Among the three types of autophagy, macroautophagy has been studied extensively and clearly. It is induced under stress stimulus, resulting in maintaining cell growth through specifically degrading damaged or superfluous organelles or leading to various human pathologies, including lung, heart, neuron, and liver disease, cancer, myopathies, aging and so on, after excessive self-degradation (Yorimitsu and Klionsky, 2005; Parzych and Klionsky, 2014; Pastuhov et al., 2016). Morphologically, the autophagosome is one of the distinct features forming by expansion as well as de novo rather than membrane budding from an already contained cargo (du Toit et al., 2018). The membrane expands and bends, and then a spherical autophagosome is generated. Using electron microscopic examination of the obverse, it has been found that the diameter of autophagosomes ranges from 0.5 to 1.5 μm in mammals and ∼0.4 to 0.9 μm in yeast (Takeshige et al., 1992; Gu et al., 2020). Besides autophagosomes, autophagy precursors (which include the cytoplasm, organelles, or bacteria) and autophagolysosomes (autophagosomes that bind to the lysosome) are also vital subcellular structures.

For microautophagy, most of the research has focused on yeast and describes its transport pathway as cytoplasmic contents are transported into the lysosome within the lysosomal membrane invagination (Marzella et al., 1981). De Duve et al. first observed microbodies in autophagy using EM (Baudhuin, 1966). Also, Sakai et al. observed the process of microautophagy using vacuoles and microbodies of yeast with fluorescent double-labeling (Itoh et al., 1998). Then, nucleus microautophagy in yeast was reported by Roberts et al. (Fernandez et al., 2003). Early on, using EM, people noticed the free-floating vesicles within the lysosome through englobing Percoll particles by way of cup-like invaginations of the lysosomal membrane (Marzella et al., 1980). Recently, nucleophagy, which targets autophagic degradation as nuclear material, has helped in elucidating the endosomal microautophagy transports detected by indirect immunofluorescence techniques (Otto and Thumm, 2020). Microautophagy is so small that TEM is needed to see it clearly. Due to the limitations of technology, we know little about microautophagy, including the inducing factors, the mechanism, and its role in the disease processes (Mijaljica et al., 2011; Sato et al., 2019; De Falco et al., 2020).

Compared with microautophagy and macroautophagy, one of the nonspecific pathways to engulf the cytoplasm, CMA, is highly specific and refers to a specific substrate (a compound formed by a pentapeptide sequence that is biochemically related to KFERQ, namely Lys-Phe-Glu-Arg-Gln and HSPA8/HSC7, the heat shock 70 kDa protein 8) and specific receptor LAMP2A, the lysosomal-associated membrane protein 2A (Chiang et al., 1989; Dice, 1990). The morphological changes of CMA are not obvious, so immunofluorescence analyses, western blots, and other biochemical methods play crucial roles in CMA assays (Wu et al., 2019).

Besides these classical autophagy pathways, the recent identification of non-canonical autophagy pathways, such as LC3-associated phagocytosis (LAP) and LC3-associated endocytosis are showing impact on cell viability (Martinez et al., 2016). This non-canonical autophagic process relies on Rubicon (rubicon autophagy regulator [RUBCN]), contributes to immunosuppression (Sil et al., 2020). And these specific pathways have been reported to underly the pathogenesis of various diseases. For example, LAP is closely related to systemic lupus erythematosus (SLE) (Bandyopadhyay and Overholtzer, 2016; Martinez et al., 2016), tumor (Cunha et al., 2018). And LC3-associated endocytosis has been shown to be protective against neuronal cell death in Alzheimer’s Disease (Heckmann et al., 2020a, b). Similar to canonical autophagy, LAP require ATG-7/-3/-5/-12/-16L for LC3 lipidation (Martinez et al., 2011, 2015). But unlike autophagy, LAP is a process requiring NADPH oxidase-2 (NOX2)5, and Rubicon5 (Martinez et al., 2015), and LAP is independent of the pre-initiation complex containing ULK1 and FIP200 and proceeds with a distinct Beclin 1-VPS34 complex lacking ATG14 (Florey et al., 2011; Martinez et al., 2011; Kim et al., 2013).

