Pyroptosis in pterygium pathogenesis

Pterygium is a common ocular disease characterized by proliferating fibrovascular tissue. Pyroptosis, a recently discovered programed cell death, is known to be associated with oxidative stress, one of the main causes of pterygia. Here, we aimed to study the role of pyroptosis in pterygium pathogenesis. The expression of nod-like receptor pyrins-3 (NLRP3), caspase-1, IL-18, and IL-1β was analyzed in 60 human pterygium tissues and 60 human conjunctival epithelium tissues using real-time quantitative polymerase chain reaction (qRT-PCR) and Western blot analysis. Human conjunctival epithelial cells (HConECs) and human pterygium fibroblasts (HPFs) were primary cultured and the level of pyroptosis-associated factors was detected. Both cells were treated with H2O2, and cell lysis was detected by lactate dehydrogenase (LDH) release assay, the expression of the factors by qRT-PCR, Western blot analysis, and immunostaining. The downstream factors IL-18 and IL-1β were measured after inhibition of caspase-1 to confirm the caspase-1-dependent pyroptosis. α-SMA and E-cadherin were detected as indicators of pyroptosis-induced myofibroblast activation in HPFs. We discovered that the expression of the factors was significantly increased in pterygium and that caspase-1-dependent pyroptosis presents in both H2O2-treated HPFs and HConECs during which the expression of these factors was significantly elevated and the elevation of downstream factors IL-18 and IL-1β was restrained after caspase-1 inhibition. α-SMA increase and E-cadherin down-regulation were detected in H2O2-treated HPFs and the changes were reversed by caspase-1 inhibition. Pyroptosis displays a role in the pathological process of pterygium formation and progression. Pyroptosis appears to be an intriguing target to prevent pterygium pathogenesis.


Introduction
Pterygium is an ocular surface disorder characterized as a wing-like shape on the nasal limbus, which hinders vision, causes inflammation, and affects cosmetic appearance.
Limbal stem cell degeneration and dissolution of Bowman's membrane occur in the histopathology of pterygium formation, accompanied by activated fibroblast growth, neovascularization, excessive proliferation of extracellular matrix, and inflammation [1].
Surgery is still the main treatment for pterygia, although it has a high risk of surgical complications such as recurrence, graft necrosis, and granuloma formation [2]. The recurrence after surgery is still common, although diverse efforts have been made to avoid it [3]. Therefore, understanding the mechanism of pterygium formation is of vital importance in improving treatment and decreasing occurrence rate.
Only after the activation of the caspase-1-dependent secretion mechanism can the procytokines be cleaved into mature IL-1β and IL-18 and secreted from cytosol and successively provoke a series of reactions [22,23].
The accumulation of reactive oxygen species (ROS) is an extensively studied model that can cause NLRP3 activation [24]. Oxidative stress is also one of the pathogenic factors in the formation of pterygium [25]. However, whether pyroptosis participates in the initiation and progression for pterygia remains unknown. In the present study, we characterized the expression of NLRP3, caspase-1, IL-18, and IL-1β in pterygia compared with normal human conjunctival epithelium. We also explored pyroptosis involvement in pterygium formation and verified its effects on human conjunctival epithelial cells (HConECs) as well as human pterygium fibroblasts (HPFs). Our data established for the first time that pyroptosis plays a role in the genesis and development of pterygia.

Human tissue samples
Sixty pterygium samples were collected from pterygium patients free from other ocular diseases in the operating room of the Eye Hospital of Harbin Medical University, China. Sixty control conjunctival epithelium samples (free of ocular diseases) were obtained from healthy donor eyes donated to the eye bank of Heilongjiang Province. Samples were immediately snap-frozen and stored at −80 • C until RNA or protein extraction. All samples were collected with informed consent and the study was approved by the Research Ethics Committee of Harbin Medical University.

HConEC and HPF primary culture and cell treatment
Another five pairs of donated normal human conjunctiva samples and surgically dissected human pterygium tissue samples were preserved in DMEM before transferring to the laboratory. The tissues were washed three times in PBS (containing 500 U/ml penicillin and 500 μg/ml streptomycin) and the conjunctiva epithelial layer/pterygium was carefully separated and disintegrated into 1-2 mm 3 pieces and seeded in Petri dishes precoated with FBS afterward. Complete culture media was added 8 h later to achieve full attachment of the tissues. The primary cell cultures were incubated at 37 • C with 5% CO 2 . The cell cultures were observed daily, and the medium was changed every 3 or 4 days. Subculturing was done when the cells reached 70-80% confluence and were spared into culture flasks containing complete culture media. Five primary cultured cell lines were established each from different patients/donors. Cells in the logarithmic growth phase were collected and co-cultured with various concentrations of H 2 O 2 for the required time periods. The experiment was performed between passages 3 and 10.

