58-F, a flavanone from Ophiopogon japonicus, prevents hepatocyte death by decreasing lysosomal membrane permeability

Lysosome membrane permeabilization (LMP) has been implicated in cell death. In the present study, we investigated the relationship between cell death and H2O2-/CCl4-induced LMP in hepatocytes in vitro and following acute liver injury in vivo. The key finding was that H2O2 triggered LMP by oxidative stress, as evidenced by a suppression of LAMP1 expression, a reduction in LysoTracker Green and AO staining, and the leakage of proton and cathepsin B/D from the lysosome to the cytoplasm, resulting in cell death. CCl4 also triggered hepatocyte death by decreasing lysosome LAMP1 expression and by inducing the accumulation of products of peroxidative lipids and oxidized proteins. Furthermore, a novel compound 5,8-dimethoxy-6-methyl-7-hydroxy-3-3(2-hydroxy-4-methoxybenzyl) chroman-4-one (58-F) was extracted from Ophiopogon japonicus and served as a potential therapeutic drug. In vivo and in vitro results showed that 58-F effectively rescued hepatocytes by decreasing LMP and by inducing lysosomal enzyme translocation to the cytosol.

The present study used the CCl 4 -induced mouse liver injury model and the H 2 O 2 -induced BNL CL.2 hepatocyte cell line injury model to test the hypothesis in vitro and in vivo, respectively, that ROS-induced LMP leads to leakage of the lysosomal contents into the cytosol and triggers cell death in hepatocytes and that the compound 58-F may play a protective role in this process. Our findings demonstrate that LMP and the consequent leakage of the lysosomal contents into the cytosol lead to cell death and that 58-F protected hepatocytes by blocking LMP. The results support our hypothesis, identifying LMP as the primary cause for ROS-induced death of hepatocytes and supporting the protective role of 58-F.

58-F Attenuates Lysosomal Damage after CCl 4 -induced Acute Liver Injury.
Haematoxylin and eosin (H&E) staining revealed normal hepatic architecture within the clear hepatic lobule and no dead hepatocytes. However, a large area of dead hepatocytes was detected in the central vein area of the CCl 4 -treated mice; the number of dead hepatocytes was significantly decreased following 58-F pretreatment (Fig. 1A). ALT is an enzyme in the cytoplasm of normal hepatocytes. When the integrity of the cell membrane is damaged, ALT is released from the hepatocytes into the serum. Therefore, ALT can be used an index of liver injury in the clinic and for in vivo studies 12 . The administration of CCl 4 to mice resulted in severe hepatotoxicity, reflected by a significantly elevated ALT level (p < 0.01). Pretreatment with 58-F prevented CCl 4 -mediated toxicity by decreasing the ALT level (p < 0.05) (Fig. 1B).
Malondialdehyde (MDA) is a final product of lipid peroxidation, so its level can be used as a biomarker of lipid peroxidation in the liver 13 . The protein carbonyl content (PCC) is used to evaluate the total level of protein oxidation. The MDA level and PCC following CCl 4 treatment both increased significantly (p < 0.01). Pretreatment with 58-F reduced both MDA levels and the PCC (p < 0.01) ( Fig. 1C-E).
LAMP1, used as a lysosomal marker, is expressed largely in the endosome-lysosome membranes of cells 3 . Both LAMP1 mRNA and protein levels decreased significantly after CCl 4 treatment (p < 0.01), and 58-F pretreatment increased both mRNA and protein levels (p < 0.01) (Fig. 1F-H). These results suggested that lysosomes mediated CCl 4 -induced hepatocyte injury.

58-F Attenuates H 2 O 2-induced Cell Viability, Death, and Restores the Inhibition of Proliferation.