All in all, these three canonical autophagy pathways could deliver the cargo to the lysosome to degrade and recycle while each of them features with different morphology. The formation of autophagosome is regarded as characteristic changes during macroautophagy, which is formed by portions of the cytosol and intact organelles (e.g., mitochondria) sequestered into a double-membrane vesicle (Mijaljica et al., 2011). By contrast, microautophagy involves the direct engulfment of cytoplasm at the lysosome surface requiring the EM for detecting, whereas CMA refers to a process delivering soluble and unfolded proteins directly across the limiting membrane of the lysosome (Massey et al., 2004).

PS (PtdSer), as a vital content in eukaryotic membranes, is the major anionic phospholipid accounting for 2–10%. PtdSer is generally not externally exposed in normal cells and the exposure of PtdSer could be a hallmark for stressed and dying cells, and a key signal for the removal of apoptotic cells through neighboring phagocytic cells (e.g., macrophages or neutrophils). That is, PtdSer exposure not only occurs in apoptosis but also in other types of cell death, such as necroptosis and autophagy (Galluzzi et al., 2012). It is an interconnected process between the exposing PtdSer and the loss of plasma membrane integrity, whose detections are discussed above. As for PtdSer, Annexin V, one of the imaging probes targeting DNA, is widely used to detect apoptosis when it comes to radionuclide imaging. It is worth mentioning that Annexin V, a nonglycosylated membrane protein probe, has been used as one of the few cell death imaging agents reaching phase II/III clinical trials (Nguyen et al., 2012). Synaptotagmin-I (SynI), synaptic vesicle-related protein, provides another point for probe design by binding to the negatively charged phospholipids PS and phosphatidylinositol in the presence of Ca²⁺ ions (Alam et al., 2010).

The pannexin (Panx) family, consisting of 3 members (Panx1 as the most extensively studied one, Panx2 and Panx3), has been reported as playing a vital role in extracellular ATP release (Baranova et al., 2004). In apoptosis, Panx1 being activated through cleavage mediated by caspase at the C terminus releases ATP as a “find me” signal, which is necessary for macrophage recruitment (Chekeni et al., 2010). Meanwhile, the activation of Panx1 channels contributes to the increase in plasma membrane permeability and the formation of Ca²⁺-permeable pores at the endoplasmic reticulum, leading to Ca²⁺ leakage and favoring mitochondrial Ca²⁺ uptake, which conveys cytochrome c to the cytosol to induce apoptosis (Chekeni et al., 2010). During research, Panx1 inhibitor carbenoxolone (CBX), pharmacological inhibition or small interfering RNA are used for indirectly inspecting the results in vivo and in vitro, suggesting PANX1 as a plasma membrane channel that could mediate the regulated release of find-me signals and selective plasma membrane permeability during apoptosis (Chekeni et al., 2010). The quantitative analysis of Panx1 could be detected through qPCR as mRNA levels for PANX1, and IHC or western blot for protein expression, whose function is analyzed using patch-clamp methods (Huang et al., 2020). In pyroptosis, caspases-3 and-7 and caspase-11 could not only contribute to cleaving the CT moiety of Panx1, resulting in channel opening and extracellular ATP release, but also induce K+ efflux and the activation of NLRP3 inflammasomes to process and subsequently release IL1β (Yang D. et al., 2015). Dahai et al. identified that Pannexin-1 worked critically for ATP-induced pyroptosis and was induced via cytosolic LPS using immunoblotting analysis with or without CBX, probenecid and trovafloxacin (one of the inhibitors of pannexin-1) (Yang D. et al., 2015). As for autophagy, extracellular ATP and engulfing dying cells in the Panx1-dependent pathway could not only recruit immune cells, but was also involved in the pathway of the inflammasome activation in macrophages in which a short hairpin (sh)RNA method for silencing pannexin-1 channels during co-incubation of macrophages with dying autophagic cells leads to the inhibition of ATP release and inflammasome activation (Ayna et al., 2012). By means of inhibitor targeting the key molecule involved in the pathway of RCDs, this reverse validation is also widely accepted, such as CBX or small interfering RNA being used to inhibit Panx1 for apoptosis detection through qPCR and ELISA.

