OxLDL induces vascular endothelial cell pyroptosis through miR-125a-5p/TET2 pathway

2.1 Cell culture and transfection

Human umbilical vein endothelial cells were obtained from Science Cell Research Laboratories (Carlsbad, CA). The cells were grown in endothelial cell medium (Sciencell Research Laboratories, Carlsbad, CA) supplemented with 10% fetal bovine serum at 37°C with 5% CO2 and 95% air. Overexpression/short hairpin RNA (shRNA) lentivirus of TET2 and the negative control (NC) were synthesized by GeneChem (Shanghai, China). MiR-125a-5p inhibitor and NC were purchased from RiboBio (Guangzhou, China). Cells were treated for 24 hr with different concentrations of oxLDL. VECs were transfected with overexpression/shRNA lentivirus of TET2 and miR-125a-5p inhibitors or NC when the cell density reached 50–70%. For experiments involving pharmacological inhibitors, total reactive oxygen species (ROS) scavenger N-acetyl-l-cysteine (NAC) and nuclear factor-κB (NF-κB) inhibitor (BAY 11–7082) were purchased from TCI Chemicals (Shanghai, China). Endothelial cells were pretreated with NAC (5 mM) or BAY 11–7082 (20 µM) for 2 hr and subsequently treated by oxLDL 100 µg/ml (Yiyuan, China) for 24 hr in the presence of inhibitors.

2.2 Measurement of intracellular ROS levels

The level of intracellular ROS was determined on the basis of the oxidative conversion of cell-permeable 2′,7′-dichlorofluorescein diacetate (DCFH-DA; KeyGen, Jiangsu, China) to fluorescent dichlorofluorescein upon reaction with hydroxyl radical, hydrogen peroxide, or peroxynitrite. Briefly, cells were incubated with control media (phosphate-buffered saline [PBS]) or oxLDL in the presence or absence of antioxidants NAC for 24 hr. Then, the cells were washed twice with cold PBS (pH 7.4) and incubated with DCFH-DA at room temperature for 30 min in the dark. VECs added 100 μg/ml oxLDL after 24 hr transfected with miR-125a-5p inhibitors, then perform ROS determination. Since TET2 has green fluorescence, we use ROS probes (DHE; KeyGen) for inspection. Dihydroethidine is an indicator of peroxide, which is oxidized in the cytoplasm and can enter the DNA of the cell, giving the nucleus a bright red fluorescence. The cells were incubated with DHE at a final concentration of 50 μM for 30 min and then washed three times with serum-free medium. The fluorescent signal was recorded by using fluorescence microscopy (Olympus Microscope IX3; Olympus, Tokyo, Japan).

2.3 Cell death assay

Pyroptotic cell death was evaluated with lactate dehydrogenase (LDH) release and Hoechst 33342/propidium iodide (PI) staining. For LDH release, cell culture supernatants were collected, and LDH activity was detected using the LDH assay kit (Nanjing Jiancheng Biology Engineering Institute, Jiangsu, China). Briefly, 25 µl cell supernatant and 25 µl substrate were mixed together and incubated at 37°C for 15 min. Then, 25 µl of 2,4-dinitrophenylhydrazine was added into the samples and incubated at 37°C for 15 min. Finally, 250 µl of 0.4 mol/l NaOH solution was added and incubated with the mixture at room temperature for 5 min. Absorbance was measured at 450 nm on a spectrophotometric microplate reader. For Hoechst 33342/PI double staining, VECs were cultured in six-well plates at a density of 5 × 10⁵ cells/well. Then, the cells were transfected with TET2 lentivirus, miR-125a-5p inhibitor, or treated with drugs. Next, the cells were stained with 10 µl Hoechst 33342 solution at 37°C in the dark for 10 min, followed by staining with 5 µl PI at 37°C in the dark for 10 min. The stained cells were observed under a fluorescence microscope (Olympus Microscope IX3).

