Epigenetic regulation of ASC/TMS1 expression: potential role in apoptosis and inflammasome function

Abstract

Cloning studies have revealed that the apoptosis-associated speck-like protein possessing a caspase-recruiting domain (ASC) and the target of methylation-induced silencing-1 (TMS) are identical proteins. ASC/TMS1 is a bipartite adaptor protein containing the N-terminal pyrin domain and the C-terminal caspase-recruitment domain. There is abundant literature on ASC/TMS1, mostly under the name TMS1, in the epigenetic regulation of apoptosis and carcinogenesis, whereas the abbreviation ASC has been adopted from studies on the assembly of inflammasomes and stimulation of inflammation. There is substantial literature emphasizing that there are common aspects in the regulation of apoptosis and inflammation, which may be related to the function of ASC/TMS1. The region of the transcription start site of ASC/TMS1 gene contains a 600-bp-long CpG island that is highly methylated and the transcription of ASC/TMS1 is repressed in several cancers. However, it is not known whether the ASC/TMS1-dependent epigenetic regulation controls the inflammasome functions and moreover whether this regulation has any role in the inflammation-mediated carcinogenesis or in the pathogenesis of age-related degenerative diseases. We will examine the mechanisms involved in the epigenetic regulation of ASC/TMS1 as well as their significance in the coordination of apoptosis and inflammasome functions. We will also review the role of aberrant methylation of ASC/TMS1 promoter in the function of inflammasomes, a major host defense system, in cellular housekeeping and carcinogenesis.

Introduction

Apoptosis is a programmed cell death mechanism that is not only activated in tissue morphogenesis during development but also in response to a variety of acute injuries and chronic degenerative diseases [1, 2]. Conversely, the repression of apoptosis can enhance carcinogenesis [3]. Recent studies have revealed that epigenetic regulation has a fundamental role in both developmental processes and cancer [4]. Apoptosis-related genes are crucial targets in the epigenetic silencing in carcinogenesis [5]. Currently, there is convincing evidence that inflammation is closely linked to carcinogenesis [6, 7]. Interestingly, both apoptosis and inflammation can be viewed as housekeeping mechanisms that maintain tissue homeostasis combating cellular damage. Inflammasomes are pattern recognition sensors for cellular stress and microbial attacks; in fact, they induce inflammatory responses that protect tissues against excessive damage [8, 9]. Since both apoptosis and inflammasomes are guardians of tissue integrity, it is not surprising that they share some regulatory mechanisms.

In 1999, Masumoto et al. [10] cloned a novel gene translating a 22-kDa protein that formed cellular aggregates, also called specks, during apoptosis induced in HL-60 cells by retinoic acid and etoposide. They termed this gene apoptosis-associated speck-like protein containing a caspase-recruiting domain (ASC). Independently, Conway et al. [11] were investigating the genes silenced by the overexpression of DNA methyltransferase, which could induce the aberrant DNA methylation in CpG islands, a feature commonly observed in cancer cells. They cloned the gene which they termed the target of methylation-induced silencing-1 (TMS1). Interestingly, the ASC and TMS1 turned out to be an identical gene, which was later also termed as PYCARD since it is a bipartite protein containing both the N-terminal pyrin domain (PYD) and the C-terminal caspase-recruitment domain (CARD) [8, 9, 12]. ASC/TMS1 is an important adaptor protein interacting with caspases via the CARD domain and inflammasome receptors via the PYD domain. Unfortunately, there is no consensus in the literature about the name; TMS1 has commonly been applied in the cancer studies on the epigenetic regulation of apoptosis, whereas ASC is used in the studies on inflammasome assembly and pyroptosis. In this review, we will examine the significance of epigenetic regulation of ASC/TMS1 adaptor in the coordinated functions of apoptosis and inflammasomes. Moreover, we will review evidence indicating that the aberrant methylation of CpG island of ASC/TMS1 might suppress the function of inflammasomes in cancer cells and thus disturb housekeeping and promote carcinogenesis.

