High Glucose Induces Human Endothelial Cell Apoptosis Through a Phosphoinositide 3-Kinase–Regulated Cyclooxygenase-2 Pathway
http://www.100md.com
动脉硬化血栓血管生物学 2005年第3期
From the Institutes of Toxicology (M.L.S., K.F.C., S.Y.L.S., S.H.L.) and Pharmacology (S.Y.L.S., W.W.L.), and the Department of Orthopedics (R.S.Y.), College of Medicine, National Taiwan University, Taipei; and the Department of Internal Medicine (F.M.H.), Tao-Yuan General Hospital, Taoyan, Taiwan.
Correspondence to Shing-Hwa Liu, PhD, Institute of Toxicology, College of Medicine, National Taiwan University, No. 1, Section 1, Jen-Ai Rd, Taipei, 10043, Taiwan. E-mail shliu@ha.mc.ntu.edu.tw
Abstract
Objectives— Diabetes mellitus causes endothelial dysfunction. The precise molecular mechanisms by which hyperglycemia causes apoptosis in endothelial cells are not yet well understood. The aim of this study was to explore the role of cyclooxygenase-2 (COX-2) and the possible involvement of phosphoinositide 3-kinase (PI3K) signaling in high glucose (HG)–induced apoptosis in human umbilical vein endothelial cells (HUVECs).
Methods and Results— For detection of apoptosis, the morphological Hoechst staining and Annexin V/propidium iodide staining were used. Glucose upregulated COX-2 protein expression, which was associated with the induction of prostaglandin (PG) E2 (PGE2), caspase-3 activity, and apoptosis. Unexpectedly, we found that PI3K inhibitors could suppress COX-2 expression, PGE2 production, caspase-3 activity, and the subsequent apoptosis under HG condition. Glucose-induced activation of PI3K resulted in the downstream effector Akt phosphorylation. PI3K inhibitors effectively attenuated the intracellular reactive oxygen species (ROS) generation and nuclear factor B (NF-B) activation. Blocking the PI3K and Akt activities with the dominant-negative vectors greatly diminished the HG-triggered NF-B activation and COX-2 expression and apoptosis.
Conclusions— These results suggest that HG, via PI3K/Akt signaling, induces NF-B–related upregulation of COX-2, which in turn triggers the caspase-3 activity that facilitates HUVEC apoptosis. Also, HG may cause ROS generation in HUVECs through a PI3K/Akt–dependent pathway.
We demonstrated for the first time that the signaling pathway of PI3K/Akt–regulated upregulation of COX-2 was involved in glucose-induced apoptosis in human endothelial cells. These findings were supported by results that PI3K inhibition prevented the glucose-caused COX-2–mediated PGE2 production, which subsequently led to the induction of caspase-3 activity and apoptosis.
Key Words: glucose ? endothelium ? cells ? apoptosis
Introduction
Diabetes mellitus can cause a wide variety of vascular complications and cardiovascular dysfunction.1 The mechanisms of hyperglycemia-related tissue damage and clinical complications still remain unclear. Apoptosis is particularly prominent in models of hyperglycemia injury, affecting a significant proportion of vascular endothelium in the tissue damage.2–4 Studies have indicated that oxidative stress can induce apoptosis, which may be regulated by different signaling pathways.5,6 Involvement of reactive oxygen species (ROS) in the induction of apoptosis by high glucose (HG) has been demonstrated in cultured human endothelial cells.7 The formation of oxygen-derived radicals might lead to an activation of transcription factors such as nuclear factor B (NF-B) transferring the activation signal into the nucleus and leading to changes in gene expression necessary for induction of apoptosis. We found previously a c-Jun amino-terminal kinase–dependent pathway was involved in the ROS-related HG-induced apoptosis in human endothelial cells.8 It has also been demonstrated that oxidative stress induced apoptosis in human aortic endothelial cells through the downregulation of bcl-2, translocation of bax, and upregulation of p53, probably through NF-B activation.9 Nevertheless, the precise molecular mechanisms by which HG causes apoptosis in endothelial cells are not yet well understood.
Prostaglandins (PGs) generated by cyclooxygenase (COX) have been implicated in hyperglycemia-induced endothelial dysfunction.10–13 A recent study has shown that HG induced oxidative stress and upregulation of COX-2, resulting in reduced NO availability and altered prostanoid profile in human endothelial cells.14 Production of PG E2 (PGE2), via COX-2 induction, has been demonstrated to be related to the induction of cell death in human osteoarthritis chondrocytes by NO.15 However, the role of COX-2 in HG-caused cell death and the molecular mechanisms regulating COX-2 expression in endothelial cells remain to be clarified. Moreover, it has been suggested that phosphoinositide 3-kinase (PI3K) might play a central role in a diverse range of cellular responses, including cell growth, survival, and malignant transformation.16,17 Recent studies have shown that activation of PI3K and Akt might play a role in regulating COX-2 expression in some cultured cells.18–20 Together, in the current study, we explored the role of COX-2 and the possible involvement of PI3K signaling in HG-induced apoptosis in human endothelial cells. We also examined the role of the PI3K pathway in ROS generation, which might regulate the HG-induced apoptosis in endothelial cells.
Methods
Cell Culture
Human umbilical vein endothelial cells (HUVECs) were obtained by collagenase treatment of umbilical cord veins as described previously.8,21 Cells were cultured on gelatin-coated dishes and propagated in M199 medium supplemented with 20% bovine calf serum (Gibco), 50 μg/mL endothelial cell growth factor, 90 μg/mL heparin, 100 IU/mL penicillin, and 0.1 mg/mL streptomycin in humidified air, 5% CO2 at 37°C. HUVECs of the third to fifth passages were used. In some experiments, transfection in HUVECs was performed using the Lipofectin reagent (BD CLONTECH) according to manufacturer recommendations with 2 μg of total DNA (or vector) per sample. The efficiency of transfection (80%) was determined by using an equal amount of a plasmid encoding the green fluorescent protein under the cytomegalovirus promoter. Plasmids containing the dominant-negative (DN)-p85 (-p85) or DN-Akt (Akt K179A)22,23 were kindly provided by Drs R.H. Chen and M.L. Kuo (Institutes of Molecular Medicine and Toxicology, National Taiwan University); DN-inhibitor B (IB)24 was provided by Dr W.W. Lin. (Institutes of Pharmacology, National Taiwan University).
Apoptosis and Caspase-3/Cleaved Caspase 3 Activity
HUVECs were collected and fixed in methanol/acetone solution for 5 minutes and washed with PBS. Fixed cells were then stained with 0.1 ng/mL Hoechst 33258 (Boehringer Mannheim) for 10 minutes in the dark to counterstain nuclei. Cells were observed and photographed under a Nikon fluorescence microscope. In some experiments, the Annexin V/propidium iodide (PI; BD CLONTECH) was used to quantify numbers of apoptotic cells. Cells were washed twice with PBS and stained with Annexin V and PI for 20 minutes at room temperature. The level of apoptosis was determined by measuring the fluorescence of the cells by flow cytometry analysis.
Also, we measured caspase-3 activity to quantify the apoptotic ratio. Caspase-3 activity was measured as described previously.8,25 Cell lysates were incubated at 37°C with 10 μmol/L N-acetyl-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin (Ac-DEVD-AMC), a caspase-3/Cleaved Caspase 3 (CPP32) substrate. The fluorescence of the cleaved substrate was measured by a spectrofluorometer with an excitation wavelength at 380 nm and an emission wavelength at 460 nm.
Cell Viability
Viability was measured using the Cell Titer 96TM AQueous cell viability assay kit (Promega). This assay is based on the cellular conversion of the colorimetric reagent MTS (3,4-(5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium salt), in the presence of electron-coupling reagent phenazine methosulfate, into soluble formazan by dehydrogenase enzymes found only in metabolically active, living cells. Formazan formation was measured on the basis of increased absorbance at 490 nm.
PI3K Activity
PI3K activity was assayed as described previously.26 Cell extracts were incubated with 2 μg of anti-p85 antibody overnight at 4°C. The immunocomplex was precipitated with protein A-Sepharose for 1 hour at 4°C and washed 3x with lysis buffer, twice with LiCl buffer, and twice with Tris-NaCl-EDTA (TNE) buffer (10 mmol/L Tris, 100 mmol/L NaCl, and 1 mmol/L EDTA). The immunocomplex was preincubated with 20 mmol/L Hepes, containing 2 mg/mL PI on ice for 10 minutes. Kinase reaction was performed by adding reaction buffer (10 μCi of [-32P]ATP, 20 mmol/L Hepes, pH 7.4, 20 μmol/L ATP, and 5 mmol/L MgCl2) for 15 minutes. The radiolabeled lipids were separated by thin-layer chromatography and visualized by phosphorimaging.
