NF-κB-HSP27 Signaling Is Essential For Resistance To Heat Stress-induced Early Stage Apoptosis In Human Umbilical Vein Endothelial Cells

BACKGROUND AND AIMSHeatstroke is a life-threatening condition that typically develops following exposure to extended periods of high temperatures. It is characterized by a rapid increase in core temperature to more than 40℃ and multiple organ dysfunction syndrome (MODS). The critical maximum temperature for the human body is between 41.6℃ and 42.0℃. Previous studies have suggested that apoptosis is a major cause of cell death in heatstroke, and that it can be induced within a few hours. It is hypothesized that endothelial cell activation/injury contributes to the pathophysiology of heat stroke, and endothelial damage has been detected in heatstroke patients. In addition, recent studies have reported that the acute phase of heat stress induces significant apoptosis in endothelial cells, and we recently reported that intense heat stress induces early apoptosis via a transcription-independent mitochondrial p53 pathway. However, the mechanisms mediating cell death in the late phase of heat stress remain unclear.NF-κB is an important intracellular signaling protein that controls the transcription of several genes involved in cell growth, inflammatory responses, cell survival, and cell apoptosis. When NF-κB is associated with inhibitory molecules of the IκB family in the cytosol, it is inactive. Correspondingly, most of the inducers that activate NF-κB use a common pathway that involves phosphorylation-induced degradation of IκB proteins. The latter includes the major protein, IκBα, which was the first protein described for this family and is also the most extensively studied IκB protein to date. Phosphorylation and degradation of IκBα requires phosphorylation of the upstream target, IκB kinase (IKK), which contains two catalytic subunits, IKKa and IKKβ. Upon release from the NF-κB/IκBα dimer, NF-κB translocates from the cytoplasm into the nucleus to bind DNA and regulate transcription.The NF-κB signaling pathway has a critical role in regulating various aspects of the apoptotic program. For example, NF-κB activation has been shown to down-regulate pro-apoptotic JNK signaling in many cell types, thereby preventing apoptosis15,16. However, in certain pathological conditions, such as ischemia, the excessive accumulation of reactive oxygen species (ROS) can induce apoptosis or necrosis by activating mitogen-activated protein kinase (MAPK) and caspase signaling cascades, and/or by disrupting mitochondrial membrane potential. NF-κB has also been shown to exert pro-survival functions by inhibiting TNF-α-induced ROS accumulation-mediated prolongation of MAPK activation and necrotic cell death in murine embryonic fibroblasts. Despite these insights, however, it remains unknown whether ROS play a critical role in heat stress-induced MAPK activation, and whether NF-κB has a role in mediating oxidative stress and MAPK signaling pathways under physiological conditions in HUVECs.Heat shock proteins (HSPs) are an evolutionarily conserved set of proteins that mediate a cell’s response to heat stress, and a subset of HSPs protects cells against an induction of cell death (including apoptosis and necrosis) in response to a variety of stresses19. In particular, HSP27 and HSP70 have been shown to contribute to the regulation of NF-κB activation, with a direct link observed between HSP27 and regulation of the NF-κB signaling pathway in cell apoptosis. For example, in macrophage-conditioned intestinal epithelial cells stimulated with interleukin-1β (IL-1β), HSP27 was shown to bind and suppress IKK to regulate NF-κB activation. Similar mechanisms have been found in keratinocytes stimulated with tumor necrosis factor-α(TNF-α) and UV irradiation, and in HeLa cells stimulated with TNF-α. Moreover, when HSP27 is overexpressed in response to various stimuli, it facilitates proteasome-mediated proteolysis via phosphorylated IκBα and enhances NF-κB activity. The latter observation is consistent with the antiapoptotic properties reported for HSP27. To investigate whether regulation of NF-κB activation by HSP27 affects heat stress-induced cell apoptosis, various experiments were performed using HUVECs as a model. As a result, a novel NF-κB signaling pathway was identified that includes HSP27 protein expression and translocation into the nucleus, the accumulation of ROS, and subsequent MAPK activation.The purpose of this study is to investigate the mechanism of apoptosis induced by intense heat stress, including signaling pathway, upstream signal molecule and interaction between the signal molecule. These studies promote the understanding of induced endothelial cell apoptosis mechanism at molecular