Biology of ASC/TMS1

Epigenetic regulation of ASC/TMS1 expression

The epigenetic regulation of gene expression is controlled by DNA and histone methylation [13, 14]. This methylomics organizes the chromatin structure; open chromatin activates transcription, whereas closed chromatin silences gene expression. CpG islands are specific genomic loci that are rich in CpG sites, locating especially in the promoters of housekeeping genes [15]. CpG islands are normally unmethylated targeting transcription factors that stimulate gene transcription. There are several mechanisms that can maintain CpG islands unmethylated or enhance their methylation by DNA methyltransferases [14–16]. Moreover, there is a close crosstalk between histone methylation and DNA methylation in the control of gene transcription in a distinct DNA locus [13, 16].

The cloning experiments revealed that there is a 600-bp-long CpG island in the region of transcription start site of the ASC/TMS1 gene [17]. Several studies have demonstrated that the methylation status of this CpG island correlates with the expression level of ASC/TMS1, i.e., the unmethylated CpG locus represents the active transcription of ASC/TMS1, whereas hypermethylated locus denotes an inactive state, as is observed in many cancer cells [11, 17, 18]. The group of Prof. Paula Vertino has probed the role of epigenetic regulation in the expression of ASC/TMS1 gene [19–21]. In the control cells expressing ASC/TMS1, the unmethylated CpG island is surrounded by the flanking regions, which are extensively methylated. In the active state, the unmethylated CpG island contains three DNase I-hypersensitive sites but no nucleosomes [17]. Lucas et al. [20] demonstrated that GA-binding proteins (GABPα and GABPβ) assembled heteromeric complexes in the unmethylated CpG region. They also observed that the occupancy of GABP complexes prevented the methylation of the CpG island and thus enhanced the expression of ASC/TMS1 gene. The 5′ end flanking domain of CpG region was marked by histone H3K4 methylation and positioned with the nucleosomes [19]. Moreover, this flanking site contained a dense acetylation of H4K16, which was induced by the histone acetyltransferase, hMOF. Recently, Kapoor-Vazirani et al. [21] demonstrated that the methyltransferase SUV420H2 trimethylated H4K20, which induced the promoter-proximal pausing of RNA polymerase II (Pol II), whereas the hMOF-mediated H4K16 acetylation activated the transcription by releasing the pausing of Pol II. The promoter-proximal pausing of Pol II is a common mechanism to control the early elongation, particularly at genes under signal-responsive regulation [22]. These studies indicated that the hMOF-induced H4K16 acetylation and SUV420H2-mediated H4K20 trimethylation exert antagonistic roles in the control of ASC/TMS1 transcription.

The aberrant methylation of CpG island of ASC/TMS1 in cancer cells was associated with (a) a decrease in the level of methylated H3K4 and acetylated H4K16, (b) an increase in the methylation of H3K9, and (c) a loss of nucleosome positioning [19]. Moreover, the methylation of CpG island repressed the recruitment of Pol II to the promoter of ASC/TMS1 and thus silenced transcription [21]. It is well known that the methylation of H3K4 in gene promoter regions is an epigenetic mark for transcriptional activation, whereas the methylation of H3K9 and H4K20 are common repressive sites [23]. In particular, the unmodified H3K4 can bind DNA methyltransferase (DNMT3), which can enhance the methylation of CpG sites, whereas methylated H3K4 can interact with proteins preventing the methylation of CpG island [16]. Moreover, the methylation of H3K9 is commonly linked to the methylation of nearby CpG sites [16, 24]. On the other hand, Kagey et al. [25] demonstrated that the transient exposure to 5-aza-2′-deoxycytidine, a common DNA methyltransferase inhibitor, enhanced the demethylation of CpG island, which was associated with a reduced level of repressive chromatin marks, such as H3K9me2 and H4K20me3, and increased the recruitment of Pol II to the ASC/TMS1 locus [25]. All these studies clearly indicated that the transcription of ASC/TMS1 gene is under epigenetic control.

