other hand, cytotoxic T cells and NK cells can directly trigger pyroptosis by cleaving GSDME or GSDMB in cancer cells via perforin-mediated release of GZMB and GZMA, respectively [35,37]. Therefore, pyroptosis and activated immunocytes could generate a positive feedback loop in anticancer immunity (Fig. 2). GSDME-mediated pyroptosis generally associates with a strong immune antitumor response [87]. GSDME upregulation in human cancers correlate positively with immune infiltration [96], and its overexpression in murine breast cancer and melanoma xenografts in immunocompetent mice significantly augmented the number of tumor infiltrating lymphocytes (TILs), mostly CD8+T and NK cells. Activated TIL cytotoxicity and increased phagocytosis by tumor-associated macrophages provoked tumor eradication [35]. Moreover, after therapeutic challenge by BRAF–MEK inhibitors GSDME-mediated pyroptosis in murine melanoma models increased T cell infiltration and improved treatment responses in vivo [85]. Consistent with the beneficial effect of GSDME pyroptosis in cancer treatment, as commented before, the GSDME targeted nanotherapies LipoDDP-DCT and BNP induced a strong immune reaction, including the release of cytokines and proinflammatory molecules, recruitment and activation of cytotoxic T cells, and dendritic cell maturation, that altogether provoked tumor eradication [91,92], Alike GSDME nanotherapies, the efficacy of the NPGSDMA3 +Phe-BF3 nanomedicine depends on the immune anti-tumor reaction. In fact, the pyroptotic killing of a small fraction of tumor cells (20%) was sufficient to activate, via IL1-β secretion, a potent anti-tumor effect (T cells, NK and macrophages) resulting in the elimination of the bulk tumor [95] (Fig. 2C). Remarkably, the GSDME tumor suppressor function in both untreated and therapy-challenged tumors seem to rely on the presence of activated cytotoxic cells. Thus, depletion of NK and CD8+T cells or knocking-out perforin reverses GSDME effects on tumor growth of untreated cancers [35] and depletion of CD4+/CD8+T-cells significantly dampens the response of xenografted melanoma cells to BRAF–MEK inhibitors. Consequently, immunocompetent (e.g., Balb/c) and not immunodeficient (nu/nu) mouse strains should be used to unveil the full effects of GSDME on cancer. In fact, differences in mouse strains may explain in part the conflicting results obtained with GSDME-positive cancer xenografts (discussed in section 2). Of note, while in vitro GSDME activation in cancer cells occurs via caspase-3 (resulting in necrosis secondary to apoptosis), in vivo cleavage by killer-cell GZMA provokes immunogenic cell death (Fig. 2), a form of cell death that is sufficient to activate an adaptive immune response [35,87]. Actually, the “vaccine” inoculation of immunocompetent mice with GSDME-overexpressing cancer cells produces the efficient killing of the secondary tumor implantation. Supporting this idea, GSDMEmediated pyroptosis triggers an immune response against residual cancer, since Gsdme-silenced murine melanomas recurred more frequently than Gsdme-expressing ones after the interruption of BRAFMEK inhibitors therapy [85]. In opposition to the intrinsic tumor suppression mediated