2.4 RNA extraction, retrotranscription, and real-time polymerase chain reaction (PCR)

MiRNA quantification was determined by using Bulge-Loop™ miRNA qRT-PCR Primer Set (one reverse-transcription (RT) primer and a pair of quantitative PCR primers for each set) specific for miR-125a-5p, which is designed by RiboBio (Guangdong, China). MiRNA expression was defined based on C_t, and relative expression levels were calculated as 2^{−[(Ct of miR-125a-5p) – (Ct of U6)]} after normalization with reference to the expression of small nuclear RNA U6. The extracted RNA was pretreated with RNase-free DNase, and 500 ng of RNA from each sample was used for cDNA synthesis primed with the specific miRNA RT primer (Supporting Information Table 3).

2.5 Luciferase reporter assays

Luciferase reporters containing wild-type or mutated miR-125a-5p plasmid were constructed using psi-CHECK2 vectors (Promega, Madison, WI). The luciferase vector (100 ng) containing the pcDNA3.1-miR-125a-5p plasmid was cotransfected with miR-125a-5p mimic, miR-125a-5p inhibitor, or NC (scramble sequence of miR-125a-5p; Supporting Information Table 1) into human embryonic kidney 293 cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) with 10 ng of Renilla luciferase reporter used as an internal control. After 48 hr, the cells were collected and lysed. Luciferase activities were determined using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s instructions.

2.6 Western blot analysis

Cells were washed twice with chilled PBS and lysed with radioimmunoprecipitation assay buffer and phenylmethanesulfonyl fluoride (94:6) on ice for 30 min and centrifuged (9,500g, 15 min). Nuclear proteins and cytoplasmic protein were extracted by Nuclear and Cytoplasmic Protein Extraction Kit (KeyGen) according to the manufacturer’s instructions. The total proteins were extracted, after which their concentrations were determined using a BCA Kit (CWBio, Peking, China). Briefly, protein samples of 20 µg/well were loaded in 8–12% sodium dodecyl sulfate polyacrylamide gel and transferred onto a nitrocellulose filter membrane and subsequently blocked by 5% nonfat milk dissolved in PBS for 2 hr. The membrane was probed with primary antibodies against TET2 (1:1,000; Abcam, MA), caspase-1 (1:500; Proteintech, Rosemont), IL-1β (1:500; Proteintech), NLRP3 (1:500; ABclonal, MA), NF-κB p65 (1:1,000; Abcam), NF-kB p65 (phospho S53, 1:1000; Abcam), IkBα (1:1,000; Abcam), Histone H3 (1:5,000; Abcam) and diluted in Tris-buffered saline with Tween^® 20 containing 2.5% skim milk buffer at 4°C overnight. Membranes were then washed five times in PBS with Tween 20 and incubated with fluorescence-conjugated antirabbit IgG secondary antibody at room temperature for 1.5 hr (1:2,000). β-Actin (Proteintech) was used as an internal control. Western blot bands were quantified using the Quantity One Software (California, CA) by measuring band intensity (area × optical density).

2.7 Immunofluorescence

Immunofluorescence staining was performed to detect the expressions of caspase-1, 5hmC, and NF-κB p65 in endothelial cells. Briefly, the cells were fixed with 4% paraformaldehyde for 30 min, penetrated by 0.1% Triton X-100 for 30 min, and then blocked with 5% goat serum for 1 hr. Subsequently, the cells were incubated with anti-caspase-1 (1:250; Abcam), anti-5hmC (1:200; Proteintech), or NF-κB p65 (1:250; Abcam) antibody at 4°C overnight, followed by incubation with Alexa Fluor-conjugated secondary antibody (Invitrogen) in the dark for 1 hr. The nuclei were stained by 4′,6-diamidino-2-phenylindole (Beyotime, Shanghai, China) for 5 min. The cells were observed under a fluorescence microscope (Olympus Microscope IX3, Tokyo, Japan).

2.8 Enzyme-linked immunosorbent assay (ELISA)

IL-1β secretion in VECs detected by ELISA. In brief, VECs were cultured in 24-well plates. Following treatment, cellular supernatants were collected and 100 μl of undiluted sample was used for ELISA analysis (NeoBioscience, Guangzhou, China). All of the measurements were performed according to the manufacturer’s protocol.