DNA fragmentation to 180-200 bp and a weight more than 50 kbp is considered as a feature that clearly distinguishes apoptosis (Walker et al., 1999). There are various biochemical techniques to detect DNA ladders, including general-use agarose gel electrophoresis or flow fluorocytometric, which benefits from not being time-consuming and allowing for individual cell analysis. According to different technical principles, there are three main routine assays that were developed to detect DNA fragmentation: DNA ladder assay, TUNEL assay, and comet assay. Firstly, the DNA ladder assay: the DNA fragments (180–200 bp) could be separated into the “DNA ladder” pattern on agarose gel electrophoresis. The DNA ladder assay is used in conjunction with commercial kits that lead to a faster, more accurate, and sensitive assay, but at a greater cost (Micoud et al., 2001). Another approach is the TUNEL assay, which is utilized as a standard histochemical method for tissues, adherent cell lines, and suspension cell lines (Cuello-Carrion and Ciocca, 1999). Different from the DNA ladder assay, the TUNEL assay is more sensitive for the use of specific markers [e.g., bromodeoxyuridine labeling (BrdU), fluorescein labeling, thymidine analogs, 5′-ethynyl2-deoxyuridine labeling (EdU) (Wu et al., 2012), etc.] combined with other techniques, such as FM, flow cytometry, or laser scanning cytometry (Darzynkiewicz et al., 2008). As for the comet assay, also named ‘single cell gel electrophoresis assay’ (SCGE), it has been developed in genotoxicity testing with rapid detection of DNA repair or damage in a single cell (Collins, 2004). According to the quantification of the fluorescent signal in the core (formed of macromolecules and unfragmented DNA) and the tail (formed of predominantly single-stranded DNA) as well as the DNA damage, comets are divided into five groups: none or very low damage, low damage, medium damage, long DNA migration, apoptotic or necrotic DNA migration (Konca et al., 2003).

Although optimized approaches have been developed, various limitations are still a challenge. For instance, DNA damage not only appears in the apoptotic process but also exists in necroptosis, leading to false-positive assay results. And if there is no DNA ladder pattern, it cannot be proven that there is no apoptosis occurring in the sample since it can result from an event of the internucleosomal cleavage of DNA occurring in late apoptosis (Cohen et al., 1992; Collins et al., 1995). There are sometimes mushrooming cases where it is absent with internucleosomal DNA degradation during apoptotic or apoptotic-like cell death so that the intensity of DNA fragment labeling in these assays will be inadequate to distinguish apoptosis (Cohen et al., 1992; Catchpoole and Stewart, 1993; Knapp et al., 1999).