by GSDME, GSDMB promotes diverse pro-tumor effects in breast cancer [55,56], and its exogenous expression in murine colon and melanoma cells does not affect tumor growth in immunocompetent mice [37]. In fact, to expose the potential GSDMB antitumor effect, the stimulation of the antitumor immune response by PD-1 immune checkpoint inhibitors is required for efficient GZMA/GSDMB-mediated tumor pyroptosis and regression [37] (Fig. 2B). This shows that triggering GSDMB cytotoxicity within tumors may require additional signals from the tumor microenvironment in order to enhance immune recognition and cancer killing. In this sense, GSDMB expression in cancer cells is upregulated by cytokines (IFNs and TNFα ) produced by TILs [37], and this may contribute to maintaining a positive feedback loop of pyroptosis (Fig. 2B). Moreover, this work indicates that pyroptosis and immunotherapy could synergize to produce a protective immune reaction. Supporting this idea, Wang et al. 2020 proved that the NP-GSDMA3 +Phe-BF 3 therapy sensitized 4T1 tumors to in vivo anti-PD-1 treatment [95] (Fig. 2C). Therefore, pyroptosis-induced inflammation, via GSDM agonists, can cooperate with immune checkpoint inhibitors, thus leading to an improved effectiveness of cancer immunotherapy. Based on these data, the current hypothesis is that GSDM-mediated pyroptosis could convert immunologically “cold” to “hot” tumors, and likely respond better to immune check-point inhibitors. 7. Fine tunning of GSDM-mediated cell death for oncologic treatment: remaining questions Although the data described above reveal a crucial role of GSDMs as novel therapeutic targets and as key determinants of chemotherapeutic and immunotherapy treatments there are still a number of questions to be resolved before GSDM-mediated oncologic therapies became a reality at the clinical level. 7.1. Controlling the specificity, intensity and timing of GSDM-mediated cancer cell death Currently, predicting the final effect of pyroptosis in therapy outcome is intricate. In order to exploit successfully GSDM-pyroptosis in cancer therapy, there are three key parameters that need to be finely controlled (specificity, timing and intensity of cancer lysis and D. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 16 [73] M.S. Kim, X. Chang, K. Yamashita, J.K. Nagpal, J.H. Baek, G. Wu, B. Trink, E. A. Ratovitski, Aberrant promoter methylation and tumor suppressive activity of the DFNA5 gene in colorectal carcinoma, Oncogene 27 (2008) 3624–3634. [74] K. Akino, M. Toyota, H. Suzuki, T. Imai, R. Maruyama, M. Kusano, N. Nishikawa, Y. Watanabe, Y. Sasaki, T. Abe, E. Yamamoto, I. Tarasawa, T. Sonoda, M. Mori, K. Imai, Y. Shinomura, T. Tokino, Identification of DFNA5 as a target of epigenetic inactivation in gastric cancer, Cancer Sci. 98 (2007) 88–95. [75] M.S. Kim, C. Lebron, J.K. Nagpal, Y.K. Chae, X. Chang, Y. Huang, T. Chuang, K. Yamashita, B. Trink, E.A. Ratovitski, J.A. Califano, D. Sidransky, Methylation of the DFNA5 increases risk of lymph node metastasis in human breast cancer, Biochem. Biophys. Res. Commun. 