level, which provide theoretical basis for clinical heatstroke prevention and treatment.METHODS AND RESULTS1. Isolated human umbilical vein endothelial cells and established thermal stress model, heat stress induced NF-κB activation and the specific mechanisms of its activationAccording to the reference we isolated from human umbilical vein endothelial cells and with the use of VE-cadherin and VEGFR-2 cell phenotype to have a identification. After HUVEC cells were grown in culture media for 48 h, the culture dishes were sealed with parafilm and immersed in a circulating water bath maintained at 43℃ to induce heat stress. After 90 min, the culture media was replaced with fresh media and the cells were further incubated at 37℃ for various periods of time (e.g.,0, 2,6, and 12 h) before being assayed. Indirect immunofluorescence studies demonstrated that p65 was redistributed from the cytoplasm to the nucleus after 6-12 h of heat stress recovery at 37℃. When nuclear and cytoplasmic extracts from the same time points were collected and analyzed by Western blot, the nuclear and cytoplasmic p65 levels were both increased. An ELISA-based TransAM NF-κB Activation kit was then used to quantify NF-κB binding to DNA. Taken together, these data suggest that heat stress induces the translocation and activation of NF-κB during the recovery period in HUVECs.To determine whether activation of NF-κB is associated with the phosphorylation of IκBα, IKK-a/p, and/or p65, as well as the degradation of IκBα, Western blot assays were performed for HUVEC extracts collected during heat stress treatment and during the recovery period following the heat stress treatment. Phosphorylation levels of IKK-α/β were high at 37℃, and were unchanged by heat stress. In contrast, levels of phosphorylated IκBα and p65 were low at 37℃, but increased after 15 min and 90min of heat treatment. The levels of phosphorylated IκBα and p65 then further increased 24 h after the heat stress treatment was completed. There were no signs of IκBα degradation observed during the heat stress treatment or during the recovery periods assayed. Thus, it appears that NF-κB activation is accompanied by phosphorylation of p65 and IκBα, and not by degradation of IκBα in HUVECs.Construction of IicB-a inhibitory mutations at phosphorylation site mutations inhibiting its phosphorylation, which did not change p65 phosphorylation levels, nor alter p65 nuclear translocation at 12 h of the heat stress recovery period. We used small interfering RNA knockdown p65 found that IκB-α expression levels also reduce in heat stress recovery (2,6,12 h). Then co-immunoprecipitation assays were performed with an anti-NF-KB p65 antibody and Western blotting revealed that NF-κB p65 co-immunoprecipitated with IκB-α both at 37℃ and 6 h after a heat stress treatment and without IκB-α dissociation, while using LPS-stimulated endothelial cells as a positive control.These results indicate that NF-κB activation during heat stress recovery follows a non-canonical signal transduction pathway in HUVECs with increased phosphorylation of p65 and IκBα, without IκBα degradation and dissociation.NF-κB activation during the heat stress recovery period has been associated with the thermolability of the NF-κB·IκBα complex2. Co-localization and functional correlation of HSP27 with NF-κB in HUVECs during the recovery period following heat stressHUVECs that were pretreated with the NF-κB inhibitor, BAY11-7082, prior to heat stress treatment were more susceptible to high levels of early apoptosis compared with the HUVECs that were pretreated with DMSO. When HUVECs were transfected with a p65-targeted siRNA, lower levels of p65 protein were detected, and these cells were more susceptible to heat stress-induced early apoptosis. Caspase-3 activity was also assayed for HUVECs that were pretreated with DMSO or the NF-κB inhibitor, BAY11-7082 (5μM), for 1 h, and for HUVECs that were transfected with p65-targeted siRNA for 48 h, then underwent a heat stress treatment (Figure 3D-E). Higher levels of caspase-3 activity were detected when NF-κB was inhibited and when levels of p65 were knocked down.HUVECs were transfected with an HSP27-targeted siRNA or an adenovirus expressing HSP27, and then were subjected to a heat stress treatment followed by a 24 h recovery period. The knockdown and overexpression of HSP27 that was initially achieved was detected in Western blot assays. Levels of apoptosis were subsequently analyzed by flow cytometry using Annexin V-FITC/PI staining. Higher levels of heat stress-induced apoptosis were detected in HUVECs following the knockdown of HSP27, while an increase in apoptosis levels were detected in cells overexpressing HSP27. A similar profile was obtained when the corresponding cell lysates