Immunoblotting
Protein levels of Akt, phospho-Akt, and COX-2 were analyzed by Western blot as described previously.26 Cell lysates were prepared, electrotransferred, and then immunoblotted with anti-Akt, anti–phospho-Akt (New England BioLabs), and anti–COX-2 antibodies (Santa Cruz Biotechnology). Detection was performed with Western blotting reagent ECL (Amersham), and chemiluminescence was exposed by the filters of Kodak X-Omat films.
Electrophoretic Mobility Shift Assay
Electrophoretic mobility shift assay (EMSA) was performed as described previously.27 The following oligonucleotide with the NF-B consensus binding sequence was used: 5'-GATCCAAGGGGACTTTCCATGGAT-CCAAGGGGACTTTCCATG-3' (Life Technologies). The consensus oligonucleotide probes were end-labeled with [-32P]ATP according to the manufacturer description. Protein–DNA complexes were separated by electrophoresis on a nondenaturing 6% polyacrylamide gel and visualized by autoradiography. For competition experiments, a 100-fold molar excess of the unlabeled oligonucleotides was added 15 minutes before incubation of nuclear extracts with the end-labeled oligonucleotides.
ROS Production
Intracellular ROS generation was monitored by flow cytometry using peroxide-sensitive fluorescent probe 2',7'-dichlorofluorescin diacetate (DCFH-DA; Molecular Probes) as described previously.8 DCFH-DA is converted by intracellular esterases to DCFH, which is oxidized into the highly fluorescent dichlorofluorescin (DCF) in the presence of a proper oxidant, and then analyzed by flow cytometry.
PGE2 Production
PGE2 levels in cell culture supernatants were determined using the PGE2 enzyme immunoassay kit (Cayman Chemical).
Statistical Analyses
Results are expressed as mean±SEM and data analyzed by Student t test. For multiple comparisons, results were analyzed by ANOVA followed by Fisher’s test. P<0.05 was considered statistically significant.
Results
COX-2 Protein Expression and Involvement in HG-Induced Apoptosis
HUVECs exposed to HG (30 mmol/L) for 8 to 12 hours increased COX-2 protein expression (4.2-fold induction; Figure 1Aa), whereas no changes were observed for COX-1 protein level (data not shown). HG-induced increase of COX-2 expression was associated with a higher release of PGE2, which could be inhibited by selective COX-2 inhibitor NS398 (10 and 20 μmol/L; Figure 1Ba). We also found that combining HG with PGE2 (1 μmol/L) showed an additive effect on HUVEC death (Figure 2A). Moreover, we examined whether COX-2–mediated PGE2 generation was involved in the HG-induced apoptosis. The inhibitory effect of NS398 on HG-induced apoptosis was demonstrated in caspase-3 activity assay (Figure 2B) and morphological characteristic Hoechst-33258 staining (Figure 2C). The HG-induced cell death could be reversed by NS398 (Figure 3A). Mannitol (30 mmol/L), used as an osmotic control, did not affect COX-2 expression or apoptosis and cell viability (Figures 1A, 2C, and 3A).
Figure 1. Aa, Expression of COX-2 protein in HUVECs exposed to control (5.5 mmol/L) and 30 mmol/L glucose (HG) or mannitol (HM) for 8 to 12 hours. Ab, Expression of COX-2 protein in HUVECs exposed to HG for 12 hours in the presence or absence of LY294002 (LY) and wortmannin (WM). In some experiments, cells were transfected with the dominant-negative forms of PI3K-p85 (DN-p85), and then the HG-induced COX-2 protein expressions were determined (Ac). Results shown are representative of at least 4 independent experiments. Ba, Bar graphs showing release of PGE2 in HUVECs exposed to HG in the presence or absence of NS398 (10 and 20 μmol/L). Bb, Bar graphs showing the effect of LY294002 (10 and 20 μmol/L) on the release of PGE2 in HUVECs exposed to HG. All data are presented as mean±SEM from 5 experiments performed in duplicate. *P<0.05 vs control; #P<0.05 vs HG alone.
Figure 2. A, Cells were treated with HG or PGE2 or HG+PGE2 for 48 hours and then cell viability detected. B, Analysis of caspase-3/CPP32 activity in HUVECs treated with HG for 48 hours in the presence or absence of NS398 (20 μmol/L). All data are presented as mean±SEM from 5 experiments performed in duplicate. *P<0.05 vs control; #P<0.05 vs HG alone. C, Apoptosis induced by HG was determined by fluorescent dye Hoechst-33258 method. Ca, Control; Cb, HG exposure for 48 hours; Cc, HG+NS398; Cd, high mannitol (HM) exposure for 48 hours. Results shown are representative of at least 4 independent experiments.
Figure 3. A, HUVECs were treated with HG for 72 hours in the presence or absence of PI3K inhibitor LY294002 (LY; 20 μmol/L) or COX-2 inhibitor NS398 (20 μmol/L), and then cell viability was detected. B, Analysis of caspase-3 activity. HUVECs pretreated with HG for 48 hours showed increase of caspase-3/CPP32 activity. LY294002 (LY; 20 μmol/L) prevented the effect of HG. All data are presented as mean±SEM from 5 independent experiments performed in duplicate. *P<0.05 vs control; #P<0.05 vs HG alone. C, Apoptosis induced by HG was determined by fluorescent dye Hoechst 33258 method. Ca, Control; Cb, HG exposure for 48 hours; Cc, HG+LY294002 (LY); Cd, HG+wortmannin (WM/100 nmol/L). Results shown are representative of at least 4 independent experiments.
PI3K Activity and Involvement in HG-Induced COX-2 Protein Expression and Apoptosis
HUVECs were cultured with HG for 12 to 72 hours in the presence or absence of PI3K inhibitors (LY294002 and wortmannin). HG-induced increased COX-2 protein expression and its related PGE2 release and cell death could be inhibited by LY294002 (10 or 20 μmol/L) or wortmannin (100 nmol/L; Figure 1Ab and 1Bb; Figure 3A). Moreover, the inhibitory effect of PI3K inhibitors on HG-induced apoptosis was demonstrated in caspase-3 activity assay (Figure 3B), morphological characteristic Hoechst 33258 staining (Figure 3C), and Annexin V/PI assay (Figure 4Ca).
Figure 4. A and B, HG activated the PI3K activity (A) and the phosphorylation of Akt (B). HUVECs were treated with HG in the presence or absence of LY294002 (LY; 10 or 20 μmol/L) for 5 to 10 (A) or 10 to 60 (B) minutes. Results shown are representative of at least 4 independent experiments. C, HUVECs were treated with HG in the absence or presence of LY294002 (LY; 20 μmol/L). Apoptosis induced by HG was determined by Annexin V/PI staining. Data represent percentage of Annexin V–positive cells (Ca). In some experiments, HUVECs were transiently transfected with DN-p85 or DN-Akt. Cell viability was detected by MTS assay (Cb). D, Analysis of caspase-3 activity. HUVECs were treated with HG in the absence or presence of LY294002 (LY; 20 μmol/L) or transfection of DN-p85 or DN-Akt. All data are presented as mean±SEM from 5 independent experiments performed in duplicate. *P<0.05 vs control; #P<0.05 vs HG alone.
Furthermore, the authentic PI3K activity of HUVECs after treatment with HG was determined. Immunoprecipitates with the anti-p85 antibody revealed a substantial increase in PI3K activity 5 to 10 minutes after HG exposure, which could be inhibited by LY294002 (10 μmol/L; Figure 4A). To further evaluate the involvement of Akt in HG-triggered responses, Akt activity was examined by determining the serine-phosphorylated status of Akt using an anti–phospho-Akt antibody. In a time-dependent manner, Akt activation was shown 10 to 60 minutes after HG treatment (Figure 4B). LY294002 markedly reduced this HG-induced response (Figure 4B).
The next aim of the investigation was to ascertain whether PI3K/Akt activity inhibition might affect the HG-induced COX-2 protein expression and apoptosis in HUVECs. To address this issue, cells were transfected with a dominant-negative form of the regulatory subunit (p85) of PI3K (DN-p85) or a dominant-negative form of Akt (DN-Akt).26 Transfection results revealed that the DN-p85 or DN-Akt transfection significantly inhibited HG-induced COX-2 protein expression (Figure 1Ac) and decreased cell viability (Figure 4Cb) and increased caspase-3 activity (Figure 4D) and apoptosis (Annexin/PI staining; Figure 5A). The control pcDNA3 transfection did not affect the HG-induced responses (data not shown).
