CREB and NF-kappaB transcription factors regulate sensitivity to excitotoxic and oxidative stress induced neuronal cell death - PubMed (original) (raw)
CREB and NF-kappaB transcription factors regulate sensitivity to excitotoxic and oxidative stress induced neuronal cell death
Jian Zou et al. Cell Mol Neurobiol. 2006 Jul-Aug.
Abstract
1. Glutamate-NMDA receptor excitotoxicity and oxidative stress are two common mechanisms associated with most neurodegenerative diseases. We hypothesize that the vital state of neurons is regulated in part by two key transcription factors, CREB and NF-kappaB. To test this hypothesis we used hippocampal-entorhinal cortex slice cultures. 2. Glutamate neurotoxicity and oxidative stress neurotoxicity, using hydrogen peroxide (H(2)O(2)) are both associated with a decrease in CREB DNA binding and an increase in NF-kappaB DNA binding. 3. Agents that modulate CREB and NF-kappaB DNA-binding activity alter neurotoxicity. Rolipram, a phosphodiesterase IV inhibitor, increased CREB DNA binding activity and decreased toxicity, whereas TNFalpha, increased NF-kappaB DNA-binding activity and increased neurotoxicity to both glutamate and H(2)O(2). Ethanol decreased CREB and increased NF-kappaB DNA-binding activity and increased neurotoxicity to both glutamate and H(2)O(2). 4. Brain-derived neurotrophic factor (BDNF) is a transcriptionally regulated trophic factor whose expression follows sensitivity to toxicity suggesting it is one of the transcriptionally regulated factors that contributes to neuronal vitality secondary to the balance of CREB-NF-kappaB-activated transcription. Together these studies suggest that neurotoxicity through glutamate-NMDA receptors or oxidative stress is dependent upon CREB and NF-kappaB DNA transcription that regulates vitality of neurons.
Figures
Fig. 1.
Reduction of CREB DNA binding by combined treatment of glutamate–TNFα–ethanol. HEC slices were prepared and cultured as described in the methods. Slices were treated with glutamate (3.3 mM) in the absence or simultaneous presence of TNFα (20 ng/mL) and/or ethanol (100 mM) for 24 h and removed for processing nuclear extracts after PI images were captured for analysis of neuronal cell death. EMSA was performed with 10 μg of nuclear extracts from the slices in each lane and a representative image from three experiments was shown in Lanes: 1. Control; 2. TNFα (20 ng/mL); 3. Ethanol (100 mM); 4. Glutamate 3.3 mM; 5. Glutamate+TNFα; 6. Glutamate+Ethanol; 7. Glutamate+TNFα+Ethanol. Image at left is one of three, with optical density measures bar graph on right, mean±standard error of the mean of three independent experiments of the integrated optical density quantified with BioQuant imaging as described in the methods. (* p<0.02 and ** p<0.03 compared to control, respectively).
Fig. 2.
Ethanol increases CREB and decreases NF-κB DNA binding. HEC slices were treated with various concentrations of ethanol for 24 h and then removed for processing of nuclear extracts. EMSA was performed with 10 μg of nuclear extracts in each lane. Treatments are shown at the top of each lane. Specificity of CREB and NF-κB binding was confirmed with 70-fold excessive unlabeled probe (not shown). Bar graph shows the integrated optical density quantified with BioQuant imaging as described in the methods. Shown are the mean±standard error of the mean of five independent experiments (_n_=5, * p<0.05). Ethanol dose-dependently decreases CREB but increases NF-κB DNA binding activities (E25=25 mM ethanol, E50=50 mM ethanol, etc.).
Fig. 3.
TNFα induced NF-κB DNA binding is increased in the presence of glutamate and ethanol. HEC slices were prepared and cultured as described in the methods. Slices were treated with TNFα (20 ng/mL) and glutamate (3.3 mM) in the absence or presence of ethanol (100 mM) for 8 h and then removed for processing nuclear extracts. EMSA was performed with 10 μg of nuclear extracts in each lane and a representative EMSA image shown. Lanes: 1. Control; 2. TNFα (20 ng/mL); 3. Ethanol (100 mM); 4. Glutamate 3.3 mM; 5. Glutamate+TNFα; 6. Glutamate+ethanol; 7. Glutamate+TNFα+ethanol. Bar graph shows the integrated optical density (mean±standard error) quantified from 3 independent experiments with BioQuant imaging program as described in the methods. Although ethanol and glutamate increase NF-kB DNA binding (Lane 3 and 4, * p<0.05 compared to Control), TNFα is many fold more efficacious (Lane 2 and 5,** p<0.0001 compared to Control). The presence of ethanol combined with TNFα and glutamate produces the most prominent induction of NF-kB DNA binding activity, which is correlated with maximal cell death. (*** p<0.0001 compared to Control; p<0.007 compared to TNF+glutamate).
