Neutrophil Elastase Inhibitors

Abstract

Introduction

Chronic obstructive pulmonary disease (COPD) constitutes a worldwide health problem. There is currently an urgent and unmet need for the development of small molecule therapeutics capable of blocking and/or reversing the progression of the disorder. Recent studies have greatly illuminated our understanding of the multiple pathogenic processes associated with COPD. Of paramount importance is the key role played by proteases, oxidative stress, apoptosis, and inflammation. Insights gained from these studies have made possible the exploration of new therapeutic approaches.

Areas covered

An overview of major developments in COPD research with emphasis on low molecular weight neutrophil elastase inhibitors is described in this review.

Expert opinion

Great strides have been made toward our understanding of the biochemical and cellular events associated with COPD. However, our knowledge regarding the inter-relationships among the multiple pathogenic mechanisms and their mediators involved is till limited. The problem is further compounded by the unavailability of suitable validated biomarkers for assessing the efficacy of potential therapeutic interventions. The complexity of COPD suggests that effective therapeutic interventions may require the administration of more than one agent such as, for instance, an HNE or MMP-12 inhibitor with an anti-inflammatory agent such as a phosphodiesterase-4 inhibitor, or a dual function agent capable of disrupting the cycle of proteolysis, apoptosis, inflammation and oxidative stress

1. Introduction

Chronic obstructive pulmonary disease (COPD) is a multi-factorial, chronic inflammatory disorder characterized by enlargement of the airspaces and airflow obstruction (1). The disorder constitutes a major health problem that affects more than 16 million people in the U.S. and is currently the fourth most common cause of death (2–3). Furthermore, COPD is continuing to increase in both prevalence and mortality, accounting for 120,000 deaths per year in the U.S. and an annual financial burden in excess of $37 billion (4). Current therapy for COPD is symptomatic and there are no drugs on the market that arrest or slow the relentless progression of the disorder (5–6). This review is intended to serve as an updated sequel to earlier reviews on neutrophil elastase inhibitors and COPD (7–9) and highlights emerging concepts and paradigms related to the pathophysiology of the disorder and potential therapeutic modalities, with particular emphasis on low molecular weight neutrophil elastase inhibitors.

2. Pathogenesis of Chronic Obstructive Pulmonary Disease

The molecular mechanisms underlying the pathogenesis of COPD are complex and poorly understood, gravely hampering the development of novel therapeutics. However, although many fundamental questions related to COPD pathogenesis (for example, the precise pathogenic mechanisms responsible for the initiation and progression of the disorder, the role(s) and the relative importance of the multiple processes and mediators involved in COPD, etc) remain poorly-defined (4), recent seminal findings have greatly illuminated our understanding of the key biochemical and cellular events associated with the disorder. These could potentially be exploited, ultimately leading to the emergence of new and effective therapeutic interventions (10).

Polymorphonuclear leukocytes (neutrophils) ordinarily serve as an important host defense against bacterial and fungal infections. Invading pathogens are phagocytosed and then killed in phagolysosomes by the combined action of reactive oxygen species generated by the NADH oxidase and the intracellular release of antibacterial proteins and proteolytic enzymes (11–13). The latter are stored in the azurophilic and specific granules of neutrophils and include the serine proteases human neutrophil elastase (HNE), proteinase 3 (Pr 3) and cathepsin G (14). Several lines of evidence suggest that neutrophil-derived oxidative and proteolytic mediators (HNE, Pr 3) released extracellularly at sites of inflammation play an important role in the pathogenesis of a range of inflammatory diseases (15), including COPD, cystic fibrosis (16–17), acute lung injury (18) and others (19). Airflow obstruction arising from the hypersecretion of mucous into the airways, a characteristic feature of COPD and other chronic lung diseases, is due to the stimulation of goblet cells by HNE and Pr 3 (20–21).