Caspases, as one of the cysteine aspartate-specific proteases, play a vital role in the early stages of apoptosis, and are regarded as general biomarkers to identify apoptosis. They are synthesized as zymogen precursors that consist of an amino-terminal domain of variable length followed by a p20 and a p10 unit containing the residues. These amino-terminal regions in initiator caspases obtain a caspase recruitment domain (including CARD; caspases 1, 2, 4, 5, 9, 11) or death effector domains (including DED; caspases 8, 10), which are essential for substrate recognition and catalytic activity. After being activated by proximity-induced auto-proteolysis or cleavage via upstream proteases in an intracellular cascade, an active heterotetramer is formed and induces the following pathway (Lamkanfi et al., 2002; Ramirez and Salvesen, 2018). According to the functions and domain architecture, the caspase-family members are classified into two parts: inflammatory (including caspases-1, -4, -5, and -11) or apoptotic (caspases-3, -6, -7, and caspases-8, -9, -10) (Van Opdenbosch and Lamkanfi, 2019). The detection of caspase activation has been used to identify the mitochondrial death pathway of apoptosis in cell-free systems (Chandra and Tang, 2009). Besides caspases, other proteins in response to apoptotic stimuli are also targeted as an indicator, such as Bid, Fas-associated protein with death domain (FADD), Bcl-2 and Bax, cytochrome c, high-mobility group box 1 (HMGβ1), nuclear factor kappa B (NFκB), poly [ADP-Ribose] polymerase 1 (PARP) and other apoptosis-related proteins (Nicholson et al., 1995; Gyulkhandanyan et al., 2012; Del Re et al., 2019; Yang L.L. et al., 2019; Zahran et al., 2020). In pyroptosis, it has been reported that caspase-1 could mediate the cleavage of the cytosolic protein gasdermin D (GSDMD), resulting in promotion of the formation of GSDMD membrane pores and cell lysis, which sheds much-needed light on the necrotic execution mechanism of pyroptosis (de Vasconcelos et al., 2019). As a caspase-independent pathway, necroptosis is also regulated through the caspase regulators in which caspase 8 could work as a potent inhibitor of necroptosis mediated through a heterotrimeric complex involving caspase 8, Fadd, and the long isoform of caspase 8- and FADD-like apoptosis regulator (Dillon et al., 2012). Caspase activity and these related proteins can be detected through flow fluorocytometry, the cleavage of an in vivo caspase substrate, FLIM and light microscopy, the Promega Caspase-Glo 3/7 assay, and ELISA, and meanwhile q-PCR is used at the level of the RNAs of the related proteins during cell death. An example of caspase detection, the Promega Caspase-Glo 3/7 assay (Promega Corp., Madison, WI, United States), is utilized for detection, adopting a proluminescent caspase substrate for caspase detection (Mfotie Njoya et al., 2018).

The integrity of the major outer-membrane protein (MOMP) is closely regulated by a group of proteins belonging to the Bcl-2 family and encoded by the BCL2 genes, which consists of pro-apoptotic and anti-apoptotic members (e.g., Bcl-2 and Bcl-xL) or other classes (e.g., BH3-only proteins with Bcl-2 homology (BH) domains only) (Cheng et al., 2001; Wei et al., 2001; Chipuk et al., 2010). Generally, it is believed that Bcl-2, Bax and Bak are of great importance in the intrinsic pathway of apoptosis and the Bcl-2/Bax ratio or the level of Bcl-2 is considered as a hall marker for apoptotic detection (Musumeci et al., 2015; Saraste and Pulkki, 2000). The extrinsic pathway of apoptosis, also known as the death receptor pathway, is closely related to the intrinsic pathway (mitochondrial pathway), since the truncated Bid (tBid) protein translocates to the outer mitochondrial membrane (OMM) to promote RCD by engaging Bax/Bak, resulting in MOMP and subsequent caspase-9 activation (Wu X. et al., 2020). Similar to second mitochondria-derived activator of caspases (SMAC) mimetics, MOMP could trigger tumor necrosis factor (NF-κB-dependent production) and coincidentally induce an alternative form of cell death, namely necroptosis (Giampazolias et al., 2017). Remarkably, the Bid is from both the cytosol and the organelle fraction. Because the form of Bid induced by anti-Fas antibodies stays in the cytosolic fraction briefly and only in small amounts, most of the Bid has been associated with the organelle fraction (Krysko et al., 2008). Both the Bid cleavage and the expression of the Bcl-2 protein family are able to be detected through western blot, ELISA, and flow cytometry.