Role of ASC/TMS1 in apoptosis and carcinogenesis

Cloning studies already revealed that the ASC/TMS1 gene regulates apoptosis via the CARD domain [10, 11]. Subsequently, a plethora of clinical studies have demonstrated that the CpG island is aberrantly hypermethylated in several cancers implying that ASC/TMS1 gene is a tumor suppressor gene [26]. For instance, there are reports that the ASC/TMS1 gene is hypermethylated in breast and lung cancers [11, 27], melanoma [28], prostate cancer [29, 30], glioblastoma multiforme [31, 32], and neuroblastoma [33]. The increased methylation level is commonly associated with the decreased expression of the ASC/TMS1 gene. Moreover, the exposure of cancer cells to DNA methyltransferase inhibitors, such as 5-aza-2′-deoxycytidine and zebularine, could restore the expression of ASC/TMS1 and induce apoptosis [18, 29–31]. In addition, two histone deacetylase inhibitors, trichostatin A and sodium butyrate, were also able to re-establish the expression of ASC/TMS1 in cancer cells [34, 35]. The methylation status of the CpG island of ASC/TMS gene is generally a sensitive prognostic marker for tumorigenesis. Given that ASC/TMS1 is a proapoptotic protein, the epigenetic inhibition of ASC/TMS1 gene in cancer cells renders them resistant to apoptotic cell death. However, ASC/TMS1 knockout mice have not revealed spontaneous tumorigenesis although they lack the expression of ASC/TMS1. Interestingly, Drexler et al. [36] demonstrated that the conditional knockout of ASC/TMS1 in keratinocytes or bone marrow-derived macrophages provoked the DMBA/TPA-induced tumorigenesis in keratinocytes but suppressed tumor formation in myeloid cells. They speculated that the tissue-specific responses could explain the absence of tumors in ASC/TMS1 knockout mice.

Currently, the exact mechanism through which ASC/TMS1 induces apoptosis is still unclear. Ohtsuka et al. [37] demonstrated that ASC/TMS1 could interact with BAX, a pro-apoptotic protein that induced apoptosis via the mitochondrial pathway (Fig. 1). They observed that BAX protein could bind to the PYD domain of ASC/TMS1. Interestingly, they revealed that ASC/TMS1 acted as a carrier protein of BAX by translocating it to mitochondria. However, a recent study has claimed that the BAX-dependent apoptosis could be linked to the activation of caspase-8 [38]. It was also demonstrated that ASC/TMS1 can interact with caspase-8 [39], in addition to the well-known inflammatory caspase-1 and caspase-5 [40, 41]. Masumoto et al. [39] observed that ASC/TMS1 interacted with caspase-8 in the ternary complex containing also NOD-like receptor NLRC4. The lack of caspase-8 rescued fibroblasts from the ASC/TMS1-induced apoptosis which indicated that caspase-8, an initiator caspase, probably could stimulate effector caspase-mediated apoptosis. Hasegawa et al. [38] reported that the ASC/TMS1-mediated apoptosis was also activated by the caspase-8-induced proteolytic maturation of BID protein. The pro-apoptotic BID could trigger the death receptor-activated, mitochondria-mediated apoptotic pathway [42]. However, there does seem to be cell-type-specific differences in caspase-8-linked, ASC/TMS1-enhanced apoptosis [38, 43]. Recently, Drexler et al. [36] demonstrated that ASC/TMS1 could display cell-type-specific functions, i.e., it has a proinflammatory role in myeloid cells whereas in epidermal keratinocytes, ASC/TMS1 acts as a tumor suppressor, probably in association with p53 activation (Fig. 1). These workers reported that UVB treatment of keratinocytes promoted a transient interaction between ASC/TMS1 and p53. The lack of ASC/TMS1 reduced the phosphorylation of p53 and subsequently decreased the expression of its target genes. It is well known that the p53 protein is a potent inducer of apoptosis [44] and thus ASC/TMS1 could potentiate the p53-dependent apoptosis. These studies imply that the reduced expression of ASC/TMS1 in cancer cells can block apoptotic cell death and enhance carcinogenesis.

Fig. 1.

Epigenetic regulation of ASC/TMS1 expression affects its interaction with inflammasome receptors and apoptosis-inducing proteins. ASC/TMS1 assembles the inflammasome complexes with AIM2, NLRP3, NLRC4, and NLRP1 receptors. These complexes activate caspase-1, which subsequently stimulates inflammatory responses. ASC/TMS1 inhibits RIP2 kinase, which context-dependently can control apoptosis and inflammation via the activation of NF-κB and caspase-1. In addition, ASC/TMS1 can interact with apoptosis-inducing BAX, BID, p53, and caspase-8 proteins and directly enhance apoptosis. COP and POP proteins inhibit the function of ASC/TMS1