370 (2008) 38–43. [76] L. Croes, K.O. de Beeck, P. Pauwels, W. Vanden Berghe, M. Peeters, E. Fransen, G. Van Camp, DFNA5 promoter methylation a marker for breast tumorigenesis, Oncotarget 8 (2017) 31948–31958. [77] E. De Schutter, L. Croes, J. Ibrahim, P. Pauwels, K. Op de Beeck, P. Vandenabeele, G. Van Camp, GSDME and its role in cancer: from behind the scenes to the front of the stage, Int. J. Cancer 148 (12) (2020) 2872–2883. [78] H. Lu, S. Zhang, J. Wu, M. Chen, M.C. Cai, Y. Fu, W. Li, J. Wang, X. Zhao, Z. Yu, P. Ma, G. Zhuang, Molecular targeted therapies elicit concurrent apoptotic and GSDME-dependent pyroptotic tumor cell death, Clin. Cancer Res. 24 (2018) 6066–6077. [79] L. Croes, M. Beyens, E. Fransen, J. Ibrahim, W. Vanden Berghe, A. Suls, M. Peeters, P. Pauwels, G. Van Camp, K. Op de Beeck, Large-scale analysis of DFNA5 methylation reveals its potential as biomarker for breast cancer, Clin. Epigenetics 10 (2018) 51. [80] J. Ibrahim, K.O. de Beeck, E. Fransen, M. Peeters, G. Van Camp, The Gasdermin e gene potential as a pan-cancer biomarker, while discriminating between different tumor types, Cancers (Basel) 11 (11) (2019) 1810. [81] J. Ibrahim, K. Op de Beeck, E. Fransen, L. Croes, M. Beyens, A. Suls, W. Vanden Berghe, M. Peeters, G. Camp, Methylation analysis of gasdermin E shows great promise as a biomarker for colorectal cancer, Cancer Med. 8 (2019) 2133–2145. [82] L. Croes, E. Fransen, M. Hylebos, K. Buys, C. Hermans, G. Broeckx, M. Peeters, P. Pauwels, K. Op de Beeck, G. Van Camp, Determination of the potential tumorsuppressive effects of Gsdme in a chemically induced and in a genetically modified intestinal cancer mouse model, Cancers (Basel) 11 (8) (2019), 1214. [83] G. Tan, C. Huang, J. Chen, F. Zhi, HMGB1 released from GSDME-mediated pyroptotic epithelial cells participates in the tumorigenesis of colitis-associated colorectal cancer through the ERK1/2 pathway, J. Hematol. Oncol. 13 (1) (2020) 149. [84] Y. Masuda, M. Futamura, H. Kamino, Y. Nakamura, N. Kitamura, S. Ohnishi, Y. Miyamoto, H. Ichikawa, T. Ohta, M. Ohki, T. Kiyono, H. Egami, H. Baba, H. Arakawa, The potential role of DFNA5, a hearing impairment gene, in p53mediated cellular response to DNA damage, J. Hum. Genet. 51 (2006) 652–664. [85] D.A. Erkes, W. Cai, I.M. Sanchez, T.J. Purwin, C. Rogers, C.O. Field, A.C. Berger, E.J. Hartsough, U. Rodeck, E.S. Alnemri, A.E. Aplin, Mutant BRAF and MEK inhibitors regulate the tumor immune microenvironment via pyroptosis, Cancer Discov. 10 (2020) 255–269. [86] D.C. Johnson, C.Y. Taabazuing, M.C. Okondo, A.J. Chui, S.D. Rao, F.C. Brown, C. Reed, E. Peguero, E. de Stanchina, A. Kentsis, D.A. Bachovchin, DPP8/DPP9 inhibitor-induced pyroptosis for treatment of acute myeloid leukemia, Nat. Med. 24 (2018) 1151–1156. [87] K. Tsuchiya, Switching from apoptosis to pyroptosis: gasdermin-elicited inflammation and antitumor immunity, Int. J. Mol. Sci. 22 (1) (2021) 426. [88] J. Cai, M. Yi, Y. Tan, X. Li, G. Li, Z. Zeng, W. Xiong, B. Xiang, Natural product triptolide induces GSDME-mediated pyroptosis in head and neck cancer through suppressing mitochondrial hexokinase-ІІ, J. Exp. Clin. Cancer Res. 40 (1) (2021) 190. [89] Y. Kong, Z. Feng, A. Chen, Q. Qi, M. Han, S. Wang, Y. Zhang, X. Zhang, N. Yang, J. Wang, B. Huang, Q. Zhang, G. Xiang, W. Li, D. Zhang, J. Wang, X. Li, The natural flavonoid galangin elicits apoptosis, pyroptosis, and autophagy in glioblastoma, Front. Oncol. 