Cell proliferation assay
Cell proliferation assay kits (CCK-8) were used in accordance with the manufacturer's instructions. HConECs and HPFs were seeded in 96-well plates, and CCK-8 solution was added after proper treatment; the plate was maintained at 37 • C for 4 h before detection. The absorbance at 450 nm was used in the evaluation with a microplate reader.

Real-time quantitative polymerase chain reaction
Total RNA was extracted from tissues or cultured cells with TRIzol reagents, and the isolated RNA was reverse-transcribed into synthesized cDNA in a 20μl reaction mixture. Real-time quantitative polymerase chain reaction (qRT-PCR) was performed according to SYBR Green qPCR Master Mix instructions. The primers were synthesized by Co Invitrogen Ltd, Beijing, China and the sequences are listed in Table 1. The housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was taken as an internal positive control standard for quantitative analysis. Data were analyzed using the 2 − C T method.

Western blot analysis
Western blot analysis was performed to detect the expression of target proteins. The total protein content was measured following the manufacturer's instructions of BCA Protein Assay Kits. Equal amounts of protein were separated with 10% SDS/PAGE; after which the protein bands were transferred onto the membrane. The membrane was incubated with target primary antibodies overnight at 4 • C with gentle shaking. Goat-anti-rabbit secondary antibody was added to identify the appropriate primary antibody. Odyssey fluorescent scanning system (LI-COR) and Quantity One software were used for the detection and analysis of the immunoreactivity. β-actin was introduced as a loading control.

Statistical analysis
All data were expressed as the mean + − standard deviation (SD). SPSS 18.0 statistical software was used to perform statistical analysis. Statistical significance was analyzed by Student's t-test after testing data for normality (according to the K-S test). P-value less than 0.05 was considered statistically significant.

Exploration of pyroptosis in the pterygium tissues
Thirty pterygium samples and thirty normal conjunctiva samples were collected and analyzed for NLRP3, caspase-1, IL-18, and IL-1β mRNA expression using qRT-PCR ( Figure 1A). The mRNA of these factors was highly expressed in pterygium samples with statistical difference (NLRP3: P<0.001; caspase-1: P<0.001; IL-1β: P<0.05; IL-18: P<0.01). Another 30 pterygium samples and 30 normal conjunctiva samples were detected for NLRP3, caspase-1, pro-IL-1β, and IL-1β protein level with Western blot analysis ( Figure 1B,C). The protein levels of these factors were highly expressed in pterygium samples compared with normal conjunctiva samples with statistical difference (NLRP3, caspase-1, and IL-1β: P<0.05). The expression level of pyroptosis-related factors was significantly up-regulated in pterygium samples compared with normal conjunctival epithelial samples, suggesting the existence of pyroptosis in the pterygium tissues.

Validation of pyroptosis in HPFs
HConECs and HPFs were primary cultured using the tissue adherent method and identified with immunofluorescence staining of vimentin ( Figure 2D). qRT-PCR was performed to analyze the mRNA expression of NLRP3, caspase-1, IL-18, and IL-1β while Western blot analysis and immunofluorescence staining were used to detect protein level of the related factors (Figure 2A-C and E-G). According to the results, HPFs held high levels of NLRP3, caspase-1, IL-18, and IL-1β mRNA expression with P<0.001, compared with HConECs. The protein expressions of NLRP3, caspase-1, IL-18, and IL-1β also presented high level in comparison with HConECs with P<0.01. Immunofluorescence images showed similar result with Western blot analysis presenting that fluorescent intensity of NLRP3, caspase-1, and IL-1β was higher in HPFs than HConECs. The elevation of pyroptosis-related factors in HPFs provided more proof in validation of pyroptosis in pterygium genesis.

