The mouse microvascular endothelial cell (MVEC) line EOMA was purchased from ATCC and cultured as we recently described (12, 13). MVECs were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco, USA), containing 10% of fetal bovine serum (Gibco, USA) and 1% penicillin–streptomycin (Gibco, USA). The cells were cultured in a humidified incubator at mixture at 37 °C with 5% CO2 and 95% air. Cells were passaged by trypsinization (Trypsin/EDTA; Sigma, USA), followed by dilution in DMEM medium containing 10 % fetal bovine serum. Sodium palmitate (Sigma) was prepared as previously described (14).

The barrier function of the endothelial monolayer was assessed by determining the transendothelial electrical resistance (TEER) at low frequency (4000 Hz). Briefly, MVEC were grown on gold electrodes (in 8W10+E arrays) till confluence and maintained in medium with 1% FBS. Cells were challenged with palmitate (50 μM) with or without simvastatin (1-10 μM). TEER was measured using electrical cell-substrate impedance sensing (ECIS) ZƟ system (Applied Biophysics). The changes of TEER for duration of 12 hours were determined. The barrier function was also confirmed by FITC-dextran transwell assay as described previously (10). MVECs were cultured in 24-well transwell plates till confluence and maintained in medium with 1% FBS and treated as indicated for 12 hr. The fluorescent intensity in each well was determined at excitation/emission of 485/530 nm using a fluorescent microplate reader (BMG Labtech).

The cell viability was assessed by lactate dehydrogenase (LDH) assay. LDH activity in the cell culture medium was determined by PierceP^tmP LDH Assay Kit (Thermo, IL, USA) according to the manufacture’s protocol. The absorbance was spectrophotometrically quantified at 490 nm using CLARIOstar microplate reader (BMG Labtech, Germany). The maximum LDH activity (LDHR_maxR) was obtained from samples treated with lysis buffer. The basal level of LDH activity (LDHR₀R) due to spontaneous LDH release was obtained from control samples. The survival was calculated as percentage of (LDHR_maxR-LDHR_sampleR) over (LDHR_maxR-LDHR₀R).

Cells were grown on eight-well chamber slides till confluence and then treated as indicated and fixed in 4% paraformaldehyde for 15 minutes. Cells were washed in phosphate-buffer saline (PBS), permeabilized in 0.3% Triton X-100/PBS for 15 min, and then incubated for 2 hours at 4°C with incubated with rabbit, rat, or mouse anti-ZO-1 (1:100; Invitrogen), VE-cadherin (1:100; Abcam), anti-TFEB (1:100; Mybiosource). Immunofluorescent staining was performed by incubating slides with Alexa Fluor 488 or Alexa Fluor 555-labeled secondary antibody (1:100, Invitrogen) for 1 hour at room temperature. The samples were fixed, mounted, and visualized using an inverted microscope (Olympus IX73 Fluoview DP-80, Japan).

Cells were lysed using sucrose buffer (20 mM HEPES, 1mM EDTA, 255 mM sucrose, cocktail o protease inhibitors (Roche), pH 7.4) and then whole cell lysates were obtained by centrifugation at 1,000 g for 5 min. The cytosolic fractions were prepared by centrifugation of whole cell lysates at 17,000 g for 30 min. After boiling for 5 min at 95 °C in a 2× loading buffer, 30 μg of total proteins were separated by a 10% or 15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The proteins of these samples were then electrophoretically transferred at 100 V for 1 hour onto a PVDF membrane (Bio-Rad, USA). The membrane was blocked with 5% nonfat milk in Tris-buffered saline-Tween 20. After washing, the membrane was probed with 1:1000 dilution of primary mouse, rat, or rabbit antibodies against anti-ZO-1, VE-cadherin, HMGB1 (Abcam), or following antibodies from Cell signaling including cathepsin B, LC3I/II, Beclin-1, phospho-p70S6 kinase, p70S6 kinase, GAPDH, β-tubulin or β-actin overnight at 4 °C followed by incubation with IRDye fluorescent dyes (IRDye 680RD and IRDye 800CW, 1:15000). The immuno-reactive bands were detected by Licor Odyssey Fc (Licor, USA) instrument methods visualized on the Odyssey Fc Imager. Densitometric analysis of the images was performed using the Image StudioP^TMP Software (Licor, USA) as we described previously (15, 16).