9 (2019) 942. [90] J. Yu, S. Li, J. Qi, Z. Chen, Y. Wu, J. Guo, K. Wang, X. Sun, J. Zheng, Cleavage of GSDME by caspase-3 determines lobaplatin-induced pyroptosis in colon cancer cells, Cell Death Dis. 10 (3) (2019) 193. [91] J.X. Fan, R.H. Deng, H. Wang, X.H. Liu, X.N. Wang, R. Qin, X. Jin, T.R. Lei, D. Zheng, P.H. Zhou, Y. Sun, X.Z. Zhang, Epigenetics-based tumor cells pyroptosis for enhancing the immunological effect of chemotherapeutic nanocarriers, Nano Lett. 19 (2019) 8049–8058. [92] P. Zhao, M. Wang, M. Chen, Z. Chen, X. Peng, F. Zhou, J. Song, J. Qu, Programming cell pyroptosis with biomimetic nanoparticles for solid tumor immunotherapy, Biomaterials 254 (2020), 120142. [93] J. Hu, Y. Dong, L. Ding, Y. Dong, Z. Wu, W. Wang, M. Shen, Y. Duan, Local delivery of arsenic trioxide nanoparticles for hepatocellular carcinoma treatment, Signal Transduct. Target. Ther. 4 (2019) 28. [94] S.G. Ahmed, A. Abdelanabi, M. Doha, G.J. Brenner, Schwannoma gene therapy by adeno-associated virus delivery of the pore-forming protein gasdermin-D, Cancer Gene Ther. 26 (2019) 259–267. [95] Q. Wang, Y. Wang, J. Ding, C. Wang, X. Zhou, W. Gao, H. Huang, F. Shao, Z. Liu, A bioorthogonal system reveals antitumour immune function of pyroptosis, Nature 579 (2020) 421–426. [96] G. Stoll, Y. Ma, H. Yang, O. Kepp, L. Zitvogel, G. Kroemer, Pro-necrotic molecules impact local immunosurveillance in human breast cancer, Oncoimmunology 6 (2017), e1299302. [97] Z. Huang, Q. Zhang, Y. Wang, R. Chen, Y. Wang, Z. Huang, Inhibition of caspase3-mediated GSDME-derived pyroptosis aids in noncancerous tissue protection of squamous cell carcinoma patients during cisplatin-based chemotherapy, Am. J. Cancer Res. 10 (2020) 4287–4307. [98] S. Jiang, H. Zhang, X. Li, B. Yi, L. Huang, Z. Hu, A. Li, J. Du, Y. Li, W. Zhang, Vitamin D/VDR attenuate cisplatin-induced AKI by down-regulating NLRP3/ Caspase-1/GSDMD pyroptosis pathway, J. Steroid Biochem. Mol. Biol. 206 (2021), 105789. [99] Y. Liu, Y. Fang, X. Chen, Z. Wang, X. Liang, T. Zhang, M. Liu, N. Zhou, J. Lv, K. Tang, J. Xie, Y. Gao, F. Cheng, Y. Zhou, Z. Zhang, Y. Hu, X. Zhang, Q. Gao, Y. Zhang, B. Huang, Gasdermin E-mediated target cell pyroptosis by CAR T cells triggers cytokine release syndrome, Sci. Immunol. 5 (2020) eaax7969. [100] F.A. Fischer, K.W. Chen, J.S. Bezbradica, Posttranslational and therapeutic control of gasdermin-mediated pyroptosis and inflammation, Front. Immunol. 12 (2021), 661162. [101] C.B. Ryder, H.C. Kondolf, M.E. O’Keefe, B. Zhou, D.W. Abbott, Chemical modulation of gasdermin-mediated pyroptosis and therapeutic potential, J. Mol. Biol. 167183 (2021). [102] S. Moussette, A. Al Tuwaijri, H.-R. Kohan-Ghadr, S. Elzein, R. Farias, J. B´ erub´ e, B. Ho, C. Laprise, C.G. Goodyer, S. Rousseau, A.K. Naumova, Role of DNA methylation in expression control of the IKZF3-GSDMA region in human epithelial cells, PLoS One. 12 (2017), e0172707. [103] J.S. Muhammad, M.N. Jayakumar, N.M. Elemam, T. Venkatachalam, T.K. Raju, R. A. Hamoudi, A.A. Maghazachi, Gasdermin D hypermethylation inhibits pyroptosis and LPS-induced IL-1β release from NK92 cells, ImmunoTargets Ther. 8 (2019) 29–41. [104] M.M. Stein, E.E. Thompson, N. Schoettler, B.A. Helling, K.M. Magnaye, C. Stanhope, C. Igartua, A. Morin, C. Washington, D. Nicolae, K. Bønnelykke, C. Ober, A decade of research on the 17q12-21 asthma locus: piecing together the puzzle, J. Allergy Clin. Immunol. 