Figure 5. The HG-induced apoptosis in HUVECs requires activation of PI3K. HUVECs were transiently transfected with DN-p85, DN-Akt, or DN-IB. A, Cells were analyzed for apoptosis by Annexin V/PI staining. The percentage of cells found in each quadrant of the dot plot is depicted. Results shown are representative of at least 4 independent experiments. B, Nuclear extracts were subjected to an EMSA. Cells were treated with HG in the presence or absence of LY294002 (LY; 20 μmol/L) for 4 hours. Quantification of the NF-B binding activity was performed by densitometric analysis. *P<0.05 vs control; #P<0.05 vs HG alone
Effects of PI3K Inhibitors on HG-Induced NF-B Activation and ROS Generation
As shown in Figure 5B, stimulation of HUVECs with HG for 4 hours showed a marked increase of NF-B activation, as determined by EMSA. LY294002 and transfection of DN-p85, DN-Akt, or DN-IB abolished this HG-induced NF-B activation. Moreover, cells transfected with a dominant-negative form of IB (DN-IB) significantly inhibited HG-induced COX-2 protein expression (Figure 6A), caspase-3 activity (HG 2.1±0.2; HG+DN-IB 1.1±0.3-fold of control; n=3; P<0.05), apoptosis (Figure 6B), and cell death (Figure 6C). Phorbol 12-myristate 13-acetate (PMA)–induced COX-2 protein expression in HUVECs could also be inhibited by DN-IB transfection (as a positive control; Figure 6A). We next investigated whether the HG-induced generation of DCF-sensitive ROS requires activity of PI3K. Treatment with HG on HUVECs for 36 to 48 hours was found to increase DCF fluorescence. LY294002 (20 μmol/L) treatment completely suppressed this HG-induced response (Figure 6D).
Figure 6. HUVECs were treated with HG in the presence or absence of transfection of DN-IB. A, Expression of COX-2 protein in HUVECs exposed to HG or PMA for 12 hours. Quantification of the COX-2 expression was performed by densitometric analysis. B, Apoptosis induced by HG was determined by Annexin V/PI staining. Data represent percentage of Annexin V–positive cells. C, HUVECs were treated with HG for 72 hours, and cell viability was detected by MTS assay. D, HG-induced intracellular DCF-sensitive ROS generation. HUVECs were exposed to HG for 36 to 48 hours in presence or absence of LY294002, antioxidants (ascorbic acid [vitamin C]; 100 μmol/L), diphenyleneiodonium (DPI; 20 μmol/L), or rotenone (ROT; 20 μmol/L). All data are presented as mean±SEM from 4 independent experiments. *P<0.05 vs control; #P<0.05 vs HG or PMA alone.
Discussion
In the present study, we demonstrate for the first time that HG causes PI3K/Akt–dependent upregulation of inducible COX-2 expression, as well as an increase of PGE2 production and increased caspase-3 activity and induction of apoptosis in HUVECs. These findings are supported by functional studies of HUVECs showing that the activity of PI3K and Akt is triggered by HG, and the inhibition of PI3K/Akt activity with LY294002 or wortmannin, or by expressing the dominant-negative p85 or Akt is capable of preventing the HG-caused COX-2–mediated PGE2 production, and subsequently leads to the inhibition of caspase-3 activity and apoptosis. Moreover, HG induces a PI3K/Akt–regulated ROS generation. Hence, activation of the PI3K/Akt pathway may represent a proximal node in the intracellular signaling leading to hyperglycemia-induced endothelial cell apoptosis.
In a number of cell and animal models, induction of COX-2 has been shown to promote cell growth, inhibit apoptosis, and enhance cell motility and adhesion.28 Paradoxically, PGE2 production via the induction of COX-2 expression has been demonstrated to relate to the induction of cell death in some cultured cells such as human articular chondrocytes15,29 and human synovial fibroblasts.30 Similarly, in the present study, we found that HG induced COX-2 protein expression in HUVECs, which was associated with a higher release of PGE2. Selective COX-2 inhibitor NS398 inhibited PGE2 production and blocked increased caspase-3 activity and apoptosis induced by HG. Combining HG with PGE2 showed an additive effect on cell death. Moreover, NF-B was known as a transcriptional regulator of expression of COX-2 gene.31 NF-B has also been demonstrated as a mediated signaling in HG (hyperglycemia) or oxidative stress-triggered apoptosis of endothelial cells.7,9 Activation of NF-B promoted p53 upregulation, which might suppress the level of bcl-2, resulting in endothelial cell apoptosis.9 We also observed that HG could trigger the NF-B activation in HUVECs, and dominant-negative IB transfection could inhibit HG-induced COX-2 protein expression and apoptosis. Therefore, these findings suggest an involvement of COX-2–mediated PGE2 production in HG-induced HUVEC apoptosis through activation of NF-B. On the other hand, in the clinical setting, hyperglycemia was usually accompanied by increased free fatty acid levels, which might also have detrimental effects on endothelial cells.32 It has been suggested that the alterations in fatty acid metabolism, impaired Akt activation by insulin, and increased caspase-3 activity preceded visible evidence of apoptosis in HUVECs incubated in a hyperglycemic medium.33 It has also been identified that saturated fatty acids induced the expression of COX-2 in monocyte/macrophage cells.34 Hence, the involvement of free fatty acid in the hyperglycemia-induced COX-2 expression and apoptosis in HUVECs may need to clarify in the future.
PI3K signal plays an important regulatory role in a diverse range of cellular responses, including cell growth, survival, and malignant transformation.16,17 It has been shown that PI3K activation has an antiapoptotic effect in endothelial cells35 and elsewhere.36 Unexpectedly, we observed that the treatment of HG in HUVECs leads to activation of PI3K and Akt, which subsequently leads to activation of NF-B and COX-2 expression, and then induces apoptosis. At present, the physiological relevance of the activation of PI3K and Akt during HG-induced HUVEC death is not clear. In accordance with our present findings, such an apoptosis-promoting effect of PI3K has been reported in other cell systems. For example, a rapid and transient activation of PI3K was critical for Fas or C6-ceramide–induced apoptosis in Jurkat cells;37 a significant increase of PI3K activity was observed during Fas-induced human neutrophil apoptosis and suggested that Fas-induced PI3K activity was a proapoptotic signal;38 and activation of PI3K/Akt was induced during pyrrolidine dithiocarbamate–induced neuronal cell death.39 Interferon-–induced apoptosis in tumor cells was mediated through the PI3K/mammalian target of rapamycin signaling pathway.40 Recent studies have also shown that activation of PI3K might play a role in regulating the COX-2 expression in some cultured cells.18–20 Therefore, the current results indicate that PI3K/Akt may be involved in the HG-induced HUVEC apoptosis, and PI3K/Akt is present upstream of NF-B activation and COX-2 expression in the HG-triggered signaling pathways.
In conclusion, our study demonstrates that activation of a PI3K/Akt–signaling pathway is required for HG-induced activation of NF-B and COX-2–mediated PGE2 production and subsequent caspase-3 activation and facilitation of apoptosis in HUVECs. Also, HG triggers ROS generation through a PI3K-dependent pathway. These findings may be relevant in understanding the possible intracellular signaling associated with diabetic hyperglycemia-induced detrimental effects on endothelial cells and provide novel therapeutic strategies to prevent the development and progression of vascular complications in diabetes mellitus.
Acknowledgments
This work was supported by research grants from National Science Council of Taiwan (NSC90-2315-B-002-009; NSC91-2314-B-002-248). The authors thank Professor Ming Liang Kuo (National Taiwan University, Taipei) for discussion regarding PI3K activity.
Received September 19, 2004; accepted December 30, 2004.
References
The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993; 329: 977–986.
Ceriello A, Quagliaro L, D’Amico M, Di Filippo C, Marfella R, Nappo F, Berrino L, Rossi F, Giugliano D. Acute hyperglycemia induces nitrotyrosine formation and apoptosis in perfused heart from rat. Diabetes. 2002; 51: 1076–1082.
Nakagami H, Morishita R, Yamamoto K, Yoshimura SI, Taniyama Y, Aoki M, Matsubara H, Kim S, Kaneda Y, Ogihara T. Phosphorylation of p38 mitogen-activated protein kinase downstream of bax-caspase-3 pathway leads to cell death induced by high D-glucose in human endothelial cells. Diabetes. 2001; 50: 1472–1481.
Zou MH, Shi C, Cohen RA. High glucose via peroxynitrite causes tyrosine nitration and inactivation of prostacyclin synthase that is associated with thromboxane/prostaglandin H(2) receptor-mediated apoptosis and adhesion molecule expression in cultured human aortic endothelial cells. Diabetes. 2002; 51: 198–203.
Buttke TM, Sandstrom PA. Oxidative stress as a mediator of apoptosis. Immunol Today. 1994; 15: 7–10.