Cell viability under various concentrations of H 2 O 2 (0, 100, 200, 300, 400, 500, 1000, 2000, and 2500 μ M) is illustrated in Fig. 2A. The cell viability was inhibited by H 2 O 2 at concentrations in the range of 300 μ M∼ 500 μ M, with a number of dead cells detected at concentrations in the range of 1000 μ M∼ 2500 μ M. The inhibition rate of 500 μ M H 2 O 2 was approximately 30∼ 50% (Fig. 2C,D). The cells were exposed to 58-F at concentrations of 0, 1, 10, 50,100 or 200 μ M for 24 h. The results revealed that 58-F concentrations from 1 μ M to 200 μ M were not cytotoxic to hepatocytes (Fig. 2B). The cells were effectively protected by 58-F at different concentrations ranging from 10 μ M to 100 μ M. The time-course for the effectiveness of 58-F (50 μ M) ranged from 6 h to 72 h of pretreatment before the addition of 500 μ M H 2 O 2 for 2 h (Fig. 2C,D). These results suggested that the 58-F compound protected hepatocytes from H 2 O 2 injury in a dose-and time-dependent manner.
Instead of thymidine, BrdU was incorporated into the nuclear DNA during the S-phase of the cell cycle to evaluate cell proliferation 14 . After exposure to H 2 O 2 , the number of BrdU-positive cells was significantly decreased (49.6% of the control). However, pretreatment with 50 μ M of 58-F significantly increased the number of BrdU-positive cells (74.3% of the control) (Fig. 2E,F).
Annexin V is a Ca 2+ -dependent phospholipid-binding protein with a high affinity for externalized phosphatidylserine, which is strictly confined to the inner leaflet of the plasma membrane facing the cytosol in viable cells. Localization of Annexin V from the inner to the outer plasma membrane is an early feature of apoptosis that occurs prior to the loss of membrane integrity 15 . Furthermore, living cells excluding PI allow for the specific detection and quantification of apoptosis.
As shown in Fig. 3A, in the untreated control, only 0.25% of the cells were early apoptotic cells, 3.01% of the cells were late apoptotic cells and 1.50% were dead cells. By contrast, in the H 2 O 2 -treated group, 3.26% of the cells were early apoptotic cells, 13.17% of the cells were late apoptotic cells and 3.36% were dead cells, suggesting an approximately four-fold increase in the number of late apoptotic or dead cells in the H 2 O 2 -treated group compared with the untreated controls.
Following 12 h of treatment with 50 μ M 58-F and 500 μ M H 2 O 2 , 1.61% of the cells were early apoptotic cells and 8.05% of the cells were late apoptotic cells. At the 36 h time point, 1.60% of the cells were early apoptotic cells and 7.57% of the cells were late apoptotic cells (Fig. 3A). These results demonstrated that 58-F protected hepatocytes from H 2 O 2 -induced apoptosis and death.
The pro-apoptotic protein Bax has a carboxy-terminal α -helix domain of the membrane-anchoring region, which is normally sequestered in an inhibitory hydrophobic groove of Bax, preventing its insertion into membranes. Upon exposure to various death stimuli, the conformation of Bax is altered, and its membrane-anchoring domain is exposed and inserted into membranes, triggering apoptosis 16 . Our results showed that after exposure to H 2 O 2 , Bax protein expression increased gradually, peaked at 8 h, and subsequently decreased at 24 h. The levels of Bax in the 58-F treated groups were less than in the H 2 O 2 group at the same time point (Fig. 3B,C).