Recently, Liu et al. [45] demonstrated that the expression of ASC/TMS1 inhibited NF-κB activity in primary melanoma cells, whereas the forced expression of ASC/TMS1 in metastatic melanoma, cells that are normally deficient in ASC/TMS1, enhanced NF-κB activity and potentiated tumorigenesis. ASC/TMS1 has several interacting proteins containing either CARD or PYD domains [26]. For instance, receptor interacting protein-2 (RIP2) contains a CARD domain [46], as well as caspase-1 [40, 41]. It is known that RIP2 is a potent enhancer of NF-κB activation [47]. Lamkanfi et al. [48] demonstrated that caspase-1 interacted with RIP2 and RIP-2-dependently it activated NF-κB. This activation was not dependent on the activity of caspase-1. Later, Sarkar et al. [49] observed that ASC/TMS1 is an inhibitor of RIP2-mediated NF-κB activation in HEK293 cells (Fig. 1). They also reported that the suppression of ASC/TMS1 expression dose-dependently enhanced NF-κB activation but simultaneously reduced IL-1β processing, evidence for the inhibition of caspase-1-dependent inflammasomes. A recent study revealed that ASC/TMS1 could context-dependently regulate the formation of inflammasome complexes or RIP2-dependent NF-κB activation [50]. Moreover, Hasegawa et al. [51] demonstrated that caspase-8 was also able to control the ASC/TMS1-mediated NF-κB activation. In conclusion, it seems that the balance between the ASC/TMS1-mediated apoptotic and inflammatory responses is under the context-dependent regulation of the complex formation between CARD and PYD domain-containing proteins, which subsequently process the specific downstream target proteins to trigger distinct cellular functions. The suppression of ASC/TMS1 expression, a phenomenon frequently observed in cancer cells, inhibits apoptosis but on the other hand, it can enhance NF-κB-dependent inflammatory responses.

Role of ASC/TMS1 in inflammasome assembly and pyroptosis

In their seminal study, Martinon et al. [40] demonstrated that ASC/TMS1 is an adaptor protein in the assembly of inflammasomes, which are key players in innate immunity defense. Inflammasomes are molecular platforms containing a pattern recognition receptor protein that recruits the inflammatory caspases, especially caspase-1, to trigger the processing of pro-IL-1β and pro-IL-18 into mature cytokines capable to stimulate inflammatory responses [8, 12, 52, 53]. NOD-like receptors (NLR), especially NLRP3, are the major guardians of tissue homeostasis and defense system against microbial attack. NLRP3 can be activated by a large variety of pathogens and cellular stress insults, generally called danger-associated molecular patterns (DAMPs). Absent in melanoma-2 (AIM2) receptors recognize specifically microbial DNA as well as cellular DNA relocated into the cytoplasm [54, 55]. In addition, NLRP1 and NLRC4 (also known as IPAF or CARD12) activate caspase-1 via ASC/TMS1 adaptor protein [40, 56, 57]. In general, NLRs bind caspase-1 directly via the CARD domains but NLRP1, NLRP3, and AIM2 do not contain CARD domain, instead they bind via the PYD domain to the ASC/TMS1 adaptor, which subsequently interacts with caspase-1 through its CARD domains [12, 40]. Although NLRC4 contains a CARD domain, its activation becomes enhanced in the presence of ASC/TMS1 [56]. ASC/TMS1 probably binds to the CARD domain of NLRC4 potentiating the activation [52, 57]. These studies indicate that ASC/TMS1 has a crucial role in the assembly and activation of AIM2, NLRP1, NLRP3, and NLRC4 inflammasomes in the stimulation of inflammatory responses (Fig. 1). The priming of inflammasomes [9] involves an increased expression of proforms of IL-1β and IL-18 and commonly also that of inflammasomal receptors, such as NLRP3 [58] and AIM2 [59, 60] in inflammatory conditions.