The protease/antiprotease imbalance hypothesis, as originally framed, postulated that damage to lung connective tissue results from the massive migration of neutrophils to the lungs during smoke-induced inflammation and the subsequent release of proteolytic enzymes. Inadequate control of the activity of these enzymes due to depressed levels of their physiological protein inhibitors leads to a protease/antiprotease imbalance in the airways, ultimately allowing the degradation of elastin, the elastic component of lung connective tissue, and other components of the extracellular matrix (22–25). Thus, the hypothesis focused primarily on neutrophil and neutrophil elastase as the primary cell and enzyme responsible for COPD. However, in addition to HNE, other cell types and enzymes derived thereof, are of critical importance in COPD (26–28). Specifically, in addition to neutrophils, the inflammation associated with COPD involves the influx to the lungs of macrophages and CD8+ T lymphocytes. Indeed, alveolar macrophages are the most abundant defense cell in the lung under normal conditions and during chronic lung inflammation. They are a major source of metallo- (macrophage metalloelastase, MMP-12) and cysteine (cathepsin S, Cat S) proteases. The capacity of cathepsin S to degrade elastin is comparable to that of HNE and Pr 3. Studies have shown that cigarette smoke-induced emphysema is closely associated with the influx of macrophages to the lungs and their enhanced elastolytic activity (29). Indeed, the extent of lung destruction in emphysematous patients is directly related to the number of alveolar macrophages and CD8+ T-lymphocytes (30–31).

Cigarette smoke inhalation inactivates histone deacetylase and results in the release of nuclear factor-κB (NF-κB) leading to transcription of neutrophil chemokines and cytokines (TNF-α and IL-8) (32). The subsequent secretion of MMP-12 by macrophages leads to the release of tumor necrosis factor-α (TNF-α), an important pro-inflammatory cytokine which activates endothelial cells and leads to an influx of neutrophils to the lungs. Subsequent release of serine proteases by neutrophils and metalloproteases by macrophages is believed to account for most matrix destruction (33).

The influx of neutrophils and other phagocytic cells is associated with chronic airway inflammation, a cardinal pathophysiologic feature in patients with COPD (34–35), which is characterized by elevated levels of interleukin-8 (IL-8), a chemotactic cytokine produced by neutrophils, alveolar macrophages, lymphocytes and epithelial cells (36–38). Indeed, the sputum concentration of IL-8 is closely related to the degree of airway obstruction in COPD (39). Several lines of evidence indicate that neutrophil serine proteases play an important role in regulating inflammation by activating pro-inflammatory cytokines (40). For instance, Pr 3 is known to activate TNF-α, IL-1β, and IL-18, while HNE induces apoptosis of lung epithelial cells and activates epidermal growth factor receptor (EGFR) and toll-like receptor-4.

Oxidative stress, a hallmark of COPD, arises from the inhalation of cigarette smoke which contains a multitude of highly reactive species (41) and the production of reactive oxygen species by inflammatory cells recruited to the lungs (45). Oxidative stress leads to infiltration of neutrophils, macrophages, and cytotoxic T lymphocytes to the lungs which release an array of proteolytic enzymes (HNE, Pr 3, cathepsin S, MMP-12) (42–43). These mediate a multitude of signaling pathways (40, 44) and participate in the degradation of lung connective tissue and other components of the extracellular matrix due to the resultant protease/anti-protease imbalance (vide supra). Furthermore, oxidative stress leads to elevated airway inflammation, alveolar cell apoptosis, and impaired lung repair. Nuclear factor erythroid-2-related factor 2 (Nrf2), a transcription factor that plays an important role in protecting the lungs against cigarette smoke-induced inflammation, oxidative stress, and alveolar cell apoptosis by upregulating multiple antioxidant and detoxification genes (45), exhibits decreased activity in COPD and may play a critical role in the pathogenesis of the disorder (46).

In summary, the preponderance of evidence indicates that COPD involves the confluence and interplay of an oxidant/antioxidant imbalance (47–48), a protease/antiprotease imbalance (42–43), apoptosis (49–51), and chronic inflammation (34–35, 52), which are interlinked by elevated levels of an array of mediators, including ceramide, nuclear factor-κB (NF-κB), histone acetyltransferases, histone deacetylases, cytokines, and chemokines (40, 53–54) (Figure 1). Thus, it is clearly evident that the convergence of multiple pathogenic processes, in particular the pivotal role played by oxidative stress in alveolar cell apoptosis, inflammation, and proteolysis of lung elastin that results in lung maintenance and repair impairment, ultimately leads to the COPD phenotype (53–57).

Figure 1.