2.9 Bioinformatics analysis

We obtained genomic and miRNA sequences from miRDB (http://mirdb.org/miRDB/), which predicts miRNA targets in various species. MiR-125a-5p binding sites were identified using the TargetScan (http://www.targetscan.org/) and miRNA (http://www.microrna.org/microrna/) web servers. Hybridization minimum free energy was calculated using the RNAhybrid (http://bibiserv.techfak.uni-bielefeld.de/bibi/Tools.html) website. We use the following websites for CpG island analysis: MethPrimer (http://www.urogene.org/cgi-bin/methprimer/methprimer.cgi) and UCSC (http://genome.ucsc.edu/cgi-bin/hgGateway).

2.10 Mitochondrial membrane potential detection

In mitochondrial dysfunction (MDF), the transmembrane potential (Δψ) is disrupted, resulting in the change in the concentration of 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolcarbocyanine iodide (JC-1) entering the mitochondria. The use of flow cytometry to detect the intensity of green/red fluorescence can reflect the mitochondrial membrane potential and indirectly reflect mitochondrial function. Briefly, the treated cells were washed twice with PBS and centrifuged at 2,000 rpm for 5 min. Cells were suspended in 500 µl of JC-1 working solution and incubated at 37°C in a 5% CO₂ incubator for 15 min. Centrifugation was performed at room temperature (2,000 rpm, 5 min); cells were harvested and resuspended in 500 µl of 1× incubation buffer. Finally, detection was achieved by flow cytometry (E_x = 488 nm; E_m = 530 nm).

2.11 Statistical analysis

The data in this study are shown as means ± SEM differences among groups were analyzed using one-way analysis of variance accompanied with Newman–Keuls multiple comparisons test (GraphPad Prism version 5.0, CA). Two-tailed Student’s t test was used for comparison between two groups; p < 0.05 was considered statistically significant.

3.1 oxLDL induces VEC pyroptosis

oxLDL is a well-known atherogenic factor (Glass & Witztum, 2001). VECs are the barriers between the blood and the vascular wall, and damage to VECs is considered the starting point of As lesions (Sitia et al., 2010). To elucidate the relationship between oxLDL and pyroptosis, we used VECs for in vitro experiments. In VECs incubated with different concentrations of oxLDL (0, 25, 50, and 100 µg/ml) for 24 hr, the expressions of caspase-1, NLRP3, and proinflammatory cytokines (IL-1β) significantly increased (Figure 1a–c). We also used immunofluorescence to examine the effect of different concentrations of oxLDL on the expression of caspase-1 in VECs (Figure 1d). During pyroptosis, pores can be formed in the cell membrane and cause the release of cellular contents and positive staining of dead cells, as can be determined by PI staining (Miao, Rajan, & Aderem, 2011). To further characterize oxLDL-induced pyroptosis of VECs, we double-stained Hoechst 33342 and PI in VECs. Our results showed that oxLDL-induced pore formation and membrane rupture, as indicated by the increased extensive PI-positive staining of cells (Figure 1e,f). To clarify the effect of different concentrations of oxLDL on the death of VECs, we used LDH release assay. The results suggest that oxLDL upregulates the release of LDH in a concentration-dependent manner (Figure 1g). These results show that oxLDL concentration-dependently induces the pyroptosis of VECs. The optimum oxLDL concentration is 100 µg/ml. Thus, we used this concentration in our follow-up experiments.

VECs were incubated with 100 µg/ml oxLDL for 0, 6, 12, and 24 hr to investigate whether oxLDL can induce pyroptosis of VECs in a time-dependent manner. Western blot results (Figure 1h–j) showed that the protein expression of NLRP3, caspase-1, and IL-1β significantly upregulated in VECs after 24 hr of incubation with oxLDL. With prolongated incubation time, the expressions of NLRP3, caspase-1, and IL-1β increased. The strongest effect was observed in the 24 hr group. To clarify the effect of different treatment times of oxLDL on the death of VECs, we used LDH release assay. The results suggest that oxLDL upregulates the release of LDH in a time-dependent manner (Figure 1k). The above results indicate that oxLDL can induce pyroptosis of VECs in a time- and concentration-dependent manner.