The mechanisms involved in the activation of NLRP3 and AIM2 have recently been the focus of inflammation research and there are several detailed reviews on that topic [8, 9, 12, 61]. Oxidative stress, lysosomal destabilization, potassium efflux, crystals, and protein aggregates are common endogenous inducers of NLRP3 inflammasomes. Moreover, NLRP3/ASC inflammasomes can be activated by high levels of saturated fatty acids [62], which indicates that NLRP3 inflammasomes have a key role in the pathogenesis of the metabolic syndrome [63, 64]. The aging process also involves stress components associated with the activation of NLRP3 inflammasomes [65]. The drug discovery programs for inflammasome inhibition, so-called inflammasome blockers, are still at the early phase of development [66]. However, it may be challenging to find specific targets since the ASC/TMS1 adaptor protein is involved in different pathways and side effects could be the problem. There are several endogenous proteins that can inhibit the function of ASC/TMS1. These are CARD-only (COPs) and PYD-only (POPs) proteins that bind to the CARD or PYD domains, inhibiting the assembly of inflammasomes [67] (Fig. 1). In addition to endogenous proteins, some viral proteins also target themselves to ASC/TMS1 in order to inhibit the function of NLRP3 and AIM2 [68].

Pyroptosis is a caspase 1-dependent, cell death mechanism that can be stimulated by many pathological insults, e.g., microbial infections and tissue injuries [69]. Apoptosis and pyroptosis are analogous cell death mechanisms but they display several fundamental differences, e.g., (a) pyroptosis enhances the inflammatory response, while apoptosis does not induce inflammation, (b) pyroptosis causes membrane swelling and rupture whereas apoptosis triggers the formation of apoptotic bodies, (c) pyroptosis is an exclusively caspase-1-dependent process, whereas apoptosis is associated with other caspases, e.g., 2, 3, and 6–10. Moreover, ASC/TMS1 has a major role in the formation of pyroptosomes, large supramolecular assemblies containing ASC/TMS1 and caspase-1 proteins [70]. Fernandes-Alnemri et al. [70] demonstrated that pyroptosomes are formed by the dimerization of ASC/TMS1 via the PYD domains, whereas the CARD domains recruit procaspase-1. They revealed that the stimulation of THP-1 macrophages by several pro-inflammatory stimuli triggered the formation of large aggregate-like structures, i.e., ASC/TMS1 specks, usually with one pyroptosome in each cell. Cheng et al. [71] have described in detail the dynamics of the speck formation in epithelial cells. The specks are protein aggregates that trigger the caspase-1-mediated pyroptotic cell death [70]. Recent studies have demonstrated that the activation of NLRP3 and AIM2 receptors can also lead to pyroptosis if the stimulus is strong enough [72, 73]. Moreover, pyroptosis can be induced without speck formation [73, 74]. Interestingly, there are also studies revealing that the activation of AIM2 and NLRP3 inflammasomes can recruit procaspase-8, instead of procaspase-1, via ASC/TMS1 and consequently trigger apoptosis [75, 76].

ASC/TMS1 can also increase the expression level of Dock2 by stabilizing its mRNA [77]. This inflammasome-independent response potentiates the Rac activation and actin polymerization in lymphocytes, which enhances their chemotaxis and migration. It seems that ASC/TMS1 also controls antigen uptake and presentation in dendritic cells [77]. However, later studies revealed that the ASC^−/− mice were also defective in Dock2 expression, and thus it seems that ASC/TMS1 does not regulate adaptive immunity responses via Dock2 pathway [78].

Significance of epigenetic regulation of ASC/TMS1

As described earlier, the ASC/TMS1 gene is a sensitive target of DNA methylation in different types of cancers, thus suppressing apoptotic cell death of cancerous cells. It is known that the demethylation of the ASC/TMS1 gene can sensitize cancer cells for stimuli, which induce apoptosis through p53 and TRAIL activation [79, 80]. Nakajima et al. [81] demonstrated that the aging process decreased the methylation status of CpG island in the ASC/TMS1 gene, which implies that the expression of ASC/TMS1 might be elevated during aging enhancing sensitivity to inflammasome activation and inflammatory responses with aging. It is known that the aging process is associated with a global genome-wide hypomethylation of DNA, a hallmark of the aging process [82]. Nakajima et al. [81] reported that physical exercise elevated the methylation level of ASC/TMS1 gene in human peripheral blood, probably by decreasing the sensitivity to inflammatory reactions following an exercise training program. Moreover, Youm et al. [83] observed that the knockout of the ASC/TMS1 gene protected mouse thymus against the age-related decline in thymic mass and cellularity, most likely by reducing apoptosis and inflammation. These observations indicate that the modulation of ASC/TMS1 expression, probably via epigenetic control, has a crucial role in the regulation of apoptosis and inflammation.