Caspase-3 plays a central role in cell death pathways. Procaspase-3 exists in cells as inactive dimers. Upon activation, procaspase-3 separates into a large and a small subunit. The active caspase-3 upregulates the downstream enzymes of the caspase family and ultimately contributes to cell death 17 . In the present study, either procaspase-3 Scientific RepoRts | 6:27875 | DOI: 10.1038/srep27875 or cleaved active caspase-3 protein levels increased and peaked at 8 h and declined at 24 h after treatment with H 2 O 2 . Treatment with 58-F significantly suppressed the increase at the same time points (Fig. 3B,D,E,F). In summary, H 2 O 2 induced both apoptotic and necrotic cell death and, therefore, inhibited cell proliferation and viability. The compound 58-F protected hepatocytes from death and promoted their proliferation. Proteins are the major target for ROS because amino acids are susceptible to oxidation. Carbonylation is the most common oxidative alteration of proteins. The results shown in Fig. 4B,C demonstrate that the amount of PCC increased progressively from 0.5 h to 24 h, although it peaked at 8 h and remained at a high level at 24 h. Similar results were obtained in vivo (Fig. 1D,E). marker that is selectively retained in the partially acidic lysosome and reflects the lysosomal membrane permeability 18 , was used to evaluate the integrity of the lysosome membrane. The intensity of green fluorescence gradually decreased with the increasing duration of H 2 O 2 treatment and reached the lowest point at 8 h. It then increased slightly at 24 h. However, the fluorescent intensity increased with the 58-F pretreatment (Fig. 5A,B). LAMP1 is highly expressed in lysosomes. Only 1-2% of the total LAMP1 can be found at the plasma membrane; its primary localization seems to be at the endosome-lysosomal membrane 18 . Our results reveal that LAMP1 protein levels gradually decreased with the increasing duration of H 2 O 2 treatment, reaching its lowest level at 8 h, with a partial restoration of LAMP1 levels observed at 24 h. By contrast, in cells pretreated with 58-F, the LAMP1 protein levels significantly increased following a similar time course as the H 2 O 2 treatment (Fig. 5D,E). Again, changes in LAMP1 mRNA expression were consistent with the levels of protein expression (Fig. 5C). The knockdown of LAMP1 expression in cells strongly diminished the intensity of green fluorescence in the lysosome (Fig. 5I,J) but also inhibited the cell viability (Fig. 5H). These data indicate that a deficiency in LAMP1 directly inhibits cell viability. Furthermore, in LAMP1 knockdown cells, the compound 58-F enhanced the fluorescence intensity of the LysoTracker Green and increased cell viability (Fig. 5H-J).

58-F Antagonizes H 2 O 2 -induced CatB/D and Proton Leakage.
To further investigate whether catalytic enzymes in the lysosome leaked into the cytosol and caused the inhibition of cell viability, we assessed both the content and the activity of cathepsin B (CatB) and D (CatD) in the cytosol and in whole cells. Surprisingly, the content of both CatB and CatD in the cytosol were extremely high after exposure to H 2 O 2 for 8 h, reaching increases of ∼ 24-fold for CatB and ∼ 31-fold for CatD ( Fig. 6A,C,D). By contrast, the content detected in the whole cells remained unchanged (Fig. 6B). The compound 58-F suppressed cytoplasmic CatB and CatD content at all of the designated time points in a time-dependent manner ( Fig. 6A,C,D).
The CatB activity in the cytosol gradually increased with the increased duration of H 2 O 2 treatment, peaking at 8 h, and then declining at 24 h (Fig. 6F). The compound 58-F suppressed cytoplasmic CatB activity at all of the designated time points (Fig. 6F).
To further analyse whether the enzymatic activity of CatB in the cytosol induced cell death, the Cat B inhibitor CA-074-Me 19 was used. As expected and consistent with the results of compound 58-F, CA-074-Me blocked the H 2 O 2-induced cell death in a time-dependent manner (Fig. 6E). AO is a cell-permeable fluorescent dye with a concentration-dependent fluorescence emission that ranges from red (when the concentration is high in the lysozyme) to green (when the concentration is low in the cytosol). AO is positively charged at a low pH, which consequently hinders the ability of the AO molecules to cross the vesicular membrane and escape into the surrounding cytoplasm. In the case of LMP, AO is released from the lysosomes into the cytosol where it emits an enhanced green fluorescence that can be monitored by fluorescence microscopy 20 . After delivery of H 2 O 2 into the cells, a significant reduction of red fluorescence and a concomitant increase in green fluorescence was observed in a time-dependent manner, suggesting that H 2 O 2 induced LMP at the cellular level (Fig. 6G,H). By contrast, the compound 58-F was associated with an increase of the red fluorescence and a concomitant suppression of the green fluorescence at all of the designated time points compared with those in the H 2 O 2 -treated cells (Fig. 6G,H).