3.2 oxLDL downregulates TET2 expression in a concentration-dependent manner

TET2 acts as a DNA demethylase and can inhibit As to mediate the protective effects of hydrogen sulfide on the oxLDL-induced dysfunction of VECs (Peng et al., 2017). Our previous study showed that TET2 upregulates the level of autophagy against ApoE^−/− mice As lesions (Peng et al., 2016), whereas the absence of TET2 accelerates the development of As (Fuster et al., 2017). Autophagy and pyroptosis are both associated with PCD and play important roles in atherosclerotic lesions. Therefore, we speculate that TET2 may be involved in oxLDL-induced VECs pyroptosis. Thus, we investigated the changes in TET2 during oxLDL-induced VECs pyroptosis, and the results suggest that oxLDL downregulated TET2 in a concentration-dependent manner (Figure 2).

3.3 Pyroptosis is curbed when TET2 is overexpressed

We conducted further experiments to further clarify the effect of TET2 on oxLDL-induced VECs pyroptosis. First, we tested the TET2 overexpression and interference efficiency. Immunofluorescence showed that the transfection efficiency was greater than 70% (Supporting Information Figure S1). The western blot results showed that TET2 overexpression and interfering lentivirus can significantly increase or reduce TET2 protein expression levels, respectively (Figure 3a,b). Further, we examined the effect of TET2 overexpression on VECs pyroptosis treated by oxLDL. As a result, in the VECs treated with oxLDL, overexpression of TET2 can inhibit the expression of pyroptosis-related proteins (NLRP3, caspase-1, and IL-1β; Figure 4a–c,e), reduce the release of LDH (Figure 4d), significantly reduce the number of PI-positive cells (Figure 4f,g). These results indicate that TET2 inhibits pyroptosis in VECs treated with oxLDL.

3.4 oxLDL increases the methylation level of VECs and can subside via overexpression of TET2

TET2, as a demethylase, is an important participant in the epigenetic modification. Epigenetic modification abnormalities are closely related to cell function and fate. Thus, we hypothesized that TET2 mitigate the oxLDL-induced pyroptosis of VECs may be associated with DNA methylation. oxLDL leads to DNA hypermethylation (including global DNA and mitochondrial DNA [mtDNA]). Subsequently, overmethylation of DNA causing MDF leads to an increase in ROS production. ROS increase causes NLRP3 inflammasome activation and promotes VEC death. Therefore, we examined the overexpression of TET2 on DNA methylation. 5hmC is a hydroxylated form of 5mc and represents the level of demethylation of DNA. Using 5hmC antibody for immunofluorescence, the results showed that oxLDL can reduce the level of 5hmC. Overexpressing TET2 can increase the level of 5hmC (Figure 5a), indicating that oxLDL can increase DNA methylation and overexpress TET2 can inhibit DNA methylation. However, the role of epigenetic regulation in oxLDL-induced pyroptosis of VECs requires more in-depth exploration, which we will explain in the discussion.

3.5 Overexpression of TET2 can improve mitochondrial membrane potential abnormalities and reduce ROS generation caused by oxLDL

Epigenetic modifications (including methylation modifications) play a crucial role in cell survival. Studies have shown that nuclear and mtDNA methylation can affect mitochondrial function through S-adenosyl methionine (Wallace & Fan, 2010). Numerous studies have shown that mitochondria are closely related to cell death, and the disruption of mitochondrial transmembrane potential (Δψ) is considered to be one of the earliest times of the apoptotic cascade. This disruption occurs before the appearance of apoptotic features in the nucleus (chromatin condensation and DNA fragmentation). Once the mitochondrial transmembrane potential collapses, apoptosis becomes irreversible. Pyroptosis is similar to chromatin condensation and DNA fragmentation. Thus, we examined the effect of oxLDL, overexpression of TET2 on mitochondrial membrane potential. Flow cytometry results showed that oxLDL significantly increased the ratio of Q3 district/Q2 district compared with the control group, indicating that oxLDL may cause mitochondrial membrane potential abnormalities. Overexpression of TET2 can reduce the ratio of Q3 district/Q2 district compared with the oxLDL-treated group (Figure 5b), indicating that the overexpression of TET2 may extenuate Δψ abnormalities induced by oxLDL. These results indicate that oxLDL-induced pyroptosis of VECs is associated with MDF.

