Simulated ischemia/reperfusion-induced p65-Beclin 1-dependent autophagic cell death in human umbilical vein endothelial cells

Myocardial ischemia/reperfusion (I/R) injury detrimentally alters the prognosis of patients undergoing revascularization after acute myocardial infarction. Our previous study demonstrated that NF-κB-induced autophagy plays a detrimental role in cardiac I/R injury using a rabbit myocardial I/R model. In this study, we sought to explore the specific mechanism of this autophagy-mediated cell damage in an in vitro simulated ischemia/reperfusion (sI/R) model using human umbilical vein endothelial cells. Our current study demonstrates that simulated I/R induces autophagy in a p65-Beclin 1-dependent manner, which can be suppressed with the inhibition of NF-κB. Furthermore, rapamycin which promotes autophagy, exacerbates sI/R-induced cell death. While 3-methyladenine rescues cell damage. Our data thus suggest that I/R promotes NF-κB p65 activity mediated Beclin 1-mediated autophagic flux, thereby exacerbating myocardial injury.

is to illuminate the associated pathway by which sI/R induced autophagy exerts effects on HUVECs. Knowledge of the role of key mediators in this pathway will help to indentify therapeutic targets for patient suffering from myocardial ischemia/reperfusion injury.

Results
sI/R promotes autophagic protein expression and autophagic flux in HUVECs. sI/R resulted in a significant increase in Beclin1 protein levels, which occurred within 4 h of reperfusion and persisted for at least 24 h (Fig. 1A). Protein expression of Beclin1 in sI/R at 4 h, 18 h and 24 h were 4.36, 4.41, and 4.28 fold, respectively, compared to the control (Fig. 1B). When autophagy is activated, cytosolic LC3-I is cleaved to proteolytic derived LC3-II. Thus, we used detection of LC3-II to estimate the abundance of autophagosomes. After "ischemia" for 30 min, a reperfusion time-dependent activation of autophagy was observed in HUVECs as measured by the conversion of LC3-I to LC3-II (Fig. 1A). sI/R-induced increases in LC3-II/LC3-I ratio at 4 h, 18 h and 24 h were 2.61, 3.87, and 7.20 fold, respectively, compared to the control (Fig. 1B). However, we had concerns that an increase in LC3-II/LC3-I ratio might not represent induction of autophagy and might instead be due to I/R repressing autophagosome fusion with lysosomes and degradation of LC3-II. Therefore, in order to determine the flux of autophagy, which reflects true autophagic activity, we carried out an LC3 turnover assay by examining the effect of disrupting lysosomal function using lysosomotropic reagent chloroquine (Fig. 1C). Chloroquine increases lysosomal pH and blocks the degradation of LC3-II, resulting in the accumulation of LC3-II. When treated with chloroquine, I/R-stimulated conversion of LC3-I to LC3-II in HUVECs increased by 28.95% more than I/R alone (Fig. 1D) (p < 0.05). sI/R-induced autophagy is NF-κB p65-dependent. To determine whether p65 was involved in the sI/R activation of autophagy, we pretreated HUVECs with pyrrolidine dithiocarbamate (PDTC) 7,12 , the specific NF-κ B inhibitor, for 1 h prior to sI/R exposure. The sI/R-induced conversion of LC3-I to LC3-II was reduced by PDTC in a dose-dependent manner. 0.25 mM and 0.5 mM of PDTC decreased sI/R-induced LC3-II/LC3-I ratio by 42.10% and 58.42% ( Fig. 2A,B). To circumvent potential side effects of the inhibitor, p65 siRNA was transfected into HUVECs to substantiate the role of NF-κ B. Consistently, knocking down of p65 dramatically sI/R-induced development of autophagic vacuoles is NF-κB p65-dependent. We examined the effect of I/R injury on the formation of autophagic vacuoles in HUVECs as measured by Monodansylcadaverine (MDC) or acridine orange staining. MDC is a fluorescent dye used to detect autophagic vacuoles, the accumulation of mature autophagic vacuoles such as autophagosomes rather than the early endosomal compartment 13 . With MDC staining, bright green dots representing autophagosomes in the cytoplasm of HUVECs increased in our stimulated sI/R model compared to the control. I/R-induced autophagosomes were decreased 47.17% and 56.60% by knockdown of p65 and PDTC(0.5 mM) treatment, respectively (Fig. 3A, upper panel and Fig. 3B). As an alternative method, using acridine orange stain, the protonated form of acridine orange accumulated in acidic compartments was characterized by red fluorescence. In the control group, cells displayed primarily green fluorescence with minimal red fluorescence indicating the absence of autophagic vacuoles, while exposure of HUVECs to sI/R produced a 4.87-fold increase in red fluorescence. This red fluorescence decreased by 47.58% with transfection of p65 siRNA and by 55.13% through PDTC treatment (Fig. 3A, lower panel and Fig. 3B). However, we were concerned about that neither MDC nor acridine orange staining is specific for autophagosomes; vesicles with positive staining may be indiscernible from lysosomes. Therefore, we additionally used electron microscopy (EM) to visualize autophagic vacuoles, which has been considered a reliable method for identifying autophagosomes (Fig. 3C,a-e). EM revealed a consistent increase in the number of autophagic vacuoles containing degraded cytoplasmic materials after sI/R (Fig. 3C,b). Silencing of p65 by siRNA diminished the phenomenon observed above in sI/R (Fig. 3C,d).