Fig. 4.
Potentiation of glutamate induced neuronal cell death by ethanol and TNFα. HEC slices were treated with glutamate (1 and 3.3 mM) in the absence or simultaneous presence of TNFα (20 ng/mL) and/or ethanol (100 mM) for 24 h. Representative PI images of HEC slices are shown from the groups as indicated. The brighter the PI labeling means more cell death (bar=500 μm). The bar graph shows the quantitative data measured from CA1 areas of the slices and expressed as the mean±standard error of the mean of 9–12 slices per group. Similar results were found with two other experiments of comparable design. (* p<0.001 compared to Control; ** p<0.004 compared to corresponding groups of Glut 1 and 3.3 mM, respectively; *** p<0.01 compared to corresponding groups of Glut+TNFα, respectively).
Fig. 5.
Antioxidant BHT inhibits NF-κB-DNA binding and TNFα-Ethanol potentiation of glutamate toxicity. HEC slices were pretreated with antioxidant BHT and followed by TNFα in the absence or presence of glutamate or other combinations for 8 h. EMSA was performed with 10 μg of nuclear extracts in each lane. Lanes: 1. Control; 2. TNFα (20 ng/mL); 3. TNFα (100 ng/mL); 4. TNFα 100 ng/mL+BHT100 μM; 5. TNFα 100 ng/mL+BHT300 μM; 6. TNFα 100 ng/mL+BHT500 μM; 7. TNFα 100 ng/mL+AP-5 50 μM; 8. Glutmate 3.3 mM; 9. TNFα 20 ng/mL+glutamate; 10. TNFα 20 ng/mL+glutamate+BHT300 μM. The bar graph shows the quantitative data measured from CA1 areas of slices that were treated with different combinations as indicated and expressed as the mean±standard error of the mean of 9–15 slices per group. The experiment was repeated with comparable design and similar results. Antioxidant BHT significantly reduced TNFα-enhanced glutamate neurotoxicity (* p<0.001 compared to Glut3.3 mM; ** p<0.0001 compared to Glut+TNFα and G+T+EtOH).
Fig. 6.
Rolipram induced increase in CREB DNA binding activity and neuroprotection in HEC slices. HEC slices were treated with PDE inhibitor rolipram for 24 h and nuclear extracts prepared. EMSA was performed with 10 μg of nuclear extracts in each lane. Lanes: 1. Control; 2. Rolipram 1 μM and 3. Rolipram 10 μM. Rolipram dose-dependently increases CREB DNA Binding. The bar graph shows the quantitation of neurotoxicity from CA1 areas of slices that were treated with glutamate in the absence or presence of 10 μM Rolipram for 24 h and expressed as the mean±standard error of the mean of 9–12 slices per group. The similar results were obtained from other 2 experiments with comparable design. (* p<0.0001 compared to glutamate).
Fig. 7.
Increased CREB DNA binding activity correlates with reduction of neuronal cell death induced by Glutamate-TNFα-Ethanol in HEC slices. HEC slices were prepared and cultured as described in the methods. Slices were treated with glutamate in absence or presence of TNFα (20 ng/mL), ethanol (100 mM, pretreatment 40 min) and/or rolipram (pretreatment 40 min). In one set of experiments the slices were removed after treatment for 8 h for nuclear extractions preparations. EMSA was performed with 10 μg of nuclear extraction proteins in each lane. Lanes: 1. Control; 2. Rolipram 500 nM; 3. Rolipram 5 μM; 4. TNFα+ethanol +glutamate; 5. TNFα+ethanol+glutamate+Rolipram 5 μM; 6. TNFα+ ethanol+ glutamate+Rolipram 500 nM and lane P:-Probe only. Neuronal cell death was determined from PI images captured after treatment for 24 h and showed in bar graph. Shown are the mean±standard error of 8–12 slices in each groups. The experiments were at least repeated with similar designs and results (* p<0.001 comparing to Glutamate; ** p<0.05 compared to TNFα+Glutamate; *** p<0.0001 comparing to TNFα+Ethanol+Glutamate).