Vaux DL, Strasser A. The molecular biology of apoptosis. Proc Natl Acad Sci U S A. 1996; 93: 2239–2244.
Du X, Stocklauser-Farber K, Rosen P. Generation of reactive oxygen intermediates, activation of NF-kappaB, and induction of apoptosis in human endothelial cells by glucose: role of nitric oxide synthase? Free Radic Biol Med. 1999; 27: 752–763.
Ho FM, Liu SH, Liau CS, Huang PJ, Lin-Shiau SY. High glucose-induced apoptosis in human endothelial cells is mediated by sequential activations of c-Jun NH(2)-terminal kinase and caspase-3. Circulation. 2000; 101: 2618–2624.
Aoki M, Nata T, Morishita R, Matsushita H, Nakagami H, Yamamoto K, Yamazaki K, Nakabayashi M, Ogihara T, Kaneda Y. Endothelial apoptosis induced by oxidative stress through activation of NF-kappaB: antiapoptotic effect of antioxidant agents on endothelial cells. Hypertension. 2001; 38: 48–55.
Ayalasomayajula SP, Kompella UB. Celecoxib, a selective cyclooxygenase-2 inhibitor, inhibits retinal vascular endothelial growth factor expression and vascular leakage in a streptozotocin-induced diabetic rat model. Eur J Pharmacol. 2003; 458: 283–289.
Gryglewski RJ, Chlopicki S, Uracz W, Marcinkiewicz E. Significance of endothelial prostacyclin and nitric oxide in peripheral and pulmonary circulation. Med Sci Monit. 2001; 7: 1–16.
Meeking DR, Browne DL, Allard S, Munday J, Chowienczyck PJ, Shaw KM, Cummings MH. Effects of cyclo-oxygenase inhibition on vasodilatory response to acetylcholine in patients with type 1 diabetes and nondiabetic subjects. Diabetes Care. 2000; 23: 1840–1843.
Tesfamariam B. Free radicals in diabetic endothelial cell dysfunction. Free Radic Biol Med. 1994; 16: 383–391.
Cosentino F, Eto M, De Paolis P, van der LB, Bachschmid M, Ullrich V, Kouroedov A, Delli GC, Joch H, Volpe M, Luscher TF. High glucose causes upregulation of cyclooxygenase-2 and alters prostanoid profile in human endothelial cells: role of protein kinase C and reactive oxygen species. Circulation. 2003; 107: 1017–1023.
Notoya K, Jovanovic DV, Reboul P, Martel-Pelletier J, Mineau F, Pelletier JP. The induction of cell death in human osteoarthritis chondrocytes by nitric oxide is related to the production of prostaglandin E2 via the induction of cyclooxygenase-2. J Immunol. 2000; 165: 3402–3410.
Krasilnikov MA. Phosphatidylinositol-3 kinase dependent pathways: the role in control of cell growth, survival, and malignant transformation. Biochemistry (Mosc). 2000; 65: 59–67.
Toker A. Protein kinases as mediators of phosphoinositide 3-kinase signaling. Mol Pharmacol. 2000; 57: 652–658.
Sheng H, Shao J, Dubois RN. K-Ras-mediated increase in cyclooxygenase 2 mRNA stability involves activation of the protein kinase B1. Cancer Res. 2001; 61: 2670–2675.
Tang Q, Gonzales M, Inoue H, Bowden GT. Roles of Akt and glycogen synthase kinase 3beta in the ultraviolet B induction of cyclooxygenase-2 transcription in human keratinocytes. Cancer Res. 2001; 61: 4329–4332.
Weaver SA, Russo MP, Wright KL, Kolios G, Jobin C, Robertson DA, Ward SG. Regulatory role of phosphatidylinositol 3-kinase on TNF-alpha–induced cyclooxygenase 2 expression in colonic epithelial cells. Gastroenterology. 2001; 120: 1117–1127.
Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest. 1973; 52: 2745–2756.
Lin MT, Lee RC, Yang PC, Ho FM, and Kuo ML. Cyclooxygenase-2 inducing Mcl-1-dependent survival mechanism in human lung adenocarcinoma cl1.0 cells. involvement of phosphatidylinositol 3-kinase/akt pathway. J Biol Chem. 2001; 276: 48997–49002.
Kuo ML, Chuang SE, Lin MT, and Yang SY. The involvement of PI3-K/Akt-dependent up-regulation of Mcl-1 in the prevention of apoptosis of Hep3B cells by interleukin-6. Oncogene. 2001; 20: 677–685.
Chang MS, Lee WS, Chen BC, Sheu JR, Lin CH. YC-1-Induced cyclooxygenase-2 expression is mediated by cGMP-dependent activations of Ras, phosphoinositide-3-OH-kinase, Akt, and nuclear factor-B in human pulmonary epithelial cells. Mol Pharmacol. 2004; 66: 561–571.
Enari M, Hug H, Nagata S. Involvement of an ICE-like protease in Fas-mediated apoptosis. Nature. 1995; 375: 78–81.
Sheu ML, Ho FM, Chao KF, Kuo ML, Liu SH. Activation of phosphoinositide 3-kinase in response to high glucose leads to regulation of reactive oxygen species-related nuclear factor-kappaB activation and cyclooxygenase-2 expression in mesangial cells. Mol Pharmacol. 2004; 66: 187–196.
Eickelberg O, Roth M, Mussmann R, Rudiger JJ, Tamm M, Perruchoud AP, Block LH. Calcium channel blockers activate the interleukin-6 gene via the transcription factors NF-IL6 and NF-kappaB in primary human vascular smooth muscle cells. Circulation. 1999; 99: 2276–2282.
Cao Y, Prescott SM. Many actions of cyclooxygenase-2 in cellular dynamics and in cancer. J Cell Physiol. 2002; 190: 279–286.
Kim SJ, Chun JS. Protein kinase C alpha and zeta regulate nitric oxide-induced NF-kappa B activation that mediates cyclooxygenase-2 expression and apoptosis but not dedifferentiation in articular chondrocytes. Biochem Biophys Res Commun. 2003; 303: 206–211.
Jovanovic DV, Mineau F, Notoya K, Reboul P, Martel-Pelletier J, Pelletier JP. Nitric oxide induced cell death in human osteoarthritic synoviocytes is mediated by tyrosine kinase activation and hydrogen peroxide and/or superoxide formation. J Rheumatol. 2002; 29: 2165–2175.
Lim JW, Kim H, Kim KH. Nuclear factor-kappaB regulates cyclooxygenase-2 expression and cell proliferation in human gastric cancer cells. Lab Invest. 2001; 81: 349–360.
Creager MA, Luscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part I. Circulation. 2003; 108: 1527–1532.
Ido Y, Carling D, Ruderman N. Hyperglycemia-induced apoptosis in human umbilical vein endothelial cells: inhibition by the AMP-activated protein kinase activation. Diabetes. 2002; 51: 159–167.
Lee JY, Sohn KH, Rhee SH, Hwang D. Saturated fatty acids, but not unsaturated fatty acids, induce the expression of cyclooxygenase-2 mediated through Toll-like receptor 4. J Biol Chem. 2001; 276: 16683–16689.
Chen JH, Hsiao G, Lee AR, Wu CC, Yen MH. Andrographolide suppresses endothelial cell apoptosis via activation of phosphatidyl inositol-3-kinase/Akt pathway. Biochem Pharmacol. 2004; 67: 1337–1345.
Kennedy SG, Wagner AJ, Conzen SD, Jordan J, Bellacosa A, Tsichlis PN, Hay N. The PI 3-kinase/Akt signaling pathway delivers an anti-apoptotic signal. Genes Dev. 1997; 11: 701–713.
Gulbins E, Brenner B, Koppenhoefer U, Linderkamp O, Lang F. Fas or ceramide induce apoptosis by Ras-regulated phosphoinositide-3-kinase activation. J Leukocyte Biol. 1998; 63: 253–263.
Alvarado-Kristensson M, Porn-Ares MI, Grethe S, Smith D, Zheng L, Andersson T. p38 Mitogen-activated protein kinase and phosphatidylinositol 3-kinase activities have opposite effects on human neutrophil apoptosis. FASEB J. 2002; 16: 129–131.
Min YK, Park JH, Chong SA, Kim YS, Ahn YS, Seo JT, Bae YS, Chung KC. Pyrrolidine dithiocarbamate-induced neuronal cell death is mediated by Akt, casein kinase 2, c-Jun N-terminal kinase, and IkappaB kinase in embryonic hippocampal progenitor cells. J Neurosci Res. 2003; 71: 689–700.