sI/R-induced expression of Beclin1 is mediated by NF-κB p65. Since activation of p65 correlated with increased Beclin1 expression in the context of I/R injury both in vivo 7 and in vitro 11 , we sought to determine whether p65 is upstream of Beclin1 in the cell signal transduction pathway of sI/R-induced autophagy. Pretreatment of HUVECs with 0.25 mM and 0.5 mM of PDTC in sI/R resulted in 71.84% and 78.41% reduction of Beclin1 mRNA respectively compared to sI/R treatment alone as detected by real time PCR (Fig. 4A). This effect was dose-dependent, with a maximal effect seen at 0.5 mM. We next investigated the effect of PDTC on protein levels of Beclin1 by Western blot. Treatment with 0.25 mM and 0.5 mM of PDTC decreased I/R-induced Figure 2. sI/R-induced LC3-I to LC3-II conversion was p65 dependent. (A) HUVECs were pretreated with 0.25 mM or 0.5 mM PDTC for 1 h before exposed to normal media (control) or I/R for 4 h. Cell lysates were subjected to immunoblotting analysis for detection of LC3B. (B) Graph showing quantification of LC3-I to LC3-II conversion from the immunoreactive bands obtained as (A). *P < 0.05 vs control. (C) HUVECs were transfected with control siRNA or p65 siRNA oligos (each at 20 nM) before exposure to normal media (control) or I/R conditions for 4 h. Immunoblotting with LC3 antibody was then performed. (D) Quantification of LC3-I to LC3-II conversion was measured from the immunoreactive bands obtained as (C), normalized to those of β-actin and the data were presented as a mean ± SD from three independent experiments. *P < 0.05 vs control. protein levels of Beclin1 by 59.62% and 84.86% respectively (Fig. 4B,C). To confirm the effect of p65 on Beclin1 in sI/R, we specifically knocked down endogenous p65 in HUVECs via RNA interference. Knockdown of p65 led to a 75.11% reduction in I/R-induced Beclin1 mRNA levels (Fig. 4D). Furthermore, in HUVECs with p65 siRNA transfection in sI/R, we measured a 79.06% reduction of p65 protein levels followed by a 43.43% reduction of I/R-induced Beclin1 protein levels (Fig. 4E,F).

Rapamycin threatens cell survival while 3-MA promotes cell survival in the context of sI/R.
WST-1 cell viability assays revealed that sI/R induced cell death in a time-dependant manner by 29.55%, 64.78% and 77.54% at 24 h, 48 h and 72 h (Fig. 5A). To verify whether p65 plays a role in sI/R-related reduction in cell survival, we assessed the effects of PDTC and p65 siRNA under sI/R conditions. PDTC (0.5 mM) treatment significantly decreased sI/R-induced cell death by 72.28%, 70.27% and 53.07% at 24 h, 48 h and 72 h. Correspondingly, p65 siRNA (20 nM) treatment significantly reduced sI/R-induced cell death by 57.97%, 54.31% and 49.05% at 24 h, 48 h and 72 h. In addition, because 3-MA is a well-known inhibitor of autophagy and rapamycin is a well-known stimulator of autophagy, we confirmed the effects of 3-MA and rapamycin on the formation of autophagic vacuoles in our sI/R system with acridine orange staining under confocal laser microscopy. Exposure of HUVECs to 3-MA in sI/R exhibited rather faint perinuclear red fluorescence compared to sI/R alone (Fig. 5B). On the contrary, HUVECs pretreated with rapamycin glowed with bright red perinuclear fluorescence. These red dots were distinct and countable under magnification. Furthermore, HUVECs pretreated with 3-MA ahead of sI/R displayed an absence of autophagic vacuoles under EM, while we observed quite the opposite phenomenon in cells pretreated with rapamycin (Fig. 5C). Cell viability assays revealed that 3-MA treatment significantly decreased sI/R-induced cell death by 29.30% at 72 h. Rapamycin treatment significantly promoted sI/R-induced cell death by 15.88% at 72 h with no statistical difference detected at 24 h and 48 h (Fig. 5D).