Confluence and interplay of major processes and mediators associated with COPD pathogenesis. Lung tissue damage resulting from the inhalation of free radicals in cigarette smoke results in oxidative stress and leads to the recruitment of neutrophils, macrophages and T lymphocytes to the lungs. Oxidative stress also leads to upregulation of ceramide, resulting in apoptosis of lung epithelial cells. Neutrophils release HNE while macrophages release MMP-12, and cathepsin S which collectively degrade lung elastin and exracellular matrix components. In addition, macrophages activate NF-κB and release various pro-inflammatory and chemotactic mediators which amplify inflammation while neutrophils release reactive oxygen species. MMP-12, oxidants in cigarette smoke, and reactive oxygen species inactivate alpha-1-proteinase inhibitor, the physiological inhibitor of HNE, resulting in a protease/anti-protease imbalance which allows the renegade proteases to degrade lung elastin. HNE inactivates TIMPs, the physiological inhibitor of MMP-12, augmenting elastin degradation. Elastin fragments are chemoattractants to macrophages and drive the progression of the disease. Alveolar wall homeostasis is eventually compromised, leading to enlargement of the airspaces and the COPD phenotype. Abbreviations: MMP-12, macrophage metalloelastase; HNE, human neutrophil elastase; TNF-α, tumor necrosis factor alpha; IL-1β, interleukin 1beta; IL-8, interleukin 8; MCP-1, monocyte chemotactic protein; MIP-1, monocyte inflammatory protein; HDACs, histone deacetylases; LTB₄, leukotriene B4; VEGFR, vascular endothelium growth fibroblast factor; NF-κB, nuclear factor kappa B; IFN-γ, interferon gamma; ROS, reactive oxygen species.

3. Neutrophil Elastase

HNE is a basic, 218 amino acid single polypeptide glycoprotein (M_r 29,500) whose primary structure shows considerable homology with proteinase 3 (54%), and cathepsin G (37%). HNE has an extended binding site and prefers hydrophobic substrates and inhibitors (58). The S₁-S₄ subsites of HNE have been mapped using peptidyl chromogenic substrates (59). The primary substrate specificity subsite S₁ of HNE shows a preference for medium size P₁ alkyl groups (isopropyl, n-propyl, isobutyl, n-butyl), however, P₁ specificity is regulated by peptide length. Several X-ray crystal structures of HNE complexed to low molecular weight inhibitors (peptidyl chloromethyl ketones) and protein inhibitors (turkey ovomucoid inhibitor, secretory leukocyte protease inhibitor domain 2) have been reported. The X-ray crystal structure of HNE consists of two homologous β-barrels, each barrel composed of six antiparallel β-sheets, and a C-terminal α-helix. The catalytic triad residues (Ser195, His57, and Asp102) are located at the junction of the two β-barrels. As is true for all serine proteases, catalysis by HNE involves three steps, namely, substrate binding, acylation of Ser195, and deacylation (59, 60).

Endogenous protein inhibitors of proteases are divided into three main classes: serpins (serine protease inhibitors) (61–62), TIMPS (tissue inhibitors of metalloproteases) (63) and cystatins (cysteine protease inhibitors) (64). The activity of HNE is primarily regulated by the serpins α-1-proteinase inhibitor (α-1-PI) (65) and monocyte/neutrophil elastase inhibitor (MNEI, also called Serpin B1) (66), and the chelonianin family of canonical inhibitors that includes secretory leukocyte proteinase inhibitor (SLPI) (67), and elafin (68–69). The activity of Pr 3 is controlled by the same inhibitors, with the exception of SLPI which does not inhibit the enzyme. α-1-PI, a blood plasma glycoprotein (M_r 53 kDa), regulates the activity of HNE and Pr 3 in the lower respiratory tract, while SLPI protects the upper airways from proteolysis (70). MNEI (SerpinB1) is a potent inhibitor of HNE, as well as PR 3 and cathepsin G, while elafin, a 6 kDa protein isolated from bronchial secretions, is a potent inhibitor of HNE and Pr 3 (68–69). α-1-Macroglobulin serves as a general plasma inhibitor of serine, cysteine and metallo- proteases. There is increasing evidence that serine protease inhibitors (α-1-PI, SLPI, elafin), besides regulating inflammation by inhibiting the proteolytic activity of proteases, also directly affect leukocyte chemotaxis and pro-inflammatory mediator release, and may contribute to defense against invading pathogens. For instance, in addition to its anti-elastolytic activity, α-1-PI may alleviate emphysema by inhibiting smoke-induced inflammation via the suppression of tumor necrosis factor-α (71). Most importantly, because macrophage metalloelastase and neutrophil elastase degrade α-1-PI and TIMPS respectively (the two enzymes degrade each other’s inhibitor), the matrix degrading capacity of these enzymes is greatly augmented in emphysema. As stated earlier, oxidative processes play multiple roles in COPD (vide supra) (45–48). For instance, the reactive center of α-1-PI has a critical methionine residue (Met-358) that serves as the primary specificity residue P₁ that is recognized and accommodated at the S₁ subsite of HNE. The inhalation of oxidants during smoking and the presence of elevated levels of endogenous oxidants released by inflammatory cells, lead to the inactivation of α-1-PI via the oxidation of Met-358 and/or Met-351 to the corresponding sulfoxide (72). Oxidized α-1-PI is incapable of preventing the proteolysis of lung elastin by HNE (73) and, furthermore, interacts directly with lung epithelial cells to release chemokines (IL-8, MCP-1, TNF-α) that attract macrophages and neutrophils into the airways (74). Cigarette smoke also inactivates SLPI and histone deacetylases. Thus, oxidative processes influence the protease/antiprotease balance by decreasing the protease inhibitor screen and by increasing the protease burden in the lungs through the recruitment and degranulation of phagocytic cells.