Pyroptosis participated in H 2 O 2 -treated HConECs
In order to explore whether pyroptosis is involved in pterygium formation, various concentrations of H 2 O 2 (0, 200, 400, 600, 800, and 1000 μM) were adopted to stimulate HConECs for 24 h. Pyroptosis was detected by morphology observation, CCK-8, qRT-PCR, Western blotting assays, and immunostaining ( Figure 3A-H). After exposure to H 2 O 2 , the viability of HConECs was suppressed according to CCK-8 result, the proliferation of HConECs presented a dose-dependent fashion of inhibition after H 2 O 2 treatment ( Figure 3A-B) and at the concentration of 1000 μM, the inhibition showed statistical difference (P<0.05). qRT-PCR results in Figure 3C demonstrated that the relative  NLRP3, caspase-1, IL-1β, and IL-18 mRNA levels were significantly increased at 1000 μM H 2 O 2 for 24 h (P<0.001), the mRNA levels were elevated for over 100% compared with control group (HConECs treated with 0 μM H 2 O 2 ). Western blot analysis in Figure 3D,E presented that the relative NLRP3, caspase-1, and IL-1β protein levels were increased at 1000 μM H 2 O 2 for 24 h, with statistical difference (P<0.001). The protein levels were elevated for over 50% compared with control group (HConECs treated with 0 μM H 2 O 2 ). mRNA and protein level of these factors after being treated with different concentrations of H 2 O 2 are presented in Figures 8 and 9 and Supplementary Figure S1. Immunofluorescence images ( Figure 3F-H) presented that fluorescent intensity of NLRP3, caspase-1, and IL-1β were significantly increased in HConECs after being treated with 1000 μM H 2 O 2 . The elevation of pyroptosis-related factors in H 2 O 2 -treated HConECs indicated that pyroptosis can be induced by H 2 O 2 and thus participates in pterygium formation.
In order to confirm the pyroptosis cause by H 2 O 2 , the release of cytosolic LDH in the supernatant after 6 h H 2 O 2 stimulation was taken as an indicator of cell lysis during pyroptosis. As shown in Figure 4A, LDH release percentage of HConECs also presented a dose-related manner in accordance with H 2 O 2 concentration. Moreover, as shown in Figure 4B, the cell lysis caused by H 2 O 2 was significantly protected by glycine (P<0.05) and obviously suppressed by caspase-1 inhibitor Ac-YVAD-cmk (P<0.05), confirming the caspase-1-dependent cell lysis in H 2 O 2 -treated HConECs. To further confirm the caspase-1-dependent pyroptosis cause by H 2 O 2 , caspase-1 inhibitor Ac-YVAD-cmk was introduced to suppress caspase-1 activation, and downstream targets IL-18 and IL-1β were investigated with qRT-PCR and Western blot. According to Figure 4C-E, the expression of IL-18 and IL-1β was significantly decreased when selective caspase-1 inhibitor Ac-YVAD-cmk was added to 1000 μM H 2 O 2 -treated HConECs.

Pyroptosis participated in H 2 O 2 -treated HPFs
To detect the function pyroptosis performed in pterygium progression, HPFs were exposed to various concentrations of H 2 O 2 (0, 200, 400, 600, and 800 μM) for 24 h. H 2 O 2 -induced pyroptosis was detected accordingly ( Figure 5A-H). The proliferation of HPFs was inhibited in a dose-dependent fashion with H 2 O 2 concentration at 24 h and the inhibition started to appear statistically different at 200 μM (P<0.001) and the inhibition showed the most marked impact after the concentration of H 2 O 2 reached 600 μM ( Figure 5A,B). When exposed to different concentrations of H 2 O 2 , the mRNA and protein levels of NLRP3, caspase-1, IL-18, and IL-1β were also increased in a dose-dependent manner. Compared with the control group, the relative mRNA and protein level of pyroptosis associated factors were increased over 200% when the H 2 O 2 concentration exceeded 600 μM ( Figure 5 C-H). mRNA and protein level of these factors after being treated with different concentrations of H 2 O 2 are presented in Figures 8 and 9 and Supplementary Figure S1. Immunofluorescence images presented that fluorescent intensity of NLRP3, caspase-1, and IL-1β were significantly increased in HPFs after being co-cultured with 600 μM H 2 O 2 . The elevation of pyroptosis-related factors in HPFs treated with H 2 O 2 indicated that pyroptosis can also be induced by H 2 O 2 in existing pterygium tissues and thus participates in pterygium progression.
For confirmation of the pyroptosis caused by H 2 O 2 in HPFs, LDH release percentage of HPFs was measured to indicate the relationship between H 2 O 2 concentration cell lysis level ( Figure 6A). As a result, H 2 O 2 -treated HPFs demonstrated a dose-dependent fashion of LDH release. Moreover, the cell lysis caused by H 2 O 2 was largely restrained by glycine (P<0.001) and almost prevented by caspase-1 inhibitor Ac-YVAD-cmk (P<0.001), as shown in Figure  6B. Caspase-1 inhibition of 600 μM H 2 O 2 -treated HPFs performed by Ac-YVAD-cmk led to a sharp decrease of downstream IL-18 and IL-1β expression, indicating that the caspase-1-dependent pyroptosis occurs in H 2 O 2 -treated HPFs ( Figure 6C-E).
To further investigate the role of pyroptosis of HPFs in pterygium progression, mRNA and protein levels of α-SMA and E-cadherin were detected after H 2 O 2 -related pyroptosis was induced. According to Figure 7A,B, mRNA and protein level of α-SMA was largely elevated after 600 μM H 2 O 2 stimulation for 24 h and the elevation could be reversed by caspase-1 inhibitor. Similarly, mRNA and protein level of E-cadherin was down-regulated after the same H 2 O 2 stimulation and the decrease was significantly prevented by caspase-1 inhibitor ( Figure 7C,D).