were analyzed for caspase-3 activity. In combination, these data suggest that NF-κB and HSP27 protects HUVECs from heat stress-induced apoptosis.Based on the anti-apoptotic properties of HSP27 during heat stress recovery, and previous observations that HSP27 and NF-κB translocate into the nucleus in response to heat stress, we hypothesized that NF-kB activation after heat stress may be linked to HSP27. Therefore, immunofluorescence studies of heat-stressed HUVECs were performed to detect the expression and localization of HSP27 and p65. Heat stress was found to stimulate the translocation of both HSP27 and NF-κB from the cytoplasm into the nucleus. To investigate the potential for interactions between NF-κB and HSP27, HUVECs were subjected to a heat stress treatment, they recovered at 37℃ for 6 h, and then whole cell lysates were prepared. Subsequent co-immunoprecipitation assays were performed with an anti-NF-κB p65 antibody and Western blotting revealed that NF-κB p65 co-immunoprecipitated with HSP27 both at 37℃ and 6 h after a heat stress treatment.To further assess the interactions between NF-κB and HSP27, HUVECs were transfected with an HSP27-targeted siRNA or an adenovirus expressing HSP27, and then were subjected to a heat stress treatment followed by a 6 h recovery period. Following the knockdown of HSP27 expression, DNA binding by NF-κB in the nucleus decreased. Although the NF-κB DNA-binding capacity did not notably increase after HSP27 overexpression, it may be that the levels of HSP27 themselves were increased after heat stress treatment. These results suggest that HSP27 may facilitate the nuclear import of NF-κB.Meanwhile, blocking NF-κB activation by the inhibitor BAY11-7082 also prevents Hsp27 from going to the nucleus. However, when the HUVECs were pretreated with BAY11-7082 before heat treatment, the nuclear and cytoplasmic levels of p65 decreased. It is possible that a feedback loop exists whereby NF-κB regulates p65 expression. Moreover, our data showed that heat stress significantly increased HSP27 protein expression during heat recovery periods, which were decreased by inaction of NF-κB with BAY11-7082 or p65siRNA. These results suggested that NF-κB signaling regulates the translocation and expression of HSP27.Next, we further analyzed whether NF-κB could regulated activation and expression of HSF-1. HUVECs subjected to a heat stress treatment at 43℃ for 15,30, or 90 min (A), or a HS at 43℃ for 90 min, followed by a recovery period at 37℃ for 0 h (R0),2 h (R2),6 h (R6), or 12 h (R12). The results showed that HSF1 phosphorylation was rapidly induced at 15min and was maintained at high levels for more than 6 h before decreasing to baseline levels after a 12 h recovery period. On the anti-HSF1 blots, slowly migrating bands were observed in cells exposed to heat shock. HSF1 is known to be a phosphorylated monomer under non-stress conditions, and heat shock induces hyperphosphorylation and trimerization of HSF1. while extracting the cytoplasm and nuclear protein detected by Western blotting to detect nuclear translocation situation of HSF1 in heat stress recovery (0,2,6,12 h). The results showed that HSF1 nuclear translocation occurred, more significant recovery 0-2 h, HSF1 was significantly lower in the nucleus at 6-12 h. HSF1siRNA were used to reduce HSF1 expression and Ad-HSF1 overexpression of HSF1, Western blotting to verify the transfection efficiency and HSP27 expression in normal and heat stress conditions, HSF1 could play a regulatory role for HSP27 expression. We used p65siRNA to detect its influence on HSF1 nuclear translocation and phosphorylation during heat stress, the results showed p65siRNA treated cells does not affect HSF1 nuclear translocation at 6h of heat stress recovery, while in p65siRNA treated cells phosphorylation of HSF1 levels did not have significant effects on phosphorylation levels of HSF1 at heat stress recovery (2,6,12 h). Therefore, we hypothesized that NF-κB may have a direct regulatory role on expression of HSP27.We checked by ChIP analysis whether NF-κB could activate transcription of HSP27 genes by binding to putative NF-κB binding sites in their promoters. We could not detect any specific binding of p65 to putative NF-κB binding sites of HSP27 promoters after heat stress. By contrast, an increased HSF1 binding to the HSP27 promoter after treatment with heat stress (HSF1as a positive control) was detected. However, after heat shock we observed an increased recruitment of RNApolⅡ to HSP27 promoters, which contain putative NF-κB binding sites. Furthermore, this increase was not observed in p65-KO cell line. These results suggest that NF-κB might regulate HSP27 at the transcriptional level. The location of NF-κB binding on the HSP27 promoters will