Thyrell L, Hjortsberg L, Arulampalam V, Panaretakis T, Uhles S, Dagnell M, Zhivotovsky B, Leibiger I, Grander D, Pokrovskaja K. Interferon alpha–induced apoptosis in tumor cells is mediated through the phosphoinositide 3-kinase/mammalian target of rapamycin signaling pathway. J Biol Chem. 2004; 279: 24152–24162.(Meei Ling Sheu; Feng Ming)
Correspondence to Shing-Hwa Liu, PhD, Institute of Toxicology, College of Medicine, National Taiwan University, No. 1, Section 1, Jen-Ai Rd, Taipei, 10043, Taiwan. E-mail shliu@ha.mc.ntu.edu.tw
Abstract
Objectives— Diabetes mellitus causes endothelial dysfunction. The precise molecular mechanisms by which hyperglycemia causes apoptosis in endothelial cells are not yet well understood. The aim of this study was to explore the role of cyclooxygenase-2 (COX-2) and the possible involvement of phosphoinositide 3-kinase (PI3K) signaling in high glucose (HG)–induced apoptosis in human umbilical vein endothelial cells (HUVECs).
Methods and Results— For detection of apoptosis, the morphological Hoechst staining and Annexin V/propidium iodide staining were used. Glucose upregulated COX-2 protein expression, which was associated with the induction of prostaglandin (PG) E2 (PGE2), caspase-3 activity, and apoptosis. Unexpectedly, we found that PI3K inhibitors could suppress COX-2 expression, PGE2 production, caspase-3 activity, and the subsequent apoptosis under HG condition. Glucose-induced activation of PI3K resulted in the downstream effector Akt phosphorylation. PI3K inhibitors effectively attenuated the intracellular reactive oxygen species (ROS) generation and nuclear factor B (NF-B) activation. Blocking the PI3K and Akt activities with the dominant-negative vectors greatly diminished the HG-triggered NF-B activation and COX-2 expression and apoptosis.
Conclusions— These results suggest that HG, via PI3K/Akt signaling, induces NF-B–related upregulation of COX-2, which in turn triggers the caspase-3 activity that facilitates HUVEC apoptosis. Also, HG may cause ROS generation in HUVECs through a PI3K/Akt–dependent pathway.
We demonstrated for the first time that the signaling pathway of PI3K/Akt–regulated upregulation of COX-2 was involved in glucose-induced apoptosis in human endothelial cells. These findings were supported by results that PI3K inhibition prevented the glucose-caused COX-2–mediated PGE2 production, which subsequently led to the induction of caspase-3 activity and apoptosis.
Key Words: glucose ? endothelium ? cells ? apoptosis
Introduction
Diabetes mellitus can cause a wide variety of vascular complications and cardiovascular dysfunction.1 The mechanisms of hyperglycemia-related tissue damage and clinical complications still remain unclear. Apoptosis is particularly prominent in models of hyperglycemia injury, affecting a significant proportion of vascular endothelium in the tissue damage.2–4 Studies have indicated that oxidative stress can induce apoptosis, which may be regulated by different signaling pathways.5,6 Involvement of reactive oxygen species (ROS) in the induction of apoptosis by high glucose (HG) has been demonstrated in cultured human endothelial cells.7 The formation of oxygen-derived radicals might lead to an activation of transcription factors such as nuclear factor B (NF-B) transferring the activation signal into the nucleus and leading to changes in gene expression necessary for induction of apoptosis. We found previously a c-Jun amino-terminal kinase–dependent pathway was involved in the ROS-related HG-induced apoptosis in human endothelial cells.8 It has also been demonstrated that oxidative stress induced apoptosis in human aortic endothelial cells through the downregulation of bcl-2, translocation of bax, and upregulation of p53, probably through NF-B activation.9 Nevertheless, the precise molecular mechanisms by which HG causes apoptosis in endothelial cells are not yet well understood.
Prostaglandins (PGs) generated by cyclooxygenase (COX) have been implicated in hyperglycemia-induced endothelial dysfunction.10–13 A recent study has shown that HG induced oxidative stress and upregulation of COX-2, resulting in reduced NO availability and altered prostanoid profile in human endothelial cells.14 Production of PG E2 (PGE2), via COX-2 induction, has been demonstrated to be related to the induction of cell death in human osteoarthritis chondrocytes by NO.15 However, the role of COX-2 in HG-caused cell death and the molecular mechanisms regulating COX-2 expression in endothelial cells remain to be clarified. Moreover, it has been suggested that phosphoinositide 3-kinase (PI3K) might play a central role in a diverse range of cellular responses, including cell growth, survival, and malignant transformation.16,17 Recent studies have shown that activation of PI3K and Akt might play a role in regulating COX-2 expression in some cultured cells.18–20 Together, in the current study, we explored the role of COX-2 and the possible involvement of PI3K signaling in HG-induced apoptosis in human endothelial cells. We also examined the role of the PI3K pathway in ROS generation, which might regulate the HG-induced apoptosis in endothelial cells.
Methods
Cell Culture
Human umbilical vein endothelial cells (HUVECs) were obtained by collagenase treatment of umbilical cord veins as described previously.8,21 Cells were cultured on gelatin-coated dishes and propagated in M199 medium supplemented with 20% bovine calf serum (Gibco), 50 μg/mL endothelial cell growth factor, 90 μg/mL heparin, 100 IU/mL penicillin, and 0.1 mg/mL streptomycin in humidified air, 5% CO2 at 37°C. HUVECs of the third to fifth passages were used. In some experiments, transfection in HUVECs was performed using the Lipofectin reagent (BD CLONTECH) according to manufacturer recommendations with 2 μg of total DNA (or vector) per sample. The efficiency of transfection (80%) was determined by using an equal amount of a plasmid encoding the green fluorescent protein under the cytomegalovirus promoter. Plasmids containing the dominant-negative (DN)-p85 (-p85) or DN-Akt (Akt K179A)22,23 were kindly provided by Drs R.H. Chen and M.L. Kuo (Institutes of Molecular Medicine and Toxicology, National Taiwan University); DN-inhibitor B (IB)24 was provided by Dr W.W. Lin. (Institutes of Pharmacology, National Taiwan University).
Apoptosis and Caspase-3/Cleaved Caspase 3 Activity
HUVECs were collected and fixed in methanol/acetone solution for 5 minutes and washed with PBS. Fixed cells were then stained with 0.1 ng/mL Hoechst 33258 (Boehringer Mannheim) for 10 minutes in the dark to counterstain nuclei. Cells were observed and photographed under a Nikon fluorescence microscope. In some experiments, the Annexin V/propidium iodide (PI; BD CLONTECH) was used to quantify numbers of apoptotic cells. Cells were washed twice with PBS and stained with Annexin V and PI for 20 minutes at room temperature. The level of apoptosis was determined by measuring the fluorescence of the cells by flow cytometry analysis.
Also, we measured caspase-3 activity to quantify the apoptotic ratio. Caspase-3 activity was measured as described previously.8,25 Cell lysates were incubated at 37°C with 10 μmol/L N-acetyl-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin (Ac-DEVD-AMC), a caspase-3/Cleaved Caspase 3 (CPP32) substrate. The fluorescence of the cleaved substrate was measured by a spectrofluorometer with an excitation wavelength at 380 nm and an emission wavelength at 460 nm.
Cell Viability
Viability was measured using the Cell Titer 96TM AQueous cell viability assay kit (Promega). This assay is based on the cellular conversion of the colorimetric reagent MTS (3,4-(5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium salt), in the presence of electron-coupling reagent phenazine methosulfate, into soluble formazan by dehydrogenase enzymes found only in metabolically active, living cells. Formazan formation was measured on the basis of increased absorbance at 490 nm.
PI3K Activity
PI3K activity was assayed as described previously.26 Cell extracts were incubated with 2 μg of anti-p85 antibody overnight at 4°C. The immunocomplex was precipitated with protein A-Sepharose for 1 hour at 4°C and washed 3x with lysis buffer, twice with LiCl buffer, and twice with Tris-NaCl-EDTA (TNE) buffer (10 mmol/L Tris, 100 mmol/L NaCl, and 1 mmol/L EDTA). The immunocomplex was preincubated with 20 mmol/L Hepes, containing 2 mg/mL PI on ice for 10 minutes. Kinase reaction was performed by adding reaction buffer (10 μCi of [-32P]ATP, 20 mmol/L Hepes, pH 7.4, 20 μmol/L ATP, and 5 mmol/L MgCl2) for 15 minutes. The radiolabeled lipids were separated by thin-layer chromatography and visualized by phosphorimaging.
Immunoblotting
Protein levels of Akt, phospho-Akt, and COX-2 were analyzed by Western blot as described previously.26 Cell lysates were prepared, electrotransferred, and then immunoblotted with anti-Akt, anti–phospho-Akt (New England BioLabs), and anti–COX-2 antibodies (Santa Cruz Biotechnology). Detection was performed with Western blotting reagent ECL (Amersham), and chemiluminescence was exposed by the filters of Kodak X-Omat films.