The release of cytochrome c from the mitochondria is a central signal in the intrinsic pathway of apoptosis mediated through OMM permeabilization. The release of cytochrome c activates caspase-3 or -7 via the apoptosome, leading to the formation of cyt-c, caspase-9 and Apaf-1, after which apoptosis is induced (Green and Kroemer, 2005; Kroemer et al., 2007). The detection of cytochrome c serves as a biomarker of apoptosis and is also important to understand certain diseases at the cellular level. To detect the level of cytochrome c, multiple existing techniques are adopted, including western blot, ELISA, high-performance liquid chromatography (HPLC), spectrophotometry and flow cytometry. As a special detection method for cytochrome c, fluorescent aptamer/carbon dot-based assays are regarded as a simple, sensitive, rapid and selective label-free assay for apoptosis detection employing the principle of a connection between the surfaces of Carbon Dots (CDs) and nucleic acid aptamer biomolecules. It is very sensitive and selective for apoptotic detection with a 25.90 nM limit in detection and a linear range of 40-240 nM (Ghayyem and Faridbod, 2018). Notably, cytosolic fractions and the organelles should be separated using a mild detergent, digitonin (concentration of 0.02%), in order to leave the mitochondria and lysosomes intact to prevent cytochrome c release from the mitochondria, resulting in artifacts of organelle preparation (Krysko et al., 2008). Moreover, cyt c could accumulate in the culture supernatant during secondary necrosis of anti-Fas-stimulated cells, and cyt c could also accumulate in the culture supernatant from the moment that the plasma membrane loses its integrity in the late necrotic phase of TNF-stimulated cells (Denecker et al., 2001).

Various factors work as key participants in necroptosis, namely in a caspase-independent pathway, such as RIP1, RIP3, and MLKL (Galluzzi et al., 2014; Ding et al., 2015; Wu X. et al., 2020). Combined with traditional methods of detection of necrosis (morphological features, the intracellular-component release, and biochemical features involved in necrosis), specific biomarkers in necrosis allow for more accurate detection of necrosis or even a measure with greater potential for clinical implications (Vanden Berghe et al., 2013; Lu et al., 2017). It has been reported that necrosis is sometimes related to upregulated RIPK1, RIPK3, or MLKL mRNA or protein expression levels in vivo in various diseases or physiological conditions (Guo et al., 2020). Activated forms of RIP1, RIP3, and MLKL have emerged as optimal biomarkers for both distinguishing necrosis and the diagnosis or prognostic assessment of diseases related to necrotic injury (He et al., 2016). Some of the specific kinase inhibitors and inducers have been widely used for research into necroptotic signals, such as drug-induced forced dimerization of RIPK1/RIPK3, a means to directly activate RIPK1 or RIPK3 (Rodriguez and Green, 2018). Tissues or cells with increased expression levels of RIPK1, RIPK3, and MLKL may indicate a predisposition to necrosis. However, there are some exceptions to challenge these biomarker indicators while non-necroptotic functions of increased RIPK1 and RIPK3 or necrosis induced without RIPK1 and RIPK3. In detail, the increased expression of RIPK1 and/or RIPK3 could contribute to inflammatory processes as an independent necrotic pathway by regulating pro-inflammatory cytokines (e.g., TNF, IL-1β, and IL-18) (Christofferson et al., 2012; Moriwaki and Chan, 2014; Wong et al., 2014). The phosphorylation status of RIPK3 on S227, and MLKL on Ser358 and Thr357, is considered to reflect activation toward necroptosis (Degterev et al., 2008; McQuade et al., 2013; Wang et al., 2020b). For this biomarker, kinase-dead RIPK3 could function as an anti-necroptotic factor, and similarly, the negative regulation also includes caspase 8, cellular FLICE-inhibitory protein (c-FLIP), chromatin immunoprecipitation (CHIP), MAPK (mitogen-activated protein kinase)-activated protein kinase 2 (MK2), pellino E3 ubiquitin protein ligase 1 (PELI1), and ABIN-1 (A20 binding and inhibitor of NF-κB) (Wang et al., 2018c).