Lysosomal acidification inhibition by H 2 O 2 Contributes to Cell Death. The vacuolar H + -ATPases
(V-ATPases) within the lysosomal membranes are responsible for establishing and maintaining a low pH. Bafilomycin A1 (BafA1) specifically inhibits the V-ATPase by binding to its proteolipid subunit and blocking the passage of protons into the lysosome, and therefore hinders lysosomal acidification 21 (Fig 7A,B). The results show that cell viability was inhibited after H 2 O 2 treatment and progressively suppressed by BafA1 at all of the designated time points (Fig. 7C). In addition, NH 4 Cl was in equilibrium with the weak base NH 3 , which is membrane permeable and enters acidic compartments to bind with protons, and therefore, it was added to artificially acidify the cytosol and alkalify the lysosome 22 (Fig. 7A,B). Similar cell viability results were found in cells treated with NH 4 Cl (Fig. 7D). Furthermore, we observed that 2.32% of cells were early apoptotic,14.32% of cells were late apoptotic and 4.50% were dead following NH 4 Cl treatment compared with control groups, in which 0.25% were late apoptotic, 3.01% were late apoptotic and 1.50% were dead cells (Fig. 7E). Together, the LMP caused by H 2 O 2 leads to lysosomal acidification inhibition and contributes to cell death.

N-acetylcysteine (NAC) is Protective against Cellular Toxicity by Inhibiting LMP. ROS produc-
tion in cells was detected using 2′ ,7′ -dichlorofluorescein diacetate (DCFH-DA), as described in the Materials and Methods, and was decreased in the presence of NAC pretreatment (Fig. 8B). In addition, LAMP1 protein levels in the cells that received NAC pretreatment were significantly increased compared with the cells that received the H 2 O 2 treatment. The content of CatB in the cytosol following NAC treatment also showed a significant decrease compared with that in the H 2 O 2 treatment group. Again, the LysoTracker Green stain was used to evaluate the integrity of the lysosome membrane, and cells treated with NAC showed an increase in the intensity of fluorescence.
As shown in Fig. 8A, after treatment with 100 μ g/ml NAC after 500 μ M of H 2 O 2 , 3.26% of the cells were early apoptotic and 13.17% of the cells were late apoptotic after 12 h, whereas in the H 2 O 2 treatment group, 0.58% of the cells were early apoptotic and 5.22% of the cells were late apoptotic (Fig. 8A). Together, these results showed that NAC protected hepatocytes from H 2 O 2 -induced apoptosis and death by inhibiting LMP.

Hepatocyte death mediated by CCl 4 -induced lysosome injury. CCl 4 is a potent hepatotoxin that
is widely used for the induction of chemical liver damage. The toxicity of CCl 4 is attributed to accumulation of ROS during its metabolism, which is involved in the pathological progress of liver diseases. Therefore, oxidative stress is one of the main mechanisms of CCl 4 -induced liver injury 23 . Our in vivo studies suggest that the levels of lipid peroxidation, protein oxidation, and serum ALT activity in hepatocytes following cell injury or death are increased after CCl 4 treatment. A number of studies have suggested that oxidized proteins that are not cleared in time lead to the loss of cell viability and persistent cell death 24 . The livers of CCl 4 -induced mice also showed large areas of dead hepatocytes. The molecular structure of LAMP1 is composed of mainly carbohydrates that form a nearly continuous coating on the inner surface of the lysosome membrane and serve as a barrier to soluble hydrolases, preventing the release of lysosomal enzymes and H + into the cytoplasm 25 . The loss of LAMP1 leads to damage to the integrity of the lysosome membrane and ultimately to the leakage of the lysosome contents. The mRNA and protein levels of LAMP1 detected in the liver were lower after CCl 4 treatment than under normal conditions, which suggested that lysosomes mediated the CCl 4 -induced hepatocyte injury/death that induced injury to the entire liver.  Following cathepsin leakage and lysosomal acidification inhibition, LMP triggers lysosomal cell death in H 2 O 2 -induced hepatocyte injury. H 2 O 2 -induced cytotoxicity is a common method used for measuring potential antioxidants. In this study, treatments with different concentrations (0-2500 μ M) of H 2 O 2 for 2 h induced a dose-dependent decrease in cell viability. The rate of inhibition achieved by 500 μ M H 2 O 2 was approximately 40%; thus, this concentration was used for the subsequent experiments. The ROS products increased gradually, peaking at 4 h, in tandem with the protein carbonyl content which peaked at 8 h. Following the oxidative damage, cell viability was inhibited, and cell proliferation was diminished, whereas apoptotic and necrotic cell death increased.