3.6 Pyroptosis is alleviated by ROS elimination

The mitochondrial membrane potential partially reflects mitochondrial function; the production of ROS mainly originates from mitochondria, and MDF can lead to increased ROS production (Desoti et al., 2012). ROS are important mediators for inflammasome activation (Guo, Callaway, & Ting, 2015). To elucidate the possible role of ROS in oxLDL-induced pyroptosis in VECs, we first detected the changes in ROS levels in VECs treated with oxLDL (100 µg/ml). Our results demonstrated that oxLDL treatment dramatically increased ROS levels, TET2 overexpression can inhibit the production of ROS (Figure 5c), which is consistent with its improved mitochondrial function. oxLDL-induced ROS also was buffered by NAC, a ROS inhibitor (Figure 6a). NAC also inhibited oxLDL-induced protein levels of NLRP3, caspase-1, and inflammatory cytokine IL-1β (Figure 6b–d,f). NLRP3 is the most important part of the inflammatory platform in pyroptosis. The N-terminal pyrin segment of NLRP3 serves as a scaffold to nucleate apoptosis-associated speck-like protein containing a caspase activation and recruitment domain (ASC), which contains a pyrin domain and a caspase activation and recruitment domain (CARD). Through its pyrin domain, ASC interacts with sensor molecules, and the CARD domain interacts with procaspase-1 and initiates procaspase-1 self-cleavage to form a caspase-1 mature body. On the one hand, activated caspase-1 recognizes the inactive IL-β and IL-18 precursors and converts them into mature inflammatory cytokines. In contrast, caspase-1 cleaves gasdermin D (GSDMD) to form 31-kDa-sized amino-terminal products (GSDMD-N) that mediate membrane pore formation. After membrane pore formation, the release of inflammatory factors is promoted, and cells swell, finally inducing pyroptosis (Afonina, Zhong, Karin, & Beyaert, 2017). Thus, the above result suggests that NLRP3 inflammasome activation depends on ROS generation. Additionally, NAC pretreatment reduced the number of host 33342 and PI double-positive cells (Figure 6g,h) LDH release in oxLDL-treated VECs (Figure 6e), suggesting the importance of ROS in oxLDL-induced VECs pyroptosis.

3.7 NF-κB involved in the protective effect of TET2 on oxLDL-induced VEC pyroptosis

Numerous studies have confirmed that ROS can activate NF-κB, NF-κB phosphorylation accounts for NLRP3 priming and the secondary step of activation (Luo et al., 2017). Thus, we further analyzed the effect of TET2 on NF-κB activity. Transfection of TET2 overexpression lentivirus led to the inhibition of IkBα degradation and NF-κB p65 nuclear translocation, whereas transfection with TET2 shRNA lentivirus significantly promoted IkBα degradation and NF-κB p65 nuclear translocation in oxLDL-treated VECs (Figure 7b–e). Furthermore, we examined the effect of TET2 on NF-κB p65 phosphorylation. Our results showed that TET2 overexpression can reduce the levels of phosphorylated Ser536 of NF-κB p65 (p-p65), which corresponds to its active form, and instead, TET2 sh lentivirus can increase the phosphorylation of NF-κB p65 (Figure 7a). Thus, these data point to an inhibitory role of TET2 in NF-κB activation in oxLDL-treated VECs and may partly explain why TET2 inhibits oxLDL-induced VECs pyroptosis.

To further verify the role of NF-κB in oxLDL-induced VEC pyroptosis, we treated NF-κB inhibitor BAY 11–7082 (20 µM) for 2 hr and then added oxLDL (100 µg/ml) for 24 hr. Western blot detected NLRP3, caspase-1, and IL-1β protein expressions levels. The results show that NF-κB inhibition can decrease the expression of NLRP3, caspase-1, and proinflammatory cytokine IL-1β (Figure 8a–d) and LDH release (Figure 8e). BAY pretreatment also reduced the number of PI-positive cells (Figure 8f,g). These results indicate that oxLDL-induced endothelial cell pyroptosis requires NF-κB activation.