142 (2018) 749–764.e3. [105] N. Saeki, R. Komatsuzaki, F. Chiwaki, K. Yanagihara, H. Sasaki, A GSDMB enhancer-driven HSV thymidine kinase-expressing vector for controlling occult peritoneal dissemination of gastric cancer cells, BMC Cancer 15 (2015) 439. [106] N. Kusumaningrum, D.H. Lee, H.S. Yoon, Y.K. Kim, C.H. Park, J.H. Chung, Gasdermin C is induced by ultraviolet light and contributes to MMP-1 expression via activation of ERK and JNK pathways, J. Dermatol. Sci. 90 (2018) 180–189. [107] R. Xi, J. Montague, X. Lin, C. Lu, W. Lei, K. Tanaka, Y.V. Zhang, X. Xu, X. Zheng, X. Zhou, J.F. Urban, K. Iwatsuki, R.F. Margolskee, I. Matsumoto, M. Tizzano, J. Li, P. Jiang, Up-regulation of gasdermin C in mouse small intestine is associated with lytic cell death in enterocytes in worm-induced type 2 immunity, Proc. Natl. Acad. Sci. U. S. A. 118 (2021), e2026307118. [108] N. Kayagaki, B.L. Lee, I.B. Stowe, O.S. Kornfeld, K. O’Rourke, K.M. Mirrashidi, B. Haley, C. Watanabe, M. Roose-Girma, Z. Modrusan, S. Kummerfeld, R. Reja, Y. Zhang, V. Cho, T. Daniel Andrews, L.X. Morris, C.C. Goodnow, E.M. Bertram, V. M. Dixit, IRF2 transcriptionally induces GSDMD expression for pyroptosis, Sci. Signal. 12 (2019) eaax4917. [109] Z. Liu, L. Gan, Y. Xu, D. Luo, Q. Ren, S. Wu, C. Sun, Melatonin alleviates inflammasome-induced pyroptosis through inhibiting NF-κB/GSDMD signal in mice adipose tissue, J. Pineal Res. 63 (1) (2017), e12414. [110] Q. Cao, D. Feng, J. He, L. Zhou, Z. Fan, Y. Chen, X. Chen, R. Jin, G. Zhou, Involvement of TFAP2A in the activation of GSDMD gene promoter in hyperoxiainduced ALI, Exp. Cell Res. 401 (2021), 112521. [111] M.S. Webb, A.L. Miller, E.B. Thompson, In CEM cells the autosomal deafness gene dfna5 is regulated by glucocorticoids and forskolin, J. Steroid Biochem. Mol. Biol. 107 (2007) 15–21. [112] W. Xu, C. Song, X. Wang, Y. Li, X. Bai, X. Liang, J. Wu, J. Liu, Downregulation of miR-155-5p enhances the anti-tumor effect of cetuximab on triple-negative breast cancer cells via inducing cell apoptosis and pyroptosis, Aging (Albany NY) 13 (2021) 228–240. [113] F. Su, J. Duan, J. Zhu, H. Fu, X. Zheng, C. Ge, Long non-coding RNA nuclear paraspeckle assembly transcript 1 regulates ionizing radiation-induced pyroptosis via microRNA-448/gasdermin E in colorectal cancer cells, Int. J. Oncol. 59 (4) (2021) 79. [114] T. Fujikane, N. Nishikawa, M. Toyota, H. Suzuki, M. Nojima, R. Maruyama, M. Ashida, M. Ohe-Toyota, M. Kai, T. Nishidate, Y. Sasaki, T. Ohmura, K. Hirata, T. Tokino, Genomic screening for genes upregulated by demethylation revealed novel targets of epigenetic silencing in breast cancer, Breast Cancer Res. Treat. 122 (2010) 699–710. [115] S. Delmaghani, J. Defourny, A. Aghaie, M. Beurg, D. Dulon, N. Thelen, I. Perfettini, T. Zelles, M. Aller, A. Meyer, A. Emptoz, F. Giraudet, M. Leibovici, S. Dartevelle, G. Soubigou, M. Thiry, E.S. Vizi, S. Safieddine, J.