ASC/TMS1 coordinates apoptosis and inflammasome functions

Given that apoptosis and inflammation are important host defense mechanisms [84, 85], it is not surprising that their regulation is coordinated by a common adaptor protein, ASC/TMS1. Although the ASC/TMS1-mediated pathways represent only one component of the complex regulation of apoptosis and inflammation, they are specifically under epigenetic regulation. The coordinated function of inflammasomes and apoptosis confers significant benefits in the maintenance of tissue homeostasis in pathological conditions (Fig. 2). Inflammasomes are sensitive at sensing insults jeopardizing cellular housekeeping, and subsequently they can trigger the secretion of cytokines and chemokines able to recruit phagocytes to the threatened tissue. The appearance of phagocytes potentiates the clearance of apoptotic bodies generated by apoptotic cell death. However, the overwhelming expression of ASC/TMS1 and inflammasome activation can lead to pyroptosis. Several studies have revealed that the expression of ASC/TMS1, in addition to NLRP3, is substantially increased in pathological conditions, e.g., after brain injuries [86], atherosclerosis [87], and infections [88]. Currently, it is well known that the expression of the ASC/TMS1 gene is epigenetically regulated in cancer but the role of DNA methylation of ASC/TMS1 in inflammatory diseases still needs to be clarified. However, a high-fat diet for over three generations induced the DNA hypomethylation of several inflammatory genes, increased their expression, and maintained chronic inflammation in mouse adipose tissue [89]. On the other hand, the ASC/TMS1-mediated coordination of apoptosis and inflammasome functions is limited to the cells where both ASC/TMS1 and inflammasome receptors are expressed. NLRP3 and AIM2 are mostly expressed in innate immunity and epithelial cells [59, 60, 90] but NLRP1 can also be expressed in many other tissues, e.g., in neurons [90]. Moreover, it is known that the infiltrating macrophages have a crucial role in the pathogenesis of obesity and metabolic syndrome [91]. Stienstra et al. [92] demonstrated that high-fat diet (HFD)-induced obesity was prevented in NLRP3 knockout mice. Interestingly, they also reported that HFD did not induce insulin resistance, liver steatosis, or adipocyte hypertrophy in ASC/TMS1-depleted mice. This indicates that the function of ASC/TMS1 in invading inflammatory cells has a fundamental role in many diseases, although it is not expressed in target tissue cells.

Fig. 2.

The expression of ASC/TMS1 coordinates the function of apoptosis and inflammasomes in carcinogenesis. The upregulation of ASC/TMS1 expression enhances apoptosis and the assembly of inflammasomes. The activation of inflammasomes induces the secretion of IL-1β and IL-18, which provoke an inflammatory response but also stimulate the recruitment of phagocytes into the tissues to clear the apoptotic bodies produced by apoptotic cell death. Increased sensitivity to apoptosis and effective clearance of apoptotic bodies prevent carcinogenesis. On the other hand, chronic inflammatory environment can enhance carcinogenesis, e.g., via reactive oxygen species and many cytokines and chemokines, which is not directly linked to the activation of inflammasomes. The epigenetic inhibition of ASC/TMS1 expression, observed in many cancers, inhibits apoptosis and exposes cells to excessive proliferation. The lack of expression of ASC/TMS1 represses the function of AIM2 and NLRP3 inflammasomes, which leads to disturbances in cellular housekeeping, e.g., pathogen recognition, and, moreover, a deficiency of caspase-1 activation can increase the level of glycolytic enzymes, which are degraded by caspase-1. Decreased cellular integrity and increased glycolytic metabolism are the hallmarks of apoptosis-resistant cancer cells. This figure is a hypothetical presentation and there can be cell type differences in the response to deficiency of ASC/TMS1 expression, as described by Drexler et al. [36] and discussed in the text

In physiological terms, it seems quite logical that the regulation of apoptosis and inflammasome function would be linked to DNA methylation. For instance, apoptosis, i.e., programmed cell death, is an important mechanism deciding the fate of the cell during morphogenesis. Given that DNA methylation is the mechanism that controls the differentiation of cells into specific tissues [93], a lack of methylation could stimulate the apoptosis of cells superfluous to tissue development. The cleansing of cell debris via the inflammasome-induced recruitment of phagocytes could prevent chronic inflammation (Fig. 2). On the other hand, the genome-wide changes in DNA methylation, observed in the aging process [94, 95], could provoke the ASC/TMS1-mediated apoptosis and impaired inflammasome function, even pyroptosis. Interestingly, epigenetic regulation could determine the fate of distinct cells in tissues via ASC/TMS1 activation, both during development and under pathological conditions, without disturbing tissue homeostasis.