In cultured cells, LMP occurred following H 2 O 2 injury 26,27 . The LMP followed by the leakage of enzymes, including CatB, triggers lysosome-induced cell death 28 , and hepatocyte death is accompanied by an increase in LMP in acute liver injury 29 . According to the cell type and the dosages of the inducing factors, lysosomal apoptosis or necrosis occurs 30 . In our study, both LysoTracker and AO staining showed that the LMP increased gradually with time and was restored after peaking at 8 h.
LAMP1 is a highly glycosylated protein located in the lysosomal membrane that controls the integrity of the lysosomal membrane 25 . Our studies found that the mRNA and protein levels of LAMP1 decreased after hepatocyte injury in vitro and in vivo. Furthermore, after knockdown of LAMP1 in hepatocytes, LMP occurred and cell viability was inhibited, which demonstrated that LMP mediated CCl 4 /H 2 O 2 -induced hepatocyte injury/death.
Under physiological conditions, proteolytic, and hydrolytic enzymes in the lysosomes were restricted to the intact lysosomal membranes. Upon LMP, H + was released into the cytoplasm resulting in acidification and cellular apoptosis 31,32 . NH 4 Cl and Bafilomycin A1 were used to evaluate the effects of lysosomal acidification inhibition on cell viability and death, respectively. The results indicated that Bafilomycin A1 further inhibited cell viability after cells were exposed to H 2 O 2 at all of the designated time points. Furthermore, the results of the NH 4 Cl treatment were consistent with that of Bafilomycin A1 and resulted in apoptosis.
In LMP, hydrolytic enzymes in the lysosomes are released into the cytoplasm 33,34 . Cathepsins, including cathepsin B/D/L, under near-neutral pH conditions also exhibited catalytic activity for a few hours 35 . CatB leakage from the lysosomes is also associated with apoptosis 36,37 . Our studies found that levels of CatB and D in the cytoplasm highly increased after exposure to H 2 O 2 for 8 h, whereas the overall cellular levels remained unchanged. Furthermore, the CatB activity in the cytoplasm increased over time following the same time course as the H 2 O 2 treatment. Moreover, the CatB-specific inhibitor, CA-074-Me increased the cell viability in the presence of H 2 O 2 . Therefore, our results confirmed that upon LMP, protons and lysosomal enzymes, including CatB/D, were translocated to the cytoplasm and triggered cell death.
Effect of 58-F on hepatocyte injury/death. The compound 58-F is a flavanone that has been newly extracted from Ophiopogon japonicus, with no published reports of its pharmacological activity with regard to hepatic protection. Here, we evaluated its effect on hepatocyte protection. The in vivo results revealed that 58-F decreased serum ALT activity, scavenged products of lipid peroxidation and oxidized proteins, and restored LAMP1 levels, suggesting that 58-F is a potentially new protective agent for hepatocytes. H 2 O 2 -induced hepatocyte damage was used to evaluate the mechanisms of 58-F on hepatocyte protection in vitro. 58-F reduced the apoptotic and dead cell numbers and prevented the inhibition of cell proliferation induced by H 2 O 2 by clearing the oxidized products of lipids and proteins, also having protective effects on hepatocytes in vitro. A large amount of H 2 O 2 diffused into the lysosome resulting in LMP 38 . In wild-type astrocytes, H 2 O 2 -induced expression of LAMP1 decreased and lysosomal LMP was decreased 39 . (-)-Epigallocatechin-3-gallate (EGCG) promotes the production of intracellular ROS upstream of LMP and cell death, and the antioxidant NAC protects against EGCG-mediated LMP by decreasing the level of ROS 40 . In addition, in Bxpc3 cells, LAMP1-shRNA significantly reduced the viability of cells and LysoTracker stain. Substantial amounts of ROS were detected with H 2 O 2 treatment, and these levels were decreased by NAC 41 . In this study, we also found that H 2 O 2 induced LMP by decreasing LAMP1 expression and NAC. Similarly, 58-F stabilized the integrity of the lysosomal membrane partly by upregulating the expression of LAMP1, following the diminishing proton and cathepsin leakage into cytoplasm.