-P. Hardelin, P. Avan, C. Petit, Hypervulnerability to sound exposure through impaired adaptive proliferation of peroxisomes, Cell 163 (4) (2015) 894–906. [116] M. Wu, Y. Wang, D. Yang, Y. Gong, F. Rao, R. Liu, Y. Danna, J. Li, J. Fan, J. Chen, W. Zhang, Q. Zhan, A PLK1 kinase inhibitor enhances the chemosensitivity of cisplatin by inducing pyroptosis in oesophageal squamous cell carcinoma, EBioMedicine. 41 (2019) 244–255. [117] F. Humphries, L. Shmuel-Galia, N. Ketelut-Carneiro, S. Li, B. Wang, V. V. Nemmara, R. Wilson, Z. Jiang, F. Khalighinejad, K. Muneeruddin, S.A. Shaffer, R. Dutta, C. Ionete, S. Pesiridis, S. Yang, P.R. Thompson, K.A. Fitzgerald, Succination inactivates gasdermin D and blocks pyroptosis, Science 369 (2020) 1633–1637. [118] J.K. Rathkey, J. Zhao, Z. Liu, Y. Chen, J. Yang, H.C. Kondolf, B.L. Benson, S. M. Chirieleison, A.Y. Huang, G.R. Dubyak, T.S. Xiao, X. Li, D.W. Abbott, Chemical disruption of the pyroptotic pore-forming protein gasdermin D inhibits inflammatory cell death and sepsis, Sci. Immunol. 3 (2018) eaat2738. [119] J.J. Hu, X. Liu, S. Xia, Z. Zhang, Y. Zhang, J. Zhao, J. Ruan, X. Luo, X. Lou, Y. Bai, J. Wang, L.R. Hollingsworth, V.G. Magupalli, L. Zhao, H.R. Luo, J. Kim, D. Sarri´ o et al. BBA - Reviews on Cancer 1876 (2021) 188635 17 J. Lieberman, H. Wu, FDA-approved disulfiram inhibits pyroptosis by blocking gasdermin D pore formation, Nat. Immunol. 21 (2020) 736–745. [120] Y. Wang, P. Shi, Q. Chen, Z. Huang, D. Zou, J. Zhang, X. Gao, Z. Lin, Mitochondrial ROS promote macrophage pyroptosis by inducing GSDMD oxidation, J. Mol. Cell Biol. 11 (2019) 1069–1082. [121] L. Hu, M. Chen, X. Chen, C. Zhao, Z. Fang, H. Wang, H. Dai, Chemotherapyinduced pyroptosis is mediated by BAK/BAX-caspase-3-GSDME pathway and inhibited by 2-bromopalmitate, Cell Death Dis. 11 (4) (2020) 281. [122] G. Luchetti, J.L. Roncaioli, R.A. Chavez, A.F. Schubert, E.M. Kofoed, R. Reja, T. K. Cheung, Y. Liang, J.D. Webster, I. Lehoux, E. Skippington, J. Reeder, B. Haley, M.W. Tan, C.M. Rose, K. Newton, N. Kayagaki, R.E. Vance, V.M. Dixit, Shigella ubiquitin ligase IpaH7.8 targets gasdermin D for degradation to prevent pyroptosis and enable infection, Cell Host Microbe S1931–3128 (21) (2021), 00386-00383. [123] C.L. Evavold, I. Hafner-Bratkoviˇ c, P. Devant, J.M. D’Andrea, E.M. Ngwa, E. Borˇ si´ c, J.G. Doench, M.W. LaFleur, A.H. Sharpe, J.R. Thiagarajah, J.C. Kagan, Control of gasdermin D oligomerization and pyroptosis by the Ragulator-RagmTORC1 pathway, Cell 184 (2021), 4495-4511.e19. [124] B. Zhou, J.Yuan Zhang, H. zi Chen, Y. li Ai, K. Cheng, R.Yue Sun, D. Zhou, J. Han, Q. Wu, Tom20 senses iron-activated ROS signaling to promote melanoma cell pyroptosis, Cell Res. 28 (2018) 1171–1185. [125] H. An, J.S. Heo, P. Kim, Z. Lian, S. Lee, J. Park, E. Hong, K. Pang, Y. Park, A. Ooshima, J. Lee, M. Son, H. Park, Z. Wu, K.S. Park, S.J. Kim, I. Bae, K.M. Yang, Tetraarsenic hexoxide enhances generation of mitochondrial ROS to promote pyroptosis by inducing the activation of caspase-3/GSDME in triple-negative breast cancer cells, Cell Death Dis. 12 (2) (2021) 159. [126] X. Zhang, P. Zhang, L. An, N. Sun, L. Peng, W. Tang, D. Ma, J. Chen, Miltirone induces cell death in hepatocellular carcinoma cell through