Fig. 8.
Time course and concentration-dependent studies of H2O2 toxicity in HEC slices. HEC slices were prepared and cultured as described in the methods. The slices were treated with H2O2 and PI images captured in hippocampal CA1. Shown are the mean±standard error of 8–10 slices per group. Similar results were found with two other experiments of comparable design. H2O2 induced cell death in concentration- (A) and time- (B) dependent manner.
Fig. 9.
Potentiation of H2O2 toxicity by TNFα and ethanol. HEC slices were treated with H2O2 (50 μM) with or without treatments as indicated for 24 h. Shown are the mean±standard error of the mean of 8 to 10 slices per group from a representative experiment of two experiments with similar designs (* p<0.01 comparing with H2O2; ** p<0.05 comparing with H2O2+TNFα; *** p<0.001 comparing to corresponding group).
Fig. 10.
Reduction of CREB DNA binding activity by treatment with H2O2 and different combinations in HEC slices. HEC slices were prepared and cultured as described in the methods. Slices were treated with H2O2 (50 μM) in the absence or simultaneous presence of TNFα (20 ng/mL) and/or ethanol (100 mM) for 12 h and removed for processing nuclear extracts after PI images were captured for analysis of neuronal cell death. EMSA was performed with 10 μg of nuclear extracts from the slices in each lane. Lanes: 1. Control; 2. H2O2 (50 μM); 3. H2O2+Ethanol (100 mM); 4. H2O2+TNFα (20 ng/mL); 5. TNFα+Ethanol+H2O2; 6. TNFα+Ethanol+H2O2+GSH and 7. TNFα+Ethanol+H2O2+PD98589 (50 μM). Bar graph shows the integrated optical density quantified with BioQuant imaging program as described in the methods. Shown are the mean±standard error of the mean of triplicate EMSA from two independent experiments (* p<0.03 comparing to control).
Fig. 11.
NF-κB DNA binding induced by combined H2O2, TNFα and ethanol. HEC slices were treated with H2O2 alone or with agents as indicated for 12 h and nuclear extracts prepared. NF-κB DNA binding activity by EMSA was performed with 10 μg of nuclear extracts in each lane. Lanes: 1. Control; 2. H2O2 (100 μM); 3. H2O2+Ethanol (100 mM); 4. H2O2+TNFα; 5. H2O2+TNFα+Ethanol; 6. H2O2+TNFα+Ethanol+GSH and 7. H2O2+ TNFα+ Ethanol+PD98589. Bar graph shows the integrated optical density quantified with BioQuant imaging program as described in the methods. Shown are the mean±standard error of the mean of triplicate from two independent experiments (* p<0.05 compared to control; ** p<0.05 compared to group of H2O2+TNFα and *** p<0.03 comparing to H+T+E).
Fig. 12.
Inhibition of H2O2 toxicity by antioxidants and Rolipram. HEC slices were treated with H2O2 in the absence or presence of MK-801 (30 μM), glutathione (GSH, 500 μM), BHT (300 μM) and rolipram (5 μM) for 12 h and then PI images were captured in hippocampal CA1. Representative psudocolor color PI images of HEC slices are shown in Control (A); H2O2 (100 μM) (B); H2O2+BHT (C) and H2O2+GSH (D). H2O2-induced cell death is significantly reduced by GSH and BHT (bar=500 μm). The bar graph shows the quantitative data measured from CA1 expressed as the mean±standard error of the mean of 6–10 slices per group. The experiments were done in triplicate with comparable design and similar results. (* p<0.001 comparing to corresponding groups).
Fig. 13.
Alterations in CREB target gene BDNF in HEC slices. HEC slices were prepared and cultured as described in the methods. Slices were treated with glutamate or H2O2 (50 μM) in the absence or simultaneous presence of TNFα (20 ng/mL) and/or ethanol (100 mM) for 4 h and removed for processing BDNF RT-PCR as described in the methods. Shown are the mean±SEM of the mean of triplicate RT-PCR measurements in a representative experiment. BDNF mRNA expression is significantly decreased with treatments, however, combined treatments shows no further reduction in BDNF mRNA expression compared to corresponding groups (* p<0.05 comparing to Control). RT-PCR products were confirmed with 1.5% argrose gel.
References
MeSH terms
Substances
LinkOut - more resources
Full Text Sources