In brief, the compound 58-F protected the integrity of the lysosomal membrane by upregulating the expression of LAMP1, which reduced cathepsin B/D and H + leakage into the cytosol, and protected hepatocytes from death. Animal experiments. In total, 24 male BALB/c mice (license number: SCXK (Shanghai) 2012-0002) were housed at the animal centre of the Shanghai University of Traditional Chinese Medicine and fed a standard pelleted diet and water. All of the experimental protocols were approved by the Shanghai University of Traditional Chinese Medicine's Animal Ethics Committee, and all of the methods in the study were performed in accordance with the approved guidelines.

Methods
The mice were initially randomized into three groups: a control group (n = 6), a CCl 4 -treated group (n = 9) and a 58-F/CCl 4 -treated group (n = 9). Before CCl 4 injection, the mice in the control and CCl 4 groups were injected intragastrically with 0.9% NaCl. The mice in the 58-F/CCl 4 -treated group were intragastrically administered 58-F at a dosage of 15 mg/kg body weight for 3 days. The mice in the CCl 4 -treated and 58-F/CCl 4 -treated group were given intraperitoneal injections of 5% CCl 4 (10 μ l/g weight), and the mice in the control group were intraperitoneally injected with olive oil (10 μ l/g weight). All of the mice were sacrificed on day 7 of the experiment. H&E staining. The 10% neutral formaldehyde fixed liver tissues were processed in an automated tissue processor that has been optimized for liver tissues, embedded in paraffin blocks and cut into 4-μ m thick sections. The sections were stained with H&E, followed by visualization of the histopathological changes under a light microscope.
Cell Culture and Short-hairpin RNA (shRNA) Transduction. The BNL CL.2 mouse embryonic hepatocyte cell line was purchased from the cell bank of the Shanghai Institutes for Biological Sciences and was routinely grown in DMEM supplemented with 10% FBS, 100 U/ml penicillin and 100 μ g/ml streptomycin.
The BNL CL.2 mouse embryonic hepatocyte cells were incubated with the LAMP1-specific shRNA (target sequences: 5′ -GCCACTGTGGGAAACTCATAC-3′ ) and lentiviral particles (pHBLV-U6-ZsGreen-Puro, Hanbio Biotechnology Co., Ltd., No 1 Building, 150 Cailun Road, Pudong New Area, Shanghai, China) at a ratio of 2 particles to 1 cell. After 48 h, the medium was replaced with 1.5 μ g/ml puromycin to improve the transduction efficiency. Successfully transduced cells included a puromycin resistance tag, and these cells were selected using 1.5 μ g/ml puromycin. LAMP1 knockdown was confirmed using quantitative RT-PCR and Western blot analysis.
All of the cells were cultured in a humidified atmosphere containing 5% CO 2 at 37 °C, unless otherwise indicated.
Cell viability assay. Cell viability was assessed by a Cell Counting Kit-8 (CCK8) assay (Dojindo, Kumamoto, Japan): 4000 cells were seeded in 96-well plates and were incubated for 24 h at 37 °C in 5% CO 2 . After incubation with different concentrations of 58-F, 100 μ M ammonium chloride (NH 4 Cl) or 50 μ M CA-074-me and 10 μ l CCK8 reagent was added to each well and incubated for 2 h at 37 °C. The absorbance at 450 nm was measured using a microplate reader (Bio-Tek, Hercules, CA, USA). The mean absorbance values from 4 wells following each treatment were used to represent the index of cell viability. The experiments were repeated at least in triplicate. The cell viability was expressed as a percentage of the control level.