4. HNE Inhibitors

The general principles related to the design of inhibitors of serine proteases in general, and HNE in particular, have been described in previous reviews (75–76). The design of HNE inhibitors has primarily focused on the attachment of a serine trap (CHO, CF₃, CF₂CF₃, CF₂COOR, CF₂(C=O)NHR, CO(C=O)NHR, CO(C=O)OR, B(OH)₂, and COHet, where Het = heterocycle) to a peptidyl recognition element that facilitates the binding of the inhibitor to the active site and the ensuing reversible formation of a tetrahedral adduct (Figure 2). Inhibitors of this type, termed transition state inhibitors, form a tetrahedral adduct with the active site serine and exploit favorable binding interactions with multiple subsites at the active site of the enzyme, consequently they are highly potent (inhibition constant, K_I, is generally in the low nanomolar range).

Figure 2.

Interaction of a serine protease with a transition state inhibitor.

A second type of inhibitors included in this review are agents that acylate the active site serine and undergo slow deacylation (termed alternate substrate inhibitors). The likelihood of exhibiting side effects is greater with acylating agents because of their higher chemical reactivity. This review includes inhibitors of HNE reported in the patent and primary literature since 2005 (Table 1).

Table 1.

Low molecular weight inhibitors of human neutrophil elastase

Entry	HNE inhibitors	IC50 (nM)
a) Transition state inhibitors
1		1600a
2		29
b) Alternate substrate inhibitors
3		15.4
4		16
5		28
6		99.3a
7		21a
c) Carbamylating agent
8		350 – 1260b
d) Mechanism-based inhibitor
9		20 – 4580b
e) Reversible competitive inhibitors
10		7340
11		3 – 32
12		< 0.3
13		2.2
14		1 – 100
15		< 100
16		0.7
17		0.26
18		0.21
19		30000
20		_c
21		<30
22		_c

^aK_I value;

^bk_inact/K_I (M⁻¹s⁻¹);

^cunavailable.

The lungs are highly perfused and enzyme inhibitors are generally hydrophobic in nature, consequently, oral or intratracheal administration of HNE inhibitors results in limited or no efficacy due to rapid clearance. To circumvent this problem, a transition state inhibitor of HNE was attached to a 25-amino acid fragment of human surfactant peptide B, yielding a construct 1 with a long lung residence time and minimal immunogenicity. Intratracheal administration of construct 1 prevented HNE-induced emphysema in a rodent model (77). In contrast, administration of the peptidyl difluoroketone inhibitor by itself failed to protect the lungs from the action of HNE.

A peptidyl trifluoromethyl ketone transition state inhibitor of HNE with a cyclic urea amide bridged carboxyl group at the N-terminus 2 has shown noteworthy efficacy in a lung hemorrhage hamster model when administered intravenously by infusion or by bolus injection (78). Compound 2 has high solubility and stability in water and shows no overt toxic effects in mice at high doses.