Electrophoretic Mobility Shift Assay
Electrophoretic mobility shift assay (EMSA) was performed as described previously.27 The following oligonucleotide with the NF-B consensus binding sequence was used: 5'-GATCCAAGGGGACTTTCCATGGAT-CCAAGGGGACTTTCCATG-3' (Life Technologies). The consensus oligonucleotide probes were end-labeled with [-32P]ATP according to the manufacturer description. Protein–DNA complexes were separated by electrophoresis on a nondenaturing 6% polyacrylamide gel and visualized by autoradiography. For competition experiments, a 100-fold molar excess of the unlabeled oligonucleotides was added 15 minutes before incubation of nuclear extracts with the end-labeled oligonucleotides.
ROS Production
Intracellular ROS generation was monitored by flow cytometry using peroxide-sensitive fluorescent probe 2',7'-dichlorofluorescin diacetate (DCFH-DA; Molecular Probes) as described previously.8 DCFH-DA is converted by intracellular esterases to DCFH, which is oxidized into the highly fluorescent dichlorofluorescin (DCF) in the presence of a proper oxidant, and then analyzed by flow cytometry.
PGE2 Production
PGE2 levels in cell culture supernatants were determined using the PGE2 enzyme immunoassay kit (Cayman Chemical).
Statistical Analyses
Results are expressed as mean±SEM and data analyzed by Student t test. For multiple comparisons, results were analyzed by ANOVA followed by Fisher’s test. P<0.05 was considered statistically significant.
Results
COX-2 Protein Expression and Involvement in HG-Induced Apoptosis
HUVECs exposed to HG (30 mmol/L) for 8 to 12 hours increased COX-2 protein expression (4.2-fold induction; Figure 1Aa), whereas no changes were observed for COX-1 protein level (data not shown). HG-induced increase of COX-2 expression was associated with a higher release of PGE2, which could be inhibited by selective COX-2 inhibitor NS398 (10 and 20 μmol/L; Figure 1Ba). We also found that combining HG with PGE2 (1 μmol/L) showed an additive effect on HUVEC death (Figure 2A). Moreover, we examined whether COX-2–mediated PGE2 generation was involved in the HG-induced apoptosis. The inhibitory effect of NS398 on HG-induced apoptosis was demonstrated in caspase-3 activity assay (Figure 2B) and morphological characteristic Hoechst-33258 staining (Figure 2C). The HG-induced cell death could be reversed by NS398 (Figure 3A). Mannitol (30 mmol/L), used as an osmotic control, did not affect COX-2 expression or apoptosis and cell viability (Figures 1A, 2C, and 3A).
Figure 1. Aa, Expression of COX-2 protein in HUVECs exposed to control (5.5 mmol/L) and 30 mmol/L glucose (HG) or mannitol (HM) for 8 to 12 hours. Ab, Expression of COX-2 protein in HUVECs exposed to HG for 12 hours in the presence or absence of LY294002 (LY) and wortmannin (WM). In some experiments, cells were transfected with the dominant-negative forms of PI3K-p85 (DN-p85), and then the HG-induced COX-2 protein expressions were determined (Ac). Results shown are representative of at least 4 independent experiments. Ba, Bar graphs showing release of PGE2 in HUVECs exposed to HG in the presence or absence of NS398 (10 and 20 μmol/L). Bb, Bar graphs showing the effect of LY294002 (10 and 20 μmol/L) on the release of PGE2 in HUVECs exposed to HG. All data are presented as mean±SEM from 5 experiments performed in duplicate. *P<0.05 vs control; #P<0.05 vs HG alone.
Figure 2. A, Cells were treated with HG or PGE2 or HG+PGE2 for 48 hours and then cell viability detected. B, Analysis of caspase-3/CPP32 activity in HUVECs treated with HG for 48 hours in the presence or absence of NS398 (20 μmol/L). All data are presented as mean±SEM from 5 experiments performed in duplicate. *P<0.05 vs control; #P<0.05 vs HG alone. C, Apoptosis induced by HG was determined by fluorescent dye Hoechst-33258 method. Ca, Control; Cb, HG exposure for 48 hours; Cc, HG+NS398; Cd, high mannitol (HM) exposure for 48 hours. Results shown are representative of at least 4 independent experiments.
Figure 3. A, HUVECs were treated with HG for 72 hours in the presence or absence of PI3K inhibitor LY294002 (LY; 20 μmol/L) or COX-2 inhibitor NS398 (20 μmol/L), and then cell viability was detected. B, Analysis of caspase-3 activity. HUVECs pretreated with HG for 48 hours showed increase of caspase-3/CPP32 activity. LY294002 (LY; 20 μmol/L) prevented the effect of HG. All data are presented as mean±SEM from 5 independent experiments performed in duplicate. *P<0.05 vs control; #P<0.05 vs HG alone. C, Apoptosis induced by HG was determined by fluorescent dye Hoechst 33258 method. Ca, Control; Cb, HG exposure for 48 hours; Cc, HG+LY294002 (LY); Cd, HG+wortmannin (WM/100 nmol/L). Results shown are representative of at least 4 independent experiments.
PI3K Activity and Involvement in HG-Induced COX-2 Protein Expression and Apoptosis
HUVECs were cultured with HG for 12 to 72 hours in the presence or absence of PI3K inhibitors (LY294002 and wortmannin). HG-induced increased COX-2 protein expression and its related PGE2 release and cell death could be inhibited by LY294002 (10 or 20 μmol/L) or wortmannin (100 nmol/L; Figure 1Ab and 1Bb; Figure 3A). Moreover, the inhibitory effect of PI3K inhibitors on HG-induced apoptosis was demonstrated in caspase-3 activity assay (Figure 3B), morphological characteristic Hoechst 33258 staining (Figure 3C), and Annexin V/PI assay (Figure 4Ca).
Figure 4. A and B, HG activated the PI3K activity (A) and the phosphorylation of Akt (B). HUVECs were treated with HG in the presence or absence of LY294002 (LY; 10 or 20 μmol/L) for 5 to 10 (A) or 10 to 60 (B) minutes. Results shown are representative of at least 4 independent experiments. C, HUVECs were treated with HG in the absence or presence of LY294002 (LY; 20 μmol/L). Apoptosis induced by HG was determined by Annexin V/PI staining. Data represent percentage of Annexin V–positive cells (Ca). In some experiments, HUVECs were transiently transfected with DN-p85 or DN-Akt. Cell viability was detected by MTS assay (Cb). D, Analysis of caspase-3 activity. HUVECs were treated with HG in the absence or presence of LY294002 (LY; 20 μmol/L) or transfection of DN-p85 or DN-Akt. All data are presented as mean±SEM from 5 independent experiments performed in duplicate. *P<0.05 vs control; #P<0.05 vs HG alone.
Furthermore, the authentic PI3K activity of HUVECs after treatment with HG was determined. Immunoprecipitates with the anti-p85 antibody revealed a substantial increase in PI3K activity 5 to 10 minutes after HG exposure, which could be inhibited by LY294002 (10 μmol/L; Figure 4A). To further evaluate the involvement of Akt in HG-triggered responses, Akt activity was examined by determining the serine-phosphorylated status of Akt using an anti–phospho-Akt antibody. In a time-dependent manner, Akt activation was shown 10 to 60 minutes after HG treatment (Figure 4B). LY294002 markedly reduced this HG-induced response (Figure 4B).
The next aim of the investigation was to ascertain whether PI3K/Akt activity inhibition might affect the HG-induced COX-2 protein expression and apoptosis in HUVECs. To address this issue, cells were transfected with a dominant-negative form of the regulatory subunit (p85) of PI3K (DN-p85) or a dominant-negative form of Akt (DN-Akt).26 Transfection results revealed that the DN-p85 or DN-Akt transfection significantly inhibited HG-induced COX-2 protein expression (Figure 1Ac) and decreased cell viability (Figure 4Cb) and increased caspase-3 activity (Figure 4D) and apoptosis (Annexin/PI staining; Figure 5A). The control pcDNA3 transfection did not affect the HG-induced responses (data not shown).