FLICA™ (Fluorescent Labeled Inhibitor of Caspases) probes were used to detect active caspase-1 enzyme as described (12). Cells were incubated with FLICA reagent (1:100 dilutions in PBS) from a FLICA™ Assay Kit (ImmunoChemistry Technologies, LLC, Bloomington, MN) for 1.5 hours at room temperature and then washes three times in PBS. The green fluorescent signal from FLICA™ probes were analyzed with a fluorescence plate reader and used to represent the relative enzyme activity of caspase-1 in cells.

MVECs cultured in eight-well chambered coverslips were treated as indicated and then incubated with 1 μM Lysotracker DND-99 (Invitrogen) in culture medium for 10 min at 37 °C. Cells were washed with fresh medium for three times and analyzed using an inverted microscope (Olympus IX73 Fluoview DP-80, Japan).

To analyze the autophagic flux, MVECs were transfected with Premo™ Autophagy Tandem Sensor RFP-GFP-LC3B (Thermo Fisher Scientific) as described previously (17, 18). After 48 h, cells were treated with 10 μM simvastatin, 10 μM chloroquine (Sigma), or 200 µM Leupeptin A (Sigma) for 18 h. Cells were fixed in freshly prepared 4% PFA, washed in PBS, and then mounted. Autophagic flux was determined by evaluating patterns of GFP and RFP dots using an inverted microscope (Olympus IX73 Fluoview DP-80, Japan).

MVECs cultured in eight-well chamber slides were treated as indicated and then incubated with 2 μg/mL acridine orange (Cayman chemical) for 17 min at 37 °C, rinsed with PBS. Cells in chamber slides were immediately analyzed and photographed. Gly-Phe-β-naphthylamide (GPN, Cayman chemical), a lysosome-disrupting agent, was used as a positive control for increased lysosome permeability. Then the intensity ratio of red-to-green fluorescence of cells was obtained by an Acoustic Focusing Cytometer (Attune NxT, ThermoFisher Scientific, USA).

Small interference RNAs (siRNAs) for TFEB gene (siTFEB) and scramble control siRNA were commercially available (Santa Cruz Biotechnology, CA). Transfection of siRNA was performed using the siLentFect Lipid Reagent (Bio-Rad, CA, USA) according to the manufacturer’s instructions.

Total RNA from cells was extracted with Aurum Total RNA isolation kit (Bio-Rad) according to the manufacturer’s protocol. One-microgram aliquots of total RNA from each sample were reverse-transcribed into cDNA by using a first-strand cDNA synthesis kit (Bio-Rad). Equal amounts of the reverse transcriptional products were subjected to PCR amplification PrimePCR™ SYBR® Green Assay on a CFX Connect Real-Time PCR Detection System (Bio-Rad). The sequences of primers for target genes are listed in Table 1.

Data are presented as means ± standard deviation. Significant differences between and within multiple groups were examined using one-way ANOVA test followed by Bonfferroni’s multiple-range test. A Students t-test was used to detect significant difference between two groups. The statistical analysis was performed by GraphPad Prism 6.0 software. P<0.05 was considered statistically significant.