Substituted benzoxazinones have attracted intense interest as inhibitors of HNE in particular (79) and proteases in general (80–83). The interaction of HNE with this class of compounds involves acylation of the active site serine followed by partitioning of the resulting acyl enzyme leading to the formation of 1H,3H-quinazoline-2,4-dione or the deacylation product (Figure 3). The preferred pathway is dependent on the nature of R (79).

Figure 3.

Mechanism of inhibition of a serine protease by substituted benzoxazinones.

A recent study has described the synthesis of benzoxazinone analogs 3 (84) which exhibit a dual mode of action by inhibiting superoxide anion generation and HNE release. Inhibition of HNE was found to be critically dependent on ring A substitution by chlorine. The results of preliminary structure-activity relationship studies involving benzoxazinone derivatives 4–5 have also been reported (85). A 5-ethyl-7-methoxy substitution pattern conferred an optimal balance between chemical stability and potency. Furthermore, rat plasma stability was found to be dependent on the nature of the piperazine and piperidine substituent.

Substituted azetidine-2, 4-diones 6 have been shown to be highly potent and selective inhibitors of HNE (86–87). Optimal inhibitory activity was observed with R₁ = R₂ = ethyl and R₃ = N¹-aryl moiety having a heteroarylthiomethyl group at the para position. A “double-hit” mechanism of action involving rapid formation of a tetrahedral adduct followed by collapse to form a transient conjugated system (a Michael acceptor) has been postulated. Subsequent reaction with an active site nucleophilic residue was proposed to result in the formation of a fairly stable enzyme-inhibitor adduct. The high reactivity of compound 6, as evidenced by its low stability in blood plasma and off-target effects, may impact adversely the therapeutic utility of this class of inhibitors.

N-Benzoylpyrazoles 7 constitute a new class of inhibitors of HNE that inhibit the enzyme by rapidly acylating the active site serine (88). Deacylation rates were found to be dependent on ring substitution. These compounds showed variable stability in phosphate buffer (pH 7.3) that was dependent on ring substitution.

The HNE acylating agent par excellence continues to be Sivelestat (Elaspol^R) (Figure 4) (89), marketed in Japan and Korea for the treatment of acute lung injury associated with systemic inflammatory response syndrome and is currently explored for a range of other indications, including sepsis associated with acute respiratory distress syndrome and disseminated intravascular coagulation (90–92).

Figure 4.

Interaction of Sivelestat with human neutrophil elastase.

A new class of carbamylating agents 8, namely where the R₃NH(C=O) portion of the inhibitor is covalently attached to the active site serine, based on the cyclosulfamide scaffold that are selective for HNE has been reported (93). The deacylation rates of carbamylated enzymes are much lower than those of acylated enzymes. The structural motif, robustness, synthetic tractability, and multiple points of diversity embodied in the cyclosulfamide scaffold are well-suited to the design of selective inhibitors of HNE and related proteases.

Mechanism-based inhibitors have been extensively used in drug design and discovery. A mechanism-based inhibitor behaves as a substrate that is processed by the catalytic machinery of an enzyme generating a reactive electrophilic species that, upon further reaction with a nucleophilic active site residue, leads to irreversible inactivation of the enzyme. Inhibitors of this type offer the promise of greater enzyme selectivity. A series of mechanism-based inhibitors 9 designed to interact with the S′ subsites of HNE has been reported (94). Inhibitor 9 is selective for HNE and was proposed to generate a Michael acceptor via an enzyme-induced 3-aza Grob fragmentation process, initiated by the attack of the active site serine at the ring carbonyl carbon.

A low micromolar reversible competitive inhibitor 10 of HNE based on the N-amino-4-imidazolidinone scaffold has been reported (95). Inhibitor 10 is devoid of any inhibitory activity toward proteinase 3 because it exploits differences in the S′ subsites of the two enzymes.