Comparison of pyroptosis factors in HConECs and HPFs after stimulation
According to CCK-8 assay, the viability of HPFs was much more susceptible to H 2 O 2 inhibition in comparison with HConECs under the same oxidative stress condition ( Figure 8A,B). qRT-PCR, Western blot analysis, and immunostaining were performed to detect the changes of NLRP3, caspase-1, IL-1β, and IL-18 expression after being treated with various concentrations of H 2 O 2 for 6, 12 and 24 h, respectively. ( Figure 8C-H and Figure 9A-F, blot images are demonstrated in Supplementary Figure S1). NLRP3 elevation appeared in the lower concentration and shorter stimulation period in both types of cells, the phenomenon coordinated with the initiation function of NLRP3 in pyroptosis. The expression of NLRP3, caspase-1, IL-18, and IL-1β all demonstrated a dose-dependent elevation pattern related to the concentration of H 2 O 2 as well as the stimulation period. The rise of pyroptosis factors occurred in HPFs at a relatively low H 2 O 2 concentration and in a remarkably early stage during co-culture. Additionally, pyroptosis factors were notably increased in HPFs at certain concentrations and during periods of H 2 O 2 stimulation, compared with HConECs exhibiting the same oxidative damage.

Discussion
Pyroptosis was observed in 1992 as Caspase-1-programed cell death and was later termed in 2001 as to differentiate it from the morphologically distinct apoptosis. The pyroptosis effector gasdermin D was discovered in 2015 as a cleavage target for caspase-1 and caspase-11 [26]. Previous study showed that pyroptosis participates in many pathological processes. By validating the elevated pyroptosis-related factors in pterygium tissues and HPFs, we deducted the possible role of pyroptosis in pterygium pathogenesis.
To elucidate the mechanism of pyroptosis in pterygium genesis, we performed further investigations. Epidemiological evidence suggests that in pterygium genesis, UV irradiation plays the most important role. UV radiation to HConECs and HPFs results in the generation of ROS [27] and causes damage in DNA, proteins, and lipids [28], which may lead to the onset and progression of pterygia. The accumulation of ROS can also up-regulate the expression of some DAMP molecules that render inflammation [29]. For pyroptosis, there is evidence that activation of ROS generation can induce NLRP3 and caspase-1 activation to trigger IL-1β as well as IL-18 production during pyroptosis [30].
Our previous study showed that pyroptosis also participated in the oxidation of human lens epithelial cells. Previous studies hinted that the accumulation of ROS can lead to pyroptosis, which may participate in the pathogenesis of  pterygia. H 2 O 2 , as a source of activated oxygen, is commonly used for in vitro cellular oxidative damage models because of its ability to permeate cellular membranes and enter the cells to cause injury [31]. We chose H 2 O 2 to simulate the oxidative injury in pterygium genesis.
We proved that in HConECs, NLRP3, caspase-1, IL-1β, and IL-18 expression was increased after treatment with H 2 O 2 , and cell lysis, characterized by LDH release, which could be inhibited by cytoprotective agent glycine, was induced. Pyroptosis, with the characteristic of rapid plasma-membrane rupture and release of proinflammatory intracellular contents, was observed in H 2 O 2 -stimulated HConECs. This phenomenon indicates that H 2 O 2 can cause increased cell permeability and fragility, and pyroptosis is induced afterward [32]. The caspase-1 inhibition experiment confirmed that caspase-1-dependent pyroptosis was induced in HConECs. These results indicate that oxidative stress participates in pterygium formation by initiating pyroptosis of HConECs.