Detection of ROS by DCFH-DA. Intracellular ROS levels were assessed using 2′ ,7′ -dichlorofluorescein diacetate (DCFH-DA). DCFH-DA is a stable, fluorogenic and non-polar compound that readily diffuses into the cells. It is deacetylated by intracellular esterases to the non-fluorescent 2′ ,7′ -dichlorodihydrofluorescein (DCFH), which is later oxidized by intracellular ROS into the highly fluorescent 2′ ,7′ -dichlorofluorescein (DCF) 42   Annexin V-FITC/PI staining with fluorescence-activated cell sorter (FACS). Cells treated with 50 μ M 58-F for 12 h or 36 h following further 2 h treatment with 500 μ M H 2 O 2 , or with 100 μ M NH 4 Cl for 2 h were harvested through trypsinization, were washed twice with cold PBS three times and were resuspended in 100 μ l of binding buffer at a density of 1.2 × l0 6 cells per ml. Then, 5 μ l of Annexin V-FITC (BD Pharmingen, Santiago, CA, USA) and 5 μ l of propidium iodide (PI) (BD Pharmingen, Santiago, CA, USA) were added to each sample in the dark for 15 min. The fluorescence was detected immediately after staining using a flow cytometer (Ex = 488 nm; Em = 530 nm). The results were presented as the percentage of cells that were viable (Ann-V − /PI − ), early apoptotic (Ann-V + /PI − ), late apoptotic (Ann-V + /PI + ) or nonviable (Ann-V − /PI + ). Cells treated with 1 μ M staurosporine for 24 h was used as the positive control for apoptosis.
Western blotting. At the end of the designated treatments, tissue homogenates and cell lysates were prepared in a buffer containing 62.5 mM Tris-HCl at pH 6.8, 20% glycerol, 2% SDS, 2 mM DTT, 100 mM PMSF and proteinase inhibitor cocktail. Protein concentrations were determined using the BCA total protein assay (Biomiga). Samples with equal amounts of protein were subjected to a 12% SDS-PAGE and transferred to a polyvinylidene difluoride (PVDF) membrane (Bedford, MA, USA). After blocking with 5% non-fat milk, the membrane was probed with the designated primary and secondary antibodies. The membranes were developed using enhanced chemiluminescence (Thermo Scientific, Waltham, MA, USA) and visualized using a FCM gel imaging device (Protein Simple, San Francisco, CA, USA).
Protein carbonyl content. The protein carbonyl content (PCC) was assayed using an Oxyblot kit (Millipore, Bedford, MA, USA). Briefly, 5 μ l of the tissue homogenates or cell lysates including 20 μ g of protein mixed with 5 μ l 12% SDS, were supplemented with 10 μ l of 2,4-dinitrophenylhydrazine in 2N HCl to derivatize the carbonyl groups from the protein side chains for 15 min at room temperature. The derivatized samples were then separated using 12% SDS-PAGE. Western blotting was performed as described above, using the 2,4-dinitrophenylhydrazine antibody provided (1:50). Analyses of the results were reported as the ratio of the protein-of-interest/β -actin. Given the presence of multiple bands, the average value of all of the bands within each lane was used to provide an overall measure of protein carbonyl content.
LysoTracker Green or Acridine Orange (AO) Staining. Ten thousand cells were plated in a 60 mm confocal plate. After the designated treatments, the cells were incubated with 75 nM LysoTracker Green (Invitrogen, Waltham, MA, USA) or 5 mg/ml AO (Sigma, St Louis, MO, USA) for 30 min at 37 °C. The fluorescence was examined using a confocal microscope with the excitation wavelength set at 488 nm. Two separate emission bands (505-570 nm and 615-754 nm) for AO staining and one for LysoTracker staining (504-511 nm) were used simultaneously, obtained and photographed.
Cytosolic/particulate separation. The method for cytosolic/particulate separation was based on the protocol described in the Cytosol/Particulate Rapid Separation Kit (BioVision, Milpitas, CA, USA). Briefly, at the end of the designated treatments, both the medium and the adherent cells were harvested into 15 ml tubes, and the samples were centrifuged at 600 × g for 5 min. The pellets were resuspended in 40 μ l of cell suspension buffer and mixed with 40 μ l of cytosol releasing buffer, including 1 mM PMSF. The cell suspension was carefully layered upon 500 μ l of an oil layer with 40 μ l of a particulate layer at the bottom on ice for 30 s. After centrifugation at 1400 × g for 1 min, the cytosol was separated from the membrane particles by the oil layer and were transferred to new tubes. A BCA total protein assay kit (Biomiga) was used to determine the protein concentration.