To examine the effects of simvastatin on endothelial hyperpermeability, we first used ECIS technology to measure the changes of TEER in response to 50 μM palmitate for 12 hours in presence or absence of simvastatin. As shown in Figure 1A, palmitate reduced the TEER of the MVEC monolayer and simvastatin significantly inhibited this response at 10 μM. In contrast, simvastatin at 1 μM has only negligible effect. FITC-dextran transwell assay was then used to confirm the cell permeability to high-molecular-weight molecules. As shown in Figure 1B, we observed significantly increased FITC-dextran molecules across the MVEC monolayer indicating the increased permeability of endothelial cell monolayers by palmitate, and simvastatin reduced this effect. The changes in the endothelial permeability by 50 μM palmitate were not associated with increased cell death because no changes in cell viability were observed (Figure 1C). Together, these results suggest that simvastatin exerts protective effects on palmitate-induced endothelial hyperpermeability.

Figure 1

Simvastatin protects palmitate-induced endothelial hyperpermeability. MVECs were treated with palmitate (PA, 50 μM) in the presence of vehicle (PBS) or simvastatin (Sim, 1-10 μM) for 12 hours. (A) EC permeability was measured by transendothelial electrical resistance (TEER). (n=4). (B) EC permeability was determined by measuring the amount of FITC-dextran that crossed the endothelial monolayer cultured on the inserts of transwell (n=4). (C) Cell viability by LDH assay (n=4). *P<0.05 vs. untreated control; #P<0.05 vs. PA alone

Endothelial hyperpermeability is associated with down-regulation of junction protein ZO-1 and VE-cadherin (10, 11). By immunofluorescence studies, we found that simvastatin dramatically reversed palmitate-induced downregulation of tight junction protein ZO-1 and adherens junction protein VE-cadherin (Figure 2A). This inhibitory effect of simvastatin on ZO1 and VE-cadherin was confirmed by Western blot analysis (Figure 2B).

Figure 2

Simvastatin blunts palmitate-induced tight-junction dysfunction. MVECs were treated with palmitate (PA, 50 μM) in the presence of vehicle (PBS) or simvastatin (Sim, 10 μM) for 12 hours. (A) Representative fluorescence images show the cell membrane fluorescence of ZO-1 (Alexa Fluor 488) and or VE-cadherin (Alexa Fluor 555) from at least three independent experiments. (B) Representative immunoblot (top) and band intensity quantification (bottom) for ZO-1 and VE-cadherin expression in response to palmitate with or without simvastatin (n=4). β-actin was an internal loading control. *P<0.05 vs. untreated control; #P<0.05 vs. PA alone.

Our recent study demonstrated that palmitate-induced endothelial hyperpermeability was associated with increased Nlrp3 inflammasome activation (10), which caused increased release of HMGB1, a novel endothelial permeability factor (12, 19). The inflammasome activity induced by palmitate was attributed to events including lysosome destabilization and consequent lysosomal release of cathepsin B but not to reactive oxygen species (ROS) (10). In the present study, we examined whether or not simvastatin prevents palmitate-induced lysosome injury and Nlrp3 inflammasome activity. Lysotracker DND-99 is a lysosomotropic dye that stains lysosomes with acidic pH, and decreases in lysotracker DND-99 intensity suggest disruptions in lysosomal function, integrity, or quantity (20). As shown in Figure 3A, palmitate significantly decreases the lysotracker staining, which was preserved by simvastatin. We also found that simvastatin abolished the cytosolic expression of cathepsin B upon palmitate stimulation (Figure 3B), suggesting that simvastatin prevents lysosomal release of cathepsin B. Thus, these data indicate that simvastatin preserves lysosome integrity and prevents cathepsin B release. We then evaluated Nlrp3 inflammasome activity by measuring two marker events including the caspase-1 activity assay and IL-18 production. It was found that simvastatin markedly suppressed the palmitate-induced activation caspase-1 (Figure 3C) and the release of and IL-18 (Figure 3D). These data confirm that caspase-1 is activated by palmitate, which is blocked by simvastatin. As shown in Figure 3E, simvastatin also dramatically inhibited palmitate-induced release of HMGB1 protein into the culture media. As a comparison, no significant effects were observed in the HMGB1 level in whole cell lysates.