Figure 5. The HG-induced apoptosis in HUVECs requires activation of PI3K. HUVECs were transiently transfected with DN-p85, DN-Akt, or DN-IB. A, Cells were analyzed for apoptosis by Annexin V/PI staining. The percentage of cells found in each quadrant of the dot plot is depicted. Results shown are representative of at least 4 independent experiments. B, Nuclear extracts were subjected to an EMSA. Cells were treated with HG in the presence or absence of LY294002 (LY; 20 μmol/L) for 4 hours. Quantification of the NF-B binding activity was performed by densitometric analysis. *P<0.05 vs control; #P<0.05 vs HG alone
Effects of PI3K Inhibitors on HG-Induced NF-B Activation and ROS Generation
As shown in Figure 5B, stimulation of HUVECs with HG for 4 hours showed a marked increase of NF-B activation, as determined by EMSA. LY294002 and transfection of DN-p85, DN-Akt, or DN-IB abolished this HG-induced NF-B activation. Moreover, cells transfected with a dominant-negative form of IB (DN-IB) significantly inhibited HG-induced COX-2 protein expression (Figure 6A), caspase-3 activity (HG 2.1±0.2; HG+DN-IB 1.1±0.3-fold of control; n=3; P<0.05), apoptosis (Figure 6B), and cell death (Figure 6C). Phorbol 12-myristate 13-acetate (PMA)–induced COX-2 protein expression in HUVECs could also be inhibited by DN-IB transfection (as a positive control; Figure 6A). We next investigated whether the HG-induced generation of DCF-sensitive ROS requires activity of PI3K. Treatment with HG on HUVECs for 36 to 48 hours was found to increase DCF fluorescence. LY294002 (20 μmol/L) treatment completely suppressed this HG-induced response (Figure 6D).
Figure 6. HUVECs were treated with HG in the presence or absence of transfection of DN-IB. A, Expression of COX-2 protein in HUVECs exposed to HG or PMA for 12 hours. Quantification of the COX-2 expression was performed by densitometric analysis. B, Apoptosis induced by HG was determined by Annexin V/PI staining. Data represent percentage of Annexin V–positive cells. C, HUVECs were treated with HG for 72 hours, and cell viability was detected by MTS assay. D, HG-induced intracellular DCF-sensitive ROS generation. HUVECs were exposed to HG for 36 to 48 hours in presence or absence of LY294002, antioxidants (ascorbic acid [vitamin C]; 100 μmol/L), diphenyleneiodonium (DPI; 20 μmol/L), or rotenone (ROT; 20 μmol/L). All data are presented as mean±SEM from 4 independent experiments. *P<0.05 vs control; #P<0.05 vs HG or PMA alone.
Discussion
In the present study, we demonstrate for the first time that HG causes PI3K/Akt–dependent upregulation of inducible COX-2 expression, as well as an increase of PGE2 production and increased caspase-3 activity and induction of apoptosis in HUVECs. These findings are supported by functional studies of HUVECs showing that the activity of PI3K and Akt is triggered by HG, and the inhibition of PI3K/Akt activity with LY294002 or wortmannin, or by expressing the dominant-negative p85 or Akt is capable of preventing the HG-caused COX-2–mediated PGE2 production, and subsequently leads to the inhibition of caspase-3 activity and apoptosis. Moreover, HG induces a PI3K/Akt–regulated ROS generation. Hence, activation of the PI3K/Akt pathway may represent a proximal node in the intracellular signaling leading to hyperglycemia-induced endothelial cell apoptosis.
In a number of cell and animal models, induction of COX-2 has been shown to promote cell growth, inhibit apoptosis, and enhance cell motility and adhesion.28 Paradoxically, PGE2 production via the induction of COX-2 expression has been demonstrated to relate to the induction of cell death in some cultured cells such as human articular chondrocytes15,29 and human synovial fibroblasts.30 Similarly, in the present study, we found that HG induced COX-2 protein expression in HUVECs, which was associated with a higher release of PGE2. Selective COX-2 inhibitor NS398 inhibited PGE2 production and blocked increased caspase-3 activity and apoptosis induced by HG. Combining HG with PGE2 showed an additive effect on cell death. Moreover, NF-B was known as a transcriptional regulator of expression of COX-2 gene.31 NF-B has also been demonstrated as a mediated signaling in HG (hyperglycemia) or oxidative stress-triggered apoptosis of endothelial cells.7,9 Activation of NF-B promoted p53 upregulation, which might suppress the level of bcl-2, resulting in endothelial cell apoptosis.9 We also observed that HG could trigger the NF-B activation in HUVECs, and dominant-negative IB transfection could inhibit HG-induced COX-2 protein expression and apoptosis. Therefore, these findings suggest an involvement of COX-2–mediated PGE2 production in HG-induced HUVEC apoptosis through activation of NF-B. On the other hand, in the clinical setting, hyperglycemia was usually accompanied by increased free fatty acid levels, which might also have detrimental effects on endothelial cells.32 It has been suggested that the alterations in fatty acid metabolism, impaired Akt activation by insulin, and increased caspase-3 activity preceded visible evidence of apoptosis in HUVECs incubated in a hyperglycemic medium.33 It has also been identified that saturated fatty acids induced the expression of COX-2 in monocyte/macrophage cells.34 Hence, the involvement of free fatty acid in the hyperglycemia-induced COX-2 expression and apoptosis in HUVECs may need to clarify in the future.
PI3K signal plays an important regulatory role in a diverse range of cellular responses, including cell growth, survival, and malignant transformation.16,17 It has been shown that PI3K activation has an antiapoptotic effect in endothelial cells35 and elsewhere.36 Unexpectedly, we observed that the treatment of HG in HUVECs leads to activation of PI3K and Akt, which subsequently leads to activation of NF-B and COX-2 expression, and then induces apoptosis. At present, the physiological relevance of the activation of PI3K and Akt during HG-induced HUVEC death is not clear. In accordance with our present findings, such an apoptosis-promoting effect of PI3K has been reported in other cell systems. For example, a rapid and transient activation of PI3K was critical for Fas or C6-ceramide–induced apoptosis in Jurkat cells;37 a significant increase of PI3K activity was observed during Fas-induced human neutrophil apoptosis and suggested that Fas-induced PI3K activity was a proapoptotic signal;38 and activation of PI3K/Akt was induced during pyrrolidine dithiocarbamate–induced neuronal cell death.39 Interferon-–induced apoptosis in tumor cells was mediated through the PI3K/mammalian target of rapamycin signaling pathway.40 Recent studies have also shown that activation of PI3K might play a role in regulating the COX-2 expression in some cultured cells.18–20 Therefore, the current results indicate that PI3K/Akt may be involved in the HG-induced HUVEC apoptosis, and PI3K/Akt is present upstream of NF-B activation and COX-2 expression in the HG-triggered signaling pathways.
In conclusion, our study demonstrates that activation of a PI3K/Akt–signaling pathway is required for HG-induced activation of NF-B and COX-2–mediated PGE2 production and subsequent caspase-3 activation and facilitation of apoptosis in HUVECs. Also, HG triggers ROS generation through a PI3K-dependent pathway. These findings may be relevant in understanding the possible intracellular signaling associated with diabetic hyperglycemia-induced detrimental effects on endothelial cells and provide novel therapeutic strategies to prevent the development and progression of vascular complications in diabetes mellitus.
Acknowledgments
This work was supported by research grants from National Science Council of Taiwan (NSC90-2315-B-002-009; NSC91-2314-B-002-248). The authors thank Professor Ming Liang Kuo (National Taiwan University, Taipei) for discussion regarding PI3K activity.
Received September 19, 2004; accepted December 30, 2004.
References
The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993; 329: 977–986.
Ceriello A, Quagliaro L, D’Amico M, Di Filippo C, Marfella R, Nappo F, Berrino L, Rossi F, Giugliano D. Acute hyperglycemia induces nitrotyrosine formation and apoptosis in perfused heart from rat. Diabetes. 2002; 51: 1076–1082.
Nakagami H, Morishita R, Yamamoto K, Yoshimura SI, Taniyama Y, Aoki M, Matsubara H, Kim S, Kaneda Y, Ogihara T. Phosphorylation of p38 mitogen-activated protein kinase downstream of bax-caspase-3 pathway leads to cell death induced by high D-glucose in human endothelial cells. Diabetes. 2001; 50: 1472–1481.
Zou MH, Shi C, Cohen RA. High glucose via peroxynitrite causes tyrosine nitration and inactivation of prostacyclin synthase that is associated with thromboxane/prostaglandin H(2) receptor-mediated apoptosis and adhesion molecule expression in cultured human aortic endothelial cells. Diabetes. 2002; 51: 198–203.
Buttke TM, Sandstrom PA. Oxidative stress as a mediator of apoptosis. Immunol Today. 1994; 15: 7–10.
Vaux DL, Strasser A. The molecular biology of apoptosis. Proc Natl Acad Sci U S A. 1996; 93: 2239–2244.
Du X, Stocklauser-Farber K, Rosen P. Generation of reactive oxygen intermediates, activation of NF-kappaB, and induction of apoptosis in human endothelial cells by glucose: role of nitric oxide synthase? Free Radic Biol Med. 1999; 27: 752–763.