Several new classes of structurally-diverse heterocyclic inhibitors of HNE have been reported in the patent literature. 1, 4-Diarylpyrimidopyridazinyldiones 11 (96) and the structurally similar 4-(4-cyano-2-thioaryl)-dihydropyrimidinones 12 (97) inhibited HNE potently with IC₅₀s in the low nM or pM range, respectively and were tested in a rat acute lung injury model. Suitably-functionalized 2-pyrazinone derivatives such as 13 (98) have nM IC₅₀ values toward HNE. 3,4,6,7-Tetrahydro-1H-pyrrolo[3,4-d]pyrimidine-2,5-diones 14 inhibit HNE and also exhibit muscarinic M3 antagonist activity (99), while the related 4-(4-cyanophenyl)-1-(3-trifluoromethylphenyl)-3,4,6,7-tetrahydro-1H-pyrrolo[3,4-d]pyrimidine-2,5-diones 15 inhibit HNE with IC₅₀ values < 100 nM and were also found to reduce blood hemorrhage in lungs of rats exposed to HNE (100). Burkamp et al have described the use of functionalized 2-pyrazinone derivatives, such as 16, in the inhibition of HNE (101). Dimeric 2-pyridinone derivatives 17 inhibit HNE with IC₅₀s in the nM to pM range, reduced HNE-induced damage in rat lungs, and inhibited rat neutrophil elastase activity in bronchoalveolar lavage fluid (102). 2-Pyridone 18 (103) is a nM inhibitor of HNE and related structural variants 19 inhibit HNE at μM levels (104). Other heterocyclic inhibitors that have been disclosed include tetrazolopyrimidines 20 (105), dihydropyridones 21 having IC50s < 30 μM (106), and dihydropyrimidone multimers 22 which were also shown to have HNE inhibitory activity in rat lungs (107). It is not clear what the current status of many of these classes of compounds is.

Although the exploration of low molecular weight inhibitors of HNE as potential therapeutics has traditionally received the most attention, recently there has been an increasing interest in the use of cyclic peptides in the design of enzyme inhibitors, including inhibitors of HNE (108–109). Of particular interest are the plant-derived mini-proteins cyclotides (110). In contrast to linear peptides, these macrocyclic, cystine-knotted peptides exhibit exceptional enzymatic and non-enzymatic stability, suggesting that the cyclotide scaffold could provide an effective means of engineering inhibitors of HNE that are devoid of the shortcomings associated with linear peptides. The potential of this approach is significantly augmented by the introduction of zicotonide, a cystine knot molecule, into the clinic (111).

5. Expert Opinion

Significant progress has been made toward our understanding of the key biochemical and cellular events associated with COPD. The multiple pathogenic mechanisms involved in the disorder are currently better clarified, and the relative significance and contribution of the mediators involved are better defined, than in the past. Nevertheless, our knowledge regarding the inter-relationships among the various processes and mediators and how they impact the initiation and progression of the disorder, is still incomplete. The problem is further compounded by the unavailability of suitable validated biomarkers (112) for assessing the efficacy of potential therapeutic interventions. The complexity of COPD suggests that effective therapeutic interventions may require the administration of more than one agent such as, for instance, an HNE or MMP-12 inhibitor with an anti-inflammatory agent such as a phosphodiesterase-4 inhibitor, or a dual function agent capable of disrupting the cycle of proteolysis, apoptosis, inflammation and oxidative stress (113–116) (Figure 1). This is a fast moving area of research that, despite the challenges that lie ahead, will continue to illuminate our understanding of the disorder, as well as identify new therapeutic targets. Exploration of these targets is likely to lead to the emergence of disease-modifying therapeutics for COPD in the not too distant future.

Article highlights box.

• COPD is the fourth leading cause of death in the US. Currently, there are no drugs on the market that arrest or slow the relentless progression of COPD.

• COPD involves the confluence and interplay of oxidative stress, a protease/antiprotease, apoptosis, and chronic inflammation.

• The release of serine proteases (HNE, Pr3) by neutrophils, and cysteine (cathepsin S) and metalloproteases (MMP-12) by macrophages is believed to account for most matrix destruction in the lungs.

• In addition to their degradative action, proteases released by inflammatory cells play an important role in modulating inflammation and activating pro-inflammatory cytokines.

• An array of selective neutrophil elastase inhibitors have been reported/patented recently. A peptidyl transition state inhibitor of HNE showed noteworthy efficacy in a lung hemorrhage hamster model without overt toxic effects. Several heterocyclic inhibitors of HNE with IC₅₀s in low nM to pM range have been reported and shown to abrogate the deleterious effects of HNE in the rat acute lung injury model.

• The complexity of COPD suggests that effective therapeutic interventions may require the administration of more than one agent to disrupt the cycle of proteolysis, apoptosis, inflammation and oxidative stress.