Discussion
Complications such as malignant arrhythmia, heart failure and other cardiovascular events resulting from ischemia-reperfusion injury are not rare in patients undergoing revascularization after acute myocardial infarction. Elucidating the mechanisms behind myocardial ischemia-reperfusion injury is critical in achieving a better clinical prognosis for these patients. Besides the roles of apoptosis, endoplasmic reticulum stress, reactive oxygen species and so on, autophagy has been implicated as contributing to the ischemia-reperfusion injury process. It is also known that NF-κ B plays a crucial role in ischemia-reperfusion injury. However, the downstream pathway is not clearly understood. In the present study, we provide evidence of a novel pathway for the induction of autophagy in response to sI/R. sI/R induced NF-κ B p65-mediated expression of Beclin1 and autophagy (Fig. 6).
During the formation of autophagosomes, Beclin 1, a mammalian ortholog of yeast Atg6, plays a very important role in the vesicle nucleation step of autophagy, and therefore serve as a marker for autophagy. Elevated levels of Beclin 1 protein were associated with an increase in autophagy phenomenon 14,15 . It is reported that Beclin 1-dependent autophagy mediates autophagic cell death during I/R 16,17 . Reduction of I/R-induced autophagy by urocortin or RNA interference-reduced Beclin 1 expression in cardiac myocytes enhanced cell survival 18 . An induction of autophagy and cardiac injury during the reperfusion phase was significantly attenuated in Beclin + /− mice 19 . However, an increase in Beclin 1 protein levels alone does not absolutely guarantee an increase in autophagy. Binding of Beclin 1 with Bcl-2 family proteins was reported to negatively regulate autophagy activity, in response to some external stimuli 15 . For example, Beclin 1-mediated autophagy implicated to play a protective role in preconditioning. Autophagy elicited by ischemic preconditioning inhibited necrosis in injured livers. Enhancement of Bcl-2 phosphorylation was associated with decreased interactions between Bcl-2 and Beclin 1 and an increased expression of microtubule-associated protein 1 light chain 3 (LC3)-II, suggesting important roles of Bcl-2 and Beclin 1 in the regulation of autophagy 20,21 . Since LC3-II is essential for autophagosome formation, the conversion ratio of LC3-II/LC3-I is widely used to monitor autophagic activity. Our present study demonstrates that Beclin 1 protein levels were increased after 4 h in sI/R and remained steadily at high levels at 24 h afterwards in sI/R. LC3-I to LC3-II conversion continued rising in a time-dependent manner in sI/R. Based on these results, we propose that Beclin 1 expression responds quickly to sI/R and is the main, but probably not the only, mediator to trigger activation of autophagy in sI/R. Moreover, in some cases, increased expression of Beclin-1 and LC3-II may not represent enhanced autophagic activity, but rather as a result of reduced degradation of autophagic vacuoles which results in either feedback stimulation of Beclin-1 or accumulation of existing LC3-II 22 . Ma, X et al. reported that impaired autophagosome clearance because of lysosomal activity inhibition contributed to cardiomyocyte death in I/R injury 23 . Therefore, it is important to distinguish autophagic flux from steady-state level of LC3 monitoring. The latter may be assessed by following LC3 in the absence and presence of autophagy inhibitors, and by examining the autophagy-dependent degradation of appropriate substrates 24 . In particular, with the evidence of an increase in LC3-II and autophagosomes in the present study, it is essential to determine whether this represents increased flux, or a block in fusion or degradation using inhibitors such as chloroquine. Chloroquine disrupts lysosomal function by increasing lysosomal pH and blocks the degradation of LC3-II, resulting in the accumulation of LC3-II. Treatment with chloroquine hence helps to determine autophagic activity, which represents the amount of LC3 delivered to lysosomes for degradation (i.e., autophagic flux). In our present study, inhibition of lysosomal activity with chloroquine further enhanced the level of LC3-II in HUVECs subjected to sI/R. Therefore, we have confirmed that sI/R not only promotes autophagic protein expression but also autophagic flux in HUVECs. Correspondingly, I/R promotes autophagic activity and causes autophagic cell death in H9c2 cells 25 . In support of our findings, based on their in vivo research, Yu, P et al. reported the protection effect of sevoflurane postconditioning in myocardial I/R through inhibition of over-activation of autophagy by suppressed elevated expressions of LC3 II/I ratio, Beclin1, Atg5 and Atg7 in I/R rat 26 .