Figure 3

Simvastatin inhibits palmitate-induced lysosome injuries and Nlrp3 inflammasome activity. MVECs were treated with palmitate (PA, 50 μM) in the presence of vehicle (PBS) or simvastatin (Sim, 10 μM) for indicated hours. (A) Representative fluorescence images show the lysotracker DND-99 staining in acidic lysosomes (PA 12 hours) from three independent experiments. (B) Western blot analysis of cathepsin B (CatB) expression in cytosolic fractions (PA 12 hours) (n=4). β-tubulin is a loading control. (C) Summarized data show the caspase-1 activation (PA 12 hours) by FLICA assay (n=4). (D) Summarized data showing IL-18 production (PA 48 hours) by ELISA (n=4). (E) Western blot documents and summarized data showing the expression of HMGB1 in either cell culture medium or cell lysate (PA 12 hours) (n = 4). *P<0.05 vs. untreated control; #P<0.05 vs. PA alone

Simvastatin dose-dependently increased the protein expression of LC3-II, an autophagy marker (Figure 4A). Simvastatin can also increase LC3-II when autophagy flux was blocked by chloroquine (Figure 4A). We noticed that simvastatin did not affect Beclin1, a modulator for autophagosome assembly. These data suggest that simvastatin might increase autophagosome biogenesis but not enhance its assembly.

Figure 4

Simvastatin induces autophagy and lysosome biogenesis. (A) MVECs were treated with simvastatin (Sim, 0, 5, 10 and 20 μM) with or without chloroquine (CQ, 10 μM) for 18 h. Representative Western blots show the expression of autophagosome marker proteins LC3II, Beclin-1, or β-actin. Summarized data show the relative expression of LC3II or Beclin-1 compared to β-actin (n=4). *P<0.05 vs. untreated control; #P<0.05 Sim+CQ vs. Sim alone. (B) MVECs were transfected with tandem RFP-GFP-LC3B and then treated with PBS control (Ctrl), simvastatin (10 μM), chloroquine (10 μM), or leupeptin A (Leup, 200 μM) for 18 h. Representative fluorescence images show the autophagosomes and autophagolysosomes from at least three independent experiments. Autophagosomes were visualized as yellow or orange puncta (RFP-GFP-LC3) in merged images, whereas red puncta (RFP-LC3) in merged images represent autophagolysosomes since acidification abolishes green fluorescence. Sample without either simvastatin or chloroquine was considered as vehicle control. (C) MVECs were treated without or with simvastatin (Sim, 10 μM) in the absence or presence of autophagosome inhibitor Spautin-1 (Sp-1, 10 μM) for 18 h. MVECs were also treated without or with simvastatin for 16 h and then GPN (200 μM) was added and incubated for additional 4 h. Representative fluorescence images show the acridine orange staining in acidic organelles (lysosomes and autophagolysosomes). Summarized data show the red fluorescence intensity of acridine orange staining by flow cytometer analysis (n=6). *P<0.05 vs. Veh Ctrl; #P<0.05 vs. Sp-1 alone. (D) and (E) Summarized data show the effects of simvastatin (Sim, 10 μM) and/or Spautin-1 (Sp-1, 10 μM) on palmitate (PA, 50 μM for 12 hours)-induced caspase-1 activation by FLICA assay or EC permeability by FITC-dextran assay (n=4). *P<0.05; n.s. (not significant)