need to be elucidated.3. ROS mediates the inhibition of MAPK phosphorylation by NF-κB activationTo examine the role of these three kinases in HUVECs during heat stress and after various periods of recovery following heat stress, HUVEC extracts were analyzed by Western blot. Phosphorylation of ERK and phosphorylation of p38 were significantly induced after 15 min at 43 oC, while phosphorylation of JNK occurred after 60 min at 43℃. During the recovery period MAPK phosphorylation was rapidly induced and was maintained at high levels for more than 4 h before decreasing to baseline levels after a 6 h recovery period. Since the decrease in MAPK levels temporally coincided with NF-κB activation, it was hypothesized that activation of NF-κB inhibited MAPK phosphorylation. Therefore, HUVECs were pretreated with BAY11-7082, SN50, or p65-targeted siRNA, then were subjected to a heat stress treatment and a 6 h recovery period. Decreasing levels of MAPK phosphorylation were observed except when NF-κB was inactivated. These data suggest that heat stress-induced phosphorylation of MAPK proteins is inhibited by NF-κB activation 6 h after a heat stress event.The antioxidant apocynin (APO) inhibited ROS accumulation in the heat stress group, but the effects were not complete in untreated HUVECs. We utilized the fluorescent dye DCFH-DA, which produces enhanced fluorescence when cells generate ROS, and analyzed fluorescent signals by flow cytometry. H2O2 was used as a positive control. Consistent with the partial inhibitory effect of APO on heat stress-induced ROS accumulation, APO only partially inhibited heat stress-induced MAPK activation. These results demonstrate that accumulation of ROS perfectly coincides with prolonged MAPK activation.The observed inhibition of MAPK activation by APO treatment in HUVECs prompted us to examine whether heat stress stimulation induces ROS accumulation in BAY11-7082-and SN50-pretreated cells and p65-depleted HUVEC cells. A substantial increase in the fluorescent signals of DCFH-DA was observed in BAY11-7082-or SN50-pretreated cells during the heat stress recovery period compared to the unpretreated group. The same results were obtained with p65-depleted HUVECs. In addition, p65-depleted HUVEC cells significantly increased the heat stress-induced the loss of mitochondrial membrane potential (△Ψ m), thereby suggesting that mitochondria may provide a permanent source of ROS in heat stressed HUVECs. Overall, the results of these experiments suggest that ROS contribute to the signaling pathway involving NF-κB activation-induced phosphorylation of MAPKs in response to heat stress.To examine whether accumulated ROS or MAPK activation participates in heat stress-induced cell apoptosis, HUVECs were stimulated with heat stress in the presence or absence of inhibitors for ROS or MAPKs. As shown in our results, treatment with APO or N-acetyl-L-cysteine (NAC) alone substantially decreased the number of cells undergoing early apoptosis. Furthermore, while HUVECs pretreated with PD98059, a specific inhibitor of ERK, exhibit an increase in cell apoptosis, HUVECs that were pretreated with specific inhibitors of JNK and p38, SP600125 and SB203580, respectively, exhibited a significant increase in cell survival. Caspase-3 activity was also assayed, and these results are consistent with the apoptosis data. Finally, We verified that these inhibitors actually inhibited MAP kinase activities by using antibodies specific for phosphorylated form of JNK, ERK, or specific substrate of p38. The three MAPK inhibitors examined also did not affect the heat stress-induced accumulation of ROS in HUVECs, thereby indicating that MAPK activation is a downstream event of ROS accumulation.CONCLUSION1. NF-κB activation during heat stress recovery follows a non-canonical signal transduction pathway in HUVECs with increased phosphorylation of p65 and IκBα, without IκBα degradation and dissociation, which together with p65 translocated to the nucleus regulating gene transcription. NF-κB activation during the heat stress recovery period has been associated with the thermolability of the NF-κB·IκBα complex.2. NF-κB p65 and HSP27 play an anti-apoptotic effect in the heat stressed HUVECs. NF-κB may modulate HSP27 expression levels and its translocation into the nucleus. In addition, HSP27 acting as a molecular chaperone may facilitate activation of this NF-κB pathway in HUVECs during heat stress recovery.3. Heat stress stimulation induces NF-κB activation, and this leads to an anti-apoptosis effect involving inhibition of heat stress-induced ROS accumulation that normally mediates the activation of MAPKs. In heat-stressed HUVECs, ERK1/2 activation has an anti-apoptotic role, while activation of JNK and p38 is pro-apoptotic role.