GSDME-dependent pyroptosis, Acta Pharm. Sin. B 10 (2020) 1397–1413. [127] X. Shen, H. Wang, C. Weng, H. Jiang, J. Chen, Caspase 3/GSDME-dependent pyroptosis contributes to chemotherapy drug-induced nephrotoxicity, Cell Death Dis. 12 (2) (2021) 186. [128] J. Lachner, V. Mlitz, E. Tschachler, L. Eckhart, Epidermal cornification is preceded by the expression of a keratinocyte-specific set of pyroptosis-related genes, Sci. Rep. 7 (1) (2017) 17446. [129] Y. Chen, T. Xiao, S. Xu, H. Gu, M. Li, X. Chen, Ultraviolet B induces proteolytic cleavage of the pyroptosis inducer gasdermin E in keratinocytes, J. Dermatol. Sci. 100 (2020) 160–163. [130] S. Grossi, G. Fenini, T. Kockmann, P. Hennig, M. Di Filippo, H.D. Beer, Inactivation of the cytoprotective major vault protein by Caspase-1 and -9 in epithelial cells during apoptosis, J. Invest. Dermatol. 140 (2020) 1335–1345.e10. [131] M. Karmakar, M. Minns, E.N. Greenberg, J. Diaz-Aponte, K. Pestonjamasp, J. L. Johnson, J.K. Rathkey, D.W. Abbott, K. Wang, F. Shao, S.D. Catz, G.R. Dubyak, E. Pearlman, N-GSDMD trafficking to neutrophil organelles facilitates IL-1β release independently of plasma membrane pores and pyroptosis, Nat. Commun. 11 (2020) 2212. [132] J. Defourny, A. Aghaie, I. Perfettini, P. Avan, S. Delmaghani, C. Petit, Pejvakinmediated pexophagy protects auditory hair cells against noise-induced damage, Proc. Natl. Acad. Sci. U. S. A. 116 (2019) 8010–8017. [133] L. Wang, K. Li, X. Lin, Z. Yao, S. Wang, X. Xiong, Z. Ning, J. Wang, X. Xu, Y. Jiang, D. Liu, Y. Chen, D. Zhang, H. Zhang, Metformin induces human esophageal carcinoma cell pyroptosis by targeting the miR-497/PELP1 axis, Cancer Lett. 450 (2019) 22–31. [134] R. Lin, H. Wei, S. Wang, Z. Huang, H. Chen, S. Zhang, J. Lin 13 (2020) 3149–3157. [135] X. Wang, H. Li, W. Li, J. Xie, F. Wang, X. Peng, Y. Song, G. Tan, The role of Caspase-1/GSDMD-mediated pyroptosis in taxol-induced cell death and a taxolresistant phenotype in nasopharyngeal carcinoma regulated by autophagy, Cell Biol. Toxicol. 36 (2020) 437–457. [136] L. Qiao, X. Wu, J. Zhang, L. Liu, X. Sui, R. Zhang, W. Liu, F. Shen, Y. Sun, X. Xi, α -NETA induces pyroptosis of epithelial ovarian cancer cells through the GSDMD/ caspase-4 pathway, FASEB J. 33 (2019) 12760–12767. [137] Y. Wang, B. Yin, D. Li, G. Wang, X. Han, X. Sun, GSDME mediates caspase-3dependent pyroptosis in gastric cancer, Biochem. Biophys. Res. Commun. 495 (2018) 1418–1425. [138] C.Cheng Zhang, C.Guang Li, Y.Feng Wang, L.Hui Xu, X.Hui He, Q.Zhen Zeng, C. Ying Zeng, F.Yi Mai, B. Hu, D.Yun Ouyang, Chemotherapeutic paclitaxel and cisplatin differentially induce pyroptosis in A549 lung cancer cells via caspase-3/ GSDME activation, Apoptosis 24 (2019) 312–325. [139] J. Zhang, Y. Chen, Q. He, Distinct characteristics of dasatinib-induced pyroptosis in gasdermin E-expressing human lung cancer A549 cells and neuroblastoma SHSY5Y cells, Oncol. Lett. 20 (2020) 145–154. [140] P. Yu, H.Yan Wang, M. Tian, A.Xue Li, X.Sha Chen, Y. Zhang, Y. Cheng, X. Luan Wang, Eukaryotic elongation factor-2 kinase regulates the cross-talk between autophagy and pyroptosis in doxorubicin-treated human melanoma cells in vitro, Acta Pharmacol. Sin 40 (2019) 1237–1244. D. Sarri´ o et al.