We next monitored the autophagic flux by expressing tandem RFP-GFP–LC3 genes. The tandem RFP-GFP-LC3B protein includes an acid-insensitive RFP and an acid-sensitive GFP. The expression pattern of GFP-RFP-LC3 constructs is dependent on the fusion of autophagosomes with lysosomes, lysosomal acidification and degradation capacity (21). As shown in Figure 4B, simvastatin elevated the number of yellow/orange puncta (autophagosomes are RFPP⁺PGFPP⁺P) and the red puncta (autophagolysosomes are RFPP⁺PGFPP^-P) in the merged image compared to control. We further determined the tandem RFP-GFP-LC3 expression pattern in merged images by two autophagic flux inhibitors. First, chloroquine neutralizes lysosomal pH and inhibits lysosome fusion with autophagosome, and thereby results in increased autophagosome accumulation and but decreased formation of autophagolysosomes. Consistently, chloroquine increased the number of yellow/orange punta and red puncta were barely observed, suggesting that it abolished the formation of autophagolysosomes and thus accumulated autophagosomes. Leupeptin is an autophagic flux inhibitor that blocks lysosome protease activity and thereby inhibits autophagolysosome degradation without affecting lysosomal pH. As expected, leupeptin increased the number of the red puncta (autophagolysosome) as well as the yellow/orange puncta (autophagolysosome) reflecting a defective autophagolysosomal degradation pathway. We noticed that simvastatin-treated cells showed more numbers of red puncta than leupeptin (Figure 4B), suggesting that simvastatin also increases lysosome biogenesis.

Considering lysosome itself as a cargo for autophagy pathway, we examined whether or not inhibition of autophagy induction could result in lysosome accumulation. Acridine orange stains acidic organelles including lysosomes and autophagolysosomes. We found that spauntin-1, an autophagy inhibitor, significantly increased acridine orange intensity, suggesting that spautin-1 blocked autophagy-dependent lysosome degradation at basal condition (Figure 4C). Simvastatin further enhanced acridine orange intensity when autophagy is inhibited, suggesting that simvastatin increased lysosome biogenesis. However, simvastatin alone had no significant effect. These data suggest simvastatin induces autophagy signaling that promotes lysosome degradation; meanwhile, simvastatin increases lysosome biogenesis. We further found that the inhibitory effects of simvastatin on palmitate-induced Nlrp3 inflammasome activity (Figure 4D) and permeability (Figure 4E) were reversed when autophagy is inhibited by spautin-1.

Transcriptional factor EB (TFEB) is a master regulator involved in autophagy signaling and lysosomal biogenesis (22, 23). As shown in Figure 5A, simvastatin markedly increased TFEB expression in the nuclei implicating that simvastatin induces nuclear translocation of TFEB and consequent transcription of genes involved in autophagy and lysosome biogenesis. We also demonstrated that simvastatin induced up-regulation of genes involved in TFEB-autophagy-lysosome signaling including TFEB, LC3, and LAMP2a, which was blocked by TFEB gene silencing (Figure 5B). Inhibition of mTOR has been shown to enhance TFEB activity (15, 24, 25). The mTOR activity is determined by the phosphorylation of S6K. However, simvastatin had no effects on S6K phosphorylation (data not shown). It is suggested that simvastatin may activate TFEB pathway in mTOR-independent manner. Consistently, TFEB gene silencing also prevented the inhibitory effects of simvastatin on palmitate-induced Nlrp3 inflammasome activity (Figure 5C) and permeability (Figure 5D).

Figure 5

Simvastatin induces activation of TFEB in MVECs. (A) Representative immunofluorescence images of TFEB show the nuclear translocation of TFEB by simvastatin (Sim, 10 μM for 2 h) from four independent experiments. (B) Real-time PCR quantification shows the effects of simvastatin (10 μM for 2 h) on mRNA levels of TFEB, LC3, and LAMP2a in MVECs that were transfected with scramble (Scr) or TFEB siRNA (siTFEB) for 24 h (n=4). *P<0.05 vs. Scr Ctrl; #P<0.05 vs. Scr+Sim. (C) and (D) Summarized data show the effects of simvastatin (10 μM) on PA (50 μM, 12 hours)-induced caspase-1 activation by FLICA assay or EC permeability by FITC-dextran assay in MVECs that were transfected with scramble (Scr) or TFEB siRNA (siTFEB) for 24 h (n=4). *P<0.05; n.s. (not significant).