Role of inflammasomes in carcinogenesis

There is abundant literature indicating that inflammation can enhance carcinogenesis [6, 96]. However, the ASC/TMS1-mediated activation of inflammasome seems to have context-dependent, opposite effects on carcinogenesis, as recently reviewed [97, 98]. The activation of inflammasomes stimulates IL-1β and IL-18 secretion, which provokes inflammatory response, e.g., by recruiting phagocytes to damaged tissue and thus enhancing the cleansing of apoptotic bodies and cell debries (Fig. 2). This inflammatory response also involves the generation of many factors, e.g., reactive oxygen species and many cytokines and chemokines, which are associated with cancer formation [6, 96]. Given that inflammasome receptors are highly expressed in several chronic inflammatory diseases, e.g., psoriasis [59] and atherosclerosis [60], which are not linked to tumorigenesis, it seems that the activation of inflammasomes do not directly induce tumor formation.

There are several studies indicating that NLRP3 inflammasomes can inhibit tumorigenesis, e.g., during chronic intestinal inflammation [99, 100]. Zaki et al. [101] demonstrated that IL-18 conferred protection against colitis-induced tumorigenesis by activating IFN-γ production. Subsequently, IFN-γ induced the STAT1-dependent tumor suppressor pathway. Allen et al. [99] observed that the presence of ASC/TMS1 and caspase-1 were required for tumor prevention in experimentally induced colitis. Many reports have revealed that the lack of ASC/TMS1 enhances tumor formation in different experimental models [36, 45, 99]. Drexel et al. [36] reported that ASC/TMS1-induced tumor suppression was associated with p53 activation. It is also known that the inhibition of ASC/TMS1 expression can increase NF-κB signaling [49], a prominent inducer of apoptotic resistance [102]. The activation of NF-κB signaling also stimulates inflammatory responses and thus could provoke the inflammation present in cancers. The deficient expression of the ASC/TMS1 protein may also jeopardize the cellular housekeeping and disturb the control of cell integrity, which enhances the apoptotic resistance commonly encountered in cancer cells (Fig. 2). Moreover, caspase-1 has many inflammation-independent targets, such as the glycolytic enzymes [103]. Therefore, the lack of ASC/TMS1 can increase the cellular glycolytic rate, which is significantly increased in cancer cells. It seems that inflammasomes have a crucial role in the defense against carcinogenesis and thus increased methylation of the CpG island of ASC/TMS1 gene can promote carcinogenesis.

Conclusions

ASC/TMS1 is an important adaptor protein since it can interact with several proapoptotic proteins as well as AIM2 and NLRP3 inflammasome receptors. Many context-dependent factors coordinate these distinct functions of ASC/TMS1 in cellular housekeeping. The promoter region of ASC/TMS1 gene contains a CpG island, a feature commonly present in housekeeping genes. Several reports have demonstrated that the level of CpG island methylation correlated with the degree of ASC/TMS1 expression. With the CpG island being highly methylated, the expression of ASC/TMS1 was blocked in many cancers. Demethylation enhanced the expression of ASC/TMS1 and induced apoptosis in cancer cells. These studies have convincingly indicated that the level of DNA methylation is a major mechanism controlling the expression of ASC/TMS1 and thus it is under complex epigenetic regulation. DNA methylation is a crucial mechanism in carcinogenesis, e.g., methylation leads to the inhibition of tumor suppressor genes. However, currently it is still unknown whether epigenetic mechanisms control the function of inflammasomes and thus inflammatory responses. Given that aging and many nutritional and environmental factors affect the DNA methylation status, it seems that the epigenetic regulation of ASC/TMS1 exerts fundamental effects not only on carcinogenesis but on the pathogenesis of many age-related diseases, which are associated with apoptotic and inflammatory components.