3.8 Prediction and verification of the miR-125a-5p target site in the TET2 3′-untranslated region (3′-UTR)

Next, we further explored the molecular mechanism of downregulation of TET2 by oxLDL. The miRNA usually binding to the 3′-UTR of target messenger RNAs (mRNAs) to inhibit mRNA translation at the posttranscriptional level may participate in the downregulation of TET2 by oxLDL. Through bioinformatics analysis, we observed that miR-125a-5p can bind to the 3′-UTR of TET2 mRNA. First, we analyzed the conservation of miR-125a-5p using the miRDB database. The results showed that miR-125a-5p is highly conserved among various species, suggesting that miRNAs are important in the evolution of species and in modulating target gene expression (Figure 9a). Then, the TargetScan, MicroRNA, and RNAhybrid databases were utilized to predict the miR-125a-5p target sites in the 3′-UTR of TET2 transcripts. The results indicated that miR-125a-5p can bind to the TET2 3′-UTR. The TargetScan and MicroRNA websites identified a binding site for miR-125a-5p within the TET2 3′-UTR (Figure 9b). In addition, a free energy score of −30.1 kcal/mol (human) was calculated in both humans and mice on RNAhybrid (Figure 9c,d), showing that miR-125a-5p may form a stable association with TET2 mRNA. Previous studies have found that miR-125a-5p is increased (Simionescu et al., 2014), while TET2 is reduced in hyperlipidemic (Peng et al., 2016, 2017). Suggesting a negative association between miR-125a-5p and TET2. Based on the support of bioinformatics and existing literature, we investigated the relationship between miR-125a-5p and TET2 in oxLDL-induced VEC pyroptosis. We performed a luciferase reporter gene assay and noted that miR-125a-5p mimic can inhibit the expression of TET2, whereas miR-125a-5p inhibitor can increase the expression of TET2 (Figure 9e–g). These data indicated that miR-125a-5p can regulate TET2 expression at the posttranscriptional level by targeting the TET2 3′-UTR.

3.9 oxLDL upregulates miR-125a-5p expression and inhibits miR-125a-5p mitigation of oxLDL-induced VEC pyroptosis

Existing research findings show that circulating miR-125a-5p levels increased in hyperlipidemic (HL) and/or hyperglycemic (HG) conditions, suggesting the contribution of these miRNAs to the atherosclerotic process (Simionescu et al., 2014). However, the relationship between oxLDL and miR-125a-5p in VEC-related pyroptosis remains unclear. Therefore, we examined the effect of different concentrations of oxLDL on the expression of miR-125a-5p in VECs by PCR. The results showed that the expression of miR-125a-5p increased with the increase in oxLDL concentration (Figure 10a). To further clarify the effect of miR-125a-5p on oxLDL-induced VEC pyroptosis, we conducted further experiments regarding the inhibition of miR-125a-5p. The result showed that miR-125a-5p inhibitor significantly reduced the expression of miR-125a-5p and miR-125a-5p mimic significantly increased the expression of miR-125a-5p (Figure 10b,c). Furthermore, transfection of miR-125a-5p inhibitor can reduce the expression of pyroptosis-related proteins (NLRP3, caspase-1, and IL-1β; Figure 10d,e,f,h) and the release of LDH (Figure 10g), significantly reduce the number of PI-positive cells (Figure 10i,j). These results demonstrate that inhibition of miR-125a-5p mitigates pyroptosis of VECs induced by oxLDL.

3.10 MiR-125a-5p inhibitor reduces ROS production by improving DNA methylation and mitochondrial function

To determine whether miR-125a-5p inhibitor inhibits oxLDL-induced VEC pyroptosis consistent with overexpression of TET2, we examined the effects of miR-125a-5p inhibitor on DNA methylation, mitochondrial function, and ROS production. The results showed that miR-125a-5p inhibitor partially abolished the inhibition of DNA demethylation by oxLDL (Figure 11a), improved mitochondrial function (Figure 11b), reduced ROS production (Figure 11c). However, miR-125a-5p inhibitor transfection had almost no effect on oxLDL-caused DNA hypermethylation, abnormal membrane potential and ROS production, when transfection of the TET2 sh lentivirus simultaneously. This indirectly suggests that miR-125a-5p inhibitor improves mitochondrial function and inhibits ROS production, at least in part dependent on increased expression of TET2.