For monitoring autophagy in vivo, the ‘core’ Atg genes knockout mouse, GFP-LC3 transgenic mouse, TfLC3 transgenic mouse, and GFP-LC3-RFP-LC3ΔG transgenic mouse are often used as animal models to detect specific protein markers (Kuma et al., 2017). These markers are experimentally adapted as follows: (1) identification of autophagy-related (ATG) genes and proteins related to the initiation of autophagy could be used to assess the activity of autophagy and to study the role of autophagy in the pathophysiological process; (2) for the LC3 (LC3B) and GABARAP family, the ratio of lipidated LC3 (LC3-II) to free LC3 is usually utilized to reflect the number of autophagosomes forming at any given time; 3) tandem fluorescent-tagged LC3 (tfLC3) reporters are used as single-molecule probes to detect autophagosomes labeled with yellow (mRFP and GFP) and autolysosomes labeled with red (mRFP only). Other detections include the GFP-LC3-RFP-LC3ΔG probe, etc. However, some of the Atg proteins may have autophagy-independent functions and the underlying mechanisms are not elucidated so far, such that the detection of autophagy is still facing challenges (Paunovic et al., 2018; Tan et al., 2019). As for specified autophagy, such as mitophagy, it is related to signaling proteins, such as TOMM20, TIMM23, LC3, and p62 (Wang et al., 2020c).

Meanwhile, the levels of beclin have been used as a marker of autophagy, but there should be more prudent use of it, since in some cases, beclin levels can be transcriptionally upregulated to wild-type levels even in beclin heterozygous or beclin allelic loss cells (Murugan and Amaravadi, 2016). The Atg family are also key proteins, but they are not a reliable reflection of autophagic flux unless combined with LC3 (Murugan and Amaravadi, 2016). According to the functions of LC3 processing for autophagosomes, both its formation and features, western blotting is widely used in autophagy monitoring with antibodies against LC3-I and LC3-II. qPCR-based approaches have also been used for detecting the mRNA levels of autophagy regulators, such as beclin-1, ATG1, DRAM, and LC3 (Kuma et al., 2017).

Regarding pyroptosis, there are some proteins used as experimental markers, including caspase-1, propidium, IL-1β, GFP, tdRFP, and LDH. Among these, PI staining is always regarded as a proxy for membrane rupture as it is often found simultaneously for these two things, namely the membrane rupture and the activation of the gasdermin pore (Jorgensen et al., 2016). Recently, researchers have evaluated the reliability of the detection as the molecules and proteins that we detected may pass through the membrane rupture, or through the gasdermin pore, or both (Kovacs and Miao, 2017). For instance, PI and IL-1β (4.5 nm; 5 nm) are small enough to pass through the gasdermin pore (10-15 nm), resulting in a situation that detected significant amounts of PI before the membrane rupture happened (Ding et al., 2016; Liu et al., 2016). On the contrary, tdRFP and LDH are larger, so they seem to be released by the membrane rupture event (Shaner et al., 2004; Russo et al., 2016).

As a non-apoptotic form of RCDs, ferroptosis is characterized by iron-dependent accumulation of toxic lipid peroxides in plasma membranes. The cystathionine-β-synthase (CBS) is a marker of transsulfuration pathway activity involved in ferroptosis (Wang et al., 2019b). The lipid peroxide level is closely related to ferroptosis, so it provides a useful means to detect ferroptosis in biological samples, such as measuring the level of cellular lipid peroxide via flow cytometry assays, immunostaining, or colorimetric assays (Uchida et al., 1993: Yagi, 1998; Drummen et al., 2002). Meanwhile, intracellular concentrations of iron and Fe²⁺ and the mitochondrial membrane potential are also used as indicators to distinguish the induction of ferroptosis (Wang et al., 2019b, 2020a). Furthermore, increased ACSL4 is required for lipotoxicity in ferroptosis, and unlike the other ACSL members, it seems to be a marker of ferroptosis since ACSL4 is remarkably downregulated in ferroptosis-resistant cells (e.g., LNCaP and K562) (Yuan et al., 2016; Wenzel et al., 2017). As for the specific detection of ferroptosis, the changes of the mitochondria are regarded as one of the most important indicators, including the morphological and biochemical changes. Increased intracellular concentrations of iron and Fe²⁺ can be monitored by kits, such as an Iron Assay Kit (Sigma Aldrich). Increased mitochondrial superoxide can be detected using a specific fluorescent probe, such as MitoSOX^TM Red Mitochondrial Superoxide Indicator for live-cell imaging (Invitrogen). Decreased mitochondrial membrane potential is measured by some kits, such as the Mitochondrial Membrane Potential Kit MAK-159 (Sigma Aldrich) for monitoring fluorescence intensity levels (λ_ex = 490/λ_em = 525 nm) and (λ_ex = 540/λ_em = 590 nm) for the ratio analysis (Wang et al., 2019b).