Transcription factor NF-κ B is an important mediator to stimulate downstream genes in response to environment stimulation outside cells. NF-κ B has been implicated in restricting autophagy in a positive 27 or negative way 28,29 in different scenarios. We hypothesize that sI/R-induced autophagy might involve the stimulation of gene spectrum such as NF-κ B. Pretreatment with PDTC and p65 siRNA transfection to suppress NF-κ B both lead to the inhibition of LC3-II expression or the disruption of autophagosome development. Our previous study indicated that NF-κ B-induced autophagy played a pivotal role in rabbit myocardial I/R injury. Our current in vitro study confirms that sI/R promotion of autophagosome formation as well as autophagic activity is mediated by NF-κ B. In support of our findings, inhibition of NF-κ B associated autophagy showed a protective effect against I/R injury in the brains of rats 30 . In addition, Shen M et al. reported ethyl pyruvate attenuates hepatic I/R injury by inhibiting the intrinsic pathway of autophagy mediated partly through downregulation of the NF-κ B axis 31 .
p65/RelA has consensus sites in the promoter with the Beclin 1 gene and p65/RelA positively modulates canonical autophagy in various human cell lines 32 . Evidence that beclin-1 is NF-κ B regulated has been previously published 33 . Furthermore, it is reported that the expression of Beclin1 is dependent upon NF-κ B 34 and inhibition of NF-κ B suppresses autophagy. However, a decrease in phosphorylation of Iκ Bα and NF-κ B/p65 and NF-κ B/p65 nuclear translocation was correlated with upregulation of Beclin 1 in SZC014-induced autophagy 35 . The research of Rui, L.X. et al. suggested that SZC014, a synthetic oleanolic acid derivative, induced autophagy in gastric cancer cells via NF-κ B pathway. Many mechanisms have been implicated that involve Beclin 1 participating in autophagy during I/R injury. It is tempting to speculate whether NF-κ B is involved in the Beclin 1 associated I/R-induced autophagy. Manipulation of wild type or transgenic Iκ Bα dominant-negative mice unveiled that Beclin-1 and LC-3 are elevated in a NF-κ B-dependent manner in the setting of myocardial I/R 36 . For the first time, our study suggested that Beclin 1 is downstream of NF-κ B in the context of sI/R. The observed increase in autophagy is NF-κ B-dependent, and likely includes direct regulation of beclin-1 mRNA and protein levels.
Whether autophagy is beneficial or detrimental to cells remains controversial 37 . For example, down-regulation of miRNA-30a alleviated cerebral ischemic injury through enhancing Beclin 1-mediated autophagy 38 . Facilitating the formation of autophagolysosomes protects the heart against second hand smoke exposure-induced myopathic changes 39 . Evidence supports that autophagy benefits heart cells during myocardial ischemia through improving myocardial energy metabolism and organelle recycling 40,41 . In agreement with these findings, delayed anesthetic preconditioning protected rat hearts from I/R injury via activation of NF-κ B and upregulation of autophagy 42 . In addition, NF-κ B-dependent increase in autophagy is indispensable in cardioprotection as ischemic preconditioning against I/R injury by acute high-fat feeding 36 . However, autophagy actually is one of the major types of morphologically distinct cell death, called typy II cell death 43 . Excessive autophagy produces lethal damage on cells 44 , confirmed by data showing that inhibition of excessive autophagy in cardiomyocytes reduces cardiac ischemia/reperfusion injury 5 . Specifically, in the context of sI/R, 3-MA rescues HUVECs by inhibiting autophagy while rapamycin promotes cell death by enhancing autophagy. Indeed, autophagy has great influence over a cell's decision to live or die. Our data support the hypothesis that sI/R threatens cell survival by stimulating NF-κ B dependant autophagy.
In brief, the present study reveals that Beclin 1-associated autophagy promotes cell death in sI/R. Activation of NF-κ B is a plausible mechanism responsible for this effect. These findings suggest the possibility that inhibition of excessive autophagic response could prevent I/R-induced myocardial injury.

Study limitations.
It is worth to point out that the limitation exists in this study. All our experiments were conducted in HUVECs. For the deference between venous endothelial cells and arterial endothelial cells, our results from HUVECs might not completely translate to arterial endothelial cells.