To further explore the relationship between pyroptosis and pterygium progression, we analyzed the expression of pyroptosis-related factors in HPFs after treatment with H 2 O 2 and a similar result is discovered in HPFs. Pyroptosis-related factors are significantly up-regulated in HPFs after H 2 O 2 stimulation and downstream factors can be restrained after caspase-1 inhibition. Moreover, we discovered that pyroptosis can be induced in HPFs with a lower concentration of H 2 O 2 in a shorter period of time, which indicates that patients with pterygia are more susceptible to oxidative stress, and pyroptosis is prone to be induced in HPFs. Pyroptosis is easily initiated once the pterygium has formed and large amount of proinflammatory cytokines are released.
NLRP3 is the initiating pyroptosis factor responding to certain cellular stress signals triggered by PAMPs and DAMPs and may activate caspase-1, thus initiating the process of pyroptosis [33]. The common causes of pterygia, such as UV radiation and other types of oxidative stress, can perform as DAMPs and trigger the process of pyroptosis. Caspase-1 plays an irreplaceable role in the process of pyroptosis, functioning as precursors of the inflammatory cytokines IL-18 and IL-1β [34]. IL-18 is a proinflammatory cytokine that participates in many biological processes, including cell fibrosis and epithelial-mesenchymal transition [35,36]. IL-1β is another pyroptosis-related cytokine that acts as an important mediator in inflammatory response and can be involved in all sorts of cellular activities, including cell proliferation, differentiation, apoptosis, EMT, and fibrosis [37]. After the occurrence of pyroptosis, IL-18 and IL-1β are activated and released in large numbers. Functional IL-18 and IL-1β cause all sorts of changes, such as fibrosis, EMT, apoptosis, and abnormal cell proliferation, all of which are included in mechanisms of pterygium genesis. Therefore, the elevation of these pyroptosis factors in HPFs, which is the main component of pterygia, may play a considerable part in the progression of pterygia.
The elevation of α-SMA and decrease of E-cadherin were observed in HPFs accompanied by H 2 O 2 -induced pyroptosis and both changes could be restrained by caspased-1 inhibition. The evidence suggests that pyroptosis in HPFs may lead to the transition of fibroblasts into activated myofibroblasts that accelerates fibroproliferation and extracellular matrix accumulation in pterygium progression. Apart from an identification of myofibroblast activation, α-SMA can produce fibrovascular tissue with higher density and solidity and lead to even severe pterygia [38]. The  loss of E-cadherin, an intercellular adhesion molecule, in stimulated HPFs is another indicator of myofibrolasts activation and may also result in a more invasive growth pattern of HPFs [39]. Therefore, the pyroptosis in HPFs induced by H 2 O 2 stimulation can give rise to the activation of myofibroblasts and the progression of pterygium.
Pyroptosis, accompanied by apoptosis and autophagy, has gained accumulating attention in recent years due to its programed cell death function in multiple biological and pathological processes. Evidence shows that multiple types of cell death are observed simultaneously in cells or tissues exposed to the same stimulus, suggesting that distinctive modes of cell death occur in an overlapping pattern after certain types of damage [40]. However, the precise mechanism of programed cell death interaction in response to a particular stimulus remains unclear. Apoptosis has been reported in pterygium pathogenesis associated with oxidative stress [41]. Thus, the functions and cross-linking reactions of apoptosis and pyroptosis in ROS-induced pterygium formation provide us an intriguing target in future study.
In summary, our results provided the first evidence that pyroptosis participates in the oxidation of HConECs and HPFs and may be involved in the pathogenesis of pterygia. The salient findings from the present study indicating the essential role of pyroptosis in pterygium formation and progression provide us a better understanding of pterygia and may unveil an innovative approach in pterygium treatment which would contribute to the reduction of medical burden worldwide.