Ho FM, Liu SH, Liau CS, Huang PJ, Lin-Shiau SY. High glucose-induced apoptosis in human endothelial cells is mediated by sequential activations of c-Jun NH(2)-terminal kinase and caspase-3. Circulation. 2000; 101: 2618–2624.
Aoki M, Nata T, Morishita R, Matsushita H, Nakagami H, Yamamoto K, Yamazaki K, Nakabayashi M, Ogihara T, Kaneda Y. Endothelial apoptosis induced by oxidative stress through activation of NF-kappaB: antiapoptotic effect of antioxidant agents on endothelial cells. Hypertension. 2001; 38: 48–55.
Ayalasomayajula SP, Kompella UB. Celecoxib, a selective cyclooxygenase-2 inhibitor, inhibits retinal vascular endothelial growth factor expression and vascular leakage in a streptozotocin-induced diabetic rat model. Eur J Pharmacol. 2003; 458: 283–289.
Gryglewski RJ, Chlopicki S, Uracz W, Marcinkiewicz E. Significance of endothelial prostacyclin and nitric oxide in peripheral and pulmonary circulation. Med Sci Monit. 2001; 7: 1–16.
Meeking DR, Browne DL, Allard S, Munday J, Chowienczyck PJ, Shaw KM, Cummings MH. Effects of cyclo-oxygenase inhibition on vasodilatory response to acetylcholine in patients with type 1 diabetes and nondiabetic subjects. Diabetes Care. 2000; 23: 1840–1843.
Tesfamariam B. Free radicals in diabetic endothelial cell dysfunction. Free Radic Biol Med. 1994; 16: 383–391.
Cosentino F, Eto M, De Paolis P, van der LB, Bachschmid M, Ullrich V, Kouroedov A, Delli GC, Joch H, Volpe M, Luscher TF. High glucose causes upregulation of cyclooxygenase-2 and alters prostanoid profile in human endothelial cells: role of protein kinase C and reactive oxygen species. Circulation. 2003; 107: 1017–1023.
Notoya K, Jovanovic DV, Reboul P, Martel-Pelletier J, Mineau F, Pelletier JP. The induction of cell death in human osteoarthritis chondrocytes by nitric oxide is related to the production of prostaglandin E2 via the induction of cyclooxygenase-2. J Immunol. 2000; 165: 3402–3410.
Krasilnikov MA. Phosphatidylinositol-3 kinase dependent pathways: the role in control of cell growth, survival, and malignant transformation. Biochemistry (Mosc). 2000; 65: 59–67.
Toker A. Protein kinases as mediators of phosphoinositide 3-kinase signaling. Mol Pharmacol. 2000; 57: 652–658.
Sheng H, Shao J, Dubois RN. K-Ras-mediated increase in cyclooxygenase 2 mRNA stability involves activation of the protein kinase B1. Cancer Res. 2001; 61: 2670–2675.
Tang Q, Gonzales M, Inoue H, Bowden GT. Roles of Akt and glycogen synthase kinase 3beta in the ultraviolet B induction of cyclooxygenase-2 transcription in human keratinocytes. Cancer Res. 2001; 61: 4329–4332.
Weaver SA, Russo MP, Wright KL, Kolios G, Jobin C, Robertson DA, Ward SG. Regulatory role of phosphatidylinositol 3-kinase on TNF-alpha–induced cyclooxygenase 2 expression in colonic epithelial cells. Gastroenterology. 2001; 120: 1117–1127.
Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest. 1973; 52: 2745–2756.
Lin MT, Lee RC, Yang PC, Ho FM, and Kuo ML. Cyclooxygenase-2 inducing Mcl-1-dependent survival mechanism in human lung adenocarcinoma cl1.0 cells. involvement of phosphatidylinositol 3-kinase/akt pathway. J Biol Chem. 2001; 276: 48997–49002.
Kuo ML, Chuang SE, Lin MT, and Yang SY. The involvement of PI3-K/Akt-dependent up-regulation of Mcl-1 in the prevention of apoptosis of Hep3B cells by interleukin-6. Oncogene. 2001; 20: 677–685.
Chang MS, Lee WS, Chen BC, Sheu JR, Lin CH. YC-1-Induced cyclooxygenase-2 expression is mediated by cGMP-dependent activations of Ras, phosphoinositide-3-OH-kinase, Akt, and nuclear factor-B in human pulmonary epithelial cells. Mol Pharmacol. 2004; 66: 561–571.
Enari M, Hug H, Nagata S. Involvement of an ICE-like protease in Fas-mediated apoptosis. Nature. 1995; 375: 78–81.
Sheu ML, Ho FM, Chao KF, Kuo ML, Liu SH. Activation of phosphoinositide 3-kinase in response to high glucose leads to regulation of reactive oxygen species-related nuclear factor-kappaB activation and cyclooxygenase-2 expression in mesangial cells. Mol Pharmacol. 2004; 66: 187–196.
Eickelberg O, Roth M, Mussmann R, Rudiger JJ, Tamm M, Perruchoud AP, Block LH. Calcium channel blockers activate the interleukin-6 gene via the transcription factors NF-IL6 and NF-kappaB in primary human vascular smooth muscle cells. Circulation. 1999; 99: 2276–2282.
Cao Y, Prescott SM. Many actions of cyclooxygenase-2 in cellular dynamics and in cancer. J Cell Physiol. 2002; 190: 279–286.
Kim SJ, Chun JS. Protein kinase C alpha and zeta regulate nitric oxide-induced NF-kappa B activation that mediates cyclooxygenase-2 expression and apoptosis but not dedifferentiation in articular chondrocytes. Biochem Biophys Res Commun. 2003; 303: 206–211.
Jovanovic DV, Mineau F, Notoya K, Reboul P, Martel-Pelletier J, Pelletier JP. Nitric oxide induced cell death in human osteoarthritic synoviocytes is mediated by tyrosine kinase activation and hydrogen peroxide and/or superoxide formation. J Rheumatol. 2002; 29: 2165–2175.
Lim JW, Kim H, Kim KH. Nuclear factor-kappaB regulates cyclooxygenase-2 expression and cell proliferation in human gastric cancer cells. Lab Invest. 2001; 81: 349–360.
Creager MA, Luscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part I. Circulation. 2003; 108: 1527–1532.
Ido Y, Carling D, Ruderman N. Hyperglycemia-induced apoptosis in human umbilical vein endothelial cells: inhibition by the AMP-activated protein kinase activation. Diabetes. 2002; 51: 159–167.
Lee JY, Sohn KH, Rhee SH, Hwang D. Saturated fatty acids, but not unsaturated fatty acids, induce the expression of cyclooxygenase-2 mediated through Toll-like receptor 4. J Biol Chem. 2001; 276: 16683–16689.
Chen JH, Hsiao G, Lee AR, Wu CC, Yen MH. Andrographolide suppresses endothelial cell apoptosis via activation of phosphatidyl inositol-3-kinase/Akt pathway. Biochem Pharmacol. 2004; 67: 1337–1345.
Kennedy SG, Wagner AJ, Conzen SD, Jordan J, Bellacosa A, Tsichlis PN, Hay N. The PI 3-kinase/Akt signaling pathway delivers an anti-apoptotic signal. Genes Dev. 1997; 11: 701–713.
Gulbins E, Brenner B, Koppenhoefer U, Linderkamp O, Lang F. Fas or ceramide induce apoptosis by Ras-regulated phosphoinositide-3-kinase activation. J Leukocyte Biol. 1998; 63: 253–263.
Alvarado-Kristensson M, Porn-Ares MI, Grethe S, Smith D, Zheng L, Andersson T. p38 Mitogen-activated protein kinase and phosphatidylinositol 3-kinase activities have opposite effects on human neutrophil apoptosis. FASEB J. 2002; 16: 129–131.
Min YK, Park JH, Chong SA, Kim YS, Ahn YS, Seo JT, Bae YS, Chung KC. Pyrrolidine dithiocarbamate-induced neuronal cell death is mediated by Akt, casein kinase 2, c-Jun N-terminal kinase, and IkappaB kinase in embryonic hippocampal progenitor cells. J Neurosci Res. 2003; 71: 689–700.
Thyrell L, Hjortsberg L, Arulampalam V, Panaretakis T, Uhles S, Dagnell M, Zhivotovsky B, Leibiger I, Grander D, Pokrovskaja K. Interferon alpha–induced apoptosis in tumor cells is mediated through the phosphoinositide 3-kinase/mammalian target of rapamycin signaling pathway. J Biol Chem. 2004; 279: 24152–24162.(Meei Ling Sheu; Feng Ming)