Based on the specific molecules involved in the pathway during NETosis, co-localization of neutrophil-derived proteins (such as myeloperoxidase (MPO) and proteinase 3 (PR3)), and extracellular DNA would suggest the presence of NETosis (Kessenbrock et al., 2009; Nakazawa et al., 2012). Additionally, citrullinated histones could be regarded as a marker for indicating NETs formation. Histones are citrullinated by PAD4 which transport from the cytoplasm to the nucleus activated via ROS generation and calcium influx (Remijsen et al., 2011; Wang and Wang, 2013). And cfDNA could also function as one form of NETs remnants which could be detected using PicoGreen^® (Zhang et al., 2014).

Caspase-3, a cysteine protease, could retain its tetrapeptide sequence DEVD (a distinct amino acid sequence of Asp-Glu-Val-Asp), which provides cleaving activity for a long time in various extracellular fluids (Hentze et al., 2001). The p20 subunits of caspase-3 and -7 have been detected in the culture medium following secondary necrosis, whereas there are no such markers but only procaspase-3 and -7 in necrosis (Denecker et al., 2001). Thus, it has been proposed as a specific releasing and circulating biomarker to indicate apoptosis in the tissues (Denecker et al., 2001). Apoptotic cells in the tissue could be measured through various technologies, such as the TdT-mediated X-dUTP nick end labeling method and IHC using antibodies against active caspase 3 or caspase-cleaved proteins to identify the degree of apoptosis (Deng et al., 2019; Kunac et al., 2019).

CK18 releases a protein cleaved via the effector caspases at two distinct sites (Asp238 and Asp396) during cell death (Ku et al., 2016). The method named the M30-Apoptosense assay (Peviva AB, Bromma, Sweden) is used to measure the caspase-cleavage CK18 at Asp396 (CK18Asp396-NE M30 neo-epitope) and to specifically discriminate between apoptotic and necrotic cell death (Cummings et al., 2008). The circulating ccCK18 was quantitated through a specific ELISA combined with the antibody, which showed a positive signal in M30 and M65 to indicate apoptosis, whereas an exclusive positive signal in M65 was used to indicate necrosis (Kramer et al., 2004). The time and temperature requirements of this test may bias the results since it requires the samples be placed immediately on ice in order to avoid artificial cell death between acquisition and processing (Greystoke et al., 2008).

High-mobility group box 1, an architectural chromatin-binding factor, could bind to DNA for increasing protein assembly targeted to specific DNA. It is a protein secreted through activated monocytes or macrophages and is passively released via necrotic or damaged cells, but it should be mentioned that HMGB-1 cannot be found in apoptotic cells even after undergoing secondary necrosis and partial autolysis, resulting in a failure to promote inflammation even if not cleared promptly by phagocytic cells (Scaffidi et al., 2002). HMGB-1 released from necrotic cells could activate the macrophages via the toll-like receptor 2 (TLR2) and toll-like receptor 4 (TLR4) pathway (Park et al., 2004). HMGB-1 can also chemoattract and activate dendritic cells (DCs) with processing and presentation of tumor antigens (Apetoh et al., 2007; Yang et al., 2007).

Other markers released from cells may also provide indicators for RCD detection, such as DNA laddering signals. The “DNA laddering” resulting from fragmented DNA is cleaved through DNase activated via the inhibitor of caspase-activated DNase (iCAT) (Krysko et al., 2008). The 166 base pairs (bp)-length cell-free DNA (cfDNA) consists mainly of nucleosome-protected DNA being released from apoptotic tumor cells into the bloodstream, which might be used in the detection of RCD, providing a target for clinical applications (Krysko et al., 2008; Ulz et al., 2016).