HUVECs simulated ischemia/reperfusion (sI/R) protocol. sI/R was initiated by incubation of
HUVECs for 30 min in an "ischemic buffer" containing 118 mM NaCl, 24 mM NaHCO 3 , 1.0 mM NaH 2 PO 4 , 2.5 mM CaCl 2 , 1.2 mM MgCl 2 , 20 mM sodium lactate, 16 mM KCl, and 10 mM 2-deoxyglucose (pH adjusted to 6.2), followed by "reperfusion" for 4 h. "Reperfusion" was accomplished by replacing the ischemic buffer with normal medium and culturing cells under normoxic conditions. We incorporated 2-deoxyglucose in our protocol to prevent cells from using their large glycogen reserves and thus truly simulate the substrate deprivation of ischemia.
In vitro siRNA transfection. Pre-designed and validated siRNA specific for NF-κ B p65, and non-targeting siRNA for control were purchased from Santa Cruz (sc-29410). HUVECs were plated to reach 30-50% confluence on the day of transfection. 20 nM siRNA duplexes were introduced with Lipofectamine RNAiMAX (Invitrogen).
Western blotting. HUVECs after treatment were lysed in chilled lysis buffer with protease inhibitor cocktail (Roche). BCA assay (Pierce) with bovine serum albumin as a standard was utilized to determine protein concentrations, and clarified lysates were boiled in SDS sample buffer. Samples were separated by 10% or 15% SDS-PAGE and transferred onto a polyvinylidene difluoride (PDVF) membrane (Millipore). Membranes were blocked in 5% non-fat milk in Tris-buffered saline (TBS) containing 0.1% Tween 20 (TBS-T) for 1 h at room temperature, incubated overnight with relevant antibodies at 4 °C, washed, probed with species-specific secondary antibodies coupled to horseradish peroxidase, and visualized using ECL chemiluminescence.
Immunofluorescent confocal laser microscopy. Autophagy is characterized by the development of autophagic vacuoles. Monodansylcadaverine (MDC) has been proposed as a tracer for autophagic vacuoles as described previously 11 . HUVECs were cultured on poly-L-lysine-coated cover slips overnight, then treated with different stimuli as described above, and rinsed with PBS. They were then stained with 50 μ M MDC at 37 °C for 1 h. After incubation, the cells were fixed for 15 min with ice-cold 4% paraformaldehyde at 4 °C, washed twice with PBS. Images were viewed and captured blindly by two observers using Leica TCS SP5 confocal laser microscopy. The formation of acidic vesicular organelles is related to cell autophagy 45 . Acridine orange-stained cells fluoresce diffuse green fluorescence, whereas the acidic compartments including autophagic vacuoles show bright red. Cells treated by different stimuli were incubated for 15 min with acridine orange (10 μ M in medium from a 10 mM stock in water), washed with PBS, and then examined under Leica TCS SP5 confocal laser microscopy.
WST-1 Cell viability assays. WST-1 cell reagent is a simple, colorimetric assay for use in measuring cell viability 46 . The assay principle is based on the conversion of the tetrazolium salt WST-1 into a colored dye by mitochondrial dehydrogenase enzymes. The soluble salt is released into the media. Within a given time period, the reaction produces a color change which is directly proportional to the amount of mitochondrial dehydrogenase in a given culture. As a result, the assay actually measures the net metabolic activity of cells. WST-1 Cell Proliferation and Cytotoxicity Assay Kit (Beyotime institute of Biotechnology) was used to assess the cell viability of the HUVECs. HUVECs were plated at 7500 cells/well into 96-well tissue culture plates and then were allowed to attach for 12 hours. Then the media was switched to ischemic buffer for 30 minutes followed by replacement with normal media in the absence or presence of rapamycin (0.5 mM) or 3-MA (10 mM) or PDTC (0.5 mM) or p65 siRNA (20 nM). WST-1 reagent was added in 24 hour increments up to 72 hours, and absorbance was recorded at 450 nm wavelength following the manufacturer's instructions. The experiments were conducted in triplicate three times. Electron microscopy. Electron microscopic analysis was performed as published previously 7 , with minor modifications. Briefly, HUVECs after treatment were fixed for 30 min with ice-cold 3% glutaraldehyde in 0.1 M cacodylate buffer, embedded in Epon, and processed for transmission electron microscopy using standard procedures. Representative areas were chosen for ultra-thin sectioning and examined on transmission electron microscope at X6, 000 magnification.